HIV Drug Resistance Among Individuals Failing Tenofovir/Lamivudine and Dolutegravir First Line Regimen in Brazil

Brazil was the first middle-income country to provide free and universal access to antiretroviral drugs to HIV infected individuals. Since 2014 local guidelines recommend that all HIV infected individuals be started on therapy regardless of CD4 count. Since January 2017, all patients are started on a DTG containing triple regimen. As of November 2018, 170,000 individuals were receiving DTG through the public health system. It is a public health priority to evaluate the risk of virologic failure and the subsequent development of INSTI resistance in these real-life settings. Our preliminary data from Brazil indicated a high virologic failure rate of 8% after 18 months of treatment TL+D. Our central hypothesis is that TDR may be associated and contribute to virologic failure with DTG in clinical practice. To test this central hypothesis, we will identify PLWH failing DTG containing regimens in Brazil. The insights generated with these studies will contribute to a more effective use of second generation INSTI in the future.

SPECIFIC AIMS We are on the verge of the global rollout of the second-generation integrase strand-transfer inhibitor (INSTI) dolutegravir (DTG), as the World Health Organization (WHO) now recommends the combination of DTG + 2 nucleoside reverse transcriptase inhibitors (NRTIs) as first- and second-line therapy. As a result, millions of people living with HIV (PLWH) will soon receive DTG. In clinical trials, which mainly included people infected with HIV-1 subtype B, the combination of DTG + 2 NRTIs demonstrated very high efficacy and acquired drug resistance was absent (first-line treatment) or very rare (second-line treatment) in the event of virologic failure. It is not clear if this high efficacy and lack of acquired drug resistance can be extrapolated to clinical practice in low- and middle-income countries, where the majority of PLWH live and where most infections are due to non-B subtypes. Furthermore, in the pivotal studies for initial treatment with DTG, candidates harboring virus resistant to NRTIs met the exclusion criteria of these studies. Although DTG failure in first line regimens in clinical trials reveal the absence of integrase resistance, it is conceivable that mutations at other HIV genomic regions such as 5´PPT (nef) could contribute to DTG lack of efficacy. Brazil was the first middle-income country to provide free and universal access to antiretroviral drugs to HIV infected individuals. Since 2014 local guidelines recommend that all HIV infected individuals be started on therapy regardless of CD4 count. Since January 2017, all patients are started on a DTG containing triple regimen. As of November 2018, 170,000 individuals were receiving DTG through the public health system. It is a public health priority to evaluate the risk of virologic failure and the subsequent development of INSTI resistance in these real-life settings. Our preliminary data from Brazil indicated a high virologic failure rate of 8% after 18 months of treatment TL+D. Moreover, acquired drug resistance was observed in first-line virologic failures (15 out of 84 investigated individuals). A relatively high rate of transmitted drug resistance (TDR) was also observed among those 84 individuals (15.6%) Our central hypothesis is that TDR may be associated and contribute to virologic failure with DTG in clinical practice. To test this central hypothesis, we will identify PLWH failing DTG containing regimens in Brazil, a model country for large-scale DTG implementation where various HIV subtypes co-circulate. The insights generated with these studies will contribute to a more effective use of second generation INSTI in the future. The specific aims of this proposal are as follows: 1. Investigate the influence of Transmitted Drug resistance, HIV clade profile and immunological and virological features among individuals failing first line regimen with Tenofovir/3TC + Dolutegravir after 24 weeks of treatment in Brazil. 2. Determine the genotypic resistance profile among individuals failing first line regimen with Tenofovir/3TC + DTG after 24 weeks of treatment in Brazil. 3. Determine which changes in the 3'-PPT are observed in viruses from patients experiencing TL+D failure and assess if this novel resistance pathway contributes to acquired drug resistance in clinical practice. The results from the proposed study can be used to optimize care of PLWH by improving treatment guidelines and drug resistance interpretation algorithms. RESEARCH PLAN SIGNIFICANCE The World Health Organization (WHO) recently changed the HIV treatment guidelines and now recommends the second-generation integrase strand-transfer inhibitor (INSTI) dolutegravir (DTG) + 2 nucleoside reverse transcriptase inhibitors (NRTIs) as first- and second-line therapy for people living with HIV (PLWH). As a result, millions of people living with HIV (PLWH) will soon