GLP-1s and HIV — what PLWH need to know in 2026.

For most of the epidemic, the weight conversation in HIV care ran in one direction. Wasting was the emergency. Gaining was the goal. Clinic scales measured survival, and a pound back on the frame was something to celebrate. That framing shaped a generation of care, and it is still lodged in a lot of us who lived through it.

Two things have changed. Modern antiretroviral therapy works so well that people living with HIV are now aging into the same cardiometabolic conditions as everyone else — plus a few that hit harder in HIV. And a class of medicines called GLP-1 receptor agonists — semaglutide, tirzepatide, liraglutide, dulaglutide, sold as Ozempic, Wegovy, Mounjaro, Zepbound, Saxenda and Trulicity — has moved from diabetes clinics into general conversation, including in HIV clinics.

So the question arrives, over and over, in support groups and waiting rooms and text threads: can I take this? This page answers that as honestly as the evidence allows in 2026.

Why people living with HIV are asking about GLP-1s

GLP-1 stands for glucagon-like peptide-1, a hormone your gut releases after you eat. It tells the pancreas to release insulin, tells the liver to ease off making glucose, slows how fast the stomach empties, and signals the brain that you have had enough. GLP-1 receptor agonists are engineered, longer-lasting versions of that signal; tirzepatide adds a second target, GIP, which is why it is called a dual agonist.6 Two features explain why they matter here: they were built for type 2 diabetes and turned out to produce weight loss well beyond what earlier medicines managed, and in people without diabetes but with cardiovascular disease and excess weight, semaglutide reduced major cardiovascular events — a finding that changed how the whole class is understood.15

Cardiometabolic health is not a side issue in HIV. A 2024 systematic review and meta-analysis pooling 102 studies and 78,700 participants found metabolic syndrome — the cluster of abdominal weight, blood pressure, blood sugar, triglycerides and HDL that predicts diabetes and heart disease — in 25.3% of people living with HIV overall, 25.6% of those on antiretroviral therapy, and 30.4% in the Americas, the highest of any region studied.4 Roughly one in four globally; closer to one in three here.

Layer on the specifics. People living with HIV carry roughly twice the cardiovascular risk of people without HIV, driven by chronic inflammation and immune activation that persist even with an undetectable viral load.7 Metabolic dysfunction-associated steatotic liver disease — fatty liver — is common. Visceral fat, the deep abdominal kind, behaves like an inflammatory organ rather than a storage closet. And some modern antiretrovirals are associated with weight gain in a way the field is still arguing about.

A word about the frame

Nothing on this page says your body is a problem to be fixed. Weight is not a moral category, and a lot of people living with HIV have been handed shame about their bodies from every direction — the wasting years, the lipodystrophy years, and now the opposite message. Some people read this evidence and decide GLP-1s are not for them; that is a legitimate answer. What is worth separating out is function. Visceral fat, liver fat, blood sugar, blood pressure and inflammation respond to treatment; a waistline is not the same thing as a cardiometabolic risk profile. If you and your clinician decide to treat something, treat that.

The drugs — what each one is actually approved for

The most common source of confusion is that the same molecule is sold under different names for different conditions at different doses. Semaglutide is Ozempic and Rybelsus for diabetes and Wegovy for weight; tirzepatide is Mounjaro for diabetes and Zepbound for weight. Insurance decisions often turn on which brand is on the prescription, not which drug is in the pen.

Two older agents round out the class. Saxenda is liraglutide, a daily injection for chronic weight management titrated from 0.6 mg to 3 mg. Trulicity is dulaglutide, a weekly injection for type 2 diabetes and for reducing major cardiovascular events in type 2 diabetes, dosed from 0.75 mg to 4.5 mg; its label notes it has not been studied in people with a history of pancreatitis and is not recommended in severe gastroparesis.6 Both produce less weight loss than semaglutide or tirzepatide — and both are sometimes what a formulary will actually cover.

