Are People With Eating Disorders Misusing GLP-1 Drugs?

 

• Among over 400 people with eating disorders in an interim analysis of an ongoing cross-sectional study, 32.1% said they have used GLP-1 drugs, and 22% reported current use.
• Overall, 10% said they have misused GLP-1 medications, and 9.9% reported using noncommercial compounded products.
• Pharmacovigilance of GLP-1 drugs is urgently needed, researchers said.

Use and misuse of GLP-1 receptor agonists were common in people with eating disorders, interim results of an ongoing cross-sectional study suggested.

Among over 400 people with eating disorders, 32.1% said they have used GLP-1 drugs, and 22% reported current use, wrote Nicholas C. Peiper, PhD, MPH, of the University of Louisville in Kentucky, and co-authors in a research letter in JAMA Psychiatry.

Notably, 10.1% said they have misused the medications, and 9.9% reported using noncommercial compounded products.

"People with eating disorders are a clinically diverse population who may be consuming GLP-1 RAs [receptor agonists] in contraindicated ways to maintain eating disorder psychopathology through rapid restriction and weight loss," the authors wrote.

GLP-1 drugs are currently indicated for the treatment of type 2 diabetes and obesity, as well as cardiovascular risk reduction, with recent expansions into treating chronic kidney disease and severe obstructive sleep apnea, Peiper and team noted. There are no indications for any eating disorders, including subtypes with higher cardiometabolic comorbidities, such as binge eating disorder or atypical anorexia nervosa.

This interim analysis provides "the first glimpse" of GLP-1 drug use in the eating disorder population, Sahib Khalsa, MD, PhD, of the David Geffen School of Medicine at the University of California Los Angeles, told MedPage Today.

Khalsa, who was not involved in the study, said that if data from the sample are nationally representative, then that suggests rates of GLP-1 agonist use among those with eating disorders are double that of the general adult population (15%), which "would be concerning."

The take-home message for the general practitioner with "even an inkling" that their patient might have an eating disorder is to assess for the problem and refer the patient to a mental health professional or dietitian, Khalsa said. If a GLP-1 drug is prescribed, clinicians should monitor for maladaptive appetite-related responses, new medical complaints, lab abnormalities, and other risky behaviors.

Peiper and colleagues also pointed out that people with eating disorders must currently navigate "a rapidly evolving risk environment of compounded GLP-1 RAs that are easily obtained without medical or regulatory oversight."

They stressed that "pharmacovigilance is urgently needed" as the market expands to include new products and formulations.

This interim analysis, conducted from December 2025 to January 2026, was from an ongoing study of people with eating disorders that began recruitment in March 2025 via "a network of federally funded studies unrelated to GLP-1 RA use, advertisements with national organizations, and targeted outreach through online recruitment platforms." Potential participants were asked to complete an online screening, and answer questions about GLP-1 drug use, if eligible.

Use of GLP-1 agonists was defined as use of semaglutide (Ozempic, Wegovy), tirzepatide (Mounjaro, Zepbound), dulaglutide (Trulicity), liraglutide (Victoza, Saxenda), or exenatide (Byetta) for diabetes or weight loss management. Misuse was defined as "taking more than the prescribed starting, titration, or maintenance dose; increasing the dose before prescribed; taking less than the prescribed dose; taking for a longer period than prescribed; disassembling or tampering with the injection equipment; or sharing with family, friends, or acquaintances without a prescription."

Peiper and team included 436 individuals who met the eligibility criteria. Mean age was 34, 94.2% were women, and 88.5% were white. Mean BMI was 28.5, and 27.2% had anorexia nervosa, 18% had binge eating disorder, 11.7% had atypical anorexia nervosa, 6.8% had bulimia nervosa, and 18% were in remission.

Psychiatric and medical comorbidities were common, including anxiety disorders, mood disorders, gastrointestinal diseases, and heart disease and hypertension. Adverse events were common among those reporting lifetime GLP-1 drug use (32.1%).

Looking at GLP-1 drug use by eating disorder subtype, the prevalence was lowest -- just over 10% -- among those with anorexia nervosa, and highest for those with atypical anorexia nervosa (over 40%) and binge eating disorder (over 50%).

