Memorandum

Novo Nordisk announces topline data from PIONEER 1: Oral semaglutide shows significant A1c-lowering vs. placebo; High-dose (14 mg daily) shows significant weight loss vs. placebo – February 23, 2018

Executive Highlights

  • Topline results from PIONEER 1 are in! Novo Nordisk’s oral semaglutide showed significant A1c-lowering vs. placebo at all three doses (3 mg, 7 mg, or 14 mg once-daily) after 26 weeks. The highest dose (14 mg) also demonstrated significant weight loss vs. placebo. Baseline A1c was 8% and baseline body weight was ~194 lbs (n=703). No p-values were included in the topline release, although Novo Nordisk did share findings from a secondary analysis looking only at patients who adhered to treatment and who did not require rescue therapy. Within this cohort, A1c dropped 0.8% with 3 mg oral semaglutide, 1.3% with 7 mg, 1.5% with 14 mg, and only 0.1% with placebo; mean weight loss was ~4 lbs on 3 mg oral semaglutide, ~6 lbs on 7 mg, ~9 lbs on 14 mg, and only ~3 lbs on placebo. Despite this being a secondary analysis, it suggests that patients who stick to an oral semaglutide regimen could experience meaningful glucose-lowering and weight loss.
  • That said, it’s unclear from the topline data what adherence was like on each dose, and adherence will be key. Novo Nordisk reported premature discontinuation due to adverse events in 2%-7% of semaglutide patients vs. 2% of placebo patients. Nausea was the most common side-effect, affecting between 5%-16% of semaglutide patients vs. 6% of placebo patients. Nausea diminished over time, but we suspect it may have lowered average adherence rates in the semaglutide groups. Fasting requirements around oral semaglutide (~6.5 hours in phase 2) could also impact adherence. Novo Nordisk management has stated that protocol for fasting is simpler in the phase 3 PIONEER program vs. phase 2, but we aren’t sure of the specifics.
  • All in all, we view the topline PIONEER 1 data as very good news. Oral GLP-1 would represent a major win for patients, and a significant stride forward in diabetes care. Ten PIONEER trials are scheduled to report results in 2018, and if all goes according to plan, the company could file an NDA for oral semaglutide in 2019. This drug would bring new innovation to the GLP-1 class and to Novo Nordisk’s GLP-1 business.

Yesterday, Novo Nordisk announced positive topline data from the PIONEER 1 trial of oral semaglutide, highlighting the candidate’s A1c-lowering efficacy, weight loss potential, and safety/tolerability. This is the first phase 3 readout on an oral GLP-1 agonist, and it marks the first of 10 PIONEER studies from Novo Nordisk that will wrap up and report results in 2018. If all goes according to plan, the company will file an NDA for oral semaglutide in 2019, with possible market launch in 2020.

In PIONEER 1 (n=703), oral semaglutide met its primary endpoint of significant A1c reductions over 26 weeks vs. placebo. Adults with type 2 diabetes were randomized to one of four treatment arms: (i) 3 mg oral semaglutide once-daily, (ii) 7 mg once-daily, (iii) 14 mg once-daily, or (iv) placebo. All three doses of the active agent showed superior A1c-lowering vs. placebo at week 26. Baseline A1c was 8% across treatment groups; margin of A1c decline with each dose was not disclosed, and we eagerly await the full PIONEER 1 results for this information. Furthermore, there’s no mention of p-values for the primary endpoint, but the announcement states that all oral semaglutide doses reached statistical significance.

Oral semaglutide at 14 mg daily demonstrated significantly superior weight loss vs. placebo (baseline body weight 88 kg, or ~194 lbs). The smaller doses showed greater weight loss vs. placebo as well, but did not meet statistical significance. Once again, p-values for weight loss were not included in the topline announcement, and we’ll be looking for these when Novo Nordisk shares full results, hopefully at an upcoming scientific conference. 

For participants who adhered to treatment/did not require rescue therapy throughout the course of the trial, A1c dropped 0.8% in the 3 mg oral semaglutide group, 1.3% in the 7 mg group, 1.5% in the 14 mg group, and only 0.1% in the placebo group (from a baseline of 8%). Mean weight loss within this cohort was ~4 lbs with 3 mg oral semaglutide, ~6 lbs with 7 mg oral semaglutide, ~9 lbs with 14 mg oral semaglutide, and only ~3 lbs with placebo (from a baseline of ~194 lbs and BMI 32 kg/m2). This was a secondary analysis of the data, but it does imply that patients who stick to an oral semaglutide regimen could see meaningful glucose-lowering and weight loss benefits.