Contraindications and label warnings across the class

Every one of these drugs is contraindicated in people with a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia syndrome type 2, and in anyone with known serious hypersensitivity.5 Labeled warnings include acute pancreatitis, acute gallbladder disease, hypoglycemia with insulin or sulfonylureas, acute kidney injury from dehydration during vomiting or diarrhea, worsening diabetic retinopathy, increased heart rate, severe gastrointestinal reactions, reports of suicidal thoughts and behavior, and pulmonary aspiration during general anesthesia or deep sedation because of delayed stomach emptying.6 That last one is practical: if you have surgery, a colonoscopy or a dental procedure with sedation scheduled, mention the medicine in advance. Published guidance for people with HIV suggests holding daily formulations at least one day and weekly formulations at least one week before elective surgery.7

Weight gain on antiretroviral therapy — what the evidence really shows

You may have been told your weight gain is just what happens when you feel better and eat again. That is part of it, and it has a name — return to health. But it is not the whole story, and the trials are specific enough to say so.

The pooled trial data

A 2020 analysis in Clinical Infectious Diseases pooled eight randomized trials covering 5,680 people starting antiretroviral therapy for the first time, with more than 10,000 person-years of follow-up. Median weight gain at 96 weeks was 2.0 kg, but the distribution matters more than the median: 48.6% of participants gained at least 3% of body weight, 36.6% gained at least 5%, 17.3% gained at least 10% — and 30.2% lost weight.1

By drug class, adjusted 96-week gain was 3.24 kg with integrase strand transfer inhibitors, 1.93 kg with non-nucleoside reverse transcriptase inhibitors, and 1.72 kg with protease inhibitors. Within the integrase inhibitors: bictegravir 4.24 kg, dolutegravir 4.07 kg, elvitegravir/cobicistat 2.72 kg. Within the nucleoside backbones: tenofovir alafenamide 4.25 kg, tenofovir disoproxil fumarate 2.07 kg, zidovudine 0.39 kg. The single strongest predictor was a baseline CD4 count below 200, worth about 2.97 kg on its own — return-to-health showing up in the data alongside, not instead of, a drug effect. Black women had the largest gains: 19.7% gained at least 10%, versus 12.4% of women who were not Black.1

ADVANCE and NAMSAL

The two trials people cite most were both done in sub-Saharan Africa and both randomized participants prospectively, which is why they carry weight. ADVANCE enrolled 1,053 people in South Africa across three arms. At week 48, mean weight gain was 6.4 kg with dolutegravir plus tenofovir alafenamide and emtricitabine, 3.2 kg with dolutegravir plus tenofovir disoproxil fumarate and emtricitabine, and 1.7 kg with efavirenz plus tenofovir disoproxil fumarate and emtricitabine. Treatment-emergent obesity followed the same order — 14%, 7% and 6% overall, and 20%, 11% and 9% among women — while viral suppression was similar or better in the dolutegravir arms.2

NAMSAL ANRS 12313 randomized 613 people in Cameroon to dolutegravir or low-dose efavirenz, both with tenofovir disoproxil fumarate and lamivudine. Median weight gain at week 48 was 5.0 kg with dolutegravir versus 3.0 kg with efavirenz, and obesity developed in 12.3% versus 5.4% — a statistically significant difference.3

Read together, three signals are consistent: the effect is real, it is largest with an integrase inhibitor paired with tenofovir alafenamide, and it falls hardest on women — particularly Black women. In one review, week-96 body mass index increases averaged 1.91 kg/m² in women versus 1.39 kg/m² in men.7

What is still unsettled

Mechanism is the open question. Proposed explanations include direct effects of integrase inhibitors on fat cell differentiation and mitochondrial function, loss of a weight-suppressing effect that efavirenz and tenofovir disoproxil fumarate may have exerted all along, and changes in appetite regulation. None is settled. What follows clinically matters more: if you gained substantial weight after starting or switching therapy, that observation is legitimate, it is documented in randomized trials, and it belongs on the table at your next visit. It is also not, on its own, a reason to stop a regimen keeping you undetectable. Weight is a manageable problem. Viral rebound is a bigger one.

The evidence in people living with HIV

Until recently, the honest answer here was "almost nothing." That has changed fast: there is now a randomized controlled trial, a multi-site trial in fatty liver disease, a real-world cohort across US HIV clinics, and a large national database analysis. All of it is small next to the general-population trials, and none of it is a cardiovascular outcomes trial. But it is real evidence.