receive DTG in low- and middle-income countries where the majority of PLWH live. Considerations of the WHO for the transition from efavirenz-based regimens to DTG-based regimens are the high efficacy of DTG-based regimens, the relatively low costs of producing generic DTG formulations, and the low risk for baseline DTG resistance and acquired DTG resistance. In clinical trials, DTG + 2 nucleoside reverse transcriptase inhibitors (NRTIs) demonstrated high efficacy in treatment naïve patients with viral suppression (\<50 c/mL) at week 48 in 90% (FLAMINGO), 88% (SPRING-2), 88% (SINGLE) and 82% (ARIA) of the participants with only 1-2% of data not in the window of \<50 c/mL. Of major importance, these trials showed that in the event of virologic failure no known resistance associated mutations (RAMs) could be detected in the integrase and reverse transcriptase gene. Similarly, DTG + 2 NRTIs showed high efficacy in treatment experienced INSTI naïve PLWH and acquired drug resistance was rare5. However, clinical trial participants are selected and closely monitored, including extra doctor's visits, more frequent viral load measurements and genotypic resistance analyses, and are switched more rapidly to other combination antiretroviral therapy (cART) in the event of virologic failure. It is not clear if this high efficacy and lack of acquired drug resistance will carry on in clinical practice in low- and middle-income countries. Therefore, it is now a public health priority to evaluate the risk of virologic failure and subsequent development of INSTI resistance in real life settings in these countries. We propose to perform such studies in Brazil, a country with 860,000 PLWH, as this country already implemented DTG regimens and thus can serve as a model for other low- and middle-income countries. Preliminary data on efficacy and acquired drug resistance of DTG + 2 NRTIs in Brazil. From January 2017, the recommended first-line regimen in Brazil is the fixed dose combination of generic tenofovir 300 mg plus lamivudine 300 mg combined with DTG 50 mg (TL+D). In 2017, 54,175 PLWH started TL+D as first-line regimen in Brazil, and 26,417 PLWH were switched from raltegravir-containing cART to TL+D, comprising 77% of individuals initiating cART in this country. 8% of individuals receiving TL+D had a viral load \>50 c/mL after 18 months of treatment which is higher than expected based on clinical trial results. In general, from the 581,064 individuals on treatment in Brazil as of December 2017, 85% presented with viral loads \< 50 c/mL (Source: MS/SVS/Departamento de IST, Aids e Hepatites Virais, Brasil). As of November 2018, 170,000 PLWH receive DTG in Brazil. Among 84 individuals who experienced virologic failure during first-line treatment with TL+D, 11 presented low level viremia (50-500 c/mL), In five patients (5.9%) major resistance associated mutations were detected at IN: 2 patients with R263Q, one with G118R, one with E138A, and one with R263R/K in IN plus K70E+M184V in RT. Six (7.1%) additional individuals presented minor IN resistance mutations: L74I/M (2 cases); G140R+G163R; V151A; V151I; T97A; E157Q; and M50I. The prevalence of the polymorphic IN mutations L101I and T124A, which are in vitro pathway for resistance, were 53.5% and 46.4%, respectively, significantly higher than in Brazilian INSTI naïve patients. Eight (9.5%) patients presented T+L RAM in the RT, including 3 with K70E+M184V. Mutations not related to TL+D (e.g., TAMs, NNRTI and PI) were present in 25.6% of the patients, significantly higher than in a representative national sample of 1,568 antiretroviral naïve patients we have tested. Therefore, the detection of IN and RT RAMs upon virologic failure indicates that selected resistance to first-line TL+D is more common than what has been reported in clinical trials. The apparent association between TL+D failure in Brazil and TDR should be better evaluated, as well as the role of IN polymorphic mutations that can emerge during DTG failure. Together, these results warrant further investigations of virologic failure and subsequent acquired INSTI resistance in this setting. Among 84 individuals who experienced virologic failure during first-line treatment with TL+D, 11 presented low level viremia (50-500 c/mL), In five patients (5.9%) major resistance associated mutations were detected at IN: 2 patients with R263Q, one with G118R, one with E138A, and one with R263R/K in IN plus K70E+M184V in RT. Six (7.1%) additional individuals presented minor IN resistance mutations: L74I/M (2 cases); G140R+G163R; V151A; V151I; T97A; E157Q; and M50I. The prevalence of the polymorphic IN mutations L101I and T124A, which are in vitro pathway for resistance, were 53.5% and 46.4%, respectively, significantly higher than in Brazilian INSTI naïve patients. Eight (9.5%) patients presented T+L RAM in the RT, including 3 with K70E+M184V. Mutations not related to TL+D (e.g., TAMs, NNRTI and PI) were present in 25.6% of the patients, significantly higher than