"Understanding the reasons for that variability, and whether that is tied to harms would be important," Khalsa said. He noted that a 2023 open-label study he co-authored suggested a "potential clinical benefit" for one subpopulation: people with presumptive binge eating disorder and comorbid obesity. In this population, using GLP-1 drugs was linked to reductions in binge-eating episodes and cravings.

Limitations of the study included its cross-sectional design, nonprobability sampling, and reliance on self-report.

"Follow-up studies will provide more precise estimates among eating disorder subtypes, evaluate demographic differences, and examine key correlates, such as patterns of misuse, sources of acquisition, reasons for use, off-label use, and marketing exposures," the authors wrote.

Disclosures

This work was supported by a grant from the University of Louisville Joint Pilot Project Program.

Peiper reported grants from Abbott Laboratories and Gilead Sciences, and personal fees from Meru Health.

Co-authors reported grants from Novo Nordisk and AbbVie, and financial interest in Be Well and Sylvia Precision Health.

Khalsa reported no conflicts of interest.

Primary Source

JAMA Psychiatry

Source Reference: Peiper NC, et al "Use and misuse of GLP-1 receptor agonists among people with eating disorders" JAMA Psychiatry 2026; DOI: 10.1001/jamapsychiatry.2026.1716.

https://www.medpagetoday.com/psychiatry/eatingdisorders/121907

'Nature Medicine' Retracts Cancer Study Linking Treatment Timing to Survival

 After a 4-month investigation, Nature Medicine retracted a study that showed that the time of day in which immunochemotherapy was given for non-small cell lung cancer (NSCLC) had an effect on survival outcomes.

In the phase III trial known as LungTIME-C01, which was published in February, Chinese researchers reported that NSCLC patients who received immunochemotherapy in the morning or early afternoon had better survival outcomes compared with those who received treatment later in the day.

Nature Medicine editors announced that they were investigating concerns about some of the study's findings and protocol weeks later.

Now, "due to the amount and nature of the problems identified, the editors no longer have confidence in the integrity of the results," they wrote in a retraction note.

In the study of 210 patients with treatment-naive disease, median progression-free survival (PFS) was 11.3 months for those who received the first four cycles of immunochemotherapy before 3 p.m. compared with 5.7 months for those treated after 3 p.m. (HR 0.40, 95% CI 0.29-0.55, P<0.001). Median overall survival was 28.0 months in the early group and 16.8 months in the late group (HR 0.42, 95% CI 0.29-0.60, P<0.001).

The journal's editors noted that concerns were raised about inconsistencies between the registration record of the trial on clinicaltrials.gov and the published version of the study protocol, and that key elements of the study -- including endpoints, eligibility criteria, sample size, and study design -- were inconsistently reported and modified over time.

Moreover, there were discrepancies between the original version of the study protocol in Chinese and the translated versions provided for peer review and publication.

There were also concerns about unexpected data patterns in the study, including:

• The shape of the PFS curve being smoother than what is most commonly observed in similar large phase III trials
• The absence of censoring in the first year of the study
• The absence of adverse events leading to treatment discontinuation throughout the trial
• Similar rates of immune-related adverse events in the trial's two arms despite the differences in therapeutic efficacy

"These concerns were compounded by the source data provided by the authors indicating that randomization was performed on the day of treatment for almost all patients, which is uncommon; and their acknowledgement of deviations from fixed-calendar imaging schedules due to COVID-19 related delays and cycle-based assessment, indicating a lack of adherence to standard RECIST [Response Evaluation Criteria in Solid Tumors] timing," the editors wrote.

According to the retraction note, a number of the study's authors agreed with the retraction, while several others failed to respond to correspondence from Nature Medicine.

https://www.medpagetoday.com/hematologyoncology/lungcancer/121913

Health Spending Projected to Hit Almost $9 Trillion in 2034

 

• National health spending is projected to reach nearly $9 trillion by 2034.
• At the same time, the percentage of Americans covered by health insurance is expected to drop, due in part to expiring subsidies for some Affordable Care Act marketplace plan enrollees.
• Medicare spending for physicians and other clinical care is also expected to decline because of a rule limiting reimbursement for skin substitutes.

Health expenditures are expected to reach $8.97 trillion by 2034, while the percentage of the population with health insurance is expected to decline for the next several years, said researchers at the Centers for Medicare & Medicaid Services (CMS).

