Observational Study Medicine ® OPEN Integrase strand-transfer inhibitors for treatment of early HIV infection A case series ∗ Ramón Teira, MDa, , Mar Gutierrezb, Pepa Galindo, PhDc, Elisa Martínez, PhDd, Pepa Muñoz, PhDe, Belén de la Fuente, PhDf, Francisco Téllez, MDg, Marta Montero, PhDh Abstract This study evaluated whether the interval from the first clinic visit until the start of antiretroviral treatment (ART) was correlated with common parameters of immunological recovery among patients with early HIV infection (EHI). We reviewed the medical records of patients with EHI who started ART using integrase strand-transfer inhibitors (ISTIs) within the first 6 months after diagnosis. Simple linear regression analyses were performed to determine whether the interval from the first visit to the start of ART was correlated with 1-year changes in CD4+ cell count, CD8+ cell count, CD4+ percentage, and CD4+/CD8+ ratio. Fifty-three patients with probable or definite EHI started ART using ISTIs between April 2014 and August 2016. Forty-nine patients completed 1 year of follow-up, including 48 men. The routes of HIV transmission were 1 case of needle sharing, 5 cases of heterosexual activity, and 43 cases of men who had sex with men. None of the immunological recovery parameters were correlated with time to the start of ART (CD4+ cell count: R = .12, P = .42; CD8+ cell count: R = .107, P = .5; CD4+ percentage: R = .14, P = .34; CD4+/CD8+ ratio: R = .23, P = .14). Furthermore, subgroup sensitivity analyses failed to detect significant correlations based on definite or probable diagnoses, treatment using elvitegravir or dolutegravir, or the time from HIV diagnosis to ART initiation. This series of EHI cases indicate that using ART with ISTI-based regimens is efficacious and well-tolerated. However, earlier initiation of treatment was not significantly correlated with common parameters of immunological recovery. Abbreviations: AHI = acute HIV infection, ART = antiretroviral treatment, EHI = early HIV infection, ISTI = integrase strand-transfer inhibitors, VL = viral load. Keywords: acute HIV infection, dolutegravir, early HIV infection, elvitegravir, integrase inhibitors 1. Introduction consideration has been given to AHI since the first US guidelines were released in 1998,[2] while “recent infection” was only added The arbitrary 6-month interval after the first entry of HIV virions to the same epigraph in the February 2013 update, when through a mucosal surface or a percutaneous event is referred to antiretroviral treatment (ART) was already recommended for all as “early HIV infection” (EHI). This interval includes a HIV-infected individuals. The debate regarding the recommen- biologically well-defined short subperiod until antibodies are dation to start ART during “early infection” had subsequently first detected in the serum (acute HIV infection; AHI) and a waned, although special considerations from the 2013 update longer, complex, and biologically ambiguous subperiod until the have remained in place until the most recent update. In this 6-month mark that is called “recent infection”.[1] Special context, pathological data from small studies have revealed a distinct and dynamic sequence of overwhelming interactions Editor: Akhilanand Chaurasia. between the HIV and the patient’s immune system, which takes The authors report no conflicts of interest place during the first few days after infection and likely influence a Hospital Universitario de Sierrallana, Torrelavega, b Hospital de la Santa Creu i the course of the subsequent chronic infection and its prognosis. Sant Pau, Barcelona, c Hospital Clínico Universitario, Valencia, d Complejo Therapeutic intervention at the earliest possible time may Hospitalario Universitario de Albacete, Albacete, e Hospital Universitario de potentially affect this sequence, which is a hypothesis that is Basurto, Bilbao, f Hospital Universitario de Cabueñes, Gijón, g Hospital Universitario de Puerto Real, Puerto Real, h Hospital Universitario y Politécnico la supported by additional data from other small but intensive Fe, Valencia, Spain. studies.[3] Three randomized clinical trials have explored this ∗ Correspondence: Ramón Teira, Hospital Universitario de Sierrallana, Calle de general hypothesis in different scenarios (especially at different Manuel Teira Fernández s/n, 39300 Torrelavega, Spain time-defined stages after the transmission, ranging from the (e-mail: ramon.teira@scsalud.es). protocol-specified exclusion of AHI cases in the ACTG A5217 Copyright © 2019 the Author(s). Published by Wolters Kluwer Health, Inc. trial, to the inclusion of a majority of AHI cases in the Primo This is an open access article distributed under the terms of the Creative SHM trial, or the almost exclusive inclusion of cases of recent Commons Attribution-Non Commercial License 4.0 (CCBY-NC), where it is permissible to download, share, remix, transform, and buildup the work provided infection in the SPARTAC) and with different outcome variables it is properly cited. The work cannot be used commercially without permission (virological in the Primo and in the ACTG trials, and from the journal. immunological in the SPARTAC).