FDA's Dapagliflozin Review To Focus On Hepatic Effects, Cancer Risks

The agency will ask its Endocrinologic and Metabolic Drugs Advisory Committee to weigh these "unexpected" safety findings despite a clean cardiovascular profile for Bristol-Myers Squibb/AstraZeneca's first-in-class diabetes drug.

Bristol-Myers Squib Co./AstraZeneca PLC's dapagliflozin seems to have skirted the cardiovascular safety concerns surrounding other agents intended for treating type 2 diabetes, but FDA is nevertheless seeking advisory committee input on a handful of "unexpected" safety issues for the first-in-class SGLT-2 inhibitor, including potential hepatic and cancer risks.

At a July 19 meeting, the agency will ask its Endocrinologic and Metabolic Drugs Advisory Committee whether sufficient evaluation as been conducted pre-approval to determine if dapagliflozin is associated with the risk of hepatotoxicity, according to agency briefing documents released on July 15.

FDA also seeks panel discussion on numeric imbalances in cases of breast and bladder cancer observed in the clinical program, and it asks the committee to weigh the clinical significance of other safety findings, including the increased risk of genital-urinary infections.

The agency has little quibble with dapagliflozin's efficacy, other than to say it is limited to patients with normal renal function or only mild impairment. However, FDA asks whether additional studies in special populations should be conducted to better characterize efficacy or whether renal function monitoring should be performed prior to and during treatment.

Novel Mechanism Of Action

Dapagliflozin is the most advanced member of the new class of oral sodium glucose cotransporter-2 inhibitors that cause renal elimination of glucose. The mechanism of action, which offers a way to lower blood glucose apart from insulin, has spurred predictions that the class would find its niche as add-on therapy at different stages of the disease (Also see "New Data Help Size Up Novel Class Of SGLT-2 Inhibitors In Diabetes" - Pink Sheet, 5 Jul, 2010.).

Bristol-Myers Squibb and AstraZeneca are seeking a broad indication for dapagliflozin: as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes.

The NDA submission included data from three Phase IIb and 11 Phase III studies evaluating dapagliflozin as monotherapy, add-on therapy to metformin, sulfonylureas, pioglitazone (Takeda's Actos) or insulin, as well as initial combination therapy with metformin. Efficacy results were comparable to existing agents, FDA said.

"The magnitude of glycemic reduction in the clinical trials has been consistent and in line with recently approved anti-diabetic drugs, such as dipeptidyl peptidase inhibitors," the Division of Metabolism and Endocrinology Products says in its clinical briefing document. "In the active-controlled trials, dapagliflozin demonstrated equivalent effects on HbA1c as near maximally effective doses of metformin and glipizide at one year."

However, the agency found efficacy to be limited in those patients with compromised renal function. The sponsors have proposed that the drug not be used in patients with moderate renal impairment, with glomerular filtration rate (GFR) less than 45 mL/min/1.73m. But the agency cited a loss of statistically significant efficacy in patients with more mild impairment, with GFR less than 45-59 mL/min/1.73 m.

The loss of efficacy that accompanies an increase in renal impairment is particularly important for type 2 diabetics, FDA reviewers say.

"It is important to note that patients with T2DM are at risk for worsening renal function over the course of their disease. Unlike treatment with other classes of anti-diabetic medications that rely on either insulin secretion or sensitivity, dapagliflozin's effect is dependent on GFR and independent of beta cell function. Therefore, secondary failure of glycemic control with dapagliflozin may represent deterioration of renal function, rather than beta cell function, and discontinuation of dapagliflozin (in contrast to the practice of adding drugs of complementary effects) may be recommended."

Cardiovascular Profile Looks Good ...

Concerns about the post-marketing cardiovascular safety of another diabetes drug, GlaxoSmithKline's Avandia (rosiglitazone), spurred new FDA guidelines in December 2008 requiring sponsors to rule out an unacceptable level of increased CV risk pre- and post-approval for type 2 treatments.

Sponsors must rule out an 80% increased risk of major adverse cardiovascular events pre-approval. If pre-approval clinical data show that the upper bound of the two-sided 95% confidence interval for the estimated risk lies between 1.3 and 1.8, companies generally will need to conduct a post-marketing trial to demonstrate that the upper bound is less than 1.3. If the pre-approval data demonstrate an upper bound below 1.3, a post-marketing CV trial generally may not be necessary (Also see "FDA Diabetes Guidance Encourages Enrollment Of Sicker Patients" - Pink Sheet, 17 Dec, 2008.).

