Antiviral Therapy 2016; 21: (doi: /IMP3024)

Transcription

1 Antiviral Therapy 2016; 21: (doi: /IMP3024) Original article Systematic review to determine the prevalence of transmitted drug resistance mutations to rilpivirine in HIV-infected treatment-naive persons Vincent Calvez 1, Anne-Genevieve Marcelin 1, Johan Vingerhoets 2, Andrew Hill 3 *, Blanca Hadacek 4, Christiane Moecklinghoff 5 1 Sorbonne Universités, UPMC Univ Paris 06, UMR_S 1136, Institut Pierre Louis d Epidémiologie et de Santé Publique, INSERM, UMR_S 1136, Institut Pierre Louis d Epidémiologie et de Santé Publique, AP-HP, Hôpital Pitié-Salpêtrière, Service de Virologie, Paris, F-75013, France 2 Clinical Virology, Janssen Infectious Diseases, Beerse, Belgium 3 MetaVirology Ltd, London, UK 4 Janssen EMEA, Issy-les-Moulineux, France 5 Janssen EMEA, Neuss, Germany *Corresponding author microhaart@aol.com Background: Transmitted drug resistance to antiretrovirals in HIV-1-infected individuals is rising in some regions and could compromise the effectiveness of first-line treatment. It is important to understand the prevalence of resistance to rilpivirine to inform treatment provision. Methods: A PUBMED/EMBASE search identified analyses of transmitted genotypic resistance to specific non nucleoside reverse transcriptase inhibitor mutations worldwide. Patients were to be HIV-1-infected and antiretroviral-naive. Rilpivirine mutations assessed were: L100I, K101E/P, E138A/G/K/Q/R, V179L, Y181C/I/V, Y188L, H221Y, F227C and M230I/L. Additionally, frequency of resistance mutations were extracted and pooled by HIV subtype from the Stanford HIV drug resistance database. Results: 138 eligible articles from 65 countries were identified (n=64,466). Among these 64,466 samples, 7 of the 9 genotypic rilpivirine mutations had a prevalence <0.1%. Two mutations were more prevalent: E138A/G/K/Q/R (0.7%, 95% CI 0.2, 1.3) and Y181C/I/V (0.3%, 95% CI 0.2, 0.4). Prevalence of E138 rilpivirine-related mutations varied between regions: highest in Latin America/ Caribbean (3.6%, 95% CI 1.0, 7.6) and in Europe (3.2%, 95% CI 0.7, 6.9). Pooled results from the Stanford database (n=52,680) correlated with these findings indicating a low prevalence of 8/9 rilpivirine mutations (<0.1%), except for E138A/G/K/Q/R (2.9%, 95% CI 1.8, 4.4). Prevalence of the mutations at E138 varied significantly by HIV subtype and was highest for subtype-c (6.1%), subtype-f (5.1%) and subtype-a (3.3%). Conclusions: The prevalence of most transmitted rilpivirine-related HIV mutations is generally low in treatmentnaive HIV-1-infected individuals (<0.1%). The prevalence of E138A/G/K/Q/R mutations is higher (0.7%) and varies according to geographical region and HIV subtype. Introduction Intensive scale-up of antiretrovirals for HIV has seen worldwide access and has led to a dramatic decrease in HIV-related morbidity and mortality [1,2]. Despite these successes, the expansion of treatment has been accompanied by a significant increase in the prevalence of both acquired and transmitted HIV drug resistance [3]. Transmitted drug resistance (TDR), in which an individual is infected by a strain of HIV already resistant to one or more drugs, may impact response to therapy, leading to virological failure and the evolution of further drug resistance [4,5]. Thus, TDR is a significant public health concern that could compromise the effectiveness of treatment campaigns. The increasing prevalence of TDR has been driven, in particular, by an increase in resistance to non-nucleoside reverse transcriptase inhibitors (NNRTIs) [6]. This is especially true in sub-saharan Africa as a result of the extensive use of efavirenz and nevirapine in mass treatment roll-out [3,6]. Rilpivirine (RPV), a secondgeneration NNRTI approved in 2011, was designed to improve the resistance profile and overcome tolerability limitations associated with first-generation NNRTIs [7] International Medical Press (print) (online) 405

