Blood (sugar), sweat, and tears: yet more work needed to get Novo Nordisk’s Tresiba approved.
Novo Nordisk has announced that the FDA has issued a Complete Response Letter (CRL) for its basal insulin analog Tresiba (insulin degludec) and franchise product Ryzodeg (insulin degludec + insulin aspart), denying marketing authorization despite a positive opinion from the FDA advisory committee. Novo Nordisk’s share value fell by 15% on the morning of February 11, 2013 as financial losses are expected in the light of this development, which could result in a launch deferral of 3–4 years or more. This delay will cost Novo Nordisk dearly, with an estimated $5bn missing from its balance sheets until 2017.
Despite already receiving approval in the EU and Japan, Tresiba’s US approval process had previously been held up by the FDA, which in July 2012 requested an additional post hoc analysis of major adverse cardiovascular events (MACE). This came after Novo Nordisk had submitted its New Drug Application (NDA), delaying the FDA’s regulatory decision until November 2012, and finally until February 2013. The likely reason for demanding a new analysis and postponing the verdict was imbalances in the risk of myocardial infarction with ST elevation, which was 3.5-fold higher in the Tresiba/Ryzodeg group with respect to the comparator group (0.21% vs 0.06%) in the initial prespecified primary meta-analysis.
Unfortunately, re-analyzing the data and including more trials, as per the regulatory body’s request, did not work in Novo Nordisk’s favor. The FDA asked that unstable angina pectoris events be excluded from the new analysis in order to avoid diluting a potential effect of insulin degludec on the risk of myocardial infarction. Moreover, the FDA also required the MACE inclusion period to be extended from 7 to 30 days after drug discontinuation.
Overall, this led to a MACE risk ratio of 1.614 in the extended new meta-analysis, meaning that patients in the Tresiba/Ryzodeg arms of the various trials appear to have had an approximately 1.6-fold higher risk of suffering a major adverse cardiovascular event compared with patients in the comparator arms. A result like this is bad news.
In the EU and Japan, the regulatory authorities looked at the trial results with less skepticism and presumably put the higher MACE risk in the Tresiba cohort down to chance, which Datamonitor believes is a more logical move than the FDA’s overly cautious reaction of rejecting Tresiba. In terms of absolute numbers, the differences in MACE are relatively small, and it has to be taken into account that patient years of exposure were quite different in the insulin degludec vs comparator groups. While this comparison is not invalid, it could be partly responsible for the differences in risk between groups. The FDA appears to have adopted a “better safe than sorry” attitude, which has major implications for Novo Nordisk, and the agency has demanded that a cardiovascular outcomes trial be conducted to ascertain Tresiba’s safety.
Datamonitor believes that the dedicated cardiovascular outcomes trial requested by the FDA will last up to 6 years, with interim results 18–24 months after trial initiation. Therefore, Tresiba’s product launch in the US will happen significantly later than planned, and Datamonitor anticipates a new launch date in the first half of 2017, assuming that post-marketing data from the EU and Japan and the interim analysis show no increase in major adverse cardiovascular event risk.
Datamonitor expects that this delay is going to cost Novo Nordisk approximately $5bn in revenues that will be missing from its balance sheets in the period from 2013 to 2017, and this grim outlook was reflected in a 15% drop of Novo Nordisk’s share value on the morning of February 11, 2013. News from generics manufacturer Mylan and biosimilars manufacturer Biocon that they are going to produce biosimilar insulin glargine, lispro, and aspart is not going to help lift Novo Nordisk’s share price, as competition in 2017 will be fiercer. As a consequence, Datamonitor expects not only a “right shift” of Novo Nordisk’s earnings curve for Tresiba and sister compound Ryzodeg, but also believes that market share and therefore peak sales will be reduced, with additional competition coming from Eli Lilly’s novel basal analog, which is currently in Phase III trials. The CRL is therefore certainly not good news for Novo Nordisk, which might have to adjust its revenue growth rate from a double-digit figure to a single-digit one for the next few years.