To: File: Fecal Antigen Assay for Diagnosis of Helicobactor Pylori (H. pylori) CAG-00006
From: Grant P. Bagley, MD, JD
Director, Coverage and Analysis Group
Mitchell I. Burken, M.D.
Medical Officer
Katherine K. Tillman, R.N., M.A.
Health Insurance Specialist
Svati B. Patel
Health Insurance Specialist
Subject: National Coverage Policy Request
Date: October 18, 1999
This memorandum serves four purposes: (1) describe and provide history of laboratory diagnosis for the ulcer-related bacteria H. pylori; (2) outline Medicare's past and current approach to coverage; (3) analyze relevant clinical literature; and (4) delineate reasons for retaining Medicare's current policy of contractor discretion.
Description and Background of H. pylori Diagnosis in the Clinical Laboratory
Until recently, the etiology of peptic ulcer disease (PUD) had been largely idiopathic, but the elucidation of H. pylori, a urease-producing bacterial species, which has a relatively high prevalence among adults, has had extensive epidemiological evidence to support its association with PUD, along with other disorders such as gastric cancer. The anti-microbial eradication of H. pylori has been demonstrated to serve as a cure for PUD and therefore, the laboratory diagnosis of this organism is a critical component of both primary diagnosis and monitoring PUD treatment.
Multiple methods have been developed to detect H. pylori due to fastidious growth requirements that make traditional culture methods impractical for routine diagnosis. Whereas biopsy material can be utilized (via histology and/or rapid urease testing), non-invasive methods have become alternatives in those clinical presentations where the patient's differential diagnosis does not warrant more extensive (i.e., endoscopic) evaluation. Serum serology (usually qualitative IgG assay) has been the traditional immunologic approach for primary diagnosis, but newer techniques, such as urea breath testing (UBT) have found roles in both initial diagnosis and post-therapeutic surveillance.
Emerging markers, such as novel immunologic formats and genotypic/amplification approaches, constitute the frontier of H. pylori diagnosis. One such immunologic format is the use of stool antigen as the target, thus creating opportunities for patients to avoid the discomforts and/or inconvenience of phlebotomy and breath sample acquisition. The current request by Meridian Diagnostics, Inc. for its Premier Platinum H. pylori Stool Assay (HpSA), approved by the Food & Drug Administration (FDA) for both primary diagnosis and evaluating eradication, is addressed in the ensuing sections.
The HpSA test utilizes a polyclonal anti-H. pylori captured antibody adsorbed to microwells. Diluted patient samples and a peroxidase conjugated polyclonal antibody are added to the wells and incubated for one hour at room temperature. A wash is performed to remove unbound material. The substrate is added and incubated for ten minutes at room temperature. A color then develops in the presence of bound enzyme. Finally, a stop solution is added and the results are interpreted visually or spectrophotometrically.
History of Medicare's Helicobactor Pylori Testing Policy
Studies have demonstrated that with the diagnosis and treatment of H. pylori infection the recurrence rates in PUD are significantly lower. Two types of diagnostic breath testing have received FDA approval. One test utilizes an ingested non-radioactive carbon 13 labeled urea. A breath sample is collected 30 minutes later and sent to an approved laboratory for testing. The second method involves administration of radioactive C-14 followed by testing for C-14 levels in a breath sample. This test can be done in any facility capable of doing radionuclide procedures. Serological tests detect a specific anti-H. pylori immune response, mostly via IgG antibodies in the patient's serum. Both breath tests and the serological tests have recently served as non-invasive alternatives to endoscopy.
In February 1997, the Technology Advisory Committee (TAC) agreed that the urea breath test measured the presence of H. pylori bacteria with high sensitivity and specificity. The necessity of such tests was brought into question. The FDA-approved label provided that physicians use the H. pylori breath test for initial diagnosis. Committee members concluded, however, that the test "would add little to an initial diagnosis as physicians would still perform an endoscopic procedure, but that the breath test had potential value as a monitor for chronic symptomatic conditions." The reimbursement for such testing was left to contractor discretion.
On January 14, 1998 the Health Care Financing Administration (HCFA) sponsored a conference on PUD. Participants included representatives from Peer Review Organizations (PROs), clinical advisors, and representatives from FDA and Centers for Disease Control (CDC). Dialogue centered on etiology, detection, diagnostic testing and treatment of PUD in the Medicare population. Strategies surrounding invasive versus non-invasive testing were discussed. Common among participants was the impression that consumers had a limited awareness of infection as a cause of ulcers and the concern that providers were hesitant to diagnose and treat H. pylori infection.
In July 1999, a formal request for coverage was received from Meridian Diagnostics, Inc. for its Premier Platinum HpSA enzyme immunoassay. Test results for the HpSA are intended to aid in the diagnosis of H. pylori infection, and to monitor response during and after therapy in adult patients. The assay gained FDA approval in December 1998. It was assigned the CPT code 87338, which will become active in January 2000.
