Menu

Report Library

All Reports

2019 Biomedtracker / Datamonitor Healthcare / Meddevicetracker Post-ADA Report

July 22, 2019

The American Diabetes Association (ADA) 79th Scientific Sessions was held in San Francisco, CA from 7–11 June 2019. This year's conference was filled with major sessions dedicated to outcomes trials and Phase III programs, notably: first details from Trulicity's REWIND (LLY, Sumitomo Dainippon) cardiovascular outcomes trial (CVOT), first conference presentation of oral semaglutide's (NVO, EMIS) PIONEER 6 CV safety study and other trials from its Phase III program, first ADA presentation of Invokana's (JNJ, Mitsubishi Tanabe, Daiichi Sankyo) CREDENCE cardiorenal outcomes trial, an update with new CV and renal details from Farxiga's (AZN, Ono) DECLARE-TIMI 58 CVOT, first details from Tradjenta's (Boehringer, LLY) CAROLINA CVOT trial against glimepiride, and first ADA presentation of Tradjenta's CARMELINA CVOT against placebo.

Trulicity's REWIND had the most new data and was the first glucagon-like peptide-1 (GLP-1) agonist CVOT to show a benefit in primary prevention patients, though key opinion leaders (KOLs) felt this was due to the trial design, and overall the results were fairly modest. The data of course feed into the competition between Eli Lilly and Novo Nordisk on GLP-1 agonists and co-agonists, which was a major topic of interest at the conference, as they seek to top each other with higher doses and novel approaches that will win out on glycated hemoglobin (A1C) reduction and weight loss, without compromising too much on tolerability.

With the strong evidence for a renal benefit with sodium-glucose cotransporter-2 (SGLT-2) inhibitors, as highlighted in presentations on Invokana's CREDENCE and analyses from Farxiga's DECLARE, a battle is also brewing between these agents and the GLP-1 agonists in the diabetic nephropathy (DN) segment, where their CV benefits are also important. Trulicity showed interesting evidence of a strong renal benefit in an updated analysis from AWARD-7 (Eli Lilly, though, also markets the SGLT-2 inhibitor Jardiance), and at its investor call, Novo Nordisk highlighted its FLOW cardiorenal outcomes trial for Ozempic, which had been announced earlier in the year, with plans to bridge the data to oral semaglutide. FLOW only started to enroll just after the conference, however, and will take some years to complete.

The major review of oral semaglutide's Phase III program, taking a full session at the conference, manifested how the drug is poised to disrupt the field of branded oral drugs, though many details were already known from prior releases of data.

In type 1 diabetes prevention, teplizumab's (PRV-031; PRVB, MGNX) Phase II trial became the first to show that an immune therapy can be used to delay type 1 diabetes, though the trial was relatively small.

On the device front, Tandem's artificial pancreas released data suggesting it could be competitive with Medtronic’s MiniMed 670G.

We highlight these and other presentations below. For the full report, please select the "Download Report PDF" button above.

Like our report? Have any questions or feedback? Please let us know at askanalyst@sagientresearch.com.

GLP-1 Agonists
  • Trulicity’s REWIND CVOT: Results from Trulicity's CVOT were disappointingly modest and underwhelming compared to expectations, with only a 12% hazard reduction on the primary endpoint.
    • While this was the first GLP-1 agonist CVOT to show a major adverse cardiovascular event (MACE) benefit in primary prevention patients, which could expand usage, KOLs and investigators we spoke with were fairly confident that the difference in primary prevention patients was just due to longer treatment duration.
    • The modest reduction in MACE was also seen in the secondary prevention subgroup, which could prompt questions whether Trulicity is as strong as options from Novo Nordisk, particularly the semaglutide franchise, though the latter only has data from smaller CV safety studies. Nevertheless, Trulicity's impact on 3-point MACE was more comparable, so the differences may be due to chance.
    • One KOL even questioned whether Trulicity would gain a primary prevention indication — though it would seem odd for it to just receive an indication for secondary prevention, which comprised a minority of the trial — but he did feel it would gain some type of CV indication. There are also regulatory questions about the borderline p-value on the primary endpoint, after adjusting for an interim analysis, which was only revealed in a table footnote in the publication, as well as a numerically worse result for the drug in the North American subgroup, though Eli Lilly officials indicted in a conference call that the hazard ratio for the US specifically was less than 1.
    • While REWIND results at least put Trulicity in the club that has had positive CV outcomes, and the convenience of its device and titration schedule help its popularity, with the modest CV benefit, Trulicity is likely to lose additional share to Novo Nordisk's metabolically stronger Ozempic, whose oral formulation should also be approved later this year and could undercut both injectable drugs (see oral semaglutide below).
      • Both Trulicity and Ozempic have higher dose studies, which will fuel the next round of competition, though the higher dose of Ozempic is also being developed to compete with Eli Lilly's tirzepatide (see below). Soon after the conference, Trulicity's pivotal high dose study reported an advantage on A1C and weight over its currently approved high dose, but details were scant.