2 V Calvez et al. RPV as a component of initial therapy has been evaluated in two large randomized clinical trials (ECHO and THRIVE) [8,9]. In these studies, non-inferior efficacy versus efavirenz was demonstrated in patients with a baseline viral load 100,000 copies/ml, and moreover, RPV was associated with a more favourable safety and toxicity profile. Due to its specific interaction with the reverse transcriptase binding site, in theory, RPV possesses a different pattern of emergence of resistance than first-generation NNRTIs [10]. Studies have shown a higher rate of resistance selection after virological failure for RPV compared with efavirenz-based treatment, particularly in patients with a viral load >100,000 copies/ml at baseline [8,11]. The development of drug mutations is more common in patients with a viral load below 100,000 copies/ml [12] Sixteen mutations have been associated with decreased RPV susceptibility and have been included in the drug resistance mutation guidelines [13]. Changes with the greatest impact on RPV susceptibility are at four codons (K101E/P/T, E138A/G/K/R, Y181C/I/V and M230L), whereas changes at another nine positions display a lower impact (V90I, L100I, V106A/I, V108I, V179F/I/L, Y188I, G190E, H221Y and F227C/L). However, in the presence of M184I, only one of two changes (either E138K or K101E) is enough to produce high-level RPV resistance. The magnitude of effect varies significantly from a 50-fold and 15-fold reduction in susceptibility (K101P and Y181I/V, respectively), down to a decrease around 2.5- to 7-fold for the remaining mutations [13]. Data relating to RPV-associated mutations in treatment-naive patients is crucial to inform the provision of treatment. As such, the aim of this study was to systematically review the available data to examine the prevalence of RPV-associated mutations in HIV-1-infected treatment-naive patients. Methods Criteria for inclusion Included studies were either observational studies, or randomized controlled trials enrolling treatment-naive patients with evaluation of RPV mutations at baseline. Observational studies included surveillance surveys, as recommended by the WHO in the WHO Threshold Survey, cross-sectional routine clinical surveillance, retrospective or prospective cohorts, and retrospective analyses of databases and blood samples. Participants were to be HIV-1-infected and antiretroviral treatment-naive and could be either newly diagnosed or chronically infected. Studies targeting the general HIV-infected population, as well as specific target groups, including pregnant women, men who have sex with men (MSM), sex-workers and adolescents, were included. There were no restrictions on study location. Studies were excluded if they sampled less than ten patients or if it was not possible to retrieve the time period in which the sampling occurred. Included studies were to have data on the prevalence of any of the specific mutations which confer resistance to the NNRTI RPV as listed in the 2014 update of the IAS-USA list of drug resistance mutations in HIV-1 [13]. These mutations are as follows: L100I, K101E/P, E138A/G/K/Q/R, V179L, Y181C/I/V, Y188L, H221Y, F227C and M230I/L. The outcome measured was the percentage of the population with each mutation. Studies were excluded if they did not contain data on individual mutations; however it was permissible for studies