Reading · Weight and Healthcare

This is the Weight and Healthcare newsletter! If you like what you are reading, please consider subscribing and/or sharing!

I’ve received a ton of reader questions about Eli Lilly’s GLP-1 weight loss pill Foundayo, which was recently granted FDA approval. In Part 1 we’ll talk about the study itself, in Part 2 we’ll break down the discussion, limitations and talk about the special treatment the government gave this drug.

Summary

A group of authors, all of whom are either taking money from Eli Lilly or employees/stockholders of Eli Lilly conducted a trial of a weight loss drug manufactured by Eli Lilly which Eli Lilly not only funded but also designed and oversaw, including with its employees guiding everything from the medical oversight, to statistical analysis, to writing the first draft. The study tested Orforglipron, brand name Foundayo, a daily GLP-1 weight loss pill. The trial tested doses of 6mg, 12mg, and 36mg (after a titration up from 1mg) against a placebo group. All groups were given the same advice around healthy eating and movement. At the end of the 72 week trial, 34.7% of the pooled drug group failed to lose 5% of their body weight, 57.4% of the pooled drug group failed to lose 10%, 76.2% failed to lose 15%, and 88.7% failed to lose 20%. While all groups, including placebo, experienced health benefits, the drug groups experienced relatively small but statistically significantly greater benefits. That said, the authors didn’t actually track behaviors and so it is not clear if the drug or if behaviors created the differences in actual health markers, rather than weight loss.

Deeper Dive

We’ll start, as we always do, with the authors. I’ve summarized the disclosed conflicts of interest (with conflicts with the drug manufacturer highlighted) and for those who appear in the OpenPayments database I’ve included information on the monetary amounts of conflicts from 2019-2025. I will say that I find it interesting that most of the authors who aren’t Eli Lilly employees are taking money from both Novo Nordisk and Eli Lilly, who are rivals in the GLP-1 space. You can find the full study disclosures here.

Sean Wharton, M.D.

Disclosures:
AbbVie, Amgen Canada, AstraZeneca, Bausch and Lomb, Biohaven Pharmaceuticals, Inc., Boehringer Ingelheim, Eli Lilly and Company (Speaking Engagement, Advisory Board), Merck, Novo Nordisk (Scienti c Advisory Board, Academic Speaking Engagement, Academic Advisory Board), Regeneron Pharmaceuticals

Dr. Wharton lives in Canada which does not offer transparency into the amount of pharma patients.

Louis J. Aronne, M.D.

Disclosures: Altimmune, Amgen Inc., AstraZeneca, Boehringer Ingelheim, Clic Bio, Eli Lilly and Company (Consultant and investigator), ERX pharmaceuticals, Intellihealth, Jamieson Wellness, Janssen Biotech, Jazz Pharmaceuticals Inc., Metsera, Novo Nordisk (Consultant / Advisory Board Member, Research Funding), Pfizer, Senda biosciences, Skye bioscience, Syntis, Verdiva Bio, Versanis, Veru Inc, Weill Cornell Medical College, Zealand Pharma A/S

OpenPayments:
General Payments: $316,287.49
Research Payments: $4,144.22
Associated Research Funding: $1,149,211.69
Eli Lilly is his top patron in all categories.

Adam Stefanski, M.D., Ph.D

Disclosures: Eli Lilly and Company (Employee and Stock Holder)
Title: Associate VP-Research & Development-Clinical Research – Cardiometabolic Health
Role: Medical role to provide scientific and clinical support during late phase drug development

Nasreen F. Alfaris, M.D., M.P.H.,6

Disclosures: Eli Lilly (Consultant)

OpenPayments:
General Payments: $10,811.50
(All from Eli Lilly)

Andreea Ciudin, M.D., Ph.D.

Disclosures: Boehringer Ingelheim, Eli Lilly and Company (Grant recipient, consultant), Novo Nordisk (consultant, education)

Dr. Ciudin lives in Spain and so is not included in OpenPaymentsKoutaro Yokote, M.D., Ph.D.