We’re certainly excited by the prospect of an oral GLP-1 agonist, and oral administration seems like the next major frontier of innovation for GLP-1 therapy. This class is growing fast from a high base, and importantly, still has ample head room for further growth – only 7% of second-line diabetes prescriptions in the US go to a GLP-1 agonist (46% still go to a sulfonylurea, which is so disappointing), and GLP-1 prescription volume represents a mere ~30% of basal insulin prescription volume globally.

Adherence, Safety, and Tolerability

  • A new diabetes drug that is difficult for patients to take regularly could struggle commercially – we’re far from this conclusion, as we’ve only seen topline results from one of 10 phase 3 studies, but we think it’s important to note. Per Novo Nordisk’s topline release, oral semaglutide was safe and well-tolerated in PIONEER 1. Nausea (mild or moderate) was the most common side-effect, affecting between 5%-16% of semaglutide-treated patients vs. 6% of placebo-treated patients. Nausea diminished over time, but we suspect it may have lowered average adherence rates in the semaglutide groups. Novo Nordisk reported premature discontinuation from the study due to adverse events in 2%-7% of semaglutide patients vs. 2% of placebo patients.
  • Fasting requirements around oral semaglutide could also impact adherence. There was no mention of this in the company announcement on PIONEER 1, but Novo Nordisk CSO Dr. Mads Thomsen previously explained how fasting protocols are simpler in the phase 3 program for oral semaglutide vs. the phase 2 program (when patients were instructed to take a dose in the morning, after ~six hours of fasting, and to refrain from eating or drinking for another ~30 minutes thereafter). The precise nature of fasting instruction in PIONEER 1 is still unclear, and this is another detail we’re eager to learn from a full results presentation. We imagine ~6.5 hours of required fasting around a dose could be perceived as burdensome for some patients, potentially offsetting the convenience of having an oral GLP-1 agonist rather than injectable. Since Novo Nordisk aims to position oral semaglutide as a competitor to other oral diabetes drugs (rather than other GLP-1 agonists), it’ll be important for ease-of-use to be on par with SGLT-2 inhibitors and maybe DPP-4 inhibitors; in other words, simply being preferred over an injectable GLP-1 option may not be enough. Dr. Thomsen emphasized on the company’s 4Q17 earnings call that adherence was strong in phase 2 despite the fasting protocol, but this won’t necessarily translate to the real world (assuming clinical trial participants are better-supported and generally more engaged). A simpler fasting requirement in phase 3 could be key, and we can’t wait to see more specific adherence numbers from PIONEER 1. Ultimately, the language around fasting that ends up on the oral semaglutide label will guide real-world adherence, treatment satisfaction, and outcomes. We’re likely still ~two years away from an FDA-approval of this oral GLP-1, but we’ll be watching adherence data with keen interest (alongside the A1c and body weight effects) as PIONEER studies continue to read out this year.

The PIONEER Clinical Program

  • With 10 phase 3 readouts coming 1Q-4Q18, 2018 could be a big year for oral semaglutide. The table below summarizes these 10 PIONEER studies, and according to Novo Nordisk, this program has enrolled >9,000 type 2 patients overall. PIONEER 2 and 7 are scheduled to report data in 2Q18. PIONEER 3, 4, and 5 are slated to read out in 3Q18, while PIONER 6, 8, 9, and 10 are expected to read out in 4Q18.

