Response to Comments: MolDX: MDS FISH

The Draft LCD provides coverage under the following indications and limitations: Indications FISH (fluorescent in situ hybridization) testing is indicated in the evaluation of patients whose bone marrow examination is suggestive of myelodysplasia (MDS) and who have had a failed or inadequate cytogenetic assessment (conventional karyotype).

Limitations:

• When the results of conventional cytogenetics are adequate, FISH testing is not reasonable and necessary and not a Medicare benefit.
• When conventional karyotyping is inadequate, Medicare will limit initial FISH testing to 4 probes (+8, -7 or del(7q), -5 or del(5q), and del(20q)).
• Reflex FISH testing may be indicated when the initial 4 probes are negative.
• FISH testing is not reasonable and necessary for the diagnosis of MDS, and is not a Medicare benefit.
• When a patient has a bone marrow suggestive of another disorder (e.g., a plasma cell disorder), MDS FISH is not indicated.
• FISH testing for MDS and a plasma cell disorder are not reasonable and necessary and not a Medicare benefit.
• Delay in diagnosis is not a legitimate reason for performing more than 4 initial FISH studies followed by step-wise reflex testing.
• Repeat FISH testing by another laboratory on the same specimen is not reasonable and necessary.

NeoGenomics Comments Regarding FISH Testing

NeoGenomics believes that FISH testing with probes (+8, -7 or del(7q), -5 or del(5q), del(20q), del(12p), i(17q), +19, and del(11q) is reasonable and necessary. These probes add significant prognostic and actionable treatment guiding information for treating physicians beyond the 4 probes proposed by Noridian and MolDX for initial FISH testing. In addition, abnormalities, particularly in TP53 (17p), ETV6 (12p), KMT2A (11q23) and MECOM (3q), can be cryptic and may be missed by conventional cytogenetics, but can be detected by FISH.^1,2,3 For these reasons, NeoGenomics urges Noridian and MolDX to offer coverage to beneficiaries for 8 FISH probes as opposed to allowing 4 probes (+8, -7 or del(7q), -5 or del(5q), and del(20q)) and reflex FISH testing when the initial 4 probes are negative.

In 2012, the Revised International Prognostic Scoring System for Myelodysplastic Syndromes (IPSS-R) was published by the American Society of Hematology.⁴ IPSS-R built upon the standard for assessing prognosis of primary untreated adult MDS patients, the International Prognostic Scoring System (IPSS) which was published in 1997. IPSS-R was derived from a much larger database than IPSS (n=7012 v. n=816) and, in their MDS Cytogenetic Scoring System, evaluated 8 cytogenetic abnormalities (+8, -7 or del(7q), -5 or del(5q), del(20q), del(12p), i(17q), +19, and del(11q)) as opposed to the four cytogenetic abnormalities evaluated in IPSS (+8, -7 or del(7q), -5 or del(5q), and del(20q)).

By evaluating these additional abnormalities IPSS-R provided greater prognostic capability than IPSS, stratifying MDS patients into 5 risk groups (Very good, median survival 5.4 years; Good, median survival 4.8 years; Intermediate, median survival 2.7 years; Poor, median survival 1.5 years; Very poor, median survival 0.7 years) versus 3 risk groups with IPSS (Good, median survival 3.8 years; Intermediate, median survival 2.4 years; Poor, median survival 0.8 years).^4,5 By stratifying patients into 5 categories versus 3, beneficiaries and their treating physicians are afforded greater prognostic resolution as to when and how to treat (or not treat).

As described in IPSS-R, del(11q) are found in 3% of MDS cases, are associated with very good prognosis and a median survival of 5.4 years. Deletions in (12p) are associated with a good prognosis and median survival of 4.8 years, while +19 or i(17q) is associated with intermediate prognosis and median survival of 2.7 years. A complex karyotype (3 abnormalities) is associated with poor prognosis and median survival of 1.5 years while a complex karyotype with greater than 3 abnormalities is associated with a very poor prognosis and median survival of 0.7 years.⁴ NeoGenomics urges Noridian and MolDX to provide coverage for these probes in addition to the four probes proposed in DL37602.