The randomized trial: HIV-associated lipohypertrophy

This is the landmark. Published in The Lancet Diabetes & Endocrinology in 2024, it was a randomized, double-blind, placebo-controlled phase 2b trial at a single US site, enrolling 108 people with HIV — 54 to semaglutide, 54 to placebo — all with controlled HIV, a body mass index of 25 or more, lipohypertrophy, and no diabetes. Participants received 32 weeks of weekly semaglutide: an eight-week titration then 1.0 mg for 24 weeks.9

The primary outcome was abdominal visceral adipose tissue, and it fell by 30.82 cm², a 30.6% reduction. Abdominal subcutaneous fat fell 42.01 cm² (11.2%) and total body fat fell 18.9%. Companion analyses reported total weight down 10.7%, plus reductions in inflammatory markers that did not track with weight or visceral fat change at all: high-sensitivity C-reactive protein down 39.9%, interleukin-6 down 18.8%, and the macrophage activation marker soluble CD163 down 12.3%.7 That last point matters in HIV specifically, where chronic inflammation is an independent driver of risk.

Safety was not silent. Eight participants in each group withdrew early, there was one semaglutide-related grade 4 elevated lipase and two possibly related cases of gallstones, and roughly one-third of the weight lost was lean body mass.9

"Semaglutide holds promise as an effective treatment for HIV-associated lipohypertrophy. The potential risk of serious adverse events deserves further scrutiny in large trials in people with HIV." — Allison Ross Eckard, Grace A. McComsey and colleagues, The Lancet Diabetes & Endocrinology, 20249

SLIM LIVER: semaglutide and fatty liver

SLIM LIVER (ACTG A5371) was a phase IIb single-arm open-label study of 51 people with HIV and metabolic dysfunction-associated steatotic liver disease, given 1 mg semaglutide weekly for 24 weeks. Intrahepatic triglyceride content fell from 12.7% to 8.5% — a relative reduction of 31.3% — with 29% of participants reaching complete resolution of steatotic liver disease. Weight fell 7.8 kg (8.1%). Eighty-two percent were on an integrase inhibitor, and viral suppression held at 97.9%.10

A muscle substudy presented at CROI 2024 is why this trial gets cited in both directions. Psoas muscle volume fell 9.3% over 24 weeks, concentrated in older participants — a 22.8% decrease in people over 60, versus 7.9% in those 40 to 60 and 2.4% in those under 40. Chair-rise time and gait speed did not worsen, and the investigators concluded physical function was preserved despite the volume loss.10 For a population already at elevated risk of frailty, that is a finding to plan around rather than shrug off.

The real-world cohort

A 2024 analysis in AIDS used the CNICS network to follow 222 people with HIV who started semaglutide across US clinical sites between 2018 and 2022. The group skewed toward metabolic illness: mean age 52.8, mean body mass index 35.5, and 76.6% with diabetes. At one year, adjusted weight change was −6.47 kg (95% CI −7.71 to −5.23), a 5.72% reduction, with HbA1c down 1.07 percentage points.8 Two details matter: being on an integrase inhibitor made no difference — −6.49 kg versus −6.38 kg, interaction p-value 0.883 — and the magnitude was comparable to STEP-2, the general-population diabetes trial, which found −6.90 kg.8 At these doses and in this population, HIV did not appear to blunt the response.

The national database picture

CROI 2025 brought the largest analysis to date: a National COVID Cohort Collaborative study across 82 US sites covering 456,558 people who started a GLP-1 receptor agonist between 2018 and 2023, including 3,955 people with HIV. Among women with HIV, tirzepatide produced 1.12 kg per month more weight loss than metformin and semaglutide 0.87 kg per month more; the differences were less pronounced in men.11

What is still missing

No GLP-1 cardiovascular outcomes trial has been done in people with HIV.7 There are no long-term data past a couple of years in this population, no data on what happens after stopping, and very little on people with advanced immunosuppression, on injectable long-acting therapy, or on adolescents and young people living with HIV. Anyone telling you the question is closed is ahead of the evidence.