The guidance's provisions applied to all drugs pending FDA review and in clinical development at the time it was issued. The new CV requirements tripped up at least one anti-diabetic agent that had already been submitted for FDA review, Takeda's DPP-4 inhibitor alogliptin (Also see "Takeda's Alogliptin Suffers Major Setback, Delaying Drug Launch Until 2012 Or Later" - Pink Sheet, 29 Jun, 2009.).

However, it appears that dapagliflozin has easily overcome the pre-approval CV hurdle laid out in the guidance.

Based upon a meta-analysis of the Phase II/III trials, the drug "does not appear to be associated with excess CV risk with an overall HR of 0.67 (95% CI: 0.42-1.08) relative to comparators for the primary composite of CV deaths, myocardial infarctions, stroke and hospitalization for unstable angina," Division of Metabolism and Endocrinology Products Director Mary Parks writes in her memo to the advisory committee.

Although the upper bound of the 95% confidence interval was below 1.3, falling shy of the cut-point for a mandatory post-approval CV outcomes trial, FDA nevertheless has decided such a study will be required.

"The applicant has proposed to conduct a CV outcomes trial whose primary objective is to demonstrate a cardio-protective effect of dapagliflozin," Parks writes. "FDA concurs with the applicant that a CV outcomes trial is necessary to better characterize the CV safety of this drug and to evaluate safety issues that have arisen in the course of this NDA review. If approved, this CV outcomes trial will be a required post-marketing trial."

But Hepatic, Cancer Concerns Worry

Although dapagliflozin's pre-marketing CV profile looks clean, "several unexpected safety issues identified in this clinical development program were of sufficient concern to FDA to merit discussion of their impact on the overall benefit-risk consideration," Parks writes.

Among these are the drug's hepatic effects. There were a total of five dapagliflozin-treated patients in the Phase IIb/III trials that met the laboratory criteria for Hy's Law (AST or ALT greater than three times the upper limit of normal and elevation of total bilirubin greater than two times the ULN). One of these was classified as a "probable" case of drug-induced liver injury.

In their review, Office of Surveillance and Epidemiology hepatologists Leonard Seeff and John Senior noted that in many cases the data were too limited to make a linkage between dapagliflozin and liver damage. Given the importance of recognizing sentinel cases of drug-induced liver injury in registrational trials, "it is prudent to gather more information on all relevant cases as part of an in-depth review of the dapagliflozin NDA in order to assess whether this agent may be hepatotoxic," they say, adding that careful hepatic monitoring should be performed in any future studies.

FDA also calls the committee's attention to numerical imbalances in cases of breast and bladder cancer among dapagliflozin-treated patients.

There were a total of seven bladder cancer cases in dapagliflozin-treated patients at the time of the four-month safety update, compared to zero in the control group. Subsequently, two more cases were reported with dapagliflozin, and one with control. "This can be extrapolated to 207 cases per 100,000 person year exposure in dapagliflozin-treated patients versus 53 cases per 100,000 person year exposure in control," the clinical review states.

The findings suggest dapagliflozin could be in for the same type of bladder cancer cloud currently hanging over Actos (Also see "Actos Gets Stronger Bladder Cancer Warning, But Escapes Black Box" - Pink Sheet, 15 Jun, 2011.).

FDA questions whether detection bias resulting from dapagliflozin's mechanism of action and related genital-urinary adverse effects could have contributed to the imbalance observed.

"More frequent assessments of urinalysis in the dapagliflozin group might result in post-baseline detection of hematuria requiring further work-up and higher rate of cancer diagnosis than control group which might not have received as extensive monitoring," FDA reviewers state.

"In this regard, it is interesting to note that concerns of bladder cancer associated with pioglitazone use arose during pre-marketing development due to non-clinical findings of bladder cancer in male rodents. This led to a prospective assessment of urine cytology in approximately 1,800 patients in clinical trials up to one year duration. Despite similar active surveillance for bladder cancer in pioglitazone and control groups, no clinical cases were detected pre-marketing. Imbalance of clinical bladder cancer risk with pioglitazone was not observed until after approval."

There also was an imbalance in the number of breast cancer cases, with nine reported in the dapagliflozin-treated group and none in the control group.

For both types of cancer, FDA concludes that the numbers observed exceeded the expected number of cases in the type 2 diabetic population. It asks the advisory committee whether any imbalance in baseline factors may have contributed to the findings and whether detection bias was a factor.

FDA also wants the committee to consider the increase in genital and urinary infections associated with dapagliflozin. Such infections have become a well-known side effect of this class of drugs, given their mechanism of action.

-Sue Sutter ([email protected])