SETBP1 and NRAS Mutations Are Frequent Events in Post-Myeloproliferative Neoplasm Acute Myeloid Leukemia (post-MPN AML) Lacking JAK-STAT Activating Mutations

Abstract Background: The BCR/ABL-negative myeloproliferative neoplasms (MPNs) include essential thrombocythemia (ET), polycythemia vera (PV), and myelofibrosis (MF) are characterized by JAK-STAT activating mutations (mutations in JAK2, CALR, and MPL) . Approximately 10% of MPN patients in chronic phase are “triple-negative” for mutations in any of the three genes. Leukemic transformation (LT) of chronic-phase MPNs carries a poor prognosis, and the biologic and genetic mechanisms underlying transformation are poorly understood. Further, although no standard therapeutic approach to LT exists, conventional anti-leukemic therapies such as induction chemotherapy and the use of hypomethylating agents have been utilized. However, there is a paucity of data comparing these approaches, as well as evaluating the impact of genomic alterations on treatment outcomes. We have sought to address these issues through detailed analysis of a well-annotated multi-center cohort of post-MPN AML patients. Methods: Next-generation sequencing was performed on 114 patients at the time of LT. Sequencing of the entire coding sequence of 585 cancer-associated genes on 84 samples collected from Memorial Sloan Kettering Cancer Center, Princess Margaret Cancer Centre, and the Myeloproliferative Diseases Research Consortium (MPD-RC) was performed. Mutational calls were made in comparison to curated matched normals. Tumor-only mutational data was obtained from an additional 30 patients sequenced using the Foundation One Heme platform. Results: Of the 114 patients analyzed, 66 (58%) had a JAK2 mutation, 8 (7%) had a MPL mutation, and 9 (8%) had a CALR mutation (Figure 1). 32 patients (28%) had TN disease. Notably, SETBP1 and NRAS mutations occurred exclusively in the TN cohort, each occurring at a frequency of 18.8% (6/32). TP53 mutations were identified in 18.8% (6/32) of TN cases and 24% (20/82) of cases with JAK-STAT mutations. ASXL1 mutations occurred frequently in both the TN cohort and the JAK-STAT mutant cohort [40% (13/32%) and 27% (22/82), respectively]. Splicing factor mutations (SF3B1, SRSF2, U2AF1, and ZRSR2) occurred in 43.8% (14/32) and 30.5% (25/82) of TN and JAKT-STAT mutant cohorts respectively; these were found to be highly prevalent in comparison to prior reports of 11% frequency in de novo AML. Further, FLT3 and NPM1 mutations, two of the most common mutations in de novo AML, occurred relatively infrequently (3.5% and Clinical factors were evaluable in 84 patients, of whom 22 had ET, 23 had PV, 19 had primary MF (PMF) prior to LT. An additional 16 patients had secondary MF or an unknown antecedent MPN defined by the presence of a JAK2 mutation. Evaluation of these four groups, without pairwise interactions, found a statistically significant one-year OS difference among them (p=0.041, Figure 2). Patients were treated with induction chemotherapy, single-agent hypomethylating agent, ruxolitinib and decitabine combination therapy on a phase I/II trial (NCT02076191), or best supportive care. A one-year OS difference was noted for patients who received anti-leukemic therapy compared to supportive care (p = 0.008, Figure 3), with improved outcomes for all three treatment modalities in comparison to supportive care alone. Overall survival did not differ between TN and JAK-STAT mutated patients (p=0.374). The presence of 3 or more mutations also did not impact OS (p=0.339). Data on remission status, allogeneic stem cell transplant status, and the impact of clinical factors on outcome on this cohort will be presented. Analysis of variant allele frequency of RAS mutant cases shows that in some cases these mutations are likely subclonal events (Figure 4). Further evaluation of clonal architecture will be presented. Conclusions: LT is characterized by a distinct mutational profile as compared with de novo AML. Further, within post-MPN AML, patients with TN disease appear to have distinct mutational events such as SETBP1 and NRAS mutations, which may have important implications for the biology of transformation. Finally, our data indicate similar OS in patients treated with induction chemotherapy and non-induction chemotherapy approaches, both of which were superior to supportive care alone. Further validation of theses observations is required in other data sets. Download : Download high-res image (311KB) Download : Download full-size image Disclosures Mascarenhas: Janssen: Research Funding; CTI Biopharma: Research Funding; Merck: Research Funding; Novartis: Other: DSMB member , Research Funding; Promedior: Research Funding; Incyte: Other: Clinical Trial Steering Committee , Research Funding. Gupta: Incyte: Consultancy, Research Funding; Novartis: Consultancy, Honoraria, Research Funding. Verstovsek: Seattle Genetics: Research Funding; Genentech: Research Funding; CTI BioPharma Corp: Research Funding; NS Pharma: Research Funding; Astrazeneca: Research Funding; Seattle Genetics: Research Funding; Blueprint Medicines Corp: Research Funding; Lilly Oncology: Research Funding; NS Pharma: Research Funding; Lilly Oncology: Research Funding; Bristol Myers Squibb: Research Funding; Astrazeneca: Research Funding; Roche: Research Funding; Celgene: Research Funding; Pfizer: Research Funding; Galena BioPharma: Research Funding; Blueprint Medicines Corp: Research Funding; CTI BioPharma Corp: Research Funding; Genentech: Research Funding; Gilead: Research Funding; Pfizer: Research Funding; Promedior: Research Funding; Galena BioPharma: Research Funding; Celgene: Research Funding; Gilead: Research Funding; Bristol Myers Squibb: Research Funding; Promedior: Research Funding; Incyte: Research Funding; Incyte: Research Funding; Roche: Research Funding. Levine: Roche: Research Funding; Roche: Research Funding; Qiagen: Equity Ownership; Celgene: Research Funding; Celgene: Research Funding; Qiagen: Equity Ownership.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.