Mark Mascolini (additions by Jules Levin)
Compared with other antiretroviral regimens, those containing protease inhibitors (PIs), especially atazanavir, raised odds of HIV escape in cerebrospinal fluid (CSF) . Major antiretroviral resistance mutations in CSF and plasma occurred more often in people with CSF escape in this 1063-person study. CSF escape cases with M184V/I mutations are likely to harbor variants with major DRMs that confer resistance to at least one ART drug and may reduce CNS ART bioavailability. Author suggests Guidelines may consider factoring in DRMs and CNS penetrance of ART regimens as a strategy to reduce intermittent CNS viral replication in people living with HIV infection.
In Q&A author stated that in mono therapy regimens they looked at there was higher odds of CNS escape. They looked preliminarily at darunavir and it was not as high as atazanavir. In Q&A author said we compared atazanavir to other boosted PIs and there was no significant difference in CSF escape. Alan Winston asked an important question in Q&A - since we use PIs often in patients with highly treatment experience who I think he said had prior treatment experience & perhaps prior drug and PI resistance, to which author said (listen to webcast) its a good question, individuals who entered this study were I think she said at least on their 2nd regimen in general, they have to look at the data to see why a patient switched. It appears there are some key unanswered questions to this study.
CSF escape--detectable HIV in CSF--occurs in 3% to 20% of antiretroviral-treated people worldwide. Researchers at Dana Farber Cancer Institute and colleagues from other centers noted several factors linked to viral escape in CSF: low nadir CD4 count, low but detectable viral load in plasma, HIV infection duration, and antiretroviral resistance mutations. They conducted this study to weigh the impact of antiretroviral therapy (ART) on probability of CSF escape in people enrolled in three cohorts in 2005-2016, and to determine the frequency of drug resistance mutations in reverse transcriptase and protease genes in HIV-infected participants on ART with CSF escape
The researchers assembled 1063 cohort members with paired CSF and plasma samples. Everyone was 18 or older and had a plasma viral load at or below 400 copies. The investigators defined CSF escape as (1) CSF viral load greater than plasma load, or (2) CSF load above 50 copies if plasma load undetectable. Through a median follow-up of 4.4 years, 77 people (7.2%) had viral escape.
Average age was similar in the total cohort and the escape group (46.2 and 45.8), and about 82% of both groups were men. A higher proportion of the escape group had HIV for more than 10 years (79.2% versus 60.9%). Compared with people without CSF escape, those with escape at time of their first escape had a lower nadir CD4 count (46 versus 94) and a lower current CD4 count (394 versus 496) but a similar median plasma viral load (50 copies). PI regimens proved more frequent in people with CSF escape (66% versus 47%), while antiretroviral central nervous system penetration-effectiveness (CPE) score was slightly lower in the escape group (7.7 and 7.9).
Individuals with escape were more likely to have symptomatic cognitive impairment. Three individuals were on an integrase inhibitor at the time of their escape. The longer a person is infected with HIV the greeter is the probability of CSF escape.
Generalized mixed-effects models considering age, nadir CD4 count, number of lumbar punctures, and plasma viral load below 50 copies linked PI therapy (versus non-PI regimens) to tripled odds of CSF escape (odds ratio [OR] 3.1, 95% confidence interval [CI] 1.9 to 5.0), PI+NRTI therapy remained an independent predictor of CSF escape (OR 3.8[95% CI: 2.0-7.6]) when analysis was restricted to visits with plasma VL <50 copies/ml.. Atazanavir regimens (versus non-PI regimens) tripled odds of CSF escape (OR 3.1, 95% CI 1.9 to 5.3), while nonatazanavir PI regimens had a slightly smaller impact (OR 2.7, 95% CI 1.6 to 4.5) when compared to other ART regimens. Lower nadir CD4 count, younger age, and plasma load above 50 copies also independently predicted CSF escape. Study author said - pharmacokinetic studies show that atazanavir levels do not consistently achieve the IC50 for wild-type HIVin the CSF.
Major reverse transcriptase and protease resistance mutations proved more frequent in CSF and plasma in people with than without CSF escape. In particular, the researchers detected the M184V/I mutation (conferring resistance to 3TC and FTC) combined with thymidine nucleoside analog mutations substantially more often in CSF and in people with CSF escape (34 of 56 CSF samples, 61%), and more often in plasma - 30% (16/55) . M184V appeared in only 3 of 43 CSF samples (7%) from people without CSF escape. in 23% (13/56) of CSF cases, M184V/I mutation was detected with at least one TAM in CSF and 15% (8/55) in plasma. Estimated CPE score proved suboptimal in people with CSF escape and M184V/I based on GSS-adjusted CPE score (GSS = genotypic susceptibility score). In this cohort the median CPE was 7 consistent with average CPE scores with standard regimens suggesting patients with M18$ were not on low CPE penetrant regimens however CPE scores do not account for resistance mutations. So they generated an adjusted CPE score that integrates information from HIV-1 genotypes providing a score for each drug based on their resistance profile. When resistance mutations were accounted for the adjusted CPE value was 6 in plasma and 3.5 in CSF, values that would be considered suboptimal CNS penetrants for ART regimens. Both resistance and inefficient ART penetration may be at play here in intermittent CNS viral replication.
The researchers suggested that "CSF escape cases with M184V/I mutations are likely to harbor variants with major drug-resistance mutations that confer resistance to at least one ART drug and may reduce CNS ART bioavailability." They believe their findings suggest that optimizing ART regimens could cut the risk of asymptomatic CSF escape.