Memorandum

Novo Nordisk releases topline results from PIONEER 8 study of oral semaglutide as add-on to insulin – October 29, 2018

Executive Highlights

  • Novo Nordisk released topline results from PIONEER 8 (n=731), supporting the safety and efficacy of phase 3 oral semaglutide as an add-on to insulin in people with longstanding type 2 diabetes (average 15 years). In the intention-to-treat analysis, all three doses of oral semaglutide (3 mg, 7 mg, 14 mg) were superior to placebo on both A1c lowering and weight loss at week 26. In an on-treatment analysis, the 14 mg dose of oral semaglutide lowered A1c by 1.4% compared to 0.0% with placebo, at 26 weeks and from a baseline 8.2%. Additionally, 3 mg oral semaglutide gave a 0.6% drop and 7 mg a 1.0% drop in A1c.

    • Oral semaglutide 14 mg also gave ~9.5 lbs of weight loss at 52 weeks, compared to ~1.3 lbs of weight gain with placebo. The 3 mg group for oral semaglutide saw ~2.2 lbs of weight loss, while the 7 mg group lost ~6.4 lbs on average. This is certainly a meaningful endpoint for insulin treated patients; mean baseline weight was ~189 lbs.

    • The 14 mg dose was associated with a 7 unit/day decrease in insulin requirements at 52 weeks. Similarly, the 7 mg dose was associated with a 6 unit/day drop and the 3 mg dose a 2 unit/day drop. The placebo group increased insulin by 10 units/day.

  • Among those randomized to oral semaglutide, 11%-23% experienced nausea vs. 7% on placebo. This was the most common and only reported adverse event; for comparison, 5%-16% of participants in PIONEER 1 experienced nausea – that is, PIONEER 8 participants seem to have experienced somewhat higher rates of this side effect. Treatment discontinuations were 7%-14% with oral semaglutide, compared to 3% with placebo and rates of 2%-7% with oral semaglutide in PIONEER 1. We wonder what baseline or other factors specific to insulin-treated patients (e.g. treatment burden) may have had an impact on these rates.

  • These results have been released on a backdrop of increasing excitement surrounding oral semaglutide (see Drs. Cliff Bailey and John Buse at EASD 2018) and anticipation for PIONEER 6 CVOT results. As Dr. Bailey outlined, oral semaglutide’s clinical profile (i.e. A1c lowering and weight loss) is second only to injectable semaglutide (Novo Nordisk’s Ozempic). The company is still targeting 2019 regulatory submission, and PIONEER 8 serves to expand the candidate’s demonstrated clinical efficacy. Before submission, however, we’ll see results for two more PIONEER studies: PIONEER 6, a smaller-scale CVOT enrolling ~3,200 and powered to demonstrate safety premarket, as well as PIONEER 9, a second head-to-head study with liraglutide to support regulatory approval in Japan. See more on Novo Nordisk’s CVOT plans below, under PIONEER Program.

On Friday, Novo Nordisk announced positive topline results from PIONEER 8, a study of phase 3 oral semaglutide vs. placebo as an add-on to insulin in people (n=731) with a long duration of type 2 diabetes (average 15 years). PIONEER 8 randomized patients already on basal, premix, or basal/bolus insulin to one of three doses of oral semaglutide (3 mg, 7 mg, and 14 mg) or to placebo, evenly across groups. Of note, average diabetes duration was 15 years. Insulin dose was capped at pre-randomized levels for the first 26 weeks, then could be adjusted unrestricted for the remaining 26 weeks.

Novo Nordisk continues to report results according to two statistical principles; the intention-to-treat analysis (i.e. the primary statistical principle) is required and considered by FDA guidance, while the on-treatment analysis (i.e. the secondary statistical principles) excludes those who discontinue treatment and require rescue medication. According to the ITT analysis, PIONEER 8 demonstrated significant and placebo-superior effects on A1c and body weight with all three oral semaglutide doses at 26 weeks.

