| |
Blood Telomere Length Changes after Ritonavir-boosted Darunavir Combined with Raltegravir or Tenofovir-Emtricitabine in Antiretroviral-Naive Adults Infected with HIV-1
|
| |
| |
The Journal of Infectious Diseases, 03 July 2018 - Natalia Stella-Ascariz*1, Rocio Montejano*1, Javier Rodriguez-Centeno1, Belen Alejos2, Christine Schwimmer3, Jose I. Bernardino1, Berta Rodes1, Clotilde Allavena4, Christian Hoffmann5, Magnus Gisslen6, Rosa de Miguel1, Andres Esteban-Cantos1, Cedrick Wallet3, Francois Raffi7, Jose R. Arribas1, and the NEAT 001/ ANRS 143 Study Group
Download the PDF here
INTRODUCTION
HIV infection leads to an accelerated immunosenescence status marked by dominant senescent and exhausted phenotypes of mature T cells with a decrease in naïve T cells 1,2. Senescent T cells have limited proliferative capacity due to telomere attrition 3. In keeping with this immunosenescence status, HIV-infected individuals have shorter blood telomere length (TL) than HIV uninfected controls 4-7.
Antiretroviral treatment (ART) partially reverses HIV associated immunosenescence. Initial control of HIV replication translates into an increase in naïve and central memory CD4 and CD8 cells that have longer telomeres. The increase in TL after initiating ART is correlated mainly with shifts of CD8 cells subpopulations towards less mature phenotypes 8,9.
In vitro studies have shown that tenofovir and abacavir -two recommended nucleos(t)ide reverse transcriptase inhibitors [N(t)RTI]- are able to inhibit human telomerase, being tenofovir the most potent inhibitor 10-12. The clinical relevance of this in vitro finding is unknown. There are no studies comparing TL changes in ART naïve participants who start treatment with N(t)RTI containing versus N(t)RTI sparing ART. For this reason, we have evaluated blood TL changes in a substudy of the NEAT001/ANRS 143 clinical trial that compared ritonavir-boosted darunavir combined with raltegravir or tenofovir disoproxil fumarate/emtricitabine in ART naive adults. Our research hypothesis was that exposure to tenofovir, in line with its in vitro activity inhibiting the telomerase, would have a negative impact on blood TL changes.
-----------------
To the best of our knowledge this is the first clinical trial that compares TL changes after initiation of two different ART strategies. .....Surprisingly, and contrary to our research hypothesis based on in vitro study results10-12, participants receiving ritonavir boosted darunavir, emtricitabine and tenofovir disoproxil fumarate had significantly higher gains in TL than those receiving a N(t)RTI sparing regimen (ritonavir boosted darunavir and raltegravir). This is the first clinical trial showing that N(t)RTI containing ART has a measurable positive impact on longitudinal TL changes, in naive HIV participants. This result suggests that ART regimens including N(t)RTIs could play an important role in initial recovery of HIV associated immunosenescence. .....Alcohol abuse has been previously associated with TL attrition in HIV negative individuals.....We have shown that in the setting of virological suppression that tenofovir has a negative impact on longitudinal TL changes22 further supporting that in naïve participants the main driver of immunosenescence is active HIV replication. .....Given these results the use of blood TL to evaluate the immunological impact of initial ART could become an interesting subject of research.
After starting ART, the main driver for immune reconstitution and a shift towards T cells subpopulations with longer TL (naïve and central memory) is the decrease of antigenic stimulation secondary to rapid control of HIV replication. In the parent NEAT 001/ANRS 143 trial although the N(t)RTI sparing regimen met non-inferiority criteria for the primary endpoint, there were important differences in efficacy in favor of the tenofovir disoproxil fumarate/emtricitabine arm in the subgroup of participants with viral loads above 100,000 HIV RNA copies/mL and/or CD4 cell counts under 200 cells/μL13. It is therefore possible that differences in TL between groups in our substudy could be due to worse virological control with the N(t)RTI sparing strategy. In our random sample of participants differences in blood TL occurred despite both groups having similar control of plasma viral replication and similar number of primary and virological endpoints. Notwithstanding, at week 96 plasma virological suppression was numerically higher -and occurred sooner- in participants receiving raltegravir than in those receiving tenofovir disoproxil fumarate/emtricitabine. A possible explanation is that plasma viral load may not completely reflect the antiviral efficacy of ART in tissues, especially in lymph nodes.
Blood telomere length changes
Blood telomere length analysis showed that at baseline there were no statistically significant differences between groups,. After 96 weeks, both groups had a gain in TL: mean TL in the total analyzed samples increased 0.028 (Figure 1 and Table S2). However, increase in TL was only significant in the tenofovir disoproxil fumarate/emtricitabine group. Additionally, the proportion of participants who had increase in TL at follow-up was 71% in tenofovir disoproxil fumarate/emtricitabine group versus 57% in the raltegravir group.
