Canagliflozin Not Linked to Fractures in Large Database Study

Miriam E. Tucker

December 31, 2018

Canagliflozin (Invokana, Invokamet, Janssen) does not appear to raise fracture risk among patients with type 2 diabetes who aren't at risk of fracture to begin with, new research suggests.

Results from a retrospective analysis of two US commercial healthcare databases were published online December 31 in Annals of Internal Medicine by Michael Fralick, MD, of Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, and the University of Toronto, Ontario, Canada, and colleagues.

"Our results are most relevant to patients with diabetes who do not have other risk factors for fracture, such as older age or previous fracture. These results should be reassuring to patients and physicians who are considering the potential risks and benefits of canagliflozin," Fralick and colleagues write.

The drug, a sodium-glucose cotransporter type 2 (SGLT2) inhibitor licensed for use in type 2 diabetes, currently carries a US Food and Drug Administration (FDA) label warning about increased risk for bone fractures.

Although a recent systematic review of clinical trial data did not identify an association between canagliflozin and increased fracture risk, the included trials were mostly small and enrolled relatively young patients with type 2 diabetes who were not prescribed insulin and did not have high rates of diabetes-related complications.

However, a 56% increased fracture risk was seen in one of two of the pivotal trials of the Canagliflozin Cardiovascular Assessment Study (CANVAS) program. The 10,142 CANVAS participants were older, about half were prescribed insulin, and a majority had cardiovascular disease. Fractures occurred as early as 12 weeks after treatment and mostly affected the upper or lower limbs. However, no increased fracture risk was seen in the other CANVAS trial.  

"Our findings...raise the question of whether the increase in fracture risk reported in CANVAS is limited to patients with high baseline risk," Fralick and colleagues say.

In an accompanying editorial to the new study, William D. Leslie, MD, University of Manitoba, Winnipeg, Canada, and John T. Schousboe, MD, PhD, Park Nicollet Clinic and HealthPartners Institute, Bloomington, Minnesota, and University of Minnesota, Minneapolis, generally praise the use of glucagon-like peptide-1 (GLP-1) agonists as active comparators and the use of propensity-score matching to minimize baseline differences as part of the study design.

They called the study "a good example of how observational data can be exploited to good advantage."

However, Leslie and Schousboe also caution, "although these data may reassure healthcare providers that prescribing canagliflozin will not increase fracture risk in their patients with diabetes, caution may still be appropriate when using this agent in older patients who have high fracture risk, with particular attention given to hydration status and fall risk."

No Increased Fracture Risk Seen Across Analyses

Fralick and colleagues used data from two commercial insurance claims databases: the deidentified Optum Clinformatics Data Mart ("Optum") and IBM MarketScan ("MarketScan"), which together include healthcare data for more than 70 million Americans. The study population comprised 92,779 patients newly prescribed canagliflozin and 101,802 patients newly prescribed a GLP-1 agonist.

Propensity score matching yielded 23,458 pairs of new users of canagliflozin or a GLP-1 agonist in Optum and 56,506 pairs in MarketScan, matched for potential baseline confounders including age, gender, and comorbid conditions including fall-related factors. (HbA1c was not included because those data weren't available for a majority of patients in either database.)

The primary outcome was a composite endpoint of pelvic fracture, hip fracture requiring surgery, humerus fracture requiring intervention (surgery, casting, or splinting), or radius or ulna fracture requiring intervention.

In MarketScan, rates of the primary outcome in the propensity-matched cohort were 2.1 per 1000 person-years for canagliflozin versus 2.3 per 1000 person-years for GLP-1 agonist, respectively, a nonsignificant difference with hazard ratio of 0.92. Similarly, in Optum those rates were 2.6 versus 2.3 per 1000 person-years in the propensity analysis, and were also not significantly different (hazard ratio, 1.13).

Combined across the two datasets, the propensity score-matched rate of the primary outcome also didn't differ significantly (hazard ratio, 0.98). There was also no difference in sensitivity analyses not counting treatment discontinuation or limiting the results to patients older than 60 years, or in any secondary outcomes, including primary outcomes without intervention and fracture sites reported in CANVAS.   

Leslie and Schousboe write, "Although this study adds to the accumulating evidence that there is probably little if any risk associated with canagliflozin use among patients with diabetes who have low fracture risk, further investigations are needed to determine whether these results apply to persons with both diabetes and high fracture risk due to advanced age, very low bone mineral density, prior fracture, frailty, or a combination of these factors."

The study was primarily funded by Brigham and Women's Hospital, Division of Pharmacoepidemiology and Pharmacoeconomics. Fralick receives funding from the Eliot Phillipson Clinician-Scientist Training Program at the University of Toronto, the Clinician Investigator Program at the University of Toronto, Canadian Institutes of Health Research Drug Safety and Effectiveness Cross-Disciplinary Training Program, and Detweiler Traveling Fellowship funded by the Royal College of Physicians and Surgeons of Canada. Leslie and Schousboe have reported no relevant financial relationships.

Ann Intern Med. Published online December 31, 2018. Abstract, Editorial

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.