CROI 2026: The Tools Are Here. The Infrastructure Is Not.
The 33rd Conference on Retroviruses and Opportunistic Infections (CROI) convened February 22nd – 25th in Denver under extraordinary tension between a pipeline of HIV prevention, treatment, and potential cure tools that could reshape the epidemic's trajectory, and a global funding crisis actively dismantling the infrastructure required to deliver any of it. As Conference Chair Nicolas Chomont of the Université de Montréal stated in the Opening Session, "we share a responsibility to defend and sustain funding for international HIV programs and research." That charge framed every session that followed.
The Funding Crisis: New Data on the Damage
The consequences of disruptions to the U.S. President's Emergency Plan for AIDS Relief (PEPFAR), the dissolution of the United States Agency for International Development (USAID), and National Institutes of Health (NIH) cuts are no longer hypothetical. The CROI session "Sleepless in Denver" presented the first systematic evidence of the damage. Ellen Brazier's survey data from the International epidemiology Databases to Evaluate AIDS (IeDEA) consortium found that across 32 countries, 47% of clinics reported disruptions in HIV services, with similar rates of disruption to medication availability, laboratory services, and clinic operations. In KwaZulu-Natal, South Africa, Lindsey Filiatreau reported that 39% of clinics experienced disruptions affecting an estimated 830,000 people living with HIV.
The damage is not confined overseas. Aaron Richterman presented data from a rapid survey across three U.S. states showing that 47% of clinics reported HIV service disruptions, including medication shortages. He also demonstrated how cuts to the Supplemental Nutrition Assistance Program (SNAP), the country's largest targeted anti-poverty program serving more than 42 million Americans, directly undermine treatment outcomes. During the 2025 government shutdown, ART adherence among people living with HIV who receive SNAP benefits dropped to as low as 40%. The connection between food security and viral suppression is well established; cutting one predictably undermines the other. As Filiatreau put it, "These things [HIV services]… can be taken away overnight, but they can't be rebuilt overnight."
Prevention: An Expanding Toolkit, a Widening Access Gap
Against this backdrop, CROI delivered a prevention portfolio that is broader and stronger than at any previous conference. Final results from the PURPOSE 1 and PURPOSE 2 trials confirmed the efficacy of twice-yearly injectable lenacapavir for pre-exposure prophylaxis (PrEP). In PURPOSE 1, which enrolled cisgender adolescent and young women in sub-Saharan Africa, HIV incidence among lenacapavir recipients was 0.07 per 100 person-years, compared to 1.98 for oral emtricitabine/tenofovir alafenamide (F/TAF) and 1.94 for emtricitabine/tenofovir disoproxil fumarate (F/TDF), with only two seroconversions among more than 2,000 participants. PURPOSE 2, enrolling men who have sex with men and gender diverse people, showed HIV incidence of 0.11 per 100 person-years for lenacapavir versus 0.92 for F/TDF, with three seroconversions among 2,179 participants.
The five total seroconversions across both studies received considerable attention, with four showing lenacapavir-associated resistance mutations that researchers believe developed during PrEP rather than being transmitted. Research into why these breakthroughs occurred is ongoing. As Gilead's Stephanie Cox stated, "We don't know why these occurred… I think the efficacy is very high." San Francisco AIDS Foundation (SFAF) Medical Director Hyman Scott, MD, MPH, added context: "The breakthrough infections are important to evaluate but are extremely rare among the thousands of study participants."
The Prévenir study's final eight-year results from France reinforced that both daily and on-demand oral PrEP are safe and effective, with overall HIV incidence of 0.11 per 100 person-years across more than 3,200 users and 13,000 person-years of follow-up. Switching between daily and on-demand use was the norm rather than the exception, with 59% of daily users changing to on-demand at least once, and 52% doing the reverse. This carries a clear message for implementation: people need flexibility, and rigid one-size prescribing undermines persistence.
