The Diets Nobody Expected to Help Cancer Treatment
Eat well. Keep the weight off. It is standard advice after almost every cancer diagnosis.
So imagine the surprise in a Montreal lab when some of the diets most likely to pile on weight helped a powerful cancer treatment work better — at least in mice. McGill University researchers say it may explain one of the strangest puzzles in cancer medicine: the so-called obesity paradox.
A paradox doctors keep running into
Obesity is linked to a higher risk of developing cancer and dying from it. Yet study after study has found something odd: cancer patients with a higher body mass index (BMI) often respond better to immune checkpoint inhibitors, a widely used immunotherapy — an association reported in cancers including lung cancer and melanoma.
Was the extra weight itself somehow helping? A team co-led by Daniela Quail of McGill's Rosalind & Morris Goodman Cancer Institute, with McGill's Logan Walsh and CR-CHUM's Bertrand Routy and Arielle Elkrief, decided to test a different suspect: the food itself, and the trillions of gut bacteria it feeds.
Twelve diets, fifteen weeks, one question
Most mouse obesity studies use one high-fat diet. This team built 12 different diets modelled on real human eating patterns — Mediterranean, Japanese, vegan, ketogenic, American, low fat, high fat, Western and fibre-enriched variations. Mice stayed on the diets for 15 weeks, in groups of 14 to 15 per diet, then were tested in a lung cancer model.
Two results stood out:
- Tumour growth did not track with body weight — and response to the anti-PD-1 immunotherapy showed no significant association with body weight or fat mass.
- Diet, not weight, predicted response. Among the obesity-linked diets, 4 out of 6 were associated with a response to treatment, compared with 2 out of 6 non-obesogenic diets. Three of the four most treatment-sensitive diets were obesity-associated.
Each diet produced a distinct community of gut bacteria, and obesity-linked diets were enriched for Lactobacillaceae, a family previously linked to favourable immunotherapy responses. "There is a synergistic relationship between the bacteria that reside in our GI tract, and the food we eat," Quail said — the food nourishes the bacteria, and together they shape the immune response.
What this means for the BMI number
If the McGill team is right, BMI may be a blunt stand-in for what actually matters: two patients can share the same BMI while eating very differently and carrying very different gut bacteria — and, potentially, responding very differently to the same drug. The goal now is to harness whatever is driving the gut effect, without the weight gain.
The honest part
This was a mouse study, not a human trial. The patient-side "paradox" comes from observational research — an association between BMI and treatment response, never proof that weight causes better outcomes. And the researchers are explicit: nobody with cancer should adopt a weight-gaining diet on the back of this work — obesity carries well-known, serious risks. Led by McGill PhD graduate Lysanne Desharnais and published in Nature, the study is a map for the next experiments — not a prescription.
The Twikup Take
The scale may have been getting credit for work done by dinner and the bacteria it feeds. Until human trials say otherwise, BMI stays a rough signal, not a verdict.
The paradox was never really about the weight. It may have been about what the weight was made of.
Image: AI-generated illustration.