Genomic Sequence Analysis Panels in the Treatment of Hematolymphoid Diseases

Language quoted from Centers for Medicare and Medicaid Services (CMS), National Coverage Determinations (NCDs) and coverage provisions in interpretive manuals is italicized throughout the policy. NCDs and coverage provisions in interpretive manuals are not subject to the Local Coverage Determination (LCD) Review Process (42 CFR 405.860[b] and 42 CFR 426 [Subpart D]). In addition, an administrative law judge may not review an NCD. See Section 1869(f)(1)(A)(i) of the Social Security Act. 

Unless otherwise specified, italicized text represents quotation from one or more of the following CMS sources: 

Title XVIII of the Social Security Act (SSA):
Section 1862(a)(1)(A) excludes expenses incurred for items or services which are not reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member. 
Section 1833(e) prohibits Medicare payment for any claim which lacks the necessary information to process the claim. 
Section 1862(a)(7) excludes routine physical examinations, unless otherwise covered by statute. 

CMS Publications:
CMS Publication 100-02, Medicare Benefit Policy Manual, Chapter 15, Section 80.1 – Laboratory services must meet applicable requirements of CLIA 
CMS Publication 100-04, Medicare Claims Processing Manual, Chapter 16, Section 40.7 Billing for Noncovered Clinical Laboratory Tests Section and 120.1 Clarification of the Use of the Term “Screening” or “Screen” 

CMS Publication 100-04, Medicare Claims Processing Manual, Chapter 30, Section 50 Advance Beneficiary Notice of Noncoverage (ABN)
CMS Publication 100-08, Medicare Program Integrity Manual, Chapter 13, Local Coverage Determinations 

CMS National Correct Coding Initiative (NCCI) Policy Manual for Medicare Services, Chapter 10, Pathology/Laboratory Services, (A) Introduction

CMS Publication 100-02, Medicare Benefit Policy Manual, Chapter 15, Section 80.6. 5 which describes the Surgical/Cytopathology Exception.

CMS National Correct Coding Initiative (NCCI) Policy Manual for Medicare Services, Chapter 10 Pathology/Laboratory Services which addresses reflex testing. 

CMS Publication 100-03, Medicare National Coverage Determinations (NCD) Manual, Chapter 1, Part 3, Section 190.3 Cytogenetic Studies.

Code of Federal Regulations:
42 CFR, Section 410.32, indicates that diagnostic tests may only be ordered by the treating physician (or other treating practitioner acting within the scope of his or her license and Medicare requirements) who furnishes a consultation or treats a beneficiary for a specific medical problem and who uses the results in the management of the beneficiary's specific medical problem. Tests not ordered by the physician (or other qualified non-physician provider) who is treating the beneficiary are not reasonable and necessary (see Sec. 411.15(k)(1) of this chapter).
 
 

Acute Myelogenous Leukemia (AML)

Indications

Genomic Sequential Analysis Panel will be considered reasonable and necessary in the evaluation of blood or bone marrow samples in the following clinical circumstances:

• Genomic Sequential Analysis Panel will be considered reasonable and necessary in the evaluation of blood or bone marrow samples for newly diagnosed or relapsed/refractory AML patients who are candidates for treatment, regardless of karyotype findings. 
• Previously diagnosed patients with AML, who have not responded to induction chemotherapy, or who have progressed following induction. The patient must be a candidate for transplantation at the time of the testing.
• Patients with AML, who have responded to treatment, either chemotherapy or transplantation, with evidence of relapse.
   
  

Myelodysplastic Syndromes (MDS)

Indications

Genomic Sequential Analysis Panel will be considered reasonable and necessary in the evaluation of blood or bone marrow samples in the following clinical circumstances:

• Patients with clinical signs or symptoms of myelodysplastic syndromes (MDS) or myelodysplastic/myeloproliferative overlap syndromes (MDS/MPN), in whom clinical, laboratory, and pathologic assessment are nondiagnostic.
• Newly diagnosed MDS or MDS/MPN patients either
• stratified by the IPSS or IPSS-R as intermediate risk, or
• in MDS with ringed sideroblasts/RARS. 

 Limitations

• Repeat Genomic Sequential Analysis Panel testing is not reasonable and necessary in MDS after initial diagnosis and risk stratification.

