Response to Comments: Molecular Pathology Procedures

 A coincident reconsideration request, outside the scope of items up for comment, had the following requests related to BCR/ABL testing in CML. 

1. The addition of "BCR/ABL is indicated in patients with suspected CML with either persistent, unexplained leukocytosis or thrombocytosis." 

2. Remove the once in a lifetime frequency limitation of BCR/ABL testing to monitor CML. 

3. Use of BCR/ABL testing for CML diagnosis before bone marrow biopsy.

1. We agree based on CML NCCN Guidelines (1).
2. We don’t agree that limitation exists. Current coverage allows for repeat testing in CML in monitoring treatment response. There is no specific lifetime limitation for these codes (81206-81208), just a general statement: "Individual has not previously received genetic testing for the disease/condition. (In general, diagnostic genetic testing for a disease should be performed once in a lifetime.) Exceptions include clinical scenarios whereby repeat testing of somatically-acquired mutations (for example, pre- and post- therapy) may be required to inform appropriate therapeutic decision-making."
3. The new diagnostic coverage does not specify test sequence.

A second coincident reconsideration request, outside the scope of items up for comment, had the following requests related to BRCA and MSI testing.

1. The addition of more generic wording for BRCA1 and BRCA2 testing to avoid the need to update each gene for each drug on a rolling basis: “BRCA1 and BRCA2 genetic testing is also required for clinical decision making involving several drugs in Medicare beneficiaries with ovarian cancer. BRCA testing is considered medically necessary when needed to determine if a Medicare covered therapy is a reasonable option given the patient's specific clinical presentation. Use of germline or somatic mutation testing should be guided by the drug labeling for the therapy being considered.”
2. Under Indications, change the language from “medically necessary for a beneficiary with a personal history” to “medically necessary for a beneficiary with a current diagnosis or personal history.”
3. Add the following wording to coverage for MSI testing under CPT code 81301: “MSI testing is also required by FDA for the clinical use of Keytruda (pembrolizumab) in a restricted population of patients. These are patients who have unresectable or metastatic solid tumors who have progressed following prior treatment and have no satisfactory alternative options. When Keytruda (pembrolizumab) is a potential clinically appropriate therapeutic choice, MSI testing is medically necessary in these patients.” The requestor also notes that because this is a wide-ranging population of advanced cancer patients, the optimal ICD-10 coding is broad, and suggests providers should be instructed to use “an ICD-10 appropriate for the tumor type and location.”

1. We agree with the concept but believe our current wording: "the results will be used to benefit the individual tested in terms of potential to guide therapeutic decision making" suffices.
2. We agree.
3. We agree. This is the first tissue-agnostic FDA drug approval (i.e., the first drug authorized for use based on a molecular biomarker rather than a traditional histopathologic diagnosis). While we have concerns with such broad coverage that doesn’t even allow for diagnosis specificity, since Medicare requires coverage of both on-label and compendia approved off-label drugs used in an anti-cancer chemotherapeutic regimen (IOM 100-2, Chapter 15, Section 50.4.5), we feel coverage of the companion diagnostic is equally required. We also agree with directing use of “an ICD-10 appropriate for the tumor type and location,” given ICD-10 specificity is impractical.

In a joint comment the College of American Pathologists (CAP) and the Association for Molecular Pathology (AMP) requested the addition of AML, MPN, and MPN/MDS diagnoses to CPT codes 81220 (IDH1) and 81121 (IDH2). “IDH2 mutations identify patients for treatment with enasidenib, a potent inhibitor of mutant IDH2 that promotes differentiation of myeloid cells in patients with advanced hematologic malignancies (2018 v1 NCCN AML guidelines, see page AML-12); IDH1/2 mutations are an independent predictor of poor prognosis in MPN acting as an essential factor in the decision whether a patient should undergo bone marrow transplant (2018 v2 NCCN myeloproliferative neoplasm guidelines). The proposed policy lists the 3 ICD-10 codes for MPN. We agree that an exact diagnosis and ICD-10 should be known for a patient previously classified as having MPN. However the ICD10 codes listed in the policy do not accommodate diseases such as AML for which current NCCN guidelines consider testing to medically necessary. In addition the ICD10 codes for diseases such as CMML which demonstrate features of both MPN and MDS have been omitted. In addition, we request addition of ICD-10 codes associated with the appropriate clinical criteria raising the suspicion of MPN and AML that trigger the oncologist’s request for this testing (e.g. D47.1 Chronic myeloproliferative disease).”

We agree based on AML and MPN NCCN Guidelines (2,3).

