CROI 2015 Program and Abstracts

Abstract Listing

Oral Abstracts

92 Majority of XDR TB Cases Are Due to Transmission in a High-HIV-Prevalence Setting N. Sarita Shah 1 ; James C. Brust 2 ; Barun Mathema 3 ;Thuli Mthiyane 4 ; Nazir Ismail 5 ; Pravi Moodley 6 ; Koleka Mlisana 8 ; Salim Allana 7 ; Jonathan Smith 7 ; Neel R. Gandhi 7

1 US Centers for Disease Control and Prevention, Atlanta, GA, US; 2 Montefiore Medical Center & Albert Einstein College of Medicine, Bronx, NY, US; 3 Columbia University Mailman School of Public Health, New York, NY, US; 4 University of KwaZulu-Natal, Durban, South Africa; 5 National Institute for Communicable Diseases, National Health Laboratory Service, Johannesburg, South Africa; 6 University of KwaZulu-Natal & National Health Laboratory System, Durban, South Africa; 7 Emory University Rollins School of Public Health, Atlanta, GA, US; 8 University of KwaZulu-Natal & National Health Laboratory Service, Durban, South Africa Background: Drug-resistant TB threatens gains made in the control of TB and HIV worldwide. In South Africa, there has been a ten-fold increase in the number of MDR and XDR TB cases over the past decade. Factors driving this rapid rise in caseload have not been fully elucidated. We sought to determine the proportion of XDR TB cases resulting from primary transmission versus inadequate treatment of MDR TB (i.e., acquired resistance). Methods: Cross-sectional study of 400 culture-confirmed patients with XDR TB diagnosed during 2011–2014 from KwaZulu-Natal province, South Africa. Participants were interviewed about demographics, previous TB and HIV history, healthcare utilization, and potential TB exposures. Medical records were reviewed for prior treatment with first- and second-line TB medications and previous hospitalizations. All M.tuberculosis isolates underwent IS 6110 RFLP genotyping and targeted sequencing of nine resistance-conferring genes. Isolates from genotypic clusters with more than 20 participants also underwent whole genome sequencing. Results: To date, 377 patients with XDR TB have been enrolled. The median age was 33 (IQR 29-43) and 57%were female. 78%were HIV-infected, with a median CD4 count of 275 cells/mm 3 (IQR 149 – 433); 90%were on antiretroviral therapy at the time of enrollment and 79% had a viral load <400 copies/ml. Participants were identified from all 11 districts in the province. Only 33% of participants had been previously treated for MDR TB, and an additional 7% had received fluoroquinolones or injectable TB medications for non-TB illnesses. Based on genotypic analysis, 87% of participants had an isolate that belonged to one of 16 clusters; only 13% had isolates that were unique. One large cluster (LAM4/KZN) accounted for 47% of participants, while the remaining cluster sizes ranged from 2–16 participants. Whole genome sequencing confirmed the highly clonal nature of the large LAM4/KZN cluster. Conclusions: In this high HIV prevalence setting, the majority of XDR TB cases occurred due to transmission of drug-resistant TB strains, rather than prior inadequate treatment of MDR TB. Analysis of epidemiologic and geospatial data also collected in this study will provide valuable insights into identifying potential locations of transmission and opportunities to intervene. TB control efforts must focus on curbing transmission through improved TB case-finding, infection control and enhanced treatment programs. 93 Linkage to HIV/TB Care in South Africa: A Randomized Trial of Health Navigators Ingrid V. Bassett 2 ; Sharon M. Coleman 1 ; Janet Giddy 4 ; Laura M. Bogart 3 ; Christine E. Chaisson 1 ; Douglas Ross 5 ;Tessa Govender 4 ; Kenneth A. Freedberg 2 ; Rochelle P.Walensky 2 ; Elena Losina 2 1 Boston University School of Public Health, Boston, MA, US; 2 Massachusetts General Hospital, Harvard Medical School, Boston, MA, US; 3 Boston Children’s Hospital, Harvard Medical School, Boston, MA, US; 4 McCord Hospital, Durban, South Africa; 5 St. Mary’s Hospital, Mariannhill, South Africa Background: Despite increases in HIV testing, only a fraction of those newly diagnosed with HIV enter care and initiate ART promptly in South Africa. Our objective was to establish the efficacy of health system navigators for improving linkage to HIV care and TB treatment completion among newly diagnosed HIV-infected outpatients in Durban, South Africa. Methods: We conducted a randomized controlled trial (Sizanani Trial, NCT01188941) among adult ( ≥ 18y) patients enrolled at 4 sites (2 hospital outpatient departments and 2 primary health clinics) prior to HIV