Last month, the annual American Diabetes Association Scientific Sessions was a busy time for Novo Nordisk; with almost three-quarters of revenue coming from their diabetes products, they had a lot of news to share and studies to publicize.
News in the clinical world, though, is different than news in the rest of the world, and indeed much of the research coming out of Novo Nordisk felt more like anti-news: “Dog bites man. All is normal.” In other words, the big releases regarding Novo Nordisk’s most promising new offerings—the GLP-1 agonist Victoza and the long-acting insulin Levemir—showed incremental improvement over existing treatments, helping more patients reach the magic HbA1c of 7.0%
The news and excitement over yet another drug that does a little bit better may seem strange from a practical standpoint, but it is crucial from a clinical and medical standpoint. Rare indeed is the polio vaccine that wipes a disease from the record suddenly and completely. Rather, medical advances often come in small steps, with an improvement here and a new mechanistic insight there.
The story of treating type 2 diabetes is much closer to the story of treating cancer than the story of the polio vaccine—a new drug comes along, and it seems to improve patient outcomes, but the change is small, and the drug must be tested and analyzed, compared at different doses and in different combinations. For chemotherapeutic agents and antidiabetic agents alike, this means multiple phases of randomized trials must be run, resulting in measurable outcomes that can be compared against a pre-defined primary endpoint. When the drug addresses a market that has no other hope, we often expect shocking results and accept some degree of risk. When the drug is a new entry into a market with a lot of other, well-tested, decently-functional options, as both Victoza and Levemir are, however, the drug company needs to show not only that the drug works, but also that it is sufficiently comparable to the existing ones in terms of safety and desired patient outcomes to be worth adding to the set of options.
So what did we hear from Novo Nordisk this year at the ADA conference?
Among the Novo-sponsored studies was an extension of a clinical trial comparing liraglutide (the generic name of the GLP-1 agonist Victoza) to sitagliptin (the DPP-4 inhibitor Januvia, from Merck). GLP-1 agonists are proteins of the incretin class that bind to GLP-1 receptors in the body; when the GLP-1 receptors are bound, they help slow food absorption in the gut, reduce appetite in the brain, and stimulate insulin release in the pancreas. DPP-4 inhibitors also act in a similar manner, but, instead of directly binding to the GLP-1 receptors, they stop the protein DPP-4 from degrading incretins like GLP-1. So, both liraglutide and sitagliptin target the same machinery inside the body, but from different angles and with different levels of GLP-1.
Dr. Bernard Zinman, from the University of Toronto, spoke about new results from the third phase of the trial that was designed to compare the two drugs. The first phase was a 26-week direct, head-to-head comparison of liraglutide and sitagliptin; the second phase was another 26-week extension under the same circumstances; and the third phase was a 26-week extension in which patients originally assigned to the sitagliptin arm were switched to treatment with liraglutide.
The first two phases had previously established non-inferiority [1], and went on to show that liraglutide resulted in a 0.6% greater drop in HbA1c as compared to sitagliptin. The news regarding the third phase was that there were additional incremental improvements seen for patients switched from sitagliptin to liraglutide. Though there were increased reports of transient nausea upon switching, patients switched to liraglutide saw an overall reduction in HbA1c (0.5%) and more were able to reach the target of an HbA1c below 7.0%. Further, there was marginally increased weight loss upon switching, and no marked increase in incidence of hypoglycemia.
In other words, some patients performed slightly better on Novo Nordisk’s Victoza than on sitagliptin. Similar results were achieved with a trial comparing Victoza to exenatide (another GLP-1 agonist, marketed as Byetta by Amylin, and administered twice daily as compared to once daily for Victoza). After an initial 26-week phase in which the two drugs were compared head-to-head, a second 14-week phase switched the exenatide patients to liraglutide. As with the sitagliptin study, of the 55% of patients who did not achieve the target HbA1c with exenatide, 32% were able to meet the goal of less than 7.0% after switching to liraglitude, with little change in the incidence of hypoglycemia or other adverse events.
Additional studies carried this initial theme— improvements were seen when patients were started on metformin and advanced to liraglutide and then eventually insulin as compared to subsets of all three drugs. Administering liraglutide helped manage weight gain associated with insulin use, and each additional treatment helped a further subset of the participants reach the target HbA1c.