NeoGenomics believes that the ability to stratify risk as outlined in the internationally accepted IPSS-R represents a major advance in our understanding of MDS in relation to IPSS and Medicare beneficiaries should be covered for the 8 FISH probes outlined in the IPSS-R Cytogenetic Scoring System.

If Noridian and MolDX adopt the verbiage from the Indication section of the draft LCD, NeoGenomics also asks Noridian and MolDX to clarify the following verbiage found under limitations in the preceding section: “Reflex FISH testing may be indicated when the initial 4 probes are negative” and “followed by step-wise reflex testing”. What is meant by step-wise reflex testing? Is there a defined protocol for which probes may be ordered, and in what sequence? We believe additional clarity should be provided in the final LCD.

NeoGenomics Comments Regarding Molecular NGS Testing

Molecular NGS testing alone (for myeloid mutations) or in combination with FISH testing is not reasonable and necessary for the diagnosis of MDS, and is not a Medicare benefit. NeoGenomics disagrees with the conclusion that “Molecular NGS testing alone (for myeloid mutations) or in combination with FISH testing is not reasonable and necessary for the diagnosis of MDS, and is not a Medicare benefit.” We believe that there are instances where NGS testing for myeloid mutations is reasonable and necessary as such testing may add significant prognostic information to inform treatment decisions.

With regard to instances where NGS testing for myeloid mutations is reasonable and necessary, NeoGenomics cites NCCN Guidelines Version 2.2018 Myelodysplastic Syndromes, which states as part of initial evaluation:

Bone marrow or peripheral blood cells may be assayed for MDS-associated gene mutations. These can establish the presence of clonal hematopoiesis, which can help exclude benign causes of cytopenias in cases with non-diagnostic morphology, but do not establish a diagnosis of MDS in the absence of clinical diagnostic criteria. Certain gene mutations (negative prognostic factors: TP53, ASXL1, ETV6, RUNX1, and EZH2; positive prognostic factor: isolated SF3B1) can refine the prognosis of MDS in patients risk stratified by the IPSS or IPSS-R and may be helpful in patients predicted to have intermediate risk. Consider molecular testing for JAK2 mutations in MDS patients with thrombocytosis .

NCCN Guidelines 2.2018 list genes TET2, DNMT3A, ASXL1, EZH2, SF3B1, SRSF2, U2AF1, ZRSR2, TP53, STAG2, NRAS, CBL, JAK2, NF1, RUNX1, ETV6, IDH1, IDH2, SETBP1, PHF6, BCOR, STAT3, and PPM1D as being clinically significant in MDS and likely to indicate clonal hematopoiesis, but state that these must be interpreted in the appropriate clinical context (e.g., cytopenias, <20% bone marrow blasts, no other AML defining criteria).

NeoGenomics supports NCCN’s position on gene mutation detection (by NGS) and urges Noridian and MolDX to provide coverage for NGS testing in cases of non-diagnostic morphology or at the very least strike this bullet from the limitations section of the LCD.

NeoGenomics respectfully seeks clarification as to whether the statement that NGS testing “alone” is not reasonable and necessary implies that NGS testing will be covered if morphology and cytogenetics are also performed or that NGS testing will never be covered for myeloid mutations?

NeoGenomics Comments Regarding FISH Testing for MDS and Plasma Cell Disorders

NeoGenomics agrees that myeloma FISH should not be used as a screening test; however, we also believe that FISH testing is appropriate for establishing prognosis in patients with plasma cell neoplasms and guiding therapy. Ordering both MDS and myeloma FISH panels is not indicated routinely, but there may be individual patients with more than one disease for whom this testing might be indicated.

NeoGenomics respectfully seeks clarification regarding the limitation that states “FISH testing for MDS and a plasma cell disorder are not reasonable and necessary and not a Medicare benefit.”

• Does this statement mean that FISH testing for both MDS and plasma cell disorder in the same specimen are not covered?
• Is FISH testing for myeloma covered?

Section titled FISH Testing the Draft LCD lists advantages of FISH over standard cytogenetics as:

A major advantage of FISH is that it can be performed on non-dividing interphase cells, affording a rapid screen for specific chromosome rearrangements or numerical abnormalities associated with hematologic malignancies. Interphase analysis can be performed on bone marrow cell suspensions routinely used for conventional cytogenetics, paraffin-embedded tissue sections, or disaggregated cells from paraffin blocks, bone marrow, blood smears and touch-preparations of cells from lymph nodes or solid tumors.

