An ongoing dialogue on HIV/AIDS, infectious diseases,
March 14th, 2021
Really Rapid Review — CROI 2021 Virtual
For a few years in the early 2010s, the Conference on Retroviruses and Opportunistic Infections (CROI) — in my opinion our premiere HIV scientific meeting — covered almost as many hepatitis C clinical trials as those on HIV. Or at least it seemed that way.
This made sense at the time — the startling success of non-interferon-based HCV treatments made for exciting news, and rapid progress. Here’s what I wrote in 2013 in another patented, copyrighted, and trademarked CROI Really Rapid Review™, for which NEJM Journal Watch receives many millions of dollars in royalties:
The results of the sofosbuvir and ledipasvir study — 100% response in both naives and prior null responders — provided one of the more exciting clinical trial results I’ve seen in years, small sample size notwithstanding.
Remember those days? Well, this CROI had barely any HCV at all, a sign of progress.
(And I was kidding about the royalties, in case you were wondering.)
Of course, in 2021 we are in the midst of a pandemic, so it’s not surprising that this CROI had plenty of COVID-19 studies, both on COVID-19 alone and on the combination of HIV and COVID-19. Here are some highlights, starting with HIV, then the HIV/COVID-19 studies, and then some interesting COVID-19 alone papers.
The links are to the presented abstracts, which requires a conference account (at least it does today) — however, as with other HIV meetings, plenty of the presented material appears on the invaluable NATAP page.
- In the NADIA study, second-line therapy with dolutegravir was non-inferior to darunavir-ritonavir, and tenofovir/FTC non-inferior to ZDV/3TC. This beautifully done randomized clinical trial included patients with extensive NRTI resistance — lots of K65R plus M184V. Though conducted in Africa, it has lessons for us all, including: 1) residual activity of NRTIs even with high-level resistance; 2) imprecision of consensus interpretations of genotype tests, 3) no need for ZDV in those with K65R (phew), and 4) the need to monitor for INSTI resistance — even though it’s rare (just 4 cases out of 235 randomized to DTG), it has major clinical implications.
- In a large, US-based clinical cohort of people with advanced HIV disease, BIC/FTC/TAF was the three-drug regimen least likely to be discontinued. Also, a higher proportion had viral suppression than on boosted darunavir. These two strategies are being directly compared in the ongoing LAPTOP study in this very population. Speaking of people with advanced HIV disease …
- A cohort of 912 patients starting ART with various regimens favored initial integrase-based treatment over regimens with NNRTIs or PIs. Benefits were lower discontinuation rates and mortality. And more on this population …
- In a small (n=101) randomized trial in people with CD4 < 100, ABC/3TC/DTG had fewer drug discontinuations than ABC/3TC plus darunavir/ritonavir. Bacterial translocation was also lower with the DTG-based regimen. Note that higher rates of discontinuation of PI-based regimens have been observed before in patients with advanced disease. Speaking of this DTG versus DRV comparison …
- In an open-label randomized trial (n=306) in treatment-naive patients, DRV/c/FTC/TAF did not meet non-inferiority versus ABC/3TC/DTG. In other words, it was not noninferior — now you can show off to your statistician friends by using that double-negative language. Notably, weight gain was a bit more with DRV than DTG. (Shrugs.)
- In treatment-naive studies of BIC/FTC/TAF and DTG plus either TAF/FTC or ABC/3TC, viral suppression was still very high, even with transmitted drug resistance. The investigators used retrospective baseline next-generation sequencing of PR, RT, and integrase, with a 15% cutoff. Importantly, pre-study standard genotyping that identified resistance to the NRTIs was exclusionary.
- Injectable cabotegravir and rilpivirine given every 8 weeks remains non-inferior to injections every 4 weeks at week 96. The current approval in the U.S. is for the 4-week strategy; approval of the 8-week approach is expected soon.
- For those switching to DTG/3TC, having M184V detected more recently increased the risk of virologic failure. This was compared to those with no historical M184V, and those in whom it had been found more than 5 years previously. Of course, this begs the question — why switch to DTG/3TC with a known M184V? Despite the ongoing activity of 3TC, this is not how the DTG/3TC regimen was studied. I wouldn’t do it.
- In a clinical cohort, predictors of 10% or more weight gain after switching to integrase-based ART were being female, underweight/normal weight, prior non-INSTI treatment, and the TDF to TAF switch. There was no difference between different INSTIs, however.
- Integrase-based ART was associated with an excess risk of cardiovascular events — but only after the first 6 months of exposure. It’s an observational study, so we cannot determine causality (just as we couldn’t with the first abacavir CVD study). And how do we explain that the effect was only temporary?
- In a database analysis of over 34,000 people starting or switching to integrase-based ART, new-onset diabetes or hyperglycemia occurred in 2.5%. This was significantly higher than non-integrase-based ART, with dolutegravir carrying the highest risk. Bictegravir was not included (database stopped in 2018), and there was very little TAF use.
- For pregnant women starting ART during pregnancy, DTG + TAF/FTC, DTG + TDF/FTC, and EFV/TDF/FTC had comparable viral suppression at 50 weeks’ postpartum. Pregnancy outcomes favored the TAF arm of the study — will our guidelines be updated soon based on this trial?
- In that study, the women in the DTG + TAF/FTC arm gained the most weight. However, this weight gain was closest to the recommended weight gain during pregnancy and also associated with a lower risk for any adverse pregnancy outcome, suggesting that pregnant women treated with EFV/TDF/FTC don’t gain enough weight.
