Impact of genotypic susceptibility score on cART outcomes during primary HIV infection
A complete list of clinical centers belonging to the INACTION study group is showed at the end of the manuscript.
This manuscript is dedicated to Andrea De Luca M.D., a great friend, physician, mentor and scientist.
Abstract
To assess the impact of genotypic susceptibility score (GSS) on combined antiretroviral therapy (cART) outcomes during primary HIV infection (PHI) we retrospectively enrolled patients with PHI diagnosed between 2008 and 2015 at 9/24 Italian Network ACuTe HIV InfectiON centers. One hundred-seventy-six patients were enrolled. Of these, 55 (32.9%) patients started with more than three drugs and 11 (7.2%) started with a GSS < 3. Regimen's GSS (per 1 point increase) (adjusted odds ratio [aOR], 4.82; 95% confidence interval [CI], 1.62-14.28; P = .005) and baseline HIV-RNA (per 1 log10 increase) (aOR, 2.02; 95% CI, 1.09-3.73; P = .025) resulted associated with early cART initiation. In conclusion, regimen's GSS resulted to be associated to the time to cART initiation during PHI.
HIGHLIGHTS
-
Italian Network ACuTe HIV InfectiON is an Italian network with the aim of studying primary HIV infection (PHI).
-
Regimen's genotypic susceptibility score (GSS) resulted to be associated to the time to combined antiretroviral therapy initiation during PHI.
-
No role of GSS was observed in the probability of virological control and first regimen discontinuation.
-
A lower GSS score was related to a X4 tropism.
1 INTRODUCTION
Primary HIV infection (PHI) is a crucial phase of HIV infection with a limited window of opportunity.1 A prompt antiretroviral treatment (combined antiretroviral therapy [cART]) initiation it is able to lead to a reduction of HIV reservoir, to a better immunological recovery and to a reduction of onward HIV transmission.2, 3 Consequently, it is mandatory to start cART as soon as possible according to the current guidelines during PHI.4 Nevertheless, the assessment of drug resistance before treatment initiation is crucial because suboptimal cART could lead to increased risk of virological failure.5 The presence of transmitted drug resistance (TDR) could be a concern not only in patients with chronic HIV infection but more so in patients with PHI. The possibility that patients with PHI could be burdened by a higher rate of TDR when compared to chronically infected patients has been postulated.6 Moreover, in the study by Wittkop et al5 it has been observed how the presence of TDR which affects the starting regimen was associated with an increased risk of virological failure, when compared with TDR which did not affect the efficacy of the regimen.
The choice of the cART in the absence of the result of the resistance test is based on regimens with high genetic barrier. In fact, there is a possibility of transmission of viral resistant strains during PHI7 although it appears to be declining in recent years after the introduction of the integrase inhibitors.8 During PHI, even though R5 viruses predominate, also X4 tropic viruses can be transmitted but their real impact on disease progression remains unclear.9
Genotypic susceptibility score (GSS) of the antiretroviral regimen built in accordance to Stanford HIV database genotypic resistance interpretation system could be a helpful tool to assess the activity of the single antiretroviral drugs composing an antiretroviral regimen.10 Moreover, GSS has been demonstrated to predict the virological outcome of antiretroviral regimens.11, 12
Aim of this study was to assess the impact of baseline GSS on early cART initiation, time to virological suppression and time to first regimen discontinuation in HIV-positive patients diagnosed during PHI.
2 MATERIALS AND METHODS
Patients with PHI diagnosed between 2008 and 2015 at 9/24 Italian Network ACuTe HIV InfectiON (INACTION) centers were enrolled in a retrospective cohort study. All patients with at least one of RT/PR/INI resistance test performed on plasma samples by Sanger technique were analyzed using the Stanford algorithm (version 8.4) and included in the final model. Estimated scores of the total antiretroviral activity of a cART regimen was analyzed with GSS. GSS drug categorization was as follow: 1 point to each drug categorized as susceptible or potential low-level resistance or low-level resistance; 0.5 point to intermediate resistance category; 0 point to the high-level resistance category. The total GSS of the regimen represents the sum of all regimen's GSS components.10 In a subset of 60 patients GP120 V3 sequence was analyzed using the Geno2pheno algorithm. Mixed X4/DM viruses were pooled with X4 tropic viruses. Patients were included in the study if they presented (a) at least one criterion for HIV infection diagnosis and concomitantly (b) at least one criterion for acute HIV infection: (a) (1) p24 reactivity, (2) HIV-RNA and/or HIV-DNA polymerase chain reaction detectable viral load; (b) (1) negative HIV-antibodies or low title reactivity at enzyme-linked immunosorbent assay third (or further) generation assay, (2) less than three positive bands at Western blot test including p24, go160/gp120, or gp41, (3) less than two positive bands at Recombinant ImmunoBlot Assay test including reactivity for gp41. Patients with Fiebig stages I-V were included.13 Patients starting cART within 3 months since PHI diagnosis were classified as early cART group. Virological suppression was defined as first HIV-RNA < 50 cp/mL from the introduction of cART.
