LAUSR

OBJECTIVES To analyse the frequency and causes of treatment discontinuation in patients who were treated with an integrase strand transfer inhibitor (INSTI), with a focus on neuropsychiatric adverse events (NPAEs). METHODS Patients in 18 HIV reference centres in France were prospectively included in the Dat'AIDS cohort. Data were collected from all patients starting an INSTI-containing regimen between 1 January 2006 and 31 December 2016. All causes of INSTI-containing regimen discontinuations were analysed, and patients' characteristics related to discontinuation due to NPAEs were sought. RESULTS INSTIs were prescribed to 21315 patients: 6274 received dolutegravir, 3421 received elvitegravir boosted by cobicistat, and 11620 received raltegravir. Discontinuation was observed in 12.5%, 20.2% and 50.9% of the dolutegravir-, elvitegravir- and raltegravir-treated patients, respectively (P < 0.001). Discontinuation for NPAEs occurred in 2.7%, 1.3% and 1.7% of the dolutegravir-, elvitegravir-, and raltegravir-treated patients, respectively (P < 0.001). In the multivariate analysis, discontinuation for NPAEs was related to dolutegravir versus elvitegravir (HR = 2.27; 95% CI 1.63-3.17; P < 0.0001) and versus raltegravir (HR = 2.46; 95% CI 2.00-3.40; P < 0.0001), but neither gender (HR for women = 1.19; 95% CI 0.97-1.46; P = 0.09) nor age (P = 0.12) was related. The association with abacavir was not retained in the final model. CONCLUSIONS Although discontinuation for side effects was less frequent with dolutegravir than with boosted elvitegravir, discontinuation for NPAEs, although rare (2.7%), was more frequent with dolutegravir. No patient characteristic was found to be associated with these side effects in this very large population.