On July 19th an FDA Advisory Committee meeting voted 9, 6 against recommending approval of dapagliflozin, Bristol-Myers Squibb/AstraZeneca's first-in-class SGLT-2 inhibitor for type 2 diabetes. Although the FDA have yet to make their final decision this is slated to happen on October 28th we expect them to follow the Advisory Committee's advice and to issue a Complete Response Letter for this drug.
Dapagliflozin which works by suppressing glucose reabsorption by the kidneys, thereby resulting in urinary excretion of excess plasma glucose appeared promising in clinical trials because its novel mechanism of action meant that it could be used in combination with other antidiabetic drugs without increasing the risk of hypoglycaemia, a key side effect associated with antidiabetic drugs. It also promotes modest weight loss a distinct benefit in a patient population that is typically overweight. Experts that we have interviewed for our Type 2 Diabetes Pharmacor report, and primary research that we have done as part of our Physician and Payer Forum reports on type 2 diabetes, indicate that dapagliflozin is likely to be used as a second or third-line oral therapy in patients who wish to delay initiating insulin therapy; combination use with insulin is also likely.
The side effect issue most talked about during dapagliflozin's clinical development was an increase in the rate of urinary tract infections, and genital tract infections (such as thrush). The latter were particularly prevalent in women, occurring in as many as 1 in 5 women taking dapagliflozin (and 1 in 12 men) in phase III trials. Although easily treatable with topical antibiotics, such infections are likely to involve an extra trip to the physician, and are clearly uncomfortable for the patient involved.
However, we did not expect this issue to delay dapagliflozin's approval, and the drug otherwise appeared to be safe and well tolerated. This view changed when a poster at the American Diabetes Association's 71st Scientific Sessions presented a pooled analysis of data from all of dapagliflozin's phase IIb/III trials. The findings from this study were further elaborated on in the briefing materials released prior to the Advisory Committee meeting, which noted that among 4310 subjects treated with dapagliflozin, nine developed bladder cancer, compared to one subject in the control group of 1962 patients. Similarly, nine cases of breast cancer were reported among patients taking dapagliflozin compared to no cases in the control group. It is noteworthy that all of these cases of cancer appeared within two years (most within one year) of treatment initiation, questioning a causal link. Furthermore, of the patients who developed bladder cancer, 6 had reported blood in their urine at the start of the trial, suggesting that the process of malignancy may have already started. Dapagliflozin's sponsors also noted in their presentations to the Advisory Committee that the question of dapagliflozin causing bladder or breast cancer was biologically implausible, and that no carcinogenicity signals had been seen in two-year rodent studies; normally the gold-standard for identifying potentially carcinogenic drugs.
Despite the small number of cases, and the factors questioning a link, the recent controversy over pioglitazone (Takeda's Actos) and bladder cancer means it was particularly unfortunate that dapagliflozin's Advisory Committee meeting had to have a section devoted to a discussion of the potential risk of bladder (and breast) cancer. A recent example showcases the potential for Advisory Committee members views to be colored by recent events: in July 2010 the Advisory Committee convened to discuss obesity drug phentermine/topiramate (Vivus Qnexa) voted not to approve the drug; it was apparent from discussion at the meeting that the negative sentiment towards this drug was shaped in part by concerns highlighted during the preceding days of rosiglitazone (GlaxoSmithKline's Avandia)'s safety.
Cancer wasn't the only issue to be raised as a specific safety concern at the Advisory Committee meeting. One patient taking dapagliflozin in the trials developed changes in liver enzymes indicative of potential severe liver injury (Hy's Law) that was judged likely to have been caused by dapagliflozin; a number of other cases were deemed unclear as to whether they were related to dapagliflozin administration. Liver toxicity is the single most common cause for safety-related drug marketing withdrawals in recent years. Although a single case of Hy's Law is unlikely to be sufficient to scupper dapagliflozin's chances of approval, it does mean that regulatory agencies are likely to want more data, to assess the potential risk better. And if it is approved, liver function tests will almost certainly be required something that neither patients nor physicians appreciate having to do.
Considering these concerns, it was not altogether surprising that the Advisory Committee voted that the benefits of the drug outweighed the risks, based on current evidence.
But what now? Considering that the risks of cancer and liver injury are far from certain, we anticipate that the FDA will follow the recommendations outlined by the CDER's Office of Surveillance and Epidemiology, namely that patients in the dapagliflozin trials continue to be followed-up, and further analysis of the data is performed, to further clarify the risks associated with the drug. We estimate that this will result in a delay of 1 to 2 years before dapagliflozin is approved in the U.S. Approval is also almost certain to be contingent on close monitoring of post-marketing data, and the implementation of a Risk Evaluation Mitigation Strategy (REMS). This delay and likely post-marketing regulatory requirements will have a significant impact on the uptake of dapagliflozin, not only because physicians and patients will be concerned about the safety of this drug, but because with canagliflozin (in phase III development with Johnson & Johnson) snapping at its heels, dapagliflozin may well lose its lucrative first-in-class status.
What is not clear, though, is if the problems faced by dapagliflozin are unique to the compound, or if they may affect the drug class as whole. It is worth noting that a large number of SGLT2 inhibitors in early stages of clinical development have been discontinued mostly for undisclosed reasons in recent years. It is easy to speculate that some of them were discontinued because of early signs of one of these safety concerns. Regardless of this, the developers of other SGLT2s remaining in development (such as Astellas ASP-1941 and Boehringer Ingelheim/Eli Lilly's empagliflozin) will be taking a long, hard look at the discussion that went on at this meeting, and at the resulting outcome when the FDA makes its decision in October. Should any particular trial requirements become apparent (such as including adjudicated cancer endpoints), then SGLT2 inhibitors that have yet to enter Phase III development may be at an advantage because the trials can be designed appropriately; those already in the final developmental stages may have to undergo additional trials.