• FISH testing can be performed on archived paraffin-embedded clot bone marrow clot sections,
• Results are available more quickly, and
• Sensitivity is superior

and continues:

However, cytogenetics is sufficiently sensitive to detect these abnormalities in most instances, such that FISH is rarely indicated.

Section titled MDS Testing Algorithm the Draft LCD states:

Many laboratories adhere to a MDS testing algorithm to determine the necessity for FISH testing. More than 20 metaphases and a resolved karyotype preclude FISH testing. Mayo Medical Laboratories (MML) specifies that “MDS FISH does not increase the detection of MDS if chromosome analysis is successful and >20 metaphases are analyzed.” ⁷ They specify that MDS FISH studies should be ordered at the discretion of the cytogeneticist if <20 metaphases are identified, if there is an unresolved karyotype, or if only 1 abnormal metaphases is indicated. MML also supports use of a FISH study with a specific probe but without chromosome analysis for follow-up of a bone marrow for a previously diagnosed MDS with a specific genetic anomaly. A number of studies support a MDS testing algorithm that a conventional karyotype is often all that is needed in the diagnostic process ^8,9,10,11 and that MDS FISH studies should only be performed when there are fewer than 20 metaphases available for analysis. The Mayo Clinic has used a diagnostic algorithm in its practice and it supports this approach. A recent published article by Mayo concludes “…supports this assumption and showed that MDS-FISH studies provide little additional value beyond conventional karyotype studies if that study is adequate (defined by at least 20 metaphases available for analysis).

The American Society of Clinical Pathology (ASCP) has endorsed this practice pattern in its practice recommendations in its “Choosing Wisely” program. ¹² The ASCP notes that the added value of MDS FISH on bone marrow is extremely low when a satisfactory karyotype is obtained (≥20 interpretable metaphases). They also note that MDS FISH can be performed post hoc in the event of an unsatisfactory karyotype.

NeoGenomics Comments Regarding FISH Testing and MDS Algorithm Testing

NeoGenomics agrees with Noridian and MolDX on the advantages of FISH. MDS FISH probes are used for diagnosis as well as prognosis. Some abnormalities are considered recurrent cytogenetic abnormalities and allow the diagnosis of MDS in the absence of morphologic criteria. As listed on page 104 table 6.03 of WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues, the finding of isolated trisomy 8 or del(20q) is not diagnostic of MDS but testing for additional FISH probes may allow one to make a definitive diagnosis of MDS. If Noridian and MolDX adopt the verbiage from the Indication section of the draft LCD, then NeoGenomics urges Noridian and MolDX to add to the final LCD, the verbiage (as described in the referenced Mayo Medical Laboratories (MML) MDS testing algorithm) that “FISH studies should be ordered at the discretion of the cytogeneticist if <20 metaphases are identified, if there is an unresolved karyotype, or if only 1 abnormal metaphases is indicated”.

Summary of NeoGenomics Comments

• Noridian and MolDX should cover 8 FISH probes (+8, -7 or del(7q), -5 or del(5q), del(20q), del(12p), i(17q), +19, and del(11q)) in instances where conventional karyotyping is not adequate as opposed to allowing 4 probes (+8, -7 or del(7q), -5 or del(5q), and del(20q)) and reflex FISH testing when the initial 4 probes are negative.
• If verbiage from the Indication section is adopted as written, Noridian and MolDX should add verbiage to LCD DL37602 stating that FISH studies may be ordered at the discretion of the cytogeneticist if <20 metaphases are identified, if there is an unresolved karyotype, or if only 1 abnormal metaphase is indicated.
• Noridian and MolDX should clarify the following verbiage found under limitations: “Reflex FISH testing may be indicated when the initial 4 probes are negative” and “followed by step-wise reflex testing” to clearly define the allowed testing protocol and FISH probes that will be covered under the proposed LCD.
• Noridian and MolDX should provide coverage for NGS testing in cases of non-diagnostic morphology or at the very least strike this bullet from the limitations section of the LCD.
• Noridian and MolDX should clarify the limitation that states “FISH testing for MDS and a plasma cell disorder are not reasonable and necessary and not a Medicare benefit.”