  • Oral semaglutide Phase III data: Striking results from oral semaglutide's Phase III program were reviewed in a major session, which had strong attendance even though it was the last day of the conference. While topline results from many of the trials have been released before, including during an investor call at last year's ADA, this year's ADA included the first conference presentations of PIONEER 2 (versus Jardiance), PIONEER 4 (versus Victoza), PIONEER 5 (moderate renal impairment), PIONEER 6 (CV safety versus placebo), PIONEER 7 (flexible dose versus Januvia), and PIONEER 8 (add-on to insulin).
    • Oral semaglutide appears to be the most effective non-insulin branded drug on A1C lowering and weight loss, except for its injectable formulation Ozempic, though some of the benefits over other drugs tested were modest or only appeared in a "trial product estimand" (essentially on-treatment, without rescue medication).
    • There were no surprises on safety, and nausea, a major side effect of GLP-1 agonists, was numerically only slightly higher than Victoza. However, the time course of the side effect was somewhat different, with the rate dropping more rapidly for Victoza.
    • In addition to gastrointestinal (GI) side effects, oral semaglutide has the disadvantage of having to be taken while fasting in the morning and 30 minutes before eating or drinking. There were mixed opinions on how much of a burden that may be for patients, and some patients may still prefer a weekly injectable GLP-1 agonist, though not having to educate patients on injectables would be a big plus for primary care physicians.
    • While oral semaglutide will undoubtedly become a major contender, various patient factors and preferences, along of course with cost, will come into play when deciding on treatment, since dipeptidyl peptidase-IV (DPP-IV) inhibitors have better tolerability and SGLT-2 inhibitors have additional benefits in heart failure and DN.
      • However, Novo Nordisk is still seeking to challenge SGLT-2 inhibitors in the important DN segment with a renal outcomes trial for Ozempic (FLOW) and plans to bridge results to oral semaglutide using renal outcomes data for the latter's CVOT. FLOW may only complete, though, in 2024, giving the SGLT-2 inhibitors, which are already recommended in guidelines for DN, some breathing room in the segment. (Please see the comment under Invokana's CREDENCE trial in the main section of the report for a more detailed discussion on this segment and the general competition among branded oral drugs.)
      • On cost, Novo Nordisk officials said in its ADA investor call that it is in the final phases of choosing a new formulation that would require less active ingredient, reducing the manufacturing cost of oral semaglutide.
    • While oral semaglutide's CV safety study had not reached statistical significance on its MACE primary endpoint, in contrast to Ozempic, it was supportive, and earlier this year Novo Nordisk submitted a New Drug Application (NDA) for a CV indication for both drugs. A combined analysis of the two CV safety studies, presented at the conference, did show a statistically significant benefit.
      • While the p-value was not given, it appeared that it could be quite small, possibly less than 0.01 but greater than 0.00125 (some metrics mentioned when the FDA looks at single studies).
      • Whether the FDA will grant the indication is highly uncertain, though, and we tend to think the company will have an uphill battle.
    • At the conference, Novo Nordisk officials indicated they hope to see 20+% weight loss with injectable semaglutide used in combination with the company’s amylin analog in obesity, though that of course is speculative.
GLP-1 Co-Agonists
  • Tirzepatide Phase II data: In a Phase II trial testing dose-escalation algorithms, more prolonged titration of glucose- dependent insulinotropic polypeptide (GIP)/GLP-1 co- agonist tirzepatide (LLY) showed lower GI adverse effects than the prior Phase IIb study, though they were still somewhat high, and rates of diarrhea were not diminished (though reported to be milder). However, the presenter said titration in Phase III is over an even longer time, which he hoped would bring overall rates down to those seen for GLP-1 agonists, though his comments appeared fairly speculative.
    • Data from the initial Phase IIb trial had suggested promising efficacy for high doses, potentially even stronger than Ozempic, but in an investor call at this year's conference, Novo Nordisk pointed to a higher dose trial of Ozempic, which it hopes will match efficacy of tirzepatide.
      • Novo Nordisk officials acknowledged that beta-cell stimulation is probably fully saturated at the current 1mg high dose of Ozempic, but believe that the 2mg dose being tested will lead to more weight loss with secondary improvements in A1C.
    • Tirzepatide's 20-week titration utilized in Phase III could be a drawback for patients who need the high dose, depending on how they perceive the efficacy. Interestingly, Novo Nordisk limited the dose of Ozempic in its higher dose study to avoid an additional titration step (the current titration is eight weeks).
    • Eli Lilly is conducting mechanistic studies to better understand the contribution of GIP, which could help differentiate the drug, since other data at the conference suggested that insulin-sensitizing effects were only partially explained by weight loss.
      • Tirzepatide also improved non-alcoholic steatohepatitis (NASH)-related markers in an analysis of the Phase IIb study, and there was anecdotal evidence for improvements unrelated to weight loss in these as well, which the company is exploring further. A Phase II trial in NASH is expected to start this year.

