Response to Comments: MolDX: Foodborne Gastrointestinal Panels Identified by Multiplex Nucleic Acid Amplification (NAATs)

The multiplex DNA extraction PCR technology is one of the most important advances in clinical medicine over the last several decades. For syndromic illnesses such as diarrhea, either acute or chronic, the ability to accurately and rapidly make a microbial diagnosis is paramount in terms of clinical management.

As you may know I specialize in gastroenterology, internal medicine, tropical and travel medicine. I have had considerable clinical and research experience in the field of travelers’ diarrhea over the past 25+ years. One of the problems in diagnosis has been the fact that the presentation of diarrhea is often protean and bacterial, parasitic and viral etiologies are often clinically indistinguishable but approached quite differently as far as treatment is concerned. Before the availability of the DNA extraction PCR, diagnosis was dependent upon doing stool cultures which often required special selective media and often took 48 to 72 hours. At our hospital, one of the largest teaching hospitals in the country, prior to this technology, only 5 bacterial pathogens were searched for, leaving the vast majority of pathogens undiagnosed. Likewise for parasitic causes, requiring ova and parasite (O & P) examination by microscopy with or without special stains is fraught with inaccuracies due to the fact that this procedure is technician dependent with propensity to either over-read or under-read. Viral pathogens have been difficult to diagnose because of lack of adequate commercial testing capability.

Efforts to limit the diagnostic possibilities to the 4 or 5 most likely pathogens is not only a bad practice but counter-intuitive. By doing this, this will reinforce the bad practices of the past few decades and we will continue to miss the vast majority of specific microbial pathogens. For travelers’ diarrhea this is particularly important as there are etiologic causes, such as Giardia, Cyclospora, and diarrheagenic E.coli which are often not looked for with traditional testing

Likewise, for community acquired diarrhea, even if the majority of cases are viral, there is value in proving a viral origin and withholding antibiotics in the interest of anti-microbial stewardship and prevention of antimicrobial resistance.

In a recent study of the BioFire GI panel compared to traditional testing methods, the cohort of patients tested on the GI panel were found to have more targeted rather than empiric therapy. In cases where antibiotics were not indicated, like in patients that present with Shiga-toxin producing E. coli, the study showed discontinuation of empiric antimicrobials 47 hours faster than traditional methods (Cybulski, 2018). In another recent study, patients tested on the BioFire GI panel should a decrease in downstream testing methods, like CT scans, x-rays, and ultrasounds compared to traditional methods likely because the physician was able to identify an infectious cause and stop looking for possible causes of diarrhea (Beal, 2018). Without the use of a comprehensive, fast, and accurate test many of the benefits observed in these studies would likely not be realized.

Another point that the draft highlights is the necessity to only detect five common known causes of infectious diarrhea, mainly bacterial causes. The identification of viral and parasitic causes of acute diarrhea are also important and can inform not only decisions to treat or not to treat, but also infection control decisions, recommendations to return to work, school, or other activities that may have an impact to public health. Additionally, studies have shown that treatment for parasitic infections can effectively shorten illness (Rossignol, 2001) and in other cases without treatment be prolonged for several months (Beal, 2018. MacKenzie, 1994) without the appropriate course of therapy. Since potential viral and parasitic causes may not be routinely tested for or have poor sensitivity, yet they make up a substantial portion of diarrheal burden in the US there presence on a multiplex NAAT is appropriate.

In addition, current practice guidelines (ACG 2016 and ISTM 2017) recommend the use of culture-independent molecular multiplex tests for the identification of potential causes of acute diarrhea because of their ability to impact patient management decisions. The guidelines offer a framework for clinicians to consider when considering what patients should be tested, treatment, as well as other patient management recommendations. The guidelines reviewed existing literature and relied on expert opinions to make recommendations that include and recommend the use of culture-independent molecular multiplex tests how they can be utilized to better inform clinicians of what could be making their patients sick.

We also know that diarrheal disease is not the benign self-limited condition we once thought it was. There is a not insignificant incidence of post-infectious sequelae such as chronic GI symptoms following acute diarrhea such as post-infectious irritable bowel syndrome (PI-IBS), post-infections functional gastrointestinal disease (PI-FGD), as well as hemolytic uremic syndrome, reactive arthritis and Guillain Barre syndrome There is a suggestion that prompt diagnosis and treatment may reduce or eliminate these post-infectious sequelae and reducing morbidity and potential mortality.

Multiplex PCR technology also has the potential benefit of reducing use of healthcare resources and reducing overall healthcare costs by providing comprehensive and accurate infectious disease results in a clinically actionable time frame. As a gastroenterologist, the need to perform endoscopic procedures in my practice has declined as a result of utilizing this diagnostic technology. I have had patients with chronic diarrhea scheduled for multiple endoscopic procedures, who with one stool specimen, were found to have pathogens such as Giardia sparing them the expense and inconvenience of these invasive procedures.

In summary, the latest draft of the LCD that impacts molecular multiplex GI panels should be reconsidered and include favorable reimbursement for such tests. Molecular multiplex GI tests have been shown to provide faster, comprehensive, accurate results that can lead to more targeted therapy/patient management decisions that lead to better patient outcomes and likely patient satisfaction. This technology has changed clinical practice and revolutionized the field of diarrheal disease. I would strongly advise against limiting the number of targets and would encourage further dialogue. I am available to speak any time convenient for you.

Based on comments received, we are expanding the covered procedures to include coverage of testing up to 11 pathogens. While it is possible that numerous additional pathogens could potentially be identified through molecular methods, we would generally expect that this would be unnecessary, either because the number of pathogens to be tested could be narrowed based on patient history, or because there is not a clear clinical utility to positive identification. The ACG Guidelines cited by the commenter above give an evaluation and treatment algorithm which in many cases would not rely on microbiologic assessment. Additionally, this guideline notes a number of shortcomings of culture-independent diagnostic modalities including the possibility of identifying the presence of nucleic acids of pathogens that are not viable and not necessarily clinically significant in a given patient. In summary, while molecular diagnostic techniques are able to identify many pathogens, the clinical utility of doing so has not been established

Thank you for this opportunity to respond to your draft local coverage determination regarding MolDX: Foodborne Gastrointestinal Panels Identified by Multiplex Nucleic Acid Amplification Tests (NAATs) (DL37709). The American Gastroenterological Association (AGA), American Society for Microbiology (ASM), Association for Molecular Pathology (AMP), Association of Public Health Laboratories, (APHL); College of American Pathologists (CAP), Infectious Diseases Society of America (IDSA), and Pan American Society for Clinical Virology (PASCV), representing multiple areas of practice, have collaborated to present the most thorough analysis for your draft local coverage determination. The members of the six organizations developing these comments are subject matter experts in diagnosis and treatment of the gastrointestinal conditions covered by this policy and its possible implementation will directly impact their practices. We are submitting joint comments because our organizations share the same concerns regarding this draft local coverage determination (LCD). Furthermore, to assist Palmetto, this letter provides specific recommended LCD language changes on medical necessity followed by supporting commentary and literature evaluations. In addition, there is a recommended configuration of ICD-10 codes, which support medical necessity language.

We recommend modifying the paragraph titled “Coverage Indications, Limitations, and/or Medical Necessity” to read as follows:

(Please note: strikethrough font indicates recommended deletion of original draft LCD language; bolded blue font indicates recommended additions)

This contractor will provide limited coverage for Gastrointestinal Pathogen (GIP)molecular assays identified by multiplex nucleic acid amplification tests (NAATs), and will limit GIP coverage up to 5 bacterial targets which represent to the top 90-95% of foodborne infections ([incidence of infection per 100,000 population]in decreasing incidence): Salmonella [15.89]; Campylobacter [12.97]; Shigella [5.53]; Cryptosporidium [3.31]; Shiga toxin producing E. coli (STEC) non-O157 [1.64] and STEC O157 [.95]. In immune competent individuals, most people with Cryptosporidium, a parasitic disease, will recover without treatment. The pathogens in some of the GIP panels are determined by the manufacturers that make them, and do not represent specific pathogens that cause a common age-based syndrome, or represent organisms that commonly are found in a specific sample type, patient population or reflect community acquired foodborne infections. considered medically necessary for therapeutic decision making. These infectious agents include Salmonella, Campylobacter, Shigella, Shiga toxin- producing E. coli (STEC) non-O157 and STEC O157, as well as enterotoxigenic E. coli, enteropathogenic E. coli, enteroaggregative E. coli, Clostridium difficile, Yersinia enterocolitica, Vibrio group (consisting of Vibrio parahaemolyticus or Vibrio vulnificus or Vibrio cholera, of which Vibrio parahemoltyicus is most common), Giardia, Cryptosporidium, norovius, rotavirus, and enteric adenoviruses. Diarrheal illnesses pose a considerable diagnostic challenge, where the history, presenting signs and symptoms, and other features are often non-specific, but effective therapy requires precise microbial identification. This clinical need has led to a new generation of diagnostic panels that cover this broad range of likely infectious etiologies. In addition, a vulnerable subset of patients, including the immunocompromised and the elderly, is particularly prone to complications. Although immune competent individuals may recover from some infections without treatment, the elderly have diminished immune function and a greater risk of mortality from GI infections. This contractor recognizes that patients may benefit from rapid diagnosis and early intervention (e.g., antibiotic therapy). Due to the substantial overlap in clinical signs and symptoms exhibited by patients with acute GI infections, the diagnostic approach and rapid analysis offered by broad GIP molecular assays, rather than the step-by-step sequential approach of single target assays, may be considered medically necessary.

In the infrequent, if not rare, situations where clinical features of the patient’s presentation indicate a specific microbial etiology and/or therapy, traditional culture methods or single target molecular assay rather than a broad GIP should be used. Because of the unique clinical circumstances of immune compromised patients, ICU patients, and HIV positive patients with diarrhea, GIP testing for bacteria, virus and parasite testing may be indicated, and thus a Medicare benefit.

We recommend modifying the paragraph titled “Summary Medicare Coverage Decision” to read as follows:

(Please note: strikethrough font indicates recommended deletion of original draft LCD language; blue font indicates recommended additions)

GIP testing is limited to no more than 5 bacterial pathogen targets. Testing for viral etiologies is not reasonable and necessary because these GI diseases are generally self- limited, virus specific therapies are not available, and patients are managed by supportive care and hydration. GIP testing is limited to the minimum number of targets needed for therapeutic decision making. The following clinical indications and contraindications characterize the role of GIP testing:

INDICATED:

1. Individuals with moderate-to-severe symptoms associated with acute diarrhea
2. Individuals with dysentery
3. Individuals with acute diarrhea lasting > 7 days
4. Immunocompromised individuals with acute diarrhea

NOT INDICATED:

1. Immunocompetent individuals with mild diarrhea, particularly of ≤ 7 days duration
2. Individuals in whom the clinical presentation of acute diarrhea suggests a specific microbial etiology (e.g. patient with antecedent antibiotic or hospitalization)

Physicians should follow current American College of Gastroenterology (ACG) clinical guidelines, unless the specific clinical situation indicates otherwise. A broad GIP molecular panel (with 6-22 targets) is indicated when a patient presents with a clinical scenario and overlapping symptoms consistent with multiple possible microbiological etiologies. When the patient history and clinical presentation suggest a specific microbial etiology and/or therapy, a broad GIP with more than 5 infectious targets is not indicated.

Although viral infections may be self-limiting, this contractor considers GIP molecular assays including GI viruses (such as norovirus) to be medically necessary to guide initial patient management through best supportive care including hydration and to avoid the use of antibiotics associated with secondary Clostridium difficile infection.

For patients in long-term care facilities, GIP molecular panels containing viruses (norovirus, rotavirus, and enteric adenoviruses) are reasonable and necessary because results inform individual patient management decisions such as isolation of the patient (which may increase risk of unrecognized delirium) , and decisions regarding appropriate use of antibiotic therapy.

Travelers with > 2 weeks of symptoms, after bacterial pathogens have been ruled out, may require traditional ova and parasite stool examination and/or specific protozoal antigen or molecular testing. Large panels including viruses and protozoa are not reasonable and necessary for usual community-acquired diarrheal illness.

There is no Medicare benefit to the use of GIP testing for national, state, or local agency tracking of diarrheal outbreaks, for epidemiologic purposes, or to confirm another etiologic test result. Once the etiology of an outbreak is identified, subsequent patient testing is generally not indicated and patients are managed empirically. However, if the clinical presentation of an individual patient varies from the outbreak prototype, a specific test for the etiologic organism may be indicated. The Medicare benefit is specifically for the clinical identification and management of disease for a given beneficiary. The Medicare benefit does not extend for purposes of the family or for community tracking or surveillance.

Rationale for the Above Language

We are supportive of Palmetto’s proposal to cover Gastrointestinal Pathogen (GIP) molecular assays identified by multiplex nucleic acid amplification tests. While we agree codes for up to five bacterial targets are applicable, we also believe that additional targets should be included, and that there are clinical scenarios where more than five targets are medically necessary. The draft LCD lists five bacterial targets that will be eligible for GIP coverage: Salmonella; Campylobacter; Shigella; Cryptosporidium; and Shiga toxin-producing Escherichia coli (STEC) O157 and non-O157. Several clinical studies have demonstrated that these targets are insufficient to identify the top 90-95 percent of pathogens that may cause diarrhea in patients (Buss, et al., 2015).

Testing for more than five infectious targets is necessary in order to achieve a diagnosis rate of 90-95 percent of pathogens. Restriction of testing to five prioritized bacterial targets would likely miss the majority of pathogens responsible for gastrointestinal infections in the elderly. In a prospective study published by the Infectious Diseases Society of America (IDSA), Salmonella, Campylobacter, Shigella, and STEC (O157 and non-O157) accounted for less than half of the gastrointestinal pathogens detected in adults up to 98 years of age with diarrhea. The other pathogens detected included enterotoxigenic E. coli, enteropathogenic E. coli, enteroaggregative E. coli, Clostridium difficile, Yersinia enterocolitica, Vibrio group (consisting of Vibrio parahaemolyticus or Vibrio vulnificus or Vibrio cholera, of which Vibrio parahemoltyicus is most common), Aeromonas sp., Entamoeba histolytica, Giardia, Cryptosporidium, adenovirus, astrovirus, calicivirus, and rotavirus (Svenungsson, et al., 2000). Additionally, in a separate study published by the American Society for Microbiology, Salmonella, Campylobacter, Shigella, and STEC (O157 and non-O157) accounted for only 24% of the GI pathogens detected in patients 65 years of age or older. At least three other pathogens were frequently detected in these patients (Buss, etal., 2015)

Request: We recommend adding coverage for the Group 2 CPT code 87506 (6-11 targets) as described in the ICD10 coding section.

American College of Gastroenterology (ACG) clinical guidelines recommends the use of broad GIP panels. Due to the substantial overlap in clinical signs and symptoms exhibited by patients with acute GI infections, a broad diagnostic approach may be required. ACG clinical guidelines recognize that “as symptoms of acute diarrhea are protean, attempts to diagnose etiologic agents or classes by symptoms are subjective at best and fraught with imprecision due to overlap in symptoms. Although features of the clinical presentation may be useful in distinguishing bacteria from protozoan causes they are often unreliable indicators of the likely pathogen responsible.” Consequently, infectious diarrhea is the second most highly-ranked syndrome in need of improved diagnostics (Blaschke et al., 2015). This is the precise rationale that makes the use of broad multiplex GI panels in clinical practice a necessity (Riddle et al., 2016). We recommend adding coverage for the Group 2 CPT code 87506 (6-11 targets) as described in the ICD10 coding section

ACG clinical guidelines also recognize the specific advantages provided by more comprehensive NAATs analysis for therapeutic decision-making over other diagnostic approaches. ACG guidelines state that "it is now possible using culture-independent molecular techniques to identify a multitude of bacterial, protozoan and viral diarrheal pathogens including some not commonly identified in clinical laboratories.” Diarrheal disease by definition has a broad range of potential pathogens particularly well suited for multiplex molecular testing. “Molecular diagnostic tests can provide a more comprehensive assessment of disease etiology by increasing the diagnostic yield compared with conventional diagnostic tests." (Riddle et al., 2016). This advantage is important for therapeutic decision-making since elderly persons are at an increased risk for severe illness and complications from infectious diarrhea, and benefit from therapeutic intervention, similar to patients with defined congenital or acquired immunocompromising conditions (Guerrant et al., 2001, Chen et al., 2015, Jagai et al., 2014, Hall et al., 2012). Another recent study confirms the benefits of these interventions.

