CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

acquisition risk, site, year of cohort entry, and time-updated ART, CD4 cell count, and history of other respective cancer. Results: In total, 27706 PWH were included. At clinic entry, median age was 33 years (IQR: 27-41), 78% were male, 37% had previous ART, and median CD4 cell count was 257 cells/ ml (IQR: 96- 457). Prevalent ADC was present in 795 PWH, 101 had NADC and 7 had both ADC and NADC. Risk of prevalent ADC and NADC varied by site. Older PWH, those with lower CD4 count (<250 cells/ml), and HIV acquisition risk factors other than heterosexual contact were more likely to present with prevalent ADC and NADC (p<0.05 for all). Those with male sex, earlier year of entry and prior ART were also more likely to present with ADC (p<0.05 for all). During a median of 3.9 (IQR: 1.0-9.4) years of follow up, there were 300 incident ADC and 264 NADC. Risk factors for incident ADC and NADC are shown in Figure 1. Conclusion: Prevalent and incident ADC were more frequent than NADC among PWH in CCASAnet. Older age and lower CD4 were associated with prevalent and incident NADC and prevalent ADC. Recent years of cohort entry and higher CD4 count were associated with decreased risk of prevalent and incident ADC. These findings continue to underscore the importance of early HIV diagnosis and treatment.

Kaposi sarcoma, colon, and Hodgkin lymphoma. We calculated age-adjusted incidence rates and standardized incidence ratios (SIR) by dividing observed cases by expected cases based on rates in the US population within region, age group, and year. We stratified these calculations by AIDS status. Results: Overall, excess risk of each cancer type varied significantly by race/ ethnicity. We examined 9 cancers (see figure), but these results will only discuss anal cancer because rates are particularly elevated among MSMWH. Non Hispanic (NH) white MSM had an overall age-adjusted incidence rate of 87.4 per 100,000 person-years (95% CI = 84.8, 89.9), significantly higher than NH Black MSM (68.8; 95%CI=66.1, 71.5), and Hispanic/Latino MSM (50.3; 95%CI=47.8, 52.7). When stratified by AIDS status, those with a prior AIDS diagnosis were 3 times higher for NH white and NH Black MSM, but 4.5 times higher for Hispanic/ Latino MSM, compared to those without. For SIRs, NH white MSMWH had 38.4 (95% CI = 36.3, 40.6) times the risk of anal cancer compared to all NH white men in the US population. NH Black and Hispanic/Latino MSM had 24.1 (95% CI = 22.3, 26.0) and 30.1 (95% CI = 28.0, 33.8) times the risk, respectively. These SIRs were greater among all races/ethnicities for MSM with a prior AIDS diagnosis compared to MSM without. Conclusion: Incidence rates and relative risks compared to the general population differed significantly by race/ethnicity. When stratified by AIDS diagnosis, SIRs and SIR ratios were significantly different between races/ ethnicities. Notably, risk of anal cancer was only 2.5 times higher for non Hispanic white MSM diagnosed with AIDS compared to HIV, but 3.5 times higher for Hispanic/Latino MSM. Hispanic/Latino MSM often had the highest SIRs suggesting a significantly higher excess risk among Hispanic/Latinos or potential under-ascertainment of cancers in the general population.