The prevention pipeline continues to expand. Merck's once-monthly oral PrEP candidate MK-8527 selected an 11 mg dose maintaining protective drug levels in at least 95% of participants, with Phase 3 EXPrESSIVE trials now enrolling. Gilead's PURPOSE 365 study, testing once-yearly lenacapavir for PrEP, is being designed. The SEARCH study showed that community health workers paired with digital tools reduced HIV incidence by 70% in rural populations, a reminder that prevention tools work best when embedded in community-driven delivery.
The problem is reach. Andrew Hill highlighted that only 2.3 million people are currently on oral PrEP, far below UNAIDS targets, and that injectable cabotegravir and lenacapavir represent just 2.9% and 0.9% of total PrEP use, respectively. We have a growing menu of prevention tools. Getting them to the people who need them is where the system breaks down.
Treatment: More Options, Longer Intervals, Patient Choice
The treatment pipeline at CROI 2026 moved toward a central goal: giving people living with HIV more choices that fit their lives. Merck presented late-breaking data from three Phase 3 trials of doravirine/islatravir (DOR/ISL), the first ever once-daily, non-INSTI two-drug regimen. In treatment-naive adults, DOR/ISL demonstrated non-inferiority to bictegravir/emtricitabine/tenofovir alafenamide (BIC/FTC/TAF), with 91.8% achieving viral suppression at Week 48 compared to 90.6%, including participants with high viral loads and low CD4 counts. The U.S. Food and Drug Administration (FDA) has set an action date of April 28, 2026 for the DOR/ISL application. For people aging with HIV who manage multiple comorbidities, a two-drug, non-INSTI regimen addresses a real clinical gap. Research presented at this same conference linked the widely used INSTI dolutegravir to neuropsychiatric effects, including blocking a brain enzyme essential for memory and emotional regulation, with one study halted for ethical reasons after participants experienced worsening symptoms. For people navigating tolerability concerns, toxicity issues, or polypharmacy, having a non-INSTI alternative with fewer active agents matters.
Gilead's ARTISTRY-1 and ARTISTRY-2 trials demonstrated that a bictegravir/lenacapavir (BIC/LEN) single-tablet regimen can maintain viral suppression for people switching from complex multi-tablet regimens (96% at 48 weeks in ARTISTRY-1) or from Biktarvy (93.5% in ARTISTRY-2). For people who have been on complex regimens for years due to resistance histories, this potential simplification addresses a real quality-of-life gap. Gilead plans to submit these results to regulatory authorities.
Long-acting injectables continued their forward march. In ViiV Healthcare's EMBRACE study, lotivibart (a broadly neutralizing antibody, or bNAb) given as an IV infusion every four months plus monthly cabotegravir maintained viral suppression in 94% of participants at 12 months. Part 2 of EMBRACE, testing lotivibart infusions every six months, is now fully enrolled. ViiV also presented early data on VH-184, a third-generation integrase inhibitor with potential twice-yearly dosing, and VH-499, a capsid inhibitor supporting twice-yearly intervals. The VOLITION study showed that 85% of treatment-naive adults opted to switch from daily pills to bimonthly long-acting cabotegravir/rilpivirine (CAB+RPV LA), with 95% maintaining suppression. Data like VOLITION's 85% opt-in rate reinforce what the HIV community has long argued: when people living with HIV are offered options that fit their lives, they take them. Payers, formulary committees, and ADAP programs should take note. Treatment is not one-size-fits-all, and coverage shouldn’t treat it as such.
Community activist Shari Margolese put it plainly at CROI's final Community Breakfast Club: "As a community we need to get much angrier about the fact that we can't get access to the drugs." Francois Venter of Ezintsha in South Africa warned that without action, "we might be sitting here again in 10 years' time" celebrating breakthroughs that never reach communities.