A recent study comparing the FilmArray GI Panel to culture-based methods demonstrated that the panel promoted targeted rather than empirical therapy and shortened the time to initiation of appropriate therapy. The study also found those patients detected by multiplex PCR are just as symptomatic and just as likely to receive antimicrobial therapy as those detected by more traditional methods. It is clear from these results that the additional positives detected by the panel are clinically important. (Cybulski et al., in press). Although GIP’s may detect multiple targets in asymptomatic and symptomatic individuals, providers and laboratory professionals routinely consider all aspects of the patient’s condition (e.g., symptom duration and severity and prior antimicrobial treatment) when interpreting the results of multiplex GI panels and act accordingly (Ramanan et al, 2018). In the hospital setting, medical professionals want to know the cause of the gastroenteritis in order to determine the appropriate level of isolation precautions, even when a targeted therapy is unavailable (e.g., for Norovirus) (Beal et al, 2018).

Testing more than five bacterial targets is necessary since alternative diagnoses other than the prioritized “5 targets” (i.e. Salmonella, Campylobacter, Shigella, Cryptosporidium, Shiga toxin producing E.coli), have been shown to benefit from specific therapy (see specific examples below). We recommend adding coverage for the Group 2 CPT code 87506 (6-11 targets) as described in the ICD10 coding section

1. Cryptosporidium parvum: Prolonged symptoms can occur even in immune-competent hosts. A randomized clinical trial study of nitazoxanide demonstrated the efficacy of anti-parasitic treatment in shortening the duration of symptoms and oocyst shedding (Rossignol et al., 2001, Dupont 2016).
2. Clostridium difficile: Delays in diagnosis are common and are associated with poor outcomes to treatment. Oral vancomycin can shorten the duration of symptoms (Guerrero et al., 2011).
3. Enteropathogenic/Enteroaggregative E.coli Infections in older adults respond to specific antimicrobial treatment (Nataro et al., 2006; Thorén et al., 1980; Wanke et al., 1998; Glandt et al., 1999).
4. Enterotoxigenic E.coli: Fluoroquinolone therapy can reduce the duration of symptoms in patients with enterotoxigenic E.coli from a mean of 3 days to 1 day (Mattila et al., 1993).
5. Yersinia enterocolitica: Symptoms including chronic diarrhea in patients up to 94 years of age can be readily treated with fluoroquinolones. (Saebø et al. 1992., Gayraud et al., 1993)
6. Vibrio: Fluoroquinolone therapy confers clinical and microbiological response in patients with gastrointestinal Vibrio infections. (Butler et al., 1993)
7. Giardia – Albendazole and metronidazole improve symptoms and shorten duration of parasite shedding. (Grandados et al., 2012)
8. Amoebic dysentery: Effective options for treatment include metronidazole, tinidazole and secnidazole. (Marie and Petri. 2013)
9. Cyclospora: Treatment reduces the duration of symptoms and parasite shedding in prolonged diarrhea in both immunocompetent and immunocompromised hosts (Hoge et al., 1995).

A comprehensive review of persistent diarrhea by the Journal of the American Medical Association (JAMA) emphasizes the challenges of a broad differential diagnosis, which includes enteropathogenic E. coli, enteroaggregative E. coli, Clostridium difficile, Aeromonas, Campylobacter, Salmonella, Shigella, norovirus, Entamoeba histolytica, Giardia, Cryptosporidium and Cyclospora (Dupont 2016). This review states that "Culture-independent sequencing diagnostic methods are now available and include a multiplex approach that allows a number of bacterial, viral and parasitic enteropathogens to be detected in a single test simultaneously. These methods are faster and have greater sensitivity than culture-based methods facilitating identification of the many pathogens that must be considered when trying to find the cause of persistent diarrhea.” There are an estimated 226,000 foodborne illnesses among U.S. adults ≥ 65 years of age, resulting in approximately 9,700 hospitalizations and 500 deaths, underscoring the burden of foodborne illness in older adults and the need for rapid diagnosis and treatment in this population.

There is evidence that using GIP assays with more than five targets would benefit patients due to the resulting incidental findings. Compelling data suggest that an incidental finding of C. difficile is beneficial, however C. difficile is one of the bacterial targets omitted from the draft LCD. Infections due to C. difficile are a significant issue for patients in long-term care facilities. A study of a long-term care facility operated by the Department of Veterans Affairs found that delays in diagnosis for C. difficile are common and associated with poorer outcomes of treatment (Guerrero et al., 2011). IDSA found that C .difficile is the main contributor to gastroenteritis-associated deaths, largely accounting for the increasing trend, with norovirus being the second most common pathogen, often as a co-infection with C. difficile (Hall et al., 2012). A lack of clinical suspicion can result in the under-diagnosis of C. difficile infection in the community setting (Reigadas et al., 2015). Growing evidence indicates that a substantial burden of C. difficile infection is community-acquired, and some of these patients lack traditional risk factors, such as antibiotic exposure or recent hospitalization (Khanna et al., 2012). In the population-based study by Khanna et al., 41% of definite C. difficile infection cases were found to be community- acquired, and nearly one-quarter of these cases had no history of recent antibiotic use. We recommend adding coverage for GIP assays with more than five targets (Group 2 CPT code 87506) to include C. difficile testing to reflect current practice and improve care for patients.

Further evidence that patients benefit from incidental detection of unsuspected pathogens is provided by a study that found that routine utilization of a multiplex molecular panel (BioFire Diagnostics, Salt Lake City, UT) for diagnosis, would have led to more rapid detection of a cyclosporiasis outbreak. In the clinical microbiology laboratory, Cyclospora testing is typically conducted using a specifically ordered, modified-acid-fast-stained fecal smear. Prior to the recognition of and during the early stages of the cyclosporiasis outbreak, de-identified specimens that tested positive for Cyclospora using the FilmArray GI panel in the research study were undetected by the clinical laboratory because modified acid-fast staining had not been not ordered by clinicians. As the medical community became aware of the outbreak, direct analysis of stools for Cyclospora became a common practice, and stool specimens enrolled in the clinical study later in the outbreak timeline were more likely to have had Cyclospora testing ordered. Routine utilization of NAAT tests consisting of more than five targets as a diagnostic tool would have led to more rapid detection of the cyclosporiasis outbreak, since Cyclospora would have been detected before anyone was specifically looking for this parasite (Buss et al., 2013).

The testing of more than five bacterial targets is a significant benefit for immunocompromised patients, including those who are elderly. Immunocompromised individuals are more susceptible to infection with enteric pathogens and are more likely to develop more severe illness and complications. The spectrum of etiologies in normal and immunocompromised hosts is similar (Guerrant et al., 2001), and gastrointestinal pathogens can cause prolonged diarrhea in both immunocompetent and immunocompromised hosts (Hoge et al., 1995, Rossignol et al.2001). One study found that the effects of aging on the immune system manifest at multiple levels that include reduced production of B and T cells in bone marrow and thymus and diminished function of mature lymphocytes in secondary lymphoid tissues. As a result, elderly individuals do not respond to immune challenges as well as the young, qualifying them as immunocompromised (Montecino-Rodriguez et al., 2013). Not expanding the infectious targets beyond those prioritized in this draft LCD may cause infections to be overlooked in susceptible patients, delaying effective treatment. Identifying and treating these patients in a timely manner would greatly benefit Medicare beneficiaries and by extension would also benefit the Medicare program as a whole. Use of this testing strategy leads to actionable diagnoses for Medicare beneficiaries.

The testing of more than five bacterial targets is a significant benefit for cancer patients with diarrhea. A recent study confirms the benefits of these interventions in cancer patients. In as study of 311 cancer patients with diarrhea who were tested with the FilmArray GI Panel multiplex PCR assay, there was a high frequency of diarrheagenic E. coli, in particular EPEC and EAEC. The patients’ clinical symptoms were similar at the time of presentation and 92% received antibiotic treatment and 87% of those treated symptoms resolved even though some of the patients had been symptomatic for weeks at the time of presentation. (Chao et al., 2017)

The testing of more than five bacterial targets is cost effective. Furthermore, a recent study has shown that use of these panels not only increases the detection of important pathogens, it also has the potential to decrease overall health care costs. When the FilmArray GI Panel’s impact on utilization in 241 emergency department or admitted patients was assessed, there a significant reduction in antibiotic utilization and abdominal or pelvic imaging studies compared to the control patients diagnosed by conventional methods. (Beal at al., 2018).

In another study of hospitalized patients (n=800), the costs associated with laboratory testing and with patient isolation were compared when using conventional enteric pathogen testing versus the Luminex GIP assay (Goldenberg et al, 2014). Goldenberg et al. report that although the GIP assay increased laboratory costs (additional £22,283), this approach reduced the overall costs of patient care by decreasing the number of days patients were placed under isolation protocols (£66,765 saved). If physicians order multiple individual stool pathogen assays, which is likely in hospital settings, the total cost could be comparable to or more expensive than that of running a single multiplex assay, making multiplex testing more cost-effective (Ramanan et al, 2018).

Request: We recommend adding coverage for the Group 2 CPT code 87506 (6-11 targets) as described in the ICD10 coding section

Request: In patients who are immune compromised, e.g. cancer patients receiving chemotherapy, transplant patients (e.g. kidney, liver or bone marrow transplant) , ICU patients, and HIV positive patients with diarrhea, we recommend coverage of the Group2 CPT code 87507 (12-25 targets) as described in the ICD10 coding section. These patients are susceptible to mixed infections including the viruses targeted in FDA-approved GIP assays.

It is our recommendation that this draft LCD should be updated to recognize that > 5 targets must be tested to reach the LCD’s stated goal of covering 90-95% of foodborne infections. The LCD should reflect ACG guidelines, which endorse the use of broad GIP panels of > 5 targets and recognize that many of the pathogens detected by multiplex assays have specific treatments that can benefit patients. Clinical scenarios require GIP molecu lar assays that consist of > 5 targets and are medically necessary for management of Medicare patients with compromised immune systems (which includes the elderly). Patients tested using GIP assays with more than five targets also benefit from incidental findings such as C. difficile. Coverage for actionable diagnoses beyond the 5 targets listed in this draft LCD will greatly benefit Medicare patients and improve their healthcare outcomes.

The experts from our organizations who reviewed this policy believe Palmetto’s view of actionable pathogens for gastrointestinal diseases is too narrow. There are other pathogens common in the Medicare population that affect the correct diagnosis and therapeutic decision- making of providers. Acute gastroenteritis is a significant cause of morbidity and mortality in the elderly.

Diagnosis of viruses that cause gastroenteritis is medically necessary since they are associated with higher morbidity/mortality rates in the elderly and medical intervention is available. Noroviruses are the leading cause of acute gastroenteritis in the United States and are a frequent cause of outbreaks in long-term care facilities (Wikswo et al., 2015, White et al., 2016). Norovirus is responsible for 10-20% of gastroenteritis hospitalizations and 10-15% of all-cause gastroenteritis deaths in older adults. Individuals 65 years of age and older are at increased risk of severe outcomes, longer length-of-stay and higher costs with attributable mortality rates 200% higher than other patient groups (Lindsay et al., 2015, Belliot et al., 2014). Although norovirus is the most common cause of gastroenteritis outbreaks in long-term care institutions, a substantial percentage of the outbreaks can also be attributed to rotavirus and enteric adenoviruses (Gaspard et al., 2015, Gerber et al. 2011). Older adults that live in retirement communities and long-term care facilities are disproportionately affected by complications of norovirus infection. Norovirus is commonly included in GIP panels but is not listed among those covered in the draft LCD’s narrow listing of covered targets. Pathogens like norovirus have specific treatment and infection control implications that can benefit patients and public health. Since noroviruses should be included in the list of infections agents it is medically necessary to identify, we recommend coverage of the Group 2 CPT code 87506 (6-11 targets) as described in the ICD10 coding section.

A correct diagnosis of norovirus affects therapeutic decision-making (i.e. is the intervention or prescribing antibiotics necessary?) For example, NAATs allowed for more rapid recognition of norovirus and C. difficile infections in elderly patients, in whom diarrhea is often initially incorrectly attributed to laxative use (Salmona et al., 2016). In this situation, a diagnosis is required before the patients are managed by supportive care and hydration. A diagnosis of a viral infection may also prevent the unnecessary use of antibiotics. This approach may, in turn, reduce the rate of C. difficile infections associated with antibiotic therapy, and is anticipated to result in cost savings for the Medicare program. A study published by ASM has shown that antibiotic exposure and C. difficile infection are closely related (Wenisch et al., 2014). The policy fails to consider the potential benefits to beneficiaries provided by NAATs that exclude or include a diagnosis with specific infection control implications (e.g., norovirus, C. difficile). The CDC recommends isolation of patients with norovirus, EPA-approved disinfection procedures, and use of personal protective equipment including masks with eye shields for at least 2 weeks after clinical resolution. This is important, as contact precautions may increase the incidence of delirium and other adverse events for the Medicare beneficiary (Day et al. 2012; Croft et al.2015). The use of sensitive diagnostic assays can allow accurate case identification, prevent secondary transmission to other Medicare beneficiaries and benefit patients by avoiding unnecessary contact isolation in those who are not infected (Mattner et al., 2015). Based on this evidence, we recommend that Palmetto provide coverage for testing of norovirus. This change would greatly benefit Medicare beneficiaries and by extension benefit the Medicare program as a whole.

As published evidence has shown, the elderly are at increased risk from gastroenteritis but Medicare beneficiaries also include a smaller patient population that is less than 65 years of age (e.g., disabled individuals, patients with amyotrophic lateral sclerosis [ALS] or end stage renal disease). It is essential that any policies regarding coverage for GIP adequately serve the entirety of the Medicare population and ensure that all are properly accounted for under any policy decisions. It is our recommendation that the draft LCD be revised to recognize that viral pathogens may cause gastrointestinal infections in patients of all ages, and that the Medicare population is at increased risk for complications, hospitalization, and death from these infections.

Coding

We support the recommendations of the ACG Clinical Guideline: Diagnosis, Treatment, and Prevention of Acute Diarrheal Infections in Adults (Riddle et al, 2016). Recommendation 3 states “Traditional methods of diagnosis (bacterial culture, microscopy with and without special stains and immune-fluorescence, and antigen testing) fail to reveal the etiology of the majority of cases of acute diarrheal infection. If available the use of Food and Drug Administration- approved culture independent methods of diagnosis can be recommended as an adjunct to traditional methods.” We note that different FDA-approved laboratory tests detect as few as 4 and as many as 22 different pathogens.

The experts from our organizations are able to describe very few situations where the clinical indicators are so strong that an accurate bedside diagnosis can be made, thus dictating testing for only a specific pathogen. One such rare example might be when hemolytic-uremic syndrome (HUS) has developed indicating Shiga Toxin-Producing E. coli (STEC) infection. Other possible scenarios whereby multi-target testing may not be necessary include outbreak investigations involving a known pathogen (ie Norwalk virus), or antibiotic exposure suggestive of clostridium difficile infection (CDI)—but in the outpatient setting even this is not a highly predictive parameter. In addition, there are only a few clinical scenarios whereby a smaller 3-5 target assay (CPT 87505) would be appropriate, as compared to a larger panel. One such rare scenario would be in a laboratory that has replaced stool bacterial culture with only molecular methods and wants to be able to detect these culturable bacteria with a 3-5 target limited panel (i.e., BD Max, Prodesse). That laboratory would have to offer other testing by non-molecular methods(parasites for example), and perhaps also offer a separate molecular test for norovirus.

ICD10 codes that support medical necessity for Group 1 CPT code 87505.

