In a matched cohort study of more than 69,000 Canadian women, risk for incident breast cancer at 1 year was 40% higher among women who had not undergone bariatric surgery, compared with those who had. The risk remained elevated through 5 years of follow-up.
The findings were “definitely a bit surprising,” study author Aristithes G. Doumouras, MD, MPH, assistant professor of surgery at McMaster University, Hamilton, Ont., said in an interview. “The patients that underwent bariatric surgery had better cancer outcomes than patients who weighed less than they did, so it showed that there was more at play than just weight loss. This effect was durable [and] shows how powerful the surgery is, [as well as] the fact that we haven’t even explored all of its effects.”
The study was published online in JAMA Surgery.
To determine whether there is a residual risk for breast cancer following bariatric surgery for obesity, the investigators analyzed clinical and administrative data collected between 2010 and 2016 in Ontario. They retrospectively matched women with obesity who underwent bariatric surgery with women without a history of bariatric surgery. Participants were matched by age and breast cancer screening status. Covariates included diabetes status, neighborhood income quintile, and measures of health care use. The population included 69,260 women (mean age, 45 years).
Among participants who underwent bariatric surgery for obesity, baseline body mass index was greater than 35 for those with related comorbid conditions, and BMI was greater than 40 for those without comorbid conditions. The investigators categorized nonsurgical control patients in accordance with the following four BMI categories: less than 25, 25-29, 30-34, and greater than or equal to 35. Each control group, as well as the surgical group, included 13,852 women.
Participants in the surgical group were followed for 5 years after bariatric surgery. Those in the nonsurgical group were followed for 5 years after the index date (that is, the date of BMI measurement).
In the overall population, 659 cases of breast cancer were diagnosed in the overall population (0.95%) during the study period. This total included 103 (0.74%) cancers in the surgical cohort; 128 (0.92%) in the group with BMI less than 25; 143 (1.03%) among those with BMI 25-29; 150 (1.08%) in the group with BMI 30-34; and 135 (0.97%) among those with BMI greater than or equal to 35.
Most cancers were stage I. There were 65 cases among those with BMI less than 25; 76 for those with BMI of 25-29; 65 for BMI of 30-34; 67 for BMI greater than or equal to 35, and 60 for the surgery group.
Most tumors were of medium grade and were estrogen receptor positive, progesterone receptor positive, and ERBB2 negative. No significant differences were observed across the groups for stage, grade, or hormone status.
There was an increased hazard for incident breast cancer in the nonsurgical group, compared with the postsurgical group after washout periods of 1 year (hazard ratio, 1.40), 2 years (HR, 1.31), and 5 years (HR, 1.38).
In a comparison of the postsurgical cohort with the nonsurgical cohort with BMI less than 25, the hazard of incident breast cancer was not significantly different for any of the washout periods, but there was a reduced hazard for incident breast cancer among postsurgical patients than among nonsurgical patients in all high BMI categories (BMI ≥ 25).
“Taken together, these results demonstrate that the protective association between substantial weight loss via bariatric surgery and breast cancer risk is sustained after 5 years following surgery and that it is associated with a baseline risk similar to that of women with BMI less than 25,” the investigators write.
Nevertheless, Dr. Doumouras said “the interaction between the surgery and individuals is poorly studied, and this level of personalized medicine is simply not there yet. We are working on developing a prospective cohort that has genetic, protein, and microbiome [data] to help answer these questions.”
There are not enough women in subpopulations such as BRCA carriers to study at this point, he added. “This is where more patients and time will really help the research process.”