Myeloproliferative Neoplasms (MPN)

Indications and Limitations of Coverage

Genomic Sequential Analysis Panel will be considered reasonable and necessary in the evaluation of blood or bone marrow samples in the following circumstances:

•  Diagnosis: Clinical signs or symptoms of myeloproliferative neoplasm (MPN) or myelodysplastic/myeloproliferative overlap syndromes (MDS/MPN) when
• clinical, laboratory, and pathologic assessment are nondiagnostic; and
• CML excluded (BCR-ABL1 negative)^1,2
• Risk Stratification: Newly diagnosed PMF not already classified as high-risk by Dynamic International Prognostic Scoring System (DIPSS) Plus^1,3,4
• Monitoring: Higher-risk MF (INT-1, INT-2, High-Risk) with progression on therapy¹

Acute Myelogenous Leukemia (AML)

Acute myeloid leukemia (AML) is a heterogeneous hematologic malignancy characterized by the clonal expansion of myeloid blasts, primarily in the peripheral blood and bone marrow. The American Cancer Society estimates that approximately 60,000 new cases of leukemia will be diagnosed in 2016, with one-third classified as acute myelogenous leukemia (AML). It accounts for the most annual deaths from leukemia in the United States. The median age of diagnosis is 67, with 54% diagnosed at 65 years or older (and approximately one third diagnosed at 75 years of age or older). Moreover, AML lies at one end of a spectrum of neoplastic myeloid diseases that includes myelodysplastic syndromes (MDS), which often progress to AML, and which are even more common in patients of advanced age, with an incidence of approximately 1/5000 patients over the age of 70.

AML is an aggressive disease that requires immediate diagnosis and treatment, with an average 5 yr survival rate of 28%, depending on a number of clinical and biologic variables, including acquired genetic alterations within the leukemic cells. Early treatment of AML generally consists of high-dose cytotoxic chemotherapy to induce remission, followed by consolidation (i.e., post-remission) chemotherapy and/or bone marrow transplantation.

Steadily accumulating genomic evidence shows that certain acquired genetic alterations within the leukemic cells are strong predictors of prognosis in AML and, accordingly, are essential factors in the decision whether a patient should undergo bone marrow transplantation (1-4). These alterations have been set aside as determinants of independent diagnostic categories in WHO AML guidelines, and as essential for AML management in NCCN guidelines (5,6).

Importantly, the indication for molecular biomarkers in AML is somewhat different from other cancers, such as non-small cell lung cancer, in that the markers themselves are often not the direct targets of treatment. In AML, these molecular genetic biomarkers are incorporated into a risk-based treatment stratification that determines whether or not to recommend transplantation.

Moreover, AML patients often have multiple combinations of these essential mutations, again in contrast to the mutually exclusive driver oncogene alterations seen in solid cancers such as non-small cell lung cancer. In AML, the clinical effect of driver mutations can be modified by the wider genomic milieu, either additively or interactively (7). Therefore, complete assessment of AML patients requires testing multiple biomarkers concurrently, rather than a sequential single-biomarker approach. In this regard, panel testing is becoming the preferred approach.

Clinical advisory committees (CACs) composed of pathologists, hematologists, oncologists, geneticists, and bioinformaticians from leading institutions worldwide have developed the International Consensus Classification (ICC) for myeloid neoplasms and acute leukemias. This classification system emphasizes that genomic sequencing panels provide valuable diagnostic and prognostic information even for patients with normal karyotypes or core binding factor cytogenetics. The ICC recognizes that many clinically significant genetic alterations are not detectable by conventional cytogenetic testing alone (9). For instance, mutations in genes such as NPM1, CEBPA, or SF3B1 can define specific disease entities or risk categories in the absence of cytogenetic abnormalities. The classification also highlights cases where genomic sequencing can reveal additional mutations that modify prognosis or treatment decisions in patients with known cytogenetic abnormalities. This consensus statement thus supports the use of genomic sequencing panels as a standard diagnostic tool for all patients with suspected myeloid neoplasms or acute leukemias, regardless of their cytogenetic profile, to ensure comprehensive genetic characterization and optimal clinical management (9).