AMP-CAP requested the addition of AML, MDS, MPN, and MPN/MDS diagnoses to CPT code 81175 and 81176 (ASXL1). “This mutation is an independent predictor of poor prognosis in MDS, MPN and in AML (except in favorable risk subtypes) acting as an essential factor in the decision whether a patient should undergo bone marrow transplant (MDS 2018 v2 NCCN guidelines, see MDS-C, MPN 2018 v2 NCCN guidelines, see MPN-D, AML 2018 v1 NCCN guidelines, see AML-A). The proposed policy lists the 17 ICD-10 codes for MDS and 3 for MPN. We agree that an exact diagnosis and ICD-10 should be known for a patient previously classified as having MDS or MPN. However the ICD10 codes listed in the policy does not accommodate AML for which current NCCN guidelines consider testing to medically necessary. In addition the ICD10 codes for diseases such as CMML which demonstrate features of both MPN and MDS have been omitted. In addition, we request addition of ICD-10 codes associated with the appropriate clinical criteria raising the suspicion of MPN, MDS or AML that trigger the oncologist’s request for this testing (e.g. D70.8 other neutropenia).”

We agree based on AML, MDS, and MPN NCCN Guidelines (2-4).

AMP-CAP requested the addition of AML and MDS diagnoses to CPT code 81334 (RUNX1). “This mutation is an independent predictor of prognosis in both AML and MDS and is an essential factor in the decision whether a patient should undergo bone marrow transplant (2018 AML and MDS NCCN guidelines, Arber DA et 2016). The proposed policy lists the 17 ICD-10 codes for MDS. We agree that an exact diagnosis and ICD-10 should be known for a patient previously classified as having MDS. However the ICD10 codes listed in the policy do not accommodate AML for which current NCCN guidelines consider testing to be medically necessary. In addition the ICD10 codes for diseases such as CMML which demonstrate features of both MPN and MDS have been omitted. In addition, we request addition of ICD-10 codes associated with the appropriate clinical criteria raising the suspicion of MDS and AML that trigger the oncologist’s request for this testing (e.g. D70.8 other neutropenia).”

We agree based on AML and MDS NCCN Guidelines (2,4).

AMP-CAP requested the addition of glioma diagnoses to CPT codes 81220 (IDH1) and 81121 (IDH2). “NCCN central nervous system guidelines make the recommendation: IDH mutation testing is required for the workup of glioma (2018 v1 NCCN central nervous system cancer guidelines, see BRAIN-F (7 of 9)). “IDH mutations define WHO grade II and III astrocytomas and oligodendrogliomas, and the secondary grade IV glioblastomas into which astrocytomas often evolve. Their presence distinguishes lower-grade gliomas from primary glioblastomas, which are IDH wild type. Detection of these mutations in a specimen that is otherwise equivocal for tumor may also be regarded as evidence that diffusely infiltrative glioma is present”

We agree that IDH1 and 2 have clinical utility in the classification and treatment of glioma in conjunction with MGMT gene testing, and as such will be adding the requested codes (5).

AMP-CAP requested the coverage of CPT codes 81105-81112 as genotyping for human platelet antigens is important for identifying woman at risk for neonatal alloimmune thrombocytopenia (NAIT). Post-transfusion purpura is an immune reaction against human platelet antigens, often occurring when a woman is sensitized during pregnancy, then subsequently receives a transfusion. We recognize that there will be very few Medicare beneficiaries for whom this testing will be clinically actionable, but still urge you to ensure these beneficiaries have access to this critical testing.”

Upon review, as stated we feel that there are so few Medicare beneficiaries that would both be pregnant and at risk for neonatal alloimmune thrombocytopenia, that at this time these codes will remain noncovered realizing that there is an appropriate possibility of coverage on appeal.

AMP-CAP requested the addition of Marginal Zone Lymphoma (MZL) diagnoses to CPT 81479 (MYD88): “The proposed policy lists the 10 ICD-10 codes for Lymphoplasmacytic lymphoma (LPL) and one code for Waldenstrom’s Macroglobulinemia. We agree that an exact diagnosis and ICD-10 should be known for a patient previously classified as having WM or LPL. However the ICD10 codes listed in the policy do not accommodate the differential diagnosis of Marginal Zone Lymphoma (MZL) versus WM/LPL. NCCN guidelines dictate using MYD88 testing to categorize WM/LPL or MZL because these lymphomas have overlapping clinical and pathological features . In addition we recommend adding ICD10 codes for diffuse large B-cell cell lymphoma (DLBCL) since an MYD88 mutation may be required for ibrutinib sensitivity.”

