Does BMI Have Prognostic Value in DLBCL? A Review Finds Mixed Results

As obesity rates rise worldwide, more attention than ever is paid to links between obesity and cancer. But those connections vary by type of cancer, and a recent review of studies involving diffuse large B-cell lymphoma (DLBCL) shows how the links between body mass index (BMI) and cancer are not always straightward.

A systematic review and meta-analysis, published in Frontiers in Nutrition June 23, 2026, by a group of authors based at universities in China, examines whether BMI predicts survival DLBCL, the most common subtype of non-Hodgkin lymphoma. Because chemotherapy doses are typically calculated from body surface area and actual body weight, the authors set out to clarify how BMI relates to overall survival (OS) and progression-free survival (PFS), given that prior studies had produced conflicting results.

The researchers searched PubMed, Embase, and the Cochrane Library through April 2026, supplemented by reference screening, ultimately identifying 23 eligible studies that covered more than 11,000 patients across the United States, Europe, Asia, and Australasia. Most studies were retrospective, and quality was rated moderate to good using the MINORS tool (Methodological Index for Non-Randomized Studies), a validated instrument that assesses quality and risk in nonrandomized studies.

The team first compared overweight and obese patients against those who were not, and they then compared underweight, overweight, and obese categories separately relative to normal weight, which are the conventional BMI categories defined by the World Health Organization.

The study’s key finding was that patients classified as overweight or obese overall had better OS (HR = 0.82) and PFS (HR = 0.83) than those who were not, and this pattern held up across subgroups restricted to R-CHOP-based regimens, multivariable-adjusted estimates, and analyses using normal-weight patients as the reference group. However, when investigators examined BMI categories individually, a more nuanced picture emerged.

Overweight status alone was clearly linked to better OS (HR = 0.77) and PFS (HR = 0.80), with no heterogeneity across studies. Obesity, by contrast, showed no significant association with either OS (HR = 0.93) or PFS (HR = 0.95). Underweight patients fared worst, with substantially higher risk of death (HR = 1.85) and progression (HR = 1.64) compared with normal-weight patients. Funnel plots and Egger's tests found no evidence of significant publication bias.

Factors That May Explain the Results

“Our research indicated that overweight status correlated with superior survival outcomes, whereas obesity lacked significant prognostic relevance,” the authors wrote. “This suggests that the relationship between BMI and DLBCL prognosis may be nonlinear, potentially even exhibiting an inverted U-shaped curve.” Survival, the authors suggest, appears concentrated in the overweight range rather than extending to obesity. They propose several biological explanations:

Adipose tissue and drug absorption. Excess adipose tissue in overweight patients may broaden the distribution of lipophilic chemotherapy agents such as doxorubicin and slow drug clearance through hepatic changes, potentially increasing drug exposure and efficacy. Overweight patients may also have greater nutritional reserve and more favorable immune cell profiles, such as elevated NK-like T cells, that support treatment tolerance.

Weight affects inflammation, toxicity. Obesity, in contrast, may involve sarcopenia, chronic inflammation, and tumor-protective effects of stressed adipocytes that offset any dosing advantage, along with cardiotoxicity risk and possible under-dosing when clinicians cap chemotherapy for safety reasons. Underweight patients likely suffer from limited drug distribution, cachexia-related muscle and fat loss, and impaired cardiac function, all of which reduce treatment tolerance and worsen outcomes.

This analysis substantially expands on a prior 2021 meta-analysis (14 studies versus 23 here) and is the first to explicitly incorporate Asian-specific BMI thresholds, finding — unlike the earlier review—a significant PFS benefit for overweight status in addition to the OS benefit.

The authors acknowledge important limitations. BMI could only be analyzed categorically rather than continuously, given how the underlying data were reported. Differing BMI classification systems across regions could introduce misclassification bias, since individual-level data were unavailable for reclassification. The underweight and obesity subgroups relied on relatively few studies, widening confidence intervals. Data on actual delivered chemotherapy dose intensity, a plausible confounder, were unavailable.

“Given its simplicity and routine availability, BMI may serve as a practical adjunctive indicator in the baseline assessment of patients with DLBCL,” the authors wrote. However, “Current evidence remains insufficient to confirm a significant prognostic impact of obesity.”

Given these limitations, “Further prospective studies are warranted to evaluate BMI as a continuous variable and to incorporate time-dependent modeling during treatment to better define its relationship with DLBCL outcomes.”

Reference

Su L, Tian Y, Li Y, et al. Evaluating the prognostic value of body mass index in diffuse large B-cell lymphoma: a systematic review and meta-analysis. Front Nutr. 2026;13:1762123. doi: 10.3389/fnut.2026.1762123