HIV and ID Observations

The Conference on Retroviruses and Opportunistic Infections (CROI) returned to Seattle this past week for its 24th meeting. It’s the 4th time CROI has been held in Seattle, an excellent city for a meeting of this size, which includes “only” 4200 people. The convention center is pleasant and user-friendly — big but not cavernous, actually encourages interactions with colleagues — and there are numerous hotels and restaurants within walking distance, plus more Starbucks per square foot than any place on the planet.

From a content perspective, the big change for CROI 2017 was the return of numerous studies on antiretroviral therapy, studies involving both approved and investigational agents. The last several years, by contrast, had relative dominance of pre-exposure prophylaxis and hepatitis C studies. With PrEP, one had the sense at the meeting that we’re now waiting for the next strategies (long-acting injectables, for example).

As for hepatitis C, well that’s been all but solved (except for the implementation part). How do you improve on 97%-plus cured. Hooray!

On to the summary, a Really Rapid Review™ of the most interesting studies at the conference (at least from one perspective, ahem). Links are to the conference website (as usual excellent), abstract #’s are in brackets, and be sure to check out some of the oral presentations here. The list is roughly organized into epidemiology/prevention, treatment, cure, and opportunistic infections. Please tell me what I missed in the comments section.

And, because I was involved in a few of the studies listed below, for full disclosure I’ve inserted a special Puppy text emoticon to draw attention to that fact. ੯ੁૂ‧̀͡u  (Make sure you click on them. Woof.)

