Failure to achieve immunological recovery in HIV-infected patients with clinical and virological success after 10 years of combined ART: role of treatment course Free

Objectives

We assessed factors, including treatment course, associated with failure to obtain a 10 year immunological response after starting first-generation PI-containing combined ART (cART).

Patients and methods

In the prospective COPILOTE cohort of HIV-infected patients started on a first-generation PI-containing regimen in 1997–99, the impact of cART history on the failure to achieve immunological response measured at 10 years was assessed by multivariate logistic regression models in the 399 patients with clinical and virological success of cART.

Results

Failure of CD4 response (CD4 >500/mm³) was associated with age ≥40 years at baseline (P < 0.001), CD4 cell counts ≤500/mm³ at month 4 (P = 0.016) or month 12 (P < 0.001) and ≥3 months of cART interruption (P = 0.016). Factors associated with failure to achieve complete immunological response (CD4 >500/mm³ and CD4:CD8 ratio >1) were CD4:CD8 ratio ≤0.8 at month 8 (P < 0.001) or month 12 (P < 0.001), ≥3 months of cumulative cART interruption (P = 0.011), ≥3 antiretroviral regimens (P = 0.009) and ≤4 treatment lines (P = 0.015). Baseline CD4 and CD4:CD8 ratio were not predictors of the 10 year immunological outcomes.

Conclusions

In this therapeutic cohort of patients starting first-generation PI-containing cART in 1997–99, poor initial immunological response had a negative impact on 10 year CD4 and CD4 plus CD4:CD8 ratio response, despite prolonged virological success. Lack of treatment interruption may improve long-term immunological outcome in HIV infection.

Introduction

Since 1996 and the availability of the first PI-based combined ART (cART), many improvements in terms of drugs' potency, barrier to resistance, tolerability, forgiveness and simplicity have been achieved. Treatment changes occur frequently in patients with HIV during the course of treatment, which is now administered lifelong, for many decades. Such changes can have numerous reasons, but are always driven by the main objective to adapt and/or optimize therapy to the patient's situation to overcome resistance and achieve sustained virological success with minimal long-term toxicity.¹ Although most patients on long-term cART achieve virological success, immunological response remains largely incomplete, which is associated with increased risk of non-AIDS clinical events.^2–4 Many studies have evaluated the determinants of not achieving CD4 cell counts >500/mm³, but few have assessed factors associated with achievement of CD4:CD8 ratio >1. A recent study from the UK identified, beyond classical host and HIV disease factors, that initiation with some antiretrovirals was associated with normalization of CD4:CD8 ratio.⁵ The impact of the treatment course and various modifications on long-term immunological outcome has not been assessed. The objective of the study was to assess the potential role, over 10 years, of treatment course and changes on long-term immunological response in a population of HIV-infected patients having started first-generation PI cART, as soon as they became available. Such patients have had a complex treatment course, with possible sequential sub-optimal regimens, treatment discontinuation(s) or change(s) driven by successive virological failures and drug toxicities.

Patients and methods

ANRS CO8 COPILOTE was a prospective cohort designed to assess long-term outcome of patients started on a PI-containing cART at 47 French clinical sites. HIV-1-infected adults were enrolled in 2003–04 if they had been started on PI from April 1997 to June 1999 and followed in the APROCO cohort for at least 5 years.⁶ The following data were collected: full treatment history since initiation of the PI-containing regimen, with dates of changes and discontinuations; plasma HIV-1 RNA; CD4+ and CD8+ cell counts; all clinical and laboratory events considered as severe; new AIDS or HIV-associated non-AIDS conditions; and deaths. Follow-up was at month 4 post-enrolment, month 8 and every 4–8 months for at least 10 years. All patients signed consent forms to participate in the study (ANRS CO8). The research study protocol was approved by an Ethics Committee (CCPPRB Paris-Cochin, Paris, France) and was conducted in accordance with the principles of the Declaration of Helsinki.

