Background
Long-term side effects of antiretroviral therapy (ART) drugs have led to the introduction in clinical practice of nucleoside reverse transcriptase inhibitor (NRTI)-sparing regimens as double or monotherapy, and their use is now recommended in specific populations according to international guidelines [
1,
2]. Indeed, based on the monitoring of surrogate markers of ART efficacy, most of these unconventional regimens, when used in switch studies, have been shown to have a non-inferior virological efficacy and a good CD4 recovery compared to standard triple drug-based therapy [
3‐
8]. The results of these studies were so encouraging that dual combinations are currently being tested in randomized clinical trials of ART-naïve patients [
9,
10]. The results of the GEMINI 1-2 international trials on ART-naïve patients with a baseline plasma viremia < 500,000 copies/mL randomized to receive either tenofovir/emtricitabine or lamivudine both combined with dolutegravir are also ongoing [
11]. If dual regimes are going to be proved non-inferior to triple combination therapy in the ART-naïve population, a single tablet of this dual regimen will likely be developed. Thus, as a consequence of the publication of the results of these studies, in order to save toxicities and resources, we could see a shift in drug production and clinical use of dual therapies [
12,
13]. However, it has been recently shown that virological suppression and CD4 cell count fail to protect from the major causes of death of persons living with HIV (PLWHIV), which are mainly non-AIDS-related events, also known as serious non-AIDS events (SNAEs) [
14]. At present, the best marker to evaluate the risk of developing SNAEs has not been determined. Interestingly, the analysis of the data of our and other cohorts have shown that, in contrast with recent data from the Antiretroviral Therapy Cohort Collaboration (ART-CC) [
15], a low CD4/CD8 ratio is a predictor of non-AIDS-related events independently from CD4 cell count [
16,
17], while other studies have shown an association of this marker with non-AIDS-defining cancers [
18] or, more recently, with pulmonary emphysema [
19].
Therefore, we hereby aimed to compare CD4/CD8 ratio changes in a group of patients who, in the presence of undetectable plasma HIV viremia, were switched to a protease inhibitor monotherapy (mono) or a dual ART, to a control group who was switched to a standard triple drug-based treatment.
Discussion
The main result of our study is that when comparing three groups of patients undergoing different switch strategies in the presence of undetectable HIV RNA, those who were switched to dual regimens showed a stabilization of the CD4/CD8 ratio, while the CD4/C8 ratio of those who were switched to a three-drug-based regimen continued to improve after the switch. This result appears not to be due to the CD4 increase, since no significant difference in the CD4 count trajectory between the two groups could be detected, but to a specific increase in CD8 lymphocyte count in participants switching to dual therapy. More importantly, because CD4 count continued to increase and plasma viral load continued by analysis design to be undetectable in all subjects, this difference in the ratio would possibly go undetected by standard monitoring of HIV RNA and CD4 count alone.
A lower CD4/CD8 ratio can be interpreted as a measure of dysregulation of a patient’s immune system, known as immunological risk phenotype, in the general population and has been clearly associated with a higher risk of AIDS and non-AIDS events in HIV-infected patients [
16,
17]. Therefore, more and more frequently, in people with undetectable HIV RNA, chronic inflammation is becoming an important long-term prognosis issue, and measuring the CD4/CD8 ratio could clinically represent a reliable tool to monitor this phenomenon.
Indeed, despite successful cART, there is evidence for continuous quantitative, qualitative, and functional defects in the CD8 compartments, although some of these defects in some cases could be reversed by early treatment [
22]. However, during chronic HIV infection, peripheral CD8 T cells persistently maintain several defects which are reflected in continuous maintenance of the immune activation parameters. It has been shown that this activation contributes to immunologic exhaustion, hyporesponsiveness of specific T cells, and perturbations in the T cell receptor repertoire. However, reasons for the persistence of elevated CD8 T cells during treatment have not been fully elucidated. Long-term therapy usually determines a significant CD4 recovery, contrasting with, despite a decrease from baseline, persistently elevated CD8 T cell counts [
23]. Previous analyses of cohort data have shown that elevated CD8 T lymphocytes at cART initiation were associated with a poor increase in the CD4 T cell count, even if the studies showed no data on CD8 activation [
22,
23].
