Ultra-rapid review: 2025 Conference on Retroviruses and Opportunistic Infections (CROI 2025)

Wow, what an interesting meeting – good and bad. The good part was the content, after all, as our leading HIV research meeting, CROI never disappoints, and this time there was indeed a lot of interesting content.

The not-so-good part , however, is the state of shock that many HIV clinicians and researchers are in. Due to the many actions of the current government, HIV research both domestically and globally has been directly affected, and there is hardly a topic or discussion that has not been affected. Some research abstracts have been retracted; some regular attendees have canceled their plans to attend meetings; and some planned projects will no longer be feasible in the current funding environment.

Now onto the regular portion of the Super Quick Review™.

BIC/FTC/TAF was superior to darunavir/cobicistat/FTC/TAF* as initial therapy for patients with advanced HIV. The importance of this study is that this population is rarely included in clinical trials—median CD4 is 41, and nearly half have HIV RNA > 500,000. The composite outcome of combined clinical and virological endpoints was not significantly different, but this result still supports BIC/FTC/TAF (or TLD) as initial therapy even for patients with advanced HIV, which is what most of us do.

Two experimental formulations of lenacapavir, given as a once-yearly intramuscular injection, achieved sufficient levels to be used for once-yearly PrEP. The researchers showed some very nice PK profiles, with levels exceeding those observed in the PURPOSE 1 and 2 trials. These data support a planned Phase 3 study.

Some information is available from HPTN 083 on subsequent treatment outcomes for people who acquired HIV during cabotegravir PrEP. That study included 47 people who acquired HIV, 22 of whom continued on INSTI-based ART and 25 on other regimens (PI- or EFV-based), with similar results in both groups. Those who had not received a CAB injection in the previous 6 months were less likely to have INSTI resistance prior to ART. A second study described 7 people who failed CAB PrEP in clinical care, 4 of whom had unreactive HIV antigen/antibody testing. Resistance was detected in some individuals only by more sensitive research assays, and its clinical significance is unclear.

In the DISCOVER trial, weight trajectories in people taking TAF/FTC for PrEP were statistically similar to those in the placebo group in iPrEx. Although limited by the cross-study comparisons, these studies of weight change in HIV-uninfected individuals are instructive because they eliminate the “restoration” effect of antiviral activity on weight.

Several “real-world” studies have shown that in clinical practice, rates of treatment failure with cabotegravir plus rilpivirine are associated with resistance and are comparable to those observed in clinical trials. Here are four: one (disclosure: co-investigator), two, three, and four. Specifically, the rate occurred in about 1% of those treated. These observational studies have limitations, most notably variability in treatment duration and the inability to capture follow-up of all patients. The fourth study mentioned above showed that 92 of 465 people (20%) from a single center discontinued CAB/RPV, with the most common reason being injection site reactions; only six people (1.3%) experienced virologic failure.

A detailed analysis of cabotegravir and rilpivirine drug levels found an association between lower concentrations and detectable viral loads >20 copies/mL. Whether low levels are associated with actual treatment failure is unclear, but I suspect that in a large enough cohort, they would be.

In a population of 81 viremic individuals at Grady Hospital in Atlanta, the use of long-acting antiretroviral therapy resulted in viral suppression in 92%. The median baseline CD4 cell count was 186, and 44% had prior HIV-related opportunistic infections. Two individuals experienced virologic failure, resistant to both cabotegravir and rilpivirine. Of note, treatment regimens were diverse, using all of our injectable options: cabotegravir/rilpivirine, ± lenacapvir, ± ibalizumab. It is exciting to see these excellent results in the urban South, which is notorious for significant barriers to optimal HIV treatment outcomes.

For those who experienced low-level viremia (50-199), those who switched therapy were more likely to achieve virologic suppression compared with those who did not switch. Note that this is contrary to the findings of most prospective intensification studies and does not match what I observe in clinical practice (except when switching from a PI to an INSTI – not reviewed here). One limitation is that this is an observational study, and there were significant differences between switchers and non-switchers in several ways that could have affected the results.

In a randomized trial, lenalidomide, given every 6 months in addition to two broadly neutralizing antibodies (bNAbs), maintained viral suppression in all but one patient. A total of 53 people received this treatment. Note that to qualify for this trial, participants had to have virus that was sensitive to both bNAbs (teropavimab + zinlirvimab, “TAB and ZAB”); only 99 of 200 screened met this criterion. This very long-acting injectable regimen is both exciting and unlikely to be widely used—there are too many barriers to successful implementation.

After stopping ART, administration of two long-acting bNAbs significantly delayed virologic rebound compared with placebo until week 20. Suppression rates of 75.0% were observed in participants who switched to bNAbs, compared with 8.8% in the placebo group. Among those who rebounded on bNAbs, some did so quickly and some much later. If the 8.8% suppression rate in the placebo group seems high to you—after all, they had stopped ART!—that’s because the study population included only suppressed HIV-infected people who started treatment at an earlier HIV stage, which is associated with lower viral reservoirs (and higher rates of viral control after interruption).

The monoclonal antibody N6LS, given every 4 months, was tested in combination with monthly cabotegravir. The good news is that most participants maintained viral suppression. The not-so-good news is that despite screening with a virus sensitive to N6LS, 4 of the 99 people who received this treatment had confirmed virologic failure—1 of whom developed a new INSTI mutation (Q148R). Can you tell I’m still not too enthusiastic about bNAbs in HIV treatment?

Among those who stopped all treatment, 4% maintained virologic control at day 84. This surprising level of viral suppression was observed in 6% of early ART initiators and 1% of late ART initiators. The median time to rebound above 50 copies/mL was 16 days. Because this study collected data from 382 people, it will provide good context to determine whether various cure interventions have an effect on the time to virologic rebound.