not to evaluate every listed mutation. Search strategy and data collection Studies for inclusion were identified through an electronic search of the PubMed database. The database was searched from 2010 until the end of April Only those articles published in English were included in the present review. This short timespan was deemed appropriate given the evolving prevalence of HIV drug resistance mutations; there were no restrictions on the time period analysed in each study. The following search strategy was used to locate the relevant studies: (HIV OR HIV-1 OR human immunodeficiency virus [Title]) AND (mutation* OR resistan* [Title]). The titles and abstracts of all identified publications were screened. The full text of all studies that were deemed potentially relevant were obtained and reviewed against the inclusion criteria. In addition to the proportion of patients with each individual mutation, data on study design, patient population and characteristics, study period and location, and the list of mutations analysed were extracted from each report. For overlapping studies, the most recent and comprehensive report was used for the analysis. Screening was performed by one individual with any queries resolved by consensus with a second person. Additionally, estimates for the frequency of RPV resistance mutations were extracted from the Stanford University HIV drug resistance database [14]. The database is a publically available online resource listing the prevalence of HIV drug resistance mutations from over 40,000 individuals inputted from more than 900 literature and GenBank references. Data was extracted for all RPV listed mutations in NNRTI-naive individuals (regardless of protease inhibitor history) in a range of HIV-1 subtypes. Data synthesis Data from individual studies were extracted and aggregated by region and year of analysis. In studies in which sampling spanned more than 1 year, the mid-point year of recruitment was used for assessment. Geographic regions International Medical Press

3 Transmitted drug resistance mutations to rilpivirine used were modified from those used by the United Nations and the World Bank as follows: Africa, subdivided in to Sub-Saharan Africa and North Africa and the Middle East; Asia, subdivided into Central Asia, South Asia, East Asia and South-East Asia; Oceania; Latin American and the Caribbean; North America; and Europe [15,16]. Statistical analysis Data were pooled by region using random effects meta-analysis according to the methods of DerSimonian and Laird [17]. Prior to analysis, data was transformed using the Freeman Turkey transformation to normalize the data and account for studies with zero events [18]. Results are presented with the corresponding Wilson confidence interval. Heterogeneity was quantified using the I-squared statistic, where a result greater than 70% indicated substantial heterogeneity. All analyses were conducted using STATA (version 12.0; Stata Corp, College Station, TX, USA). Results As shown in Figure 1, the search strategy returned 644 titles that were screened for inclusion. Of these, 284 were found to be of possible relevance and the full text was obtained. In total, 138 articles were included in the final analysis; the main reasons for exclusion are shown in Figure 1. The 138 articles included data from 65 countries, comprising a total of 66,195 individuals with HIV-1 infection (Table 1; individual studies shown in Additional file 1). The majority of studies originated from Africa (42; number of persons =7,419), followed by Europe (33; n=43,612), Asia (28; n=9,586), Latin America and the Caribbean (24; n=2,486) and North America (11; n=3,092); no