GLP-1s and antiretroviral therapy — the interaction question

The theoretical worry is straightforward. GLP-1 receptor agonists slow gastric emptying and reduce gastric acid secretion. Some antiretrovirals need food, or need an acidic stomach, to be absorbed properly. So does the combination lower drug levels enough to risk viral rebound?

The reassuring part is pharmacology. GLP-1 receptor agonists are peptides broken down by endopeptidases — semaglutide by neutral endopeptidase — not by the cytochrome P450 enzymes that drive most antiretroviral interactions. A peer-reviewed review in HIV Medicine concluded that because they are metabolized this way, GLP-1 agonists "do not generate major drug–drug interactions with most drugs, including ARVs," and that theoretical considerations plus the available clinical observations support prescribing semaglutide and liraglutide in people with HIV, with no indications of concern about efficacy, safety or pharmacological interactions with antiretrovirals.12

What the interaction checkers say

The University of Liverpool's HIV Drug Interactions database is the free, traffic-light-coded tool most HIV clinicians and pharmacists actually use. Its current entry for bictegravir/emtricitabine/tenofovir alafenamide lists "No Interaction Expected" against semaglutide, liraglutide, dulaglutide, tirzepatide and exenatide, and the same verdict appears for ritonavir with semaglutide — though there the underlying evidence quality is flagged as very low, meaning the prediction rests on pharmacology rather than dedicated studies.13 That distinction is easy to lose. "No interaction expected" is a prediction, not a trial result — a reasonable one that specialists work from, and also the reason monitoring beats assumption.

The specific regimens to flag

Rilpivirine. The Edurant label is unambiguous: one 25 mg tablet once daily with a meal. Drugs that raise gastric pH may decrease rilpivirine concentrations and cause loss of virologic response, and antacids are restricted to at least two hours before or four hours after the dose.14 A medicine that suppresses appetite, slows the stomach and reduces acid runs against every one of those requirements. The HIV Medicine review names oral rilpivirine and atazanavir specifically as the two antiretrovirals requiring caution and close monitoring with GLP-1 agonists, because both need low gastric pH for optimal absorption.12

This is not a hard prohibition. It is a reason to make sure the meal actually happens — a real meal, not a cracker — and a reason to check a viral load sooner than you otherwise would after starting or escalating the dose. Atazanavir carries the same logic and the same monitoring.

Protease inhibitors generally. The same review advises extra caution when starting a GLP-1 agonist in people taking protease inhibitors who already have risk factors for heart rate variability, given the class effect on heart rate.12 Oral semaglutide carries its own rule that has nothing to do with HIV: Rybelsus must be taken at least 30 minutes before any other oral medication.13 If you take antiretrovirals in the morning, that changes your routine — and routine is what keeps adherence intact.

Practical takeaway: bring your full medication list to whoever writes the GLP-1 prescription, ask them to run it through the Liverpool checker in front of you, and ask for a viral load check about three months after starting.

Cardiovascular data — strong in the general population, pending in HIV

The reason GLP-1s are treated as cardiovascular drugs and not just weight drugs is one trial.

SELECT

SELECT randomized 17,604 people — 8,803 to weekly semaglutide 2.4 mg and 8,801 to placebo — all aged 45 or older with a body mass index of 27 or higher and established cardiovascular disease, and none with diabetes. Over a mean 39.8 months, the primary composite of cardiovascular death, non-fatal heart attack or non-fatal stroke occurred in 6.5% on semaglutide versus 8.0% on placebo: a hazard ratio of 0.80 (95% CI 0.72–0.90), P<0.001. Weight fell 9.39% versus 0.88% at week 104, and discontinuation for adverse events was 16.6% versus 8.2%.15

A 20% relative reduction in major cardiovascular events, in people without diabetes, is the finding that moved this class into cardiology. But SELECT included essentially no one identified as living with HIV — and people living with HIV carry roughly double the cardiovascular risk, with inflammation as a major driver.7 There is no equivalent trial in this population.