We agree with addition of the MZL diagnoses based on the latest B-Cell Lymphomas NCCN Guidelines (6). We couldn’t find the same consensus for MYD88 testing for ibrutinib sensitivity and no source was cited in the comment.

AMP-CAP requested the addition of diagnoses to CPT code 81335 (TPMT). “The proposed policy lists 2 ICD-10 codes for acute lymphoblastic leukemia. However these patients routinely receive a bone marrow transplant and may be treated with 6-MP therapy in relapse. Consequently we request consideration of ICD10 codes associated with bone marrow transplant. In addition 6-MP is used to control autoimmune diseases such as Crohn’s disease and ulcerative colitis.”

We agree with the additional ALL related diagnoses based on the latest ALL NCCN Guidelines (7), and the autoimmune diagnoses (8), both in the context of 6-MP treatment.

Implementing Genomics In Practice (IGNITE): “We disagree with the conclusions in the LCD as the evidence supports that genotyping for IFNL3 (IL28B) variation (rs12979860) is the strongest baseline predictor of response to PEG-interferon-alpha-containing regimens in HCV genotype 1 patients. Patients with the favorable response genotype (rs12979860 CC) have increased likelihood of response (higher sustained viral response rate) to PEG-interferon-alpha-containing regimens as compared to patients with unfavorable response genotype (rs12979860 CT or TT). We do acknowledge that newer treatment regimens are replacing PEG-interferon therapies. Given the level of evidence identified in the CPIC evaluation and the guideline recommendations for modification of treatment based on the IFNL3 (IL28B) status, we believe the criterion that a test result have an impact on the patient’s management (i.e., clinical utility) has been met.”

NGS disagrees. Per UpToDate: “Several clinical features that were predictors of response to interferon-based regimens are no longer relevant to combination direct-acting antiviral (DAA) regimens…Polymorphisms in the IL28B gene, which encodes interferon lambda 3, effectively predicted responses to treatment with interferon-based therapies and accounted for a significant proportion of the differential response observed in patients of certain races, such as patients of African descent. In contrast, neither non-CC IL28B genotype nor race has consistently been associated with lower sustained virologic response (SVR) rates in multiple trials and cohort studies of contemporary DAA combination regimens. Although some studies have suggested a limited impact of IL28 genotype or race on SVR rates with DAA regimens, the magnitude of the impact is small when appropriate regimens are used and not sufficient enough to recommend IL28B genotype testing in routine clinical practice (9).” In addition, there is no evidence that when the IL28 testing is negative, that clinicians still do not use the PEG-interferon-alpha-containing regimens despite the unfavorable response genotype.

Implementing Genomics In Practice (IGNITE): “The WHO recommends testing of drugs to predict for risk of hemolysis in G6PD deficient individuals if the drugs are to be prescribed in areas of high prevalence of G6PD deficiency. Given the level of evidence identified in the CPIC evaluation and the WHO and FDA recommendations for modification of treatment based on the G6PD status, we believe the criterion that a test result have an impact on the patient’s management (i.e., clinical utility) has been met.”

NGS disagrees. While initial and even confirmatory testing for G6PD deficiency when certain high-risk drugs are used is appropriate (CPT 82960 (Glucose-6-phosphate dehydrogenase (g6pd); screen) and CPT 82955 (Glucose-6-phosphate dehydrogenase (g6pd) quantitative), the use of molecular/genetic/DNA methods is not established. General screening, not to be confused with testing immediately before prescription of high-risk drugs, is not a Medicare benefit.

Regarding 81346 (TYMS), IGNITE commented that “Thymidylate synthase (TYMS) catalyses the methylation of dUMP to dTMP. As the sole de novo source of thymidylate in the cell, it is an important target for drugs such as 5-fluorouracil and methotrexate. Since there is not strong evidence to support clinical utility, we agree that TYMS testing should not be covered at this time.”

We agree and appreciate a comment not only where there is disagreement.

We received a comments requesting coverage for tests performed outside of NGS’s jurisdiction: 81521 (Mammaprint), 81541 (Prolaris), and 81551 (Confirm MDX).

There is a claims processing rationale to assume that claims for Laboratory Developed Tests (LDTs) not performed in NGS’s jurisdiction are unlikely to be correctly payable by NGS. The lab jurisdiction administrative process, rather than a clinical review of the evidence, is the reason for denial.  We will segregate these codes into a distinct list within the LCD or in an attached article to make this clearer.

We received one comment requesting coverage of CPT 81450 (CancerTYPE ID).

This service was not open to comment, and in any event, is an LDT not performed in NGS’s jurisdiction (see response to #13).

References