diagnosis. HIV-infected participants underwent TB screening with sputummicroscopy and culture and were randomized to receive a health system navigator intervention or usual care. Participants in the usual care arm received care according to each clinic’s practice. Participants in the navigator arm had a baseline interview using a strengths-based case management approach to assist in identifying barriers to entering care and devising solutions, and then received scheduled phone calls and text messages over 4 months. The primary outcome was 3 months on ART and alive for those ART-eligible at baseline or completion of 6 months of TB treatment. We assessed outcomes via medical records at study sites and the South African death registry 9 months after enrollment. Results: Of 6,536 subjects screened, 4,903 were eligible and enrolled; 49%were female and mean age was 34 years (SD 13). 1,899 (39%) subjects were HIV-infected, of whom 1,146 (60%) were ART eligible; 369 (19%) were co-infected with TB. In the usual care arm, 197 (21% of HIV-infected and 37% of ART eligible) reached the primary composite study outcome, compared to 212 (22% and 34%, respectively) in the navigator arm (p=0.60). 250/1,899 (13%) HIV-infected subjects died during the study period; there was no difference in death rates between study arms. Conclusions: Less than 40% of ART-eligible individuals newly diagnosed with HIV in Durban had evidence of taking ART for ≥ 3 mo or completing TB treatment at the study sites during the nine months after diagnosis. This RCT did not provide evidence that a health navigator-based intervention improves ART initiation or TB treatment completion. Low rates of engagement and retention in care, coupled with the lack of efficacy of the navigator intervention, highlight the urgency of identifying more effective strategies for improving HIV and TB care outcomes. 94 Is It Safe to Stop Cotrimoxazole in Adults on ART: COSTOP, a Noninferiority RCT Paula Munderi 1 ; Jonathan B. Levin 1 ; Zacchaeus Anywaine 1 ; Ronnie Kasirye 1 ; Anatoli Kamali 1 ; Andrew J. Nunn 2 ; Heiner Grosskurth 3 On behalf of COSTOP ResearchTeam 1 MRC/UVRI Uganda Research Unit on AIDS, Entebbe, Uganda; 2 University College London, London, United Kingdom; 3 London School of Hygiene and Tropical Medicine, London, United Kingdom Background: In Uganda, cotrimoxazole (CTX) prophylaxis is recommended as part of a package of comprehensive care. The benefits of continuing CTX in African patients who have regained immune competence through ART are not known particularly in the light of potential disadvantages such as co-toxicity with ART and increased pill burden leading to possible diminished adherence to HIV treatment The objective of COSTOP was to assess the risks and benefits of discontinuing CTX in patients achieving sustained immune restoration. Methods: From January 2011 HIV-infected patients aged >18 years, stable on ART with confirmed CD4 restoration to 250 cells/mm3 and above who consented to join the double-blind trial were randomised 1:1 to one oral tablet of 960 mg of CTX daily or matching placebo. Co-primary outcome measures were (1) time to first CTX-preventable event excluding malaria, (2) time to the occurrence of the first grade 3 or 4 haematological adverse event (AE). Patients attended study clinics monthly during the first 3 months post- randomisation and 3-monthly thereafter for a range of 1 – 3 years. An endpoint review committee adjudicated the efficacy endpoints. The analysis of the efficacy assesses the non-inferiority of the outcome in the placebo arm compared to the control; non-inferiority required the upper limit of the 90% confidence interval to be less than a 25% increased risk of an endpoint event on the placebo arm. The primary safety analysis assesses the reduction in haematological adverse events in the placebo arm. The study had a minimum of 80% power based on assumptions concerning endpoints in the CTX arm. Results: 2180 patients were enrolled from two sites in SW Uganda; 74% female, median age 41, CD4 count 518, months on ART 48. 93.3% patients completed at least a year in the study. In the per protocol analysis a total of 124 (54 CTX, 70 placebo) first CTX preventable adjudicated events occurred, hazard ratio adjusted for site and CD4 stratum (aHR) 1.35 (90% CI 1.00,1.81), a difference of 0.9/100 person years. These findings were confirmed in ITT and sensitivity analyses. The incidence of first grade 3 or 4 haematological adverse events was reduced in the placebo arm, aHR 0.70, 95% CI 0.59, 0.82.

Oral Abstracts

131

CROI 2015