"The main [cost] drivers are high utilization growth across those services, and notably rapid growth in retail prescription drug spending," Jacqueline Fiore, PhD, an economist in the national health statistics group at the CMS Office of the Actuary, said during a reporter briefing. "Significant legislative provisions from the Inflation Reduction Act and the One Big Beautiful Bill Act are anticipated to affect health spending and insurance coverage trends through 2028. Together, these legislative provisions play a role in reducing the insured population through 2028. And finally, the federal government's share of health costs is projected to increase through 2034, mainly from Medicare."

Overall, "health spending growth is projected to average 5.4% [annually] over the projection period, and is expected to grow more rapidly than the gross domestic product," she continued. This means that healthcare's share of the economy will increase from 18.0% in 2024 to 20.6% in 2034. The health expenditure projections were reported by Fiore and colleagues in Health Affairs.

But as spending on healthcare rises, private health insurance spending growth is projected to slow from 8.2% in 2025 to 6.3% in 2026, "in large part as a result of a projected decline in private health insurance enrollment," the authors noted.

The reduction is due in large part to the expiration of enhanced premium tax credits geared toward helping patients buy health insurance policies on the Affordable Care Act's health insurance exchanges. The expiring tax credits "contribute to a projected decline in direct-purchase insurance of 3.7 million enrollees," and means the population with health insurance is expected to drop from 91.7% in 2025 to 90.8% in 2026, increasing the number of uninsured people by 3 million, they explained.

In terms of who will be funding healthcare, "over the course of 2025 to 2034, Medicare is projected to experience the highest spending growth, 7.7% per year on average, among the major sources of funding, reflecting faster per-enrollee spending growth and faster enrollment growth, partially driven by demographic factors," Fiore said. "For the major spending categories, retail prescription drugs is the fastest growing major spending category over the projection period. This is driven by projected elevated growth and utilization in 2025 and 2026, partly attributable to continued demand for high-cost retail prescription drugs, particularly for people enrolled in Medicare and private health insurance."

Medicare physician spending will be affected in the next few years by a rule adopted in the fiscal year 2026 final Physician Fee Schedule payment rule to address a problem Medicare was having with overpayment for skin substitutes used in wound healing. As MedPage Today reported, the Office of the Inspector General (OIG) at HHS found that spending on skin substitutes reached more than $10 billion in 2024 -- more than 15% of the $66 billion in total spending for all Medicare Part B drugs that year, according to the HHS watchdog.

In response, a new rule adopted by Medicare is expected to reduce skin substitute payments by more than 90%. "This change is expected to exert downward pressure on Medicare spending for physicians and other professional providers who use these products in patient care. As a result, overall other professional services spending growth is expected to slow from 5.8% in 2025 to 2.0% in 2026," Fiore and colleagues wrote.

They also projected that during the years 2029-2034, spending on physician care and other clinical services would average around 5.5% annually, a figure that is based on historical trends, according to co-author John Poisal, deputy director of the national health statistics group within the CMS Office of the Actuary.

"We oftentimes find ourselves, if there's a lack of better information, looking to the history," he told MedPage Today. "You drive a car by looking in the rearview mirror in a lot of ways to see where you've been and use that to help where you're going."

In contrast with the moderating trend later on, rates of growth in Medicaid physician and clinical services spending are projected to have accelerated from 4.2% in 2024 to 9.4% in 2025, according to the researchers. (Numbers for 2024 and 2025 are projections because even though those years have passed, it takes a while for researchers to do their analysis.) In addition, they wrote, "for medical goods, total Medicare prescription drug spending growth is projected to have remained elevated at 12.6% in 2025 (compared with 12.9% in 2024), as Medicare beneficiaries continued their high utilization of GLP-1s for various indications, coupled with increased prices of expensive oncology drugs used to treat certain types of cancers, including blood, renal, liver, and thyroid cancer."

Disclosures

The authors reported no conflicts of interest.