The primary end-point was to assess the impact of GSS on early cART initiation. The secondary end-points were to assess the impact of GSS on time to virological suppression and time to first regimen discontinuation.
Logistic regression analysis was performed to identify factors associated to early cART initiation. Cox regression analysis was performed to determine predictors of time to virological suppression after cART initiation and time to first-line regimen discontinuation.
P < .05 was considered statistically significant. All statistical analyses were performed using SPSS version 13.0 software package (SPSS Inc, Chicago, IL).
The study was approved by the Ethics Committee (EC) of Monza-Brianza for San Gerardo Hospital (coordinating center) on June 2014 and then by the EC of each participating center. Written informed consent was obtained from all enrolled subjects.
3 RESULTS
A total of 176 patients were enrolled. Baseline characteristics are shown in Table 1. During the study period 167 (94.9%) of the patients started a cART regimen, of whom 125 (74.9%) patients started during the first 3 months since HIV diagnosis. All patients who started an antiretroviral regimen during the time of observation had at least three drugs in their cART regimen. Of these, 55 (32.9%) patients started with more than three drugs. The GSS was available for 153 patients, of whom 11 (7.2%) started with a GSS < 3. Patients with a X4 tropic virus have lower GSS score than patients infected by a R5 tropic virus (3 vs 3.5; P = .012).
| Population characteristics (n = 176) | N (%) or median (IQR) |
|---|---|
| Age, years | 37 (28-45) |
| Male sex | 153 (86.9) |
| Risk factor | |
| Heterosexual | 54 (30.7) |
| Homo/bisexual | 112 (63.6) |
| IDU | 6 (3.4) |
| Other/unknown | 4 (2.3) |
| Months from last negative HIV test | 7.9 (3.3-21.2) |
| Fiebig stage | |
| I | 6 (3.4) |
| II | 22 (12.5) |
| III | 28 (15.9) |
| IV | 59 (33.5) |
| V | 61 (34.7) |
| CDC stage | |
| A | 133 (75.6) |
| B | 15 (8.5) |
| C | 1 (0.6) |
| Unknown | 27 (15.3) |
| CD4, cells/mm3 | 453 (329-603) |
| HIV-RNA, log10 | 5.76 (5.07-6.42) |
| First cART regimen (n = 167) including: | |
| NRTI | 164/167 (98.2) |
| NNRTI | 29/167 (17.4) |
| PI | 120/167 (71.9) |
| INSTI | 61/167 (36.5) |
| Entry inhibitor | 23/167 (13.8) |
| Early cART initiation (within 3 mo from the diagnosis) | 125/167 (74.9) |
| First cART regimen >3 drugs | 55/167 (32.9) |
| Available V3 loop sequence | 60 (34.1) |
| X4 | 17/60 (28.3) |
| R5 | 43/60 (71.7) |
| FPR% | 27.7 (9.4-61.0) |
| FPR > 60 | 16/60 (26.7) |
| Available integrase sequences | 50 (28.4) |
| % resistance | 0/50 (0) |
| Available RT-PR sequences | 165 (93.8) |
| GSS first regimen (n = 153) | 3 (3-4) |
| GSS first regimen < 3 | 11/153 (7.2) |
- Note: During the study period 167 (94.9%) of the patients started a cART regimen, of whom 153 had available sequences to assess the GSS.
- Abbreviations: cART, combined antiretroviral regimen; CD, cluster of differentiation; CDC, Center for Disease and Control; FPR, false positive rate; GSS, genotypic susceptibility score; IDU, intravenous drug users; IQR, interquartile range; N, number; PR, protease; R5, CCR5 tropic virus; RT, reverse transcriptase; X4, CXCR4 tropic virus.