We hypothesize that participants receiving darunavir and raltegravir could have persistent HIV replication in lymph nodes - due to lower tissue concentrations- than participants receiving darunavir and tenofovir disoproxil fumarate/emtricitabine. This persistent antigenic stimulation in lymph nodes would maintain the stimulus for T cells to differentiate to mature phenotypes with shorter TL. In naïve participants with high levels of HIV replication a higher penetration of tenofovir disoproxil fumarate/emtricitabine in lymph node tissue may overcome the inhibitory effect of tenofovir upon telomerase10-
Multivariable Estimative Analysis
In our estimative analysis, exposure to tenofovir disoproxil fumarate/emtricitabine had a positive impact on TL change. After 96 weeks, tenofovir disoproxil fumarate/emtricitabine exposed participants had a gain in mean blood TL adjusted by baseline TL that was 0.031 superior to raltegravir exposed participants (p=0.009) (Figure 2). This effect was not significantly confounded by age, gender, race, time since HIV diagnosis, baseline HIV RNA, nadir or baseline CD4 cell count, baseline CD8, baseline CD4/CD8, tobbaco and alcohol consumption, statins or hepatitis C. These results were unchanged when TL was analyzed as a binary variable -TL shortened/not shortened-(data not shown).
In the univariate analysis treatment with tenofovir disoproxil fumarate/emtricitabine , younger age and no current alcohol consumption were significantly associated with a gain in mean TL among all participants. We found no significant associations with tobacco, gender, race or statin treatment. Regarding HIV related factors, we found no significant associations of mean TL gain with time since HIV diagnosis, nadir or baseline CD4, baseline HIV RNA and hepatitis C virus coinfection (Table 3)
In the multivariable analysis, independent predictors of gain in TL were baseline TL (p < 0.001), treatment with TDF/FTC (p = 0.005) and no current alcohol consumption at baseline (p = 0.026). There was a trend (p =0.097) towards younger age also being associated with higher gains in TL (Table 3)
DISCUSSION
Surprisingly, and contrary to our research hypothesis based on in vitro study results10-12, participants receiving ritonavir boosted darunavir, emtricitabine and tenofovir disoproxil fumarate had significantly higher gains in TL than those receiving a N(t)RTI sparing regimen (ritonavir boosted darunavir and raltegravir). This is the first clinical trial showing that N(t)RTI containing ART has a measurable positive impact on longitudinal TL changes, in naive HIV participants. This result suggests that ART regimens including N(t)RTIs could play an important role in initial recovery of HIV associated immunosenescence
Although overall mean blood TL increased after two years regardless of ART, this was only statistically significant in participants randomized to tenofovir disoproxil fumarate/emtricitabine. Difference in mean gain in TL was also statistically significant in favor of tenofovir disoproxil fumarate/emtricitabine, and the proportion of participants with increases in TL was 14% higher than the N(t)RTI-sparing group. In our estimative analysis the adjusted difference between groups in mean TL changes was 0.031 which represents a 4.2% of the baseline mean blood TL for the whole population. One recent study in 51 intravenous drug users showed that three months after HIV seroconversion the TL in PBMC measured by quantitative PCR decreased 13%15. Therefore, it is likely that the difference between arms in recovery of blood TL after starting ART is important.
The positive impact of tenofovir disoproxil fumarate/emtricitabine on TL gain was not confounded by baseline or week 96 variables. In our predictive analysis the only factors associated with a statistically higher gain in TL were tenofovir disoproxil fumarate/emtricitabine and no current alcohol consumption with younger age approaching significance. Alcohol abuse has been previously associated with TL attrition in HIV negative individuals 16.
To the best of our knowledge this is the first clinical trial that compares TL changes after initiation of two different ART strategies. Two small prior studies have reported that participants starting ART experienced increases in mean TL8,17 driven by a uniform increase in the TL of CD8 T cells that correlated with a decrease in mature memory cells. Changes in the TL of CD4 T cells were more inconsistent and variable. Given these results our hypothesis for the observed differences in TL between both strategies is that participants receiving tenofovir disoproxil fumarate/emtricitabine experienced larger increases in the TL of mainly CD8 cells and that this increase represents a shift towards less mature T8 cell phenotypes with longer TL. Support for this hypothesis comes from the fact that six months after starting ART there is a decrease in proportion of CD28-CD8+ that characteristically have shorter TL and an increase in central memory T cells that have longer TL9. Interestingly the other predictive factor of lower TL gain in our study -alcohol consumption- increases CD8+ T-cell immunosenescence in simian immunodeficiency virus-infected rhesus macaques 18.