The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours provides limited direct evidence. This expert consensus statement and narrative review, while not a systematic review or meta-analysis, emphasizes the increasing integration of genetic and molecular data in the diagnosis and classification of hematologic neoplasms. The document highlights a trend towards comprehensive genetic profiling, noting that most acute myeloid leukemia (AML) types with defining genetic abnormalities no longer require a 20% blast threshold for diagnosis. It also discusses the complex molecular landscapes revealed by next-generation sequencing in diseases such as diffuse large B-cell lymphoma (DLBCL), which exhibits approximately 150 recurrently mutated genetic drivers with a mean of 8% mutated per patient (10). The document underscores the importance of genetic analyses in various hematologic malignancies, citing examples such as the detection of PML::RARA fusion in over 90% of specific AML cases using FISH, and the essential role of TP53 mutation and IGHV region somatic hypermutation analyses in chronic lymphocytic leukemia (CLL) prognosis. However, the document does not provide quantitative data specifically addressing the utility or frequency of genomic sequencing panels in patients without normal karyotype or core binding factor cytogenetics. As an expert consensus statement, this document synthesizes current knowledge to establish a standardized classification system for hematologic malignancies, reflecting the evolving importance of genetic profiling in this field (10).

Myelodysplastic Syndromes (MDS)

The myelodysplastic syndromes (MDS) represent a spectrum of clonal bone marrow diseases, with heterogeneous presentations that typically include one or more cytopenias, defective differentiation of blood cell progenitors into mature functional cells, and an increased rate of progression to acute myeloid leukemia (AML). These secondary AML cases carry a worse prognosis than de novo AML cases. Furthermore, there are myeloid neoplasms that share overlapping characteristics with both MDS and myeloproliferative neoplasms (MPNs), such as chronic myelomonocytic leukemia (CMML). The World Health Organization (WHO), has designated these diseases separately as MDS/MPNs, distinct from either MDS or MPNs (1).

According to the 2016 National Comprehensive Cancer Network (NCCN) Guidelines, the overall incidence of MDS is approximately 5/100,000 per year, primarily in adults. MDS is rare in patients under the age of 40, but much more common in older patients, with incidence of 30/100,000 among ages 70-79, and 60/100,000 in patients 80 years and older (2).

MDS treatment can range from surveillance/observation to high dose chemotherapy and bone marrow transplantation, with the principal determining factors being the patient's overall health and co-morbidities, and prognostic categorization.

MDS has historically been classified by a combination of traditional laboratory techniques, such as demonstration of stable cytopenias by complete blood count, microscopic examination of a bone marrow biopsy, and bone marrow cytogenetic studies. Other than the clinical feature of the number of cytopenias and specific cytogenetic changes found recurrently in MDS, all other diagnostic criteria in MDS rely upon light microscopy findings. 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 ring sideroblasts. Low risk MDS is associated with dysplasia affecting only one cell lineage, with or without ring sideroblasts, and isolated large deletions involving chromosome 5 (5q-). High risk disease is associated with dysplasia across multiple lineages, increased blast percentages, and complex karyotype. With the exception of SF3B1 mutations (see below), no specific mutations are incorporated into the current diagnostic criteria of MDS.

However, evidence has steadily accumulated over the first two decades of the 21st century showing that certain specific acquired genetic alterations within the myeloid cells are strong predictors of prognosis in MDS and, in the case of SF3B1, are necessary diagnostic markers as well (3-9). In addition, other somatic alterations may support the diagnosis of MDS in certain contexts. MDS can be challenging to diagnose, due to the subjective morphologic assessment of dysplasia, and a multitude of benign reactive conditions that can manifest as peripheral cytopenias and cytologic atypia that are especially prevalent in the elderly population. In this regard, the demonstration of these clonal molecular alterations in clinically and/or morphologically ambiguous cases can help establish a diagnosis of MDS and expedite therapy earlier in the disease course, before progression to a more overt, and life-threatening, condition. Accordingly, a number of specific genetic alterations are included in the NCCN guideline recommendations as necessary for the diagnosis and management of patients with MDS.

Forty-seven different gene mutations have been identified as recurring findings in MDS, including TET2, SF3B1, ASXL1, DNMT3A, SRSF2, RUNX1, TP53, U2AF1, EZH2, ZRSR2, STAG2, CBL, NRAS, JAK2, SETBP1, IDH1, IDH2, and ETV6. While some are more common than others, no single gene has been reported in more than approximately one-third of cases. Most of these are useful as adjunctive diagnostic markers for clinically/microscopically ambiguous cases, to help establish a more firm diagnosis and, potentially, as markers of clonal disease that can be used to monitor for disease progression and response to interventions.