REPRIEVE — what an HIV-specific cardiovascular trial looks like

REPRIEVE is the useful comparator, because it is the trial the HIV field actually ran. It randomized 7,769 people with HIV on antiretroviral therapy, aged 40 to 75, at low-to-moderate cardiovascular risk, to pitavastatin 4 mg daily or placebo. It was stopped early for efficacy in March 2023. Major adverse cardiovascular events occurred at 4.81 versus 7.32 per 1,000 person-years — a hazard ratio of 0.65 (95% CI 0.48–0.90), a 35% reduction — with the composite of events or death down 21%. Median LDL cholesterol fell from 107 to 74 mg/dL. Muscle symptoms occurred in 2.3% versus 1.4%, and new diabetes in 5.3% versus 4.0%.16

Two things follow. REPRIEVE proves an HIV-specific cardiovascular prevention trial is possible at scale, which is the strongest argument for running one with a GLP-1. And more immediately: if you are 40 to 75 and living with HIV, a statin is an evidence-based intervention available right now, with REPRIEVE's results incorporated into 2026 statin recommendations for people with HIV.16 A GLP-1 conversation should not displace a statin conversation.

Lipoatrophy, lipohypertrophy, and why this history matters

Anyone who was in HIV care in the late 1990s and 2000s hears "fat redistribution" and flinches. Naming what happened explains both why the visceral fat research is framed the way it is and why some people are wary of any drug that shrinks fat.

Two different problems that got one name

Beginning around 1998, as combination therapy spread, clinicians described a syndrome of body shape change: fat lost from the limbs, buttocks and face — lipoatrophy — sometimes alongside fat accumulating in the abdomen, upper back and breasts — lipohypertrophy. They were lumped together as "lipodystrophy," but they are separate problems with separate causes and separate treatments.

Lipoatrophy was closely linked to the thymidine analogue nucleoside reverse transcriptase inhibitors — stavudine most of all, zidovudine to a lesser degree. Facial wasting was the part that broke people: visible, legible to strangers as an HIV diagnosis, and a documented driver of disclosure fear, isolation and treatment discontinuation. Replacing those drugs cut the incidence sharply, but stopping did not reliably reverse what had already happened. Lipohypertrophy, meanwhile, did not go away. Central visceral fat accumulation remains common on modern antiretroviral therapy, and it drives inflammation and cardiometabolic risk rather than being a cosmetic footnote — which is exactly why it was chosen as the primary endpoint of the semaglutide randomized trial.9

Tesamorelin — the option that came first

Tesamorelin, a growth-hormone-releasing factor analogue, is the only medication approved in the United States specifically for excess abdominal fat in people with HIV and lipodystrophy. It is sold as Egrifta SV and, since March 2025, as Egrifta WR — a reformulation dosed at 1.28 mg subcutaneously once daily, with one reconstituted vial lasting seven days. The label's limitations of use are explicit: long-term cardiovascular safety has not been established, and it is "not indicated for weight loss management as it has a weight neutral effect." It is contraindicated in active malignancy, in disruption of the hypothalamic-pituitary axis, and in pregnancy.17 (The FDA label uses the phrase "HIV-infected adult patients"; that is the regulatory wording, not ours.)

The clinically meaningful difference is what each drug does to which fat. Tesamorelin reduces visceral fat while largely preserving subcutaneous fat, which is why published guidance still describes it as the preferred option where lipoatrophy is part of the picture. GLP-1 agonists reduce both: in the randomized trial, visceral fat fell 30.6% but subcutaneous fat also fell 11.2%.7 If you carry facial or limb fat loss from the stavudine era, that history is a specific and legitimate reason to discuss tesamorelin before, or instead of, a GLP-1.

Insurance realities in 2026

This is where most GLP-1 conversations actually end, and the landscape shifted significantly in 2026. List prices run in the range of a thousand dollars a month before rebates — Medicaid paid roughly $1,000 gross per GLP-1 prescription in 2024 — so coverage is not a detail.19

Medicare

The old rule was simple and harsh: Medicare Part D is statutorily prohibited from covering drugs used for weight loss, so semaglutide was covered as Ozempic for diabetes and not as Wegovy for obesity. That statutory prohibition still exists.