The proposed policy lists ICD-10 codes, A01.00-A05.3 as the only codes that are used to support medical necessity. Based on the available evidence we suggest that the A01.00-A05.3 ICD-10 codes plus greater than 80 additional ICD-10 codes listed in the table below support medical necessity for Group 1 CPT code 87505. The ICD-10 coding system contains detailed disease classification pertaining to gastrointestinal infections, many codes of which are relevant to the draft policy that is proposed. We request that additional ICD-10 codes be added to this policy including, but not limited to, the following list:

ICD10 codes that support medical necessity for Group 2 CPT code 87506. Based on the evidence presented in the body of the letter we propose that ICD10 codes that are appropriate for Group 1 CPT code 87505 are also appropriate for Group 2 CPT code 87506. We request that additional ICD-10 codes be added to this policy including, but not limited to, the following list:

ICD10 codes applicable to Group 2 CPT code 87506 for LCD DL37709

ICD10 codes that support medical necessity for Group 2 CPT code 87507. In the table below we also advocate for additional ICD-10 codes to support medical necessity for Group 2 CPT code 87507 in immune compromised patients, such as transplant patients and cancer patients receiving chemotherapy. Not all laboratories have adopted the large syndromic panels, and there are still small panels available and FDA cleared which would conform to ACG guidelines. Therefore we also request coverage for Group1 CPT code 87505 and Group 2 CPT code 87506 if an ICD-10 code associated with immune compromised patients is used in conjunction with those CPT codes.

We respectfully ask that you consider these comments, which were prepared by expert panels including members of AGA, ASM, AMP, APHL, CAP, IDSA, and PASCV who provide services to Medicare beneficiaries covered by Palmetto. Without hesitation, we are willing to be of assistance in providing additional clinical information, references, contacts, or whatever is needed to assist you with this draft LCD.

We appreciate the commenters’ detailed remarks. Based on these remarks and re-evaluation of the evidence and guidelines, we are modifying the LCD to allow testing for the presence of 6-11 pathogens, to allow for testing of additional organisms, particularly Clostridium difficile. It is not clear that testing for more than 11 pathogens is necessary following an appropriate history and exam, nor is it necessarily able to provide clinically actionable information regardless of the immune status of the patient being tested. As such, testing for more than 11 pathogens will remain a non-covered service.

From my perspective, the multiplex GI panels have been a revolutionary advance for my patients with diarrhea. The more sensitive assays are allowing laboratories to make a specific microbiological diagnosis more rapidly and five times as often as before, which is allowing targeted treatment that is more likely to work as well as the withholding of empirical treatment when the cause turns out to be a virus that won’t respond to antibiotics. Specific and effective treatments are available for many of the targets included on the new multiplex panels, including not only Salmonella, Shigella, Campylobacter and C. difficile, but also less well-appreciated pathogens including ETEC, EPEC, EAEC, Yersinia enterocolitica, Vibrio spp., Giardia duodenalis, Entamoeba histolytica, Cryptosporidium parvum and Cyclospora cayetanensis.

Since the clinical signs and symptoms of the various causes of acute gastroenteritis are largely overlapping, I believe it makes much more sense to order a “syndromic panel” rather than to expect clinicians to list all of the possible infectious causes of diarrhea, as they are sure to leave some important ones out. I am not finding that the use of a culture-independent diagnostic test is interfering with public health efforts, as the laboratory is performing targeted cultures whenever a molecular test detects a culturable pathogen, so the public health lab is still getting their isolates. To the contrary, the laboratory is now detecting more foodborne pathogens that would have been missed by less sensitive culture-based approaches, so public health efforts are enhanced. It has also been a pleasant surprise to find that the increased costs of the assay reagents are largely offset by a reduced laboratory workload and more rapid turnaround time, so the net financial impact on the laboratory has not been substantial.

When these tests first became available, an important question in my mind was whether the additional diagnoses made possible by multiplex panels were clinically relevant. I therefore decided to conduct a prospective multi-center study to compare the clinical features of patients who were detected by conventional methods with those who were only detectable by the more sensitive PCR-based tests. After analyzing 1,887 consecutive specimens and their clinical correlations, the results were clear: patients detected by culture or by PCR were comparable in nearly all respects and just as likely to be assessed by their physicians to warrant antimicrobial therapy. The difference is that the more rapid test allowed clinicians to provide targeted rather than empirical treatment, and more rapid treatment can lead to a shorter duration of symptoms as well as a shorter duration of organism shedding which should limit transmission in the community. We also observed that clinicians were willing to withhold antibiotic treatment when only viruses were detected, and some of these patients would likely have received empirical antibiotics if the PCR assays had not been available. Our paper has been accepted for publication by the journal Clinical Infectious Diseases, and I am attaching an advance copy in case it is useful to you.

As promised, here is a sub-analysis of patients ≥65 years of age from our study, which I have prepared with the help of Dr. Rob Cybulski. At the end it includes a recommendation for testing indications in these patients. I hope that the additional information is useful to you. Please let me know if anything requires clarification.

Clinical Impact of the FilmArray GI Panel in Patients ≥65 Years of Age

Patients ≥65 years of age made up 20.2% (382 of 1887) of the total patient population tested during the period of our study (January-September 2017). A total of 52 older patients (13.6%) were diagnosed with an infectious cause for their gastroenteritis, including 22 cases in which patients were found to be coinfected with two or more pathogens. The spectrum of pathogens detected by the FilmArray? was comparable for patients ≥65 years of age and the overall population, with only Clostridioides difficile detection rates varying significantly (p < 0.0001) between these groups (Figure). These observations are consistent with the fact that our study population was comprised predominantly of outpatients (82%), and patients with community- acquired C. difficile infections tend to be younger than those with hospital-acquired infections (Khanna et al., 2012).

Unable to display image in this format.

The distribution of pathogens detected within patients ≥65 years of age and the total population showed some variation (Table 1) that was consistent with previous studies. Campylobacter, Salmonella, Shigella/EIEC and STEC (including O157:H7) accounted for only 18.7% of the total pathogens detected. Viral pathogens (particularly Sapovirus and Norovirus) made up a larger proportion of total detections among ≥65 y/o patients (24%) compared to the population as a whole (15%). The diarrheagenic E. coli species Enteropathogenic E. coli (EPEC) and Enteraggregative E. coli (EAEC) were also more prevalent in the older population (37% vs. 30%). This is noteworthy, as pathogens such as norovirus, sapovirus, EPEC and EAEC are widely considered to be pediatric pathogens and may be underrecognized and underappreciated as pathogens in older adults (Nataro et al., 2006; Gaspard et al., 2015; White et al., 2016). C. difficile and parasitic pathogens were less frequently encountered in patients ≥65 years of age.

The use of the FilmArray nearly doubled the sensitivity of laboratory detection of bacterial pathogens common to both the FilmArray and traditional stool culture in patients ≥65 years of age (Table 2). Two-thirds of the patients ≥65 years of age in whom treatable targets were identified received antimicrobial therapy, which was comparable to the treatment rate in the overall study population. The median time from specimen collection to the initiation of antimicrobial therapy was 21 hours. The FilmArray detected 35 additional cases of bacterial infection caused by diarrheagenic E. coli or C. difficile, neither of which is recoverable by conventional culture. One parasitic infection (Giardia lamblia) was detected by FilmArrayÔ when the ordering physician did not order an Ova & Parasite Exam or Giardia Stool Antigen and would presumably have been missed if a panel had not been ordered. The patient received targeted antiparasitic therapy. Eighteen cases of viral gastroenteritis were diagnosed with the FilmArray, and the University of Washington laboratory-developed Viral Gastrointestinal Pathogen (PCR) Panel detected the same pathogen in 1 of the 2 cases in which it was ordered. In the other 16 cases, viral-specific testing was not ordered, and the diagnoses would presumably have been missed. Although viral gastroenteritis is not treated by antimicrobial therapy, the detection of a viral pathogen allows the avoidance of empirical antibiotics, which carry a risk of C. difficile, and also triggers infection control precautions in hospitals and long term care facilities to prevent other residents from becoming infected (Mattner et al., 2015). In terms of clinical acuity, patients ≥65 years of age with detected pathogens were very similar to the total patient population with regard to clinical features (Table 3). When patients’ medical records were scored for the presence or absence of symptoms characteristic of acute gastroenteritis, patients ≥65 years of age had a mean number of features similar to that of the total population (3.1 versus 3.3), suggesting comparable disease severity. Older patients had an increased likelihood of hospitalization compared to the population as a whole, although the difference did not achieve statistical significance. This supports previous observations which indicate that older persons are at increased risk for severe illness and complications from acute gastroenteritis as a consequence of reduced gastric acidity, impaired intestinal motility and immunocompromise (Belliot et al., 2014; Chen et al., 2015; Lindsay et al., 2015).

In summary, as in the overall population, we found that the FilmArray GI Panel improved diagnostic sensitivity in older patients (≥65 years of age) with acute diarrhea, identified patients with clinical features comparable to those of younger patients, and enabled clinicians to make more timely and targeted therapeutic decisions.

Per recent guidelines from the American College of Gastroenterology (Riddle et al., 2016), specific indications for multiplex diagnostic testing in older patients with acute diarrhea are the presence of dysentery, moderate-to-severe disease or symptoms lasting longer than 7 days. In addition, the guidelines note that testing may be indicated in vulnerable populations at increased risk for severe disease, which includes many elderly patients as well as those with characterized immune deficiencies or recent foreign travel.

Table 1. Comparison of Positive Detections in ≥65 y/o and Total Patient Populations

Unable to display image in this format.

Notes. P-values refer to comparison of percentage positive among total specimens tested by stool culture and FilmArray.

Abbreviations. POS, total positive; %TOT, percentage positive of total tested; %POS, percentage positive of total positives; n/a; not applicable; n.s., not significant

1. FilmArray cannot distinguish between Shigella and Enteroinvasive E. coli (EIEC)
2. FilmArray cannot distinguish between pathogenic and nonpathogenic strains of Yersinia enterocolitica
3. Detected by culture/immunoassay or O157:H7-specific FilmArray GI target
4. Total STECs, including O157:H7
5. Non-O157:H7 STEC isolates are not recoverable using standard culture techniques

Table 2. Comparison of FilmArrayÔ versus Stool Culture in the ≥65 y/o Patient Population

Unable to display image in this format.

Notes. P values refer to comparison of percentage positive among total specimens tested by stool culture and FilmArray.

Abbreviations. POS, total positive; %TOT, percentage positive of total tested; %POS, percentage positive of total positives;

n/a; not applicable; n.s., not significant

1. FilmArray cannot distinguish between Shigella and Enteroinvasive E. coli (EIEC)
2. FilmArray cannot distinguish between pathogenic and nonpathogenic strains of Yersinia enterocolitica
3. Detected by culture/immunoassay or O157:H7-specific FilmArray GI target
4. Total STECs, including O157:H7
5. Non-O157:H7 STEC isolates are not recoverable using standard culture techniques

Table 3. Clinical Characteristics of ≥65 y/o Patients Compared to the Total Patient Population

Unable to display image in this format.

We appreciate the detailed and thoughtful response. We have reviewed the copy of the manuscript. It is important to note that this manuscript does not look at mid-term or long term outcomes and focuses heavily on short term process of care outcomes, such as antibiotic prescription.

Because of this we believe that the scope of the scope of the study does not include the necessary information to change our coverage decision. It is worth noting that the use of empiric antibiotics for infectious gastrointestinal illness is generally not recommended by the Infectious Disease Society of America in their 2017 guideline except in very specific circumstances. Even when an organism is identified, the guidelines do not necessarily recommend targeted antibiotic treatment for all organisms. While these guidelines may not be universally applicable to all patients, we believe that these guidelines reflect broadly applicable evidence-based used of antibiotics in the treatment of infectious diarrhea. As such, for the purposes of applying the results of this study to inform our coverage decision (though not necessarily by itself change the decision), we would at minimum require data detailing the indications for the use of empiric and directed antibiotic treatment in this study.

Shane, AL, Mody, RK, Crump, JA et al, 2017. 2017 Infectious Diseases Society of America clinical practice guidelines for the diagnosis and management of infectious diarrhea. Clinical Infectious Diseases, 65(12), pp.e45-e80.

While it is clear that high-order multiplex testing is not indicated for all patients presenting with diarrhea, the availability of multiplex panels for enteric infection have improved my ability to care for patients with diarrhea in a number of ways.

The American College of Gastroenterology recommends a diagnostic work up for patients presenting with diarrhea who have either a) dysentery; b) moderate to severe disease or c) symptoms lasting > 7 days. These recommendations are clear and clinically appropriate and patients falling into these categories can be infected with a range of pathogens, including viruses, bacteria and parasites.

While ideally a physician would be able to evaluate a patient's history, symptoms, and duration of illness and prioritize testing for enteric pathogens that would lead to a diagnosis in the least number of steps, it has been shown in several studies that this is not clinical reality. Symptoms of gastrointestinal illness are not pathogen-specific; even statistical learning tools that have been used to evaluate criteria that could be used to predict viral vs. bacterial vs. protozoal causes of infection have fallen short, suggesting that it is not possible to know with adequate certainty what disease is present. And while manifestations of disease can be similar, the treatments are not. Even amongst bacterial causes of diarrhea, antibiotics are appropriate for some but dangerous for others. And the use of antibacterials for viral or protozoal disease is always inappropriate.

In addition to the therapeutic implications of diagnosing a specific bacterial or protozoal cause of diarrhea, diagnosis of a viral cause of diarrhea also impacts care in a number of ways. These include decreased use of antibiotics that may have unwanted side effects, including dysbiosis and C. difficile infection, prolonging symptoms and leading to further interventions. Prolonged illness itself, possibly due to secondary dysbiosis, but without a viral diagnosis can lead to further testing and unnecessary treatment. Diagnosis of viral gastroenteritis allows for supportive care measures to be initiated as well as appropriate measures for infection control at home, in the community or in the hospital setting. Outside of multiplex panels, routine testing for viral gastroenteritis is not routinely available.

For patients with severe or prolonged symptoms, conventional testing for enteric pathogens can be drawn out and costly. Prior to the introduction of multiplex panels, standard testing involved first a culture, with a turn-around time of several days; protozoal testing was sometimes included from the start, but often followed a negative bacterial result, leading to return visits and continued disease. In addition, many conventional tests are less sensitive than PCR and thus pathogens may be missed. The "diagnostic odyssey" for prolonged gastroenteritis was frustrating for patients and clinicians and often resulted in over or under treatment of disease.

Patients with immune compromising conditions are at risk for extended illness from a range of gastrointestinal pathogens, with implications for care of both the enteric illness and further treatment for their underlying disease. They are at high risk for both bacterial illness, including C. difficile, but also severe and unremitting viral infection.

Multiplex testing for gastrointestinal pathogens does not solve all our problems, and stewardship of such testing by clinicians and laboratorians is clearly needed. Repeat testing and follow up test of cure using a multiplex panel is generally unnecessary. Interpretation of results can be complex, but we are learning more everyday. Despite some issues, however, the availability of multiplex panels including a range of pathogens has improved the care we give our patients, both in ways that are measurable by cost, but also in patient and provider satisfaction. I hope that this type of testing can remain a covered benefit for the patients that I care for.

Thank you for the comment. The ACG guideline on the management of diarrhea does not provide clear guidance on the pathogens for which a patient should be evaluated. For the purposes of selecting targeted antibiotic therapy, the draft LCD will allow testing for those organisms for which the ACG guideline provides specific targeted antibiotic treatment recommendations. As regards viral gastrointestinal illness, we are unaware of evidence demonstrating that broad panel testing can be used to lead to better outcomes.

I am sure you have received valuable information and data from sources, much more qualified than I am, addressing these Drafts. The purpose of this email is to emphasize our desire and willingness to continue to collaborate and work towards the most appropriate reimbursement model for Multiplex Syndromic testing. One that truly benefits the Medicare beneficiaries while having a positive impact on the quality and cost of care. Please advise if we can provide further information and additional data to assist in this endeavor. We have appreciated this opportunity to have these types of discussions as well as your openness to engage us in them.

Thank you for this willingness. We continue to be open to covering tests with established analytical validity, clinical validity, and clinical utility.

Mercy Health Systems includes over 40 hospitals spread across Missouri and the adjoining states. The laboratories of our hospitals work in coordination, under the guidance of our Pathology Specialty Council (comprised of pathologists with broad expertise) and Subject Matter Expert groups (comprised of experienced clinical lab scientists) to determine which lab tests best meet the needs of our patients. We have chosen to offer the BioFire FilmArray GI Pathogen Panel in our laboratories because we believe this assay benefits the health of the individual patient, the health of our communities, and the fiscal health of our system.