Left alone with the Novo Nordisk news, then, I would find myself neutral; the studies seem promising, Victoza seems like a good enough drug, but these are all very incremental steps. A little bit better management, a little bit at a time.
What then makes Victoza matter? Why bother adding it to an already lengthy list of treatments for type 2 diabetes? Is marginal improvement in a clinical trial enough to warrant educating physicians and patients about yet another option?
As it turns out, the studies and clinical trials do not alone address the real benefits of Victoza and other products in the Novo Nordisk pipeline. To hear about those, I turned to Dr. Alan Moses, Global Chief Medical Officer of Novo Nordisk.
Prior to joining Novo Nordisk in 2004, Dr. Moses spent over 25 years in professional and academic endocrinology, including five years as the senior vice president and chief medical officer of the Joslin Diabetes Center. Given his years of experience and clinical insight, I asked Dr. Moses about the other half of the Victoza story, the half not included in the clinical results—what did patients think of Victoza? I expected positivity from Dr. Moses—he does, after all, work for Novo Nordisk—but I was surprised by the scale of his response; patients responded remarkably well to Victoza, and he had never seen such a spontaneous, positive response from patients to a drug. “This is one drug that based on anecdotal reports appears to work better in clinical practice than in clinical trials,” he said.
Some credit for the superior results in practice may belong to the doctors’ discretion in prescribing Victoza, but, according to Moses, the primary reason for the success of the drug with patients was the ease of use. Victoza must be injected, but unlike the other GLP-1 agonist on the market, Byetta, Victoza is a once-a-day injection, given irrespective of mealtime, easing the burden both physically and logistically for patients. A further advantage of Victoza over other options on the market that is large for patients but small for clinical trials, Dr. Moses noted, was the modest weight loss. As an aggregate figure in a study, the weight loss is moderate at best; but for individual patients, even a few kilograms feels significant.
This emphasis on ease of use is, notably, not an accident according to Moses. Victoza is indicative of the patient-centered ethos that has driven Novo Nordisk since its founding. The Nordisk Insulinlaboratorium, which would eventually become what we know as Novo Nordisk today, was established in 1923 by August Krogh, whose wife suffered from type 2 diabetes. After learning about Frederick Banting and Charles Best’s successful extraction of insulin from dog pancreases in 1922, Krogh started Nordisk in order to manufacture insulin for his wife and those like her with diabetes. What began as a literal marriage to the end-user has now become a metaphorical marriage, but the principle remains: “It’s about drugs that work, and it’s about drugs that work for the patient,” says Dr. Moses.
Ease for the patient thus drives much that’s in the pipeline for Novo Nordisk. Oral insulins and oral GLP-1 agonists are unfortunately still on the drawing board; Lotte Bjerre Knudsen, who led the team that created the GLP-1 agonist known as Victoza, explains that the difficulty of creating oral insulins and GLP-1 agonists is that they are just chains of amino acids—so when the proteins enter the gastrointestinal system, it’s just like a steak entering, and it will require a lot of molecular manipulation to protect the proteins from degradation and digestion.
Barring orals, then, Novo Nordisk is fast on its way to intermediate advances to make drug administration easier for patients. Trials have begun to test mixes of insulin and Victoza, which would minimize the number of injections patients on both drugs would require. The first challenge—physically mixing the two agents—has proven quite possible, as Novo Nordisk’s target insulin, the ultra long-acting insulin degludec, is very stable in solution, forming tight dihexamers which form soluble large multihexamers after injection, before they separate into absorbable insulin monomers very slowly after injection. And the second challenge—making sure patients dose and respond correctly to the mix—appears feasible, given how well patients have responded to the liraglutide and insulin administered simultaneously, but with separate injections. Further, even beyond ease of use, there is much promise for the combination treatment, as the post-prandial effects of Victoza have a tendency to smooth the curves of blood glucose, making it easier for patients to stay in range and avoid hypoglycemia.
But, wait—before we get carried away by the benefits, what about the risks I’ve heard about? More specifically, what about the thyroid cancer warnings?
