CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

treated and cured, and <25% of those who could benefit from newer MDR-TB treatments receive them. Methods: We compared prices for 7 medicines prioritized in current WHO guidelines: bedaquiline, delamanid, linezolid, moxifloxacin, levofloxacin, cycloserine, and amikacin. Price data were gathered from national price sources for 41 countries in North America, Europe, the Middle East, South East Asia, India, and South Africa, and converted to the price for a monthly of treatment at standard doses. For all medicines except amikacin, only solid oral forms were used. Results: Monthly prices of MDR-TB medicines are displayed in the Table, for selected countries. As context, we include estimates of generic prices calculated in a previous analysis, based on analysis of the wholesale prices of the active pharmaceutical ingredients and other costs of production. For moxifloxacin, levofloxacin, linezolid, and amikacin, price data were available for more than 30 of the 47 countries. Data were limited for the other medicines, with prices of bedaquiline available only for 15 countries, cycloserine 9, and delamanid 4. Prices for a month of treatment ranged $94-5,273 for bedaquiline, $3,070–6,614 for delamanid, $7–4,856 for linezolid, $4–3,526 for amikacin, $48–2,501 for cycloserine, $1–674 for levofloxacin, and $4–206 for moxifloxacin. These were in most cases significantly higher than previously estimated generic prices (e.g. $8-17/month for bedaquiline, $5-16/month for delamanid). Conclusion: Prices of key MDR-TB medicines remain very high in many countries. The low availability of pricing data for bedaquiline, delamanid, and cycloserine may reflect unavailability in many countries. Global price differences are large for MDR-TB medicines, both for patented medicines (bedaquiline and delamanid, with linezolid recently off-patent) and to a lesser extent generics (all others). Attention is drawn to the high prices charged in ex-Soviet bloc countries that, despite now being classed as high-income countries or upper-middle- income in the case of Bulgaria, have relatively underresourced health systems and high burdens of MDR-TB (e.g. Bulgaria, Latvia, Lithuania, Slovenia).

frequencies (9-16kHz). We estimated standardized weekly AG exposure as: {prescribed daily AG dose (mg) x frequency of dosing per week} ÷ weight (kg). Cox proportional hazard and logistic regression were used for multivariable adjustment. Results: Of 936 participants, 54%were male; mean age was 36 years; and 75%were HIV coinfected at baseline. Comparing patients with high (≥75 mg/ kg/week) versus low (<75 mg/kg/week) AG exposure, the adjusted hazard (aHR) of regimen cessation due to ototoxicity was 1.33 (p=0.006), and for audiometric hearing loss was 1.34 (p=.038). Baseline hearing loss (aHR=1.71, p<.001) and age (aHR=1.02, p=.031) were also associated with increased hazard of hearing loss. Predictors of ototoxicity in the final prediction model included: standardized weekly AG exposure, HIV status, CD4 count, age, serum albumin, BMI, and baseline hearing loss. This model demonstrated moderate discrimination (AUC=0.72) and good calibration (χ2[8]=6.10, p=.64) at normal frequencies and better discrimination (AUC=0.81) at ultrahigh frequencies that might represent early manifestations of AG ototoxicity. Discrimination for AG regimen cessation due to ototoxicity (among 671 patients without baseline audiometric data) was weaker (AUC=0.60). Using a cutoff of 85% predicted probability of hearing loss, the positive predictive value was 100%, and the negative predictive value was 41%. Conclusion: This model identifies patients at high risk for AG-induced hearing loss and may inform clinical guidelines regarding which patients to prioritize for injectable-free regimens.

Poster Abstracts

741 RIFAMPICIN RESISTANCE ACCURATELY IDENTIFIED BY CLUSTERING ULTRA MELTING TEMPERATURES Gabriel D. Eisenberg 1 , Lesley Scott 1 , Puleng S. Marokane 2 , Pedro da Silva 2 , Kyle Fyvie 1 , Wendy Stevens 1 1 University of the Witwatersrand, Johannesburg, South Africa, 2 National Health Laboratory Service, Johannesburg, South Africa Background: South Africa (SA) has performed >1.2 million GeneXpert(Gx) MTB/RIF Ultra(Ultra) tests since October 2017 in the national tuberculosis(TB) program. All Ultra test results are stored in a central data warehouse and all operational data is accessible through C360(supplier dashboard) installed on each Gx. Ultra applies melting curve analysis (Tm) to differentiate wild type from rifampicin resistance (RR) mutations with 97% unambiguity on well characterised strains. This analysis describes that only using the Tm can accurately identify the RR conferring mutation obtained from Ultra test results. Methods: Ultra test results with rpoB Tm’s were extracted between October2017 to April2018 and k-means was used to cluster Tm’s into different RR mutation groups. Random forest(RF) imputation was applied to the wild type and RR Tm’s to assign probable Tm’s to incomplete values (1 to 3 rpoB probe values missing) thus optimizing clustering performance. K-means was applied to RR Tm’s and then to resulting clusters, labelled according to known Tm profiles and visualised with t-distributed stochastic neighbour embedding(t-

740 DEVELOPMENT AND VALIDATION OF A PREDICTIVE MODEL FOR AMINOGLYCOSIDE OTOTOXICITY

Hyejeong Hong 1 , David Dowdy 2 , Kelly E. Dooley 3 , Chakra Budhathoki 1 , Hae-Ra Han 1 , Jason Farley 1 1 Johns Hopkins University, Baltimore, MD, USA, 2 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 3 Johns Hopkins University School of Medicine, Baltimore, MD, USA Background: Individuals treated for drug-resistant tuberculosis (DR-TB) with aminoglycosides (AGs) in resource-limited settings often experience permanent hearing loss, but there is no practical and cost-effective means to identify those at higher risk. We sought to develop a prediction model of AG-induced hearing loss among patients initiating DR-TB treatment in South Africa. Methods: We nested this analysis within a cluster randomized trial of nurse-led case management in 10 South African TB hospitals. All participants ≥13 years old received kanamycin or amikacin. We performed clinical and audiometric evaluations at treatment initiation. Hearing loss was defined as a poorer hearing threshold compared to baseline. We developed the model using data from 265 patients at hearing frequencies from 250Hz to 8kHz and validated the model using data from 114 separate patients at both 250Hz-8kHz and ultrahigh

CROI 2019 285