Cure, Comorbidities, and the Equity Question
On the cure front, the RIO trial's Phase B results offered genuine encouragement. Among the 28 people who received a placebo in Phase A and were then given bNAbs teropavimab and zinlirvimab, 54% had prolonged viral remission after stopping antiretroviral therapy (ART), with two still off treatment after more than a year. Because ART was stopped six months after the bNAb infusions, these results point to an immunological "vaccinal effect" rather than direct viral suppression. A cure remains distant, but these are the kinds of incremental, well-designed steps that build the evidence base forward.
CROI also highlighted the growing urgency of managing comorbidities in aging populations living with HIV. The POPPY study found that depression affects 32.4% of people living with HIV over 50, linked to inflammatory markers rather than psychosocial factors alone. A study of over 1,500 men showed that the combination of age 65 and over, HIV, and metabolic syndrome produced significantly worse cognitive impairment than any factor alone. Metabolic syndrome is the modifiable factor in that triad, which means clinicians and patients can act on it now. CROI data on GLP-1 receptor agonists like semaglutide point toward a potential tool for doing so: in a study of people living with HIV-associated lipohypertrophy, semaglutide produced a 19% reduction in total body fat, a 31% decrease in visceral adipose tissue, and a nearly 50% drop in C-reactive protein, a key inflammatory marker. Separate data presented at the conference found that semaglutide did not worsen depression in people living with HIV, and that people with moderate or severe depression at baseline actually showed improvements. These findings warrant dedicated research into how GLP-1 therapies can address the long-term health consequences of chronic inflammation and aging with HIV. If the evidence continues to support their role, GLP-1 receptor agonists should be evaluated for inclusion in the HIV standard of care, with appropriate insurance and program coverage to match.
The equity question came into sharp focus with data from the ENCORE cohort. Black trans women in the U.S. had an HIV incidence of 15.5 per 1,000 person-years, compared to 1.4 for White trans women. Poverty, houselessness, and lack of insurance were significant drivers, and only 4% of trans women in the cohort used long-acting injectable PrEP. Dr. Sari Reisner of the University of Michigan warned that the current administration's erasure of gender identity data from federally funded datasets will make it harder to monitor disparities and determine what works. We cannot close gaps we refuse to measure, and they know that.
What Comes Next
CROI 2026 produced a clear picture: we have tools that can change the course of the HIV epidemic, and the systems required to deliver them are being actively undermined. The path forward requires specific action. We must defend and restore funding for PEPFAR, NIH, and the global HIV infrastructure that makes science reach people. State ADAP programs and payers must expand coverage for long-acting prevention and treatment options and remove administrative barriers that delay access. To do that, they need to be adequately funded. PrEP implementation must embrace the full range of validated modalities, from daily oral to on-demand to injectable to monthly oral, with flexibility built into delivery. Care for people aging with HIV must shift toward whole-person approaches that integrate cognitive screening, metabolic risk management, and mental health support alongside viral suppression. And we must protect the community-led surveillance and data collection that allows us to see and respond to disparities, especially for trans and gender-diverse communities.
Wes Sundquist of the University of Utah, who helped develop the science behind lenacapavir, reflected at CROI on the decades-long journey that produced this tool. He warned that while the field now has "a really powerful new tool in the arsenal," forces are blocking its use. "It will be a human tragedy," he said, "if we don't overcome those." The science has done its part. The rest is a question of political will, policy design, and whether we as a community can sustain the pressure long enough for these tools to reach the people who need them.
The Great American Recovery Needs More Than a Slogan
On February 2, 2026, Health and Human Services Secretary Robert F. Kennedy Jr. took the stage at SAMHSA's Prevention Day to announce a $100 million pilot program addressing homelessness and addiction, alongside a meaningful expansion of medication access for families affected by opioid use disorder. In the same speech, Kennedy characterized harm reduction as a "non-effective intervention" that "enabled future drug use." The contradiction captures the current state of American addiction policy: genuine progress on biomedical treatment access undermined by ideological rejection of the evidence-based strategies needed to keep people alive long enough to access that treatment.