  • Janssen ends partnership with Hanmi: Soon after the conference, Janssen dropped its partnership with Hanmi for HM12525A, a GLP- 1/glucagon dual agonist, because it did not meet targets for glycemic control.
SGLT-2 Inhibitors
  • Invokana’s CREDENCE data: While the main results of Invokana's CREDENCE DN trial had already been released, this first ADA presentation of the data took place in a major session, and new details on CV and renal outcomes by secondary and primary prevention subgroups were also released.
    • Several speakers at various sessions at the conference said while it was reassuring that CREDENCE did not find an increase in amputations, it does not completely eliminate concerns raised by CANVAS, which led to a black box warning. Hence, we expect the issue will still be a drag on sales.
    • Unlike SGLT-2 inhibitor CVOTs, CREDENCE showed a benefit on MACE in primary prevention patients, though looking at the data more closely, the primary prevention group was at quite high risk, probably similar to patients with established CV disease.
    • Though the benefit on MACE and renal outcomes was even stronger in patients with moderate renal impairment than less severe disease, physicians will need to balance that with declining glycemic control at that stage, as with all in the SGLT-2 inhibitor class, but not so much for GLP-1 agonists. Glycemic control with SGLT-2 inhibitors is of course better at earlier stages, but a couple of experts said there was not so much concern about renal progression at that point, though there are other reasons to use the class.
      • Highlighting potential competition with the GLP-1 agonists in this space, updated data at the conference from Trulicity's AWARD- 7 trial in patients with moderate to severe renal impairment found a substantial reduction in a renal composite outcome compared to basal insulin, despite lack of a significant difference in A1C or blood pressure. Moreover, as noted above, Novo Nordisk has initiated a renal outcomes trial for Ozempic and plans to bridge results to oral semaglutide using renal outcomes data for the latter's CVOT.
      • We currently forecast usage of SGLT-2 inhibitors in DN patients in the US to grow 80% or more, limited by competition with oral semaglutide. Please see the comment under CREDENCE for a detailed discussion of competition among the branded drugs.
    • A couple of presentations at the conference challenged prevailing views on the mechanism for the renal benefit of SGLT-2 inhibitors, with one raising questions whether a reduction in tubuloglomerular pressure was indeed the most important effect.
DPP-IV Inhibitors
  • Tradjenta’s CAROLINA CVOT: While it was already announced that CAROLINA did not show a CV benefit for Tradjenta against the sulfonylurea glimepiride, it also failed to show any evidence of longer-term A1C preservation and only showed a modest benefit on weight.
    • The presentation highlighted Tradjenta's well-known advantage on hypoglycemia, but it was interesting to see that the hypoglycemia risks occurred even at lower doses of glimepiride, and while appearing early on during titration, also persisted throughout the trial.
    • Nevertheless, the trial could well reassure physicians about the use of generically available sulfonylureas given the lack of much difference on other measures.