Victoza currently carries a black-box warning as a result of the fact that it causes thyroid C-cell tumors at clinically relevant doses in rats and mice. To date, though, there has been no evidence that this is also the case for humans, so I asked Dr. Moses to what degree thyroid cancer is a concern, and if it is safe for patients with pre-existing thyroid problems like thyroiditis. The increased incidence of thyroid cancer, he explained, seemed to be a problem limited specifically to rodents, dependent on a biological pathway that is not present in primates. Novo Nordisk, in order to clarify the mechanisms that cause thyroid cancer in rodents, has just completed several new studies to be published soon. According to Novo Nordisk’s research, GLP-1 is a critical regulator of calcitonin in rodents. GLP-1 agonists (not just Victoza) stimulate the release of calcitonin in rodent thyroid C cells. Thus far, Novo Nordisk and others have found no evidence that the same is true in humans or other primates [3]. That doesn’t prove, of course, that there is no risk of thyroid cancer in humans, and so the warning in the label will likely remain, but it does give support to the belief that this is a problem unique to rodent biology, and not a threat to humans.
What, then, about the more subtle threat of beta-cell stimulators like liraglutide, studies show that type 2 diabetes begins with insulin resistance, but progresses towards increased beta-cell stress, toxicity, and eventually death [3]. Might a stimulating agent like GLP-1 cause endoplasmic reticulum stress in the beta-cells, resulting in a faster track towards the death of the cells?
Unfortunately, Novo Nordisk does not yet have evidence from long-term clinical trials that this does not happen; however, what they have been able to learn from in vitro (that is, on a cell plate or in a test tube) and animal studies indicates that overall there is actually a positive effect on beta cells. According to Dr. Moses, because GLP-1 agonists stimulate insulin only in response to glucose, while at the same time slowing the exit of glucose out of the gastric system, the overall effect of liraglutide is to ease the burden of the beta cells, allowing them periods of rest between meals as they respond to glucose in the body.
So, no evidence yet of increased beta cell death, no prevailing threat of thyroid cancer, easier to use than Byetta, and marginally helpful in clinical trials. Taken all together, I must admit, Victoza seems a better-than-neutral entry into the crowded type 2 diabetes space. That said, I am likely biased by the strong case Dr. Moses made for Novo Nordisk’s commitment to diabetes—the goal of the company, he said, has always been to “defeat diabetes.” As the second largest funder of diabetes research after the NIH, Novo Nordisk is instrumental in research towards both treatment and a cure. “This company goes all in,” says Dr. Moses, and in its dedication to diabetes and diabetics has created “the biggest diabetes sandbox in the world,” with patient-centered research at its core.
Plus, I think their little blue bull logo is cute.
- Pratley RE, Nauck M, Bailey T, Montanya E, Cuddihy R, Filetti S, Thomsen AB,Søndergaard RE, Davies M; 1860-LIRA-DPP-4 Study Group. Liraglutideversussitagliptin for patients with type 2 diabetes who did not have adequate glycaemiccontrol with metformin: a 26-week, randomised, parallel-group, open-label trial. Lancet. 2010 Apr 24;375(9724):1447-56. Erratum in: Lancet. 2010 Jul24;376(9737):234. PubMed PMID: 20417856.
- Hegedüs L, Moses AC, Zdravkovic M, Le Thi T, Daniels GH. GLP-1 and calcitonin concentration in humans: lack of evidence of calcitonin release from sequentialscreening in over 5000 subjects with type 2 diabetes or nondiabeticobesesubjects treated with the human GLP-1 analog, liraglutide. J ClinEndocrinolMetab. 2011 Mar;96(3):853-60. Epub 2011 Jan 5. PubMed PMID: 21209033.http://www.ncbi.nlm.nih.gov/pubmed/21209033
- Johnson JD, Luciani DS. Mechanisms of pancreatic beta-cell apoptosis in diabetes and its therapies. AdvExp Med Biol. 2010;654:447-62. Review. PubMed PMID: 20217509.http://www.ncbi.nlm.nih.gov/pubmed/20217509
Comment: Timing of dose not dependent upon meal timing. Good.
Question: However, is effect time dependent? Specifically, if my serum glucose readings are typically elevated first thing in the morning, before eating, would dose before bedtime lower them? You comment that the beta cells may get some rest at night, but mine would seem to be more likely under stress then. If my beta cells are asleep at night, maybe a dose then would help protect them then from the cytotoxic effect of my higher readings. Is there any evidence of a diurnal/nocturnal variation in the effect?