February 2019: HIV and STDs

Integrase inhibitors may have a lower barrier to resistance in HIV-2 infection.  The authors, who manage a large cohort of folks with HIV-2 in Spain, retrospectively studied 44 patients given INSTIs as either initial therapy or rescue therapy after a previous treatment failure.  All patients received two NRTIs, and half of the 26 patients given an INSTI as rescue therapy also received a boosted protease inhibitor.  Clinicians prescribed raltegravir in 28 cases, dolutegravir in 9 cases, and elvitegravir in 7 cases.  After a median 12 months of followup, 16/18 patients (89%) starting an INSTI as initial therapy achieved and maintained an undetectable HIV-2 viral load.  However, among the patients given INSTIs as a rescue therapy, only 17/26 (65%) achieved an undetectable viral load over a median 13 months of followup.  Twelve patients (27%) developed INSTI resistance mutations during the study, and these were frequently associated with new reverse transcriptase and/or protease mutations.  Most of the treatment failures occurred with raltegravir (which has a lower barrier to resistance than dolutegravir), but failure with dolutegravir occurred in 3/9 cases and in one case was associated with emergence of dolutegravir resistance (via the combination of mutations R263K and E92G).

This high rate of INSTI failure among the treatment-experienced patients – and particularly dolutegravir failure – stands in contrast to high success rates seen in HIV-1.  Based on these data, I would be quite cautious trying dual therapy in HIV-2 infection (e.g. dolutegravir-lamivudine or dolutegravir with boosted darunavir– obviously we wouldn’t do dolutegravir-rilpirivine because the NNRTIs are not efficacious for HIV-2), and would probably stick to the higher-barrier INSTIs like dolutegravir and bictegravir when choosing to use that class. 30753573

Women’s susceptibility to gonorrhea infection is associated with menstruation and mediated by elevation of vaginal pH.  The female reproductive tract’s immune defenses are pretty dang impressive – there aren’t many other places where you can find a more or less direct communication between sterile and non-sterile sites in the human body.  In addition to your ubiquitous adaptive and innate immune systems, the vagina continually sheds its epithelium to mechanically remove pathogens, secretes antimicrobial peptides like defensins and lactoferrin, and produces mucus that both A) acts as a physical barrier between pathogens and the cells of the vaginal mucosa and b) recruits and sustains Lactobacillus spp., which contribute to this symbiotic relationship by producing lactic acids that lower the vaginal pH to levels inhospitable to most other microorganisms. Pretty cool, right?

So the authors of this study, noting that between a quarter and a third of women are able to resist N. gonorrhea infection when having vaginal intercourse with a gonorrhea-infected man, looked for the factors contributing to these women’s STD resistance by prospectively studying female contacts of men with gonorrhea to compare those who did vs did not also have the infection.  The enrolled women provided information about their sex practices, menstrual cycle, and use of douching and contraceptives, and additionally underwent physical and laboratory examinations.  The authors performed both cervical cultures and PCR for gonorrhea on each woman to determine whether they had the infection.

Sixty-one women participated in the study (mean age 25; range 18-39), of whom 62% had acquired gonorrhea.  Comparing the women who did versus did not acquire gonorrhea, there were no differences in the groups’ sexual practices, use of douching, and use of contraceptives. Eight participants had a Trichomonas coinfection, which was not associated with acquisition of gonorrhea; on the other hand, 7/8 patients with Chlamydia coinfection had contracted gonorrhea as well.  The incidence of bacterial vaginosis was 28%, and BV wasn’t associated with having gonorrhea.

So what was associated with having acquired gonorrhea?  First, cervical discharge, which was more often present in the infected (42% vs 17%; p<0.05%), no surprise given that’s a sign of N. gonorrheae infection.  More importantly, vaginal pH over 5.0 was strongly associated with increased risk of having acquired gonorrhea (OR 20.4; p=0.017), and this risk increase was linear, with an additional 3.6-fold increase in risk of gonorrhea for each successive 1.0 increase in the vaginal pH (p=0.022).  This association was stronger the closer the measurement of vaginal pH was to the date of exposure, which lends even more credence to the association.  In multivariate analysis, active menstruation was strongly associated with acquiring gonorrhea (OR 77.3; p=0.032), as was, to a lesser extent, exposure during ovulation (OR 23.7; p=0.08).

