CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

397 NEXT-GEN VIRAL OUTGROWTH ASSAYS AS PROXIES FOR CLASSIC QVOA TO MEASURE HIV RESERVOIR Mars Stone 1 , Daniel S. Rosenbloom 2 , Peter Bacchetti 3 , Xutao Deng 1 , Melanie Dimapasoc 1 , Sheila M. Keating 1 , Douglas D. Richman 4 , John W. Mellors 5 , Steven G. Deeks 3 , Janet Siliciano 6 , Nicolas Chomont 7 , Roger Ptak 8 , Michael P. Busch 1 1 Blood Systems Research Institute, San Francisco, CA, USA, 2 Columbia University, New York, NY, USA, 3 University of California San Francisco, San Francisco, CA, USA, 4 University of California San Diego, San Diego, CA, USA, 5 University of Pittsburgh, Pittsburgh, PA, USA, 6 Johns Hopkins University, Baltimore, MD, USA, 7 Université de Montréal, Montreal, QC, Canada, 8 Southern Research Institute, Frederick, MD, USA Background: Evaluations of HIV curative interventions require efficient assays that reliably quantify the latent reservoir (LR) of replication-competent HIV-1. The “classic” quantitative viral outgrowth assay (QVOA) monitors limiting-dilution co-culture of resting CD4+ T cells and stimulated donor PBMCs for 3 weeks by p24 Ag ELISA. While regarded as a “gold standard,” QVOA is prohibitively resource- and labor-intensive for widespread use. We compared five induced outgrowth assays employing PCR or ultrasensitive p24 readout of short-term resting CD4+ T cell cultures (“next-gen assays”) to assess their suitability as scalable proxies for QVOA. Methods: Next-gen viral outgrowth assays were performed and compared to classic QVOA results on samples derived from single leukapheresis collections from 5 ART-suppressed HIV+ participants and one HIV- control, with 3 aliquots of cryopreserved cells and one fresh sample tested at each lab. A certain amount of variation between split samples is inevitable due to Poisson sampling variation, so we used Markov chain Monte Carlo methods to estimate extra- Poisson variation at the aliquot, batch, and lab levels. Models also estimated the effect of using frozen versus fresh samples. Results: Next-gen assays had similar estimates of variation to QVOA, with random variation at aliquot, batch, and lab levels having overlapping credible intervals. Overall, RNA-based assays reported higher IUPM than classic p24-based assays. Assaying split samples in the same batch had 2.5-fold extra- Poisson variation (95% CI 2.1 – 3.5) for next-gen assays. Assay performance by two separate labs increased total extra-Poisson variation to 3.4-fold (95% CI 2.6 – 5.4). Frozen storage did not substantially alter IUPM (–18%, 95% CI –52% – +39%). Within this cohort, two of the next-gen assays using short- term stimulation and PCR or ultrasensitive p24 readout had moderately high correlation with all four classic assays (R2 > 0.5 for all four comparisons, R2 > 0.8 for at least one comparison). Conclusion: The data offer cautious support for use of next-gen assays as proxies for more laborious outgrowth-based co-culture QVOA, while providing greater sensitivities and dynamic ranges. Measurement of LR in eradication strategies would benefit from development of assays that are high-throughput and scalable.