The scope of the crisis is not in dispute. According to the White House fact sheet accompanying the Great American Recovery Initiative, 48.4 million Americans, or 16.8% of the population, live with substance use disorder. Nearly eight in ten did not receive treatment in 2024. These numbers should focus policymakers on removing every barrier between people and care. Instead, the administration is simultaneously expanding some pathways while actively dismantling others.
The Biomedical Frontier
One area of genuine scientific promise involves glucagon-like peptide-1 (GLP-1) receptor agonists, medications originally developed for diabetes and obesity that are showing unexpected potential for treating addiction. These drugs target the brain's mesolimbic reward pathways, and emerging research indicates they may modulate the dopamine neurotransmission involved in addictive behaviors.
The implications are significant. As the British Journal of Pharmacology notes, no FDA or EMA-approved medications currently exist for cocaine or stimulant use disorders. This treatment gap disproportionately affects marginalized communities, including LGBTQ populations where methamphetamine use remains a significant concern intersecting with HIV and HCV transmission.
Early evidence is encouraging. A large observational study using the VA database found that people with alcohol use disorder who used GLP-1 medications had a 50% lower rate of alcohol bingeing compared to those not on the medications. People with opioid use disorder on these medications had a 40% lower rate of overdose. Clinical trials are now underway for multiple substance use disorders, including a trial specifically enrolling people with both cocaine use disorder and HIV.
"This research is very important because alcohol and drug addiction are major causes of illness and death, yet there are still only a few effective treatment options," Dr. Lorenzo Leggio of the National Institute on Drug Abuse and National Institute on Alcohol Abuse and Alcoholism noted in October 2025.
The critical question is access. As Penn Medicine researchers have observed, "many who struggle with addiction are multiply marginalized, making access to these medications a potential concern." The VA study data came largely from older white males, and robust research across demographics remains necessary. Breakthrough treatments mean little if the people who need them most cannot obtain them.
Meaningful Progress
Credit where due: the administration has taken concrete steps to expand medication access for opioid use disorder. On February 2, the Administration for Children and Families announced that buprenorphine, methadone, and naltrexone now qualify as prevention services eligible for Title IV-E funding. States and tribes can receive a 50% federal match to provide these medications to parents when children are at imminent risk of entering foster care. The policy reflects sound reasoning: keeping families together through effective treatment generally serves children better than separation.
The December 2025 reauthorization of the SUPPORT Act extended substance use disorder programs through fiscal year 2030 after the original legislation had languished since its 2023 expiration. The bill passed with strong bipartisan support, 366-57 in the House and by unanimous consent in the Senate.
There is also useful historical precedent from the first Trump administration. In May 2020, HHS Office for Civil Rights Director Roger Severino secured an agreement with West Virginia establishing that people in recovery using medication-assisted treatment are entitled to ADA protections. "People in recovery from opioid use disorder should never be stigmatized for seeking appropriate medical treatment that can save their lives," Severino stated at the time. That principle should guide current policy.
Where Policy Contradicts Evidence
Against these advances stands a pattern of actions that undermine the stated goal of connecting people with treatment.
The Substance Abuse and Mental Health Services Administration has lost approximately one-third of its roughly 900 employees over the past year. In January 2026, the administration briefly cancelled nearly $2 billion in SAMHSA grants before bipartisan backlash forced a reversal within 24 hours. Providers report an environment where planning for the future feels impossible.
The administration proposed folding SAMHSA into a new "Administration for a Healthy America." Congress rejected this in the FY2026 LHHS appropriations package and added structural protections requiring 60 days' advance notice before HHS reorganizations affecting CDC functions and three days' notice before grant terminations. These guardrails exist because they proved necessary.
On harm reduction, the gap between evidence and policy is particularly troubling. The July 2025 executive order "Ending Crime and Disorder on America's Streets" directed SAMHSA to defund "so-called harm reduction" programs. A subsequent SAMHSA letter drew an explicit line between acceptable overdose reversal tools like naloxone and the "ideological concept of harm reduction."
