CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

SNE). Algorithm performance was determined using root mean square error (RMSE), where a minimum Tm shift to identify a mutation is 2.3°C. Results: A total 52431 tests were extracted, 5354 positive(10.2%), 18(0.3% hetero resistant and excluded from analysis), 339 (6.3%) RR and 37(0.7%) were RR indeterminate. 121 RR results had incomplete Tm values. The attached figure shows clustering of RR Tm’s: mutations(n=280); potential wild types (n=5 to 24 owing to skewing by RF) and unknown (n=35). Average RMSE was 0.3 °C for known Tm profiles. Conclusion: Ultra and Xpert are clinically used as a qualitative diagnostic for TB and screen for RR. Ultra generates additional information over Xpert through Tm’s which can definitively identify RR conferring mutations and thus provide valuable information for individual patient care and population based surveillance. The algorithm tested here shows good accuracy (<=”” div=””>

expected budget impact as a result of the introduction of the shortened/BDQ MDR-TB regimen in South Africa. Results: By 2023, an annual estimated 5791, 6489, and 6670 patients will be successfully treated under the standard, standard/BDQ, and shortened/BDQ regimens, respectively. The cost/successful outcome for the standard regimen will be $8327 (annual $48million nationally). Introducing BDQ to standard- length treatment results in a similar cost per successful treatment ($8372), but higher total annual costs due to BDQ price ($54million/year). This increase in total cost can be offset by a simultaneous switch to a BDQ containing shortened regimen, $5232/treatment success, and impose an annual cost of $35million (28% savings compared to standard length therapy). Conclusion: Despite the increased cost of BDQ, the move from a 24-month to a 9-12 month regimen is predicted to result in a net decrease in cost along with improved outcomes for MDR/RR-TB patients in South Africa.

Poster Abstracts

743 NEURAL TUBE DEFECTS, HIV, AND ANTIRETROVIRALS: BIRTH-DEFECT SURVEILLANCE IN UGANDA Linda Barlow-Mosha 1 , Daniel M. Mumpe 1 , Dhelia Williamson 2 , Diana Valencia 2 , Robert Serunjogi 1 , Joyce N. Matovu 1 , Philippa Musoke 1 1 Makerere University–Johns Hopkins University Research Collaboration, Kampala, Uganda, 2 CDC, Atlanta, GA, USA Background: Neural tube defects (NTDs) are one of the most common congenital malformations affecting births worldwide. The estimated NTD prevalence in Africa is 12 per 10,000 live births [95% confidence interval (CI) 5-75]; but data are limited. The impact of antiretroviral therapy (ART) during pregnancy on the risk for birth defects is unknown; therefore, ongoing surveillance is needed for pharmacovigilance. Methods: A hospital-based surveillance programwas established at four hospitals in Kampala, Uganda to provide estimates of the baseline prevalence of selected birth defects and assess potential associations with HIV status and ART use. All live births and stillbirths, regardless of gestational age, were included. Data were collected from hospital records, maternal interviews, photographs, and narrative descriptions of birth defects (BD). Births were examined by trained midwives and confirmed by a BD specialist. Prevalence (Wilson 95% CI) and adjusted odds ratio (AOR) estimates for potential risk factors of NTDs using logistic regression with site as an effect modifier are reported for births from August 2015-December 2017. Results: A total of 69,767 births were included in surveillance. Median maternal age was 26 years (IQR=22-30), 51.3% had their first antenatal visit after the first trimester, 9.6% (6,725/69,767) were HIV-infected with 95.2% (6,399/6,725) on ART. The majority of HIV-infected women were on an efavirenz-based ART (80%), 16% on nevirapine-based ART and 4% received other ART regimens. Overall, 62 births were affected with NTDs, giving a prevalence of 8.9 (6.8-11.4) per 10,000 live births. Spina bifida (n=34) was the most common type of NTD with prevalence (95%CI) of 4.9 (3.4-6.8) per 10,000 live births, followed by anencephaly (n=16) with 2.3 (1.3-3.7) and encephalocele (n=12) with 1.7 (0.9- 3.0). There was no significant difference in NTD prevalence (95%CI) among HIV- infected [5.9 (0.1-11.8)] and HIV un-infected women [9.2 (6.8-11.6)]; AOR 0.75 (95% CI:0.2-2.2), p= 0.61. NTDs were not significantly associated with maternal age, HIV status, ART, or parity. Anencephaly was more common among females

742 THE COST AND IMPACT OF MDR-TB TREATMENT GUIDELINE CHANGES IN SOUTH AFRICA Sithabiso D. Masuku 1 , Ribka Berhanu 2 , Craig Van Rensburg 1 , Nobert Ndjeka 3 , Sydney Rosen 2 , Lawrence Long 2 , Denise Evans 1 , Brooke E. Nichols 2 1 Health Economics and Epidemiology Research Office, Johannesburg, South Africa, 2 Boston University, Boston, MA, USA, 3 South African National Department of Health, Pretoria, South Africa Background: The current standard regimen for treating multi-drug and rifampicin-resistant tuberculosis (MDR/RR-TB) in South Africa is 24 months, requires daily injectable aminoglycoside (IA) treatment during the first 6 months, has significant side-effects, and poor treatment outcomes. Recent evidence supports the use of a shortened regimen (9-12 months), as well as the replacement of the IA with the well tolerated, higher-cost, oral bedaquiline (BDQ). To determine affordability of a switch in the MDR/RR-TB regimen for the South African Government, we compared the budgetary impact of the use of BDQ in the current MDR/RR-TB regimen or in the shortened regimen. Methods: We developed a Markov model to analyze the impact of changes in the treatment regimens on patient outcomes and total costs over a five-year time horizon (2019-2023). The model utilized the South African RR-TB case register (EDRweb) from 2013-2015 to define treatment outcomes. The standard regimen scenario allows for switching (14% in the first month) to BDQ when toxicity to the IA occurs, as per current practice. Costs were estimated from the provider perspective using costs incurred by 137 MDR-TB patients at Helen Joseph Hospital in Johannesburg and included laboratory tests, staff, consumables, equipment, overhead, and 2018 drug prices. Improvement in mortality and successful outcomes were assumed for those receiving BDQ. Results are reported as the cost of reaching a successful outcome (cured or completed treatment), calculated as total cost/successful outcomes and

CROI 2019 286