We appreciate the comment. Noridian understands that the draft policy originally limited coverage to an initial 4 probes, with possible additional testing if the initial 4 probes were negative. Rather than limiting the use of additional probes to cases in which the initial 4 probes were negative, we will modify the coverage to allow the use of additional probes where the diagnosis is still in question, which we would generally expect to be the case if the initial 4 probes are negative. Additionally, we will reiterate that we would expect FISH to be used uncommonly regardless of the number of probes.

As regards the matter of NGS testing for the diagnosis of myelodysplastic syndrome, we will remove this statement from the LCD. Medicare has published an NCD regarding NGS use.

Per the LCD, below are the Indications for FISH testing:

Indications: FISH (fluorescent in situ hybridization) testing is indicated in the evaluation of patients whose bone marrow examination are suggestive of myelodysplasia (MDS) and who have had a failed or inadequate cytogenetic assessment (conventional karyotype).

Per these indications, FISH testing is indicated in the evaluation of patients whose bone marrow examination are suggestive of myelodysplasia (MDS) and who have had a failed or inadequate cytogenetic assessment.

Per the FY 2018 ICD-10 Official Guidelines for Coding and Reporting, Section IV Diagnostic Coding and Reporting Guidelines for Outpatient Services, H. Uncertain diagnosis: Do not code diagnoses documented as "probable", "suspected," "questionable," "rule out," or "working diagnosis" or other similar terms indicating uncertainty. Rather, code the condition(s) to the highest degree of certainty for that encounter/visit, such as symptoms, signs, abnormal test results, or other reason for the visit.

For those times in which the patient's bone marrow exam are suggestive of MDS and whose cytogenetic assessment is inadequate, for coding purposes, the highest degree of certainty will be the symptoms, signs, abnormal test results, or other reason for the visit. Please consider adding the following ICD 10 codes for those situations in which the patient's bone marrow exam is suggestive of MDS and whose cytogenetic assessment is inadequate.

Thank you for the comment. While we understand that for many patients, one of the myelodysplastic syndrome diagnosis codes will not be usable, refractory anemia, which is currently a billable ICD-10 code, is virtually always present in myelodysplastic syndromes. As such, we do not believe that additional codes, which have less specificity, are necessary.

I think there are 2 types of MDS patients, those for which supportive care until death is proper therapy. These are older very debilitated patients with multiple comorbidities. These patients do not need rigorous testing on a bone marrow.

Younger patients for whom transplant is being considered are different. Then goal of testing is to determine who is high risk to proceed urgently to transplant, those who will need transplant when there is evidence of clonal evolution, hopefully detected just as FISH progression or molecular progression and then tracking patients post transplant for minimal residual disease for additional cells from the donor.

The problem is the coding of both categories of patients on claims will likely be the same. Since traditional Medicare does not have preauthorization, doing this as review of claims makes the proposed LCD problematic.

Additionally there is the patient with unexplained blood abnormalities where we are trying to decide between clonal vs nonclonal disorders. Therapies are radically different but these patients are often coded as MDS, even if that in the end is not what they have. We have only one shot to get these therapies right. Patients do not want multiple bone marrows and in rural states, these are not testing that is done in every community. Testing is not under control of local pathologist but more likely under control of clinician-hematologist and or hematologist-oncologist.

The other problem is patients with myeloma and lymphoma being considered for an autologous stem cell transplant. If we miss a therapy related MDS, these patients with autologous transplant almost always develop AML post transplant.

I recognize Noridian's desire to control costs. As I pay a lab bill for my college student son who is very healthy who someone ordered a PTH level and vitamin D level, I understand Noridian's concerns. However, restrictive LCD for FISH and molecular testing will create strong potential for beneficiaries to receive substandard and possibly incorrect care, potentially shortening their lives.

Thank you for the comment. The LCD is not intended to govern coding but rather to govern coverage of services.

