NICE Snubs Novo With Draft Lira Guidance

You'd think that even NICE would have welcomed with open arms a new, effective treatment option for the UK's nearly 3 million diabetic patients--particularly one that not only lowers blood sugar, but also causes weight loss.

Nah. NICE is playing hardball (again). Its draft recommendation for Novo Nordisk's GLP-1 analog liraglutide (Victoza), although not nearly so damning as last week's preliminary guidance around Sprycel and Tasigna, suggests that the drug be used only in limited circumstances by the National Health Service. 'Limited' means a) as part of triple therapy regimens only (that is, for patients already on metformin and a sulfonylurea, or metformin and a thiazolidinedione, where the above combos aren't quite doing the trick), b) among patients with a body mass index equal to or higher than 35kg/m2 (that's high) and c) only at the lower, 1.2mg daily dose.

Ok, so patients that aren't quite so obese as 35kg/m2 could also qualify for the drug, "if it is considered that the drug's use could help to achieve levels of weight loss that could be beneficial in treating other conditions caused by being obese", the NICE press release concedes.

But a quick read of the full appraisal consultation document reveals that even among those tightly-defined patient groups, drug treatment should only be maintained if the individual loses at least 3% of body weight after six months (that's not far off the threshold for an actual weight-loss drug) and sees a reduction in blood sugar levels of at least 1 percentage point.

It seems, at first glance, to be a case of "damned if you do, damned if you don't" for Novo. After all, weight loss is supposed to be a nice side-effect of the diabetes drug--one that could lead to lower incidence of co-morbitidies--yet it's apparently being used to limit its reimbursement.

Although NICE concedes that liraglutide "may have some advantages over insulin...in particular its effect on weight," it appears to rule out the higher dose point blank, and is asking for further analyses on the cost-effectiveness of liraglutide on patients with a lower BMI, which it says were not presented.

NICE was also unsatisfied with the extent of the data comparing liraglutide in triple therapy with other combinations of oral drugs, including for instance DPP-4 inhibitors, and felt that the LEAD-1 trial comparing liraglutide to rosiglitazone used an insufficiently high dose of the glitazone.

It's impossible to cover all the various permutations and combinations in diabetes therapy, however--and Novo did submit data from six trials in over 4000 patients. All except one of the trials showed that the drug reduced HbA1c levels significantly better than comparators, which included rosiglitazone (Avandia), sitagliptin (Januvia, the DPP-4 inhibitor), insulin glargine (Lantus), placebo, and Lilly/Amylin's exenatide (Byetta, the only other GLP-1 out there). The exception: the LEAD-2 trial, comparing liraglutide with glimepiride (Amaryl), which showed up liraglutide's weight advantage although no significant difference in blood sugar lowering.

Perhaps NICE's conclusion isn't such a surprise, though, considering the fate of Byetta in the UK (which you can read about in NICE's Type 2 diabetes guideline document). In brief, Byetta's "not recommended for routine use in Type II diabetes", with the same six-month benefit hurdles as described above, deemed as "expensive" and "not cost-effective for an unselected population as compared to commencing human insulin therapy."

As such, Novo's VP Europe, Viggo Birch, admitted that "given what they [NICE] have done for exenatide, this is what you'd expect in the first round." Still, "I'm not happy with it," he continued. "We have a much better product [than exenatide] and much better data." Novo's looking for reimbursement as second-line therapy, "at least for important patient subgroups," and a relaxation of the BMI-based restriction, since this, Birch claims, "isn't fair; it was plucked out of the sky."

There's a crack in the door, though. NICE will hold another meeting on March 18, allowing Novo and others time to comment and submit further data supporting wider use of the drug. And in its full diabetes document, NICE does acknowledge, with regard to Byetta, some uncertainty as to whether GLP-1s would be deemed cost-effective if the [health economic] model "fully reflected the negative quality of life issues of insulin, including fear of hypoglycaemia, and the costs of support and patient education for modern intensity of insulin dose titration" and added that the "more obese require much higher insulin doses, such that insulin costs alone can easily exceed those of exenatide."

Still, reading between the lines, we suspect this preliminary appraisal for Victoza is a call not just for more analysis, but also for a cost-sharing scheme of some description to help smooth the drug's passage past the cost-effectiveness watchdog.

That's not likely, according to Birch. "It [a cost-share scheme] is not really on the radar for this product right now." Perhaps with reason: after all, Victoza's hardly the most expensive drug to cross NICE's desk--the low dose costs GBP 954 per year, compared to tens of thousands of pounds for some cancer treatments.

But given the growing prevalence of diabetes, "NICE is probably wary of giving free rein to something that [may become] so huge" comments one analyst.

The question is whether its hard-ball stance on Victoza ideally balances the concern over escalating drug costs, with that of the escalating costs of the disease itself (plus consequences), estimated to eat up about 10% of the health care budget.

image by flikrer Roland used under a creative commons license