Analysis of the Medical Evidence
Both the package of materials submitted by the requestor, in tandem with MEDLINE-generated recent peer-reviewed publications, demonstrate that HpSA has a potential role in the day-to-day management of PUD patients with suspected and/or documented H. pylori infection. In particular, the following three recent studies demonstrate comparable HpSA effectiveness with respect to other current diagnostic tests. No other studies (in addition to the ones described below) have reported comparative inferiority to currently available assays:
Makristathis et al. (1998) used a patient sample of 100 individuals (of whom 63 were known to be H. pylori-positive) in order to realistically reflect actual adult prevalence rates. With respect to both primary diagnosis and monitoring eradication, HpSA had similar sensitivity and specificity to polymerase chain reaction (PCR) testing, which is a powerful molecular diagnostic tool. The only concern with this particular study is the use of different independent gold standards for positive H. pylori status (using histology and culture) vs. negative status (using urea breath test and serology).
Trevisani et al. (1999) drew upon a larger sample of 270 adults (mean age = 56) who had been referred for esophagogastroduodenoscopy. 154 patients were being evaluated for primary diagnosis and the remaining 116 for monitoring eradication. Using a uniform gold standard (histology + rapid urease testing), performance parameters were reported to be at least 90%, except for a relatively lower specificity (82%) of HpSA for monitoring eradication. Please note that follow-up testing appeared to confirm that there were legitimate numbers of false-positive results, contributing to a lower specificity.
Vaira et al. (1999) was a multi-center European diagnostic trial, using a sample of 501 adults (mean age = 52) from 11 centers, undergoing gastroscopy, 107 of whom were available for monitoring eradication. Using a histology + urease or culture only as the gold standard, HpSA had equivalent sensitivity/specificity to urea breath test (UBT) for both primary diagnosis and monitoring eradication.
Several references have already illustrated the comparable sensitivity, specificity and predictive values of other (non-HpSA) assays (Al Assi [1999], Cutler [1995], Fennerty [1998], Megraud [1997], Rollan [1997], Sternberg [1997], Thijs [1996] and Wilcox [1996]), thus inviting HCFA to surpass the scope of such "performance-oriented" studies and incorporate the concept of clinical utility. Non-invasive H. pylori testing has a well-circumscribed role (CDC [1997]) in the evaluation of patients who:
Present with signs and symptoms most suggestive for gastric/duodenal ulcers;
Follow resection of early gastric cancer; and
Have low-grade gastric MALT lymphoma.
However, there are other pertinent questions for which consensus in the clinical community would be most helpful:
What is the role of H. pylori testing and treatment in the context of new-onset dyspeptic presentations (not due to non-steroidal anti-inflammatory drugs), in which there may be a wider differential diagnosis than PUD (Soll [1996])?
Based upon the above question, what is the ability of non-invasive testing to defer the need for more invasive testing, and in what types of patient subgroups would such a technology substitution question be most relevant?
Might extensive detection/eradication of H. pylori be associated with longer-term adverse shifts in the overall microbial balance of the gastrointestinal tract, which might offset shorter-term therapeutic benefits (Blaser [1999])?
Thus, in the absence of current resolution of these issues, it appears premature to reach a conclusion with regards to clinical utility.
Retaining Current Contractor Discretion
Given FDA-approval, coupled with documented effectiveness and comparability in peer-reviewed literature, there is not a basis for a national non-coverage policy. Therefore, it should be deemed "reasonable and necessary" to enable providers to both diagnose and monitor eradication of H. pylori for the indications (CDC [1997]) mentioned above.
However, there are two critical factors that counterbalance the benefit of a national coverage policy specifically configured for HpSA:
Such a "snapshot" national coverage policy would fail to account for the dynamic changes which are occurring in the development of non-invasive H. pylori diagnostics (including genotypic approaches currently being researched - Megraud [1997]). HCFA would be better served to defer consideration of such an expansive policy until greater maturation is achieved across multiple assay formats.
The previously described need for enhanced clarity regarding the clinical utility of H. pylori testing would detract from making a national coverage policy.
Continued contractor discretion offers the most optimal solution for addressing the current coverage request from Meridian Diagnostics, Inc. Encouraging technology diffusion at a local level, driven by clinical expertise at contractor advisory committees, will create an opportunity for stool antigen testing to obtain an appropriate diagnostic foothold. As any new markers emerge, there will be a continuous opportunity for Medicare contractors to compare HpSA against both existing and new techniques, such that a blend of performance and clinical utility considerations can result in more informed policy-making. The flexibility of the national coverage process allows for the introduction of future scientific and clinical knowledge which can be properly harnessed at the appropriate time when a national policy on the spectrum of H. pylori diagnostics becomes applicable.