in a representative national sample of 1,568 antiretroviral naïve patients we have tested. Therefore, the detection of IN and RT RAMs upon virologic failure indicates that selected resistance to first-line TL+D is more common than what has been reported in clinical trials. The apparent association between TL+D failure in Brazil and TDR should be better evaluated, as well as the role of IN polymorphic mutations that can emerge during DTG failure. Together, these results warrant further investigations of virologic failure and subsequent acquired INSTI resistance in this setting. APPROACH Study design: This is a prospective nested case control study comparing the baseline HIV profile of individuals experiencing virologic failure to TL+D regimen after 24 weeks of treatment initiation (cases) to randomly selected 2:1 control individuals with viral load bellow detection limits 24 weeks after treatment initiation. Identification of PLWH who fail DTG in Brazil PLWH will be included at 5 large clinical sites in Brazil: one collection site will be in the far South of Brazil where 85% of strains are subtype C, the northeast of Brazil where 30% of samples are subtype F, the city of Santos, South-east of Brazil where 50% of strains are BF recombinant forms (CRF\_28, CRF\_29, and unique recombinant forms), Sao Paulo and Rio de Janeiro. At each site a research nurse will identify PLWH that will initiate a TL + D regimen, and baseline samples will be collected and send to the Central Laboratory (R Diaz, PI) after an informal consent. At this time, epidemiological, virological and immunological data will be collected from each patient, and stored in a Central database. We expect to recruit 2,500 patients initiating antiretroviral therapy in a 24-week period. Those recruited patients will be oriented to perform a plasma viral load after 12 and 24 weeks of recruitment. We expect that by week 24, 8% of recruited individuals will reach the primary end point of the study which is the antiretroviral virologic failure, and Blood samples from these individuals will be also sent to the Diaz laboratory for genotypic resistance testing. Cases and controls will be defined as: * Cases: HIV infected individuals with subtype C, subtype F, or BF recombinant forms who use TL+D for at least 24 weeks and have a confirmed plasma viral load \> 200 copies/mL. We expect to include samples from 200 individuals experiencing virologic failure as defined above. * Controls: HIV infected individuals with subtype C, subtype F, or BF recombinant forms who use TL+D for at least 24 weeks and have a confirmed plasma viral load \< 50 copies/mL. We will randomly select 400 samples from control individuals in order to characterize by genomic sequencing the HIV strains present at baseline, before treatment initiation. Based on the total number of PLWH receiving DTG containing cART at these three sites, our preliminary results showing a virologic failure rate of 8%, and assuming that 10% will be lost to follow-up, we will recruit 2,500 PLWH in order to obtain 200 individuals experiencing virologic failure. Based on our preliminary results, it is safe to conservatively assume that 10% of these patients will harbor TDR at baseline. Coordination of inclusion of PLWH at the clinical sites: In Brazil, all HIV infected individuals are entitled to free access to antiretroviral therapy. As of November 2018, 170,000 individuals were receiving DTG through the public health system. Participants for the present study will be recruited in 5 large clinical units, all highly experienced in the conduct of multicenter studies. In each site a specially dedicated research nurse will be hired. He/she will be responsible for identifying, recruiting, consenting, and ensuring follow-up of study participants. Database management: A private repository will be set up on GitHub (https://github.com/) that will give access to remote researchers (physicians and nurses) and to Central coordination for data management and analysis at Diaz´s Retrovirology Laboratory. We will prepare a short form in Excel format for the remote researchers to use to enter the data and serve as an electronic Case Report Form. For every patient, remote researchers will enter the needed data, and all the entries will be combined into a database and bring the individuals files down onto our computers at the Retrovirology Lab to be stored offline. The analyses will be performed by Dr. James R Hunter from the Retrovirology Lab. Anticipated results and alternative approaches: We do not anticipate difficulties in recruiting the expected number of cases and controls. Nonetheless, expanding to other sites should not represent any major challenge. This is because the laboratories in each of the five sites are responsible for performing viral loads for several primary care facilities in their respective cities. Thus, the same procedures could be expanded to recruit participants from primary care facilities located in the proximity of the main hospitals.