A prospective patient registry to monitor safety, effectiveness, and utilisation of bedaquiline in patients with multidrug-resistant tuberculosis in South Korea

Background Multidrug-resistant tuberculosis (MDR-TB) represents a major public health concern, with an ongoing need for new effective treatments. Bedaquiline is an oral diarylquinoline that has shown encouraging treatment success and culture conversion rates in MDR-TB.Methods A South Korean patient registry was set up across 19 centres between 2016 and 2018 for the prospective collection of data from patients with MDR-TB who received either a bedaquiline-containing or a non-bedaquilinecontaining regimen. Treatment was at the physician's discretion (bedaquiline use requiring approval by special committee) and was based on patient characteristics, disease status, and local treatment guidelines.Results The safety population included 172 patients (88 bedaquiline and 84 non-bedaquiline). The mean (standard deviation, SD) duration of follow-up was 24.3 (9.5) months. Mean (SD) durations of treatment were 5.4 (1.8) months in bedaquiline-treated patients and 15.7 (6.7) months in the non-bedaquiline group. Treatment success (cured and treatment completed according to WHO 2013 treatment outcome definitions) was achieved by 56.3% of bedaquilinetreated and 45.2% of non-bedaquiline-treated patients. Sputum culture conversion rates were 90.4% and 83.7% with and without bedaquiline, respectively. Diarrhoea and nausea were the most frequently reported treatment-emergent adverse events (TEAEs) in the bedaquiline group (27.3% [24/88] and 22.7% [20/88], respectively). The most frequent bedaquiline-related TEAEs were prolonged QT interval (10.2%; 9/88), and diarrhoea and nausea (9.1% each; 8/88). QT interval prolongation was reported in 19.3% (17/88) of bedaquiline-treated and 2.4% (2/84) of non-bedaquilinetreated patients, but bedaquiline was not discontinued for any patient for this reason. There were 13 (14.7%) and three (3.6%) deaths in the bedaquiline-treated and non-bedaquiline groups, respectively. Review of fatal cases revealed no unexpected safety findings, and no deaths were bedaquiline-related. The most common cause of death was worsening cancer (three patients). Patients in the bedaquiline group tended to have poorer baseline risk profiles than nonbedaquiline patients and were more likely to have relapsed or already failed second-line treatment. Interpretation of mortality data was complicated by high rates of loss to follow-up in both groups. Conclusions The South Korean registry findings support previous risk/benefit observations and the continued use of bedaquiline as part of combination therapy in patients with MDR-TB.