Clinical Presentation and Induction Outcomes Among Hispanic and Non-Hispanic Patients with Acute Promyelocytic Leukemia

Background: Untreated acute promyelocytic leukemia (APL) is the most aggressive form of acute myeloid leukemia (AML) with a median survival of <1 month. With treatment, survival can exceed 95%. Despite excellent outcomes overall, population-based studies report inferior survival among Hispanic versus (v) non-Hispanic (NH) patients. At Columbia University Medical Center (CUIMC), over 50% of the catchment population is Hispanic. Leveraging this highly diverse cohort, we evaluated clinical outcomes among Hispanic and NH patients with newly diagnosed APL.

Methods: Patients treated for newly diagnosed APL at CUIMC from 2003-2023 were retrospectively assessed. Patient, disease, and treatment data were summarized by treating providers and collected from electronic medical records (EMR). Patient-reported ethnicity was categorized as Hispanic or NH. Baseline comorbidities were assessed by the Charlson Comorbidity Index (CCI) and performance status (PS) was measured with the Eastern Cooperative Oncology Group (ECOG) scale. Risk stratification was based on the Sanz scoring system. Coagulopathy was defined as fibrinogen <150 mg/dL. The primary endpoint was overall survival (OS) by ethnicity, defined as the time from APL diagnosis to the time of death or last contact. Secondary endpoints included relapse, time from diagnosis to polymerase chain reaction (PCR) clearance, and incidence of bleeding, differentiation syndrome, and neutropenic fever. Group differences were tested via chi-square, Fisher’s exact, and log-rank tests and were presented as odds ratios (OR) with 95% confidence intervals (CI). A p-value <0.05 was significant.

Results: Median follow-up was 5.2 years (y). A total of 62 patients, 16-80y (mean 50y), were included and 39% (N=24) were Hispanic. At presentation, Hispanic (v. NH) patients were younger (mean 40y v. 56y, p<0.001) with lower CCI (median 2 vs. 4, p=0.002). A higher proportion of Hispanic patients had high risk disease (48% v. 29%, p=0.17). Specifically, Hispanics had higher white blood cell counts (WBC, 10⁹/L) (median 10 v. 2, p=0.045) and lower platelet counts (10⁹/L) (median 20 v. 36, p=0.036). Initial hemoglobin, blast count, and PS were similar. Hispanic patients were more likely to exhibit a coagulopathy (83% v. 58%, p=0.046) at presentation and receive chemotherapy (70% v. 40%, p=0.033). Despite being more likely to receive all-trans retinoic acid (ATRA) within 48 hours of diagnosis (100% v. 73%, p=0.008), Hispanics experienced more bleeding events (50% v. 24%, p=0.033; OR 4.0 [95% CI: 1.3-13.0], p=0.018) and differentiation syndrome (54% v. 45%, p=0.469). Given these complications, Hispanic patients had longer induction lengths of stay (median 50 v. 35 days, p=0.171). Rates of neutropenic fever, relapse, time to PCR clearance, and OS did not differ by ethnicity.

Conclusions: Despite a younger age at diagnosis, fewer comorbidities, and timely ATRA administration, Hispanic patients with APL were more often high-risk and had more induction toxicities with longer admissions compared to NH patients. While bleeding and prolonged admissions may be related to Hispanic patients having higher risk disease and presenting with concurrent coagulopathies, it may also be due to a greater utilization of myelosuppressive chemotherapy in this cohort. Alternatively, there may be inherent variability in tumor biology or social determinants of health that are driving higher risk disease and poorer outcomes in Hispanic patients. This real-world study is the first to illustrate the greater severity of hematologic abnormalities and higher incidence of APL-related complications endured by Hispanic patients, thus highlighting the importance of considering APL in the differential diagnosis of Hispanic patients with suspected AML and initiating urgent treatment. While our study was not powered to detect a difference in OS, it is possible that the observed differences in early, life-threatening toxicities between Hispanic and NH patients with APL may be driving survival disparities reported at the population level. Further investigation is therefore needed to explain disparate responses to standard therapies and yield a more precise means of prognostication in this vulnerable patient population.