Obesity: Disease or surrogate marker?

People get sick and die from the diseases triggered by obesity, not from obesity itself. Studying outcomes from the use of GLP-1s should also be on the negotiating table.

“Of all the forms of inequality, injustice in health is the most shocking and the most inhuman because it often results in physical death.”

— Dr. Martin Luther King, March 25, 1966

Dr. King spoke those words at a press conference just before addressing the Medical Committee for Human Rights convention in Chicago. Today, on the day set aside to commemorate his life and teachings, I offer those words as the touchstone for how this Substack column will cover health and health care over the next four years.

Now, on to today’s column.

The outgoing Biden administration on Friday placed the blockbuster obesity drug semaglutide atop the list of drugs slated for price negotiations later this year. The incoming Trump administration, whose HHS designee Robert F. Kennedy Jr. has expressed disdain for the drug, will decide if it stays there.

The drug, branded as Ozempic for diabetes and Wegovy and Rybelsus for weight loss, has already generated more than $50 billion in sales for its Danish manufacturer, Novo Nordisk. That is more than the company spent on research and development over the past three decades combined, undermining arguments by Big Pharma that high prices are needed to convince companies to invest in R&D. Physicians only discovered the drug’s effect on reducing appetite when it turned up as a side effect of a new Novo Nordisk diabetes drug, for which there were already many alternatives.

The drug’s average cost across the three brands is currently about $1,000 a month. An independent analysis of their medical benefits by the Institute for Clinical and Economic Review (ICER) found the prices needed to be cut in half to bring their costs in line with their benefits for individuals’ health.

There’s no doubt that semaglutide and similar drugs in the GLP-1 class shrink waistlines, which has significant health benefits. Obesity, designated a disease by the American Medical Association in 2013, is associated with increased incidence of type 2 diabetes, heart attacks, strokes, colon and breast cancer and many other diseases.

But obesity’s designation as a disease itself was controversial from the start. While it is correlated with higher incidence of illness, just like high blood pressure and elevated cholesterol are correlated with increased heart attacks and strokes, many obese people suffer from none of those diseases. Doctors also feared emphasizing drug therapy for obesity would distract people from pursuing better diets and regular exercise, a surer if more difficult to follow path to better health.

The Food and Drug Administration guidance for developing new drugs has a designation for factors associated with greater incidence of disease. They are called surrogate markers. When the FDA approves a drug based on a surrogate marker, it usually requires follow-up confirmatory trials to show that using the drug actually reduces incidence of disease.

Obesity’s designation as a disease meant that a drug could be approved simply by bringing weight down. No follow-up trials are required. When semaglutide branded as Wegovy was first approved in 2021, weight loss was the only indication on its label. Semaglutide branded as Ozempic had previously been approved to reduce blood sugar among people with diabetes.

Novo Nordisk, to its credit, launched a separate clinical trial to show that losing weight by taking semaglutide also improved health. The follow-up SELECT trial (Semaglutide Effects on Heart Disease and Stroke in Patients With Overweight or Obesity) showed semaglutide reduced heart disease, which led to the FDA giving Novo Nordisk the right to make that claim on the Wegovy label. The American Heart Association celebrated by claiming in a press release that it reduced heart disease by 20%.

It’s worth taking a closer look at the medical benefits shown in that trial. Millions of overweight people (not obese) will be taking this drug and other GLP-1s to look good, feel better and feel better about themselves. For many, the health benefits will be a side effect.

Relative v. absolute risk reduction

One of the first rules I learned when I began reporting on clinical trials was to investigate the absolute benefit of a new drug, not just its relative benefit. If a drug reduced risk of contracting a disease by 20%, what was the original risk? It was important to report the number needed to treat for one person to achieve those benefits to give readers better perspective on their magnitude.

Here’s a hypothetical example to explain the math. If a new drug is brought to market that reduces non-fatal heart attacks to 20 per 10,000 people taking the drug from 30 per 10,000 given a placebo, that is a relative risk reduction of 33% (30 minus 20 =10; 10 divided by 30 = 33%). Sounds pretty good!

But what is the absolute reduction in risk? Without the drug, there would be 1 heart attack per 333 people at risk of a heart attack. By taking the drug, that risk is reduced to 1 non-fatal heart attack per 500. That’s an improvement. But the risk to begin with wasn’t very large.

Now let’s turn to the SELECT trial, whose results were published in the New England Journal of Medicine in 2023. The company enrolled more than 17,600 obese and near-obese patients with a history of serious heart disease (all had previous heart attacks or strokes or had peripheral artery disease). Half took semaglutide. Half were given a placebo.

The trial’s primary measure (called the endpoint) combined deaths from heart attacks with non-fatal heart attacks and strokes. After two years, 6.5% of those taking semaglutide (569 out of 8,803) suffered one of those events compared to 8.0% (701 out of 8,801) among those on placebo. That’s an 18.8% reduction, which was rounded up to 20% in the AHA press release.

That sounds pretty good, and for FDA purposes, was sufficient to gain approval for reducing heart disease. However, the results were less than stellar if you looked at deaths alone. Deaths from heart attacks or strokes accounted for less than 40% of all negative events in both arms of the trial, with 39 more in the placebo arm. Those results were not statistically significant, meaning the slight reduction in heart attacks deaths in the semaglutide arm could have been due to chance.

Now let’s look at the results in terms of health care system costs at current prices. This non-significant outcome meant physicians would need to treat 226 obese patients with a previous history of heart disease for two years to prevent a single death. At $12,000 a year for Wegovy, treating 226 patients for two years would cost Medicare, private insurers and patients about $5.4 million. If we throw in the reduction in second but non-fatal heart attacks and strokes (thus making the overall results statistically significant), it would reduce the number needed to treat to prevent one cardiovascular event to 67 with total drug cost of about $1.6 million at current prices.

No wonder ICER found that the price of semaglutide and other GLP-1s would need to be drastically reduced before they met any rational cost-effectiveness standard.

Measure outcomes

One final thought. The patient population in the clinical trials were very sick people with multiple chronic conditions. Almost all were already taking drugs to treat high blood pressure, high cholesterol and elevated blood sugar. All had suffered previous cardiovascular events.

I suspect the patient population that has sent Wegovy demand soaring in recent years is far healthier than those who entered the clinical trials, who were carefully selected because of their serious illnesses and were most likely to benefit from immediate weight loss. Going forward, the only criteria for a prescription is being overweight or obese, not being sick. Moreover, doctors are allowed to prescribe weight-loss drugs off-label to people who only want to drop a few pounds.

This will help some people avoid advancing to more serious health conditions, no doubt. But it will also greatly inflate the cost side of these new drugs compared to their actual medical benefits.

Should the Trump administration decide to enter negotiations over the price of semaglutide drugs, it should seriously consider including as a condition for obtaining Medicare payment measuring the real-world outcomes from use of these new wonder drugs. The Centers for Medicare and Medicaid Services should require the drug companies to collect data on the subsequent health care needs and outcomes for everyone who gets a prescription. They should be required to offer those de-identified medical records to independent researchers so they can measure just how much medical benefit was achieved by use of the drugs beyond people simply losing weight.

Only then will we have a true measure of their actual benefit to the Medicare population and have a solid basis for setting their price