LabCorp would appreciate clarification on a few areas surrounding implementation of the determination:

1. How will Palmetto identify that the tests submitted on the claim are MDS FISH studies? The group 1 CPT/HCPCS codes identified in the draft determination for billing FISH studies are non-specific to MDS. There are other clinical justifications for FISH testing. For example, a patient with undiagnosed AML might come to us with cytopenia codes. We find AML and perform appropriate FISH probes. Do we run the risk for denial based on the incoming ICD10 codes, or would FISH be approved because we made a diagnosis of AML?
2. How will Palmetto identify when a karyotype is inadequate? The laboratory performing the MDS FISH studies may not be the same laboratory that performs the cytogenetic studies. Therefore, the claim for the MDS FISH studies may not include sufficient information to identify whether or not conventional karyotyping was completed.
3. How are labs performing MDS FISH expected to triage orders directed specifically for MDS FISH by an ordering pathologist or clinician? As a provider of hematopathology services LabCorp has educated clinicians and pathologist on the limitation of MDS FISH for several years and have published our clinical pathways showing that we do not recommend using MDS FISH as an aide in providing an initial diagnosis of MDS. However, we are still obligated to perform those test ordered by the requesting physician, or at the very least notify and/or seek permission for the cancellation of a test order. Performing laboratories will carry an inordinate amount of the burden of implementing the determination.

LabCorp agrees with the determination that it is incorrect to bill CPT codes 88365-88377 for FISH for cytogenetic analysis.

1. The draft LCD identifies which CPT codes are to be used based on whether the study is ancillary to cytogenetic studies (uses codes 88271-88291) or adjunct to surgical pathology or cytopathology (use codes 88365-88377), and the final determination states that use of 88365-88377 is not appropriate for cytogenetic studies. LabCorp supports this opinion and we have implemented the following rationale into how we code our FISH testing based on 1) the specimen type and 2) the actual techniques implemented in the lab.
2. The CPT codes 88367-88377 are described as Morphometric analysis, in situ hybridization. After review of AMA materials, LabCorp has interpreted that this family of codes require a pathologist to perform the following activities:
   1. Determine the appropriate areas of tumor to evaluate from review of the H&E-stained slide.
   2. Examine positive control tissue known to contain cells that express the probe/probes set to verify that the stain is optimized. Also examine the negative control to check for nonspecific binding and false positive staining.
   3. Then examine the patient sample by fluorescence microscopy. Interpret the probe/probe set staining pattern and quantification of the signals, and determine their significance in their histologic and cellular locations.
3. The process by which FISH studies are performed on hematologic specimens like those submitted for MDS FISH, including bone marrow aspirate and peripheral blood, does not provide a pathologist the appropriate material to perform a morphometric analysis. The most common technique used in labs today is to lyse away the cell membrane and/or stain the nuclei with DAPI to create a clear contrast for reading the FISH probe signals. This practice eliminates the ability to identify particular cells of interest/non-interest on which to base the FISH signal scoring.
4. LabCorp bills using the CPT codes 88271-88291 for hematologic (blood and bone marrow) specimens where morphometric analysis of the cells/nuclei being evaluated is not possible given current techniques implemented in the lab.
5. LabCorp has determined that the CPT codes 88367-88377 should only be used for formalin fixed paraffin embedded specimens or potentially smears from a fine needle aspirate where morphometric analysis can be performed on the ISH slide or with correlation to an H&E to ensure scoring of the ISH signal is only in the neoplastic or other cells of interest.

The scope of the LCD should be limited to FISH testing in MDS.

1. The limitations section of the determination includes the following statement “Molecular NGS testing alone (for myeloid mutations) or in combination with FISH testing is not reasonable and necessary for the diagnosis of MDS, and is not a Medicare benefit”. The statement appears to provide a negative determination on the use of NGS in the setting of MDS.
2. LabCorp agrees that molecular NGS testing only has clinical utility when taken in the context of key clinical findings including presence of cytopenias and a complete morphologic review of a bone marrow biopsy.
3. After review of the cited evidence in the determination, LabCorp believes these guidelines lean toward the use of molecular NGS testing as an aide in providing an accurate diagnosis for MDS.
4. The current version of the NCCN guidelines for MDS while stressing the limitations of NGS studies alone, also state the following:
1. A Table of 23 MDS related genes is referred to as a list of “gene mutations likely to be somatic (acquired, not congenital) and disease-related and therefore presumptive evidence of MDS.
2. “Certain Gene mutations (negative prognostic factors: YP53, ASXL1, ETV6, RUNX1 and EZH2: positive prognostic factor: Isolated SF3B1) can refine the prognosis of MDS in patients risk stratified by the IPSS or IPSS-R and may be helpful in patients predicted to have intermediate risk”.
5. The determination makes reference to the IPSS and IPSS-R not requiring mutational status in their calculations. The IPSS was published in 1997 and the APSS-r in 2012. Both publication dates precede the availability of scalable cost efficient mutational screening. Since the publication of the IPSS and IPSS-R, several studies have shown the addition of mutational status and enhanced value of this schema. This was documented in the 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia where the authors stated, “The number and types of specific mutations are strongly associated with disease outcome in MDS, and the addition of mutation data improves the prognostic value of existing risk-stratification schemes in MDS.”
6. LabCorp believes that molecular NGS studies in MDS provide valuable diagnostic and prognostic information when combined with key clinical information and should be considered a Medicare benefit when used in combination with evaluation for cytopenias, a complete morphologic review of the bone marrow, and when the results are correlated with key risk factors used in either the IPSS or IPSS-R.