Indeed, in Caby’s study, 50% of individuals with a ratio < 1 despite a normalized CD4 count (> 500 cells/uL) still displayed a CD8 count that remained abnormally high (> 1000 cells/uL) [
24]. Moreover, only individuals with a ratio ≥ 1.5 achieved an apparently normal median CD8 count when compared to healthy HIV-seronegative individuals [
24]. Furthermore, after 8 years of suppressive cART, only one-third of patients of the French Hospital Database on HIV cohort achieved CD4/CD8 restoration [
23]. Encouragingly, Saracino et al. have shown that patients with more than 15 years of cART had a progressive increase in the CD4/CD8 ratio which never reached a plateau, but patients included in that analysis were receiving triple therapy [
25]. Our data suggest that treating patients with less than three drugs might lead to a stop of this virtuous trend.
What could explain the CD8 increase observed in our patients who were switched to dual therapies? One possible explanation is that two drugs could not suppress HIV as well as three drugs can. A small residual viremia, probably in lymphoid tissues, could thus trigger the production of proinflammatory and/or homeostatic cytokines that, in turn, would stimulate and/or maintain CD8 T cell proliferation/activation. In our analysis, everybody had an HIV RNA ≤ 50 copies, but we could not create a finer classification below this threshold with the available information on the assays used. In addition, we have shown that viral blips > 50 copies/mL were rare, and their rate of occurrence was similar between the mono, dual, and triple regimen groups, while those > 200 copies/mL were actually more frequent among those who were in the triple regimen. Moreover, it has been previously described that even more sophisticated markers of residual HIV replication, such as HIV -DNA, did not differ between patients switched to dual or monotherapy compared to those continuing triple therapy [
26,
27]. However, it is possible that viral replication restarts in lymphoid tissues, and it could be detected by new sophisticated techniques like digital droplet polymerase chain reaction (PCR) or functional virological assays [
28] or by analyzing cells resident in those anatomic districts—assays that typically are not available in routine clinical studies. Indeed, recent data from the Collaboration of Observational HIV Epidemiological Research in Europe (COHERE) have shown that, in HIV controllers, the decrease of the ratio and the increase in CD8 lymphocyte counts preceded by 5 years the end of virological control. Moreover, CD8 lymphocyte counts increased significantly in those who experienced loss of virological control, whereas they remained stable in the other groups [
29]. Currently it is very unlikely to see randomized clinical trials which last longer than 144 weeks, which is too short a time to verify the COHERE observation in the context of randomized data. Moreover, very rarely are CD8 lymphocyte count data reported in clinical trials. Helleberg et al. have shown that in patients receiving cART for 10 years a value of CD8 lymphocyte count which stays above 1500 cells/uL is associated with increased non-AIDS-related mortality (mortality rate ratio 1.82; CI 1.09–3.22) [
22].
The changes in ratio and in CD8 were not detected in subjects switched to monotherapy. Indeed, although there was a trend in the univariable analysis, it was not confirmed with the multivariable linear regression and mixed models analysis. One possible explanation could be the lower number of subjects who were switched to monotherapy, but another could be that their better immunological profile was definitively better at baseline.
We are well aware of the limitations of our study. First, among dual therapies we could not examine protease inhibitor (PI) and integrase strand transfer inhibitor (INSTI)-based combinations separately (not many people received dolutegravir either); second, a median follow-up of 2 years could not allow us to evaluate either the persistence of this phenomenon or the impact of the strategies on the onset of non-AIDS events; third, we did not have more detailed immunological data than the counts themselves, and it would be extremely relevant to stratify the analysis according to specific CD8 subpopulations and CD38/HLDR+ expression. Finally, since our analysis is retrospective and non-randomized, it only accounts for measured fixed confounders at the time of the therapy switch. Thus, we cannot rule out bias being introduced by unmeasured confounders or time-dependent confounders that have not been appropriately controlled for. In particular, even in the controlled setting of undetectable HIV RNA, patients switching to a reduced drug regimen (less than three) instead of another three-drug regimen might be determined by the factors that we did not measure.
After more than 20 years since the introduction of cART because of the use in clinic of new potent and better tolerated drugs, there is now interest in reassessing the ideal number of antiretroviral drugs which need to be prescribed. In the absence of data from randomized studies, our results appear to be relevant to this debate. Current research on the clinical effectiveness of dual cART regimens is focused on non-inferiority of the virological outcome and on saving toxicity due to reducing drugs in the regimens. Nevertheless, it is possible that dual regimens have an unfavorable impact on the CD4/CD8 ratio, and this possibility must be thoroughly investigated before implementing novel drug treatment strategies including less than three drugs.