[4–6] Therefore, we aimed to Medicine (2019) 98:35(e16866) determine whether, in an unselected clinical setting, the early Received: 16 January 2019 / Received in final form: 7 May 2019 / Accepted: 25 introduction of integrase strand-transfer inhibitors (ISTI) July 2019 provided an easily detectable immunological benefit. This http://dx.doi.org/10.1097/MD.0000000000016866 treatment has several theoretical advantages, including the 1 Teira et al. Medicine (2019) 98:35 Medicine elimination of resistance testing and the rapid control of viral 3. Results replication (based on reductions in the HIV viral load). Fifty-three patients fulfilled the inclusion criteria between April 2014 and August 2016. The exclusions were 1 patient who was 2. Methods prescribed dolutegravir but decided not to take it and 3 patients We reviewed the medical records of patients who attended who were lost to follow-up after 0 months, 2 months, and 7 participating clinics and fulfilled the 2 following criteria: months. The remaining 49 patients completed the 1-year follow- up, although only partial data were available for baseline CD4+ 1. A definite or probable diagnosis of “early infection”: count (46 patients), CD8+ count (42 patients), VL (47 patients),  Definite EHI based on (i) a negative serological test result and CD4+ percentage (48 patients). The vast majority of patients during the 6 months before the diagnosis or (ii) a were male (48 patients, 98%). The routes of HIV transmission simultaneous negative serological test result and a positive were needle sharing (1 patient, 2%), heterosexual relations (5 HIV-RNA test result. patients, 10.2%), and men who had sex with men (43 patients,  Probable EHI based on (i) no negative serological test result 87.8%). The median age was 37 (interquartile range: 31–39). A during the previous 6 months plus (ii) an epidemiological definite diagnosis of EHI was made in 29 cases (59.2%). The and clinical syndrome suggestive of acute HIV infection plus treatment drugs were elvitegravir (38 patients, 77.6%), dolute- an HIV-RNA plasma concentration (VL) of >100,000 gravir (9 patients, 18.4%), and raltegravir (2 patients, 4.1%). copies/mL. The median interval from the baseline visit to the start of ART 2. Started ART using an ISTI plus 2 nucleos(t)ide analogs during was 12 days (interquartile range: 0–18 days), and 16 patients the first 6 months after the diagnosis of HIV infection (based started ART on the day of the first visit. on the first positive result for anti-HIV antibodies or plasma Table 1 and Figure 1 show the data for the variables of interest HIV-RNA). and their scatterplots. No significant correlation with the interval The outcome variables were defined as the approximately 1- to the start of ART was observed for the changes in CD4+ cell year changes in the patients’ CD4+ cell count, CD8+ cell count, count (R = .12, P = .42), CD8+ cell count (R = .107, P = .5), CD4+ CD4+ percentage, and CD4+/CD8+ ratio. The baseline visit was cell percentage (R = .14, P = .34), and the CD4+/CD8+ ratio defined as the first clinic visit and the 1-year measurement was (R = .23, P = .14). Furthermore, the subgroup sensitivity analyses obtained at the visit that was closest to 1 year after the baseline failed to detect significant correlations for definite or probable visit. The main explanatory variable was defined as the interval diagnoses, treatment using elvitegravir or dolutegravir, or the (days) between the baseline visit and the start of ART. We also time from HIV diagnosis to the start of ART. The median change evaluated the interval between the diagnosis of HIV and the start in CD4+ cell count was +227/mL for patients who started ART at of ART. Moreover, we performed subgroup sensitivity analyses baseline and +343/mL for the remaining patients (P = .254, based on variables of theoretical interest (e.g., specific ISTI drugs Mann–Whitney U test). Finally, we found no correlation between and whether the diagnosis was definite or probable). Patients age and any of the 4 specified outcome variables. CD4+ counts were included if they continued ART for at least 1 year, regardless increased by a median of 328 cells/mL in the subgroup of patients of whether they were still receiving the original treatment with ages above the median and by 243 cells/mL in the subgroup regimen, as we aimed to evaluate the effect of the interval to the with ages below it (P = .45, Mann–Whitney U test) start of ART, rather than the effects of specific drugs or drug Five patients changed their prescribed treatments during the