Time and again, we have seen the benefits of these rapid assays for individual patients. The coverage guidance implies that the symptoms and signs of the infections assayed in GI pathogen panels are completely distinct. If this were true, GI pathogens could be diagnosed by history and physical exam alone, which is definitely not the case. We have used the GI pathogen panel with great success to diagnose GI illnesses. A middle aged female with three days of diffuse abdominal pain and non-bloody diarrhea in the context of complex medical issues including lupus was found to have cyclospora diagnosed by GI pathogen panel. A parasite was not expected and the diagnosis saved the patient further costly work up and guided appropriate treatment. A young child with a few days of bloody diarrhea was found to have salmonella by GI pathogen panel, diagnosed days ahead of when results would be produced by traditional bacterial culture. Appropriate antibiotic treatment could be started immediately. Knowing that antibiotics are contraindicated in some forms of bloody diarrhea, such as Shiga toxin-producing E. coli, having a rapid diagnosis allowed the physician in this case to proceed with appropriate treatment confidently. We have a multitude of such stories, since going live with the GI Pathogen Panel in 2016. Once a diagnosis is established, even if it is a viral illness with no treatment other than supportive care, we can stop looking for another cause of the patient’s illness and move forward with supporting the patient through their illness.

The GI pathogen panel benefits the health of all our patients and our communities. By rapidly identifying GI pathogens, we are able to provide timely information to our clinicians and public health officials that allows for patients to be appropriately removed from high exposure settings, such as daycare, and contact tracing to be initiated. Early diagnosis allows for other affected individuals to be quickly identified and treated and for further infections to be prevented. Considering all patients who had either the GI pathogen panel or traditional GI pathogen testing, we found that patients who underwent GI pathogen panel testing had a shorter mean length of stay (2.3 days versus 3.8 days for traditionally diagnosed patients, n = 7,008 over 2016 and 2017).

Perhaps our most high profile example of the public health impact of the GI pathogen panel was demonstrated this summer, when we detected a cyclospora outbreak through utilization of the GI pathogen panel. Other area hospitals, testing with traditional bacterial cultures, were not seeing an increased incidence of cyclospora simply because they weren’t testing for it.

Cyclospora, a parasitic infection, is not detected by bacterial stool culture and requires a separate ova and parasite screen when traditional methods are employed. Testing with the GI pathogen panel, we identified over twenty cases of cyclospora that were confirmed by the state laboratory. On July 12, 2018, the St. Louis County Department of Public Health released an announcement regarding the increased incidence of cyclospora in the region. Although the epidemiologic investigation remains ongoing, this notification caused McDonalds to pull their salads from the Missouri market. Recognition of this outbreak would have been greatly delayed or overlooked if not for our use of the GI Pathogen Panel.

While critics of rapid multiplexed GI pathogen panels are quick to argue that the panel is more expensive than traditional culture, this is not true in our experience. When added together, the cost of a bacterial stool culture, a microscopic ova and parasite screen, and viral testing are greater than the cost of a GI pathogen panel. Additionally, we have found that the GI Pathogen Panel results in improved utilization of our healthcare resources. When considering direct variable and fixed costs (including labor, medical supplies, medication, physician costs, and capital) as well as indirect costs (indirect physician cost) for the 7,008 patients who underwent GI diagnostic testing in 2016 and 2017, we found that patients who had a BioFire FilmArray GI Pathogen Panel had an average of $2,743.64 LESS in cost per patient compared to patients tested using traditional methods.

As a health system, we recognize that both the rapid multiplex pathogen panel and traditional approaches are costly and we have set up guidelines to curb unnecessary testing. GI Pathogen Panel testing is only allowed on patients who have been admitted for less than 24 hours. When the GI Pathogen Panel is ordered on an inpatient, the ordering clinician must consult with a pathologist regarding the indication for testing. This policy has been effective in preventing inappropriate testing and the majority of tests ordered on inpatients are cancelled after consultation with a pathologist.

We all must be stewards of limited healthcare resources and we appreciate your commitment to seriously evaluate the utility of rapid GI pathogen diagnostic assays. Failing to reimburse for this important and effective diagnostic tool will hurt our patients and our communities.

Our limited coverage decision for multiplex panels to detect gastrointestinal pathogens refers only to Medicare coverage, not to broad usefulness. Our coverage decision is based upon our assessment of the evidence indicating to us that large multiplex panels are not reasonable and necessary under the Medicare benefit. As reviewed in the draft LCD, the Medicare benefit does not cover epidemiologic testing and testing for infection control. As such, the ability of these panels to be helpful in such scenarios does not make them coverable under the Medicare benefit.

I am writing to you now regarding the most recent LCD drafts for Palmetto and other subscribers to the MolDx program regarding molecular GI testing. As you know, the latest draft is not favorable for reimbursement intended for multiplex GI nucleic acid amplification tests (NAATs) and, as a clinician utilizing tests like the BioFire GI panel, I wanted make a case as to why multiplex testing for infectious gastroenteritis is a leap forward in testing methodologies and has significant impact on patient care and likely patient satisfaction.

The multiplex DNA extraction PCR technology is one of the most important advances in clinical medicine over the last several decades. For syndromic illnesses such as diarrhea, either acute or chronic, the ability to accurately and rapidly make a microbial diagnosis is paramount in terms of clinical management.

As you may know I specialize in gastroenterology, internal medicine, tropical and travel medicine. I have had considerable clinical and research experience in the field of travelers’ diarrhea over the past 25+ years. One of the problems in diagnosis has been the fact that the presentation of diarrhea is often protean and bacterial, parasitic and viral etiologies are often clinically indistinguishable but approached quite differently as far as treatment is concerned. Before the availability of the DNA extraction PCR, diagnosis was dependent upon doing stool cultures which often required special selective media and often took 48 to 72 hours. At our hospital, one of the largest teaching hospitals in the country, prior to this technology, only 5 bacterial pathogens were searched for, leaving the vast majority of pathogens undiagnosed. Likewise for parasitic causes, requiring ova and parasite (O & P) examination by microscopy with or without special stains is fraught with inaccuracies due to the fact that this procedure is technician dependent with propensity to either over-read or under-read. Viral pathogens have been difficult to diagnose because of lack of adequate commercial testing capability.

Efforts to limit the diagnostic possibilities to the 4 or 5 most likely pathogens is not only a bad practice but counter-intuitive. By doing this, this will reinforce the bad practices of the past few decades and we will continue to miss the vast majority of specific microbial pathogens. For travelers’ diarrhea this is particularly important as there are etiologic causes, such as Giardia, Cyclospora, and diarrheagenic E.coli which are often not looked for with traditional testing. Likewise, for community acquired diarrhea, even if the majority of cases are viral, there is value in proving a viral origin and withholding antibiotics in the interest of anti-microbial stewardship and prevention of antimicrobial resistance.

In a recent study of the BioFire GI panel compared to traditional testing methods, the cohort of patients tested on the GI panel were found to have more targeted rather than empiric therapy. In cases where antibiotics were not indicated, like in patients that present with Shiga-toxin producing E. coli, the study showed discontinuation of empiric antimicrobials 47 hours faster than traditional methods (Cybulski, 2018). In another recent study, patients tested on the BioFire GI panel should a decrease in downstream testing methods, like CT scans, x-rays, and ultrasounds compared to traditional methods likely because the physician was able to identify an infectious cause and stop looking for possible causes of diarrhea (Beal, 2018). Without the use of a comprehensive, fast, and accurate test many of the benefits observed in these studies would likely not be realized.

Another point that the draft highlights is the necessity to only detect five common known causes of infectious diarrhea, mainly bacterial causes. The identification of viral and parasitic causes of acute diarrhea are also important and can inform not only decisions to treat or not to treat, but also infection control decisions, recommendations to return to work, school, or other activities that may have an impact to public health. Additionally, studies have shown that treatment for parasitic infections can effectively shorten illness (Rossignol, 2001) and in other cases without treatment be prolonged for several months (Beal, 2018. MacKenzie, 1994) without the appropriate course of therapy. Since potential viral and parasitic causes may not be routinely tested for or have poor sensitivity, yet they make up a substantial portion of diarrheal burden in the US there presence on a multiplex NAAT is appropriate.

In addition, current practice guidelines (ACG 2016 and ISTM 2017) recommend the use of culture-independent molecular multiplex tests for the identification of potential causes of acute diarrhea because of their ability to impact patient management decisions. The guidelines offer a framework for clinicians to consider when considering what patients should be tested, treatment, as well as other patient management recommendations. The guidelines reviewed existing literature and relied on expert opinions to make recommendations that include and recommend the use of culture-independent molecular multiplex tests how they can be utilized to better inform clinicians of what could be making their patients sick.

We also know that diarrheal disease is not the benign self-limited condition we once thought it was. There is a not insignificant incidence of post-infectious sequelae such as chronic GI symptoms following acute diarrhea such as post-infectious irritable bowel syndrome (PI-IBS), post-infections functional gastrointestinal disease (PI-FGD), as well as hemolytic uremic syndrome, reactive arthritis and Guillain Barre syndrome There is a suggestion that prompt diagnosis and treatment may reduce or eliminate these post-infectious sequelae and reducing morbidity and potential mortality.

Multiplex PCR technology also has the potential benefit of reducing use of healthcare resources and reducing overall healthcare costs by providing comprehensive and accurate infectious disease results in a clinically actionable time frame. As a gastroenterologist, the need to perform endoscopic procedures in my practice has declined as a result of utilizing this diagnostic technology. I have had patients with chronic diarrhea scheduled for multiple endoscopic procedures, who with one stool specimen, were found to have pathogens such as Giardia sparing them the expense and inconvenience of these invasive procedures.

In summary, the latest draft of the LCD that impacts molecular multiplex GI panels should be reconsidered and include favorable reimbursement for such tests. Molecular multiplex GI tests have been shown to provide faster, comprehensive, accurate results that can lead to more targeted therapy/patient management decisions that lead to better patient outcomes and likely patient satisfaction. This technology has changed clinical practice and revolutionized the field of diarrheal disease. I would strongly advise against limiting the number of targets and would encourage further dialogue. I am available to speak any time convenient for you.

Thank you for the comment. A notable limitation of testing for gastrointestinal pathogens is the inability to use the information to alter patient care in a way that will necessarily lead to a better outcome. The comment raises the point that nucleic acid amplification testing allows for targeted rather than empiric therapy. This is likely true, but it is not clear that for many patients or infections that targeted therapy is reasonable and necessary. Our coverage policy has been created to allow providers to test for a broad enough range of organisms to test for those organisms that are both clinically likely and the identification of which will lead to clinically actionable information. Draft LCD reflects our understanding of what information is currently clinically useful. We will reconsider this LCD if we become aware of new relevant evidence.

Diatherix Eurofins is pleased to submit its comments to Palmetto GBA regarding the decision to limit coverage for molecular multiplex Gastrointestinal Pathogen assays to 5 bacterial targets associated with foodborne illnesses.

Gastroenteritis

There are many causes of gastroenteritis, yet the symptoms are similar. These common symptoms include vomiting, diarrhea, and abdominal discomfort. However, variances in the cause (whether viral, bacterial, or both) produce a significant difference in the range of both severity and duration of symptoms. Not only is the clinician challenged in associating the patient's symptoms with a particular etiology, non-infectious causes are also prevalent and include improper diet, malabsorption syndromes, various enteropathies, and inflammatory bowel disease.1

When the etiology of bacterial gastroenteritis is examined, it is clear that each agent has specific pathogenic mechanisms that can produce a spectrum of mild to severe disease. Some (like the pathogenic Escherichia coli) employ several mechanisms that may begin with mucosal ulceration and abscess formation. The initial infection may progress via an inflammatory cascade and result in an enteric fever syndrome. With the aforementioned disease spectrum that can be caused by the pathogenic E. coli group, the differentiation of the specific etiology in the group is highly significant. Moreover, a diagnostic test that can accurately predict the specific organism is not only necessary, but essential.2 When invasive infections occur, antimicrobial therapy may limit the adjacent and systemic spread of the bacteria and, also, may reduce the production of toxins that control enteral and extra-enteral cellular processes. Appropriate interventional steps may control significant fluid loss and progression of the infection.1 It is important to note that culture technology cannot reliably and quickly distinguish between the pathogenic E. coli.3 In addition, culture cannot easily distinguish non-culturable causes of infectious diarrhea such as viruses and protozoa that may also be present as co-infections. Delays in intervention and specific treatment not only exacerbate the patient's symptoms, but, also may prolong recovery and promote disease spread. 1

As implied in the proposed Local Coverage Determination (LCD) draft that addresses the use of multiplex NAATs, most of these gastroenteritis syndromes are "generally self-limited". The fact is that they often are not. Gastroenteritis syndromes caused by species of pathogenic

1. coli, Salmonella, Shigella, Campylobacter, Aeromonas, Clostridium, Vibrio, and Yersinia produce a significant spectrum of diseases and their prevalence has doubled in recent years (See Table 1). 4 · 5 It should be of particular interest to the CMS agency that of the more than 17,000 gastroenteritis related deaths tabulated by the CDC, C. difficile accounted for 83% of them in adults older than 65 years. Highly sensitive and specific molecular assays that detect the pathogenic strains of C. difficile are incorporated into the current American Gastroenterology Association's practice guidelines and the clinical impact of these sensitive and specific assays has been significant.4

Notably missing in the LCD draft dialogue is the significance of co-detection when evaluating gastroenteritis. Viruses in combination with other viruses or bacteria can be a significant contributing factor in gastroenteritis; producing exacerbation of symptoms, a protracted clinic course, and increased hospital admissions. Patients with coinfection represent a subgroup of patients with acute gastroenteritis that may require heightened levels of treatment. Dual infections raise the question as to whether a single pathogen is responsible for the illness or whether several pathogens act in synergy-5 7; combinations of organisms frequently seen in gastroenteritis are shown in Table II.

In children, the impact of coinfections on the severity and duration of gastroenteritis is receiving increasing amounts of attention. For example, mixed
rotavirus infections occur frequently and include other viruses, bacteria, and protozoans. A number of clinical studies have shown that children infected with rotavirus and a bacterial pathogen have more severe diarrhea and/or dehydration as well as an extended course of the disease. It is important to note that rotavirus infections (singly or in combination with another pathogen) produce approximately 611,000 deaths per year in children worldwide. Better access to more comprehensive diagnostics such as multiplexed molecular assays would provide better insight into the need for more comprehensive treatment regimens to reduce the morbidity and mortality of these often dual infections.8

In addition, the LCD draft suggests that the multiplex tests currently available as FDA approved platforms or as LDTs is a "one size fits all" panel approach that is not restricted to specific populations. In both FDA approved platforms and LDT approaches to developing multiplex platforms for the diagnosis of gastroenteritis, the pathogens incorporated into the test panels are those that produce disease in a variety of patient populations. Given that patients who seek medical treatment are usually symptomatic, these highly sensitive and specific assays have high positive and negative predictive value as a diagnostic tool and provide the clinician with actionable data on which to base their treatment. Furthermore, rapid diagnostics and aggressive treatment regimens have a positive impact on pathogen focused therapy and epidemiological control mechanisms that limit the spread of highly virulent strains of both bacteria and viruses. It is true that several of these diseases are self-limited, but others are invasive and produce a spectrum of disease that is treatable. Relevant and timely diagnostics often provide life-saving results.

Finally, neither culture independent diagnostic tests nor culture can reliably distinguish active infection from colonization; particularly with infections that involve heavily colonized mucosal surfaces. Since there are no established thresholds for bacterial or viral burdens that are associated with true infection of mucosal surfaces, the diagnosis is based upon disease symptoms. Molecular assays are very sensitive and this sensitivity can detect DNA or RNA fragments for weeks following some infections. The key diagnostic element for the clinician is whether the patient is symptomatic or not. Specific bacterial or viral loads that are indicative of infection have not been established and are not likely given the variations seen in a patient's immunologic status. Relying on a single laboratory test (whether culture or molecular) to determine if a patient is infected is beyond the scope of the laboratory at the present time.

