CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

391 REBOUND OF HIV-1 IN CEREBROSPINAL FLUID AFTER TREATMENT INTERRUPTION Richard W. Price 1 , Maria Bednar 2 , Laura P. Kincer 2 , Rebecca Hoh 1 , Magnus Gisslén 3 , Sarah B. Joseph 2 , Serena S. Spudich 4 , Steven G. Deeks 1 , Ronald Swanstrom 2 1 University of California San Francisco, San Francisco, CA, USA, 2 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 3 University of Gothenburg, Gothenburg, Sweden, 4 Yale University, New Haven, CT, USA Background: HIV-1 persists in the body despite years of suppressive antiviral therapy. One well characterized viral reservoir is in resting CD4+ T cells. Under certain circumstances an independently replicating viral population can be detected in the CNS as indicated by differing viral sequences in blood and CSF. We examined rebound virus in the blood and CSF in 9 participants who chose to interrupt therapy. Methods: Nine men interrupted therapy (TI), most in the context of early systemic failure (Price and Deeks J. Neurovirol. 2004 PMID: 14982739). Multiple blood and CSF samples were collected before and after TI over a period of up to 41 weeks, and clinical and laboratory data were recorded. Viral RNA was extracted from the plasma and CSF at each time point and viral populations examined using deep sequencing, with the sampling depth defined and sequencing errors corrected using Primer ID. Results: In 7/8 participants with clinical laboratory data available there was a significant increase in WBCs in the CSF (i.e. pleocytosis) typically between 1-2 months after TI. The peak in WBCs in the CSF often corresponded with a peak in CSF viral load (VL), reaching viral RNA levels equivalent to those in the blood. At all analyzed time points for the 9 participants before and during the increase in VL in the CSF, the viral population in the CSF was well mixed with the viral population in the blood. Two frequently seen features of the CSF viral population were transient expansion of a single genotype (clonal amplification) of an R5 T cell-tropic virus, and differential penetration of one viral population present in the blood into the CSF relative to a second blood viral population. There was no evidence for a distinct viral population emerging from the CNS based on sampling of the CSF in these participants. Conclusion: In this study of 9 participants we did not observe any evidence of a CNS-specific lineage in the CSF after TI in spite of extensively sampling the viral populations using deep sequencing. The large majority of the virus observed in the CSF appears to be the result of infected CD4+ T cells trafficking into the CNS from the blood rather than from ‘release’ from a CNS reservoir after TI. It is not clear why there is a spike in pleocytosis during virus rebound. These results do not preclude a CNS reservoir but rather point to the difficulty of observing such a reservoir in the presence of actively circulating CD4+ T cells transporting virus from the blood into other compartments. 392 COMPARTMENTALIZED HIV REBOUND IN THE MALE GENITAL TRACT AFTER ART INTERRUPTION Sara Gianella 1 , Antoine Chaillon 1 , Tae-Wook Chun 2 , Caroline Ignacio 1 , Colin Kovacs 3 , Erika Benko 3 , Sanja Huibner 3 , Rupert Kaul 3 1 University of California San Diego, La Jolla, CA, USA, 2 National Institute of Allergy and Infectious Diseases, Bethesda, MD, USA, 3 Maple Leaf Medical Clinic, Toronto, ON, Canada Background: If strategies currently in development succeed in eradicating HIV reservoirs in peripheral blood and lymphoid tissues, residual sources of virus may remain in anatomic compartments, including the genital tract. To design effective eradication strategies, it is crucial to determine to what extent compartmentalized HIV reservoirs contribute to viral rebound after antiretroviral therapy (ART) interruption. Methods: Paired blood and semen samples were collected from 12 individuals enrolled in a randomized, double-blind, placebo-controlled clinical trial of HIV-MAG DNA vaccine prime, rVSVN4CT1gag booster vaccine in people living with HIV (PLWH) who began ART during acute or early infection (NCT01859325). At the first available time-points following viral rebound, we sequenced HIV-1 env (C2-V3), gag (p24), and pol (reverse transcriptase) regions amplified from cell-free HIV RNA in blood and seminal plasma using the MiSeq Illumina platform. Comprehensive sequence and phylogenetic analyses were performed to look for evidence of viral compartmentalization (Fst test), diversity (Shannon entropy measures) and unique viral subpopulations in seminal plasma as compared with blood plasma.

