Overweight patients less likely to die of colorectal cancer

BMI at the time of colorectal cancer diagnosis is associated with mortality risk, according to results of a retrospective observational study.

Patients categorized as underweight or class II to III obese at the time of and following diagnosis experienced an increased risk for all-cause mortality and colorectal cancer-specific mortality compared with participants categorized as having a normal weight.

In contrast, overweight patients experienced a decreased risk for all-cause and colorectal cancer-related mortality.

“Overweight and obesity have been identified as risk factors for many health conditions, but for people with colorectal cancer, some extra weight may provide protections against mortality,” Candyce H. Kroenke, ScD, research scientist at Kaiser Permanente Division of Research, said in a press release. “Our study, which represents the largest cohort of colorectal cancer patients with the most comprehensive data regarding patient weight before, at time of and following diagnosis supports the notion of the ‘obesity paradox.’”

Because previous analyses have been limited by selection bias, reverse causality and collider bias, Kroenke and colleagues examined associations between BMI and outcomes for 3,408 patients (50.3% women) aged 18 to 80 years with stage I to III colorectal cancer from the Kaiser Permanente Northern California population.

Median collection time of “at-diagnosis” BMI — defined as the WHO–classified BMI measurement closest to diagnosis, either before or after — was 0 months (range, –5.3 to 6).

Median follow-up was 4.5 years.

At diagnosis, 67 patients were underweight (BMI < 18.5), 1,050 were normal weight (BMI 18.5-24.9), 1,214 were overweight (BMI, 25-29.9), 671 were stage I obese (BMI, 30-34.9), and 406 were stage II (BMI, 35-29.9) or III obese (BMI 40).

Seeking further analysis, researchers expanded the WHO classifications for normal-weight participants into low normal (n = 553; BMI 18.5-22.9) and high normal (n = 497; BMI 23-24.9) and overweight participants into low overweight (n = 787; BMI 25-27.9) and high overweight (n = 427; BMI 28-29.9). Participants in the low-normal subgroup served as the reference in the expanded analysis.

In total, 617 participants died, 411 as a result of colorectal cancer.

Results of multivariate analysis using the expanded classification categories showed underweight patients had a significantly increased risk for all-cause mortality (HR = 2.65, 95% CI, 1.63-4.31) and colorectal cancer-specific mortality (HR = 3.18, 95% CI, 1.78-5.69) compared with the low-normal group.

Patients classified as class II or III obese also demonstrated a trend toward increased risk for all-cause mortality (HR = 1.33, 95% CI, 0.89-1.98) and colorectal cancer-specific mortality (HR = 1.12, 95% CI, 0.61-2.06).

Risk for all-cause mortality was lower among patients in the high-normal group (HR = 0.77, 95% CI, 0.56-0.1.06), low-overweight group (HR = 0.75, 95% CI, 0.55-0.74) and high-overweight group (HR = 0.52, 95% CI, 0.35-0.77). Participants classified as stage I obese showed no statistical difference.

Post-diagnosis BMI data — recorded 15 months following diagnosis — were available from 3,157 participants. After a median follow-up of 3.5 years, 482 participants died, 317 of whom died of colorectal cancer.

Risk for all-cause mortality increased significantly for participants in the underweight group (HR = 3.38, 95% CI, 2.19-5.2) and decreased for all BMI groups higher than low normal.

Researchers acknowledged that a lack of data regarding physical activity and the inability to control for detailed treatment information may have impacted the results.

“The current findings, and previous and ongoing research on the obesity paradox, suggest that recommendations for the ideal weight range associated with the best outcomes after a cancer diagnosis may not be the same as the ideal weight range to prevent cancer,” Bette J. Caan, DrPH, research scientist at Kaiser Permanente Division of Research, said in a press release. “Just as treatment differs by cancer, ideal weight recommendations may vary according to cancer site.”

The “complex” association between BMI and cancer survival can be a true obesity paradox, or can be related to methodologic issues and collider bias, Andrew G. Renehan, PhD, FRCS, and Matthew Sperrin, PhD, both of University of Manchester in the United Kingdom, wrote in an accompanying editorial.

“These findings should not alter current clinical practice or recommendations,” Renehan and Sperrin wrote. “Although we appreciate that the interpretation of the association between BMI and mortality in patients with cancer is considerably more challenging than that between BMI and incidence cancer risk, we have a duty to appropriately convey this complexity to our patients and populations. The observations of Kroenke et al might be paradoxical, but they are readily explained through several methodologic mechanisms.” – by Nick Andrews

Disclos ure: The researchers and Renehan report no relevant financial disclosures.