Myeloproliferative Neoplasms (MPN)

The myeloproliferative neoplasms (MPNs) represent a group of rare clonal bone marrow diseases, have a median age at onset of 65-70 years, and heterogeneous presentations that typically include overgrowth of one or more of the myeloid cell lineages in the marrow, with increased circulation of mature forms in the peripheral blood, and an increased rate of progression to acute myeloid leukemia (AML). Symptomatology varies between the different diseases, typically related to the specific proliferating cell lineages.

MPNs can be subdivided in two main categories based upon the presence or absence of BCR-ABL1: chronic myeloid leukemia (CML) and non-CML MPNs.² The main non-CML MPNs, also termed classical MPNs, including essential thrombocythemia (ET), primary myelofibrosis (PMF), and polycythemia vera (PV), had traditionally been defined by laboratory criteria such as thrombocytosis, marrow fibrosis, and erythrocytosis, respectively. Non-CML MPN treatment can range from observation to targeted therapies to hematopoietic stem cell transplantation (HCT), with the principal determining factors being the patient's overall health and co-morbidities, presence of fibrosis or increased blasts, molecular and prognostic categorization. Other less common diagnostic entities within the non-CML MPNs include chronic neutrophilic leukemia (CNL) and chronic eosinophilic leukemia (CEL). 

While the definitions of the entities within MPN are fairly distinct, in practice, phenotypic overlap is common, and definitive classification can be challenging based solely on clinical grounds and traditional laboratory tests, such as complete blood count (CBC) and bone marrow biopsy. At diagnosis, the discrimination from reactive conditions is often critical, and the demonstration of several specific clonal molecular alterations in clinically or morphologically ambiguous cases can expedite an MPN diagnosis before progression to a more overt, life-threatening, condition. Accordingly, one or more MPN-restricted, driver mutations are included in practice guideline recommendations as necessary for the diagnosis or management of patients with MPN or MPN-like conditions, including Janus kinase 2 (JAK2), calreticulin (CALR), and myeloproliferative leukemia virus (MPL).^5-7 Driver mutations are usually mutually exclusive.

JAK2 V617F mutations in exon 14 are the sine qua non of PV, reported in >90% of cases. Another 2-3% have missense mutations or small insertion/deletions in a secondary hotspot of JAK2, in exon 12.^8,9 The JAK2 V617F mutation is not specific within the MPNs, however, and has also been reported in approximately 40-60% of cases of ET and PMF. The JAK2 mutation is not present, however, in CML or in benign conditions that lead to secondary erythrocytosis and its presence is invaluable in this differential diagnosis.

Mutations in CALR exon 9, by contrast, are encountered in approximately 20-35% of patients with ET and PMF and are not characteristic of PV. Two alterations predominate: a 52 bp deletion (Type 1) that is more common in PMF or post-ET myelofibrosis, and a 5 bp insertion (Type 2) that is more common in ET, associated with more indolent disease and lower risk of thrombosis, despite an association with extremely high platelet counts. Beyond enabling a specific diagnosis, the presence of a CALR mutation portends a favorable prognosis in PMF (overall survival (OS) 17.7 years vs. 3.2 years with no mutations (“triple-negative” patients), and these patients have a lower rate of progression to AML over 10 years (9.4% vs. 34.4%).¹⁰ CALR mutation also is associated with improved overall survival (82% vs. 56% at 4 years) and non-relapse mortality (7% vs. 31%) following transplantation for PMF.¹ The significance of CALR mutations in ET is dependent upon the type of mutation, as mentioned above, and less clear.

MPL mutations are another sequence alteration seen infrequently in ET and PMF, and predicts an increased risk of transfusion dependence in PMF.¹⁰ MPL mutations, like JAK2 and CALR, portends a favorable prognosis across MPN, when compared to triple negative MPN (approximately 10%) that lack mutations in any of these three genes.¹

