Cytokine receptor and JAK/STAT pathway mutations in acute myeloid leukemia: Prevalence and clinical impact.
Key claim: Cytokine receptor and JAK/STAT pathway mutations are uncommon in AML (~10% combined), but blast-phase MPN carries the worst prognosis (13% response rate), while de novo AML harboring JAK2/CALR/MPL mutations shows intermediate outcomes driven largely by co-occurring high-risk features rather than the mutations themselves.
Abstract
BACKGROUND: Mutations in cytokine receptor and JAK/STAT; Cy-JAK/STAT) signaling genes drive myeloproliferative neoplasms (MPNs) but remain incompletely characterized in acute myeloid leukemia (AML). The authors evaluated the prevalence, clinical presentation, and prognostic significance of Cy-JAK/STAT pathway mutations in patients with AML and compared outcomes across disease ontogeny. METHODS: In total, 971 adults with newly diagnosed AML who were treated at the Cleveland Clinic between January 2015 and September 2023 were retrospectively analyzed. Baseline next-generation sequencing was queried for eight Cy-JAK/STAT pathway genes: CALR, JAK1, JAK2, JAK3, MPL, SH2B3, STAT3, and STAT5B. Outcomes included composite-complete remission (complete remission or complete remission with incomplete hematologic recovery), event-free survival, and overall survival. In a prespecified subgroup of patients who had baseline molecular testing results available (n = 621), those who had blast-phase MPN (n = 68), de novo AML with JAK2/CALR/MPL mutations (n = 23), and de novo AML without Cy-JAK/STAT mutations (n = 530) were compared. RESULTS: Mutation frequencies for the eight genes were as follows: JAK2, 43 of 625 patients (6.9%); CALR, nine of 623 (1.4%); SH2B3, eight of 392 (2.0%); JAK3, five of 439 (1.1%); MPL, one of 639 (0.2%); and STAT5B, one of 390 (0.3%); no JAK1 or STAT3 mutations were detected. The median follow-up among survivors was 40.9 months, and the median overall survival was 12 months. JAK2-mutated AML was associated with higher cardiovascular comorbidity, lower complete remission/complete remission with incomplete hematologic recovery rates (32% vs. 54%), and inferior overall survival (p < .05), but not on multivariable analysis. In subgroup analyses, patients who had blast-phase MPN had the lowest response rate (13%) and the poorest survival; whereas patients who had de novo AML with JAK2, CALR, or MPL mutations had intermediate outcomes versus those who had mutation-negative, de novo AML (41% vs. 57% complete remission). CONCLUSIONS: Cy-JAK/STAT pathway mutations are uncommon in AML, but each mutation has distinct biologic and clinical characteristics. The worse survival observed in patients with JAK2-mutated AML largely appeared to be driven by the co-occurrence of high-risk characteristics. Blast-phase MPN was associated with a dismal prognosis.
Why this matters
This large-cohort observational study quantifies just how dismal blast-phase MPN outcomes remain (13% response rate) even in the modern era, sharpening the decision-layer argument that AP/BP transformation is a point of maximum urgency for transplant referral and that conventional AML-style induction is insufficient as bridging. It also clarifies that canonical MPN driver mutations (JAK2/CALR/MPL) in de novo AML are not themselves the prognostic driver — co-occurring high-risk lesions are — which matters for risk stratification when distinguishing true blast-phase MPN from de novo AML with incidental JAK/STAT mutations.