CROI 2020 Abstract eBook

Abstract eBook

Oral Abstracts

Background: Fewmultinational, multisite studies have evaluated predictors of TB infection in MDR-TB household contacts (HHC). Methods: From 10/2015-4/2016, ACTG and IMPAACT networks conducted A5300/I2003, a cross-sectional observational study of adults with pulmonary MDR-TB and their HHCs, in high TB-burden countries in preparation for a randomized trial. Among HHCs≥15 years of age without TB disease, index case (IC), household (HH), HHCs, and TB exposure characteristics were evaluated for association with TB infection based on interferon gamma release assay (IGRA) status (QuantiFERON Gold/Gold-in-Tube). HHCs<15 years have previously been reported on. Logistic regression using generalized estimating equations was used for testing. Results: 278 ICs enrolled at 16 sites in 8 countries had 712 HHCs≥15 years of age. 36% of ICs were HIV-infected and 10% had unknown HIV status. 59 (8%) HHCs age≥15 years were HIV-infected and 436 (64%) female (16 pregnant). 686 had determinate, 4 indeterminate, and 22 no IGRA results. Factors independently predictive (p≤0.05) in multivariable models are shown in the Table. Of 686 HHCs with determinate results, 471 (69%, 95% confidence interval: 65-73%) were positive; prevalence varied with age: 59% in 15-<25, 76% in 25-<50, and 68% in ≥50 years (p<0.001). Cavitations on CXR, smear status, and duration of IC TB treatment were not associated with HHC TB infection prevalence. TB infection prevalence increased when a HHC had self-reported or a medical history of COPD/asthma (83% vs 69%, p=0.039), spent more nights/week with the IC (61%, 68%, 70% for 0-2, 3-5, 6-7 nights, respectively, p=0.05) but not by sleeping proximity. Compared to HHCs never incarcerated and not substance or alcohol users (66%), HHCs previously incarcerated had the highest prevalence of TB infection (95%); HHCs never incarcerated using substances or alcohol were also more likely to have TB infection (84%)(p<0.001). Smoking in the household (77% vs 64%, p=0.02) and lower quality exterior wall materials (see definition in Table) were associated with increased TBI prevalence (77% vs 67%, p=0.009). Conclusion: Over 2/3rd of HHCs age≥15 in HHs of adult MDR-TB patients had evidence of TB infection, confirming the importance of household contact investigation. HHs with poorer quality homes and HHCs highly exposed to IC, ever incarcerated or currently using substances or alcohol, or with COPD/asthma require particular attention to identify all TB infected HHCs.

