Piotr Nowak [1], Stefan Lindbäck [2], Andreas Palmborg [2], Hatef Darabi [3], Magnus Gisslen [4], Bo Hejdeman [5], Anders Sönnerborg [1].
Affiliates: [1] Department of Infectious Diseases, Karolinska University Hospital/Unit of Infectious diseases, Department of Medicine Huddinge, Karolinska Institute, Sweden. [2] Janssen Pharmaceutical Companies of Johnson&Johnson. [3] Quantify, Stockholm, Sweden. [4] Department of Infectious Diseases, Sahlgrenska Academy, University of Gothenburg, Sweden. [5] Department of Infectious Diseases, Venhälsan-Södersjukhuset, Stockholm, Sweden.

Rilpivirine (RPV), a non-nucleoside reverse transcriptase inhibitor (NNRTI) launched in 2012, is perceived as having a low barrier to development of resistance and concern has been raised regarding the risk of virologic failure (VF) and development of drug resistance associated mutations (DRM) when used in clinical practice.

Data on all patients that have been treated with RPV in Sweden between January 2012 and 31st of October 2016 were collected from the national InfCareHIV database. VF was defined as having two consecutive viral loads (VL) ≥200 HIV-1 RNA copies/ml or one VL ≥200 copies/ml followed by RPV discontinuation after more than 6 months of treatment in treatment naïve and at any time point in treatment experienced patients. All patients that discontinued RPV during the first 6 months of treatment were individually assessed regarding reason for discontinuation to detect eventual early viral failure episodes. Patients with missing VL data during or close to RPV treatment period were excluded in the calculation of VF rate.

During the observation period 937 patients started RPV treatment; 190 treatment naïve and 747 treatment experienced patients. Of the patients 634 (68%) were male and 303 (32%) were female. Transmission route was MSM in 46%, heterosexual in 45%, IV drug use in 2% and unknown/missing in 7%. Mean age was 42 years in treatment naïve and 48 years in treatment experienced patients. In treatment naïve patients none had AIDS diagnosis at start of treatment, 5 patients (3%) had a VL >100 000 copies/ml and 7 patients (4%) had a CD4+ cell count <200×106/l. In treatment experienced patients 95 (13%) had an AIDS diagnosis, 2 (100 000 copies/ml and 26 (3%) had a CD4+ cell count <200×106/l at start of treatment.
After a mean/median RPV treatment duration of 31/35 months, 3/188 (1.6%) treatment naïve patients met the definition of VF. In treatment experienced patients the mean/median treatment duration was 29/32 months and 15/743 (2.0%) met the definition of VF. When patients with missing VL data were calculated as virologic failures (conservative approach) corresponding VF rates were 5/190 (2.6%) in treatment naïve and 19/747 (2.5%) in treatment experienced patients. The time to VF was 819, 867 and 920 days, respectively, in treatment naïve patients. The median (range) time to VF in treatment experienced patients was 230 (50-1321) days. 14/18 patients had genotypic resistance tests performed at time of VF; results showed no detectable RAMs in 6 patients and in another 6 patients new treatment emergent NNRTI and/or N[t]RTI RAMs were detected with or without pre-existing DRMs. In the 2 remaining patients one had an isolated transmitted K103N mutation and one had pre-existing thymidine analogue mutations (TAMs).

Virologic failure was rare and only seen in 18/937 (2%) of all patients treated with RPV in Sweden 2012-Oct 2016 in an on-treatment analysis where patients with missing values were excluded. New treatment emergent DRMs were detected in 6 of 14 patients with genotypic resistance test performed at VF.