Obesity Effect on Cancer Risk Twice That Currently Thought

Current estimates of the effect of obesity in increasing the risk for cancer are based on observational studies and may be underestimating the effect. The risk may be twice as large as is currently thought, says a group of international researchers who report a new approach to estimating the risk.

They conducted a systematic analysis of genetic data using a Mendelian randomization (MR) approach.

"The importance of these analyses is that they suggest that the effect of being overweight on cancer risk has been underestimated in the past and that obesity plays an even more important role in cancer than previously suggested," commented coauthor Richard Martin, BMedSci, PhD, professor of clinical epidemiology at the University of Bristol Medical School, United Kingdom, in a statement.

The study was published online July 25 in the International Journal of Epidemiology.

"We compared the genetic MR estimates of the association between BMI [body mass index] and cancer risk with the estimates from classical cohort studies for eight common cancer types that have been linked to obesity," explain corresponding author Paul Brennan, PhD, International Agency for Research on Cancer, Lyon, France, and colleagues.

"[Our] MR-based results [show] the cancer burden explained by elevated BMI for the six cancers [analyzed] would be almost 8% in high-income countries, as opposed to 3% based on previous estimates for these cancers, suggesting that the cancer burden has been substantially underestimated for the majority of obesity-related cancers," they conclude.

Germline Genetic Variants

The authors explain that, for a variety of reasons, it is difficult to identify factors that may give rise to cancer. These reasons include limitations in epidemiologic study design and inherent problems of confounding and reverse causation.

In contrast, MR uses germline genetic variants as instruments to assess risk factors of interest.

MR estimates are therefore less susceptible to confounding errors and more accurately reflect exposure to the risk factor of interest across a lifespan.

For their study, the team compared genetic MR estimates of the association between BMI and cancer risk with estimates arrived at by the World Cancer Research Fund (WCRF).

For each of the six cancer sites analyzed, "the MR estimate for a 5-unit increment in BMI was notably higher than the WCRF estimate," the investigators report.

For example, the MR estimate was approximately twofold higher for kidney cancer, endometrial cancer, ovarian cancer, and esophageal cancer than WCRF estimates.

For pancreatic and colorectal cancer, the MR estimate for every 5-unit increase in BMI was more than fourfold higher than WCRF estimates, the researchers add.

For breast and lung cancer, MR estimates of BMI risk were discordant with those from previous observational studies.

"In particular, previous observational studies have reported a modest protective effect of elevated BMI for pre-menopausal breast cancer and a modest increase in risk for post-menopausal breast cancer," the authors point out.

In contrast, MR analyses indicated that elevated BMI has a protective effect against breast cancer in both pre- and postmenopausal women.

Similarly, MR estimates for lung cancer indicate that there is an increased risk for lung cancer overall with higher BMIs as well as for both squamous and small cell lung cancer.

Again, this is in contrast with previous observational evidence, which has shown an inverse association between being overweight and developing lung cancer.

The investigators point out that recent evidence indicates that individuals with higher BMIs are more likely to smoke and to smoke more heavily than persons with lower BMIs.

"This would imply that obesity is an unrecognized yet important causal factor in lung cancer, albeit mediated by its influence on smoking patterns," the authors suggest.

As for breast cancer, MR results suggest that there is a "more complex association" between higher BMI and breast cancer risk than earlier studies might have suggested.

Indeed, there is increasing evidence that weight gain during adulthood is a more important determinant of postmenopausal breast cancer than weight per se.

Relative Risk Estimates

The authors explain that the observed differences in relative risk estimates between those arrived at by MR analyses and estimates from traditional observational studies may in part be explained by the ability of MR to avoid bias by reverse causation and residual confounding.

However, because MR estimates are based on genetic "proxies" of BMI, the more important reason why MR analyses provide more accurate estimates of BMI on cancer risk is that they are able to reflect the risk of accumulated exposure to elevated BMI across a lifespan, the authors also suggest.

"MR studies have proven instrumental in improving our understanding of obesity in cancer aetiology," Brennan and colleagues state.

The study was supported by the World Cancer Research Fund. The authors have disclosed no relevant financial relationships.

Int J Epidemiol. Published online July 25, 2019. Abstract

Follow Medscape on Facebook, Twitter, Instagram, and YouTube