What changed is a workaround. Beginning July 1, 2026, the Medicare GLP-1 Bridge covers Foundayo tablets, Wegovy (injection or tablet) and the Zepbound KwikPen for people with Part D, at a $50 monthly copay, with prior authorization and a provider certification that the medicine is being used alongside a diet and exercise program. It runs through December 31, 2027. Eligibility requires being 18 or older, plus one of:18

• a body mass index of 35 or higher;
• a body mass index of 30 or higher with diastolic heart failure, uncontrolled high blood pressure, or stage 3a or worse chronic kidney disease; or
• a body mass index of 27 or higher with prediabetes, a prior heart attack or stroke, or symptomatic peripheral artery disease.

Two catches. The $50 does not count toward your deductible or out-of-pocket limit and cannot be lowered by Extra Help. And people with type 2 diabetes, moderate-to-severe sleep apnea or fatty liver disease are not Bridge-eligible — their Part D plan is already supposed to cover a GLP-1 for those conditions, so that is the route.18

A separate demonstration, the BALANCE Model, was designed to negotiate GLP-1 prices for Medicare and Medicaid. CMS announced in April 2026 that it would not implement BALANCE in Medicare in 2027; the Medicaid side began rolling out May 1, 2026. Neither program removes the underlying statutory ban.19

Medicaid

Medicaid coverage depends entirely on your state, and it has been moving in the wrong direction. As of January 2026, 13 state Medicaid programs covered GLP-1s for obesity — down from 16 in October 2025, after California, New Hampshire, Pennsylvania and South Carolina eliminated coverage. North Carolina dropped it in October 2025 and reinstated it in December.19

Coverage for obesity is optional for states. Coverage is required when the drug is prescribed for an FDA-approved indication Medicaid must cover — type 2 diabetes, cardiovascular risk reduction with Wegovy since March 2024, and moderate-to-severe obstructive sleep apnea with Zepbound since December 2024 — plus coverage for children under EPSDT.19 So the indication written on the prescription can determine whether the claim pays: a conversation to have with your prescriber, not a reason to misrepresent a diagnosis. For scale on why states keep flipping, Medicaid GLP-1 prescriptions grew from about one million in 2019 to more than eight million in 2024, with gross spending going from roughly $1 billion to roughly $9 billion.19

Commercial insurance and AIDS Drug Assistance Programs

Commercial plans almost always require prior authorization: a documented body mass index at or above threshold, a diagnosis code matching the drug's approved indication, and documentation of a supervised lifestyle attempt, often three to six months. Some require step therapy through an older, cheaper agent first. Denials are frequently procedural rather than clinical — which is why they are frequently reversible.

AIDS Drug Assistance Programs are the piece most people do not think to check. ADAP is funded under Part B of the Ryan White HIV/AIDS Program, and each state runs its own formulary and eligibility rules.20 Formularies are not limited to antiretrovirals — many include treatments for the conditions that come with long-term HIV. Whether GLP-1s are on your state's list is a question with a real answer, and it is worth asking.

Compounded GLP-1s — what changed, and why it matters

For a stretch in 2023 and 2024, when Ozempic and Mounjaro were on the FDA drug shortage list, compounding pharmacies were permitted to make copies. A large telehealth industry grew up around that window. That window is closed: the FDA declared the tirzepatide shortage resolved on October 2, 2024 and the semaglutide injection shortage resolved on February 21, 2025. Enforcement discretion for 503A compounding pharmacies ended March 10, 2025 for tirzepatide and April 28, 2025 for semaglutide — the latter following an April 24, 2025 federal court decision — with 503B outsourcing facilities following on March 19 and May 22, 2025. Neither drug is on the shortage list or the 503B bulk substances list now.21

The FDA's operative standard is whether a compounded product is "essentially a copy" of an approved drug. It has said it does not intend to act against a compounder filling four or fewer prescriptions per month of such a product, and that semaglutide combined with vitamin B12 may still count as essentially a copy if the strength is within 10% of an approved one.21 Those are narrow lanes, not a general permission.