Disclosures: Astellas Pharma, Bayer yakuhin, Daiichi Sankyo Company LTD, Eli Lilly Japan (speaking, consulting), KOWA COMPANY, LTD., Mitsubishi Tanabe Pharma Corporation, Nippon Boehringer Ingelheim Co., Ltd., Novartis Pharma, Novo Nordisk (advisory board, speaking), Pfizer, Sanofi, Sumitomo Dainippon Pharma Co., Ltd., Taisho, Takeda Pharmaceutical Company, Limited

Dr. Yokote lives in Japan and is not included in OpenPayments

Bruno Halpern, M.D., Ph.D.,

Disclosures: Boehringer Ingelheim, Currax Pharmaceuticals LLC, Eli LIlly (Advisory Board), Novo Nordisk (Advisory Board)

Dr. Halpern lives in Brazil and is not included in OpenPayments

Alpana P. Shukla, M.D.

Disclosures: Eli Lilly and Company (Clinical Trial-Site Principal Investigator), Novo Nordisk (Clinical Trial Site Principal Investigator), Sun Pharmaceuticals

OpenPayments

General Payments: $4,883.47
Research Payments: $57,313.20
Associated Research Funding: $1,583,074.76

Eli Lilly is the top patron in every category

Chunmei Zhou, M.S.

Disclosures: Eli Lilly (employee and stock holder)
Title: Director
Role: Project statistician

Lisa Macpherson, M.S.P.H.

Disclosures: Eli Lilly (employee and stock holder)
Title: Director of Statistics
Role: Serve as the statistical lead for this study

Sheryl E. Allen, M.D.,

Disclosures: Eli Lilly (employee and stock holder)
Title: Executive Director
Role: Medical Director for Attain 1

Nadia N. Ahmad, M.D., M.P.H.

Disclosures: Eli Lilly (employee and stock holder)
Title: Associate Vice President
Role: Combined Management and Subject Matter Expert role, leading late phase development programs (Phase 3 trials) for ob*sity medications

Suzanne R. Klise, B.S.

Disclosures: Eli Lilly (employee and stock holder)
Title: Executive Director - Clinical Research Scientist
Role: Serve as medical lead for Phase 3 clinical research

The Basics

As always, quotes from the study will be indented, you can skip them and still get the gist of the piece.

This was a 72 week, double blind placebo controlled trial of Orforglipron, brand name Foundayo, a GLP-1 weight loss pill created by Eli Lilly. The trial was funded by Eli Lilly, but they did way more than fund it:

The trial sponsor, Eli Lilly, designed the trial and oversaw its conduct. Trial investigators were responsible for data collection. The sponsor performed site monitoring, data collation, and data analysis. The investigators and authors worked under confidentiality agreements with the sponsor. The authors, with assistance from sponsor-funded medical writers, wrote the first draft of the manuscript. All the authors had access to the data and analyses, interpreted the data, critically reviewed the manuscript, approved the decision to submit it for publication, and vouch for the accuracy and completeness of the data and for the fidelity of the trial to the protocol.

Participants
The study include people 18 and older with a

“BMI of at least 30 or who have a BMI between 27 and 30 and who have at least one ob*sity-related complication, including hypertension, dyslipidemia, cardiovascular disease, or obstructive sleep apnea, and a history of at least one patient-reported unsuccessful dietary effort to lose body weight.”

You’ll notice that “ob*sity-related complications” are health conditions that people of all sizes get that get called “ob*sity-related” when higher-weight people have them. They are also continuing to use the BMI threshold of 27. This is curious because BMI is a deeply problematic and unhelpful metric and, in fact, the only thing it has going for it is precision - per BMI, the “overw*ight” category starts at a BMI of 25, yet these GLP-1 trials consistently start at 27, suggesting to me that they looked at the data from their Type 2 Diabetes trials and cherry-picked this BMI range to get the largest effect during the trial period.

