What can body mass index tell us about immunotherapy outcomes?

Among cancer patients receiving immunotherapy, those in higher body mass index (BMI) groups tended to live longer and also had more immune-related side effects. The two findings came from separate pooled comparisons in a 2021 systematic review and meta-analysis. They describe patterns across groups, not a personal benefit or side-effect forecast from a BMI score.

BMI is a measure calculated from a person’s weight and height. The review asks whether that simple measure tracked with outcomes during checkpoint immunotherapy, and finds that the answer varied across the studies and patient groups it combined.

A pooled pattern across several cancers

The researchers identified 771 potentially relevant publications and included 20 studies in their analyses of survival and immune-related adverse events, often shortened to irAEs. The cancers represented included melanoma, lung cancer and renal cell carcinoma. Nivolumab, pembrolizumab and atezolizumab were among the common immune checkpoint inhibitors in the studies, and most participants were from the United States.

Most articles used 25 kg/m² as the cutoff for a higher BMI, though definitions varied. The review pooled comparisons between higher- and lower-BMI groups rather than testing one shared threshold across every cancer and treatment. Its literature search covered studies published through June 2020; the review itself was published on January 29, 2021.

Survival estimates compare groups over time

For overall survival, the pooled hazard ratio was 0.71, with a 95% confidence interval of 0.59 to 0.85. For progression-free survival, it was 0.76 (95% CI 0.65 to 0.88). In both comparisons, estimates below 1 mean the higher-BMI groups had a lower estimated hazard of the measured event during follow-up than the lower-BMI groups. Overall survival concerns death; progression-free survival tracks time before cancer progression or death.

A hazard ratio is not a percentage-point increase in the chance of living longer, nor does it say how many extra months a particular person might live. It summarizes a relative comparison between groups across follow-up. The pooled estimates point in the same direction, but the researchers also found substantial variation among the survival results from individual studies (p < 0.001). The overall averages therefore combine results that did not line up uniformly.

Side effects were a separate outcome

For irAEs, the pooled odds ratio was 2.54 (95% CI 1.12 to 5.79) in the higher-BMI comparison. An odds ratio compares the odds of an event between groups; it is not a statement that each patient in the higher-BMI group had 2.54 times the probability of a side effect. Without the underlying event rates, this estimate does not give an individual’s absolute chance of experiencing one.

The confidence interval is wide, so the size of the difference is less precise than the headline estimate may suggest. The review found higher odds of immune-related adverse events alongside the survival pattern, but it did not establish why the side-effect comparison differed. These outcomes should be read separately: a group-level survival association does not measure toxicity, and the toxicity estimate does not explain the survival pattern.

The average shifted across patient groups

The review’s threshold analyses show why a single pooled number can be difficult to apply across studies. In comparisons using a BMI threshold of at least 30, pooled hazard ratios were 0.64 for overall survival (95% CI 0.48 to 0.86) and 0.73 for progression-free survival (95% CI 0.58 to 0.92). In comparisons using at least 25, the estimates were 0.72 for overall survival (95% CI 0.50 to 1.03) and 0.75 for progression-free survival (95% CI 0.53 to 1.06). The latter confidence intervals include 1, so those pooled estimates were less conclusive.

These are different pooled comparisons based on the thresholds used in the studies; they do not follow the same people as their BMI rises from 25 to 30. The estimates therefore do not establish a dose-response rule in which a higher BMI reliably means a larger treatment benefit.

Results also differed across the review’s subgroups. Its BMI–survival association appeared in the male subgroup but not the female subgroup, and the authors did not observe an association in renal cell carcinoma. The evidence centered on several cancer types, but these differences show that the overall average did not describe every subgroup in the same way.

BMI may reflect more than one thing

A weight-to-height measure does not identify the biological reason an outcome differs. The authors discussed performance status—how well a patient is functioning during illness—as one possible factor that BMI could partly stand in for. They also raised obesity-related inflammation and the possibility that weight-based immunotherapy dosing could contribute to side effects. The pooled comparisons did not isolate any of these explanations as the cause of the paired findings.

The comparison also answers a narrower question than whether BMI predicts who benefits specifically from immunotherapy. The included estimates compared higher- and lower-BMI groups receiving immunotherapy; they did not establish that treatment works better for one BMI group on immunotherapy than it would on a different treatment. Nor do they show that changing a person’s BMI would improve treatment outcomes. BMI here is an associated measure in a set of studies, not a treatment or weight-management recommendation.

How this could change immunotherapy planning

A validated BMI-related measure could help oncology teams discuss likely treatment benefit and prepare for immune side effects before checkpoint therapy. That use would depend on evidence that the measure adds reliable patient-level information across relevant cancers and patient groups, and distinguishes an immunotherapy-specific benefit from general prognosis. If those conditions are met, the measure could support more informed treatment conversations and side-effect preparation.

Key Sources

• Primary source: The predictive value of body mass index on prognosis and adverse events of cancers treated with immunotherapy: a systematic review and meta-analysis.
  DOI or Journal Page · Review or Meta-Analysis · 2021-01-29