• New HIV infections in the USA continue to decline [30]. From 2008–2014, the annual number of HIV infections dropped 3.6% per year, and the percentage of persons with undiagnosed infection declined at around the same rate. That’s the good news. The bad news is that new infections continue to increase in young black and Hispanic men who have sex with men. For the record, we’re down to 37,600 cases/year in the USA, and I expect this will continue to drop since PrEP use didn’t really start until 2014.
• Among participants in the HIV Outpatient Study, the proportion of time spent with HIV RNA > 1500 has dropped dramatically since 2000 [31]. This higher proportion of diagnosed, treated, and suppressed patients must be the primary reason why HIV incidence has declined in the USA.
• Doxycyline administered as post-exposure prophylaxis reduced the incidence of chlamydia and syphilis among participants in the IPERGAY PrEP study [91LB]. No effect on gonorrhea. Significantly more GI adverse events in doxycycline recipients. Interesting results — but definitely not enough evidence yet to make this a policy recommendation, a fact stressed by the lead author, Jean-Michel Molina.
• A case of PrEP failure occurred despite good adherence and no transmitted resistance [953]. The case illustrates the likely importance of “inoculum” in PrEP failure — the person had a very high number of potential exposures. It’s worth looking at the full poster presentation, as the clinical and laboratory manifestations of acute HIV infection were unusual, possibly influenced by being on TDF/FTC.
• In treatment-naive patients, the investigational NNRTI doravirine was non-inferior to boosted darunavir [45LB]. 84% vs 80% < 50/copies at week 48, with a trend toward doravirine being better with very high (>500,000) baseline HIV RNA. Favorable aspects of this treatment included once-daily dosing, few CNS side effects, and a better metabolic profile that darunavir. Only 1 of 364 doravirine-treated patients developed resistance, a low number for NNRTI strategy; none did in the darunavir arm. And note the abbreviations — “DOR” vs “DRV”! A coforumulated DOR/TDF/3TC tablet is in development. ੯ੁૂ‧̀͡u
• In virologically suppressed patients with no history of resistance, dolutegravir plus rilpivirine maintained virologic suppression as well as continuing current therapy [442]. 95% < 50 in each study arm at week 48. Only one of 513 patients in the DTG + RPV arm failed with resistance (she clearly stopped therapy). There was a slightly higher rate of discontinuations due to adverse events in switchers (3% vs <1%), likely at least in part due to the open label design. Note that this is a rare success for an NRTI-free treatment strategy, a fact emphasized by Jose Arribas in his outstanding oral presentation on treatment simplification. A coformulated DTG/RPV tablet is in development, to be used in follow-up studies.
• In a phase 2 double-blind clinical trial, bictegravir + TAF/FTC was comparable to dolutegravir + TAF/FTC [41]. Responses 97% and 91% at week 48, respectively; more details in the published paper. Bictegravir is an investigational once-daily, unboosted integrase inhibitor with activity against wild-type and most resistant viruses and few predicted drug interactions. A coformulated BIC/TAF/FTC tablet is in development — and being used in the phase 3 program, which is fully enrolled.  ੯ੁૂ‧̀͡u
• In a single-arm study, dolutegravir plus lamivudine maintained virologic suppression [458]. Study included 110 patients with no history of failure and who tolerated an 8-week “lead in” of DTG + 2NRTIs first. One virologic failure (no resistance), one almost failure with low-level viremia. Comparative clinical studies ongoing. Study is called “LAMIDOL”, which sounds like a skin cream.
• Dolutegravir monotherapy increases risk of virologic failure and resistance [451LB]. So we now have several studies that show this strategy to be a bad idea. To quote the abstract: “The genetic barrier of DTG monotherapy is insufficient to allow for maintenance monotherapy.” Here are more details on the resistance pathways in several other studies. Let’s stop doing this, ok? Thoughtful editorial to this effect by Joel Gallant says the same thing, though much more eloquently.
• Transmitted drug resistance did not reduce antiviral efficacy in this large treatment study [43]. This is a surprising finding, as several prior studies have shown that transmitted NNRTI resistance does reduce the activity of these regimens. One possible explanation is the greater potency of TDF/FTC vs AZT/3TC.
• Case report of DTG resistance in a treatment-naive patient [500LB].  Baseline HIV RNA 1,970,000 and CD4 78. The fact that CROI accepted a single case of this event shows how incredibly rare this is.
• Darunavir but not atazanavir associated with an increased risk of cardiovascular events [128LB]. It’s a DAD analysis, and the effect (a 53% increase) was comparable to lopinavir and indinavir. Is this a class effect, ameliorated by atazanavir’s bilirubin-raising properties?
• More discontinuations due to neuropsychiatric side effects on DTG than RAL or EVG/c [651]. Concomitant ABC/3TC with DTG appears to be a strong contributor. Note that rates of d/c due to adverse events overall did not differ between the drugs.
• High DTG plasma concentrations associated with more neuropsychiatric adverse events [426]. Study done in Japan — could tolerability be linked to BMI or pharmacogenomic issues?
• More adverse events in EFV rapid metabolizers [384]. Study done in Botswana, and runs counter to analysis of suicidality done in (mostly) US-based studies, where those with genotypes associated with slow metabolism appeared to be at higher risk.
• EFV use as a first-line agent in the USA has dropped dramatically [454]. Now less than 10% start it as first line, a stark contrast from the pre-integrase era.
