CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

without NPCs at baseline. NPCs were based on verbatimmedical history; terms were coded and NPCs were defined as those within the MedDRA v22 system organ class Nervous System Disorders or Psychiatric Disorders. The primary objective was assessment of virologic response (HIV-1 RNA <50 copies/mL) at Week 48 by intent-to-treat FDA snapshot analysis in patients with or without NPCs in each treatment arm. Results: Among 725 patients in AMBER, 88 (D/C/F/TAF) and 99 (control) had NPCs. Overall, psychiatric comorbidities (125/187 [67%]) were more common than neurologic comorbidities (81/187 [43%]). Patients with NPCs, vs those without, were more likely to be black (17% vs 9%), from North America (37% vs 18%), use nicotine (52% vs 45%), and be drug users (26% vs 14%). Patients with NPCs had higher rates of early study discontinuation vs those without NPCs (D/C/F/TAF, 10% vs 5%; control, 10% vs 7%), which was largely driven by loss to follow-up. High virologic response rates (85-93%) were achieved at Week 48 regardless of NPCs (Table); while patients with NPCs had numerically lower response rates, no patients in either arm discontinued due to lack of efficacy or developed darunavir, primary protease inhibitor, or TAF resistance. Rates of discontinuation due to related adverse events (AEs) were low regardless of NPCs (Table). Patients with NPCs did not experience a higher incidence of neurologic or psychiatric AEs related to D/C/F/TAF. The most common (≥5%) neurologic AE, regardless of treatment arm or NPCs, was headache. For patients with NPCs, the most common (≥5%) psychiatric AEs were anxiety and depression (D/C/F/TAF), and depression and insomnia (control); no psychiatric AEs met this threshold among patients without NPCs. Conclusion: In AMBER, the presence of NPCs did not preclude virologic response in either treatment arm. Patients with NPCs were not at added risk of discontinuing due to AEs and did not experience a higher incidence of neurologic or psychiatric AEs related to D/C/F/TAF.

440 CSF CXCL-10 IS ASSOCIATED WITH THE PRESENCE OF LOW-LEVEL CNS HIV DURING ART Albert Anderson 1 , Bin Tang 2 , Florin Vaida 2 , Oluwakemi Okwuegbuna 2 , Daniel McClernon 3 , Reena Deutsch 2 , Suprateek Kundu 4 , Mariana Cherner 2 , Debralee Cookson 2 , Melanie Crescini 2 , Igor Grant 2 , Ronald J. Ellis 2 , Scott L. Letendre 2 1 Emory Center for AIDS Research, Atlanta, GA, USA, 2 University of California San Diego, La Jolla, CA, USA, 3 bioMONTR Labs, Research Triangle Park, NC, USA, 4 Emory University, Atlanta, GA, USA Background: The central nervous system (CNS) is a reservoir of HIV persistence during antiretroviral therapy (ART). Our group and others have demonstrated that both HIV RNA by single copy assay (SCA) and HIV p24 antigen by single molecule digital ELISA can be detected in cerebrospinal fluid (CSF) during ART. However, these markers require specialized protocols and are not always quantifiable during ART. Therefore, surrogate markers of HIV CNS persistence are needed that are widely available and readily quantifiable. Methods: We performed a cross-sectional analysis of persons with HIV (PWH) on combination ART with both plasma and CSF HIV RNA <50 copies/ml by conventional PCR. In addition to HIV RNA by SCA and p24 antigen by digital ELISA, we measured CSF CXCL10 and sCD30, immune activation markers that may reflect HIV persistence in the CNS. We also measured CSF neurofilament light chain (NFL) and neuron specific enolase (NSE), markers that reflect neuronal damage. Results are reported in pg/ml, with comparisons made with Wilcoxon rank sum. Logistic regression was performed with CSFHIV+ as outcome. Results: 66 adult PWH with virologic suppression on ART were analyzed. 19 (29%) were CSFHIV+ (positive by either SCA or p24). CSFHIV+ participants did not differ from those without detectable CSF HIV (CSFHIVneg) in terms of age, gender, race, current/nadir CD4+ T-cell count, CSF total protein, or duration of current ART regimen (all p>0.2). CXCL10 was significantly higher in the CSFHIV+ group compared to the CSFHIVneg group (median= 411 [IQR= 344-640] versus median= 312 [IQR= 205-468], p=0.008). In contrast, sCD30 was not significantly different (p=0.43) between the two groups (median= 8.97 [IQR= 4.05-14.58] in CSFHIV+ versus median= 7.04 [4.19-10.76] in CSFHIVneg). There was no significant difference in NFL between the two groups (p=0.85), but there was a trend towards higher NSE values (p=0.096) in the CSFHIV+ group. In logistic regression accounting for the effect of detectable plasma HIV by SCA, increasing IP-10 concentration (Odds Ratio= 1.33, 95% CI= 1.01-1.77) remained significantly associated with CSFHIV+. Conclusion: In this study of PWH on suppressive ART, there was a significant relationship between CSF CXCL10 and the presence of low-level HIV in CSF. CSF CXCL10 merits further study as a candidate marker of CNS persistence that may be useful in the evaluation of HIV eradication interventions. 441 USE OF D/C/F/TAF WITH NEUROLOGIC/PSYCHIATRIC COMORBIDITIES: AMBER SUBGROUP ANALYSIS Keith J. Dunn 1 , Richard Simonson 1 , Donghan Luo 1 , Jiyun Cai 2 , David Anderson 1 1 Janssen Scientific Affairs, LLC, Titusville, NJ, USA, 2 Janssen Research & Development, LLC, Spring House, PA, USA Background: Patients with human immunodeficiency virus (HIV)–1 and neurologic or psychiatric comorbidities (NPCs) may face challenges with HIV-1 care. Methods: The phase 3 AMBER trial (ClinicalTrials.gov: NCT02431247) enrolled treatment-naïve, HIV-1–infected adults who were randomized 1:1 to receive once-daily darunavir/cobicistat/emtricitabine/tenofovir alafenamide (D/C/F/ TAF) 800/150/200/10mg or control (D/C+F/tenofovir disoproxil fumarate). Here we report a subgroup analysis evaluating efficacy/safety in those with and