studies were found from the Oceania region. In terms of sample size, the majority of the total population were from studies carried out in Europe (65.9%). Methods of sampling varied widely between the included studies. With respect to the population, studies included patients of varying HIV-1 subtypes, those with recent and chronic infection, and often analysed specific subpopulations including pregnant women, MSM and youths. The sampling frame utilized was predominantly consecutive enrolment, however other methods, such as random selection were also common. For sequencing, the majority of studies used plasma samples, however, Figure 1. Flow chart of included studies Studies identified through PubMed search (n=644) Excluded as title irrelevant (n=360) Full text retrieved (n=284) Excluded for following reasons (n=146): Patients not ART-naive: 63 Irrelevant topic: 51 Article not in English: 12 Unable to access article: 9 No data on individual mutations: 5 Date of sampling unknown: 4 Pre-ART era: 1 Duplication: 1 Included in final analysis (n=138) Antiviral Therapy

4 V Calvez et al. Table 1. Studies included in the review, by region and year of survey All years Total number of studies 21 (25,136) 54 (27,812) 66 (13,247) 138 a (66,195) Total number of countries represented Africa 2 (101) 16 (5,196) 24 (2,122) 42 (7,419) Sub-Saharan Africa 2 (101) 15 (5,159) 21 (2,026) 38 (7,286) Middle East and North Africa 0 1 (37) 3 (96) 4 (133) Asia 1 (1,349) 8 (1,425) 19 (6,812) 28 (9,586) South Asia 0 2 (78) 1 (44) 3 (122) East Asia 1 (1,349) 3 (490) 11 (2,640) 15 (4,479) South-East Asia 0 3 (857) 7 (4,128) 10 (4,985) Latin America and the Caribbean 2 (111) 9 (1,201) 14 (1,174) 24 a (2,486) North America 2 (1,244) 6 (1,292) 3 (556) 11 (3,092) Europe 14 (22,331) 15 (18,698) 5 (2,583) 33 a (43,612) Data are number of studies (number of samples) unless otherwise indicated. a Sum of year groups does not equal total number of studies in all years as one European study was split into all three year group categories and one Latin America and the Caribbean study was split into two year groups. some used dried blood spot sampling, and specific tools used for extraction and sequencing varied. Resistance to RPV The pooled results suggest that resistance to RPV is rare with the prevalence of mutations varying from 0.0% (95% CI 0.0, 0.0) to 0.7% (95% CI 0.3, 1.4) overall (Table 2). With regard to specific mutations, prevalence of the L100I, K101E/P, V179L, Y188L, H221Y, F227 and M230I/L mutations were consistently low across the different regions (Table 2). Prevalence of the E138A/ G/K/Q/R mutation varied worldwide, as shown by the high level of between-study heterogeneity observed (I 2 =87.1%; Figure 2). The pooled results suggest that the prevalence of E138 RPV-related mutations were highest in Latin America and the Caribbean (3.6%, 95% CI 1.0, 7.6%) and in Europe (3.0%, 95% CI 1.3, 5.2%). Within Europe, there was a significant level of heterogeneity observed (I 2 =84.4%) and the prevalence varied from 0.0% to 7.4% in individual studies. The pooled results suggest that the prevalence of Y181C/I/Y is 0.3% (95% CI 0.2, 0.4%), however, it was particularly high in North America (2.0%, 95% CI 0.5, 2.6%). Of the 263 samples with mutations at position E138, 80% were E138A, 10% were E138K, 9% were E138G, 1% were E138Q and none was E138R. There were too few samples available to compare the percentage with the different E138 mutations by region. Similarly, the results of the Stanford resistance database indicate a very low prevalence of L100I, K101E/P, V179L, Y181C/I/V, Y188L, H221Y, F227C and M230I/L across HIV-1 subtypes (Table 3). The pooled