• TAF superior to TDF after 144 weeks [453]. The most important safety finding is that 0/866 on TAF vs 12 (11 if you remove “bladder spasm”)/867 on TDF developed clinically significant renal issues, a difference that is highly statistically significant. ੯ੁૂ‧̀͡u
• Resistance to integrase inhibitors remains extremely rare [478]. In a 9 site survey, baseline (transmitted) integrase resistance was 0.04%; post-treatment was 0.4% — both very good news.
• Should we be sending baseline integrase genotypes before starting therapy [493]? Based on this modeling study, the answer is a resounding NO, for three reasons:  1) the prevalence of transmitted integrase resistance is fortunately still very low (see above); 2) DTG-based regimens will likely succeed even when resistance to first-generation INSTIs is present; and 3) even if virologic failure occurs, it will delay use of a salvage regimen only briefly. ੯ੁૂ‧̀͡u
• Many novel drugs in various stages of development. These include (deep breath) ibalizumab [449LB] (anti-CD4 monoclonal antibody), PRO140 [437] (different humanized anti-CD4 monoclonal antibody), UB-421 [450LB] (yet another monoclonal antibody that blocks viral entry), elsulfavirine [452LB] (NNRTI with 8-day half-life), MK-8591 [435] (potent and very long acting NRTI), GS-PI1 [433] (an unboosted once-daily protease inhibitor), GS-9131 [436] (NRTI with activity against essentially all NRTI-resistant isolates), and GS-CA1 [38] (capsid inhibitor that in vitro is more potent than any other antiretroviral agents). Wow that’s a lot!
• Even very early HIV treatment does not prevent virologic rebound after stopping ART [124].  Eight started treatment during “Fiebig I acute infection” (VL+, p24-, Ab-), which is known to induce a lower HIV reservoir, yet all rebounded by 9 weeks.
• A prolonged virologic remission in a patient undergoing a stem cell transplant [319]. The patient continued suppressive ART for 784 days after transplant, and had no detectable virus (HIV RNA or DNA) using many techniques. Rebound occurred after 288 days off ART, longer than in the “Boston patients”. Note that donor was CCR5 wild-type, not CCR5 negative.
• Latency-reversing agent plus HIV immunization may induce “viremic control” [119LB]. Strategy tested was “MVA.HIVconsv” vaccination, then three doses of romidepsin, then another vaccination; 5 of 13 maintained low-level viremia after treatment interruption. Importantly, study subjects started ART soon after HIV acquisition, hence were likely to have low viral reservoirs.
• TLR7 agonist induced “viral remission” in two of four Indian Rhesus macaques [338LB]. One of the agents (GS-9620) is currently in human trials.
• Does ART with integrase inhibitors increase the risk of immune reconstitution inflammatory syndrome (IRIS)? The answer is yes, based on two studies conducted in non-TB endemic regions ([731] and [732]). As DTG-based regimens are increasingly adopted world-wide, we should expect IRIS incidence to increase. And count me in the group that thinks this is a good thing — IRIS is a marker for faster virologic response and immunologic recovery. (Not everyone agrees.)
• Pre-emptive prednisone reduced the risk of TB IRIS in patients with advanced HIV disease [81LB]. All patients (CD4 < 100, diagnosed with TB) received ART and TB therapy. The study compared prednisone 40 mg/day for 2 weeks then 20 mg/day for 2 weeks versus placebo. Incidence of IRIS was 47% in placebo arm, 33% in prednisone arm (p=0.02). Incidence of infectious complications or malignancies did not differ between study groups.
• In cryptococcal meningitis, a single high dose (10 mg/kg) of liposomal amphotericin reduced CSF fungal burden as fast as standard dosing (3 mg/kg/day for 14 days) [82]. Based on these encouraging early results, this strategy will be evaluated in a randomized clinical endpoint study.
• Researchers intensively studied 5 patients with paradoxical CD4 decline on suppressive ART [226]. 4 of 5 had CD4 < 50 (other was 74). Unlike immuologic non-response, these patients did not have elevated inflammatory markers or immune activation, but did have higher levels of homeostatic cytokine IL-7. The title says it all:  “a novel, rare and perplexing immunologic outcome.” (I’ve seen this once in my career, for the record.)
• For neurosyphilis, intravenous penicillin and procaine penicillin plus probenecid achieved similar clinical outcomes [750]. Although limited by the observational design, there is so little modern data on treatment of neurosyphilis this poster is still of great interest.
• Widespread use of HCV therapy in Holland has reduced the incidence of acute HCV [137LB]. An excellent example of treatment as prevention.
• No dose adjustment should be required when glecaprevir and pibrentasvir are administered with either ECF-TAF or ABC/3TC/DTG [413]. Approval of this pan-genotypic HCV treatment is expected soon.
• American ginseng did not reduce HIV-related fatigue [669]. We can add this to the (long) list of studies that fail to show benefits from homeopathic non-FDA approved herbal remedies. Plus, I just like typing “American ginseng.”

If you’re interested in watching webcasts of interesting plenary or symposium talks, I can strongly recommend Demetre Daskalakis on ending the HIV epidemic in NYC (he has more energy than the rest of the 4199 CROI attendees combined), Charlie Flexner on long-acting ART (he dispelled several “myths”), and Carl June on how cellular treatments of cancer might apply to the HIV cure effort.

One caveat to this last talk is that Dr. June neglected to mention that there’s huge difference in acceptable risk between someone with refractory metastatic cancer versus stable HIV infection. So I’m mentioning it for him.

You’re welcome.

See you at next year’s CROI in Boston!

[youtube https://www.youtube.com/watch?v=FSvNhxKJJyU]