Patients with favourable clinical and virological response at 10 years, i.e. without new AIDS-defining events in the last 18 months of follow-up and with a plasma HIV-1 RNA <50 copies/mL at the last follow-up visit and with no more than one viral blip (defined as HIV-1 RNA between 50 and 500 copies/mL) in the last 18 months of follow-up, were included in the analysis. Primary outcome was failure to achieve CD4 response at 10 years, defined as CD4 cell counts >500/mm³ in the last 12 months with no more than one value <500/mm³ in the last 18 months. Secondary outcomes were: (i) failure to achieve a CD4:CD8 ratio >1; and (ii) failure to achieve a complete immunological response defined as the combination of CD4 cell counts >500/mm³ and the CD4:CD8 ratio >1 in the last 12 months, with no more than one value <500/mm³ for CD4 and <1 for the CD4:CD8 ratio in the last 18 months. These endpoints were categorized using all values of CD4 and CD8 available during the last 18 months of follow-up. cART course during follow-up was summarized by the number of different regimens, of different classes of antiretrovirals, of different antiretrovirals within each class, of antiretrovirals received, as well as the number of treatment lines received by each patient. A line was defined as any change in a cART combination, e.g. a patient with three successive lines of PI-containing therapy was classified as having received one regimen and three lines of treatment. Duration of treatment interruption was also included. We calculated the impact of ART history on the immunological response measured at 10 years, by using multivariate logistic regression models, taking into account other potential predictors of long-term immunological response available at baseline and during follow-up. Covariates with P < 0.20 in univariate analysis were selected. Then a backwards selection was used to compute the final model. Fixed covariates in the model were HIV-1 RNA measures at each timepoint, cART-naive indicator at baseline and characteristics of treatment history.

Results

Among the 719 patients enrolled in the COPILOTE cohort, 610 had at least 10 years of follow-up (median = 10.0 years, IQR = 10.0–10.1). Others withdrew participation (n = 92) or died (n = 17). Among the 591 patients (96.9%) with clinical response, 561 had at least two HIV-1 RNA measures in the last 18 months. Favourable virological response was seen in 399 (71.1%) participants (Figure S1, available as Supplementary data at JAC Online). Patients' characteristics were similar for the total COPILOTE cohort and the 399 patients in virological success (Table S1). The median number of CD4 and CD8 measures during the last 18 months of follow-up was 5 (IQR = 4–5). Failure to achieve a CD4 response (CD4 >500/mm³) was seen in 132 patients (33.1%), failure to obtain a CD4:CD8 ratio >1 in 306 participants (76.7%) and absence of complete immunological response (combined CD4 and CD4:CD8 ratio measures) in 319 participants (80.0%). The proportion of patients failing to achieve a CD4:CD8 ratio >1 and a complete immunological response were quite closed, analyses found concordant results for both sub-groups (data not shown), and only results for the first (CD4 response) and third (complete immunological response) ones are presented. Table 1 shows demographics, baseline and ART characteristics of patients according to the immunological response. In multivariate analysis, factors independently associated with failure to achieve immunological response were older age, non-immunological response during the first year of follow-up, in addition to some treatment history characteristics (Table 2). A longer duration of treatment interruption was significantly associated with a higher risk of failure of both CD4 and complete immunological responses (Table 2). CD4 response was not associated with number of treatment classes, regimens or lines received. Failure to achieve a complete immunological response was also associated with a higher number of different antiretroviral regimens and lower number of lines of therapy (Table 2). Baseline CD4 count was not associated with the 10 year immunological outcome when taking into account initial first year (month 4, 8 or 12) immunological response.