In the on-treatment analysis (i.e. the secondary statistical principle), 14 mg oral semaglutide gave a 1.4% A1c reduction vs. 0.0% with placebo, at 26 weeks and from a mean baseline A1c of 8.2%. The 7 mg dose gave a 1.0% reduction in A1c, and the 3 mg dose gave a 0.6% drop. At week 52, mean A1c reductions were 1.2%, 0.8%, and 0.5%, respectively, vs. 0.0% with placebo. These results support that the efficacy of oral semaglutide is maintained in people with longstanding diabetes and on top of insulin, adding another dimension to the phase 3 candidate’s demonstrated efficacy profile.

Body weight fell by ~9.5 lbs (4.3 kg) with 14 mg oral semaglutide after 52 weeks, also in the on-treatment analysis, compared to weight gain of ~1.3 lbs (0.6 kg) with placebo – certainly meaningful for insulin-treated patients. The 7 mg oral semaglutide dose gave ~6.4 lbs (2.9 kg) of weight loss at 52 weeks and the 3 mg dose ~2.2 lbs (1.0 kg). Weight loss was from a mean baseline body weight of ~189 lbs (85.9 kg).

Among those receiving 14 mg oral semaglutide, an impressive 64% achieved an A1c <7.0% in the on-treatment analysis. This metric was achieved by 36% of those on 3 mg oral semaglutide, 47% on 7 mg oral semaglutide, and 10% on placebo. The 14 mg dose of oral semaglutide was associated with a 7 unit/day drop in mean total insulin dose at 52 weeks, compared to a 6 unit/day drop with the 7 mg dose, a 2 unit/day increase with the 3 mg dose, and a 10 unit/day increase with placebo.

Safety and Tolerability

  • As in previous PIONEER readouts, oral semaglutide was well-tolerated overall. As expected, mild to moderate nausea was the most common adverse event: 11%-23% of those randomized to oral semaglutide experienced nausea, compared to 7% of those on placebo. No data was given on GI side effects generally, but these numbers are roughly on-par with other injectable GLP-1 agonists and previous oral semaglutide data. For comparison, in PIONEER 1, nausea occurred at a rate of 5%-16% in patients on oral semaglutide vs. 6% of patients on placebo, so this particular GI side effect appears somewhat more common in the PIONEER 8 population. Treatment discontinuations occurred at a rate of 7%-14% in the oral semaglutide groups compared to 3% with placebo. This compares to discontinuation rates of 2%-7% with oral semaglutide in PIONEER 1 – that is, meaningfully higher. We wonder if, among patients on insulin, the addition of oral semaglutide treatment presented a more significant treatment burden than it might have otherwise, leading to higher discontinuation rates; other baseline characteristics could also have an impact. It’s not clear how many treatment discontinuations were due to adverse events.

Competitive Landscape

  • We continue to be very impressed with oral semaglutide’s clinical profile, particularly the 14 mg dose, and the field seems to be increasing in excitement for the first-ever oral GLP-1. For example, earlier this month at EASD 2018, Dr. Cliff Bailey demonstrated that oral semaglutide’s efficacy on A1c lowering and weight loss is second to only Ozempic in the GLP-1 agonist class. Additionally, Dr. John Buse prominently highlighted oral semaglutide in a rundown of the diabetes candidates he is most excited about. Overall, our interpretation of PIONEER data is that 14 mg oral semaglutide especially offers meaningful efficacy improvements over comparators both within and outside the GLP-1 agonist class, as well as an edge on weight loss in most cases.