At present, there is evidence supporting CD8 count monitoring as optional in patients with satisfactory virological and immunological control during ART [
30], and this is currently taken into account by international guidelines [
2]. However, immunological monitoring based only on the predictive role of a low CD4 count on the risk of developing clinical events may underestimate the role of the CD8 count as a surrogate of a proinflammatory state.
Acknowledgements
We acknowledge the following regarding this research:
Icona Foundation Study Group.
Board of directors
A d’Arminio Monforte (President), A Antinori, A Castagna, F Castelli, R Cauda, G Di Perri, M Galli, R Iardino, G Ippolito, A Lazzarin, GC Marchetti, CF Perno, G Rezza, F von Schloesser, P Viale.
Scientific secretary
A d’Arminio Monforte, A Antinori, A Castagna, F Ceccherini-Silberstein, A Cozzi-Lepri, E Girardi, S Lo Caputo, C Mussini, M Puoti.
Steering committee
M Andreoni, A Ammassari, A Antinori, C Balotta, A Bandera, P Bonfanti, S Bonora, M Borderi, A Calcagno, L Calza, MR Capobianchi, A Castagna, F Ceccherini-Silberstein, A Cingolani, P Cinque, A Cozzi-Lepri, A d’Arminio Monforte, A De Luca, A Di Biagio, E Girardi, N Gianotti, A Gori, G Guaraldi, G Lapadula, M Lichtner, S Lo Caputo, G Madeddu, F Maggiolo, G Marchetti, S Marcotullio, L Monno, C Mussini, S Nozza, M Puoti, E Quiros Roldan, R Rossotti, S Rusconi, MM Santoro, A Saracino, M Zaccarelli.
Statistical and monitoring team
A Cozzi-Lepri, I Fanti, L Galli, P Lorenzini, A Rodano, M Shanyinde, A Tavelli.
Biological bank INMI
F Carletti, S Carrara, A Di Caro, S Graziano, F Petrone, G Prota, S Quartu, S Truffa.
Participating physicians and centers
Italy: A Giacometti, A Costantini, V Barocci (Ancona); G Angarano, L Monno, C Santoro (Bari); F Maggiolo, C Suardi (Bergamo); P Viale, V Donati, G Verucchi (Bologna); F Castelli, C Minardi, E Quiros Roldan (Brescia); T Quirino, C Abeli (Busto Arsizio); PE Manconi, P Piano (Cagliari); B Cacopardo, B Celesia (Catania); J Vecchiet, K Falasca (Chieti); L Sighinolfi, D Segala (Ferrara); P Blanc, F Vichi (Firenze); G Cassola, C Viscoli, A Alessandrini, N Bobbio, G Mazzarello (Genova); C Mastroianni, I Pozzetto (Latina); P Bonfanti, I Caramma (Lecco); A Chiodera, P Milini (Macerata); A d’Arminio Monforte, M Galli, A Lazzarin, G Rizzardini, M Puoti, A Castagna, G Marchetti, MC Moioli, R Piolini, AL Ridolfo, S Salpietro, C Tincati, (Milano); C Mussini, C Puzzolante (Modena); A Gori, G Lapadula (Monza); N Abrescia, A Chirianni, G Borgia, R Orlando, G Bonadies, F Di Martino, I Gentile, L Maddaloni (Napoli); AM Cattelan, S Marinello (Padova); A Cascio, C Colomba (Palermo); F Baldelli, E Schiaroli (Perugia); G Parruti, F Sozio (Pescara); G Magnani, MA Ursitti (Reggio Emilia); M Andreoni, A Antinori, R Cauda, A Cristaudo, V Vullo, R Acinapura, G Baldin, M Capozzi, S Cicalini, A Cingolani, L Fontanelli Sulekova, G Iaiani, A Latini, I Mastrorosa, MM Plazzi, S Savinelli, A Vergori (Roma); M Cecchetto, F Viviani (Rovigo); G Madeddu, P Bagella (Sassari); A De Luca, B Rossetti (Siena); A Franco, R Fontana Del Vecchio (Siracusa); D Francisci, C Di Giuli (Terni); P Caramello, G Di Perri, S Bonora, GC Orofino, M Sciandra (Torino); M Bassetti, A Londero (Udine); G Pellizzer, V Manfrin (Vicenza); G Starnini, A Ialungo (Viterbo).