Regimen's GSS (per 1 point increase) (adjusted odds ratio [aOR], 4.82; 95% confidence interval [CI], 1.62-14.28; P = .005), baseline HIV-RNA (per 1 log10 increase) (aOR, 2.02; 95% CI, 1.09-3.73; P = .025) and the calendar year 2012-2015 vs 2008-2011 (aOR, 9.38; 95% CI, 3.40-25.89; P < .001) resulted associated with early cART initiation (Table 2). No association was observed with regimen GSS and time to virological suppression during PHI (Table S1). Regimen GSS (per 1 point increase) resulted associated to first regimen discontinuation in the univariate analysis [hazard ratio, 2.21; 95% CI, 1.44-3.37; P < .001), but the significance was lost in the multivariate model (adjusted hazard ratio, 1.15; 95% CI, 0.59-2.21; P = .683) (Table S2).
| Univariate analysis | Multivariate analysis | |||
|---|---|---|---|---|
| OR (95% CI) | P value | aOR (95% CI) | P value | |
| Age (per 10-y increase), y | 0.97 (0.70-1.35) | .872 | ||
| Sex, male vs female | 1.73 (0.67-4.42) | .256 | ||
| Risk factor | ||||
| Heterosexual | Ref | .021 | Ref | .275 |
| Homo/bisexual | 2.39 (1.14-4.99) | .646 | 1.81 (0.63-5.22) | .499 |
| IDU | 1.73 (0.17-17.80) | .911 | 2.43 (1.19-31.98) | .352 |
| Other/unknown | 1.15 (0.10-13.56) | 0.19 (0.01-6.38) | ||
| Months from last negative HIV test, (per 1-mo increase) | 1.01 (0.99-1.02) | .619 | ||
| Calendar year at diagnosis: 2012-2015 vs 2008-2011 | 6.53 (3.05-14.00) | <.001 | 9.38 (3.40-25.89) | <.001 |
| Fiebig stage, IV-V vs I-III | 1.10 (0.52-2.32) | .797 | ||
| Symptomatic acute HIV infection | 1.99 (0.72-5.52) | .186 | ||
| STD | 0.83 (0.25-2.79) | .758 | ||
| Baseline laboratory results | ||||
| CD4 (per 100 cell/mm3 increase), cells/mm3 | 0.89 (0.76-1.05) | .171 | 0.97 (0.78-1.21) | .805 |
| CD8 (per 1 cell/mm3 increase), cells/mm3 | 0.99 (0.97-1.02) | .552 | ||
| CD4/CD8 > 1 | 0.99 (0.26-3.80) | .993 | ||
| HIV-RNA, per 1 log10 increase | 1.98 (0.31-2.97) | .001 | 2.02 (1.09-3.73) | .025 |
| X4 vs R5 | 1.01 (0.18-5.81) | .988 | ||
| FPR > 60 vs FPR ≤ 60 | 0.92 (0.16-5.30) | .927 | ||
| First regimen's GSS, per 1 point increase | 5.20 (2.21-12.25) | <.001 | 4.82 (1.62-14.28) | .005 |
| First regimen's GSS, <3 vs ≥3 | 0.62 (0.17-2.23) | .461 | ||
- Note: The multivariate model was built by entering in the model all variable with a P < .05 in the univariate analysis and adjusting for clinically significant variables (baseline CD4 cell count). The final model was built on the 153 patients with an available GSS.
- Abbreviations: aOR, adjusted odds ratio; cART, combined antiretroviral regimen; CD, cluster of differentiation; CI, confidence interval; FPR, false positive rate; GSS, genotypic susceptibility score; IDU, intravenous drug users; OR, odds ratio; R5, CCR5 tropic; STD, sexually transmitted diseases; vs, versus; X4, CXCR4 tropic.
4 DISCUSSION
In our study we observed a correlation between GSS and the time of cART initiation in patients with PHI. This finding can partially be explained by the high percentage (32.9%) of patients who started the cART with more than three drugs. Even though there are no current evidence favouring the start of a cART with more than three drugs,4, 14 many clinicians still use to start the treatment with a multitarget drug regimen waiting for the result of the resistance test and after that simplifying the regimen. It is possible that a multitarget drug regimen could have been chosen by the physician in charge due to the fear of TDR reported in patients acutely infected when compared with chronically infected ones.6, 7 Even though, no association between first regimen GSS during PHI and time to virological control was observed in our study. A promptly revision of the cART is mandatory after the result of the resistance test to avoid the exposure to suboptimal antiretroviral regimen during PHI.