Primary Source

Health Affairs

Source Reference: Fiore JA, et al "National health expenditure projections, 2025-34: strong utilization growth initially, legislative impacts later" Health Aff 2026; DOI: 10.1377/hlthaff.2026.00642.

https://www.medpagetoday.com/publichealthpolicy/healthpolicy/121920

'ALS Testing Expected to Drive Sharp Rise in Clinic Visits'

 

• Genetic ALS testing may sharply increase specialized clinic visits as more relatives learn their carrier status.
• Projected gene-related ALS cases and asymptomatic carriers may nearly triple in the U.S. by 2035.
• Researchers warned that ALS centers should prepare for rising demand and surveillance needs.

Genetic testing of people with a family member diagnosed with amyotrophic lateral sclerosis (ALS) will greatly increase the number of clinic visits to specialized ALS centers over the next decade, a population model showed.

Based on estimates of four common gene variants -- SOD1, C9orf72, FUS, and TARDBP -- 2,704 people in the U.S. had a symptomatic gene-related form of ALS in 2026 and 10,944 people were asymptomatic gene carriers, reported Jennifer Morganroth, MD, of Massachusetts General Hospital in Boston, and co-authors.

By 2035, those numbers are expected to rise to 7,474 symptomatic ALS patients and 26,111 asymptomatic gene carriers, the researchers wrote in Neurology Genetics.

Fewer than 50 extra clinic visits were needed annually per ALS center in most states in 2026, with 12 states needing 50 to 99 additional clinic visits and no states surpassing 100. By 2035, only six states would remain below 50 visits per ALS center annually; 22 states would have 50 to 99 visits, 18 states would have 100 to 199, and three would exceed 200.

"As more at-risk relatives get genetic testing, the testing will identify more gene carriers, so specialized ALS centers must plan for the long-term care of more people," Morganroth said in a statement.

"Anticipating the clinical needs of people with a genetic risk for ALS, and which states may see the greatest increases in this patient population, is essential for improving care and ensuring that clinics are ready as new therapies become available," she continued. "This will become increasingly important as gene-targeted therapies, biomarker monitoring, and preventive trials continue to emerge."

The first gene-targeted ALS treatment -- tofersen (Qalsody), an antisense oligonucleotide that targets mRNA to reduce SOD1 protein synthesis -- was approved in 2023 and has slowed decline in some patients with SOD1-ALS. Other drugs, like investigational jacifusen for FUS-ALS, are being studied. Consensus guidelines recommend that genetic testing should be offered to all people with ALS, regardless of their family history.

ALS clinics already contend with growing numbers of patients with symptomatic ALS amid deep funding cuts, Morganroth and co-authors noted.

"While current ALS center capacity may meet present needs, many clinics are already operating at or near full capacity caring for symptomatic patients, and recent funding cuts to the Muscular Dystrophy Association and ALS Association further threaten sustainability," they wrote.

"These financial and operational pressures underscore the fragility of the current system and how even modest increases in patient volume could challenge care delivery," they added.

In their analysis, the researchers used two complementary methods to estimate ALS prevalence by state. In one approach, they applied race-specific ALS prevalence and incidence rates from a population-based study in the metropolitan Atlanta region to 2023 census demographic data. In the other, they used state-level ALS prevalence and incidence estimates derived from the 2011-2018 National ALS Registry, which identified cases in Medicare, Veterans Health Administration, and Veterans Benefits Administration records.

Morganroth and colleagues calculated the number of gene-positive cases using published frequencies of pathogenic variants, assuming 7% of ALS cases were attributable to C9orf72 mutations, 2% to SOD1, 0.5% to FUS, and 0.4% to TARDBP. They estimated an average of 4.25 carrier relatives per proband and one annual ALS center visit for each asymptomatic carrier.

The researchers acknowledged that their projections, anchored to a modeled first year of 2026, are not precise forecasts. They also noted that ALS center counts were based on the I AM ALS directory, which may not include unaffiliated neuromuscular centers.

The study excluded de novo variants, assumed in-state care, and did not account for family mobility across states, Morganroth and colleagues pointed out.

"Future work should integrate laboratory and registry data, demographic-specific penetrance, family mobility, and observed testing behavior, and evaluate the feasibility and effectiveness of dedicated carrier clinics and telehealth-based surveillance," they wrote.

Disclosures

Morganroth reported no conflicts of interest.