What this means for you

Compounded drugs are not FDA-approved, and the FDA states plainly that they may be of substandard quality or otherwise unsafe — there is no premarket review of potency, purity or sterility the way there is for Ozempic or Zepbound.21

There is also a specific HIV problem with the telehealth model. A weight-loss subscription service that never sees your chart does not know you are on rilpivirine, does not know your CD4 history, does not order your viral load, and does not talk to your HIV clinic. Fragmented care is how adherence breaks. If cost is what is pushing you toward a compounding route — and for many people it genuinely is — take that to your HIV clinic first and ask about manufacturer patient assistance, the $50 Medicare Bridge copay if you have Part D, and ADAP.

Side effects — the common, the serious, and the HIV-specific

Gastrointestinal, which is to say: most of it

Nausea, vomiting, diarrhea, constipation, abdominal pain, burping and reflux are the dominant side effects, and they are why people stop. In SELECT, 16.6% of the semaglutide group discontinued for adverse events versus 8.2% on placebo, and gastrointestinal reasons accounted for 10.0% versus 2.0%.15 In the HIV trials, gastrointestinal effects were mostly grade 1 — mild.10

Most of this improves with slow dose escalation, smaller meals, and time. It also carries a specific risk in HIV: vomiting and diarrhea are how doses get missed and how dehydration turns into acute kidney injury, a labeled warning across the class. If you cannot keep an antiretroviral dose down, call your clinic — do not wait it out.

The serious but uncommon

Acute pancreatitis and acute gallbladder disease are labeled warnings for all of these drugs.5 The HIV randomized trial recorded one grade 4 elevated lipase and two possibly related gallstone cases among 54 people on semaglutide — small numbers, but not zero.9 Severe, persistent upper abdominal pain, especially radiating to the back or with vomiting, needs same-day evaluation.

Other labeled concerns: hypoglycemia with insulin or a sulfonylurea, worsening diabetic retinopathy, increased resting heart rate, reports of suicidal thoughts and behavior, and aspiration under anesthesia. Semaglutide has also been observationally linked to non-arteritic anterior ischaemic optic neuropathy, a rare cause of sudden vision loss — so any abrupt vision change is an urgent call.7

Muscle and face — the part HIV changes

Weight loss on these drugs is not all fat. In the general-population STEP-1 trial, lean mass accounted for roughly 40% of total weight lost, and in the HIV lipohypertrophy trial about one-third of the loss was lean body mass.7 SLIM LIVER measured it directly: psoas muscle volume down 9.3% at 24 weeks, with a 22.8% decrease in people over 60 — while chair-rise and gait speed held steady.10

For people aging with HIV, who face elevated rates of frailty and sarcopenia, this deserves a plan rather than a footnote: resistance training two to three times a week, adequate protein, and someone actually tracking function — grip strength, chair rise, walking speed — not only weight. Ask for that at the start, not after six months.

The facial volume loss people call "Ozempic face" is the same phenomenon where it shows most, and in HIV it lands differently, because facial wasting was for years the most visible marker of a diagnosis. In the randomized trial, subcutaneous fat — including in the face — fell 11.2% alongside the visceral reduction.9 If that history is yours, say so out loud in the appointment.

Who is a candidate — and who probably is not

Published guidance for people with HIV mirrors the general-population thresholds: a body mass index of 30 or higher, or 27 to 29.9 with a weight-related condition, after a documented attempt at 5% weight loss through lifestyle change over three to six months. A type 2 diabetes diagnosis opens an easier door, because that indication is the one insurers cover most readily. Established cardiovascular disease plus overweight or obesity is the indication SELECT created.7

Reasons this class is off the table, or needs a hard second look:

• Absolute: a personal or family history of medullary thyroid carcinoma or MEN 2; known serious hypersensitivity; pregnancy or planning pregnancy in the near term.5
• Strong caution: a history of pancreatitis; active gallbladder disease; severe gastroparesis; active eating disorder; significant existing lipoatrophy; advanced frailty or sarcopenia.7
• Needs a plan first: a regimen built on rilpivirine or atazanavir; a history of adherence disruption during periods of illness; upcoming surgery.12

There is also the question nobody asks out loud. If your weight is not causing metabolic problems — normal blood sugar, normal lipids, normal blood pressure, no fatty liver, no sleep apnea — the case for a lifelong injectable with a lean-mass cost is much weaker, whatever the scale says. "Because I want to be smaller" is a human reason; it is not the reason these drugs were studied. And bring insurance navigation into the room early: Ryan White case managers, ADAP enrollment workers and AIDS service organization benefits counselors do prior authorizations and appeals as their actual job, usually faster than a clinic, and free.20

Florida specifics

Florida has a concrete piece of good news here that a lot of people do not know.