One elite controller lost virological control after 32 years, but contributed a large number of samples to research during that time. The researchers found that in this individual, despite an undetectable viral load, the balance between the viral reservoir and the immune response was dynamic. They used the term “outlier elite controllers” to describe those who had controlled HIV for more than 20 years; many, like this woman, are also outliers for their generous contributions to HIV research.

Two randomized clinical trials showed that a single-tablet doravirine/islatravir (DOR/ISL) can maintain viral suppression comparable to that of the baseline regimen. In one study, the starting regimen was BIC/FTC/TAF; this was double-blind. The other trial was open-label and compared DOR/ISL with any baseline antiretroviral therapy. Although it was encouraging that this one-tablet-a-day treatment did not produce the lymphocyte suppression seen with higher doses of islatravir, there did not appear to be any clear benefit from switching. A comparison of this combination with BIC/FTC/TAF in treatment-naïve patients is ongoing; it is similar to the previously published study, except that the islatravir dose is lower (0.25 mg per day).

Switching from DTG/3TC to BIC/FTC/TAF had comparable effects on weight, cholesterol levels, and systemic inflammation (IL-6) as continuing DTG/3TC. In some ways, this was a study that “fortified” DTG/3TC by adding TAF, if we assume that BIC=DTG and FTC=3TC. This is in contrast to the DYAD study, where patients were on BIC/FTC/TAF from the beginning. In summary – aside from the PASO DOBLE results (which I’m still struggling to understand), there doesn’t seem to be much difference between the two treatments when it comes to metabolic and inflammatory profiles.

An experimental integrase inhibitor, given on days 1 (baseline), 4, and 7, induced a viral load reduction of more than 2 logs, comparable to that observed with dolutegravir monotherapy. This drug, currently known as VH-184, is active against many INSTI-resistant strains and is being developed for long-acting therapy.

There are also two potential cases of HIV cure—one from Chicago and one from Oslo. Both had hematological malignancies and received cells from donors with the CCR5-delta 32 gene mutation, which confers resistance to HIV. Both have stopped antiretroviral therapy, although the Chicago case initially experienced viral rebound (and restarted ART). The key question, of course, is what these cases can teach us about HIV cure without resorting to this extremely toxic (and expensive) procedure.

In a randomized clinical trial, semaglutide stabilized epigenetic aging compared with placebo in HIV-infected people with lipodystrophy. In simple terms, “epigenetic aging” is a molecular marker of biological aging rather than calendar age; HIV accelerates this process in some people. These results (and many others) made me wonder, what else can’t these drugs do?

In an open-label randomized trial, letermovir for CMV suppression resulted in some unexpected benefits in inflammation, immune activation, CD4 counts, and even one test of functional status. Why unexpected? Because the study was actually stopped early due to an unexpected increase in sTNFR2 (usually a bad sign, I guess). These unexpected benefits raise the question of whether CMV suppression improves true clinical outcomes—an important thing to determine, since letermovir is currently very expensive, even for a short course, and is hardly an intervention that clinicians should be employing right now.

Compared with placebo, tecovirimat did not shorten the time to clinical resolution in Mpox. In this study, the time-to-improvement curves did overlap; this is indeed a negative finding. Viral clearance was not faster, either. While there is tremendous enthusiasm for this treatment approach in the summer of 2022, the results of two independent studies show the critical importance of conducting randomized trials in unproven therapies.

In a large North American cohort, the incidence of hepatitis C infection tripled between 1995-2005 and 2015-2023 among MSM who had sex with men and did not inject drugs. In contrast, among PDUs, the incidence remained stable (but high). The authors do not provide an explanation, but I suspect that the “U=U” messaging and the use of PrEP have undoubtedly contributed to the increase in condomless sex among MSM.

The HepB-CpG vaccine (trade name Heplisav-B) provides better and longer-lasting serological protection than the standard HepB-alum vaccine. This includes people who did not respond to previous HBV vaccines and those who have not been vaccinated against HBV before. 3 doses is better than 2, but it is unclear to me whether this benefit is worth the extra dose as standard treatment, since the commonly recommended regimen is 2 doses one month apart.

Brief COVID-19 section

For household contacts of COVID-19, ensitrelvir reduced infection incidence by 67% over five days compared with placebo. This is the first antiviral drug to show protection in COVID-19 post-exposure prophylaxis (molnupiravir and Paxlovid were ineffective), likely due to the drug’s very long half-life. It is already approved for treatment in Japan and Singapore; hopefully we will soon get approval for this post-exposure prophylaxis in the United States.

In the ACTIV-6 randomized clinical trial, metformin did not shorten the time to symptom resolution in adults with mild to moderate COVID-19. The study population was primarily healthy, especially when compared with other COVID-19 treatment studies. It will be interesting to see if any secondary endpoints of this study favor treatment, such as reductions in viral load or subsequent development of long COVID-19.

In a Phase 2 study, SNS812, an aerosolized broad-spectrum antiviral small interfering RNA (siRNA), significantly accelerated the time to resolution of several key symptoms of COVID-19. It also improves virus clearance time. The treatment is notable not only for its mode of administration (inhalation), but also for its predicted activity not only against SARS-CoV-2 variants but potentially against other coronaviruses as well.

Monsoon (2019) – A Quiet Film About Love, Culture, and Finding Where You Belong

Mysterious Ways (New Zealand)

Land of Storms (Viharsarok, 2014) – Between Love and Violence in Rural Hungary

Layla (UK)

We Three Queens: Season 1 & 2 (Canada) (Web Series)