After starting ART, the main driver for immune reconstitution and a shift towards T cells subpopulations with longer TL (naïve and central memory) is the decrease of antigenic stimulation secondary to rapid control of HIV replication. In the parent NEAT 001/ANRS 143 trial although the N(t)RTI sparing regimen met non-inferiority criteria for the primary endpoint, there were important differences in efficacy in favor of the tenofovir disoproxil fumarate/emtricitabine arm in the subgroup of participants with viral loads above 100,000 HIV RNA copies/mL and/or CD4 cell counts under 200 cells/μL13. It is therefore possible that differences in TL between groups in our substudy could be due to worse virological control with the N(t)RTI sparing strategy. In our random sample of participants differences in blood TL occurred despite both groups having similar control of plasma viral replication and similar number of primary and virological endpoints. Notwithstanding, at week 96 plasma virological suppression was numerically higher -and occurred sooner- in participants receiving raltegravir than in those receiving
tenofovir disoproxil fumarate/emtricitabine. A possible explanation is that plasma viral load may not completely reflect the antiviral efficacy of ART in tissues, especially in lymph nodes. Three recent studies have reported that compared to tenofovir disoproxil fumarate and emtricitabine, both darunavir and raltegravir have lower concentrations in lymph node tissue19-21. We hypothesize that participants receiving darunavir and raltegravir could have persistent HIV replication in lymph nodes - due to lower tissue concentrations- than participants receiving darunavir and tenofovir disoproxil fumarate/emtricitabine. This persistent antigenic stimulation in lymph nodes would maintain the stimulus for T cells to differentiate to mature phenotypes with shorter TL. In naïve participants with high levels of HIV replication a higher penetration of tenofovir disoproxil fumarate/emtricitabine in lymph node tissue may overcome the inhibitory effect of tenofovir upon telomerase10-12. We have shown that in the setting of virological suppression that tenofovir has a negative impact on longitudinal TL changes22 further supporting that in naïve participants the main driver of immunosenescence is active HIV replication.
An alternative explanation for our findings is that differences in blood TL indicate an increase in T cells with shorter TL in participants treated with darunavir and raltegravir due to better control of virological replication and earlier decrease of immune activation. Individuals chronically infected with HIV have low proportions of CD28− CD8+ T cells expressing CD57 which are characterized by very short telomeres23. After ART initiation the proportion of CD28− CD8+ T completing terminal differentiation and expressing CD57 increases,24. Therefore, is possible that participants treated with darunavir and raltegravir do not experience blood TL increase due to lower immune activation and increasing numbers of CD28- CD57+ CD8+ T cells. We consider this possibility less likely: firstly because overall results of NEAT 001 indicate lower efficacy of the darunavir and raltegravir regimen, and secondly because the net effect o
An alternative explanation for our findings is that differences in blood TL indicate an increase in T cells with shorter TL in participants treated with darunavir and raltegravir due to better control of virological replication and earlier decrease of immune activation. Individuals chronically infected with HIV have low proportions of CD28− CD8+ T cells expressing CD57 which are characterized by very short telomeres23. After ART initiation the proportion of CD28− CD8+ T completing terminal differentiation and expressing CD57 increases,24. Therefore, is possible that participants treated with darunavir and raltegravir do not experience blood TL increase due to lower immune activation and increasing numbers of CD28- CD57+ CD8+ T cells. We consider this possibility less likely: firstly because overall results of NEAT 001 indicate lower efficacy of the darunavir and raltegravir regimen, and secondly because the net effect of successful ART is to increase TL (as has been seen in several studies8,17, including our prospective cohort of virologically suppressed participants)22
Significant differences in blood TL occurred despite similar changes in CD4 and CD8 cell counts and similar CD4/CD8 ratios. In virologically suppressed participants with CD4 cell counts above 500 cells/μL, the CD4/CD8 ratio is correlated positively with the frequency of T cells with longer telomeres (naïve T cells, central memory CD8 and transitional memory CD8) and negatively with the frequency of T cells with shorter telomeres (effector memory and terminally differentiated cells)25. However, in our study despite the large difference observed in TL by treatment arm there were no differences in CD4/CD8 ratio or in time to achieve a CD4/CD8 ratio above 0.4, a cutoff that in one study identified individuals with prominent immunosenescence25. Our data suggest that the CD4/CD8 ratio may not be sensitive enough to identify differences in the distribution of T cell subpopulations with different TL within the first two years of initial ART. In our study we unveil important TL differences between two different ART strategies despite similar control of viral replication in blood and similar changes in CD4, CD8 and CD4/CD8 ratios. Given these results the use of blood TL to evaluate the immunological impact of initial ART could become an interesting subject of research
The main limitation of our study is that we did not determine TL on specific subsets of T cells. Consequently, we cannot prove at this time our hypothesis that blood TL changes are driven by modifications in T cell subpopulations. The other limitation is the lack of samples beyond week 96. Without these long-term samples it is not possible to evaluate the long-term evolution of the observed differences between study arms.
In summary, in antiretroviral naïve participants, N(t)RTI sparing ART using ritonavir boosted darunavir and raltegravir was associated with lower longitudinal gains in blood TL than N(t)RTI containing ART using ritonavir boosted darunavir, emtricitabine and tenofovir disoproxil fumarate. These results suggest that N(t)RTI containing ART produces a more rapid initial recovery from HIV associated immunosenescence.
|
|
| |
| |
|
|
|
