Assessing Disparities in a Chicagoland Cohort of Patients with Myelofibrosis

Introduction: Amongst the classic Philadelphia-chromosome negative chronic-phase myeloproliferative neoplasms (MPNs), primary myelofibrosis (PMF) and post-polycythemia vera/post-essential thrombocythemia myelofibrosis (also known as secondary myelofibrosis (SMF)) carry the poorest long-term outcomes. Our group previously evaluated drivers of disparate outcomes in a Chicagoland cohort of patients with acute myeloid leukemia (AML) (Abraham et al Blood 2022; Palmer et al, Blood Adv 2024); we aimed to investigate potential drivers of outcomes in myelofibrosis (MF) with a focus on disease characteristics, access to standard therapies, rates of allogeneic transplant (allo-HCT), and complication rates. Methods: Adult patients (pts) with MF diagnosed at 5 Chicagoland centers from 2011 onward were included. Disease characteristics, Dynamic International Prognostic Score System + (DIPSS+), Mutation-Enhanced International Prognostic Scoring System 70+ version 2 (MIPSS70+ v2.0), JAK inhibitor use, allo-HCT rate, venous thrombosis rate, arterial thrombosis rate, and accelerated/blast-phase (AP/BP) progression were analyzed for the entire cohort and by race/ethnicity. Overall survival (OS) was calculated by Kaplan-Meier analysis. Results: Our total cohort comprised of 496 pts with MF; 352 were non-Hispanic White (NHW), 83 were non-Hispanic Black (NHB), 29 pts were Hispanic, and 32 reported a race/ethnicity not represented by these (Other). Forty-five percent of the cohort was female and the median age of the whole cohort was 66 years old (yo). Median age by race/ethnicity was 66yo for NHW, 63 for NHB, and 70 for Hispanic (p=0.11). Amongst the NHW pts 64% had PMF while 36% had SMF; NHB and Hispanic pts had higher rates of PMF (80% and 86%, (p=0.004). Driver mutation testing was available for 443 pts; 70% had a JAK2 mutation, 18% had a CALR mutation, and 5% had a MPL mutation with no significant difference between NHW, NHB, and Hispanic pts. DIPSS+ was evaluable in 479 pts (341 NHW, 79 NHB, 29 Hispanic, 30 Other); 16% had low (L), 30% had intermediate-1 (I-1), 42% had intermediate-2 (I-2), and 12% had high (H) risk disease. For NHW pts 16% had L, 32% had I-1, 42% had I-2, and 11% had H risk disease. For NHB pts 23% had L, 25% had I-1, 42% had I-2, and 10% had H risk disease. For Hispanic pts 10% had L, 24% had I-1, 62% had I-2, and 3% had H risk disease (p=0.059). MIPSS70+ v2.0 was evaluable in 425 pts (299 NHW, 74 NHB, 27 Hispanic, 25 Other); 5% had very low (VL), 29% had low (L), 25% had intermediate (I), 32% had high (H), and 9% had very high (VH) risk disease. There was no significant difference in MIPSS70+ v2.0 risk score between NHW (5% VL, 29% L, 25% I, 31% H, 9% VH), NHB (8% VL, 28% L, 24% I, 31% H, 8% VH), and Hispanic (0% VL, 26% L, 19% I, 44% H, 11% VH) pts (p=0.9218). JAK inhibitors were used in 54% of the entire cohort; the rate of JAK inhibitor use was greatest for NHW and Hispanic pts compared to NHB pts (57% and 59% vs. 37%, respectively (p=0.012). There was no significant difference in allo-HCT rate (23% NHW, 12% NHB, 14% Hispanic) although the rates were numerically lower in NHB and Hispanic pts when compared to NHW pts (p=0.524). We then compared rates of thrombotic complications, and there was no significant difference between the groups. The venous thromboembolic rate was 21% for NHW, 29% for NHB, and 7% for Hispanic pts (p=0.08). Arterial thrombosis rate was 18% for NHW, 16% for NHB, and 14% for Hispanic pts (p=0.908). Fourteen percent of pts had progression to AP/BP with no significant difference between NHW, NHB, and Hispanic pts. Median OS was also analyzed for the cohort. Median OS was 7.69 years for NHW pts, not reached for NHB pts, and 9.48 years for Hispanic pts with no significant difference. After controlling for age and gender, NHB pts had a 44% greater hazard for all-cause mortality when compared to NHW pts (p=0.07) while Hispanic pts had a 14% lower hazard (p=0.66). Conclusions: In our cohort, NHB and Hispanic pts had higher rates of PMF in comparison to SMF when compared to NHW pts. Distribution of prognostic scores were similar across those three groups; despite this, JAK inhibitor usage rate was significantly lower in NHB pts. Lastly, when controlling for age and gender, NHB pts had a greater hazard for mortality when compared to NHW although this was not statistically significant. Insurance type and census tract level analyses will be incorporated in future analyses to further assess potential drivers of disparate outcomes.