PIONEER Phase 3 Program for Oral Semaglutide

Trial

Estimated Enrollment

Comparator/Design

Timeline

PIONEER 1

703

Placebo

Topline results announced February 2018; Completed December 2017

PIONEER 2

816

Lilly/BI’s Jardiance (empagliflozin)

Expected to complete March 2018

PIONEER 3

1,860

Merck’s Januvia (sitagliptin)

Expected to complete March 2018

PIONEER 4

690

Novo Nordisk’s Victoza (liraglutide)

Expected to complete March 2018

PIONEER 5

324

Moderate renal impairment

Expected to complete May 2018

PIONEER 6

3,176

CVOT

Expected to October 2018

PIONEER 7

500

Flexible dose escalation

Expected to complete March 2019

PIONEER 8

720

Insulin add-on

Expected to complete August 2018

PIONEER 9

240

Placebo and liraglutide in Japan

Expected to complete August 2018

PIONEER 10

455

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

Expected to complete July 2018

Novo Nordisk’s GLP-1 Business

  • Novo Nordisk has shown remarkable commitment to innovation in GLP-1, with market-leading Victoza (liraglutide once-daily), just-launched Ozempic (injectable semaglutide once-weekly), and now phase 3 oral semaglutide. The semaglutide molecule appears to be highly-potent, leading to profound glucose-lowering, weight loss, and even cardio/renal protection – to this end, we’re especially curious for PIONEER 6 CVOT results (anticipated in 4Q18), which will reveal whether oral administration of the agent confers similar CV and renal benefits as seen in SUSTAIN 6 (the pre-market CVOT for injectable semaglutide). Notably, Novo Nordisk will also initiate the larger, post-market SOUL CVOT for injectable semaglutide in mid-2018. And, the company has plans for the SELECT CVOT of injectable semaglutide in obesity, also to start sometime this year – talk about a robust clinical program around one molecule!
  • As outlined during Capital Markets Day last year, oral semaglutide could reach the commercial market just in time for Victoza’s patent expiry (around 2022 or 2023), helping to sustain and drive Novo Nordisk’s GLP-1 business. Ozempic prescription volume is also expected to outpace Victoza prescription volume around 2020.

  • If approved, oral semaglutide will be positioned earlier in the course of diabetes development, so that it competes with other oral agents more so than injectable GLP-1s. Lilly management has expressed skepticism that oral semaglutide will be able to compete with an oral therapy like SGLT-2 Jardiance (empagliflozin), especially if it’s priced on par with existing GLP-1s, which might make it a lower value proposition in the eyes of payers relative to SGLT-2s. We’re more optimistic ourselves, though we acknowledge there are key tolerability/adherence issues to look for as more PIONEER data comes out. For one, Novo Nordisk hasn’t announced any final decisions on pricing strategy for oral semaglutide, so any statement on “value proposition” can only be speculation right now. PIONEER 2 comparing the drug head-to-head vs. Jardiance will be telling, as will the other head-to-head trials ongoing (PIONEER 3 vs. Merck’s DPP-4 inhibitor Januvia, PIONEER 4 vs. Victoza, and PIONEER 10 vs. Lilly’s once-weekly GLP-1 agonist Trulicity). We maintain that a safe/effective oral GLP-1 product would be a significant victory for patients, a significant stride forward in diabetes care. We imagine there are ways to mitigate adherence issues, particularly those related to fasting, with simpler instructions and strong real-world patient education. Given Novo Nordisk’s clear commitment to GLP-1 and to semaglutide (both oral and injectable), we’re confident the company could develop these straightforward instructions alongside regulatory agencies, and could promote real-world education and awareness.

Close Concerns Questions

Q: When can we expect full results from PIONEER 1? We wouldn’t be surprised to see this at ADA 2018 in Orlando.

Q: What were the p-values on primary endpoint?

Q: What were the adherence rates in each arm of the study (3 mg, 7 mg, 14 mg, or placebo)? How was adherence defined in PIONEER 1? How exactly was “adhering to treatment” and “not requiring rescue therapy” defined?

Q: Were there dose-dependent effects observed in PIONEER 1? How did margin of A1c-lowering compare with low-, middle-, and high-dose oral semaglutide? How did adherence and nausea (among other side-effects) differ across groups? This could be important in determining optimal doses for real-world patients, balancing efficacy with tolerability.

Q: How will oral semaglutide be priced, assuming it is eventually approved and launched commercially? It may be early to speculate – as of now, an NDA is planned for 2019, which means the product wouldn’t be available in pharmacies until 2020 at the earliest – but it’s hard to ignore cost/reimbursement questions given the challenging pricing environment around diabetes drugs (and the bar for new therapies is only rising).

 

-- by Payal Marathe and Kelly Close