combinations on immunological recovery. All analyses involved first year of follow-up. Four patients had started receiving simple linear regression analysis after determining that the elvitegravir and switched treatments because of side effects (3 dataset fulfilled the requisite conditions. However, we failed to cases) or virologic failure (1 case). One patient had started detect significant correlations in the main and secondary analyses receiving raltegravir and switched treatment based on a change in and elected to not perform subsequent multivariate modeling. hospital policy. All 49 patients had VLs of <200 copies/mL after This study was approved by the Ethics Committee of Clinical 1 year of treatment, including 44 patients with <20 copies/mL Investigation of Cantabria. and 2 patients with 20–50 copies/mL. Table 1 Summary of baseline and outcome variables, overall and by subgroups of interest. Baseline CD4+ Baseline CD8+ Baseline CD4+ Baseline log10 Change CD4+ Change CD8+ Change CD4+ Change count /mm3 count /mm3 percentage Viral load count /mm3 count /mm3 percentage CD4+/CD8+ Global 439 (263–561) 981 (589–1573) 22 (14–25) 5.39 (4.88–6.05) 279 (134–498) –63 (–514–329) 10.5 (4.0–15.0) .37 (.15–.62) Definite (n = 29) 456 (290–568) 944 (531–1359) 24 (16–27) 5.38 (4.93–6.34) 327 (200–560) –57 (–501–322) 11.50 (4.75–19.00)) .46 (.23–.69) Probable (n = 20) 377 (233–528) 1031 (620–1991) 19 (12–23) 5.55 (4.74–5.89) 254 (102–393) –70 (–674–1138) 8.00 (3.00–14–00) .24 (.08–.40) Elvitegravir (n = 38) 454 (277–557) 890 (617–1626) 20 (14–25) 5.40 (4.97–5.96) 277 (156–393) –70 (–324–353) 11.50 (4.75–14.25) .32 (.14–.60) Dolutegravir (n = 9) 441 (300–712) 1196 (399–1716) 24 (19–31) 5.20 (4.62–6.40) 495 (194–735) 5 (–616–582) 7.00 (5.00–14.00) .39 (.18–.89) <500 CD4+ cells/mm3 336 (178–440) 758 (521–1353) 16.50 5.64 (4.99–6.18) 257 (142–481) –63 (–491–318) 12.00 (6.00–14.50) .34 (.16–.61) at BL (n = 29) (12.25–23–75) >500 CD4+ cells/mm3 610 (557–811) 1138 (623–1805) 26.00 5.09 (4.51–6.10) 372 (90–575) –60 (–711–572) 12.00 (7.00–14.00) 0,38 (.06–.67) at BL (n = 17) (22.75–31.25) All values are medians and interquartile range in brackets. 2 Teira et al. Medicine (2019) 98:35 www.md-journal.com Figure 1. Scatterplots of changes of CD4+ cell counts (A), CD8+ cell counts (B), CD4+/CD8+ ratios (C) and CD4+ cell percentages (D) against time in days since the first visit to the clinic. 4. Discussion histopathological data, it is not reported whether the median This series of EHI cases revealed that ART using ISTI-based CD4+ peripheral cell counts changes differed across Fiebig-stage regimens was efficacious, well-tolerated, and led to high levels based subgroups of patients. This is important, as the peripheral of treatment adherence and retention. However, we did not compartment retains greater capacity for recovery, based on detect any benefit from earlier treatment initiation based on numerous clinical trials involving patients starting ART in the improvements in the common parameters of immunological chronic phase. Another well-characterized cohort of patients recovery. It is important to note that these patients were treated with AHI in San Diego has revealed that the likelihood of as part of standard clinical care and away from any peripheral immunologic recovery improves when ART is started experimental milieu. as early as possible, although the strongest driver of this recovery Advances in our understanding of the unique pathogenetic was the reduced immunological deterioration before the start of events during AHI and EHI have provided strong support for the treatment.[12] Based on those findings, it is unsurprising that our earliest possible initiation of ART.[1] However, the available patients who started ART when their CD4+ cell counts were clinical evidence is less compelling, as most studies have been >500/mL reached even higher levels after 1 year of treatment, restricted to small heterogeneous case series that revealed very with a median value that was within the range defined in the San modest benefits based on second-line surrogate immunological Diego study as “recovered”. However, the absolute magnitudes markers[7,8] and virological markers.[9,10] Data from a well- of the increases in CD4+ cell count, CD4+ percentage, and CD4 characterized cohort of patients with AHI in Thailand indicates +/CD8+ ratio were similar between our subgroups of patients that the transition to Fiebig stages IV/V involves irreversible with starting CD4+ cell counts of ≥500/mL and <500/mL, and damage to the central immune system compartment.