result for E138A/G/K/Q/R suggests a pooled prevalence of 2.9% (95% CI 1.9, 4.4%), however this varied dramatically by subtype. The highest E138- related mutation prevalence was observed in subtype- C (6.1%), subtype-f (5.1%), subtype-a (3.3%) and subtype-d (3.0%). Discussion This large systematic review included the genotypic results from over 65,000 antiretroviral-naive patients across 65 countries worldwide. The pooled results suggest that the prevalence is low for all RPV-related mutations identified in the 2014 IAS-USA update of drug resistance mutations in HIV-1 [13]. The most prevalent mutations worldwide were E138A/G/K/Q/R and Y181C/I/V at 0.7% (95% CI 0.2, 1.3) and 0.3% (95% CI 0.2, 0.4), respectively. The pooled analysis from the Stanford database validated these findings, indicating a low prevalence of RPV mutations (95% CI 0.0, 0.1%) except for E138A/G/K/Q/R which was higher than that observed in the systematic review (2.9%, 95% CI 1.8, 44). In the systematic review, the prevalence of E138A/ G/K/Q/R mutations varied significantly by geographic location. Prevalence was highest in Latin America and the Caribbean and across Europe. In the analysis of the Stanford database, the prevalence of E138 mutations varied by HIV subtype and was highest for subtype-c (6.1%), subtype F (5.1%) and subtype A (3.3%). Subtype C is prevalent in sub-saharan Africa, India and Brazil, and subtype-f is endemic in South and South-East Asia. Subtype A is prevalent across Central and East Africa, and parts of Eastern Europe [19]. Thus, whilst the high prevalence of E138 mutations observed in these subtypes somewhat overlaps the prevalence observed across regions, the results from the systematic review and the Stanford International Medical Press

5 Transmitted drug resistance mutations to rilpivirine Table 2. Pooled estimates of the prevalence of specific RPV drug resistant mutations among antiretroviral-naive HIV-1-infected individuals, L100I K101E/P E138A/G/K/Q/R V179L Y181C/I/V Y188L H221Y F227C M230I/L Overall (all countries) 27/38,269; /45,186; /13,149; 0.7 4/6,848; /65,181; /36,391; 0.0 9/9,265; 0.0 2/9,208; /42,567; 0.0 (0.0, 0.0) (0.0, 0.1) (0.3, 1.4) (0.0, 0.0) (0.2, 0.4) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) Africa 3/4,597; /6,940; /1,616; 0.6 0/130; 49/7,419; 0.2 2/4,597; 0.0 0/1,153; 0.0 0/1,096; 0.0 2/7,033; 0.0 (0.0, 0.0) (0.0, 0.1) (0.1, 1.5) (0.0, 0.3) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) Sub-Saharan Africa 2/4,464; /6,897; /1,579; 0.7 0/93; 48/7,286; 0.2 2/4,464; 0.0 0/1,097; 0.0 0/1,040; 0.0 2/6,900; 0.0 (0.0, 0.0) (0.0, 0.1) (0.1, 1.6) (0.0, 0.4) (0.0, 0.0) (0.0, 0.1) (0.0, 0.1) (0.0, 0.0) Middle East and North Africa 1/133; 0.2 1/133; 0.1 0/37; 0/37; 1/133; 0.2 0/133; 0.0 0/56; 0/56; 0/133; 0.0 (0.0, 2.6) (0.0, 2.3) (0.0, 2.6) (0.0, 1.5) (0.0, 1.5) Asia 5/7,691; /9,437; /5,967; 0.1 2/3,688; /9,586; 0.1 6/7,708; 0.0 3/4,727; 0.0 2/4,727; 0.0 3/9,437; 0.0 (0.0, 0.0) (0.0, 0.1) (0.0, 0.5) (0.0, 0.0) (0.0, 0.3) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) South Asia 0/122; 0.0 2/122; 1.2 0/44; 0/44; 0/122; 0.0 0/122; 0.0 0/75; 0/75; 0/122; 0.0 (0.0, 0.2) (0.0, 4.5) (0.0, 1.7) (0.0, 1.7) (0.0, 1.7) East Asia 4/4,893; /5,012; 0.1 3/3,764; 0.0 0/1,915; /5,161; 0.0 4/5,012; 0.0 0/2,493; 0.0 0/2,493; 0.0 1/5,012; 0.0 (0.0, 0.1) (0.0, 0.2) (0.0, 0.1) (0.0, 0.0) (0.0, 0.1) (0.0, 0.1) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) South-East Asia 1/2,676; /4,303; /2,159; 1.4 2/1,729; 30/4,303; 0.4 2/2,574; 0.0 3/2,159; 0.0 2/2,159; 0.0 2/4,303; 0.0 (0.0, 0.0) (0.0, 0.1) (0.9, 2.0) (0.2, 0.7) (0.0, 0.0) (0.0, 0.2) (0.0, 0.1) (0.0, 0.0) Latin America and the Caribbean 