Conclusion

We implore Palmetto GSA to reconsider its decision to provide limited coverage for molecular multiplex Gastrointestinal Pathogen assays for up to 5 bacterial targets associated with foodborne illnesses. The advent of molecular diagnostics has provided a glimpse of the importance of mixed infections and will provide a better understanding of the epidemiology of diarrheal disease and the contribution of mucosal immunology in the development and management of gastroenteritis in children and adults. These new technologies offer a more nuanced and accurate picture of the pathogenesis of infectious diarrhea.

Table I. Pathogens Producing Common Gastroenteritis Syndromes*

*Rapid diagnostic tools provide the clinician with advance warning of a potential developing complication.

Table II. Association of different microorganisms and their frequency⁶*

*Routine detection of coinfections is possible with multiplexed PCR testing

References:

1. http://emedicine.medscape.com/article/176400-overview.
2. Kaper et al. Pathogenic Cscherichia coli. Nature Reviews in Microbiology. Vol 2; Feb, 2004.
3. 3. http://dx.doi.org/10.1016/8978 -0-12-800886-7.00012-1.

1. Surawicz et al. Guidelines for Diagnosis, Treatment, and Prevention of C/ostridium difficile Infections. 2013. Amer. Jour. Gastro. Vol 108: 478-498.
2. https://www.cdc.gov/mmwr/preview/mmwrthml/ss6412a1.htm.
3. Valentini et al. Coinfection in acute gastroenteritis predicts a more severe clinical course in children. Eur. J. Clin. Microbial. Infect. Dis. (2013) 32: 909-915.
4. Feghaly et al. Viral Co-infections are Common and are Associated with Higher Bacterial Burden in Children with C. difficile Infections. 2013 Dec; 57(6): 813-816.
5. Grimprel et al. Rotavirus Disease: Impact of Coinfections. Ped Infect Dis Vol 27 (No. 1). Jan 2008.

Thank you for the comment. A notable limitation of testing for gastrointestinal pathogens is the inability to use the information to alter patient care in a way that will necessarily lead to a better outcome. This comment specifically brings up the issue of co-infection contributing to a more severe disease course and increasing the intensity of treatment needed. However, based on current treatment guidelines, the type and intensity of care given to patients for gastrointestinal illnesses is generally based on symptomatic and physiologic measures of illness severity rather than identification of the underlying causative organism. The LCD is designed to allow limited testing so that organism identification can be done to identify organisms, the identification of which would be expected to change management.

Luminex Corporation appreciates the opportunity to comment on the above-referenced Noridian Healthcare Solutions ("Noridian") draft LCD related to Gastrointestinal Pathogen (GIP) molecular assays identified by multiplex nucleic acid amplification tests (NAATs). Luminex manufacturers two IVD gastrointestinal pathogen panels, xTAG® Gastrointestinal Pathogen Panel (GPP) and VERIGENE® Enteric Pathogens Test (EP). These commercial panels have been cleared by the FDA for the detection of GI pathogens from stool samples. Both Luminex in vitro diagnostic (IVD) tests, xTAG® GPP and VERIGENE® EP, are relevant to the Noridian draft LCD.

Our interest in the draft LCD pertains to how it limits coverage for GIP moleculaT assays. We seek to ensure that the appropriate utility of the test is realized and is consistent with nationally recognized medical specialty society guidelines.

1. Medical Specialty Society Clinical Guidelines

The Noridian proposed LCD is inconsistent with expert panel recommendations that were based on evaluation of the scientific literature that was used to develop specific clinical guidelines. In April 2016, the American College of Gastroenterology (ACG) published the revised clinical guideline, Diagnosis, Treatment, and Prevention of Acute Diarrheal Infections in Adults. In the revised clinical guideline, ACG issued a strong recommendation for the use of FDA-cleared culture-independent methods of diagnosis, at least as an adjunct to traditional methods.1

The ACG clinical guideline supports the indications for testing as "Laboratory testing for an infectious etiology of diarrhea may be indicated in community-acquired diarrhea of>7 days duration or travel-related diarrhea or diarrhea with signs or risk factors for severe disease". Both the xTAG® GPP and VERIGENE® EP panels are factors for severe disease". Both the xTAG® GPP and VERIGENE® EP panels are discussed in the ACG clinical guidelines as FDA-cleared laboratory tests for enteric pathogens. 

The limited coverage in the draft LCD is inconsistent with the ACG clinical guidelines. Luminex submits this request to Noridian to expand GIP coverage to include testing for viral and parasitic pathogens, where clinical circumstances meet the indications for testing in the ACG guidelines.

2. ICD-10 Codes that Support Medical Necessity

Group 1:

The proposed policy considers only bacterial pathogen-related gastroenteritis ICD-10 codes as support of medical necessity for Group 1 code 87505 (3-5 targets). Luminex requests Noridian revise the ICD-10 codes that support medical necessity to include the following viral and parasitic codes under group 1:

A07.1 Giardiasis (lambliasis)
A07.2 Cryptosporidiosis
A07.8 Other specified protozoal intestinal disease
A08.0 Rotaviral enteritis
AOS.11 Acute gastroenteropathy due to Norwalk agent
AOS.2 Adenoviral enteritis
AOS.8 Other specified intestinal infections

Group 2:

The proposed policy considers only immunocompromised and bacterial pathogen-related gastroenteritis ICD-10 codes from group 1 as support of medical necessity for Group 2 codes 87506 and 87507. A study on kidney transplant recipients evaluated the microbiological diagnosis of severe dia11'hea in kidney transplant recipients using multiplex PCR assays since specific treatment (ciclosporin and mycophenolate mofetil combination) increases the risk factor of developing Norovirus infection.2 The authors concluded that molecular tools significantly improved the detection of single and multiple enteric infections by comparison to traditional techniques and may be the key element in the management of severe acute dianhea in transplant recipients.

Luminex requests Noridian revise the ICD-10 codes that support medical necessity to include the following transplant recipient status codes in addition to the bacterial codes from group 1 and the above proposed viral and parasitic codes:

Z94.0 Kidney transplant status
Z94.1 Heart transplant status
Z94.2 Lung transplant status
Z94.3 Heart and lungs transplant status
Z94.4 Liver transplant status
Z94.5 Skin transplant status
Z94.6 Bone transplant status
Z94.7 Corneal transplant status
Z94.81 Bone marrow transplant status
Z94.82 Intestine transplant status
Z94.83 Pancreas transplant status
Z94.84 Stem cells transplant status
Z94.89 Other transplanted organ and tissue status

3. Aetna Clinical Policy Bulletin (CPB): Polymerase Chain Reaction Testing: Selected Indications

In April 2017, Aetna revised the above-referenced CPB and considers polymerase chain reaction (PCR) testing medically necessary for the gastrointestinal pathogen panel for certain indications. 3

Aetna covers gastrointestinal pathogen panels for the following indications: I) community-acquired diarrhea of ≥: 7 days duration; 2) travel-related diarrhea; and 3) diarrhea with signs or risk factors for severe disease (fever, bloody diarrhea, dysentery, dehydration, severe abdominal pain, hospitalization and/or immunocompromised state).

The Aetna medical policy cites the ACG guidelines as a reference for this coverage benefit.

4. Clarification of Pathogens Listed in the xTAG GPP and Verigene EP Assays

We reviewed the list of pathogens in the proposed policy and provide the following clarifications for the pathogens listed in the assays.

xTAG® Gastrointestinal Pathogen Panel (GPP):

The following pathogens are not listed in the Noridian draft LCD: Vibrio cholera, Adenovirus 40/41, and Entamoeba histolytica. These pathogens are included in the FDA-cleared device. The FDA document K140377 lists the pathogens Vibrio cholera, Adenovirus 40/41, and En/amoeba histolytica as included in this FDA­ cleared device.4 This document supersedes the

other FDA documents and verifies that these pathogens are included in the GPP assay.

Verigene® Enteric Pathogens Nucleic Acid Test (EP):

The following viruses are not listed in the Noridian draft LCD: Norovirus GI/GU, and Rotavirus A. The FDA document K142033 confirms that Norovirus GI/GU and Rotavirus A are included in this FDA-cleared device.5 This document supersedes the other FDA documents and verifies that these pathogens are included in the EP assay.

Luminex requests Noridian revise the draft policy to include the above-referenced pathogens.

5. Revision Request for the EP Assay

Luminex Corporation manufactures the FDA-cleared Verigen®e Enteric Pathogens

Nucleic Acid Test. The Noridian draft LCD lists Nanosphere as the manufacturer. Luminex requests Noridian revise the policy to reflect Luminex as the manufacturer of the EP assay.

References:

1. Riddle MS, DuPont HL, Connor BA. ACG Clinical Guideline: Diagnosis, treatment, and prevention of acute diarrheal infections in adults. Am J Gastroenterol. 2016;111(5):602-622 .
2. Coste JF, et al. J. Clin. Microbiol. June 2013 vol. 51no.61841-1849.
3. Aetna Clinical Policy Bulletin (CPB): Polymerase Chain Reaction Testing: Selected Indications (0650). Available at http://www.aetna.com/
4. xTAG® Gastrointestinal Pathogen Panel (OPP) 51O(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY. FDA document K140377
5. Verigene® Enteric Pathogens Nucleic Acid Test (EP) 510(k) SUBSTANTIAL

EQUIVALENCE DETERMINATION DECISION SUMMARY. FDA document K142033.

Thank you for the comment. The explanation in the LCD text underlying the coverage decision agrees that the appropriate range of organisms for which to test in individuals who are immunocompromised may be broader than the appropriate organisms for which to test in immune competent individuals.

Some of the diagnoses that the comment suggests be added to group 1 would be indicated particularly in immunosuppressed beneficiaries. Since the draft LCD currently allows testing a for a large number of organisms in immunosuppressed patients, but it does not allow for testing for testing a lesser number of organisms in an immunosuppressed patient, we have added the ICD-10 codes which would indicate an immunosuppressed status to the list of group 1 codes.

The comment also recommends the addition of a number of ICD-10 codes to the list of group 2 covered diagnoses. We agree that transplantation can often be associated with immunosuppression. As such, we have included the transplantation diagnoses that would generally be associated with an immunosuppressed status.

BioFire agrees that the FilmArray GI Panel should only be used when it is reasonable and medically necessary. BioFire is in favor of appropriate use of diagnostic testing and believes that rapid and comprehensive diagnostic testing leads to better patient care and overall lower health care costs. It is reasonable to believe that better patient care is achieved when the health care provider can rapidly arrive at the patient diagnosis and treatment plan. The ideal diagnostic test would support this goal by providing a fast, accurate and comprehensive test result in a clinically actionable time frame.

While molecular panel assays entail greater reagent costs, overall health care cost savings can be achieved by reducing additional tests and medical procedures that are performed while the clinician tries to determine the cause of the illness, by reducing the number of health care encounters for follow-up evaluation, by reducing hospital length of stay (including ER time), by reducing the number of secondary cases, and by preventing adverse events associated with misdiagnosis and mistreatment. In addition, appropriate use of infection control practices can reduce the risk of nosocomial infections (by using appropriate contact precautions for patients with positive test results) and reduce cost and improve patient care by eliminating or reducing unnecessary patient isolation (for patients with negative test results or by identification of a pathogen that does not require isolation).

In fact, a poster presented at the recent AACC meeting (Beal, August 2017) found that implementation of the FilmArray GI Panel for pediatric and adult inpatients increased diagnostic yield from 6.9% to 32.8% and improved time to result from a mean of 54.75 hours to 8.94 hours when compared to traditional clinician ordered tests. Then compared to a matched historical control group, implementation of the FilmArray GI Panel lead to a reduction in the number of additional stool tests (3.02 vs 0.58, p=0.001), a trend toward shorter duration of antibiotics (2.12 days vs 1.54 day, p=0.06), significantly fewer imaging studies (0.39 vs 0.18, p=0.0002) and a reduction in length of hospital stay after sample collection (3.9 days vs 3.4 days, p=0.04). Reduced length of stay was more pronounced for the adult population (4.3 days vs 3.6 days, p=0.01).

The draft LCD describes limiting coverage for Gastrointestinal Pathogen (GIP) molecular assays to detection of 5 bacterial targets, suggests that testing for viral etiologies is not reasonable and necessary and limits testing of parasites to travelers with >2 weeks of symptoms with a negative test for bacterial pathogens. BioFire disagrees with these recommendations and finds that they are not supported by the current medical literature or by medical necessity.

Current testing for infectious gastroenteritis is suboptimal and requires clinicians to be intimately familiar with laboratory test composition and methods. Several studies have shown that clinician test ordering practices for gastroenteritis are problematic, in part due to the complexity of which pathogens are covered by what test (Hennessy 2004; van den Brandhof, 2006; McNulty, 2014; Polage, 2011). In addition, stool testing suffers from low diagnostic yield, technical complexity and long time to result. As a result, clinicians often order multiple tests for the same stool sample, or perform testing sequentially until a causative pathogen is identified. As an example, a study conducted at a children’s hospital (Stockmann, 2015) found that a median of 3 tests (range 1-10) were ordered per stool sample.

The FilmArray GI Panel was designed specifically to meet the unmet medical need of providing a fast, accurate and comprehensive diagnostic test for the identification of pathogens known to cause gastroenteritis. The composition of the FilmArray GI Panel was determined by a careful review of the medical literature and in consultation with medical and laboratory experts. Researchers at BioFire compiled lists of possible pathogens that were then carefully vetted by medical experts. Pathogen-specific assays were included on the FilmArray GI Panel only if the pathogen was known to cause gastroenteritis and if the panel of medical experts considered them to be medically indicated in the diagnosis of infectious gastroenteritis. In contrast, the menu for traditional microbiology tests (e.g., bacterial culture, O&P examinations) are often based on laboratory testing methods, not on medical necessity or the needs of the patient. For example, the draft LCD suggests that the Cryptosporidium is a common cause of food borne gastroenteritis, but the LCD only includes testing for 5 bacterial pathogens and states that ‘most people with Cryptosporidium…will recover without treatment’. In fact, Cryptosporidium can cause very protracted diarrhea, even in immunocompetent individuals, and a prospective, randomized, double-blind placebo-controlled clinical trial showed that nitazoxanide is effective for treatment of cryptosporidiosis (Rossignol, 2001). The rationale for exclusion of Cryptosporidium from the LCD appears to has more to do with the pathogens recovered by stool culture than by medical necessity and good patient care.

It is also widely recognized that there are many causative agents of gastroenteritis and that they cannot be reliable determined by clinical presentation. In fact, the American College of Gastroenterology (ACG) recently published a guideline for the diagnosis, treatment and prevention of acute diarrheal infections in adults (Riddle, 2016). This guideline includes the following quote,

‘As symptoms of acute diarrhea are protean, attempts to diagnose etiologic agents or classes are subjective at best and fraught with imprecision due to overlap in symptoms. Although features of the clinical presentation may be useful in distinguishing bacterial

from protozoan causes, they are often an unreliable indicator of the likely pathogen responsible. As with any syndromic disorder, there can be considerable overlap in symptoms caused by various agents.’

As opposed to using the composition or number of pathogens of any given multiplex panel as a guideline for determining medical necessity, the American College of Gastroenterology (ACG) and the Mayo Clinic have published clinical guidelines for the diagnosis of diarrheal illness (Riddle, 2016; http://www.mayomedicallaboratories.com/it-mmfiles/Laboratory_Testing_For_Infectious_Causes_of_Diarrhea.pdf). These guidelines acknowledge that most cases of gastroenteritis are self-limiting and that patients should be treated with rehydration and supportive care without the need to perform any microbiologic testing. The guidelines offer clinical criteria to determine when a microbiologic assessment should be performed. These are limited to;

• Patients that are at a high risk of spreading the disease (ACG)
• During known or suspected outbreaks (ACG)
• Patients with bloody diarrhea, with moderate to severe disease (defined as requiring alterations to normal function), and those with symptoms lasting > 7days (ACG and Mayo Clinic).

In addition, recommendation 3 of the ACG guideline states that;

“Traditional methods of diagnosis (bacterial culture, microscopy with and without special stains and immunofluorescence, and antigen testing) fail to reveal the etiology of the majority of cases of acute diarrheal infection. If available, the use of Food and Drug Administration-approved culture independent methods of diagnosis can be recommended at least as an adjunct to traditional methods.”