This LCD expresses MolDX's coverage policies not claims processing policies. Medicare administrative contractors may adjudicate claims in accordance with coverage criteria given in the LCD as well as can be achieved given the limited information available on a claim. However, LCDs are not restricted to specifying coverage requirements based on information that can be submitted on a claim.

Karyotype adequacy may be assessed on an individual basis in line with generally accepted principles. Factors that may be considered could include the number of metaphases, as described in the LCD.

Medicare administrative contractors do not govern how laboratories triage their services or how laboratories choose to implement compliance programs, though we are happy to speak with individual labs who are working to strengthen their own compliance.

As regards the matter of NGS testing for the diagnosis of myelodysplastic syndrome, we will remove this statement from the LCD. Medicare has published an NCD regarding NGS use in cancer.

We are submitting joint comments because at this time both of our organizations share the same perspective regarding this draft LCD. We appreciate the effort that has gone into the creation of this proposed LCD, and we offer the following recommendations for Palmetto's consideration.

FISH testing detects alterations in 5-10% of cases negative by conventional karyotyping that inform diagnosis, prognosis, and/or therapy. In cases that are positive by conventional karyotyping, FISH testing detects additional alterations that may impact clinical management above and beyond those that are detected by conventional karyotyping alone.

FISH testing is utilized to monitor the patient's clinical course over time and there are many alterations that are required to establish a diagnosis of MDS or acute myeloid leukemia above and beyond the 4 probes referenced in the draft LCD. For example, according to the 2016 WHO classification for hematologic disorders, in high grade MDS, detection of an aberration typically seen in acute myeloid leukemia would necessitate upgrading a diagnosis of MDS to acute myeloid leukemia.⁴ This has very significant diagnostic, prognostic, and treatment implications.

Patients who are treated for other malignancies including plasma cell disorders may subsequently develop myeloid disorders such as MDS. This is outlined in the WHO book.

As a result, the draft LCD is overly restrictive and will negatively impact patient access to medically necessary FISH testing.

Summary of Evidence 

1. dLCD statement, third paragraph: These include the number of cell lineages (i.e., platelets, red blood cells, white blood cells) affected by dysplasia, the percentage of immature "blast" cells, and the presence or absence of a characteristic pattern of iron deposition in immature red blood cells called ringed sideroblasts.

Recommendation: Change the language to, the dysplastic changes on one or more cell lineages of megakaryocytes, erythrocytes and granulocytes; increased myeloblasts; and/or presence of ringed sideroblasts.

2. dLCD statement, third paragraph: Low risk MDS is associated with dysplasia affecting only one cell lineage, with or without ringed sideroblasts, and isolated large deletions involving the short arm of chromosome 5 (5q-). High risk disease is associated with dysplasia across multiple lineages, increased blast percentages, and complex karyotype.

Recommendation: Change "large deletions" to "deletions", and change "short arm" to "long arm."

Cytogenetic Testing (Chromosome Analysis)

1. dLCD statement, first paragraph: The identification of a chromosomal abnormality strongly supports the diagnosis of MDS and has important prognostic implications.

Recommendation: Change the language to, "The identification of clonal cytogenetic abnormalities, except for +8, del(20q) and -Y, can serve as presumptive evidence of MDS."