In contrast to the favorable prognosis conferred by MPL, CALR, and JAK2, a number of other not MPN-restricted, non-driver, mutations are considered "high risk" for progressive disease and are associated with both a shorter OS and leukemic-free survival in PMF, including ASXL1, TET2, EZH2, SRSF2, SF3B1, SH2B3, U2AF1, TP53, IDH1, and IDH2.^1,5,6 These co-mutated, myeloid genes additively contribute to phenotypic variability and shifts, as well as progression to more aggressive disease.⁶ Mutations in at least one of these genes confers a shortened median survival (81 vs. 148 months),¹¹ and the presence of 2 or more of these mutations reduces median survival from 12.2 years to 2.6 years.¹² However, a study of 570 patients with PMF demonstrated a counterbalancing effect of combining a low-risk mutation (CALR) with a high-risk mutation (ASXL1); the median OS was longest in CALR(+)/ASXL1(-) patients (10.4 years), shortest in CALR(-)/ASXL1(+) patients (2.3 years), and intermediate (5.8 years) when both are present or absent.³ In triple-negative patients, these non-driver mutations serve as a major diagnostic criteria for both PMF and ET. WHO and NCCN guidelines recommend analysis of mutations in these genes to assist in establishing an MPN diagnosis in the absence of mutations in MPL, CALR, and JAK2.^1,2

Moreover, patients with progressive MPNs are candidates for several different therapies, including hypomethylating agents, induction chemotherapy, HCT, and clinical trials. Selecting between these options is a complex determination that includes several factors, and NCCN recommends the analysis of mutations in ASXL1, EZH2, IDH1/2, SRSF2, and TP53 (category 2A) to assist in risk stratification and treatment planning in select PMF patients.¹ Monitoring response to treatment is recommended every 3-6 months with additional molecular testing reserved for MF patients with INT-1-risk or INT-2-risk/high-risk disease to aid in decision-making regarding allogeneic HCT.¹ 

Acute Myelogenous Leukemia (AML)
The spectrum of genetic abnormalities that are relevant in AML is broad, and includes specific sequence variants within genes, copy number changes, and structural variants such as chromosomal translocations. Smaller scale mutations require a molecular diagnostics method (e.g., sequencing) for analysis, while larger scale chromosomal abnormalities may be analyzed using either molecular diagnostics or cytogenetics (e.g., FISH, karyotype) methods. Molecular diagnostics and cytogenetic testing play a complementary role in helping refine prognosis, particularly in cytogenetically intermediate risk normal karyotype AML (NK-AML), or those with core binding factor where KIT mutations help refine the prognosis (6,8). The following molecular genetic biomarkers are considered necessary for diagnosis and management of AML.

Table 1 Biomarkers that require a molecular diagnostics method (either via panel or individually):

Table Note: Table 1 lists biomarkers that require molecular diagnostics for detection in patients with acute myeloid leukemia (AML). The alterations in these genes are associated with varying prognostic risk categories, ranging from favorable to poor, and are essential for determining prognosis and guiding therapy selection. The National Comprehensive Cancer Network (NCCN) assigns category 2A to biomarkers, indicating uniform NCCN consensus that the intervention is appropriate despite a low level of evidence. Mutations with an asterisk are recognized as a distinct diagnostic category with poor prognosis by the World Health Organization (WHO) 2016 AML Classification. As more genetic variants are demonstrated to lead to safe and effective therapy selection, meeting Medicare coverage guidelines, they may be added to this table.
 
Table 2 Biomarkers that can be assessed by either a molecular diagnostics method (panel only) or by a cytogenetics method:

Myelodysplastic Syndromes (MDS)

Five of these genes (TP53, EZH2, ETV6, RUNX1, ASXL1) have established independent value as prognostic determinants, with mutations in any one of these five genes indicating a poor risk status, worsened overall survival, and a prognosis equivalent to one step worse in the Revised International Prognostic Scoring System (IPSS-R), all classified under the NCCN 2A category. Similarly, mutations in several other genes including DNMT3A, U2AF1, SRSF2, CBL, PRPF8, SETBP1, and KRAS are also associated with poor risk and decreased survival, and they are also categorized under NCCN 2A.

Mutations in TET2 have been associated with an improved response rate to hypomethylating agents, yet do not impact response duration or overall survival, and thus, alterations in this gene are recommended in NCCN 2A as determinants for treatment with this class of agents. Newly categorized under the "Poor risk (AML-MR)" without specific NCCN guidelines (N/A) are the genes BCOR, IDH1, IDH2, ZRSR2, and STAG2, based on findings from the WHO 5th edition classification of haematolymphoid tumors.