This statement highlights the well-known complexity of stool testing (consisting of several different and technically complex test methods) and the low diagnostic yield. As stated, even after performing several different tests, identification of the causative agent of disease is often not achieved. Indeed, the current IDSA Practice Guidelines for the Management of Infectious Diarrhea (Guerrant, 2001) reported the yield for stool cultures to be between 1.5 to 2.9% with a cost per positive result of $952 to $1200. In addition, a properly preformed O&P assay requires the collection and testing of three different stool samples. This is a method that is known to be technically difficult, to have low sensitivity, to be improperly used and to have a low diagnostic yield (1.4%, Polage, 2011). In fact, the reason that the use of multiplex molecular assays has increased is that they fill an unmet clinical need. Multiplex molecular tests were designed with a broader menu precisely because current testing methods suffer from low sensitivity and diagnostic yield.

In addition to appropriate antimicrobial therapy, implementation of appropriate infection control measures is extremely important for appropriate management of patients with infectious gastroenteritis. In a recent retrospective study (Rand, 2015), the FilmArray GI Panel was used to test frozen stool samples that had previously been tested for rotavirus and C. difficile for infection control purposes. The study showed that 22% of the samples contained pathogens that should have required infection control measures, including norovirus, rotavirus and C. difficile. Of these patients, 60% were under no or inadequate contact precautions, for a total of 109 patient days. Conversely 24.5% of the patients with negative results by the FilmArray GI Panel were unnecessarily placed under contact precautions for a total of 181 patient days. This study illustrates that without a rapid comprehensive test result, contact precautions are not rationally applied, leading to both an increased risk of nosocomial infections and unnecessary costs and morbidity associated with inappropriately applied contact precautions. These are important factors when considering the care of patients in hospitals and in long term care facilities. A separate study (Goldenberg, 2014) found that use of a CIDT multiplex panel for hospitalized patients was cost-effective because the increased cost of the laboratory testing was more than off-set by the cost saving for elimination of unneeded contact precautions.

According to the current IDSA guidelines (Guerrant, 2001), identification of the causative agent of gastroenteritis is important to appropriate treatment decisions. For example, antibiotics are contraindicated when Shiga toxin-producing E. coli (STEC, increased risk of HUS) and C. difficile (worsening of infection) are the causative agents. In contrast, antibiotics are indicated for cases of severe or prolonged gastroenteritis caused by Campylobacter, Vibro, Plesiomonas and Salmonella in infants. Further, while not typically indicated, antibiotics have been shown to shorten the duration of illness for gastroenteritis caused by Shigella, Enterotoxigenic E. coli (ETEC), Enteroaggregative E. coli (EAEC) and Enteropathogenic E. coli (EPEC). This is an important consideration, especially when treating elderly patients that are at higher risk of poor outcomes. In all cases, the recommended antibiotic therapy (drug and dosage) is dependent on the pathogen and the patient’s presentation. Use of antibiotics is not appropriate when gastroenteritis is caused by a viral pathogen and pathogen-specific therapies may be indicated for the treatment of parasitic infection (antibiotics or antiparasitic agents) and C. difficile disease (metronidazole or oral vancomycin). As identification of etiologic agents in infectious gastroenteritis is important to patient management, it is medically necessary and should be a covered benefit.

The use of culture-independent multiplex panels has greatly increased the diagnostic yield for stool testing due both to increased test sensitivity and an expanded test menu. Studies using the FilmArray GI Panel identified a pathogen in ~ 40-50% of stool samples (Spina, 2015; Buss, 2015; Stockmann, 2016). The FilmArray GI Panel is unique in that it also offers the advantage that the results are available within about one hour of test initiation, is technically easy to perform and requires very little technologist hands-on time.

Previous to this DRAFT LCD, BioFire had established a specific billing code (Z-code) that carried a lower reimbursement rate compared to the multiple pathogen CPT codes. The rationale for the specialized code was that the original test evaluation relied on a valuation of a traditional molecular testing method (Luminex XTag) that requires substantially more laboratory resources, including additional laboratory instrumentation (nucleic acid extraction instrument, PCR thermocyclers and the Luminex specific instrument), more technologist hands-on time and more peripheral supplies. The FilmArray GI Panel is a simple-to-use system that only requires about 2 minutes of technologist time, does not require any additional laboratory equipment (only the FilmArray instrument), and virtually everything that is needed to perform the test is included in the reagent test kit.

Use of FilmArray GI Panel to test patients that meet the clinical criteria described by medical experts (ACG, Mayo Clinic) along with a test specific reimbursement would offer the benefit of fast and accurate pathogen identification at a reasonable cost.

Historically, physicians were not required to select specific tests for individual pathogens. Samples were sent to the microbiology laboratory for ‘culture’ and the laboratory was expected to identify any pathogen. The use of immunoassays and molecular methods rely on organism-specific antibodies or primers and therefore required the clinicians to select pathogen-specific tests. Newer molecular technologies have allowed a return to simultaneous evaluation of each sample for a comprehensive set of pathogens associated with the specific clinical syndrome. However, reimbursement for IVD PCR assays are currently based on per target models that are out of date and that no longer make economic sense. BioFire encourages establishing a reimbursement level that is based on an economic assessment of the specific technology. In this way, patient care can be improved by providing comprehensive diagnostic information to patients suffering from gastroenteritis in a clinically actionable time frame without adversely impacting the cost of health care.

In this section, BioFire provides responses to the specific information presented in the Local Coverage Determination.

[MolDX has changed the formatting below using a mix of italics and plain text to avoid the use of red text, which was part of the commenter’s original comment.]

Coverage Indications, Limitations, and/or Medical Necessity

This contractor will provide limited coverage for Gastrointestinal Pathogen (GIP) molecular assays identified by multiplex nucleic acid amplification tests (NAATs), and will limit GIP coverage up to 5 bacterial targets which represent the top 90-95% of foodborne infections ([incidence of infection per 100,000 population]in decreasing incidence): Salmonella [15.89]; Campylobacter [12.97]; Shigella [5.53]; Cryptosporidium [3.31]; Shiga toxin producing E. coli (STEC) non-O157 [1.64] and STEC O157 [.95]. In immune competent individuals, most people with Cryptosporidium, a parasitic disease, will recover without treatment. The pathogens in some of the GIP panels are determined by the manufacturers that make them, and do not represent specific pathogens that cause a common age-based syndrome, or represent organisms that commonly are found in a specific sample type, patient population or reflect community acquired foodborne infections. Because of the unique clinical circumstances of immune compromised patients, ICU patients, and HIV positive patients with diarrhea, GIP testing for bacteria, virus and parasite testing may be indicated, and thus a Medicare benefit.

BioFire Response – Gastroenteritis has infectious and non-infectious causes. Infectious causes include bacteria, viruses and parasites and while many infections are foodborne, infections can also result from contaminated water sources and from person-to-person transmission. The intended use statement for the FilmArray GI Panel is ‘from individuals with signs and symptoms of gastrointestinal infection’. Testing coverage should not be limited to food-borne pathogens. However, even if it is limited, according to CDC, the leading causes of foodborne deaths, hospitalization and illnesses are (www.cdc.gov/foodborneburden/questions-and-answers.html)

• Nontyphoidal Salmonella, Toxoplasma, Listeria, and norovirus caused the most deaths.
• Nontyphoidal Salmonella, norovirus, Campylobacter, and Toxoplasma caused the most hospitalizations.
• Norovirus caused the most illnesses. Although norovirus usually causes a mild illness, norovirus is a leading cause of foodborne deaths because it affects so many people and can cause severe illness in the elderly.

In the UK, the most common causes of community gastroenteritis are norovirus (16.5%), sapovirus (9.2%), Campylobacter (4.6%) and rotavirus (4.1%) (Tam, 2012). There are also non-infectious causes, such as irritable bowel disease or side effects of medications, and it is often desirable to rule-out infectious causes of gastroenteritis when evaluating non-infectious causes.

The rationale to limit diagnostic testing to the 5 listed bacteria does not provide adequate information for the diagnosis of gastroenteritis. It ignores the viral agents, such as noroviruses, which are the most common etiologic agents. In the majority of these cases, the bacterial tests will be negative and then additional tests (e.g., individual viral PCR assay, EIA assays for viruses, O & P examinations) will be ordered, resulting in suboptimal patient care, delaying appropriate patient management, and increasing the risk of transmission.

Background

Traditionally, stool testing algorithms required physicians to consider which specific pathogens that might be associated with individual cases of gastroenteritis, and choose a testing scheme that ensured that all the appropriate pathogens were targeted. In the setting of community-acquired diarrheal illness, large foodborne GIP testing panels for parasites and viral etiologies is not reasonable and necessary because these GI diseases are:

• Generally self-limited,
• Virus specific therapies are not available, and
• Patients are managed by supportive care and hydration.

Travelers with >2 weeks of symptoms, after bacterial pathogens have been ruled out, may require traditional ova and parasite stool examination and/or specific protozoa antigen or molecular testing.

BioFire response – The ACG and Mayo Clinic guidelines also state that most cases of gastroenteritis are self-limiting and suggest that no microbiology work up should be performed unless the patient meets specific clinical criteria or risk factors. BioFire agrees that physicians should choose a testing scheme that ensures that all the appropriate pathogens are targeted. If the clinical evidence suggests a specific pathogen, then a specific test should certainly be ordered. However, in many cases, the clinical presentation is not suggestive of a specific pathogen and a comprehensive, accurate and fast result increases the chance of identifying the causative agent of disease at a time when it is most likely to benefit the patient. The idea that tests for viral pathogens should not be a covered benefit simply because there is no specific therapy is short sighted. Medicare covers many tests for detection of viral pathogen for which there is no specific treatment (i.e., individual norovirus PCR assay, EIA assays for rotavirus). That is because patient management is more than prescribing a medication. Identification of a viral pathogen confirms that antibiotics are not necessary and directs clinicians to focus on supportive care and hydration. Ruling out a comprehensive list of gastrointestinal pathogens can aid in the proper diagnosis and treatment of non-infectious gastroenteritis, such as irritable bowel syndrome or can be used to remove patients from unnecessary and costly contact precautions. Additionally, for immune compromised persons, particularly in the transplant setting, the rapid identification of a pathogen can be life-saving and prevent nosocomial transmission.

Background (Cont.)

Medicare specifies that testing must be reasonable and necessary for the specific needs of a given patient. Large panels that represent a “one size fits all” approach to testing without regard for a patient’s medical history, time of year, clinical setting, and patient symptoms are not reasonable and necessary, and thus not a Medicare benefit. A “one size fits all” panel approach is not restricted to specific population subgroups, such as neonates, pediatrics, or adults, does not differential between community-acquired vs traveler source of infection, and does not differentiate the needs of select patient populations such as the ICU patient or immunocompromised patients. In addition, while identification of specific viruses may be of interest in an outbreak or epidemiologically, clinical management is not predicated on viral test results, and are thus not reasonable and necessary.

BioFire response – The composition of the FilmArray GI Panel was determined by a careful review of the literature and in consultation with medical experts. The panel was designed to address the low diagnostic yield, slow time to result, and the reality that most stool samples require testing using multiple different tests often without identification of the causative pathogen. This delays time to results, wastes resources and results in sub-optimal patient care because the treatment plan must be determined without key diagnostic information. The FilmArray GI Panel is meant to be similar to a culture test, in which the laboratory is asked to look for a comprehensive set of pathogens that can cause a specific clinical syndrome (in this case, gastroenteritis). Traditional testing has such a low yield and the time to result is so slow, that the value of testing is often questioned (Guerrant, 2001) and the cost per positive result is very high. BioFire encourages physicians to carefully consider a particular patient’s medical history, time of year, clinical setting, and patient symptoms before ordering the FilmArray GI panel and in interpreting the FilmArray GI panel results.

Background (Cont.)

This contractor recognizes that GIP assays are closed systems, without random access for physician directed, patient-specific testing. However, some laboratories elect to use GIP panel tests but report only the specific tests ordered by the physician. In other words, and the laboratory “blinds” unnecessary test results or utilize disclaimers in their reporting and bill only for the medically necessary test results. Other laboratories report results of all tests in the panel which adds unnecessary cost to the healthcare system when reimbursement is directly related to the number of organisms in the panel. The FDA approved/cleared assays discussed below are comparable with coverage limited to bacterial organisms for acute diarrhea, with justification of medical necessity recorded in the patient’s medical record.

BioFire response – As stated above, BioFire does not believe that reimbursement should be directly related to the number of organism on a panel. Instead, the economics of the individual test system combined with the added medical value of clinically actionable diagnostic results should be used to determine an appropriate reimbursement level. Current reimbursement for the FilmArray System is based on technology assessment for traditional PCR based assays. These current models do not apply to easy-to-use sample-to-answer systems. The reimbursement structure should be modified so that these valuable tests are available for appropriate patient care. BioFire already initiated the process of assigning a specific Z-code with a reduced reimbursement rate to address this issue. We would welcome the chance to discuss an appropriate reimbursement rate for the FilmArray GI Panel.

Nucleic Acid Amplified Probe Technique (NAAT) for Identification of Microorganisms: Tests performed by NAAT uses a microorganism’s DNA or RNA to directly identify specific bacteria, viruses, and/or protozoa rather than standard microorganism detection techniques such as bacterial culture, microscopy with and without stains, direct fluorescent antibody testing, rapid antigen testing, qualitative and quantitative immunoassay for identification of antigens or toxins from stool and single-plex PCR assays. Multiplex NAAT tests are included in the larger grouping of culture-independent diagnostic tests (CIDT). CIDT includes but is not limited to simplex direct probe and amplified probe techniques. This technology offers same day results in a matter of hours rather than 2-3 days of time-consuming and labor intensive bacterial cultures and immunoassays for processing stool specimens. CIDT are touted as providing a more comprehensive assessment of disease etiology by increasing the diagnostic yield compared with conventional diagnostic tests permitting earlier initiation of appropriate therapeutic agents targeted to the detected pathogen(s), if any, rather than empirical therapy until culture results are available. CIDT testing is not without its challenges; latent infections or colonization cannot be distinguished from active, clinically significant infections. Additionally, fragments of nucleic acids from dead microorganisms may cloud organism identification, complicating clinical interpretation, and potentially, clinical management. In a CIDT comparative study, mixed infections were identified in 13-21% of positive prospective stool samples compared to only 8.3% by routine (culture/immunoassay/microscopy) methods.¹ In another recent study, 32.9% of the FilmArray GI Panel-positive specimens were found to contain more than one potential pathogen.² The significance of detecting coinfections may be difficult to understand, as the clinical implications of specific pathogen combinations are not well documented or understood. Many GI pathogens can be shed asymptomatically or for prolonged periods of time after symptoms subside, further complicating the interpretation of positive results. For example, Salmonella spp. and norovirus can be shed for weeks to months after symptoms subside. Asymptomatic infection with Cryptosporidium spp. or G. lamblia is common in children.² High rates of asymptomatic carriage of enteropathogens, often identified as a co-infection in large microbial panels, create diagnostic confusion by the interpreting clinician.³

BioFire response – BioFire agrees with the information provided in this section. With all new medical technology, there are new questions raised by new information. However, it is simply true that patients with gastroenteritis often have co-infections and more than one pathogen may contribute to the symptoms. These are areas of active investigation and the clinical implications will be clarified via patient outcomes studies.

Nucleic Acid Amplified Probe Technique (NAAT) for Identification of Microorganisms: (Cont.)

From a public health and epidemiologic point of view, CIDT testing does not provide the culture isolates that are needed for antimicrobial susceptibility testing, serotyping, subtyping and whole genome sequencing that are critical for monitoring trends, detecting clusters of illness and investigating outbreaks. For Salmonella, the inability to distinguish serotypes will prevent tracking of important changes in incidence by serotype, and markedly limit detection and investigation of outbreaks (not a Medicare benefit). For Shiga toxin producing E. coli (STEC), because identification of serogroups requires culture, it is not known which STEC-positive CIDT result represents 0157 vs non-0157.⁴

BioFire response – Per the logic describe in this LCD, the impact to public health should not be a consideration of medical necessity because it is not a medical benefit to the specific patient. However, rapid outbreak detection is one of the important benefits of the increased sensitivity and menu of CIDT molecular panels. In fact, during the 8 month clinical trial for the FilmArray GI Panel, two different outbreaks of gastroenteritis were detected (Cyclospora in Nebraska (Buss, 2013) and Shigellosis in Rhode Island (Prakash, 2015). In fact, the following quote is made by public health officials in a recent publication about the impact of CIDT to foodborne-disease surveillance (Shea, 2017).