2/2,390; /2,390; /317; 3.6 0/125; 17/2,449; 0.3 2/2,390; 0.0 0/207; 0.0 0/207; 0.0 2/2,238; 0.0 (0.0, 0.0) (0.0, 0.5) (1.0, 7.6) (0.1, 0.6) (0.0, 0.0) (0.0, 0.9) (0.0, 0.9) (0.0, 0.1) North America 6/2,221; /3,092; 0.3 4/864; 0.1 0/864; /3,092; /2,186; 0.4 1/864; 0.0 0/864; 0.0 0/2,649; 0.0 (0.0, 0.3) (0.0, 0.7) (0.0, 1.3) (0.0, 0.1) (0.5, 2.6) (0.0, 0.8) (0.0, 0.2) (0.0, 0.1) (0.0, 0.0) Europe 11/21,370; /23,327; /4,385; 3.0 2/2,041; /42,635; /19,510; 0.0 5/2,314; 0.0 0/2,314; 0.0 8/21,210; 0.0 (0.0, 0.0) (0.0, 0.1) (1.3, 5.2) (0.0, 0.0) (0.2, 0.7) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) (0.0, 0.0) Total number of individuals (studies) 38,269 (123) 45,186 (130) 13,149 (45) 6,848 (18) 65,181 (139) 36,391 (119) 9,265 (38) 9,208 (37) 42,567 (126) Data are the number of cases/number of samples; pooled estimate (95% CI) unless otherwise indicated; results were not pooled if there were <3 studies reporting data (marked with a dash). Studies underlined had an I 2 value indicating considerable between-study heterogeneity (I 2 70%). RPV, rilpivirine. Antiviral Therapy

6 V Calvez et al. Figure 2. Prevalence of E138A/G/K/Q/R mutation by region 8 Prevalence of E138A/G/K/Q/R mutations, % n=1,579 Sub-Saharan Africa n=3,764 East Asia n=2,159 South-East Asia n=317 Latin America/ Caribbean n=864 North America n=4,385 Europe n=13,149 a Pooled estimate Graph shows the prevalence of E138A/G/K/Q/R mutations by region with the corresponding 95% CI. Shown are the number of individuals included in each analysis. a For regions with less than three studies, results were not pooled individually. The pooled estimate includes the data from these regions (Middle East and North Africa, n=37; South Asia, n=44). Table 3. Prevalence of specific RPV drug resistant mutations among NNRTI-naive HIV-1 individuals from the Stanford drug resistance database, by HIV-1 subtype Sample Mutation prevalence, % (95% CI a ) Subtype size, n L100I K101E/P E138A/G/K/Q/R V179L Y181C/I/V Y188L H221Y F227C M230I/L Subtype-A 4,195 0 (0.0, 0.1) 0 (0.0, 0.1) 3.3 (2.8, 3.9) 0 (0.0, 0.1) 0 (0.0, 0.1) 0 (0.0, 0.1) 0.1 (0.0, 0.2) 0 (0.0, 0.1) 0.1 (0.0, 0.2) Subtype-B 27,392 0 (0.0, 0.0) 0.1 (0.1, 0.1) 2.0 (1.8, 2.2) 0 (0.0, 0.0) 0.1 (0.1, 0.1) 0.1 (0.1, 0.1) 0 (0.0, 0.0) 0 (0.0, 0.0) 0 (0.0, 0.0) Subtype-C 8,671 0 (0.0, 0.0) 0.1 (0.0, 0.2) 6.1 (5.6, 6.6) 0 (0.0, 0.0) 0.1 (0.0, 0.2) 0 (0.0, 0.0) 0 (0.0, 0.0) 0 (0.0, 0.0) 0 (0.0, 0.0) Subtype-D 1, (0.0, 0.4) 0.4 (0.1, 0.8) 3.0 (2.2, 4.0) 0 (0.0, 0.2) 0.1 (0.0, 0.4) 0 (0.0, 0.2) 0.2 (0.0, 0.5) 0 (0.0, 0.2) 0 (0.0, 0.2) Subtype-F (0.0, 0.5) 0 (0.0, 0.5) 5.1 (3.6, 6.9) 0 (0.0, 0.5) 0 (0.0, 0.5) 0 (0.0, 0.5) 0 (0.0, 0.5) 0 (0.0, 0.5) 0.1 (0.0, 0.8) Subtype-G 1,593 0 (0.0, 0.2) 0.3 (0.1, 0.7) 2.4 (1.7, 3.3) 0 (0.0, 0.2) 0 (0.0, 0.2) 0.2 (0.0, 0.5) 0 (0.0, 0.2) 0 (0.0, 0.2) 0.1 (0.0, 0.5) Subtype-AE 5,640 0 (0.0, 0.0) 0 (0.0, 0.0) 0.8 (0.6, 1.1) 0 (0.0, 0.0) 0.1 (0.0, 0.2) 0 (0.0, 0.0) 0 (0.0, 0.0) 0 (0.0, 0.0) 0.1 (0.0, 0.2) Subtype-AG 2,959 0 (0.0, 0.1) 0.1 (0.0, 0.3) 2.7 (2.2, 3.4) 0 (0.0, 0.1) 0.1 (0.0, 0.2) 0 (0.0, 0.1) 0 (0.0, 0.1) 0 (0.0, 0.1) 0 (0.0, 0.1) Pooled estimate b 0.0 (0.0, 0.0) 0.1 (0.0, 0.2) 2.9 (1.8, 4.4) 0.0 (0.0, 0.0) 0.1 (0.0, 0.1) 0.0 (0.0, 0.1) 0.0 (0.0, 0.0) 0.0 (0.0, 0.0) 0.0 (0.0, 0.1) Studies underlined had an I 2 value indicating considerable between-study heterogeneity (I 2 70%). a Exact 95% CI; in the case of no events, the two-sided 97.5% CI was used to calculate the upper interval. b Results of random-effects meta-analysis, prevalence % (95% CI). NNRTI, non-nucleoside reverse transcriptase inhibitor; RPV, rilpivirine. database do not correlate completely. A limitation of this analysis is that the