Impact of BMI On Treatment-Related Toxicity in United States Veterans with Diffuse Large B-Cell Lymphoma (DLBCL) Receiving CHOP+/−R.

Obesity is associated with an increase in all-cause mortality, much of which is due to cancer. Recently, we demonstrated improved survival in overweight and obese Veterans with DLBCL. To better understand this survival disparity, we explored the frequency of dose-reductions and treatment-related mortality (TRM) as they relate to BMI in this population.

We identified a cohort of DLBCL patients diagnosed between 1998 and 2008 in the Veterans Health Administration (VHA) central cancer registry. Additional data on height, weight, treatment drug, dose, date of diagnosis, date of death, and comorbidities was collected from administrative records or via chart abstraction. Only patients treated with CHOP +/− R within the VHA health system were included in the study cohort. Patients with primary cutaneous, CNS, or HIV-related DLBCL were excluded. Body Mass Index (BMI) was used to categorize patients as normal weight (BMI 18.5 to <25), overweight (BMI 25 to <30), or obese (BMI >30) in accordance with the WHO classification. First cycle dose reduction was defined as an average relative dose-intensity of doxorubicin and cyclophosphamide <85% of expected dose during the 1^st cycle of chemotherapy. Treatment-related mortality (TRM) was defined as death within 30 days of last treatment cycle. Logistic regression modeling was used to explore the relationship between BMI groups and TRM after controlling for other factors.

Of the 3,227 initial patients, 1,622 met inclusion criteria. Patient characteristics are listed in Table 1. Significant differences in age, stage, Charlson comorbidity score, and presence of B-symptoms were noted between BMI groups. Consistent with our previous work, Cox modeling indicated that overweight and obese patients had better overall survival compared to the non-obese after controlling for these factors (HR, 0.80; 95% CI, 0.69 to 0.94 and HR, 0.79; 95% CI, 0.65 to 0.95, respectively). There were significant differences in the proportion of patients with 1^stcycle dose reductions and use of G-CSF among the 3 groups, with obese patients appearing to undergo more dose-reductions and receiving less G-CSF support. There were also significant differences in TRM among the 3 groups, with overweight and obese patients experiencing less TRM. In multivariate logistic regression controlling for age, stage, LDH, Charlson comorbidity score, and G-CSF use, being obese was associated with decreased TRM (OR 0.57, 95% CI 0.35 to 0.93). Overweight BMI was associated with a trend towards decreased TRM (OR 0.71, 95% CI 0.48 to 1.06).

Recent ASCO guidelines recommend against empiric dose-reduction and support full weight-based chemotherapeutic dosing in obese patients. Despite higher rates of empiric dose-reduction and lower rates of G-CSF support, obese patients within our cohort have improved overall survival. This is due in part to a decreased risk of TRM in the obese, even after controlling for baseline differences between BMI groups. Further analysis of the association between BMI, treatment-related toxicity, and dose-intensity characteristics are warranted within this cohort. Given the observational nature of the data and the two-way relationships between dose-intensity, treatment-related toxicity, and long-term outcomes, such analysis must be approached with caution.

Carson:Genentech: Consultancy, Speakers Bureau.