Previous studies have demonstrated a link between BV, elevated vaginal pH, and infections with gonorrhea or chlamydia, but the literature has been inconsistent on this point, and this study really suggests that it’s the change in pH, and not BV itself, that likely confers susceptibility to acquiring gonorrhea.  I suspect race may be a confounding factor here, as previous studies have shown that the vaginal microbiome varies by race and in particular that women of African descent are more likely to have a polymicrobial vaginal microbiome rather than one dominated by Lactobacillus.  Since the study prior to this one also failing to show an association between BV and gonorrhea or chlamydia coinfections was done in mostly black women (15976598), it seems likely to me that some women have vaginal bacterial symbionts other than Lactobacillus that aren’t disrupted by BV.  It would be interesting to know whether BV is associated with elevated vaginal pH among all women, or only in those whose microbiome is predominated by Lactobacillus at baseline. 30308531

Is it safe to prescribe doravirine to a patient who has previously taken other NNRTIs?  Doravirine is a new and relatively high-barrier to resistance NNRTI, which as a class has a lot to recommend it – namely, few side effects, few medication interactions, no adverse effects on the lipid profile, and friendly dosing.  Unfortunately, for all those reasons – and because NNRTIs were some of the first “3rd drugs” used in effective combination antiretroviral therapy, a large portion of people living with HIV are NNRTI-experienced.  To make matters worse, you may miss NNRTI resistance mutations in someone who currently isn’t on NNRTI therapy (because the mutations can be archived) or who has achieved viral suppression on another regimen (because you can’t get an RNA genotype – and the DNA proviral genotype isn’t 100% sensitive). So, is it safe to use doravirine in these people?

The authors assessed doravirine resistance in 6893 NNRTI-experienced patients who underwent genotype testing for a virologic failure; of these, 64% had previously received efavirenz, 54% nevarapine, 8% rilpirivine, and 7% etravirine.  They defined intermediate doravirine resistance as any of V106A/M, Y188C/H, V108I, or K103N plus P225H, and high-level doravirine resistance as any of Y188L, M230L, G190E, V106A/M plus F227L, or V106A/M plus L234I.  The incidences of resistance were 13% (intermediate) and 7% (high-level).  In multivariate analysis, exposure to efavirenz (OR 1.52) and etravirine (1.91) was associated with doravirine resistance, whereas doravirine resistance was lower among patients previously exposed to rilpivirine (OR 0.39).  These associations were mediated by the presence of the Y188L, the most commonly encountered doravirine resistance mutation.

So in summary, previous use of NNRTIs is associated with doravirine resistance, although the overall incidence of resistance is low.  Efavirenz and etravirine are the highest risk previous exposures, whereas prior use of rilpivirine does not contribute much to risk of doravirine resistance. 30769200

Speaking of NNRTIs, there’s a new multicenter retrospective study showing the efficacy and safety of dual therapy with darunavir-ritonavir plus rilpirivine.  The study included a total 161 patients (median age 49 years; median duration of ART 14 years; 29% with a prior diagnosis of AIDS and 37% with a prior history of virologic failure).  A quarter of patients had a baseline viremia, which ranged from 50-1000 copies/ml.  By week 24 after the switch to dual therapy, 88% of patients remained on darunavir-ritonavir and rilpivirine without virologic failure; when excluding patients who switched regimens due to reasons other that virologic failure, the efficacy was 95%.  So, while certainly not all dual therapies are created equal (Truvada still isn’t a complete regimen, y’all), I think the data is pointing toward the adequacy of any two-drug combination that includes either dolutegravir or darunavir-ritonavir. 30819101

Which is better for 2nd line therapy after treatment failure, dolutegravir or lopinavir-ritonavir?  I’m pretty sure everyone who takes care of folks with HIV could correctly guess the answer to this, but hey, now there’s a randomized control trial.

Over a two-year period, this multicenter study enrolled 627 patients who had experienced virologic failure after at least 6 months of therapy with an NNRTI and two NRTIs, randomizing them 1:1 to receive two NRTIs plus either dolutegravir 50mg daily or lopinavir-ritonavir (given either 800mg-200mg daily or 400mg-100mg twice daily).  The primary outcome was the proportion of patients who achieved and maintained viral suppression through week 48 of therapy.  This was intended as a non-inferiority trial for dolutegravir, with a margin of 12%.  That’s WAY too generous if you ask me – with how well-tolerated modern ART is, I wouldn’t willing prescribe an regimen with more than 5% difference in failure rate versus the best available alternative.  It’s also funny, because the implicit assumption is that dolutegravir is going to be inferior to lopinavir-ritonavir, which I think most experts would agree is backwards. But that was the study design.

Anyway, at week 48, 261 (84%) of the patients in the dolutegravir arm had achieved and maintained viral suppression versus 219 (70%) of the patients in the lopinavir-ritonavir arm (adjusted difference 13.8%, 95% CI 7.3-20.3).  One way of interpreting this data is to say that dolutegravir was superior to lopinavir-ritonavir (because the 95% confidence interval doesn’t cross zero), which the authors did.  Another interpretation, which the authors also stated, is that dolutegravir is non-inferior to lopinavir-ritonavir (because the 95% confidence interval also does not cross the pre-specificied threshold of -12%), which again is silly because who would have expected lopinavir-ritonavir to outperform dolutegravir?  Finally, one might reasonably argue that lopinavir-ritonavir does not meet criteria for non-inferiority versus dolutegravir (because the 95% confidence interval does cross 12% - which was the author’s pre-specified margin in the other direction).  This is all to say that dolutegravir ought to be used in preference to lopinavir-ritonavir when patients in resource-limited settings fail their first-line, NNRTI-based regimens. 30732940