SF3B1 mutations, which play a role in the diagnosis of MDS with ringed sideroblasts (previously known as refractory anemia with ringed sideroblasts, or RARS), now fall under the NCCN 2A category, marking them as low-risk but significant for diagnostic purposes. The revised WHO classification establishes this diagnosis with the presence of at least 15% ringed sideroblasts by light microscopy or at least 5% when an SF3B1 mutation is also present.

The spectrum of genetic abnormalities relevant in MDS remains broad, encompassing specific sequence variants within a significant number of genes, as well as a wide range of aberrations in others. Despite the feasibility of single gene assays, the practicality of such testing is limited by the multitude of hotspot variants and the number of relevant genes. Comprehensive coverage of the entire gene is often required, making MDS an appropriate indication for multiplexed sequencing, typically performed by next-generation sequencing. The genes listed, particularly those categorized under NCCN 2A and those newly included under the N/A category from the WHO 5th edition, are considered necessary biomarkers for the diagnosis and management of select MDS cases.

In total, the required nucleotide or small insertion/deletion variants occur at numerous specific loci across multiple genes. These variants are crucial for determining prognosis and guiding therapeutic decisions. The genes listed include both those associated with risks, such as poor prognosis or favorable responses to specific treatments, as well as diagnostic categories. As additional genetic variants are demonstrated to effectively guide therapy selection and thus meet Medicare coverage guidelines, more genes may be added to the table.

Myeloproliferative Neoplasms (MPN)

 The following molecular genetic biomarkers are considered necessary for diagnosis and management of select MPNs.  

In total, essential nucleotide or small insertion/deletion variants occur across several genes, each playing a crucial role in determining prognosis and guiding therapy in myeloproliferative neoplasms (MPNs). Key genes such as JAK2, CALR, MPL, and others including ASXL1, EZH2, IDH1/2, SRSF2, and TP53 are noted for their diagnostic and prognostic implications. Additional genes like TET2, DNMT3A, U2AF1, and more, although not assigned an NCCN category, are vital for diagnosing conditions such as primary myelofibrosis and assessing disease progression. As research advances, demonstrating that additional genetic variants safely and effectively guide therapy, these genes may also meet Medicare coverage guidelines, prompting further updates to the diagnostic table.

N/A

N/A

 Myeloproliferative Neoplasms (MPN)

1. Mesa R, Jamieson C, Bhatia R, et al. Myeloproliferative Neoplasms. NCCN Clinical Practice Guidelines in Oncology. 2017;Version 2.2018.
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(20):2391-2405.
3. Tefferi A, Guglielmelli P, Lasho TL, et al. CALR and ASXL1 mutations-based molecular prognostication in primary myelofibrosis: an international study of 570 patients. Leukemia. 2014;28(7):1494-1500.
4. Tefferi A. Management of primary myelofibrosis. UpToDate. 2017.
5. Vainchenker W, Kralovics R. Genetic basis and molecular pathophysiology of classical myeloproliferative neoplasms. Blood. 2017;129(6):667-679.
6. Rumi E, Cazzola M. Diagnosis, risk stratification, and response evaluation in classical myeloproliferative neoplasms. Blood. 2017;129(6):680-692.
7. Vannucchi AM, Harrison CN. Emerging treatments for classical myeloproliferative neoplasms. Blood. 2017;129(6):693-703.
8. Tefferi A, Lasho TL, Guglielmelli P, et al. Targeted deep sequencing in polycythemia vera and essential thrombocythemia. Blood Advances. 2016;1(1):21-30.
9. Passamonti F, Elena C, Schnittger S, et al. Molecular and clinical features of the myeloproliferative neoplasm associated with JAK2 exon 12 mutations. Blood. 2011;117(10):2813-2816.
10. Rumi E, Pietra D, Pascutto C, et al. Clinical effect of driver mutations of JAK2, CALR, or MPL in primary myelofibrosis. Blood. 2014;124(7):1062-1069.
11. Vannucchi AM, Lasho TL, Guglielmelli P, et al. Mutations and prognosis in primary myelofibrosis. Leukemia. 2013;27(9):1861-1869.
12. Guglielmelli P, Lasho TL, Rotunno G, et al. The number of prognostically detrimental mutations and prognosis in primary myelofibrosis: an international study of 797 patients. Leukemia. 2014;28(9):1804-1810.

• Provider Education/Guidance