Poster Abstracts

442 CLINICAL FEATURES AND OUTCOMES OF PATIENTS STOPPING DTG FOR NEUROPSYCHIATRIC SYMPTOMS Andrea Calcagno 1 , Alberto Borghetti 2 , Maurizio Milesi 1 , Jessica Cusato 1 , Mariacristina Tettoni 1 , Antonio D'Avolio 1 , Giovanni Di Perri 1 , Simona Di Giambenedetto 2 , Stefano Bonora 1 1 University of Torino, Torino, Italy, 2 Catholic University of the Sacred Heart, Rome, Italy Background: Dolutegravir is a safe and effective integrase strand transfer inhibitor used in combination in people living with HIV (PLWH). In several, but not all, cohorts a high rate of discontinuation for neuropsychological side effects (NPS) has been reported: age, female gender, older age and abacavir co-administration have been suggested as potential risk factors while pharmacological and genetic features are still under study. Aim of this analysis is to describe the clinical features and outcomes of patients stopping DTG for NPS. Methods: In a cohort study involving two Italian outpatient clinics we enrolled patients starting DTG and recorded clinical, therapeutic, pharmacokinetic and pharmacogenetic features. The study was approved by the two Ethics Committees and patients signed a written informed consent. In this analysis we focused on patients stopping DTG for NPS in terms of pre-existing psychiatric comorbidities and outcomes after drug withdrawal. Symptoms were clinically assessed and no/partial/complete resolution was recorded. Results: 112 (out of 561) patients stopped DTG after a median follow up of 27 months (18-37): 66 for NPS. They were mostly sleep disorders (30.3%), headache (27.3%), anxiety (25.8%), depression (18.2%), psychosis (4.5%), vertigo (4.5%) and confusion (3%). Pre-existing psychiatric comorbidities were reported in 21 subjects (31.8%) mostly anxiety/depression in 24.2%: the latter was associated with the discontinuation for worsening depression (p=0.021, OR=4.4) but

CROI 2020 154