On April 2 2009, the FDA had an Advisory Committee meeting to discuss Novo-Nordisk's application to sell their new diabetes drug, liraglutide. The FDA routinely turns to Advisory Committees for input about the safety and/or efficacy of drugs they are reviewing, but then the FDA makes the final decision about whether a new drug should be approved after mulling over the Advisory Committee's comments as well as reviewing all the data provided by the drug's manufacturer. The FDA always has two fundamental concerns before approving a new drug: does the drug work, and is the drug safe.
Liraglutide, which has the proposed brand name Victoza, belongs to the class of diabetes drugs that has been called either "glucagon-like peptide-1 (GLP-1) analogs," or "incretin mimetics." These drugs work by enhancing insulin secretion by the pancreatic beta cells, suppressing glucagon secretion, and slowing gastric emptying. In addition to exenatide (brand name Byetta), which is already available, other similar compounds are under development by other companies, including albiglutide and taspoglutide, and perhaps others. Liraglutide, if approved, would have one advantage over Byetta: it only needs to be given once daily whereas Byetta requires a twice-daily administration.
The first GLP-1 analog to be approved by the FDA, Byetta, came under scrutiny about a safety issue after it was approved, because of a concern that it might increase the risk of pancreatitis: I've written about this several times previously, and I have wondered if an increased risk of pancreatitis might be a "class effect" of this group of drugs. Novo Nordisk found during studies of liraglutide that the rate of pancreatitis was 2.2 in liraglutide-treated subjects and 0.6 in nonliraglutide subjects per 1,000 years of subject exposure. They indicate that they plan to put a safety statement about pancreatitis into liraglutide's label.
At the recent Advisory Committee meeting, FDA's asked the panel two other questions about liraglutide's safety: one about cardiovascular safety, and one about thyroid tumors.
The Advisory Committee felt that liraglutide was safe from a cardiovascular standpoint and voted 8 to 5 that liraglutide does not increase cardiovascular risks. That's a split decision, and may lead the FDA to propose further studies of this risk; Novo-Nordisk has indicated willingness to set up a cardiovascular outcome trial.
But the much bigger question, in my mind, was the FDA's concern about thyroid tumors. There were findings of an unusual form of thyroid tumor in rodent studies of liraglutide, and thyroid cancers were seen in humans in liraglutide's clinical trials. Rodents, both rats and mice, developed benign (non-cancerous) thyroid C-cell tumors (AKA adenomas) and had increased levels of a hormone made in the C-cells, calcitonin. The FDA also
pointed out that in clinical studies of liraglutide, five patients taking liraglutide developed papillary thyroid cancer, whereas there was only one case among those not receiving the drug. The FDA said most of the reported thyroid cancers were "very small" and that small thyroid tumors are commonly found in the general population. Novo Nordisk, in their
briefing document for the Advisory Committee, said that "the proliferative C-cell findings represent a rodent specific phenomenon." The committee, however, was worried, voting 6 to 6 (with one member abstaining) that the cancer data in animals were sufficient not to recommend approval.
Novo Nordisk claimed that "data and analyses do not suggest any treatment-related effect of liraglutide" on the development of neoplasms; the Advisory Committee agreed, unanimously supporting the approvability of liraglutide with regard to risk of papillary thyroid cancer.
The clear difference of opinion among the members of the Advisory Committee about the findings in animals and whether they apply to humans might be enough to persuade the FDA to hold off on approval of liraglutide, or if the FDA does decide to approve it, to add a demand in the label that physicians should obtain thyroid testing, either blood testing for calcitonin, or thyroid ultrasound studies. The FDA might also demand that more studies will be needed to be done by Novo Nordisk to investigate the thyroid situation further, and demand the studies before approving the drug, or afterwards.
As this is being written, we don't yet know whether the FDA will decide to approve liraglutide, and if so, with what conditions. Since it's a "me-too" drug, with no clear advantage over Byetta except once-daily dosing, it's not really a big deal either way. Unless you own stock in Novo Nordisk, of course. Or unless it's approved and you later get thyroid cancer.
[Editor's Note: Liraglutide was approved by the FDA in January, 2010.]