A group of researchers assessed the association between Glucagon-like peptide 1 receptor agonists (GLP-1RAs) (drugs that mimic GLP-1 hormone to lower blood sugar) and the risk of thirteen obesity-associated cancers (OACs) in patients with type 2 diabetes (T2D), compared to those prescribed insulin or metformin.
Study
The study utilized the TriNetX platform to access deidentified electronic health records (EHRs) of 113 million patients from 64 healthcare organizations across the United States (US), representing diverse demographic groups. The platform supports patient-level analyses and has been employed in various retrospective cohort studies.
EHR data included demographics, diagnoses, medications, procedures, laboratory tests, genomics, visits, and socioeconomic information, with oncology-specific data from cancer registries. Self-reported sex, race, and ethnicity data were standardized, with missing values marked as "unknown."
The study included 1,651,452 T2D patients without a history of the 13 OACs, prescribed GLP-1RAs, insulin, or metformin between March 2005 and November 2018. Patients were categorized into exposure (GLP-1RA/no insulin or GLP-1RA/no metformin) and comparison groups (insulin/no GLP-1RA or metformin/no GLP-1RA). Each OAC was examined separately in propensity-score-matched groups for relevant covariates. Follow-up continued until outcome occurrence, death, loss to follow-up, or 15 years.
The 13 OACs included esophageal, breast, colorectal, stomach, kidney, ovarian, endometrial, gallbladder, pancreatic, thyroid cancers, hepatocellular carcinoma (a type of liver cancer that originates in the liver cells), meningioma (a tumor that forms on the membranes covering the brain and spinal cord), and multiple myeloma (a cancer of plasma cells in the bone marrow). Cox proportional hazard and Kaplan-Meier survival analyses compared event rates, calculating hazard ratios (HRs) and 95% confidence intervals (CIs). Statistical analyses were performed using the platform's built-in functions in R, Python, and Java. Data collection and analysis concluded on April 26, 2024.
Findings
The study analyzed 1,651,452 patients with T2D, with a mean age of 59.8 years, including 827,873 men and 775,687 women. The racial composition included 0.4% American Indian or Alaska Native, 4.0% Asian, 17.0% Black, 0.8% Native Hawaiian or Other Pacific Islander, and 60.6% White participants. To compare GLP-1RAs with insulins, 1,093,728 patients without a prior diagnosis of OACs were included. The GLP-1RA/no insulin group consisted of 48,983 patients, while the insulin/no GLP-1RA group had 1,044,745 patients. The GLP-1RA group was younger, had a higher proportion of women and White participants, and had higher rates of family history of cancer, obesity, cancer screening, and prior use of other antidiabetic agents.
The analysis showed that GLP-1RAs were associated with a significantly lower risk of 10 out of 13 OACs compared to insulins. These included gallbladder cancer (HR, 0.35), meningioma (HR, 0.37), pancreatic cancer (HR, 0.41), hepatocellular carcinoma (HR, 0.47), ovarian cancer (HR, 0.52), colorectal cancer (HR, 0.54), multiple myeloma (HR, 0.59), esophageal cancer (HR, 0.60), endometrial cancer (HR, 0.74), and kidney cancer (HR, 0.76). Although the HR for stomach cancer was less than 1, it was not statistically significant (HR, 0.73). GLP-1RAs did not significantly affect the risk of postmenopausal breast cancer or thyroid cancer. The cumulative incidences of colorectal and liver cancers were lower in the GLP-1RA group compared to the insulin group. The mean follow-up time for colorectal cancer was approximately 2075 days for the GLP-1RA group and 1982 days for the insulin group, while for liver cancer, it was about 2023 days for the GLP-1RA group and 2038 days for the insulin group.
In comparing GLP-1RAs with metformin, the study included 888,525 patients without a prior diagnosis of OACs. The GLP-1RA/no metformin group had 32,365 patients, and the metformin/no GLP-1RA group had 856,160 patients. Propensity-score matching was used for each OAC outcome. GLP-1RAs were not associated with a lower risk of colorectal cancer, gallbladder cancer, and meningioma compared to metformin but were linked to an increased risk of kidney cancer. The cumulative incidences of colorectal and liver cancers showed no significant difference between GLP-1RAs and metformin. The mean follow-up time for colorectal cancer was approximately 1967 days for the GLP-1RA group and 2102 days for the metformin group, while for liver cancer, it was about 1971 days for the GLP-1RA group and 2130 days for the metformin group.
Conclusion
To summarize, analyzing over 15 years of EHRs from a US cohort of over 100 million individuals, researchers found that GLP-1RAs significantly reduced the risk of 10 out of 13 OACs in patients with T2D compared to insulins. These included esophageal, colorectal, kidney, pancreatic, gallbladder, ovarian, endometrial, and liver cancers, as well as meningioma and multiple myeloma. Compared to metformin, GLP-1RAs showed non-significant risk reductions for some OACs but increased risk for kidney cancer.
Source:
https://www.news-medical.net/news/20240708/GLP-1-drugs-found-to-lower-cancer-risk-in-diabetes-patients.aspx