‘Culture-independent diagnostic tests (CIDTs) are improving diagnostics in ways that greatly benefit patient care. CIDTs can test for an array of clinically important infections, including respiratory, bloodstream, and enteric infections, more quickly and effectively than other methods can. Many of the benefits and challenges to CIDT implementation were discussed at a CIDT Forum organized by APHL and CDC in April 2012 (1). APHL lauds the benefits of CIDT technology while suggesting steps all players can take to avoid unintended negative consequences (2).’

Public health officials recognize that new technologies must be developed to replace older methods that rely on recovery of culture isolates (e.g., sequencing). However the CDC, APHL, clinical laboratories and IVD manufacturers also recognize that it is critical that the public and private sectors collaborate to find ways to obtain isolates when a CIDT yields a positive result. BioFire’s VP of Medical and Scientific Affairs (Dr. Christine Ginocchio) is actively working with CDC on this issue and BioFire fully supports the “Potential Short-Term Solutions” that include efforts to:

1. Encouraging clinical laboratories to work with public health laboratories to continue culturing and isolating the harmful bacteria from ill people with positive CIDTs. Patient specimens with a positive CIDT for Salmonella, Shiga toxin-producing coli, and Shigella should be cultured to isolate the bacterial strain. Selected laboratories should also do this for Campylobacter.
2. CDC is considering ways to make follow-up cultures easier and cheaper for clinical laboratories.
3. CDC is working closely with the APHL, public health officials, regulatory agencies, diagnostic laboratories, CIDT kit manufacturers, and clinicians to make sure that cultures are obtained in clinical labs when CIDTs are positive, or the positive specimen from the patient is provided to public health laboratories so they can culture it.
4. CD alive so they can be cultured if the test is positive. BioFire is in compliance with this request as our samples are tested from Cary-Blair transport medium that supports preserving viable organisms.
5. CDC is adapting surveillance systems like the Foodborne Diseases Active Surveillance Network, or FoodNet (https://www.cdc.gov/foodnet/index.html), to include infections diagnosed only by CIDTs.

In some cases, the clinical laboratories perform the culture and in other cases the sample is sent to the public health laboratory for culture recovery. In both cases, the culture can be limited to recovery of a specific organism and does not require a full stool culture work-up. This method has been reasonably successful at supplying culture isolates for public health needs and has increased the ability to detect food borne outbreaks.

As a potential long term solution the CDC is working with partners to develop advanced testing methods (next generation sequencing) that will not require bacterial isolates to provide information needed by public health officials. These tests may also provide additional information to healthcare providers about the pathogen’s potential for antibiotic resistance or likelihood of causing serious illness.

FDA-approved GIP Assays: Five FDA approved GIP assays are currently on the market, and all are closed system tests that do not allow random access for physician’s to select likely etiologic agents of diarrhea. These include:

• Hologic/Gen-Probe’s ProGastro SSCS:
  • Targets identified:
    • Salmonella,
    • Shigella,
    • Campylobacter (C. jejuni and C. coli only, undifferentiated) nucleic acids, and
    • Shiga toxin 1 (stx1) /Shiga toxin 2 (stx2) genes (STEC typically harbor one or both genes that encode for Shiga toxins 1 and 2)
  • TAT (turn-around time) - 4 hr.

• BD Diagnostics’ BD MAX Enteric Bacterial Panel (EBP):
  • Targets identified:
    • Campylobacter spp. (jejuni and coli),
    • Salmonella spp.,
    • Shigella spp.,
    • Enteroinvasive E. coli (EIEC),
    • Shiga toxin 1 (stx1)/Shiga toxin 2 (stx2) genes (found in STEC, as well, as Shigella dysenteriae
  • TAT – 3-4 hr.

• Nanosphere’s Verigene Enteric Pathogens (EP):
  • Targets identified:
    • Campylobacter Group (comprised of C. coli, C. jejuni, and C. lari),
    • Salmonella species,
    • Shigella species (including S. dysenteriac, S. boydii, S. sonnei and S. flexneri),
    • Vibrio Group (comprised of V. cholerae and V. parahoemolyticus),
    • Yersinia enterocolitica,
    • Shiga toxin I gene and Shiga toxin 2 gene virulence markers, Shiga toxin producing E coli (STEC)
  • TAT – 2 hr.

• Luminex’s xTAG Gastroenterology Pathogen Panel (GPP):
  • Targets identified:
    • Campylobacter (C. jejuni, C. coli and C. lari only)
    • Clostridium difficile (C. difficile) toxin A/B
    • Cryptosporidium (C. parvum and C. hominis only)
    • Escherichia coli (E. coli) O157
    • Enterotoxigenic E. coli (ETEC) LT/ST
    • Giardia (G. lamblia only) (aka G. intestinalis and G. duodenalis)
    • Norovirus GI/GII
    • Rotavirus A
    • Salmonella
    • Shiga-like Toxin producing E. coli (STEC) stx 1/stx 2
    • Shigella (S. boydii, S. sonnei, S. flexneri and S. dysenteriae)
  • TAT - <5 hr.

• Biofire Diagnostic’s FilmArray GI Panel:
  • Targets identified
    • Campylobacter (C. jejuni/C. coli/C. upsaliensis),
    • Clostridium difficile (C. difficile) toxin A/B ,
    • Plesiomonas shigelloides,
    • Salmonella,
    • Vibrio (V. parahaemolyticus/V. vulnificus/ V. cholerae), including specific identification of Vibrio cholerae,
    • Yersinia enterocolitica,
    • Enteroaggregative Escherichia coli (EAEC),
    • Enteropathogenic Escherichia coli (EPEC),
    • Enterotoxigenic Escherichia coli (ETEC) lt/st,
    • Shiga-like toxin-producing Escherichia coli (STEC) stx1/stx2 (including specific identification of the E. coli O157 serogroup within STEC),
    • Shigella/ Enteroinvasive Escherichia coli (EIEC),
    • Cryptosporidium,
    • Cyclospora cayetanensis,
    • Entamoeba histolytica,
    • Giardia lamblia (also known as G. intestinalis and G. duodenalis),
    • Adenovirus F 40/41,
    • Astrovirus,
    • Norovirus GI/GII,
    • Rotavirus A,
    • Sapovirus (Genogroups I, II, IV, and V)
  • TAT -1-2 hr.

All targeted viruses included in GIPs are more prevalent in young children than in adults. In one study, sapovirus was detected in 10% of all specimens from children >1 year old and 7.4% of specimens from children between 1 to 5 years of age.² enteropathogenic E. coli (EPEC), historically associated with developing countries, are known to cause both acute and persistent diarrhea in young children in the US and were identified in one study in 24.8% of all samples collected from children <1 year of age, and 37% of all samples from children between age of 1 and 5 years. EPEC strains can also be found in healthy children and adults, thus confounding its significance when identified in symptomatic children and adults.

BioFire Response: Although high rates of viral infections are found in children a recent systematic review of 225 articles regarding the burden of gastroenteritis in China identified high rates (∼29%) of norovirus infections in adults and the elderly and in children aged 6-35 months (∼22%) (Zhou, 2017). An additional study conducted in Alberta, Canada identified viruses in diarrheal stools from 17.0% of persons tested, with norovirus GII the most frequent (8.0%), followed by sapovirus (4.3%), rotavirus (2.0 %), astrovirus (1.8 %), norovirus GI (0.9 %), and norovirus GIV (0.1%) (Leblanc, 2017). The prevalence of mixed viral infections in individuals with diarrhea was 2.8% (n = 11). Children from 1 to 5 years of age accounted for the highest prevalence of positive stools, followed by elderly individuals (≥70 years). Severe viral gastroenteritis in the elderly can lead to rapid dehydration, and the failure to rapidly detect an outbreak in chronic long term care facilities can lead to increased transmission, hospitalizations, morbidity and mortality (Rajagopalan, 2016). A review of the literature demonstrated that older adults are at increased risk of severe norovirus-associated health outcomes, hospitalization rates were higher, the length of stay was longer, the disease was more severe, and patients incurred greater costs than younger patients (Lindsay, 2015). Mortality rates among individuals 65 years and older with norovirus were approximately 200% higher as compared to children <5 years, emphasizing the need for rapid detection for older adults and to prevent institutional spread.

FDA-approved GIP Assays: (Cont.)

Similarly, the interpretation of C. difficile toxin A/B detection is also complicated, especially in children <1 year old. The American Academy of Pediatrics does not recommend routine testing for C. difficile in children <1 year of age and suggests that positive C. difficile results be interpreted with suspicion in children <3 years old.⁵

BioFire Response - Recent studies have documented a considerable burden of C. difficile infections in the community setting (Chitnis, 2013). Many of these patients lack traditional risk factors such as recent hospitalization or antibiotic use (Khanna, 2012). The use of GI panels has increased the detection of unsuspected C. difficile infections when clinicians were seeking a different diagnosis (Stockmann, 2014).
 

Indications for Foodborne GI Testing Acute diarrhea, often called gastroenteritis, can be defined as the passage of a greater number of stools of decreased form from the normal lasting < 14 days. Acute diarrhea is generally associated with clinical features of nausea, vomiting, abdominal pain and cramps, bloating, flatulence, fever, passage of bloody stools, tenesmus and fecal urgency. It is the leading cause of outpatient visits, hospitalizations, and lost quality of life occurring domestically and those traveling abroad. The Centers for Disease Control and Prevention (CDC) has estimated 47.8M cases occurring annually in the US with an estimated healthcare cost upwards of US$150M.⁶ Detection of microbial pathogens associated with GI disease may be important in certain populations, such as immunocompromised hosts, the critically ill and individuals with prolonged disease that is refractory to treatment. Over a 20 year period, some foods that have been linked to food-borne outbreaks including milk (Campylobacter), shellfish (noroviruses), unpasteurized apple cider (Escherichia coli O157:H7), raw and undercooked eggs (Salmonella), fish (ciguatera poisoning), raspberries (Cyclospora); strawberries (Hepatitis A virus); and ready-to-eat meats (Listeria).⁷ Although the etiologic agents responsible for about 80% of GI illnesses are unidentified or otherwise unspecified, Norovirus and Salmonella spp (non-typhoidal) are currently the most commonly identified pathogens associated with food-borne disease in the US and account for 5.5 and 1.0 million cases each year, respectively.⁸ Clostridium perfringens, Campylobacter and Staphylococcus aureus follow norovirus and Salmonella spp. in decreasing frequency in domestically acquired foodborne illnesses. Healthcare- and antibiotic-associated diarrhea is also problematic, with the major causative pathogen being toxin-producing Clostridium difficile.⁹ In the US, >300,000 cases of C. difficile are diagnosed annually, with associated costs of >$1 billion.

BioFire Response – The prevalence numbers provided above are not consistent with the information provided in the LCD coverage determination. The prevalence in this section is more in line with the published literature. Indications for Foodborne GI Testing (Cont.)

In 2015, the number and incidence of confirmed infections per 100,000 population were reported for Salmonella (15.89), Campylobacter (12.97), Shigella (5.53), Cryptosporidium (3.31), Shiga-toxin producing Escherichia coli (STEC) non-O157 (1.64), STEC O157 (.95), Vibrio (0.39), Yersinia (0.29), Listeria (0.24) and Cyclospora (0.13).⁴ Among confirmed infections, the vast majority were diagnosed only by culture. Compared with incidence in 2012-2014, the incidence of confirmed infections was significantly higher for STEC non-O157 (40% increase) and Cryptosporidium (57% increase). No significant changes were observed in 2015 for other pathogens compared with the previous 3-year averages.⁴ In addition to the 20,107 confirmed cases of infection, there were 3,112 positive CIDT case reports. In general, the incidence of most foodborne bacterial pathogens and for Cryptosporidium is highest among children aged <5, except for Listeria and Vibrio for which the highest incidence is among persons aged ≥ 65 years.¹⁰

BioFire Response – The prevalence reported in the above paragraph are at odds with the previous paragraph and reflect the limited number of pathogens that are considered. Norovirus is the most common cause of gastroenteritis and the most common cause of hospitalization and death caused by gastroenteritis. Inconsistencies in the rates reported are directly related to the great variety of testing methods, the varying sensitivity of the test methods, what pathogens are reportable and the lack of standardized comprehensive testing. Low rates are some pathogens are a direct result of a simple lack of routine diagnostic testing.

Indications for Foodborne GI Testing (Cont.)

Many episodes of acute diarrhea are self-limited and require fluid replacement and supportive care. Oral rehydration is indicated for patients who are mildly to moderately dehydrated. IV fluids may be required for more severe dehydration. Routine use of antidiarrheal agents is not recommended because many of these agents have potentially serious adverse effects, particularly in infants and young children. Antimicrobial therapy is warranted only for patients with severe disease or for individuals with immune systems are severely weakened from medications and other illnesses.¹¹ Laboratory testing algorithms for infectious causes of diarrhea generally agree that testing is NOT warranted for community-acquired diarrhea of <7 days duration without signs or symptoms of severe (fever, bloody diarrhea, dysentery, severe abdominal pain, dehydration, hospitalization and immunocompromised state) disease. In general, when community-acquired diarrhea persists for ≥7 days, or the diarrhea is travel-related, or there are signs/symptoms of severe disease, GIP testing may be warranted. Additional directed testing may be indicated if the GIP results are negative and diarrhea persists. No additional testing is indicated for GIP-positive result unless the clinical pictures changes. Clostridium difficile molecular testing is warranted on health-care associated diarrhea with onset after the 3rd inpatient day or after recent antibiotic use.

BioFire response – This information is consistent with using patient signs and symptoms, not test menu to determine when testing is medically necessary. If the testing for this patient population is limited to the listed bacterial pathogens, less than 3% of cases would have a pathogen identified (Guerrant, 2001).

Summary Medicare Coverage Decision:

GIP testing is limited to no more than 5 bacterial pathogen targets. Testing for viral etiologies is not reasonable and necessary because these GI diseases are generally self-limited, virus specific therapies are not available, and patients are managed by supportive care and hydration. Travelers with >2 weeks of symptoms, after bacterial pathogens have been ruled out, may require traditional ova and parasite stool examination and/or specific protozoa antigen or molecular testing. Large panels inclusive of viruses and protozoa are not reasonable and necessary for community-acquired diarrheal illness.

There is no Medicare benefit for GIP testing for national, state or local agency tracking of diarrheal outbreaks, for epidemiologic purposes, or to confirm another etiologic test result. Once the target etiology of an outbreak is identified, subsequent patient testing is generally not indicated and patients are managed empirically. However, if the clinical presentation varies from the outbreak prototype, a specific test for the causative organism may be indicated. The Medicare benefit is specifically for the clinical identification and management of disease for a given beneficiary. The Medicare benefit does not extend for purposes of the family or community tracking or surveillance.

Limitations

A GIP test panel is a single service with a single unit of service (UOS =1). A panel cannot be unbundled and billed as individual components regardless of the fact that the GIP test reports multiple individual pathogens and/or targets. The panel is a closed system performed on a single platform, and as such, is a single test panel with multiple components (UOS=1).

If C. difficile is not included in a GIP panel, testing for C. difficile may be reasonable and necessary when ordered in addition to a GIP bacterial pathogen panel and supported by documentation in the medical record.

References

Buss SN, Alter R, Iwen PC, Fey PD. Implications of Culture-Independent Panel-Based Detection of Cyclospora cayetanensis. 2013 Nov;51(11):3909.

Buss SN, Leber A, Chapin K, Fey PD, Bankowski MJ, Jones MK, et al. Multicenter Evaluation of the BioFire FilmArray Gastrointestinal Panel for Etiologic Diagnosis of Infectious Gastroenteritis. Journal of Clinical Microbiology. 2015;53(3):915–25.

Chitnis AS, Holzbauer SM, Belflower RM, Winston LG, Bamberg WM, Lyons C, et al. Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011. JAMA Intern Med. 2013 Jul 22;173(14):1359-67. doi: 10.1001/jamainternmed.2013.7056.