  • To our knowledge, Novo Nordisk is still targeting 2019 for oral semaglutide submission to FDA and EMA, which means this first-in-class oral GLP-1 agonist could be available to real-world patients as soon as 2020. For context, we don’t expect another oral GLP-1 agonist to reach the market within the next ~five years; the closest in-class competitor is vTv’s phase 2b TTP273. However, it remains to be seen where in the diabetes treatment algorithm Novo Nordisk will aim to position oral semaglutide. Evidence is now in place that oral semaglutide offers a clinical profile that can compete handily with injectable GLP-1 agonists. However, with two injectables already on the market, we expect Novo Nordisk will strive to position oral semaglutide as an earlier treatment intensification option, alongside other second-line oral agents (DPP-4 and SGLT-2 inhibitors) as outlined during the company’s 2017 Capital Markets Day. Given strong evidence from PIONEER 3 (vs. sitagliptin) and PIONEER 2 (vs. empagliflozin), we think oral semaglutide is in a very strong position clinically – though much more remains to be determined on pricing and reimbursement. That said, this new PIONEER 8 data also supports the use of oral semaglutide as an intensification option on top of a variety of insulin regimens, so it’s conceivable that Novo Nordisk would promote its use across the spectrum of diabetes treatment.

The PIONEER Program

  • Only two studies from the ten-trial PIONEER program for oral semaglutide (see table below) have yet to read out. These include PIONEER 6, a smaller-scale CVOT enrolling ~3,200 and powered to demonstrate safety premarket, as well as PIONEER 9, a second head-to-head study with liraglutide to support regulatory approval in Japan. Topline results for both are expected within 2018. These PIONEER 8 results follow readouts from PIONEER 1 (vs. placebo), PIONEER 2 (vs. Lilly/BI’s Jardiance), PIONEER 3 (vs. Merck’s Januvia), PIONEER 4 (vs. Novo Nordisk’s own Victoza), PIONEER 5 (in moderate renal impairment), PIONEER 7 (vs. Merck’s Januvia, with dose adjustment), and PIONEER 10 (vs. Trulicity in Japan). There’s no doubt that PIONEER 6 results are both eagerly anticipated and have broad implications for Novo Nordisk’s entire GLP-1 agonist business. Critically, during Novo Nordisk’s 2Q18 update, CSO Dr. Mads Thomsen outlined how if PIONEER 6 manages to demonstrate CV superiority over placebo, FDA would likely accept the combination of PIONEER 6 and SUSTAIN 6 (for injectable semaglutide) data as evidence of the molecule’s CV benefit. Otherwise, a larger CVOT for oral semaglutide will support a CV indication for both formulations. Of course, PIONEER 6 was designed to demonstrate CV safety, not superiority, so there’s no strong expectation that superiority will be achieved. However, SUSTAIN 6 was also designed for safety and managed to demonstrate superiority for injectable semaglutide, so we’re certainly not ready to rule the possibility out. Superiority would be an enormous win for Novo Nordisk and people with diabetes alike.

Phase 3 PIONEER Program

Trial

Estimated Enrollment

Comparator/Design

Timeline

PIONEER 1

704

Placebo

Completed December 2017; Topline results announced February 2018; Full results at ADA 2018

PIONEER 2

816

Lilly/BI’s Jardiance (empagliflozin)

Completed March 2018; Topline results announced May 2018

PIONEER 3

1,864

Merck’s Januvia (sitagliptin)

Completed March 2018, Topline results announced June 2018

PIONEER 4

711

Novo Nordisk’s Victoza (liraglutide)

Completed March 2018; Topline results announced June 2018

PIONEER 5

324

Moderate renal impairment

Completed May 2018; Topline results announced August 2018

PIONEER 6

3,176

CVOT

Completed September 2018

PIONEER 7

504

Flexible dose escalation

Completed March 2019; Topline results announced June 2018

PIONEER 8

731

Insulin add-on

Completed August 2018; Topline results announced October 2018

PIONEER 9

230

Placebo and liraglutide in Japan

Completed August 2018

PIONEER 10

336

Lilly’s Trulicity (dulaglutide) as an add-on to oral agents in Japan

Completed July 2018; Topline results announced September 2018

 

--by Ann Carracher and Kelly Close