[11] the magnitude of these increases were not correlated with the However, despite the detailed array of immunological and intervals from the diagnosis or first visit to the start of ART. 3 Teira et al. Medicine (2019) 98:35 Medicine Nevertheless, it is unrealistic to target early-stage population- Supervision: Ramon Teira Cobo, Pepa Galindo, Elisa Martínez, wide interventions, based on the methods of the Thailand and Pepa Muñoz, Francisco Téllez, Marta Montero. San Diego studies, and a real-world approach must target some Validation: Pepa Muñoz. compromise between a highly proactive research setting and real Writing – original draft: Ramon Teira Cobo. clinical practice. The results of the randomized SPARTAC trial, Writing – review & editing: Ramon Teira Cobo, Mar Gutierrez, which recruited EHI cases with broad time-from-infection Pepa Galindo, Elisa Martínez, Pepa Muñoz, Belén de la intervals, suggest that only very modest benefits were observed Fuente, Francisco Téllez, Marta Montero. in delayed immunological deterioration (the primary endpoint) among patients who were randomized <12 weeks after References seroconversion.[5] Our results do not contradict the very plausible interpretation that very early initiation of ART during AHI can [1] Panel on antiretroviral guidelines for adults and adolescents. Guidelines for the use of antiretroviral agents in adults and adolescents living with have a significant beneficial effect on immunological preservation HIV. Department of Health and Human Services. Available at http:// or recovery. However, based on our data from real clinical www.aidsinfo.nih.gov/ContentFiles/AdultAdolescentGL.pdf [access practice, it appears that physicians are not likely to encounter date April 23, 2018]. these patients during the “very early” window for starting ART, [2] Centers for Disease Control Prevention. Report of the NIH Panel to Define Principles of Therapy of HIV Infection and Guidelines for the Use and that the treatment options at the initial encounter can likely of Antiretroviral Agents in HIV-Infected Adults and Adolescents. be based on those for chronic infection. MMWR, 1998; 47(No. RR-5): (1–83) Our study has some limitations derived from its retrospective [3] Cohen MS, Shaw GM, McMichael AJ, et al. Acute HIV-1 infection. N nature. However, we believe that the common sources of bias in Engl J Med 2011;364:1943–54. observational studies are minimal in ours and that they are [4] Hogan CM, DeGruttola V, Sun X, et al. The setpoint study (ACTG A5217): effect of immediate versus deferred antiretroviral therapy on unlikely to have influenced the overall conclusions. All the virologic set point in recently HIV-1 infected individuals. J Infect Dis participating clinics share a software application for HIV clinical 2012;205:87–96. care (AC&H, AMVACH, Suances Spain) in which all patients [5] Grijsen ML, Steingrover R, Wit FNM, et al. No treatment versus 24 or 60 been cared-for at each clinic are registered and their data weeks of antiretroviral treatment during primary HIV infection: the randomized primo-SHM trial. PLoS Med 2012;9:e1001196. prospectively recorded after their informed consent has been [6] The SPARTAC trial investigatorsShort-course antiretroviral therapy in given. Therefore, selection and information biases should be of primary HIV infection. N Engl J Med 2013;368:207–17. no concern. As for confounding, it is unlikely that an unexpected [7] Guadalupe M, Reay E, Sankaran S, et al. Severe CD4+T-cell depletion in and uncontrolled variable that could have influenced the decision gut lymphoid tissue during primary human immunodeficiency virus type or time to initiate ART would have had an effect on the outcome 1 infection and substantial delay in restoration following highly active antiretroviral therapy. J Virol 2003;77:11708–17. variables selected for our study. [8] Malhotra U, Berrey MM, Huang Y, et al. Effect of combination antiretroviral therapy on T-cell immunity in acute human immunodefi- ciency virus type 1 infection. J Infect Dis 2000;181:121–31. Author contributions [9] Strain MC, Little SJ, Daar ES, et al. Effect of treatment, during primary Conceptualization: Ramon Teira Cobo, Elisa Martínez, Pepa infection, on establishment and clearance of cellular reservoirs of HIV-1. J Infect Dis 2005;191:1410–8. Muñoz, Belén de la Fuente, Marta Montero. [10] Hamlyn E, Ewings FM, Porter K, et al. Plasma HIV viral rebound Data curation: Ramon Teira Cobo, Mar Gutierrez, Pepa following protocol-indicated cessation of ART commenced in primary Galindo, Elisa Martínez, Pepa Muñoz, Belén de la Fuente, and chronic HIV infection. PLoS One 2012;7:e43754. Francisco Téllez, Marta Montero. [11] Schuetz A, Deleage C, Sereti I, et al. Initiation of ART during early acute Formal analysis: Ramon Teira Cobo. HIV infection preserves mucosal Th17 function and reverses HIV-related immune activation. PLoS Pathog 2014;10:e1004543. Investigation: Mar Gutierrez. [12] Le T, Wright EJ, Smith DM, et al. Enhanced CD4+ T-cell recovery Methodology: Pepa Muñoz, Francisco Téllez, Marta Montero. with earlier HIV-1 antiretroviral therapy. N Engl J Med 2013;368: Project administration: Ramon Teira Cobo, Marta Montero. 218–30. 4