2. dLCD statement, first paragraph: "In decreasing order of frequency, the most frequent chromosomal abnormalities associated with MDS are: +8, -7 or del(7q), -5 or del(5q), and del(20q)."

Recommendation: Change the chromosomal abnormalities to "-7 or del(7q), -5 or del(5q), +8 and del(20q)."

3. dLCD statement, second paragraph: "Depending on the application, detection of structural chromosome changes, resulting in a loss or gain of genetic material by these methods, is estimated to be limited to those of 4-6 mb (megabase) in size."

Recommendation: We recommend changing the language to, "Depending on the application, detection of structural chromosome changes, such as deletion or translocation, is limited to those of 5-10 mb (megabase) or above in size."

FISH Testing dLCD statement, first paragraph: "FISH testing is a method by which an assessment is made for the presence, absence, relative position and/or copy number of specific DNA segments by fluorescence microscopy."

Recommendation: We recommend changing the language to, "FISH testing is a method that can be used to detect gene location, copy number changes, and/or gene rearrangement by fluorescence microscopy."

Indications and Limitations of Coverage

1. dLCD statement: "FISH (fluorescent in situ hybridization) testing is indicated in the evaluation of patients whose bone marrow examination are suggestive of myelodysplasia (MDS) and who have had a failed or inadequate cytogenetic assessment (conventional karyotype)."

Recommendation: We recommend modifying the first sentence to read, "FISH testing is indicated in the evaluation of patients whose bone marrow examination is suggestive of a myeloid disorder such as MDS and/or when there is clinical suspicion of MDS by the treating oncologist."

Limitations

1. dLCD statement, bullet #6: "FISH testing for MDS and a plasma cell disorder are not reasonable and necessary and not a Medicare benefit."

Recommendation: We recommend this item be modified to state, "FISH testing should be limited to MDS or plasma cell disorder based on clinical and pathologic findings, but not both."

2. As the only CPT codes specifically mentioned in this dLCD are those within the "Cytogenetic Studies" category, it would appear that Palmetto's intent is that this particular dLCD should not address the "side issue" of the appropriateness of molecular mutation testing in the evaluation of MDS and we certainly agree with that intent. Had Palmetto indeed intended to address mutation testing for MDS in this particular dLCD (often done by next-generation sequencing), some "Molecular Pathology" CPT codes (such as 81450) would undoubtedly have been included-and they were not.

Recommendation: Given this universal intent to not extend this FISH-specific payment policy to other non-cytogenetic testing modalities, we strongly recommend the deletion of specific statements that may be misconstrued as a possible formal Medicare payment policy for mutation-based molecular testing in MDS.

In particular, we recommend that the “Limitations” section of the dLCD be revised to delete the statement:

“Molecular NGS testing alone (for myeloid mutations) or in combination with FISH testing is not reasonable and necessary for the diagnosis of MDS and is not a Medicare benefit.”

3. Medicare will only cover up to four 4 FISH studies (+8, -7 or del(7q), -5 or del(5q), and del(20q)) on initial evaluation. This decision is inconsistent with established guidelines. NCCN guidelines specify MDS-associated karyotype to include del(5q), del(20q), +8, or -7/del(7q). WHO guidelines specify an MDS diagnosis to include both 1) an abnormal karyotype +8, del(20q) with dysplasia or increased blast count or 2) a specific MDS-associated karyotype [eg, del(5q), -7/del(7q), isochromosome 17q or t(17p)]. The choice of FISH probes may be different based on the clinical background of the patient. For example, therapy-related MDS (t-MDS) comprises 10-20% of all MDS. Unlike primary MDS, +8 and del(20q) is very rarely present in t-MDS as a sole abnormality, so FISH for +8 and del(20q) has a very limited or no value in t-MDS, neither diagnostic or prognostic. On the other hand, TP53 deletion is a common finding in t-MDS (in 25-30% of t-MDS patients) and is also a very important prognostic marker.

Recommendation: We recommend the coverage of 4 FISH studies (+8, -7 or del(7q), -5 or del(5q), del(20q) or isochromosome 17q or t(17p)) on initial evaluation with the choice of probes based on the patient's clinical history.