X4 tropic viruses had a GSS lower when compared with R5 tropic viruses. In a previous study an increased risk of harboring drug resistant viruses was not observed for X4 tropic viruses when compared with R5 tropic viruses.15 Whereas, we did not find any association between viral tropism and cART investigated outcomes, thus the impact on HIV disease progression remains to be clarified.16
Our study accounts for several limitations. First, the retrospective nature of the study could have affected the analysis due to missing data. Second, the integrase sequences were available only for 50 patients. Nevertheless, we did not observe any integrase region resistance in the 50 patients tested in our cohort; accordingly, for the GSS calculation integrase inhibitors were considered as fully susceptible as supported by literature until 2015.17 This assumption could be not applicable to more recent years. In fact, due to the widespread use of integrase inhibitors, transmitted resistance during PHI could be observed.18 Third, only 60 patients had an available V3 loop sequence. Consequently, all 23 patients treated with maraviroc had an available V3 loop sequence, thus we were able to calculate their GSS.
According to our findings, regimen's GSS resulted to be associated to the time to cART initiation during PHI. Together with an expected better outcome with an earlier cART, the GSS evaluation by clinicians has been used before cART start, at least during 2008-2015.
ACKNOWLEDGMENT
We thank all the clinicians and patients involved in the INACTION study.
CONFLICT OF INTERESTS
AG and IdB do not declare any competing interest. MF received speakers’ honoraria and support for travel to meetings from Bristol-Myers Squibb, Gilead, Merck Sharp & Dohme, ViiV Healthcare, Janssen-Cilag, and fees for attending advisory boards from Bristol-Myers Squibb and Gilead. SN received travel grants and fees for presentations from ViiV, Gilead and Janssen. EF received consultancy fees and speaker's honoraria from, Gilead, ViiV Healthcare, Janssen-Cilag and Merck Sharp & Dohme. BMC received grants, travel grants and speaker's honoraria from Abbvie, Bristol-Myers Squibb, Gilead, ViiV, Janssen-Cilag and Merck Sharp & Dohme. GM received personal fees from Gilead Sciences and ViiV Healthcare. CM participated to Advisory boards supported by Gilead, ViiV, BMS, janssen, MSD, Angelini, Abbvie, received grants from ViiV, MSD, Gilead, Janssen and participated to speaker's bureau supported by Gilead. AA received grants, personal fees and nonfinancial support from Gilead Sciences, ViiV Healthcare and Bristol-Myers Squibb, grants and personal fees from Janssen-Cilag, personal fees from Merck Sharp and Dohme and personal fees and nonfinancial support from Abbvie. GE received consultancy payments and speaking fee from Bristol-Myers Squibb, Gilead, ViiV Healthcare, Merck Sharp Dohme, ABBvie and Janssen. GM received consultancy and/or speakers’ fees from Abbott, Bristol-Myers Squibb, Gilead Sciences, Janssen, Merck Sharp & Dohme and ViiV. AB received consultancy payments, speaking fees, travel and research grants from Bristol-Myers Squibb, Gilead, ViiV Healthcare, Merck Sharp Dohme, ABBvie and Janssen. AM received consultancy payments, speaking fees, travel and research grants from Bristol-Myers Squibb, Gilead, ViiV Healthcare, Merck Sharp Dohme, ABBvie and Janssen. SR received consultancy payments, speaking fees, travel and research grants from Bristol-Myers Squibb, Gilead, ViiV Healthcare, Merck Sharp Dohme, ABBvie and Janssen.
INACTION STUDY GROUP PARTICIPATING PHYSICIANS AND CENTERS
Andrea Gori, Antonio Muscatello, Alessandra Bandera (Ospedale Maggiore Policlinico Milano); Giuseppe Tambussi, Silvia Nozza, Marco Ripa, Raffaele Dell'Acqua (H San Raffaele Milano); Andrea Antinori, Carmela Pinnetti (INMI Spallanzani Roma); Andrea Calcagno, Giancarlo Orofino, Ilaria De Benedetto, Micol Ferrara (Torino); Cristina Mussini, Vanni Borghi, Federica Carli (Modena); Benedetto Maurizio Celesia (Catania); Lucio Cosco, Carlo Torti (Catanzaro); Gabriella d'Ettorre (Umberto I Roma); Antonio Di Biagio (Genova); Emanuele Focà, Eugenia Quiros-Roldan (Brescia); Antonina Franco (Siracusa); Diego Ripamonti, Franco Maggiolo (Bergamo); Roberto Gulminetti, Massimiliano Fabbiani (Pavia); Sandro Piga, Marzia Garau, Marco Campus (Cagliari); Stefano Rusconi, Tiziana Formenti, Arianna Gabrieli, Alessia Lai, Andrea Giacomelli, Cecilia Bonazzetti (H Sacco Milano); Giulia Marchetti, Camilla Tincati (H San Paolo Milano); Antonella Cingolani (H Gemelli Roma); Giordano Madeddu (Sassari).