- When started late in pregnancy, dolutegravir achieved viral suppression more rapidly than EFV. In this randomized trial of 237 women, there were three in utero HIV transmissions in the DTG arm, and one post-partum transmission in the EFV arm.
- A 4-month daily regimen of rifapentine and moxifloxacin was noninferior to the standard 6-month regimen for tuberculosis treatment. The participants also received isoniazid and pyrazinamide for the first 8 weeks. Anything that shortens TB treatment is a real advance!
- In highly treatment-experienced patients, oral lenacapavir induced a 1.93 log viral load decline (vs. 0.29 in placebo) after 14 days when added to a failing regimen. At this point, treated patients switched to subcutaneous lenacapavir given every 6 months, along with an optimized background regimen. The study is ongoing (and includes a non-randomized arm), and 19/26 patients have viral suppression. Two experienced viral rebound with emergent resistance. Not much information about the background regimens used with this novel agent.
- The investigational NNRTI MK-8507 has a PK profile that supports once-weekly dosing. It will be paired with islatravir for treatment of HIV. The resistance profile is likely to be similar to doravirine.
- Further evaluation of daily versus on-demand PrEP with TDF/FTC shows a low incidence of HIV in both strategies. One wonders whether TAF/FTC would be this effective given on-demand — suspect it would be, but it has not been tested this way.
- Laboratory evaluation of cabotegravir PrEP failures in HPTN 083 was challenging. And that’s an understatement! The issue is that CAB delayed detection of infection using standard HIV testing algorithms (point-of-care antibody testing, standard 4th generation antigen/antibody tests). Also, a small number failed despite adequate adherence and measured CAB levels, with development of INSTI resistance. Should we be using viral load tests to monitor our patients on CAB-based PrEP? Likely yes.
- A cost-effectiveness analysis of long-acting CAB for PrEP suggested that it would not be cost-effective if priced similarly to cabotegravir-rilpivirine. Though CAB is more effective than generic TDF/FTC for HIV prevention, the estimated $17,000/year cost difference puts the cost-effectiveness ratio at more than $1 million per quality-adjusted life-year. A limitation of this analysis, of course, is that we don’t know yet what CAB for PrEP will cost!
- As both a pill and as a drug-eluting implant, islatravir moves forward as a long-acting PrEP option. The pill projects to be once-monthly; the implant could be yearly.
- In consecutive sampling of 1076 people with HIV (PWH) in Spain, 8.6% were seropositive for SARS-CoV-2. Those on TDF/FTC were less likely to test positive — even after controlling for underlying comorbidities (an adjustment not made in this group’s previously published paper).
- In a large national cohort with nearly 600,000 cases of COVID-19, PWH and solid organ transplant recipients were more likely to be hospitalized than controls. Both groups had a high burden of comorbidities — which increases the risk of adverse outcomes from COVID-19 for all groups, including those with HIV.
- A single infusion of bamlanivimab and etesevimab given to outpatients with COVID-19 at high risk for severe disease reduced the risk of clinical progression (hospitalization or death). There were no deaths in the treatment arm, 10 in placebo. Viral load decline and clinical recovery were also faster with treatment.
- Compared to placebo, the oral antiviral molnupiravir reduced the proportion of patients who had positive SARS-CoV-2 viral cultures by day 5. Unusual endpoint, but the results do demonstrate an antiviral effect. Several phase 3 clinical trials are ongoing, both in the inpatient and outpatient setting.
- Banlanivimab was effective in preventing COVID-19 in skilled nursing facilities. Those who did get COVID-19 despite treatment had lower viral loads and more rapid resolution of symptoms compared to the placebo group. While vaccination will undoubtedly be the mainstay of preventive strategies for COVID-19, in certain settings monoclonal antibodies may have a role.
- In another study of COVID-19 prevention, casirivimab plus imdevimab given subcutaneously to asymptomatic household contacts of active cases prevented 100% of symptomatic infections. Among those who did acquire COVID-19, viral loads were lower and PCR positivity was of shorter duration compared to placebo. Note the subcutaneous route of administration in this study, as it obviated the need for IV access — a big advance given the difficulty of arranging these treatments under the current Emergency Use Authorization.
- For 23 patients with refractory COVID-19 due to B-cell deficiency, convalescent plasma led to clinical recovery in 20. This population — receiving rituximab or similar anti-CD20 agents — has proven particularly vulnerable to COVID-19, often with prolonged symptoms and persistently high levels of virus.
- Treatment of hepatitis C with sofosbuvir/velpatasvir achieved cure in 95% of patients with minimal monitoring. And I mean minimal! (The study was called “MINMON”.) This single-arm study with 399 participants dispensed all tablets for a 12-week course at entry, had no on-treatment laboratory monitoring, and had just one week 22 visit to assess for cure (SVR). Aside from two phone contacts for adherence monitoring, that was it — amazing.
- Didn’t see all the plenaries or invited talks — attending a virtual conference can be strangely exhausting — but among those I did see, can highly recommend Dr. Shahin Lockman on HIV treatment in pregnancy, Dr. Jose Arribas on dual versus triple ART, and Linda-Gail Bekker on long-acting options for HIV prevention. Bravo!
You’ll note I started and finished this post mentioning hepatitis C studies — which due to our success in treatment now have a very small footprint at CROI. Let’s hope the same happens to COVID-19 in the not-so-distant future!
And since this CROI was supposed to be Chicago and bring in people from all over the country (and the world), here’s a wonderful tour around North American accents, with of course a stop in the land of da Bears — right around 3’40”.