In BCR-ABL1 Negative Myeloproliferative Neoplasms the Detection of JAK2exon12, MPLW515, CBL, KITD816V, FIP1L1-PDGFRA Mutations Are Closely Linked to Specific Entities, Whereas the JAK2V617F, TET2, or EZH2 Mutations Demonstrate a Broader Diversity: Patterns From Diagnostic Reports of 18,547 Patients Analyzed

Abstract 458

The first mutation detected in BCR-ABL1 negative myeloproliferative neoplasms (MPN) was JAK2V617F that revolutionized diagnostics of MPN during the last five years. However, although this genetic marker is useful to discriminate MPN from reactive disorders, it is not specific for one entity. In addition, approximately 5% of all polycythemia vera (PV) and 50% of essential thrombocytosis (ET) and primary myelofibrosis (PMF) are not JAK2V617F mutated. In these entities other activating mutations, e.g. MPLW515 mutations or JAK2exon12 mutations, cover additional small proportions of patients without JAK2V617F mutation. To further improve the molecular genetic characterization of MPN research focuses on the identification of novel mutations and, recently, CBL, TET2, and EZH2 genes were identified to be mutated in MPN. We here report on our single centre experience in applying these markers in a daily diagnostic work flow comprizing a total cohort of 18,547 cases with suspected MPN that were investigated between 8/2005 und 8/2010 with individual patient specific combinations of these markers as soon as published. Thus, the most frequently tested marker was JAK2V617F that was applied in 17,027 pts. In 6,622/17,027 (38.9%) a definite diagnosis of MPN could be made or confirmed on the basis of the detection of JAK2V617F mutation. More detailed, the percentage of JAK2V617F positive cases varied depending on the suspected diagnoses: In patients with cytomorphologically confirmed or suspected ET 581/891 (65.2%) were JAK2V617F positive, in PMF: 168/290 (57.9%), in PV: 800/942 (84.9%), in MPN-U: 51/212 (24.0%), in CMML: 38/383 (9.9%), in “MPN” not further specified by the referring physician: 4741/11249 (42.1%), and in those with unexplained leukocytosis/thrombocytosis/splenomegaly or suspected hematologic malignancy: 139/2492 (5.6%). Many of the before mentioned cases were suspected MPN and therefore analyzed for both JAK2V617F and BCR-ABL1. Thus, in 9,924 pts BCR-ABL1 and JAK2V617F testing were performed in parallel. As such, in 541/9,924 (5.5%) analyses BCR-ABL1 positive CML was identified and 3,558 cases were JAK2V617F mutated (35.9%). Only 8 pts were BCR-ABL1/JAK2V617F double positive (0.08%), thus this is a very rare event. In cases with JAK2V617F negative PV in a second step JAK2exon12 mutation was analyzed and 27/147 (18.3%) were tested positive. JAK2V617F negative ET or PMF were analyzed in a second step for MPLW515 mutations. In ET 24/258 (9.3%) and in PMF 14/164 (8.5%) cases were tested positive. JAK2exon12 or MPLW515 were never concomitantly detected with JAK2V617 in our cohort (parallel assessments: n=3,769). PCR for detection of FIP1L1-PDGFRA was performed in 1,086 cases with suspected HES/CEL or unclear eosinophilia but only 26 (2.4%) were tested positive and a CEL could be diagnosed. However, in 36/130 (27.7%) FIP1L1-PDGFRA negative cases a KITD816V mutation was detected and thus a diagnosis of mastocytosis could be established. In addition, confirmation of mastocytosis was achieved in further 326/731 (44.6%) pts with suspected mastocytosis, three of these pts had a JAK2V617F mutation in addition. Further analyses were recently done on selected well characterized cohorts of MPN: CBL mutations were analyzed in 623 cases and tested positive in 54 (8.7%): 26/199 CMML (13.0%), 1/25 PMF, 27/293 MPN-U (9.2%), but never were detected in ET (n=61) or PV (n=45). TET2 sequencing detected mutations in 56/191 (29.3%) of pts analyzed: ET: 6/28 (21.4%), PMF: 4/12 (33.3%), PV: 10/31 (32.3%), CMML: 17/22 (77.3%) cases, MPN-U: 17/86: (19.8%), HES: 1/9 cases, Mastocytosis: 1/3 cases. Thus, TET2 mutations are widely spread in different entities and were frequently associated with other mutations: JAK2V617F: n=16, JAK2exon12: n=1, MPLW515: n=2, CBL: n=5, FIP1L1-PDGFRA: n=1, KITD816V: n=1, and EZH2: n=2. Finally, EZH2 sequence analysis detected mutations in 4/68 (5.9%) cases (1/16 PV, 2/11 PMF, 1/17 MPN-U, 0/20 ET, 0/4 CEL). In conclusion, these data show that the analysis of molecular mutations greatly improved the diagnostic work up of MPN in the last 5 years. The detection of some mutations (JAK2exon12, MPLW515, CBL) are useful to further subclassify MPNs. Others (JAK2V617F, TET2, EZH2) are widely distributed and are helpful for classification and also to discriminate MPN from reactive disorders. The individual power of each marker for prognostication in MPN remains to be defined in future studies.