CROI 2020 Abstract eBook

Abstract eBook

Oral Abstracts

Oral Abstracts

69

CONJUNCTIVAL CANCER IN PEOPLE LIVING WITH HIV: THE SAM STUDY Tafadzwa G. Dhokotera 1 , Mazvita Sengayi 2 , Lina Bartels 1 , Frédérique Chammartin 1 , Serra L. Asangbeh 1 , Victor Olago 2 , Elvira Singh 2 , Matthias Egger 1 , Julia Bohlius 1 1 Institute of Social and Preventive Medicine, Bern, Switzerland, 2 National Health Laboratory Service, Johannesburg, South Africa Background: Conjunctival cancer has been associated with HIV in sub-Saharan Africa but, the evidence on its epidemiology is scarce. According to the 2014 National Cancer Registry (NCR) report, the incidence of eye cancer was 1.13 and 1.25 per 100 000 of the population in males and females respectively. We aimed to determine the incidence of conjunctival cancer amongst people living with HIV (PLWHIV) and the associated risk factors. Methods: The South African HIV Cancer Match (SAM) study used privacy preserving record linkage to create a large cohort of cancer in PLWHIV from national laboratory and NCR data 2004-2014. We used the ICD-O-3 coding to identify conjunctival cancers. We calculated crude incidence rates of conjunctival cancer and used Cox regression to obtain hazard ratios (HR) of CD4 cell count, age, sex and calendar period, stratified by province. Results: Over 12 547 950 person-years of follow-up, 1 359 incident conjunctival cancer cases were diagnosed in the SA cohort of 4 766 614 PLWHIV. Approximately 94% (n=1 274) of conjunctival cancers were squamous cell carcinomas. The median age at entry into the cohort was 33 years (Interquartile Range [IQR]: 26-41 years) and 38 years (IQR: 33-44 years) at cancer diagnosis. The median CD4 cell count at baseline was 294 cells/µl (IQR: 159-467 cells/µl). There was an upward trend in CD4 cell counts across the years from a median of 240 cells/µl in 2004 to 340 cells/µl in 2014. The crude conjunctival cancer IR was 11.0 per 100 000 person-years (95% Confidence Interval [CI]: 10.2-11.4). Being male, lower CD4 cell count, earlier calendar period and older age were all associated with higher rates of conjunctival cancer (Table 1). Conclusion: To our knowledge, this is the largest epidemiological study of conjunctival cancer in PLWHIV ever done. Our results indicate that immunodeficiency as indexed by lower CD4 counts, immune senescence and prolonged UV light exposure (both indexed by age) are strongly associated with conjunctival cancer risk. The decrease in incidence in more recent calendar periods might reflect increased ART coverage across time and initiation of ART at higher CD4 cell counts. Our analysis suggests that effective HIV control is essential for the prevention of conjunctival cancers. We recommend symptom screening and communication of conjunctiva cancer risk to PLWHIV as well as their clinicians

70 CLEARANCE OF HPV ANAL HIGH-GRADE INTRAEPITHELIAL LESIONS WITH LOW-DOSE POMALIDOMIDE Mark Polizzotto 1 , David Van Bockel 1 , Carmella Law 2 , Jennifer Roberts 3 , Susanne Just 1 , Griselda Buckland 1 , Simon Comben 2 , Mary Poynten 1 , Richard Hillman 2 , Richard Gilson 4 , Sarah Pett 5 , Anthony Kelleher 1 , Sean Emery 1 , for the SPACE Study Group 1 Kirby Institute, Sydney, NSW, Australia, 2 St. Vincent's Hospital, Sydney, NSW, Australia, 3 Douglass Hanly Moir Pathology, Macquarie Park, Australia, 4 Mortimer Market Centre, London, UK, 5 MRC Clinical Trials Unit at UCL, London, UK Background: People with HIV (PLWH) have an increased risk of anal cancer. This is preceded by high-grade squamous intraepithelial lesions (HSIL). Spontaneous clearance of HSIL is associated with systemic T-cell response to human papillomavirus (HPV) oncogene E6. Pomalidomide may enhance immune responses to HPV and be therapeutic in HSIL. Methods: This phase II single centre study (NCT3113942) recruited participants with persistent (>12 months) biopsy-proven anal HSIL. Therapy was oral pomalidomide 2mg 21/28 days for 6 months. PLWH were eligible if on ART with viral load (VL)<200 copies/mL and CD4 count >200 cells/µL. Primary outcome was response at end therapy (CR defined as histological clearance; PR as ≥50% reduction in area); secondary included response after a further 6 months without therapy. Immune activation was assessed by flow cytometry. Antigen- specific CD4+ T-cell responses to HPV16 E6 and E7 were assessed by OX40 assay. Results: 26 participants enrolled, 24 evaluable for response. All male; median age 54 (range 41-74). 10 (38%) PLWH: median CD4 700 cells/uL (320-1070), all HIV VL <20 copies/mL. All AIN3 HSIL, median duration HSIL 37 months (15-86), median octants 2 (0.5-5); HPV16 in 55%; multiple high risk HPV types in 50%. Overall response (CR+PR) was 52% (CI: 31-73) at end therapy, increasing to 63% (95% CI 40-81) after 6 months observation. Responses were comparable in PLWH. Adverse events (AEs) were mild and self-limited, including cytopenias, constipation, rash, with no idiosyncratic AEs in PLWH. HIV suppression was maintained. Over 137 cycles (c), attributable grade (g) 3/4 events were g3 neutropenia (4c) and g3 angina (1c). Systemic CD4+ T-cell responses to HPV E6 but not E7 increased during therapy, peaking day 14: baseline 0.06%, IQR 0.01 – 0.12%), median increase day 14 0.13% (IQR: 0.02 – 0.26%), p=0.001. CD4+ and CD8+ cell activation (CD38, HLA DR, CD38+HLA-DR) increased during therapy. Conclusion: Low dose pomalidomide was well tolerated and induced durable continuing clearance of anal HSIL of multiple genotypes in even in chronic extensive disease irrespective of HIV status. Induction of HPV-specific CD4+ responses and immune activation support an immunological mechanism of action.

23

CROI 2020