Background: Latently infected resting CD4+ T cells (rCD4+) contribute to HIV persistence in individuals receiving long term suppressive ART. The low frequency of latently infected cells among the rCD4+ population presents challenges in quantifying the latent reservoir, which is critical to evaluate HIV eradication strategies. Several assays have been developed to quantify HIV persistence, however all present limitations; PCR-based assays overestimate the HIV reservoir, while QVOA has been shown to underestimate it. To support the development of effective HIV eradication therapeutics, there is an urgent need to develop a robust and precise assay that more accurately quantifies the frequency of infected cells carrying replication-competent HIV. To address this need, we modified QVOA to include a digital p24 endpoint that not only allowed for a shorter assay duration, but also increased sensitivity. Methods: rCD4+ cells were enriched from cryopreserved PBMC from 5 ART-suppressed, HIV+ individuals provided by the Reservoir Assay Validation and Evaluation Network (RAVEN). Standard QVOA was performed, with culture supernatant collection on days 8, 12, and 20. IUPM were calculated by assessing the frequency of p24 positive culture supernatants at each time point as determined by Quanterix Simoa digital p24 system or ELISA. Results: The digital p24 endpoint readily detected HIV+ samples that were below the limit of detection by ELISA. Analysis revealed patterns of viral growth kinetics at p24 levels not detected by ELISA, supporting identification of replicating virus present in QVOA that are missed by standard approaches. While ELISA efficiently detected robustly replicating virus with greatest sensitivity at day 20, digital p24 enabled enhanced detection as early as day 8 with greatest sensitivity at day 12 (Table). Overall, use of digital p24 coupled with QVOA reduced assay duration and increased the estimated size of the latent reservoir (~7-fold) compared to day 20 IUPMmeasured by ELISA. Conclusion: Digital p24 permits early and more sensitive detection of HIV outgrowth in QVOA enabling the duration of the assay to be reduced by 8-12 days, coupled with an approximate one log increase in the calculated IUPM. These data support that digital p24 offers an approach to improve the sensitivity of QVOA in quantifying the HIV reservoir, a critical component for effective development of HIV eradication strategies. 399 QUANTIFICATION OF REPLICATION COMPETENT LATENT HIV-1 IN GALT AND SEMEN Anwesha Sanyal 1 , Ian McGowan 1 , Ma Somsouk 2 , Charles Rinaldo 1 , Nicolas Sluis- Cremer 1 , Phalguni Gupta 1 1 University of Pittsburgh, Pittsburgh, PA, USA, 2 University of California San Francisco, San Francisco, CA, USA Background: The latent replication-competent HIV-1 reservoir represents a major obstacle to a cure. This reservoir has been well-characterized in resting CD4+ (rCD4+) T cells in blood. However, few studies have quantified the frequency of replication-competent latent HIV-1 in tissue, primarily due to the limitation of the sensitivity of current assays. We recently developed a reporter cell-based assay to quantify inducible, replication-competent latent HIV-1 in the blood (Sanyal et al., Nat. Med. 2017. 23:885) which is sensitive, requires only a small blood volume, and is less labor intensive than other available technologies. The goal of this study was to adapt this assay (termed TZA) to quantify replication competent HIV-1 in GALT and semen from virally suppressed individuals. Methods: The TZA was adapted to quantify inducible replication-competent HIV-1 in tissue samples. Using this assay, we quantified inducible latent virus in total CD4+ T cells from the blood and semen of 2 infected individuals on ART, and from the blood and GALT of 6 infected individuals on ART. Results: We routinely isolated 2-3×106 CD4+ T cells of high purity from rectal tissue using the gentleMACSTM Octo Dissociator (Miltenyi). We found in 6 patients that there was more replication competent HIV-1 in the GALT than

Poster Abstracts

Estimated extra-Poisson variation (fold), Posterior Median (95% credible interval) Classic QVOA (4 Assays) Next-gen QVOA (5 Assays)

Aliquot Level

1.5 (1.1-2.1) 1.8 (1.0-2.4)

2.4 (2.0-3.1) 1.1 (1.0-1.8)

Batch Level (alone)

Aliq + Batch

2.0 (1.6-2.7)

2.5 (2.1-3.4)

Assay-Level (alone) Aliq + Batch + Assay

1.5 (1.0-2.5)

2.5 (1.0-4.4)

2.3 (1.8-3.5)

3.7 (2.9-5.7)

Frozen Effect

+4% (-44%, +97%) -18% (-50%, +39%)

398 QVOA COUPLED WITH DIGITAL P24 ANALYSIS ENHANCES HIV RESERVOIR QUANTIFICATION Elizabeth R. Wonderlich 1 , Yury V. Kuzmichev 1 , Carol Lackman-Smith 1 , Marie K. Mankowski 1 , Christine N. Raney 1 , Hayley P. Madeira 1 , Jiayi Wei 1 , Rebecca M. Bernbaum 1 , Krupa Subramanian 1 , Rachel L. Jarrett 1 , Ellen R. Nordgren 1 , Mars Stone 2 , Michael P. Busch 2 , Roger Ptak 1 , Deanna Kulpa 3 1 Southern Research Institute, Frederick, MD, USA, 2 Blood Systems Research Institute, San Francisco, CA, USA, 3 Emory University, Atlanta, GA, USA

CROI 2018 139