Is Dapagliflozin Really Tied to Bladder Cancer?

November 9, 2015

Researchers sift through risk factors and clinical trial findings to determine if the SGLT2 inhibitor increases bladder cancer risk.

Whether dapagliflozin can increase the risk for bladder cancer has been an open question. Materials presented to the FDA during the dapagliflozin approval process suggested a signal for bladder cancer, but not for other SGLT2 inhibitors. On the other hand, diabetes itself has been linked to increased risk for bladder cancer, creating a chicken or egg scenario.

A recent pooled analysis of 21 clinical trials, though, has suggested no increased risk for bladder cancer with dapagliflozin. The study was published in Diabetes Therapy, and was sponsored by Astrazeneca, the maker of dapagliflozin.

In the study, researchers did a pooled analysis of 21 phase 2b/3 clinical trials of dapagliflozin (>9000 patients, up to 4 years’ duration). Patients on dapagliflozin (n = 5936) had a mean duration of diabetes of 7.0 years, mean baseline HbA1c 8.21%, and 43.3% had a history of smoking. Those on placebo (with or without background medication) or active control (n = 3403) had a mean duration of diabetes of 7.6 years, mean baseline HbA1 of 8.14%, and 46.3% had a history of smoking.

Key Results:

• Bladder cancer cases:

♦ 9/5936 dapagliflozin-treated patients,

♦ 1/3403 comparator-treated patients

♦ IRR of 5.168 (95% CI: 0.677, 233.55)

• All bladder cancer cases had clinical factors associated with the disease:

♦ All were ≥60-year-old males

♦ 8/10 were current/former smokers

♦ 8/10 had microscopic hematuria before or within 6 months of starting treatment

The authors pointed out several findings that suggest dapaglflozin is not linked to increased risk for bladder cancer. First, dapagliflozin can increase the risk of urinary tract infections (UTIs), and one theory holds that UTIs may contribute to bladder cancer. The authors noted, though, that 7/10 patients with reported bladder cancer did not have urinary events like UTIs before diagnosis. Treatment-related UTIs could also lead to more testing in patients on dapagliflozin, increasing the likelihood for detecting pre-existing bladder cancer in this group. In addition, all cases were reported within two years of starting dapagliflozin, a relatively short period of time given the latency period for bladder cancer, they pointed out. Finally, there was no detailed workup for hematuria before randomization or during dapagliflozin treatment, so the trials could have enrolled patients with pre-existing bladder cancer.

Preclinical studies also did not support the idea that increased urinary volume may contribute to bladder cancer, and showed that SGLT2 receptors are not expressed in bladder tissue.

“[T]he totality of evidence to date does not suggest a causal relationship between dapagliflozin and bladder cancer, and the observed effect in clinical trials may be a chance finding,” first author Agata Ptaszynska, MD, of Bristol-Myers Squibb, and colleagues, “However, given the observed imbalances, any firm conclusion cannot be made without additional clinical data, and as such accruing bladder cancer events are being carefully investigated in ongoing trials of dapagliflozin.”

Take Home Points

• A pooled analysis of 21 clinical trials has suggested that dapagliflozin is not linked to increased risk for bladder cancer.

• All bladder cancer patients had clinical risk factors for the disease.

• Because a number of explanations could account for the higher number of bladder cancer cases among patients treated with dapagliflozin, ongoing trials are looking at this issue.

Dr. Cohen reports consulting for AstraZeneca and Bristol-Myers Squibb (dapagliflozin), Johnson & Johnson (canagliflozin) and Boehringer Ingelheim (empagliflozin). Drs. Ptasynska and Reilly are employees of Bristol-Myers Squibb. Drs. E Johnsson and K Johnsson are employees of AstraZeneca.

Reference: Ptaszynska A, et al. Assessing bladder cancer risk in type 2 diabetes clinical trials: the dapagliflozin drug development program as a ‘case study’. Diabetes Ther. 2015 Sep; 6(3): 357-375.