Work

Body Mass Index and Clinical Associations in Australasian Lymphoma Patients: A Lymphoma and Related Diseases Registry Study

Year: 2024

Type: conference-abstract

Abstract: Introduction: The prevalence of overweight and obesity is increasing rapidly in most countries including Australia (Phelps et al., 2024). Underweight, though much less common, is also associated with adverse health outcomes. Body mass index (BMI) despite known limitations is widely used as a surrogate marker of adiposity. There is poor understanding of baseline clinical risk profiles and implications of BMI in patients with lymphoma. Recent international guidelines indicate that cancer therapy doses should not be modified for obesity alone (Griggs et al., 2021). There are no specific recommendations for low BMI patients. Methods: This study examined BMI in relation to clinical features and outcomes in lymphoma patients diagnosed from 01 January 2016 to 23 January 2024 from the Australasian Lymphoma and Related Diseases registry. Four key lymphoma subtypes were included; Hodgkin lymphoma (HL), diffuse large B cell lymphoma (DLBCL), follicular lymphoma (FL) and chronic lymphocytic leukaemia (CLL). BMI (kg/m2) was grouped according to standard WHO categories; underweight (<18.5), normal (18.5-24.9), overweight (25.0-30.0) and obese (> 30.0). An exploratory analysis comparing BMI groups for baseline clinical features, treatment categories and outcomes was performed. Categorical variables were described in frequencies and percentages with comparisons of p-values using the chi-squared test. Continuous variables were described with medians using the interquartile range (IQR). Kaplan-Meier analysis was conducted for progression-free survival (PFS) and overall survival (OS) for histological subgroups according to BMI. A p value of <0.05 was considered significant. Results: 3602 eligible patients were included; 102 (2.8%) were underweight, 1190 (33%) were normal weight, 1251 (34.7%) were overweight and 1059 (29.4%) were obese. For all subtypes, the underweight group represented less than 5% of patients. There were differences in the age profile of HL patients based on BMI; the small underweight population was younger (median 26 years (y)), while overweight and obese patients were older (median 42y and 41y respectively) (p<0.001). Underweight patients had poorer ECOG performance status for all subtypes, apart from CLL where numbers were too small for comparison. The proportion of HL with advanced stage disease was enriched in the underweight group (61.3% compared to 48% normal weight, p=0.007). There was no significant difference in stage by BMI for any other histologic subtypes. There were no significant differences in the proportions of patients receiving different treatment regimens across BMI categories in any subtype (HL p=0.85, DLBCL p=0.09, FL p=0.69, CLL p=0.24). Initial dose reductions were infrequent (< 9% in all types of lymphoma) and there was no association between the likelihood of a dose reduction and BMI; (HL p=0.22, DLBCL p=0.71, FL p=0.18). For DLBCL a significantly higher proportion of the underweight group (22.2%) received delayed treatment, compared to the other BMI groups (p=0.001). This was not observed for any of the other lymphoma subtypes. Patients with an overweight or obese BMI had no significant decrement in overall survival (OS) or progression free survival (PFS), compared to those with a normal BMI for any of the subtypes, with median follow-up of 72 months. However, in the multivariable analyses adjusting for age and sex, patients with an underweight BMI demonstrated an inferior OS for HL compared to normal weight (adjusted hazard ratio [adjHR]=2.7, 95% confidence interval [CI]=1.02-7.16, p=0.05) and inferior PFS (adjHR=1.81, 95%CI=1.14-2.90, p=0.01) and OS (adjHR=1.62, 95%CI=1.04-2.54, p=0.03) for DLBCL, compared to patients with a normal BMI. Conclusions: In our study of 3602 Australasian lymphoma patients, we found few differences in baseline clinical characteristics for overweight and obese patients compared to those with a normal weight. Initial dose reductions were infrequent with no correlation with BMI. There was no evidence of a reduction in OS and PFS for obese and overweight patients compared to normal weight BMI patients in any histologic subtype. The minority of underweight patients showed enrichment of adverse prognostic features and inferior outcomes in DLBCL and HL. Our findings support continued standard dosing in overweight and obese patients; and identify poor outcomes requiring careful management in underweight populations.

Source: Blood

DOI: 10.1182/blood-2024-206277