The mean age of the patients was 45 years; 64.2% were women; and 56.5% were white, 28.6% were Asian, and 8.6% were Black, 37.6% were Hispanic or Latino. At randomization, the mean body weight was 103.2 kg and the mean BMI was 37.0. A total of 46.0% of the patients had a BMI of less than 35, and 36.0% of the patients had “prediabetes”.

Procedures

The patients were randomly assigned to Orforglipron doses of 6mg, 12mg, 36mg, or placebo. They were assigned 3:3:3:4.

I’m going to interrupt the study analysis to clear up some misunderstandings I’ve seen floating around. I’ve seen well-meaning people saying things like “the 6mg minimum dose of Foundayo is 12 times higher than the 0.5mg minimum dose of semaglutide!” I understand why people are confused but, also, that’s not how it works. The dosages between drugs are not comparable. In fact, the dosages between injection and pill are also not comparable, even if the medication is the same. While we’re on the subject, the medication in Novo Nordisk Wegovy pill is semaglutide, the same medication that is in the Wegovy injection, but the medication in Eli Lilly’s Foundayo pill (Orforglipron) is NOT the same medication as in Eli Lilly’s injectable Zepbound (tirzepatide). Moving on…

Orforglipron is a capsule that is taken daily. As is the case with all GLP-1s, whether used for actual health benefits or weight loss, there was a titration up from a low dose. Subjects were started at 1mg dose and then titrated up each week until they got to the dose their group was assigned (6 mg at 8 weeks, 12 mg at 12 weeks, and 36 mg at 20 weeks.) The goal of the low starting dose (often called the subtherapeutic dose) is to get the body used to the medication and hopefully reduce side effects as the dose is titrated up. Because this is a weight loss dose, the medication is relentlessly uptitrated to the assigned dose (unlike GLP-1s for Type 2 Diabetes where the goal is to give the minimum dose to reach desired glycemic management and minimize side effects.) Since weight loss is simply a side effect of these drugs, the goal of dosing/titration is to maximize the side effect of weight loss.

Each participant also received “individualized lifestyle counseling focused on a healthy, balanced diet combined with physical activity.” This is good news in that they did not recommend caloric restriction (like other trials have,) however there was no mechanism to track whether/how participants practiced this advice and thus no way to determine if behavior changes, and not the drugs, or weight loss were responsible for health benefits.

The primary end point was “percent change in body weight from baseline to week 72” and “Multiplicity-adjusted secondary end points were the percentage of patients who had a reduction in body weight of at least 5%, 10%, 15%, or 20% at week 72, along with the change from baseline to week 72 in waist circumference, systolic blood pressure, non–high-density-lipoprotein (HDL) cholesterol, and triglycerides. Additional secondary end points included changes in glycemic measures, diastolic blood pressure, and other lipid measures. Changes in body composition from baseline to week 72 were assessed for a subgroup of patients by means of dual-energy x-ray absorptiometry (DXA)”

They determined that “a sample size of 3042 patients would provide the trial with at least 90% power to show the superiority of individual doses of Orforglipron over placebo for the primary end point.” The study started with 3127 people. Administration of Orforglipron or placebo was continued throughout the 72-week trial period in 75% of the participants (2344 patients). The dropout rate in the drug groups increased slightly with dose with 21.9% of the 6mg group failing to stay on the drug for the full 72 week trial, 22.5% of the 12mg group failing to stay on the drug for the 72 week trial, and 24.4% of the 36-mg group failing to stay on the drug for the full trial. In the placebo group, 29.9% failed to stay on the placebo for the full trial.

This deserves some unpacking because these are fairly high dropout rates in both the drug and placebo groups, but not necessarily for the same reason. It’s important to remember that people who joined these trials wanted to join a weight loss trial. In the placebo group, withdrawal can be because they don’t experience any of the expected side effects or any weight loss and so they realize they are part of the placebo group and they drop out of the study. In the weight loss groups, you would expect more motivation to stay in the group, but they are also experiencing side effects. In this case, 13.8% of the placebo group withdrew due to “patient’s decision” while 8.5-8.9% of the drug group listed this reason. 5.1%-10.3% of the drug group withdrew due to “adverse events” (with higher numbers at higher doses) while only 2.6% of the placebo group did.