Florida ADAP covers GLP-1 agonists

The Florida ADAP formulary dated April 2026 lists semaglutide (tablet and injection), liraglutide (injection) and dulaglutide (injection) under the category "Antidiabetic / GLP-1 receptor agonist." Tirzepatide is not on the list. The formulary's separate "anti-obesity" category contains only naltrexone extended-release/bupropion extended-release and orlistat — no GLP-1s. Tesamorelin is not listed. Pitavastatin, the statin used in REPRIEVE, is.20

Read the category label carefully, because it is the whole story: these agents are on the Florida list as antidiabetic drugs, so the realistic route is a diabetes or blood-sugar indication, not weight management. Ask your ADAP pharmacist to confirm current criteria, because formularies change — Biktarvy and Descovy were themselves restored to the Florida list effective July 1, 2026. Eligibility requires HIV-positive status, need for HIV medications, income at or below 400% of the federal poverty level, and being uninsured or having inadequate prescription coverage; the state ADAP Help Desk is 844-381-2327.20

Florida Medicaid

Florida is not among the 13 state Medicaid programs identified as covering GLP-1s for obesity as of January 2026.19 Coverage for the mandatory indications — type 2 diabetes, cardiovascular risk reduction, obstructive sleep apnea — still applies. The operative document is Florida Medicaid's Preferred Drug List, set by the Pharmaceutical and Therapeutics Committee and revised periodically; because it changes, verify the current version with your plan or pharmacist rather than any summary, including this one.

Where to start locally

Florida's HIV/AIDS Patient Care Program, run through county health departments and Ryan White Part B networks, is the entry point for ADAP enrollment and case management. If you receive care at a Ryan White clinic, your case manager can pull your ADAP status, check the formulary, and start a prior authorization the same week — and your local AIDS service organization likely has a benefits specialist who has already fought a GLP-1 denial.

Start here

If you want to move on this, here is the order that wastes the least time.

1. Write down what you actually want to change. Blood sugar? Liver fat? Waist? Sleep apnea? Cardiovascular risk? Be specific — the answer determines the drug, the indication code, and whether insurance pays.
2. Bring it to your HIV clinician and primary care. Your HIV clinician owns the interaction question, your viral load and your lipoatrophy history; primary care or endocrinology usually owns the prescription and the titration. Make sure one chart has the whole list.
3. Get labs before you start. Weight, waist, blood pressure, HbA1c or fasting glucose, full lipid panel, kidney function, liver enzymes, and a current viral load and CD4 count. Without a baseline you cannot tell whether this worked.
4. Settle coverage before you settle on a drug. Ask which GLP-1s your plan covers and under which diagnosis; if you have Medicare Part D, ask whether the GLP-1 Bridge or your regular benefit applies;18 and ask your ADAP pharmacist what is on the current formulary.20
5. Build the muscle plan on day one. Resistance training two to three times a week, protein at every meal, and a baseline measure of function so decline is visible if it happens.10
6. Recheck the viral load roughly three months in — sooner if you are on rilpivirine or atazanavir, or if nausea has cost you doses.14
7. Decide in advance what "not working" looks like. If your target measure has not moved after several months at a full tolerated dose, or side effects are costing you adherence, that is information — not failure. Stopping is a legitimate outcome.

One last thing, plainly. The HIV community spent decades being told what its bodies meant — first that thinness was death, later that any visible change was a diagnosis on display. You are allowed to be interested in these medicines. You are also allowed to decide they are not for you. Either way, you deserve a clinician who answers the question with evidence instead of a lecture — and the evidence in people living with HIV now exists, is genuinely encouraging, and is still incomplete.