Additional Recommendation to Proposed Policy

1. Although this dLCD does not mention the role of FISH-based testing in the post-diagnostic setting for monitoring responses to therapy, we recommend that such repeat testing of post-treatment follow-up samples be specifically covered (when standard cytogenetic testing is inadequate). The clinical utility of FISH-based testing for monitoring the "cytogenetic response" to therapy in MDS patients is specifically endorsed by NCCN guidelines³ and the MDS International Working Group.

ICD-10 Codes

The proposed policy lists 12 ICD-10 codes for MDS. We agree that a specific diagnosis and ICD-10 should be known for a patient with pathologically-confirmed MDS. However, the ICD10 codes listed in the policy do not accommodate MDS/MPN diseases such as CMML which demonstrate features of both MDS and MPN. In addition, given that the clinical presentation of MDS, and thus the clinical justification for a bone marrow biopsy, is quite often very non-specific (cytopenias of various lineages), FISH-based testing is often necessary (when conventional cytogenetic karyotyping is inadequate) to distinguish MDS from benign/toxicologic/immunologic causes of cytopenias. As the treatment of a patient with a confirmed MDS diagnosis is radically different from that of a patient with a non-malignant etiology of cytopenias, the ability to rule out an MDS diagnosis is of great clinical relevance. We therefore strongly recommend that this MDS-FISH dLCD specifically include additional ICD10 codes corresponding to clinical conditions (cytopenias) that mimic MDS - and would thus be the appropriate diagnostic code in a patient in whom the bone marrow biopsy does not confirm a specific MDS diagnosis. The use of an "MDS" ICD10 code in a patient who did meet MDS diagnostic criteria -  but did need bone marrow FISH to "rule out" MDS (after an inadequate conventional cytogenetic study) - would obviously be inappropriate.

We recommend inclusion of additional ICD-10 codes for MDS-mimic conditions that would fulfill criteria for this policy. These additional codes include, but may not be limited to, those listed below:

MPN/MDS (Myelodysplastic/myeloproliferative neoplasms) as defined by the WHO

1. National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology, Myeloproliferative Neoplasms (Version 2.2018). Available at https://www.nccn.org/professionals/physician_gls/pdf/mds.pdf Accessed: April 9, 2018.
2. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood. 2016 127:2391-2405; doi:10.1182/blood-2016-03-643544.
3. Cheson BD1, Greenberg PL, Bennett JM. Blood. 2006 Jul 15;108(2):419-25. DOI: 10.1182/blood-2005-10-4149).
4. Swerdlow SH, Campo E, Harris NL, et al, eds. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: IARC Press; 2017. World Health Organization Classification of Tumours; vol 2.
5. Makishima H, Rataul M, Gondek LP, et al. FISH and SNP-A karyotyping in myelodysplastic syndromes: improving cytogenetic detection of del(5q), monosomy 7, del(7q), trisomy 8 and del(20q).Leuk Res. 2010 Apr;34(4):447-53. doi: 10.1016/j.leukres.2009.08.023. Epub 2009 Sep 15. PMID: 19758696 [PubMed - indexed for MEDLINE]
6. Coleman JF, Theil KS, Tubbs RR, Cook JR. Diagnostic yield of bone marrow and peripheral blood FISH panel testing in clinically suspected myelodysplastic syndromes and/or acute myeloid leukemia: a prospective analysis of 433 cases.Am J Clin Pathol. 2011 Jun;135(6):915-20. doi: 10.1309/AJCPW10YBRMWSWYE. PMID: 21571964 [PubMed - indexed for MEDLINE]
7. Mukherjee S, Sathanoori M, Ma Z, Andreatta M, Addition of chromosomal microarray and next generation sequencing to FISH and classical cytogenetics enhances genomic profiling of myeloid malignancies. Cancer Genet. 2017 Oct;216-217:128-141. doi: 10.1016/j.cancergen.2017.07.010. Epub 2017 Aug 14. PMID: 29025587.
8. Arber DA, Orazi A, Hasserjian R, The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood. 2016 May 19;127(20):2391-405. doi: 10.1182/blood-2016-03-643544. Epub 2016 Apr 11. Review. PMID: 27069254.

Thank you for the comment. We are modifying the coverage criteria such that it will allow for the use of additional probes if the diagnosis remains in question following the first 4 probes, which we would expect to generally occur in cases in which the first 4 probes were negative.