results from the systematic review could not be analysed by HIV sub-type, and the results from the Stanford database could not be analysed by region. The results from the systematic review and the Stanford database both showed a very low prevalence of all RPV-associated mutations except for those at position E138. For effective treatment delivery, it is important to understand the effect harbouring specific TDR mutations has on the clinical outcomes of treatment. While data on RPV resistance is being rapidly amassed, no randomized trial to assess the use of RPV-containing regimens among patients with key baseline RPV mutations has been conducted. In both the ECHO and THRIVE clinical trials, patients with International Medical Press

7 Transmitted drug resistance mutations to rilpivirine baseline TDR were excluded from randomization, and presently RPV is only recommended in treatment-naive patients without RPV resistance mutations [8]. Available evidence suggests that presence of the E138A/G/K/Q mutations may reduce RPV susceptibility by about two- to threefold [20 22]. Importantly, the combination of E138K and the nucleoside reverse transcriptase inhibitor mutation M184I further reduces RPV susceptibility to around fivefold, and the combination appears sufficient to bring about virological failure of patients on regimens containing RPV [20,21]. Y181C, the most prevalent Y181 mutation observed in this review reduces RPV susceptibility by around threefold [20]. The K101P mutation is likely to have the largest effect on treatment outcome, reducing RPV susceptibility by around 50-fold. This analysis suggests however that the prevalence of K101P is currently very low worldwide (K101E/P 0.0%, 95% CI 0.0, 0.01). The present study has a number of limitations. Foremost, the analysis included all studies evaluating RPV mutations and was not discriminative based on study type. This means that the methods significantly differed between studies with respect to the studied populations, the sampling frame and the laboratory methods used. As such, the estimates may not be representative nationally. In conclusion, prevalence of TDR to RPV is low worldwide. Even for the most prevalent mutations identified, E138A/G/K/Q and Y181C/I/V, occurrence was rare and the effect of the mutation was limited, reducing RPV susceptibility by two- to threefold. While there is limited evidence regarding treatment outcome in patients harbouring RPV resistance mutations, RPV-based regimens should not be used in such cases in order to prevent virological failure and increased prevalence of RPV related mutations. Baseline testing for drug resistance should be used, where feasible, for patients starting first-line NNRTI-based treatment, consistent with current international treatment guidelines. Acknowledgements We would like to thank Bryony Simmons (Imperial College London, London, UK) for help in preparing this manuscript. Disclosure statement CM, BH and JV are employees of Janssen. AH has received consultancy payments from Janssen. The other authors (VC and A-GM) do not declare any conflicts of interest. Additional file Additional file 1: A table of studies included in the meta-analysis of transmitted RPV-associated drug resistance can be found at com/uploads/documents/3697_calvez_addfile1.pdf References 1. Mocroft A, Vella S, Benfield T, et al. Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet 1998; 352: Gilks CF, Crowley S, Ekpini R, et al. The WHO publichealth approach to antiretroviral treatment against HIV in resource-limited settings. Lancet 2006; 368: Gupta RK, Jordan M, Sultan B, et al. Global trends in antiretroviral resistance in treatment-naive individuals with HIV after rollout of antiretroviral treatment in resourcelimited settings: a global collaborative study and metaregression analysis. Lancet 2012; 380: Booth C. Chapter 10 Transmitted resistance. In Geretti A (editor). Antiretroviral resistance in clinical practice. London: Mediscript, Little SJ, Holte S, Routy J, et al. Antiretroviral-drug resistance among patients recently infected with HIV. N Engl J Med 2002; 347: World Health Organization. WHO HIV drug resistance report Switzerland, Geneva: WHO (Accessed 1 June 2015.) Available from drugresistance/report2012/en 7. Ripamonti D, Bombana E, Rizzi M. Rilpivirine: drug profile of a second-generation non-nucleoside reverse transcriptase HIV-inhibitor. Expert Rev Anti Infect Ther 2014; 12: Cohen CJ, Molina J, Cassetti I, et al. Week 96 efficacy and safety of rilpivirine in treatment-naive, HIV-1 patients in two Phase III randomized trials. AIDS 2013; 27: Behrens G, Rijinders B, Nelson M, et al. Rilpivirine versus efavirenz with emtricitabine/tenofovir disoproxil fumarate in treatment-naive HIV-1-infected patients with HIV-1 RNA 100,000 copies/ml: week 96 pooled ECHO/ THRIVE subanalysis. AIDS Patient Care STDS 2014; 28: Das K, Bauman J, Clark AJ, et al. High-resolution structures of HIV-1 reverse transcriptase/tmc278 complexes: strategic flexibility explains potency against resistance mutations. Proc Natl Acad Sci U S A 2008; 105: Porter DP, Kulkarni R, Fralich T, Miller M, White K. 96-week resistance analyses of the STaR study: rilpivirine/ emtricitabine/tenofovir DF versus efavirenz/emtricitabine/ tenofovir DF in antiretroviral-naive, HIV-1-infected subjects. HIV Clin Trials 2015; 16: Parczewski M. Rilpivirine in the light of the genotypic drug resistance data. HIV & AIDS Review 2013; 12: Wensing AM, Calvez V, Gunthard H, et al Update of the drug resistance mutations in HIV-1. Top Antivir Med 2014; 22: Rhee SY, Gonzales M, Kantor R, Betts B, Ravela J, Shafer R. Human immunodeficiency virus reverse transcriptase and protease sequence database. Nucleic Acids Res 2003; 31: United Nations. Composition of macro geographical (continental) regions, geographical sub-regions, and selected economic and other groupings (Accessed 5 June 2015.) Available from m49/m49regin.htm 16. World Bank. Regions (Accessed 4 June 2015.) Available from Antiviral Therapy

8 V Calvez et al. 17. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: Rücker G, Schwarzer G, Carpenter J, Olkin I. Why add anything to nothing? The arcsine difference as a measure of treatment effect in meta-analysis with zero cells. Stat Med 2009; 28: Buonaguro L, Tornesello ML, Buonaguro FM. Human immunodeficiency virus type 1 subtype distribution in the worldwide epidemic: pathogenetic and therapeutic implications. J Virol 2007; 81: Rimsky L, Vingerhoets J, Van Eygen V, et al. Genotypic and phenotypic characterization of HIV-1 isolates obtained from patients on rilpivirine therapy experiencing virologic failure in the Phase 3 ECHO and THRIVE studies: 48-week analysis. J Acquir Immune Defic Syndr 2012; 59: Kulkarni R, Babaoglu K, Lansdon E, et al. The HIV-1 reverse transcriptase M184I mutation enhances the E138Kassociated resistance to rilpivirine and decreases viral fitness. J Acquir Immune Defic Syndr 2012; 59: Xu HT, Colby-Germinario S, Asahchop E, et al. Effect of mutations at position E138 in HIV-1 reverse transcriptase and their interactions with the M184I mutation on defining patterns of resistance to nonnucleoside reverse transcriptase inhibitors rilpivirine and etravirine. Antimicrob Agents Chemother 2013; 57: Accepted 11 January 2016; published online 13 January International Medical Press