Goldenberg SD, Bacelar M, Brazier P,Bisnauthsing K, Edgeworth, JD. A cost benefit analysis of the Luminex xTAG Gastrointestinal Pathogen Panel for detection of infectious gastroenteritis in hospitalised patients. J Infect. 2015 May;70(5):504-11.

Guerrant RL, Van Gilder T, Steiner TS, Thielman NM, Slutsker L, Tauxe RV, et al. Practice guidelines for the management of infectious diarrhea. Clin Infect Dis. 2001 Feb 1;32(3):331-51.

Hennessy TW, Marcus R, Deneen V, Reddy S, Vugia D, Townes J, et al. Survey of physician diagnostic practices for patients with acute diarrhea: clinical and public health implications. Clin Infect Dis. 2004 Apr 15;38 Suppl 3:S203-11.

Khanna S, Pardi DS, Aronson SL, Kammer PP, Orenstein R, St Sauver JL, et al. The epidemiology of community-acquired Clostridium difficile infection: a population-based study. Am J Gastroenterol. 2012 Jan;107(1):89-95.

Leblanc D, Inglis GD, Boras VF, Brassard J, Houde A. The prevalence of enteric RNA viruses in stools from diarrheic and non-diarrheic people in southwestern Alberta, Canada. Arch Virol. 2017 Jan;162(1):117-128.

Lindsay L, Wolter J, De Coster I, Van Damme P, Verstraeten T. A decade of norovirus disease risk among older adults in upper-middle and high income countries: a systematic review. BMC Infect Dis. 2015 Oct 14;15:425.

McNulty CA, Lasseter G, Verlander NQ, Yoxall H, Moore P, O'Brien SJ, et al. Management of suspected infectious diarrhoea by English GPs: are they right? Br J Gen Pract. 2014 Jan;64(618):e24-30.

Murphy CN, Fowler RC, Iwen PC, Fey PD. Evaluation of the BioFire FilmArray® Gastrointestinal Panel in a Midwestern Academic Hospital. Eur J Clin Microbiol Infect Dis. 2017 Apr;36(4):747-754.

Polage CR, Stoddard GJ, Rolfs RT, Petti CA. Physician use of parasite tests in the United States from 1997 to 2006 and in a Utah Cryptosporidium outbreak in 2007. J Clin Microbiol. 2011 Feb;49(2):591-6.

Prakash VP, Leblanc L, Alexander-Scott NE, Skidmore J, Simmons D, Quilliam D, et al. Use of a Culture-Independent Gastrointestinal Multiplex PCR Panel during a Shigellosis Outbreak: Considerations for Clinical Laboratories and Public Health. Journal of Clinical Microbiology. 2015Jul;53(3):1048–9.

Rajagopalan S, Yoshikawa TT. Norovirus Infections in Long-Term Care Facilities. J Am Geriatr Soc. 2016 May;64(5):1097-103.

Rand KH, Tremblay EE, Hoidal M, Fisher LB, Grau KR, Karst SM. Multiplex gastrointestinal pathogen panels: implications for infection control. Diagnostic Microbiology and Infectious Disease. 2015;82(2):154–7.

Riddle MS, DuPont HL, Connor BA. ACG Clinical Guideline: Diagnosis, Treatment, and Prevention of Acute Diarrheal Infections in Adults. Am J Gastroenterol. 2016 May;111(5):602-22.

Rossignol JF, Ayoub A, Ayers MS. Treatment of diarrhea caused by Giardia intestinalis and Entamoeba histolytica or E. dispar: a randomized, double-blind, placebo-controlled study of nitazoxanide. J Infect Dis. 2001 Aug 1;184(3):381-4.

Shea S, Kubota KA, Maguire H, Gladbach S, Woron A, Atkinson-Dunn R, et al. Clinical Microbiology Laboratories' Adoption of Culture-Independent Diagnostic Tests Is a Threat to Foodborne-Disease Surveillance in the United States. J Clin Microbiol. 2016 Dec 28;55(1):10-19.

Spina A, Kerr K, Cormican M, Barbut F, Eigentler A, Zerva L, et al. Spectrum of enteropathogens detected by the FilmArray GI Panel in a multicentre study of community-acquired gastroenteritis. Clinical Microbiology and Infection. 2015;21(8):719–28.

Stockmann C, Rogatcheva M, Harrel B, Vaughn M, Crisp R, Poritz M, et al. How well does physician selection of microbiologic tests identify Clostridium difficile and other pathogens in paediatric diarrhoea? Insights using multiplex PCR-based detection. Clinical Microbiology and Infection. 2015;21(2).

Stockmann C, Pavia AT, Graham B, Vaughn M, Crisp R, Poritz MA, et al. Detection of 23 Gastrointestinal Pathogens Among Children Who Present With Diarrhea. Journal of the Pediatric Infectious Diseases Society. 2016Apr.

Tam CC, O’Brien SJ, Tompkins DS, Bolton FJ, Berry L, Dodds J, et al. Changes in Causes of Acute Gastroenteritis in the United Kingdom Over 15 Years: Microbiologic Findings From 2 Prospective, Population-Based Studies of Infectious Intestinal Disease. Clin Infect Dis 2012 54 (9): 1275-1286.

van den Brandhof WE, Bartelds AI, Koopmans MP, van Duynhoven YT. General practitioner practices in requesting laboratory tests for patients with gastroenteritis in the Netherlands, 2001-2002. BMC Fam Pract. 2006 Oct 2;7:56.

Zhou HL, Zhen SS, Wang JX, Zhang CJ, Qiu C, Wang SM, et al. Burden of acute gastroenteritis caused by norovirus in China: A systematic review. J Infect. 2017 Jun 17. pii: S0163-4453(17)30208-6.

We appreciate the detailed comments and analysis of the LCD. We are expanding indications in our LCD to allow up to 11 pathogens to be tested if Clostridium difficile is one of them.

The comment above makes the point that many pathogens involved in diarrheal and foodborne illness are associated with adverse events. However for many pathogens we are aware of no evidence showing that pathogen identification will necessarily predict outcomes reliably in an individual patient, and as since the treatment in many cases is similar supportive treatment guided by an individual’s signs and symptoms regardless of the underlying pathogen, positive pathogen identification does not necessarily provide clinically actionable information. The draft LCD covers testing to as to allow pathogen identification where it may be clinically actionable. The specific pathogens for which testing is recommended in IDSA guidelines for treatment decisions (mentioned in the above comment) are coverable under this LCD.

I am the founder of OmniPathology, a physician owned pathology laboratory specialized in GI pathology. I am a fellowship trained Gastrointestinal Pathologist working with many Gastroenterologist reviewing their patients’ biopsies. I receive consultations from my pathologist colleagues asking for my expert opinion in difficult GI cases. I am also involved in training pathology residents and fellows at UCLA and Harbor UCLA Medical Centers.

I am writing to you in regards to your proposed LCD DL37350 Proposed LCD Title MolDX: Foodborne Gastrointestinal Panels Identified by Multiplex Nucleic Acid Amplification Tests (NAATs).

Titled: Coverage Indications, Limitations, and/or Medical Necessity

“This contractor will provide limited coverage for Gastrointestinal Pathogen (GIP) molecular assays identified by multiplex nucleic acid amplification tests (NAATs), and will limit GIP coverage up to 5 bacterial targets which represent the top 90-95% of foodborne infections ([incidence of infection per 100,000 population]in decreasing incidence): Salmonella [15.89]; Campylobacter [12.97]; Shigella [5.53]; Cryptosporidium [3.31]; Shiga toxinproducing E. coli (STEC) non-O157 [1.64] and STEC O157 [.95].”

Background on the FilmArray GI Panel

The FilmArray GI Panel tests for 22 gastrointestinal pathogens—13 bacteria, 4 parasites, and 5 viruses. The test setup requires 2 minutes of hands-on time and delivers results in about an hour. At the time of completion, a report is generated that indicates the result for each target. Identifying the etiological agents of my patient’s gastrointestinal infection using the FilmArray GI Panel enables my clients (gastroenterologists and primary care physicians) as the ordering physician to make an informed therapy decision and prescribe targeted treatment based on the pathogens causing the diarrheal condition. The patients receive treatment for infections much more quickly than if a conventional stool culture had been ordered in place of the FilmArray GI Panel.

Additionally, early infection identification not only assists in determining the appropriate therapy but addresses the unnecessary prescribing of antibiotics for viral conditions.

Despite the high incidence of gastrointestinal infection, there are limited clinical guidelines available for the diagnosis and treatment of patients with suspected gastrointestinal infection. Furthermore, traditional testing methods present a number of challenges in diagnosing the causative infectious agent. Challenges with traditional testing methods include:

• time-consuming
• low-yield
• insensitive/unspecific
• complex to perform
• collection, storing and transport of multiple specimens on multiple days is cumbersome and highly unpleasant for patients/family members adding to delay and noncompliance with advised testing)

Rationale for Clinical Use

As a result of the current diagnostic and treatment paradigm for suspected gastrointestinal infection many physicians choose empirical therapy with broad-spectrum antibiotics as a first-resort, or may prescribe an antimicrobial that is ultimately ineffective against the particular pathogen, which may lead to drug resistance and a prolonged disease state. In addition, many gastrointestinal infections can result in post-illness complications including Guillain-Barré syndrome, irritable bowel syndrome, and reactive arthritis.

Unlike traditional stool testing methods, the FilmArray GI Panel is fast, accurate, and comprehensive.

As an example, we recently had a patient, a medical student who just came back from a volunteer trip in Africa. He was suffering from severe diarrhea and was in town visiting his family for the weekend before going back to school. His family brought him to one of my clients, who ordered the test on a Friday. We ran the test which was positive for Enteroaggregative and Enterpathogenic E. coli. My client was able to prescribe appropriate antibiotic treatment and the patient was already improving before flying back to his school on Sunday night. Without the ability to rapidly provide results, this patient would have not been able to receive treatment in such a short time. Furthermore, knowing the exact infectious organisms allowed for a specific antibiotic coverage and avoided the empiric antibiotic treatment approach.

In the microscopic review of biopsies taken from patients with chronic diarrhea, the pathologist encounters morphologic changes that, at times, can be nonspecific. S/he would frequently report the findings descriptively providing the endoscopist with a differential diagnosis that includes long lasting infectious process and inflammatory bowel disease. In many cases primary presentation of IBD can lack features of chronic injury and ruling out an infectious process helps the clinician narrow down the diagnosis. As you can see, with the help of such a comprehensive panel, the clinician obtains better quality of information.

Therefore, to limit the compensation to 5 organisms, in addition to being arbitrary, it would be a disservice to the patients. Furthermore, there is no option to test for less organisms in these panels and if the option exists, it would be impossible to predict which ones to choose. We live in a diverse society where people travel to and from different parts of the world, and relying on what is “common” and “uncommon”, in this case, would lead us into an unscientific guessing process.

We agree that many organisms could cause a given patient’s symptoms. However, for many organisms positive identification does not necessarily provide clinically useful information, and the LCD is intended to cover information with clinical utility. The case example given above does not sound representative of the Medicare population. If research is available demonstrating the use of broad testing in a population representative of the Medicare population, we would be willing to reconsider this coverage decision.

As CAC representative from California Medical Association and representing California GI specialty societies also, I have comments I hope Noridian will consider re several draft LCDs

DL37350 Proposed LCD Title MolDX: Foodborne Gastrointestinal Panels Identified by

Multiplex Nucleic Acid Amplification Tests (NAATs)

This is what you propose as

Coverage Indications, Limitations, and/or Medical Necessity

“This contractor will provide limited coverage for Gastrointestinal Pathogen (GIP) molecular assays identified by multiplex nucleic acid amplification tests (NAATs), and will limit GIP coverage up to 5 bacterial targets which represent the top 90-95% of foodborne infections ([incidence of infection per 100,000 population]in decreasing incidence): Salmonella [15.89]; Campylobacter [12.97]; Shigella [5.53]; Cryptosporidium [3.31]; Shiga toxinproducing E. coli (STEC) non-O157 [1.64] and STEC O157 [.95].”

Although your observations about the testing and cited references appear appropriate, it is very nearsighted and clinically burdensome to limit coverage to 5 pathogens and will discourage adequate and thorough evaluations of the range of pathogens causing significant gastroenteritis. Normally-well individuals with severe symptoms deserve prompt accurate diagnosis since whether to treat with antibiotics, and if so with what choice of antibiotic, can be an urgent decision that can’t await weeks of symptomatic management, educated guesses if there is a better way, or shotgun broad spectrum antibiotics. Even now C difficile testing is advocated for community based acute diarrhea since it is increasingly prevalent and pathogen exposures and antibiotic history is often incomplete from our patients.

The high risk individual might have a much wider potential of infectious agents and will sometimes require antibiotics even where these might not be employed in the immunocompetent. In addition, if a more narrowed choice of antibiotic can be made, or if a pathogen is confirmed which does not benefit from antibiotic treatment, not only do we spare the patient cost and hassle but also risk, of which C difficile is of course now a major scourge.

Compared to paying claims for several separate CPT codes for standard bacterial cultures for the range of bacterial targets, a claim for C difficile toxin or other testing (PCR, GH or combined) and sometimes several O&P exams and giardia antigen exam, (I see all of this commonly ordered by primary physicians and by emergency department physicians), herea single test run with one specimen that can be readily gotten to a lab through UPS/FEDEX and can have a one hour turnaround (eg Biofire) is a great advance.

Our own group has discovered within the past 2 weeks examples of entameba, vibrio, variant E coli, concurrent E coli and C difficile and other pathogens, knowledge of which affected treatment choices.

I expect hospital emergency departments will utilize this test once adopted by hospital based labs and utilize it to triage those acutely ill with gastroenteritis, where the positive diagnosis of a viral syndrome versus C difficile versus one of the “5 bacterial targets” can make a major difference. Likewise, patients with 4-6 weeks of ongoing symptoms, where inflammatory bowel disease becomes a consideration, can now be much better ascertained in many cases to in fact have infectious cause, avoiding need for colonoscopy in some cases and mistaken therapies with mesalamine or corticosteroids.

To limit payment arbitrarily—I stress arbitrarily—to the lower tier of MolDx payment and not recognize the actual range of testing done is inappropriate or punitive to labs trying to service the real needs of clinicians for their Medicare beneficiaries.

I would further question this policy, considering that the ordering physician is NOT asking for 5 targets but is requesting the entire panel. The laboratory is just carrying out what the ordering physician requested. This differs from the situation where a pathologist might receive a pathologic specimen (88013) and make the decision about the extent of specialty stains or immunohistochemistry might be performed—here, it is the pathologist judgment and choice.

We urge then that the limitation of coverage to this technology NOT be implemented.

We agree with the commenter that the identification of some organisms can facilitate appropriate targeted treatment sooner. The draft LCD is intended to allow clinicians to test for those organisms. In addition, in recognition of the fact that Clostridium difficile is also an infectious organism, the identification of which has clear treatment implications, we are allowing testing for up to 11 organisms if Clostridium Difficile is one of them.

Multiplex Analysis of Gastrointestinal Infectious diseases

This analysis did not focus on Clostridium difficile colitis. Sadly we are moving to panel testing. I do not do this testing in absence of diarrhea. These infections are rare in my patients, but my patients may shed antigen for awhile. Viral testing of immunocompromised patient is warranted as it necessitated isolation precautions with norovirus and for rotavirus which does not always clear in my patients, oral immunoglobulin therapy is warranted. CMV colitis we make dx on invasive biopsy but if we move to stool testing cmv colitis in immunocompromised, immunosuppressed patient, the treatment for cmv colitis is IV ganciclovir and high dose IV immunoglobulin. Additionally we will often test for viral etiology of diarrhea before doing colonoscopy of thrombocytopenic immunocompromised stem cell transplant patient soas to rule out viral infection before doing invasive biopsy looking for graft versus host disease. C difficile is a problem for immunocompromised or immunosuppressed patients and frequent testing for this is warranted.

Thank you for the comment. We agree that testing for a wide range of gastrointestinal pathogens, including Clostridium difficile in immunocompromised patients is warranted. The specific details regarding covered CPT codes in the LCD does allow for this, and we are making the text of the LCD explicit to acknowledge that testing for Clostridium difficile may be reasonable and necessary and to acknowledge that testing for an expanded range of organisms may be reasonable and necessary in patients who are immunosuppressed.