HIV RNA rebound in semen occurred significantly later (median of 66 vs 42 days post ART interruption) and reached lower levels (164 vs 16,224 copies/ ml). Paired sequence data were available for 5 participants. All presented compartmentalized viral rebound between blood and semen (Fst, p≤0.05 for all genes). Phylogenetic analysis confirmed the presence of compartment-specific monophyletic HIV RNA populations in at least one HIV region in 3 out of the 5 participants in longitudinal time-points, suggesting that rebound originated within genital compartment rather than migrating from blood (Figure 1 panel A). Interestingly, despite early ART start, genetic diversity after adjusting for variant frequency was higher in semen compared to blood in all three coding regions (Significant for gag and pol (<0.01) but not in env (p=0.06)), Figure 1 panel B. Conclusion: HIV reservoirs in the genital compartment might contribute to viral rebound in PLWH interrupting ART. Higher diversity in the genital compartment illustrates viral compartmentalization and distinct evolutionary dynamics. Reservoirs in all anatomic compartments need to be actively targeted to achieve a complete functional cure.

Poster Abstracts

393 SEEKING SUPPRESSION IN HAVARTI: VIREMIA & T CELLS AFTER VEDOLIZUMAB & ATI IN HIV/ART

Michaeline McGuinty 1 , Jonathan Angel 1 , Ashok Kumar 2 , Richmond Sy 1 , Sanjay Murthy 1 , Donald Kilby 1 , Nancy Tremblay 3 , Elizabeth Lavoie 4 , Stephanie Burke Schinkel 3 , Siddappa N. Byrareddy 5 , D. William Cameron 1 , for the HAVARTI CTNPT031 Study Team 1 University of Ottawa, Ottawa, ON, Canada, 2 Children’s Hospital of Eastern Ontario, Ottawa, ON, Canada, 3 Ottawa Hospital Research Institute, Ottawa, OT, Canada, 4 University of Ottawa, Ottawa, OT, Canada, 5 University of Nebraska Medical Center, Omaha, NE, USA Background: HAVARTI is a dose-ranging trial of vedolizumab and analytical treatment interruption (ATI) in HIV/ART. Methods: Eight healthy HIV+ adults on ART for 2-10 years had vedolizumab given 3 times in 6 weeks before, and 4 times in 14 weeks after ATI, at 300mg doses in 4 (Group 1) and 150mg in 4 (Group 2). Monthly follow-up for adverse events (AE), plasma viremia (pVL) and T cell count outcomes informed clinical judgement for ART retreatment. Results: Groups had similar mean age, nadir CD4, pre-ART pVL, ART duration & baseline CD4 count. No serious clinical or severe laboratory AE occurred. One case had non-sustained pVL suppression <40 copies/mL in 2 sequential measures. CD4 T-cell count response varied, but none had sustained CD4 <350 cells/µL. Percent a4b7+ CD4 T cells in rectal mucosa decreased in Groups 1 and 2 respectively from 46.90 ±23.30 and 30.63 ±9.86 before, to 2.77 ±1.73 and 3.05 ±2.47 after vedolizumab. Group 1 pVL rebounded in 3 of 4 at 2 weeks, and all 4 at 6 weeks into ATI. pVL doubling time (T2) from ATI week 2 to 6 was 7.67 ±4.41 days, to a peak pVL level below pre-ART in each by mean 1.21 ±0.56 log10 copies/mL, sustained on average to 22 weeks, before a consistently rising pVL trajectory after 26 weeks, 12 weeks after last vedolizumab dosage. Group 2 pVL rebounded in 1 of 4 at 2 weeks, and all 4 at 6 weeks into ATI. T2 was 2.58 ±0.79 days, to week 6 peak pVL above pre-ART in 3 of 4 by mean 0.26 ±1.37, falling belowmean pre-ART pVL to 14 weeks, with consistent rising trajectory onset after 18 weeks, 4 weeks after last vedolizumab dosage. Conclusion: Viremia rebound appears attenuated in group 1 compared with group 2, supported by individual consistency of much slower T2 (p=0.057), and much lower pVL peak compared to pre-ART (by 1.47 log10), sustained 2 months longer after last vedolizumab dose. This difference is corroborated by

Results: Vaccine had no effect on kinetics and magnitude of HIV RNA rebound in blood plasma (Sneller et al, STM 2017). Compared to blood,

CROI 2019 141