Poster Abstracts

757

Increased Cancer Risk With Low CD4 Counts Persists Despite Over 2 Years of Virological Suppression Jennifer F Hoy , for the RESPOND and D:A:D Study Groups Alfred Hospital, Melbourne, Australia Background: Antiretroviral therapy (ART) leads to a reduction in AIDS-related events and lower CD4 counts are associated with some non-AIDS events. The impact of long term virological suppression on non-AIDS related cancers (NADC) is unclear. We determined whether the most recent CD4 count was an independent predictor of incident cancer risk in people with HIV who had virologic suppression (VS) for at least 2 years. Methods: Individuals from the D:A:D and RESPOND cohort collaborations who achieved 2 years of VS on ART were included. Follow-up was from baseline (date of VS for 2 years) until the earliest of a first cancer event, confirmed virological failure (>200 copies/mL) or cessation of ART for >2 months, final follow-up, or administrative censoring date (D:A:D: 2/1/2016; RESPOND: 12/31/2021). Multivariable Poisson regression was used to assess associations between cancer incidence (total, AIDS-defining cancer (ADC), NADC, infection-related, smoking-related and BMI-related cancer) and time updated CD4 count (<350, 350-499, 500-749 and >750 cells/µL) stratified by pre-ART nadir CD4 count and adjusted for confounders determined a priori. Results: Overall, 51,622 people with VS were included (median [IQR] baseline age 44 years [37, 51], CD4 count 536 cells/µL [376, 729], nadir CD4 count 238 cells/µL [112, 386], 72% male, 36% current smokers). There were 2152 incident cancers during a total of 321,126 person-years of follow-up (PYFU), median 6 years [2.9, 9.5]) (incidence rate (IR)/1000 PYFU 6.70 [95% confidence interval 6.42, 6.99]). This included 276 ADC (0.86 [0.76, 0.97]/1000 PYFU), and 1876 NADC (5.84 [5.58, 6.11]). There were 721 infection-related (2.24 [2.08, 2.41]), 927 smoking-related (2.89 [2.7, 3.08]), and 491 BMI-related (1.53 [1.4, 1.67]) cancers, which were not mutually exclusive. After adjustment, there was a significant reduction in the adjusted IR ratio (aIRR) by higher time-updated CD4 count for all cancers (overall and by type) stratified by nadir CD4 count (Figure). No significant interaction between time-updated CD4 count and time-updated age, or calendar periods was present. Conclusion: Despite being virologically supressed on ART for >2 years, individuals with poorer immune recovery (CD4 <500 cells/µL) continue to experience a significantly higher incidence of all cancer groups. This underscores

756

Prevalent and Incident Cancers in a Cohort of People With HIV in Latin America Valeria I Fink 1 , Carina Cesar 1 , Shengxin Tu 2 , Bryan E. Shepherd 2 , Claudia P. Cortes 3 , Guilherme Calvet 4 , Juan Sierra-Madero 5 , Eduardo Gotuzzo 6 , Diana Varela 7 , Jessica L. Castilho 2 , for the Caribbean, Central and South America Network for HIV Epidemiology (CCASAnet) 1 Fundación Huésped, Buenos Aires, Argentina, 2 Vanderbilt University, Nashville, TN, USA, 3 University of Chile, Santiago, Chile, 4 Instituto Nacional de Infectologia Evandro Chagas, Rio de Janeiro, Brazil, 5 Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico, 6 Universidad Peruana Cayetano Heredia, Lima, Peru, 7 Instituto Hondureño de Seguridad Social, Tegucigalpa, Honduras Background: Cancer is a leading cause of morbidity and mortality among people with HIV (PWH). Earlier HIV diagnosis and timely access to antiretroviral treatment (ART) have led to longer life expectancies among PWH, resulting in an increase in aging-related comorbidities, including non-AIDS-defining cancers (NADC). While the epidemiology of NADC in PWH has been described in high resource settings, less is known in low- and middle-income settings. We evaluated the prevalence and incidence of NADC and AIDS-defining cancers (ADC) within the Caribbean, Central and South America network for HIV epidemiology (CCASAnet) and factors associated with these malignancies. Methods: PWH ≥18 years old enrolled at CCASAnet sites in Argentina, Brazil, Chile, Honduras, Mexico and Peru from 2000-2020 were included. Prevalent (diagnosed any time before or <90 days after clinic entry) and incident (diagnosed ≥90 days after clinic entry) NADC and ADC were collected. Clinical and demographic factors associated with prevalent and incident ADC and NADC were evaluated using multivariable logistic regression and multi-state Cox proportional hazards models, respectively. Models included age, sex, HIV

CROI 2024 225