Dapagliflozin, a New Drug for T2D, Rejected by an FDA Advisory Committee   (July 24, 2011)

On July 19, 2011, the FDA's Endocrinologic and Metabolic Drugs Advisory Committee met and discussed whether to approve a new drug for type 2 diabetes (T2D), dapagliflozin.  The FDA frequently calls upon its Advisory Committees ("AdComms") to review the scientific data presented by the manufacturer, and to advise the FDA about whether to approve the drug. The FDA usually follows the advice of the AdComms, but not always. The agency is scheduled to make their decision on dapagliflozin by October 28.

Dapagliflozin is the first drug in the class of sodium-glucose co-transporter 2 (SGLT2) inhibitors to be presented to the FDA for marketing approval in the United States. Dapagliflozin and the other SGLT2 inhibitors work by a novel mechanism of action: they cause loss of glucose into the urine, and thereby decrease both blood glucose and A1C. I wrote about them previously (SGLT2 inhibitors -- an upcoming class of medications for diabetes).

The data for dapagliflozin was submitted by the co-developers (Bristol-Myers Squibb and AstraZeneca) and "included data of up to two years in duration and involved approximately 6,000 individuals in 40 clinical studies. The trials were designed to evaluate the safety, tolerability and efficacy (as measured by HbA1c) of dapagliflozin in the approximately 4,200 patients who received dapagliflozin and were at various stages of the continuum of type 2 diabetes. In accordance with FDA guidelines, the New Drug Application also included data assessing the cardiovascular safety of dapagliflozin in adults with type 2 diabetes" (quoted from a press release).

The clinical trials showed that dapagliflozin increases the risk of infections in the urinary tract and genitals, presumably as the extra sugar in the urine makes the area more hospitable to microorganisms.  In women, the most common such event was vulvovaginal fungal infection. In men, pruritus (itching) was the most common event. The huge surprise in the safety data, however, was that breast and bladder cancers were found at higher rates in patients taking dapagliflozin. There were nine (0.2% of the total population, 0.4% of the female population) patients in the dapagliflozin group and none in the control with breast  cancer. There were a total of nine cases of bladder cancer in dapagliflozin-treated patients versus one in patients treated with control. These numbers are obviously small, making it hard to draw definitive conclusions. And the manufacturers argued that many of the cancers occurred too soon to have been caused by the drug.

The FDA asked the AdComm to review five questions about dapagliflozin: (1) what to do about the problem of reduced efficacy of dapagliflozin in patients with renal disease, (2) whether there should be more studies of the risk of liver injury from dapagliflozin, (3) the risk of breast and bladder cancer from dapagliflozin, (4) other safety findings, including risk of genitourinary infections and bone safety concerns, and (5) whether the data support approval of dapagliflozin for T2D.

The AdComm voted 9 to 6 that dapagliflozin should not be approved because of these safety concerns. News stories since the vote indicate that the committee members agreed that more study of the possible cancer risks and other safety questions would be needed, and most of the panelists thought that the additional studies needed to evaluate these risks should  be done before approval, even though that might delay approval by years.  Some members of the advisory panel thought that teasing out those possible safety problems would be best accomplished by using post-marketing studies and patient registries. It was also commented that the clinical trials did not include enough minorities and very old patients, two groups that are prone to developing T2D.

What will the FDA do? First, I'd expect that they will take longer than the required deadline to make their decision. October 28 is the putative deadline, but the FDA has procrastinated on decisions on other diabetes drugs, and probably will do so again with this hot potato.

The easy way out for the FDA is simply to accept the recommendation of the majority of the AdComm, and delay approval of dapagliflozin until more studies are done. The studies would be expensive and of long duration, and dapagliflozin would not be approved for many many years, if ever. It would also cast a long shadow on other SGLT-2 inhibitors now in development: to assure the regulators that they don't have equivalent problems with their versions of this class of medications.

The alternate path for the FDA is to approve use of dapagliflozin with multiple restrictions (for instance, a strongly worded warning not to use in women prone to breast cancer), as well as demands for more post-approval studies.

The FDA recently put another diabetes drug on super-severe restrictions (Avandia), allowing patients already on Avandia to continue on it, but essentially disallowed new starts, restricting new use to "patients not already taking rosiglitazone who are unable to achieve glycemic control on other medications and, in consultation with their health care professional, decide not to take pioglitazone for medical reasons." Whether a new patient or a patient already on Avandia, the physicians need to document in the medical record that the patient has been informed of "complete risk information."

Depending on how the Avandia restrictions are working out, the FDA might choose to approve dapagliflozin with somewhat similar restricitions. That would allow the manufacturers to make a little money selling their product, but the restrictions would mean that dapagliflozin would never become a blockbuster as every manufacturer hopes.

There's a third and intriguing possibility: the FDA might not approve the use of dapagliflozin, but European and/or other health authorities might approve it. This disconnect has happened before with other diabetes drugs: for example, vildagliptin (Galvus), a DPP-4 inhibitor, wasn't approved in the US because of concern about skin lesions and kidney problems seen in animal studies, but it is approved in Europe and Australia.

I'll guess that the FDA will not approve dapagliflozin, but that's only a guess. We'll have to wait and see.

[Editor's Note: As of mid-March, 2012, the FDA still had not acted to approve or deny the NDA for dapagliflozin.]