FDA panel advises against approval of dapagliflozin

The FDA's Endocrinologic and Metabolic Drugs Advisory Committee voted 9-6 that dapagliflozin, an investigational sodium-glucose co-transporter 2 inhibitor, should not be approved as an adjunct to diet and exercise for the management of type 2 diabetes.

Several panel members noted that more data are needed, in light of safety concerns linking dapagliflozin (Bristol-Myers Squibb, AstraZeneca) with breast and bladder cancers. Among 11 randomized, controlled, phase 3 clinical trials, researchers found nine cases of bladder cancer in patients receiving dapagliflozin compared with one case in control groups. Similarly high was the incidence of breast cancer, with five cases occurring in the treatment arm vs. one case control groups. A separate FDA review of the data linked dapagliflozin with a fourfold increased risk for breast cancer and fivefold increased risk for bladder cancer. One case of probable drug-induced liver injury that met criteria for Hy's Law also raised red flags.

"The issue of hepatotoxicity is always one that is a little concerning to me," committee member Doris B. Strader, MD, associate professor of medicine at the University of Vermont, said during the meeting. "This is only one case, but it does raise some issues about monitoring. …We can't dismiss breast and bladder cancer as minor or irrelevant."

Unanswered questions

Although research did not suggest that dapagliflozin increases risk for fracture or other negative effects on bone health, several panel members noted that results from short-term trials, ranging 24 weeks to 2 years, may not be long enough to properly assess this outcome. Possible adverse events such as dehydration and imbalanced nutrition were also discussed, although the panel generally felt these may not be significant. Researchers who studied dapagliflozin noted a doubling of genital infections, including vulvovaginitis and balanitis, and urinary tract infections among patients in treatment arms.

"The list of things needed for postmarketing surveillance was too long," panel member Edward W. Gregg, PhD, chief of the epidemiology and statistics branch, division of diabetes translation, CDC, said.

Findings raised uncertainty about the efficacy of dapagliflozin in patients with renal impairment as well. Data suggest that the drug was ineffective in patients with estimated glomerular filtration rates less than 60 mL/min. The committee questioned whether more accurate cut-off points for the drug's use could be established and how monitoring for efficacy should be conducted as renal function declines with a patient's age.

A randomized controlled trial involving at least 30,000 patients would be required to quell the committee's concerns, according to panel discussion.

Optimism remains

Among those committee members who favored approval of dapagliflozin, the consensus was that the adequate amount of data required to accurately assess a relationship between the drug and cancers and serious hepatic injury could not be collected in premarketing trials.

Gregg, along with several other committee members, however, pointed out that gathering this information, even premarket approval, may not take as long as expected.

"With data being collected now and with ongoing studies, perhaps a medium-sized trial, rather than a large trial, is enough to at least tell us whether the experiences from these data bases were aberrations, background noise or something else," Gregg said.

Overall, despite concerns, the advisory panel appreciated the novel approach to diabetes management that dapagliflozin offers. The fact that it is an oral agent that can be taken at any time daily and does not attempt to regulate insulin are all big bonuses, committee members said.Dapagliflozin is the first SGLT2 inhibitor to go before the advisory panel. Bristol-Myers Squibb and AstraZeneca plan to make the drug available in 5-mg and 10-mg doses. Canagliflozin (Johnson & Johnson), another agent in this drug class, has also undergone phase 3 clinical trials.

While the FDA is not required to follow the recommendations of the advisory committee, it usually does. - by Melissa Foster

The 6–9 vote by the FDA advisory panel regarding dapagliflozin was a vote against approval. But, the sentiment of the committee was more positive than the vote implied. Many of the advisors liked the drug profile and its ability to lower glucose and weight without excess risk for hypoglycemia. But, the ill-defined, fourfold to fivefold increased risks for breast and bladder cancer and the one case of probable drug-induced hepatic toxicity (by Hy's Law) gave pause to many that resulted in the negative vote for approval. Most panel members seemed to think that with more study, those risks could be better defined and the drug would be ultimately approvable. What seemed necessary to the advisors was a 30,000-patient, long-term, randomized controlled trial to define cancer and hepatic toxicity.

– Alan J. Garber, MD, PhD
Endocrine Today Chief Medical Editor

Disclosure: Dr. Garber reports no relevant financial disclosures.

Today, the bar is set very high in terms of getting FDA approval for a new drug. This plays a major role in the FDA's decisions regarding potential adverse effects, especially in light of the recent history of finding significant adverse effects with new and older drugs during the past 5 to 10 years.

Moreover, if you look at the specifics of the 9-6 vote, it is not surprising because a lot of emphasis was placed on the report of liver toxicity that occurred in one patient. Even more concerning was the 9-1 incidence of bladder cancer and 5-1 incidence of breast cancer in patients receiving dapagliflozin vs. control patients. Epidemiologists and statisticians may fault that interpretation because there were generally twice as many patients in the treatment groups compared with the control groups. Consequently, even if it looks like the incidence is 9-1 or 5-1, these numbers would probably be different if there were more controls. In addition, there is also a suggestion that some of the cancer may have occurred earlier than the actual period of treatment. Taking this into account, some discussion suggested that these statistics may be imperfect. However, it is always a concern when you have adverse effects, and with effects as serious as cancer and liver toxicity one has to be much more careful in determining whether they are real or not. I think the FDA panel was saying that these [effects] need to be looked at more seriously and will require more information before considering approval.

That said, I personally feel that this family of drugs - SGLT2 inhibitors - is very exciting, and I believe that they are going to be very valuable in the future. It is a new class of drugs and, therefore, can be used in combination with drugs that we already prescribe because we realize that some of the medications that we have used to date are showing some significant adverse effects.

- Derek LeRoith, MD, PhD
Endocrine Today Editorial Board Member

Disclosure: Dr. LeRoith is a consultant for Bristol-Myers Squibb and AstraZeneca.

The mechanism of action of dapagliflozin in inhibition of the renal transporter SGLT2 is attractive in that it allows noninsulin-mediated glucose-lowering and, hence, is less likely to cause hypoglycemia. The agent also has modest benefits in terms of weight loss and blood pressure control. However, the question of urinary infection has always been an issue with the glycosuria caused by this approach. In addition, during the FDA evaluation and the past 1 to 2 months, a number of other potentially serious problems with the drug were discovered. The agent appears to have markedly reduced efficacy with moderately decreased renal function and may have adverse renal effects in people with albuminuria but a normal or only mildly reduced glomerular filtration rate. Most worrisome is the newly released information about bladder cancer, breast cancer and liver function abnormalities. Dapagliflozin's parent compound, phlorizin, has been known to be associated with liver function abnormalities, but this agent has been thought not to exhibit such toxicity. The more than fourfold increase in breast and bladder cancer is also of concern.

Unfortunately, an immense clinical trial would be required to adequately assess these adverse events. Perhaps the manufacturers will be able to perform animal toxicity studies, specifically focusing on models of bladder cancer, breast cancer, liver disease and albuminuria, to address the questions of whether particular toxicity might be seen among people with certain characteristics. The SGLT2 inhibitors, then, seem to have a potentially useful mechanism of action, and therefore deserve further investigation. The FDA decision will not end the investigation of this class of drugs, but the committee's responsible recommendations do show how tremendously difficult it may be to bring out the promise of this approach.

- Zachary T. Bloomgarden, MD
Endocrine Today Editorial Board Member

Disclosure: Dr. Bloomgarden reports consulting arrangements with Boehringer Ingelheim, Bristol-Myers Squibb, Zimmer, Novo Nordisk, Pfizer and Eli Lilly.

Follow EndocrineToday.com on Twitter.