One co-author reported relationships with the NIH, ALS Association, ALS Finding a Cure, Ionis, Empire ALS Alliance, Target ALS, Merck, Mitsubishi Tanabe, Corcept, Biogen, Ionis, Prilenia, uniQure, Insmed, the Muscular Dystrophy Association, NEALS, the American College of Medical Genetics, ML BridgeBio, Neurizon, GenieUS, Leidos, Shinogi, and Regeneron. He also provided medical-legal consulting for several groups and held editorial roles with Muscle and Nerve and the Journal of Neuromuscular Disease.

Primary Source

Neurology Genetics

Source Reference: Morganroth J, et al "Preparing amyotrophic lateral sclerosis clinics to provide longitudinal care for individuals carrying ALS risk variants" Neurol Genet 2026; DOI: 10.1212/NXG.0000000000200406.

https://www.medpagetoday.com/neurology/generalneurology/121922

Long-Term CAR-T Results in Lymphoma Confirm Early Promise

 

• Almost a third of patients with aggressive lymphoma remained cancer free 10 years after treatment with CAR T-cell therapy.
• The long-term data indicate that CD19-directed CAR T-cell therapy may cure some patients with relapsed/refractory lymphoma.
• Emerging technologies could substantially improve treatment outcomes with CAR T-cell therapy.

Ten years after receiving a single dose of CD19-targeted CAR T-cell therapy, almost a third of patients with aggressive lymphomas and no treatment options remained lymphoma free, updated results from a landmark trial showed.

Overall, 12 of 38 patients remained lymphoma-free at 10 years, half with heavily pretreated large B-cell lymphoma (LBCL) and half with relapsed/refractory follicular lymphoma. No relapses occurred after 5.4 years of follow-up, and no patient with follicular lymphoma relapsed after 2.7 years. The 10-year overall survival was 17% among 24 patients with LBCL and 50% for the 14 with follicular lymphoma.

The results show that CD19-directed CAR T-cell therapy not only can achieve long-term remissions but may possibly cure some patients, reported Marco Ruella, MD, of Penn Medicine in Philadelphia, and colleagues in the New England Journal of Medicine.

The results exceeded even the most optimistic expectations at the start of the trial.

"I thought none of [the patients] would be alive at 5 years, and it turned out that most of them were," co-author Stephen Schuster, MD, also of Penn Medicine, told MedPage Today. "Once I saw the 5-year data, I thought 'Oh my god, maybe -- I hate to say cure -- but maybe we've cured patients.' We addressed that question by saying, well, if you live as long as an age- and gender-matched American and you die of unrelated causes and your lymphoma is in complete remission, then that's essentially a cure. Sure enough, our patients matched what you would kind of expect, so I think we can safely say we're curing some patients, and I think we'll cure more in the future."

Another CAR T-cell researcher had high hopes early on, especially after seeing data from single-center studies conducted at the University of Pennsylvania and elsewhere.

"We were super excited and hopeful that [CAR T-cell therapy] would lead to long-term remissions, or at least high remission rates for patients without other options," said Frederick Locke, MD, of Moffitt Cancer Center in Tampa, Florida, who helped organize and coordinate multicenter trials. "The durability was unclear. This study is just now at 10 years and we treated our first patient in 2015, so about 10 years ago. We were hopeful, and we couldn't have been more pleased with the results we saw. Now the therapies are FDA approved and more widely available, although we need to do more work to make them even more available."

Study Background

The Philadelphia study involved the CAR construct known as CTL019, subsequently named tisagenlecleucel (Kymriah). Investigators enrolled 49 patients from January 2014 to September 2019, and data cutoff occurred Oct. 1, 2025. Eleven patients did not receive the CAR T-cell therapy, five because of rapid disease progression, five because the manufacturing yield was inadequate for the specified dose, and one patient withdrew consent.

The overall 10-year lymphoma-free survival was 38% (defined as the time from CAR T-cell infusion to relapse or death), including 32% in the patients with LBCL and 47% for those with follicular lymphoma. Most treatment failures or relapses occurred within the first year after infusion, consistent with previous reports. An analysis that included deaths from any cause showed a 10-year progression-free survival of 17% in the patients with LBCL and 29% in the follicular lymphoma group. The 10-year incidence of non-relapse-related death was 18%, decreasing to 14% with exclusion of deaths related to COVID.