Prevention of Type 1 Diabetes
  • Teplizumab Phase II data: Teplizumab's Phase II trial was the first to show that an immune therapy can be used to delay type 1 diabetes, though the study was small and the findings tentative.
    • The drug only appeared to delay the condition by two years, but in light of the daily burden of disease management, particularly for children, the investigator felt the difference was clinically meaningful. Patients are also still being followed (LIFT study) to see if there will be a group that does not develop diabetes even longer term.
    • The company is pursuing a breakthrough designation and is hoping to submit to the FDA with this one study, but the prospects seem tenuous, given the small size and past failures of the drug. A Phase III study in newly diagnosed type 1 diabetes is expected to report in 2022.
Insulins
  • LY900014 Phase III data: Ultra-rapid acting insulin LY900014 (LLY) only demonstrated non-inferiority with Eli Lilly's own Humalog in pivotal type 1 and type 2 diabetes studies when both were given at mealtime, and it was somewhat disappointing that it did not perform any better on A1C or have a more significant advantage on hypoglycemia.
    • The results were not too dissimilar from Fiasp (NVO), though the latter had a slight statistically significant advantage on A1C over its comparator NovoLog when given at mealtime in type 1 patients (LY900014 just missed superiority in its type 1 study).
    • Post-meal dosing of LY900014 also met non-inferiority criteria to mealtime Humalog on A1C in type 1 patients (based on the upper bound of the confidence interval), but its slightly worse performance was statistically significant and an endocrinologist in the audience was quite critical of the investigator's positive portrayal of that option. Post-meal Fiasp had performed more similarly to its comparator NovoLog in type 1, though the differences between the findings of these trials are small.
    • The presentations highlighted an advantage over lispro on post-prandial glucose during a meal test in both type 1 and 2 diabetes, but an expert acknowledged that not enough data were shown to know where it may have been worse, given that A1C values did not differ much. Disappointingly, in glucose testing with real meals, an investigator said LY900014 had a post- meal advantage only during the first meal of the day.
    • Similarly, while LY900014 had less hypoglycemia in the late post-prandial period in the type 1 study, likely related to its shorter duration, that did not make a significant difference for overall hypoglycemia, and in the type 2 study, it had more hypoglycemia between 1 to 4 hours after the meal.
    • While the data made it unclear what the advantage of LY900014 would be, experts had some suggestions on how or why they might use it, with one noting the ultra- rapid insulins are safer in type 2 and the post-prandial benefits could be important for long-term vascular outcomes, and another saying he might use LY900014 in patients having difficulty taking mealtime insulin, even though the studies did not find an A1C benefit. Still, one acknowledged that Fiasp has not taken off in the market, suggesting similar questions about whether it has much advantage.