Discussion

In this therapeutic cohort of HIV-infected patients who started cART with first PI-containing cART, patients with no clinical progression and with virological control after 10 years of follow-up, had a relative high rate of incomplete immunological recovery, not only for absolute CD4 cell counts (33.1%), but also even more importantly for CD4:CD8 ratio (76.7%). The proportion of 23.3% of CD4:CD8 normalization at 10 years in our study is in line with results of two large cohorts, ICONA and UK CHIC, where the estimated Kaplan–Meier probability of obtaining a CD4:CD8 ratio of ≥1 by 5 years after cART initiation was 29.4% and 18.3%, respectively.^5,7 Our study had a different approach than other cohorts, attempting to identify factors of failure to achieve immunological response in patients with clinical and virological success on cART. Our results confirm those from cohorts with shorter follow-up (2.77 and 4.20 years),^8,9 factors associated with poorer immunological recovery being patients' age >40 years at PI initiation, and initial immunological response. The UK CHIC study, from a cohort of 9257 subjects starting cART after 2000, identified host, HIV disease and antiretroviral factors, such as initiation of cART with an NRTI backbone other than tenofovir/emtricitabine, to be associated with normalization of the CD4:CD8 ratio.⁵ However, this study had a shorter follow-up (median = 3 years), and did not take into account changes of regimens. The original contribution and strength of our study is the precise recording of treatment lines and regimens during a long follow-up allowing assessment of the impact of characteristics of treatment history on immunological response. We found that the cumulative duration of treatment interruption, higher number of treatment regimens and lower number of treatment lines, were significantly associated with reduced probability of normalization of CD4 cell counts or the CD4:CD8 ratio in the long term.¹⁰ Age was a predictor of CD4, but not of complete immunological response. This might be explained by only minor changes in the balance between the CD4 and CD8 lymphocyte subsets with ageing except in the elderly.¹¹ Neither baseline CD4 nor baseline CD4:CD8 ratio were predictive of failure to achieve immunological response. Our results confirm that the ‘original sin’ and ‘persistent sequelae’ of late initiation and/or having a poor immune status—low CD4 cell counts and/or CD4:CD8 ratio—at cART initiation¹² can be partly counterbalanced in the long term by a good early (month 12) immunological response. Mechanisms of the negative long-term impact of first year CD4 response could be multifactorial and include genetic background, age, thymic output, viral reservoir size, residual viraemia and co-infections such as hepatitis B or C, cytomegalovirus, Herpes simplex virus.^13–17 From a clinical standpoint, CD4 and the CD4:CD8 ratio evaluation 1 year after cART initiation in patients starting with low CD4 cell counts is the clue, as not reaching normal values of CD4 and CD4:CD8 ratio at this timepoint is strongly associated with absence of 10 years immunological recovery and should thus alert clinicians to the possibility of HIV-associated non-AIDS complications.¹⁸ Association of treatment interruption with a lower rate of CD4 recovery confirms previous results of a large cohort with 7 years of follow-up.¹⁹ We provide here additional information by showing that up to 10 years, a total interruption >3 months is irrevocably deleterious for immune recovery. The characteristics of treatment course had a more direct impact on complete immunological response suggesting that the CD4:CD8 ratio is more sensitive than the CD4 response to day-to-day closed management of cART. We hypothesize that the use of more various regimens over time, required by virological failure, but also intolerance and/or adherence problems, may have led to periods of sub-optimal therapy with deleterious consequence on long-term immunological outcome. On the other hand, the higher rate of immunological response in patients exposed to more lines of therapy might indicate that their treatment has been optimized according to individual situation and new available therapies.

Our study has several limitations. Only the patients surviving and not lost to follow-up could be assessed after 10 years of follow-up. Only baseline CD4, at initiation of cART, not CD4 nadir was available. However, in ART-naive or pre-treated patients having previously received mono or dual NRTI regimens, baseline CD4 is probably very similar to the CD4 nadir. All patients from our cohort were commenced on PI regimens, per inclusion criteria of the APROCO cohort, initiated in 1997, at the time when the first cART regimens became available on a large scale. The PIs initially prescribed in our cohort, mainly indinavir, saquinavir and nelfinavir, the first-generation PIs, are considered by today standards sub-optimal, and were associated with higher toxicities and pill burden than more recent PIs. Whether our results, particularly the negative impact of treatment interruption as short as 3 months, will also be seen with more modern drugs with higher potency and forgiveness and newer regimens will need future investigation. With these new and potent antiretrovirals currently in the clinic, the impact of treatment course on immunological response could be less critical, given that treatment success is achieved early and sustained.

In conclusion, in this therapeutic cohort of patients with HIV with clinico-virological success after 10 years of potent cART, the immunological response remains sub-optimal, particularly for CD4:CD8 ratio normalization, particularly in older patients, in those who do not achieve early immunological response, and in those with a total duration of treatment interruption ≥3 months or who have received more than three different regimens. Further research on long-term cohorts with new antiretrovirals is needed to assess whether poor short-term immunological response will still have an impact on long-term outcome, and to elucidate the mechanisms that lead to a persistently low rate of immunological response despite viral suppression.