This framing ignores the government's own evidence. In December 2025, the VA published an analysis of its harm reduction programs describing syringe services programs as "one of the most effective public health interventions ever devised." The data: SSPs decrease new HIV and HCV infections by up to 67%, increase the likelihood of achieving abstinence five-fold, and "do not enable or increase drug use, nor do they cause increases in crime."
The FY2026 appropriations bill maintains Section 525, the longstanding prohibition on using federal funds for sterile needles or syringes outside narrow outbreak exceptions. Report language frames harm reduction through an abstinence-first lens, treating harm reduction and recovery as opposing forces when the evidence shows they are complementary. Meeting people where they are is how you eventually connect them with treatment.
The Syndemic Reality
These policy contradictions have real consequences for communities facing intersecting epidemics. Syringe services programs are foundational infrastructure for preventing HIV and HCV transmission among people who inject drugs. Cutting STI prevention funding by $10 million while syphilis and congenital syphilis remain at historically high levels makes no public health sense.
The approach to homelessness reveals similar contradictions. The July 2025 executive order abandons Housing First, the evidence-based model that prioritizes stable housing as a foundation for recovery. In its place, the order directs agencies to prioritize jurisdictions that enforce bans on urban camping, loitering, and open-air drug use when awarding federal grants. It encourages states to expand involuntary civil commitment and conditions housing assistance on participation in behavioral health treatment. The Bipartisan Policy Center notes this approach may invite Fair Housing Act lawsuits, since conditioning housing on treatment could constitute discrimination against people with disabilities, including those with substance use disorder.
HHS’s $100 million STREETS Initiative operates within this enforcement-first framework. Kennedy described the model as finding people on the street and moving them "from crisis to detox treatment to housing to employment." Housing comes after treatment compliance, not before. The National Alliance to End Homelessness has been direct in its assessment: "Deinstitutionalization did not cause homelessness, and re-institutionalization will not solve it."
The 2024 Point-in-Time count recorded over 770,000 people experiencing homelessness, an 18% increase from the previous year and the largest annual jump ever recorded. Those most affected include people with mental illness or substance use disorder, LGBTQ youth, and veterans, as Harvard's Howard Koh has noted. A $100 million pilot serving eight cities cannot address a crisis of this scale, particularly when the broader policy framework criminalizes the people it claims to help.
Access barriers to existing treatments compound the problem. The Cato Institute reports that 80% of U.S. counties have no opioid treatment programs, and only 600,000 of the 8 million people meeting criteria for opioid use disorder received methadone in 2024. The bipartisan Modernizing Opioid Treatment Access Act would have enabled primary care prescribing of methadone; it was not reintroduced in the current Congress.
The Path Forward
The promise of emerging treatments like GLP-1 agonists cannot be realized without the infrastructure to deliver them. A breakthrough medication for stimulant use disorder means nothing to someone cycling between encampments and emergency rooms because Housing First was abandoned in favor of treatment mandates they cannot access. Flat funding for SAMHSA, restrictions on harm reduction, and criminalization of homelessness create gaps that no medication can bridge.
"If we want to create a world where there's opioid recovery, we need to also offer affordable housing and access to affordable food and improved access to health care," Dr. Sadie Elisseou of Harvard told Behavioral Health Business. This syndemic framing should guide policy. It currently does not.
The administration cannot simultaneously expand medication access, gut the agency responsible for treatment infrastructure, restrict the harm reduction programs that keep people alive and connected to care, and criminalize the circumstances of those most in need of help. These policies do not form a coherent strategy. They form a contradiction.
Congress rejected the administration's most extreme proposals through the passage of the L-HHS funding package, but holding ground is not progress. Advocates should monitor SAMHSA implementation closely, push for evidence-based harm reduction funding that aligns with the VA's proven model, defend Housing First against ideological attack, and ensure that new treatments reach marginalized communities rather than only those with private insurance and stable housing.
The tools to address substance use disorder exist. What remains absent is a policy framework that treats people who use drugs as deserving of care rather than punishment. Until that changes, the Great American Recovery will remain a slogan, not a strategy.