  • SAR341402 Phase III data: SAR341402 (SNY), a biosimilar mealtime insulin, performed reassuringly similarly to its reference product, Novo Nordisk's insulin aspart (NovoLog), in a Phase III trial of type 1 and type 2 diabetics. There were some differences on secondary measures, but these could have been due to chance.
Non-Alcoholic Steatohepatitis (NASH)
  • Additional data on insulin sensitizers: Insulin sensitizers MSDC-0602K (Cirius Therapeutics) and MK-3655 (MRK) had incremental updates from their early-stage studies, and the report reviews their data. The drugs are interesting as the insulin-sensitizing diabetes drug pioglitazone is already recommended in US guidelines for NASH.
    • MSDC-0602K is hoped to target a receptor the company believes is responsible for efficacy of pioglitazone, while avoiding its peroxisome proliferator-activated receptor (PPAR) gamma side effects. MK-3655 targets a specific fibroblast growth factor 21 (FGF21) receptor complex in hope of avoiding side effects of less selective FGF21 agonism, and was compared directly to pioglitazone.
    • Early signals of improving liver function tests and markers of fibrosis are encouraging for both, albeit tentative, with MK-3655 additionally showing improvements in liver fat that appeared stronger than pioglitazone.
    • Both also improved glycemic control that appeared at least comparable to pioglitazone, which could be an added benefit in diabetes patients, though it remains to be seen if their glycemic effects will be as strong in actual diabetes patients with less well-controlled A1C than in the studies.
    • However, on the downside, both also increased weight (for MK-3655, the gain was less than for pioglitazone), which could be due to insulin sensitization, and both need larger studies to make sure that they indeed avoid the significant side effects of related drugs.
    • Despite the encouraging results observed with the insulin sensitizers, their late entry onto the crowded NASH market will leave them at a disadvantage. There are multiple pipeline agents with different mechanisms in development to treat NASH, so the insulin sensitizers will have to exhibit a strong ability to reduce fibrosis to compete. They could find success in combination therapy with complementary mechanisms to augment their effects on fibrosis regression.
Dyslipidemia
  • Opinions on Vascepa: Enthusiasm for Vascepa (AMRN), which contains the omega-3 fatty acid ethyl eicosapentaenoic acid (EPA), was mixed, both among cardiologists and endocrinologists. Details from the drug's positive CVOT had generated a great deal of interest at last year's American Heart Association conference.
    • Cardiology KOLs at the ADA had differing opinions on whether the mineral oil comparator could have led to more negative results for the control group in the CVOT, with one KOL questioning whether an FDA advisory committee panel would even vote in favor of a CV indication.
    • Concerns involved understanding how the drug would fit in when patients were on so many medications already, in addition to pill burden. And for diabetics, due to affordability issues with multiple branded drugs, there could well be competition with diabetes drugs that also have a CV benefit.
Devices
  • t:slim X2 data: Tandem's artificial pancreas (AP), t:slim X2 with new Control-IQ technology, demonstrated a greater time in range than a control sensor- augmented pump and comparable to data seen for Medtronic’s MiniMed 670G, the first approved hybrid closed-loop insulin delivery system. If approved, Tandem hopes to launch the updated device in the second half of 2019.
    • Tandem's AP is the first with automated correction boluses, though competition is not far behind, with Medtronic planning to introduce this feature with its next- generation MiniMed 780G, expected to launch by April 2020.
    • At an ADA investor conference call, Medtronic officials said they hoped the 780G will increase time in range to even higher levels. Medtronic still lags on eliminating routine fingersticks (as achieved by the Dexcom G6 CGM used with the Tandem AP) and an interoperability designation from the FDA for use with system components from various manufacturers, though in the call the company talked about potential progress in these areas.
This report also has additional data/news on:
  • Direct PK/PD comparison of Toujeo (SNY) versus Tresiba (NVO)
  • Preclinical data on AstraZeneca's weekly insulin
  • NASH studies of SGLT-2 inhibitors
  • Speculation on the mechanisms for the benefit of SGLT-2 inhibitors in heart failure
  • Update on progress of G-protein-coupled receptor 40 (GPR40) agonist HD-6277 (Hyundai)
  • Inhaled insulins Dance 501 (Dance) and Afrezza (MNKD)
  • Possible path forward for HDV-L (Diasome)
  • Chinese studies of PPARalpha/gamma/delta pan-agonist chiglitazar (Shenzhen)
  • Hopes for novel marketing of inexpensive drugs (in relation to NuSirt's NS-0200)
  • A GLP-1/FGF21 dual agonist (Duke University; preclinical)
  • Combination treatment with GLP-1/oxyntomodulin/peptide YY (Imperial College)
  • Preclinical data on liver-targeted mitochondrial uncouplers (Yale)
  • Information on suspended drugs atrasentan (ABBV) and LY3325656 (LLY)
  • Real-world evidence for Abbott's FreeStyle Libre.
For our disclosures, please read the Biomedtracker Research Standards.

Indications Covered: Acute Coronary Syndrome (ACS)
Chronic Heart Failure - Reduced Ejection Fraction (Chronic HFrEF)
Diabetes Mellitus, Type I
Diabetes Mellitus, Type II
Diabetic Nephropathy
Dyslipidemia / Hypercholesterolemia
Non-Alcoholic Steatohepatitis (NASH)

 Additional Resources: