CROI 2020 Abstract eBook

Abstract eBook

Oral Abstracts

71 INCREASED CANCER RISK WITH LOWER CD4/CD8 RATIO AMONG ADULTS WITH HIV IN NA-ACCORD Jessica L. Castilho 1 , Aihua Bian 1 , Cathy Jenkins 1 , Chad J. Achenbach 2 , W. C. Mathews 3 , Greer Burkholder 4 , M J. Gill 5 , Janet Tate 6 , Angel M. Mayor 7 , Mari M. Kitahata 8 , Michael J. Silverberg 9 , Richard D. Moore 10 , Keith M. Sigel 11 , Staci Sudenga 1 , for the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) of IeDEA 1 Vanderbilt University, Nashville, TN, USA, 2 Northwestern University, Chicago, IL, USA, 3 University of California San Diego, San Diego, CA, USA, 4 University of Alabama at Birmingham, Birmingham, AL, USA, 5 Southern Alberta Clinic, Calgary, AB, Canada, 6 VA Connecticut Healthcare System, West Haven, CT, USA, 7 Universidad Central del Caribe, Bayamon, Puerto Rico, 8 University of Washington, Seattle, WA, USA, 9 Kaiser Permanente Northern California, Oakland, CA, USA, 10 Johns Hopkins University, Baltimore, MD, USA, 11 Icahn School of Medicine at Mt Sinai, New York, NY, USA Background: Persons living with HIV (PLWH) are at increased risk for a number of cancers. Altered immunity is one proposed mechanism driving this excess risk of cancer. Low CD4/CD8 ratio in treated PLWH is associated with immune senescence, activation, and inflammation which may contribute to carcinogenesis. However, there is no clinical consensus of which ratio values that best predict cancer risk. This study examined whether low CD4/CD8 ratio predicted cancer risk (excluding non-melanoma skin cancer (NMSC)) in the North America AIDS Cohort Collaboration on Research and Design (NA-ACCORD). Methods: Adults without a history of any cancer (excluding NMSC) prior to cohort entry and with ≥1 CD4/CD8 value in 12 NA-ACCORD cohorts between 1998-2016 were included. Cancer outcomes were validated in each cohort. Risk of cancer and 6-month-lagged CD4/CD8 ratio were evaluated in multivariable, time-updated Cox proportional hazard models adjusting for age, sex, race, hepatitis C virus coinfection, lagged CD4 count (cells/mL) and lagged, log- transformed HIV RNA (copies/mL). Models for any and for specific cancers were censored at earliest occurrence of death, other cancer diagnoses, loss to follow- up (gap in care ≥ 1.5 years), end of cohort observation period or December 31, 2016. Observation time was censored during periods of missing laboratory data. Continuous variables, including CD4/CD8, were modeled using restricted cubic splines with 4 knots. Results: Among 75,161 PLWH, there were 5046 incident cancer diagnoses. Most frequent cancers were lung cancer (n=714), non-Hodgkin lymphoma (NHL, n=459), Kaposi sarcoma (KS, n=440), and anal cancer (n=375). Median age at cohort entry was 43 years, 90%were male, and 44%were white. The median CD4/CD8 ratio during the observation period was 0.49 (interquartile range: 0.27-0.79). Adjusted hazard ratios for CD4/CD8 ratio and any cancer and specific cancers are shown in the Figure. For any cancer and specific cancers, non-linear CD4/CD8 ratio was inversely associated with cancer risk in adjusted models (p< 0.001). Conclusion: Low CD4/CD8 ratio was consistently associated with increased cancer risk, independent of CD4 count and HIV RNA. Further research into the causes of CD8 cell inflation and persistent immunologic disturbance in PLWH is needed. CD4/CD8 ratio may serve as useful clinical biomarker for cancer risk in PLWH.

Oral Abstracts

72 WHOLE-BODY PET IMAGING OF THE HIV RESERVOIR USING RADIOLABELED VRC01

Timothy J. Henrich 1 , Denis Beckford-Vera 1 , Enrique Martinez-Ortiz 1 , Maya Aslam 1 , Cassandra Thanh 1 , Shreya Kumar 1 , I-Wei Katherine Wu 1 , Rebecca Hoh 1 , Robert Flavell 1 , Youngho Seo 1 , Marion Pardons 2 , Nicolas Chomont 2 , John R. Mascola 3 , Steven G. Deeks 1 , Henry VanBrocklin 1 1 University of California San Francisco, San Francisco, CA, USA, 2 Université de Montréal, Montreal, QC, Canada, 3 Vaccine Research Center, NIAID, Bethesda, MD, USA Background: HIV largely resides in anatomical regions that are inaccessible to routine sampling, and non-invasive methods to understand the tissue-wide burden of HIV are urgently needed. We report on the first-in-human PET- magnetic resonance (MR) imaging studies of HIV infection using a radiolabeled HIV Env-specific mAb, 89Zr-VRC01. Methods: PET-MR imaging using 89Zr-VRC01 (t 1/2 =73h) was performed on 5 viremic participants (plasma HIV-1 RNA ranging from 3,459 to 789,705 c/mL), 4 ART-suppressed participants (duration of ART ranging from 3.5 to 280 months), and 5 uninfected controls. PET-MR imaging was performed 2h, 6h, 24h, 72h (day 3), and, in a subgroup of 6 participants, 132h (day 6) following a single 1 mCi injection of 89Zr-VRC01. Radiotracer maximum and mean standard uptake values (SUVmax, SUVmean) adjusted for blood pool background signal were quantified for various lymphoid and other anatomical regions of interest. Results: Adjusted 89Zr-VRC01 SUVs were significantly higher in inguinal and axillary lymph nodes, nasal-associated lymphoid tissue (NALT), and bone marrow in viremic participants compared with uninfected controls (all P<0.05). SUVmax/mean (NALT, bone marrow) and SUVmean (inguinal lymph node) were significantly higher in ART-suppressed individuals compared with uninfected controls, and generally lower than in viremic participants. The greatest differences between SUVs in HIV-infected and control participants were observed 72h after tracer injection, although differences in tracer uptake in inguinal lymph node tissue were observed up to 6 days following tracer injection (Figure). 89Zr-VRC01 inguinal lymph node SUVmax in viremic and ART-suppressed participants positively correlated with the frequency of p24 expressing cells measured by flow cytometry in fine needle aspirates (p=0.017). 89Zr-VRC01 tracer uptake in lymphoid tissues was lower in participants who were on suppressive ART for longer periods of time. Conclusion: HIV envelope-specific PET imaging was able to detect differences between HIV-infected individuals, including those on suppressive ART, and uninfected participants. Importantly, PET tracer uptake correlated with measures of HIV protein expression in tissue. These data suggest that PET imaging of HIV-infected cells has the potential to localize and quantify multiple anatomical HIV reservoirs in a wide range of HIV persistence and curative studies.

24

CROI 2020