Funding

The ANRS CO8 COPILOTE cohort is funded by ANRS (Agence Nationale de Recherches sur le Sida et les hépatites virales) and CMIT (Collège des Universitaires de Maladies Infectieuses et Tropicales), which received research grants from Abbott, Boehringer-Ingelheim, Bristol-Myers Squibb, GlaxoSmithKline, Gilead Sciences, Pfizer and Roche. Other support: Sidaction Ensemble contre le Sida.

The funders were responsible for financing the cohort, but did not interfere with data analysis and interpretation and dissemination of the scientific results.

Transparency declarations

None to declare.

Author contributions

F. R. and C. L. coordinated the ANRS CO8 COPILOTE cohort. F. R., C. L. and S. H. conceived the study. F. R., C. F. and A. A. prepared the statistical analysis plan. A. A. did the statistical analyses, F. R., A. A., C. F. and C. L. wrote the report. All authors revised and approved the report.

Supplementary data

Figure S1 and Table S1 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).

Acknowledgements

We dedicate this study to the memory of Solange Andreani, for her professional expertise and support during the 10 years of conduct of the cohort. We thank Adélaïde Perrier for help with statistical analysis. We would also like to thank all participating patients, nurses and physicians at the clinical sites.

Composition of the ANRS CO8 APROCO-COPILOTE Study Group

Scientific Committee

Steering Committee: Principal Investigators: C. Leport and F. Raffi, Methodology: G. Chêne and R. Salamon, Social Sciences: J. P. Moatti, J. Pierret and B. Spire, Virology: F. Brun-Vézinet, H. Fleury and B. Masquelier (deceased), Pharmacology: G. Peytavin and R. Garraffo.

Other Members: D. Costagliola, P. Dellamonica, C. Katlama, L. Meyer, D. Salmon and A. Sobel.

Events Validation Committee

L. Cuzin, M. Dupon, X. Duval, V. Le Moing, B. Marchou, T. May, P. Morlat, C. Rabaud and A. Waldner-Combernoux.

Project Coordination

L. Hardel and P. Reboud.

INSERM-ANRS Representatives

S. Couffin-Cadiergues and L. Marchand.

Methodological, Statistical and Other Contributions

A. Assuied, P. Carrieri, S. Habak, F. Couturier, C. Jadand, A. Perrier, M. Préau and C. Protopopescu.

Clinical Centres (Investigators)

Amiens (Pr J. L. Schmit), Angers (Dr J. M. Chennebault), Belfort (Dr J. P. Faller), Besançon (Pr C. Chirouze, Pr N. Magy-Bertrand and Pr P. Humbert), Bordeaux (Pr M. Dupon, Pr M. Longy-Boursier, Pr P. Morlat and Pr D. Neau), Bourg-en-Bresse (Dr P. Granier), Brest (Pr S. Ansart), Caen (Pr R. Verdon), Compiègne (Dr D. Merrien), Corbeil Essonnes (Dr P. Chevojon), Créteil (Pr Y. Levy and Pr A. Sobel), Dijon (Pr L. Piroth), Garches (Pr C. Perronne), Lagny (Dr E. Froguel), Libourne (Dr J. Ceccaldi), Lyon (Pr C. Chidiac), Meaux (Dr V. Grégoire), Montpellier (Pr J. Reynes), Nancy (Pr T. May), Nantes (Pr F. Raffi), Nice (Pr J. G. Fuzibet and Pr P. Dellamonica), Orléans (Dr P. Arsac), Paris (Pr E. Bouvet, Pr F. Bricaire, Pr P. M. Girard, Pr S. Herson, Pr C. Leport, Dr J. Monsonego, Pr G. Pialoux, Pr O. Sain, Pr D. Salmon and Dr P. Sellier), Poitiers (Pr P. Roblot), Reims (Pr F. Bani-Sadr), Rennes (Pr C. Michelet), Saint-Etienne (Pr F. Lucht), Saint-Mandé (Pr T. Debord), Strasbourg (Pr T. Martin), Toulon (Dr J. P. De Jaureguiberry), Toulouse (Pr B. Marchou) and Tours (Pr L. Bernard)

Sponsor

The French National Institute for Health and Medical Research-France Recherche Nord&Sud Sida-HIV Hépatites (INSERM-ANRS).

References

Author notes