Oral Abstracts

138 ISONIAZID TO PREVENT MTB INFECTION IN HIV-EXPOSED UNINFECTED INFANTS Sylvia LaCourse 1 , Barbra A.Richardson 1 , Lisa M. Cranmer 2 , Elizabeth Maleche Obimbo 3 , Daniel Matemo 4 , Alex J. Warr 5 , Jaclyn Escudero 1 , John Kinuthia 4 , Thomas Hawn 1 , Grace John-Stewart 1 1 University of Washington, Seattle, WA, USA, 2 Emory University, Atlanta, GA, USA, 3 University of Nairobi, Nairobi, Kenya, 4 Kenyatta National Hospital, Nairobi, Kenya, 5 Baylor College of Medicine, Houston, TX, USA Background: HIV-exposed uninfected infants (HEU) in TB endemic settings are at high risk of Mycobacterium tuberculosis (Mtb) infection and TB disease, even in the absence of known Mtb exposure. For infants, progression from primary infection to TB disease can be rapid; whether isoniazid (INH) prevents primary Mtb infection in HEU is unknown. Methods: We conducted a non-blinded RCT comparing 12 months daily INH (10 mg/kg) vs. no INH to prevent Mtb infection among HEU infants enrolled at 6 weeks of age in Kenya between August 2016-September 2019. Initially Mtb infection was assessed by interferon gamma release assay (IGRA, QFT-Plus) at 12 months. After February 2018, tuberculin skin test (TST, >10 mm) was added as a composite primary endpoint per DSMB recommendations. Results: Three-hundred HEU infants were enrolled (150 in each arm). Median age was 6 weeks (IQR 6.0-6.6), 158 (52.7%) were male, and 282 (94.0%) received BCG. All mothers were on antiretroviral (ARV) therapy; 297 (99.0%) infants received ARVs for HIV prevention. Two-hundred twenty-four (74.7%) mothers had received programmatic isoniazid preventive therapy (IPT) and 32 (10.7%) mothers reported history of TB. Excluding two HIV-infected children, retention was 96.0%with 286 participants completing the study (1 withdrew, 1 died, 10 were lost-to-follow up). Of 263 (88.3%) with available QFT and/or TST results, 3/246 (1.3%) were QFT-positive and 25/186 (13.4%) were TST-positive. Prevalence of Mtb infection (by QFT or TST) was 7.7% (10) in the INH and 13.5% (18) in the no INH arm (7.0 vs. 13.5 per 100PY, HR 0.55 [95% 0.25-1.17], p=.118). There was a trend for decreased risk of positive TST among children randomized to INH (8.9% vs. 18.3%, 5.6 vs. 12.7 per 100PY, HR 0.48 [95% 0.20-1.06], p=.06). Severe adverse events (primarily non-TB hospitalizations) were similar between arms (INH 12.7% [19] vs. no INH 10.0% [15], p=.72). There were no INH-related serious adverse events. One child was diagnosed with TB (INH arm), and one died of non-TB related causes (No INH arm). Conclusion: Rates of primary Mtb infection were approximately 2-fold lower in children randomized to INH, though this did not reach statistical significance, with a stronger trend of reduced TST-positivity. Secondary endpoint analyses including IFN-γ-independent immune markers in QFT-Plus supernatants to indicate Mtb infection are ongoing. 139 PREDICTORS OF TUBERCULOSIS INFECTION IN MDR-TB HOUSEHOLD CONTACTS ≥15 YEARS OLD Soyeon Kim 1 , Amita Gupta 2 , Xingye Wu 3 , Michael D. Hughes 3 , Rodney Dawson 4 , Vidya Mave 2 , Nagalingeswaran Kumarasamy 5 , Elizabeth Smith 6 , Roxana Rustomjee 7 , N. Sarita Shah 8 , Anneke Hesseling 9 , Susan Swindells 10 , Gavin Churchyard 11 , for the A5300/I2003 PHOENIx Feasibility Study Team 1 Frontier Science & Technology Research Foundation, Inc, Amherst, NY, USA, 2 Johns Hopkins University, Baltimore, MD, USA, 3 Harvard T.H. Chan School of Public Health, Boston, MA, USA, 4 University of Cape Town, Cape Town, South Africa, 5 Voluntary Health Services, Chennai, India, 6 DAIDS, NIAID, Bethesda, MD, USA, 7 DAIDS, NIAID, Rockville, MD, USA, 8 CDC, Atlanta, GA, USA, 9 Stellenbosch University, Cape Town, South Africa, 10 University of Nebraska Medical Center, Omaha, NE, USA, 11 The Aurum Institute, Johannesburg, South Africa

140 DIAGNOSTIC AND THERAPEUTIC CHALLENGES ARISE WITH EARLY HIV INFECTION ON PrEP Michael J. Peluso 1 , Monica Gandhi 1 , Susa Coffey 1 , Heather Hartig 1 , Susan P. Buchbinder 2 , Michael P. Busch 3 , Christopher D. Pilcher 1 , Hyman Scott 2 , Stephanie E. Cohen 2 , Darpun Sachdev 2 , Pierre Crouch 4 , Diane V. Havlir 1 , Steven G. Deeks 1 , Timothy J. Henrich 1 , Sulggi Lee 1 1 University of California San Francisco, San Francisco, CA, USA, 2 San Francisco Department of Public Health, San Francisco, CA, USA, 3 Vitalant Research Institute, San Francisco, CA, USA, 4 San Francisco AIDS Foundation, San Francisco, CA, USA Background: The impact of PrEP during HIV acquisition may alter reservoir establishment, viral load set points, and immune responses. Some individuals on PrEP may remain negative by screening assays while still becoming infected. Characterization of such individuals is needed to define how to diagnose early infection in this context.

47

CROI 2020