Obesity Is An Adverse Prognostic Factor For Overall and Disease-Free Survival In Adult Acute Promyelocytic Leukemia But Not In Acute Myeloid Leukemia: A Pooled Analysis From Four Alliance Prospective Studies

There are mounting data associating obesity with an increased risk of developing acute myeloid leukemia (AML) and acute promyelocytic leukemia (APML). However, the role of obesity in the outcome of patients with AML and APML has not been extensively evaluated. In this study, we assess the effect of obesity in relapse and survival rates of clinical trial patients with AML and APML.

Data on patients ≥18 years from 4 prospective clinical trials from the Cancer and Leukemia Group B (Alliance) were pooled for this analysis (n=2,093). This reflects exclusion of 72 patients deemed ineligible or non-evaluable, and 8 patients without height or weight data. Three studies were in de novo AML: 9621 (n=393), 10503 (n=541) and 19808 (n=714), and one study was in de novo APML: 9710 (n=445). BMI was calculated following the formula (BMI=weight/height²) and categorized according to WHO criteria as underweight/normal (BMI <25 kg/m²), overweight (BMI 25-29.9 kg/m²) and obese (BMI 30+ kg/m²). AML and APML cohorts were analyzed separately. Baseline BMI was evaluated in relation to clinical characteristics, including age, gender, ECOG performance status (PS), and race/ethnicity. Univariate and multivariable logistic regression models were used to assess relationships of these factors with obesity. Analysis of clinical outcomes included overall survival (OS) in all patients, and disease-free survival (DFS) in those patients who achieved a complete response. The impact of obesity was evaluated using Kaplan-Meier methods and log-rank tests. Multivariate Cox regression analyses were used to assess prognostic impact of obesity while adjusting for other factors such as age, gender, PS, race and ethnicity on OS and DFS. P-values <0.05 were considered statistically significant.

In the AML cohort (n=1,648), median age was 46 years (range: 18-66 years), 53% were men, 87% had PS 0-1, 17% were non-white and 7% Hispanic. For the APML cohort (n=445), median age was 43 years (range: 18-80 years), 52% were male, 82% had PS 0-1, 18% were non-white and 11% Hispanic. The proportion of obesity at study entry for AML and APML was 38% and 50%, respectively. APML patients experienced a superior OS than AML patients (5-year OS: 81% vs. 37%). In AML, age, gender and race were all significantly related to whether or not patients were obese in univariate and multivariate settings. In APML, only age and race were significantly associated with obesity. Median follow-up time was 6.8 years for AML and 8.5 years for APML patients. In the AML cohort, 1058 of 1648 patients have died, and 813 of 1246 patients evaluable for DFS had an event. For APML, 102 of 445 patients died and 106 of 401 patients evaluable for DFS had an event. DFS and OS distributions were not significantly different in obese vs. non-obese AML patients (p=0.8 and p=0.6, respectively). However, obese APML patients had a significantly shorter OS than non-obese patients (HR=1.61, p=0.02; 5-year OS rates: 76% vs. 85%, respectively). Similarly, obese APML patients tended to have shorter DFS than non-obese patients (HR=1.49, p=0.05; 5-year DFS rates: 74% vs. 82%, respectively). Only focusing on those APML patients between 18-60 years of age, differential OS and DFS in obese patients was more pronounced with corresponding worse survival (OS: HR=2.13, p=0.003; DFS: HR=1.75, p=0.015).

Obesity did not influence OS and DFS in AML. However, in a multivariate analysis, obese APML patients had significantly inferior OS and DFS than non-obese patients. Multivariate analysis suggests that the adverse impact of obesity in APML is independent of age, sex, performance and race/ethnicity. Previous studies have suggested that obesity is a risk factor for developing APML; we now show that obesity at diagnosis also confers a worse prognosis. Reasons for our findings could include potential pharmacological issues, such as relative dose intensity, which needs further research. Moreover, it will be necessary to determine if our findings will be replicated in patients with APML treated without chemotherapy as is now becoming the standard of care for those who present with white blood cell counts <10,000/uL.

No relevant conflicts of interest to declare.