With respect to safety, two patients had persistent grade 2 or 3 neutropenia. No patients developed late anemia or thrombocytopenia. Nine patients developed second primary cancers, consisting of three cases of treatment-related acute myeloid leukemia, two cases each of prostate cancer and non-small cell lung cancer (both involving smokers), and one each of mycosis fungoides and melanoma. The 10-year cumulative incidence of second primary cancer was 21%, as compared with a 10-year cancer risk of 11% for a matched population from the Surveillance, Epidemiology, and End Results database.

Best Yet to Come

Schuster has high expectations for continued improvement in outcomes with CAR T-cell therapy. Development of more effective therapies for lymphoma, such as bispecific antibodies, is one factor contributing to the optimism.

"We've learned that the most important factor for successful outcome, particularly in large B-cell lymphoma, is disease control prior to CAR T-cell therapy," said Schuster. "Patients [whose disease] is actively growing are the ones more likely to fail. Those who have at least stable disease or partially responding disease to any kind of therapy prior to CAR T cells are likely to have a better response."

Increasingly sophisticated assays for circulating tumor DNA will help determine the extent and depth of response to preceding therapies to guide decisions about continuing to CAR T-cell therapy. Recent studies have shown that the Bruton's tyrosine kinase inhibitor ibrutinib (Imbruvica), given before and after CAR T-cell infusion, enhances T-cell activation toward more cytotoxic T cells, which could make CAR T-cell therapy more effective.

Ongoing studies to develop rapid culturing techniques offer the potential to speed up the CAR T-cell production process and also produce more effective and durable therapy, said Schuster. The next generation of studies will use less differentiated T cells that are capable of serial cancer cell killing and then regenerating themselves. A shortened processing time would be especially beneficial for patients with aggressive B-cell lymphomas. A recent study showed that only 25% of patients potentially eligible for CAR T-cell therapy actually received the treatment, owing to access issues, production issues, and patient-related characteristics.

Further on the horizon, off-the-shelf allogeneic CAR T-cell products appear to be a viable alternative to autologous cells. Preliminary studies suggest results could rival those of conventional autologous products.

"That totally gets away from manufacturing, and that would also include patients that would fail manufacturing because they have bad lymphocyte counts," said Schuster.

Also further out, so-called in vivo CAR T-cell therapy offers potential to get around obstacles posed by the production process, said Locke.

"The CAR T cells are made within the patient's own body, instead of taking the T cells out and delivering the vector to the CAR T cells and expanding them in a laboratory and then reinfusing them into the patient," he said.

Another evolving technology that could eliminate some of the production and processing obstacles involves lipid nanoparticles that deliver the "CAR machinery" to T cells.

"We've seen some pretty incredible clinical results in China and Australia," said Locke. "We need to get those therapies to the United States and tested more rapidly. The regulatory and other hurdles are different in China and Australia. We'd like for the U.S. to be the first place to test these therapies. But we're still excited for our patients that we're seeing great results."

Said Schuster, "All these things are converging, I believe, to give us a future that will show us much better results than these results that are now 10 years old."

Disclosures

The study was supported by the Richard Berman Family Innovations Center in CLL and Lymphomas.

Ruella disclosed relationships with AbClon, Acera, Bayer, Beckman Coulter, Biocartis, Bristol Myers Squibb, Curiox Biosystems, the Gerson Lehrman Group, GlaxoSmithKline, Lumicks, ModeX, NanoString Technologies, ONI, PeproMene, Sana Biotherapeutics, and Vittoria Bio.

Schuster disclosed relationships with AbbVie, ADC Therapeutics, AstraZeneca, BeiGene, Bristol Myers Squibb, Caribou Bioscience, Genentech, Genmab, Janssen, Kite Pharma, MorphoSys, Novartis, and Vittoria Bio.

Locke disclosed relationships with E.R. Squibb & Sons, Pfizer, Iovance Biotherapeutics, Janssen, Novartis, Kite Pharma, Gilead Sciences, and Celgene.

Primary Source

New England Journal of Medicine

Source Reference: Ruella M, et al "Ten-year outcomes after CAR T-cell therapy for B-cell lymphomas" N Engl J Med 2026; DOI: 10.1056/NEJMoa2518035.

https://www.medpagetoday.com/hematologyoncology/lymphoma/121923