Weight-related findings

Average weight loss was:
6mg: 7.5%
12mg: 8.4%
36mg: 11.2%

If we dig a bit deeper:

Failed to lose 5%
6mg: 39.4%
12mg: 36.5%
36mg: 28.2%

Failed to lose 10%
6mg: 66.7%
12mg: 60%
36mg: 45.4%

Failed to lose 15%
6mg: 84.9%
12mg: 79.7%
36mg: 64%

Failed to lose 20%
6mg: 93.6%
12mg: 91%
36mg: 81.6%

Additional Treatment Outcomes

Here they report that the drug “significantly improved cardiometabolic risk factors” including “systolic blood pressure, non-HDL cholesterol, and triglycerides.”

To measure these, they pooled all of the dose groups together which is… a choice that we’ll discuss more in part 2. For now, they reported that the difference in systolic blood pressure was a pooled average of -5.7mmHg (−5.0 to -6.3), while the placebo group averaged −1.4 (-.05 to −2.4). The difference in non-HDL cholesterol in the pooled drug groups was −6.7 (-5.8 to −7.6), while the placebo group averaged −1.9 (−0.2 to -3.6). The difference in triglycerides in the pooled drug group was −14.8 (-13.3 to -16.3) while the placebo group averaged −3.8.

They also point out that Orforglipron was associated with improvements in diastolic blood pressure, other lipid fractions, high-sensitivity C-reactive protein… as well as levels of glycated hemoglobin, fasting glucose, and fasting insulin”

The fasting glucose changes aren’t a big shocker - remember these aren’t really weight loss medications, they are type 2 diabetes medications with a side effect of weight loss. This also begs the question - if this group of people were given a different Type 2 Diabetes medication, would they have seen similar improvements?

Regardless, both the drug and placebo groups showed improvement in all areas with relatively small, but statistically significant, differences between the pooled drug group and the placebo group, though there was some overlap between the individual drug groups and the placebo groups. As a reminder, statistical significance is not a measure of the size or importance of the difference, it just means that, statistically, it’s more likely that the difference was due to the medication than due to chance (more about statistical significance here). Except there is a big confounder here - behaviors.

I have often said that there should be weight-neutral comparator groups in these trials. Now, this wasn’t actually weight-neutral since it was counseling that was done under the guise of a weight-loss trial, but at least they didn’t recommend caloric restriction. Again, an issue with the trial is that both the drug and placebo groups are given advice about food and movement (to be clear, there are many other aspects of health but it seems that food and movement are always the myopic focus of this kind of “counseling”.) Since both groups are given the same advice about food and movement, we are meant to believe that this will control for whether or not the drug/drug-induced weight loss is actually creating any health benefits. But as my stats professor used to say “if you didn’t track it, you can’t claim it” and this is definitely the kind of thing she was talking about.

For that control mechanism to work, we would have to know if the actual behaviors were the same in each group. Let’s look at one of many possible examples. If the placebo group realized that they were on the placebo, then they may have been less motivated to exercise may have made different food choices than the people on the drug, who may have felt more motivated to exercise. If that’s the case, increased movement and different food choices in the drug group could be responsible for all the differences in actual health benefits we see here. Of course we have no idea if that happened. It’s not enough to give food and movement “counseling,” these trials should actually track the behavior they are “counseling” people about, especially since there is a body of research that finds that behaviors can have significant impacts on health.

So that’s the study, in Part 2 we’ll break down the authors’ interpretations of the results and look at the very special treatment the government gave this drug very special treatment.

If you appreciate the work I do here, you can support my ability to do it by becoming a free or paid subscriber!

Liked the piece? Share the piece!

Reader Question - What is Foundayo? Part 1 - Study Basics

Comments