| |
Impact of Antiretroviral Treatment Containing Tenofovir Difumarate on the Telomere Length of Aviremic HIV-Infected Patients
|
| |
| |
Download the PDF here
JAIDS Sept 1 2017 - Montejano, Rocio MD*; Stella-Ascariz, Natalia MS; Monge, Susana MD‡; Bernardino, Jose I. MD*; Perez-Valero, Ignacio MD*; Montes, Maria L. MD*; Valencia, Eulalia MD*; Martin-Carbonero, Luz MD*; Moreno, Victoria MD*; González-Garcia, Juan MD*; Arnalich, Francisco MD*; Mingorance, Jesus PhD; Pintado Berniches, Laura MS; Perona, Rosario MD; Arribas, Jose R. MD*
Senescence, Telomerase, Aging HIV - (08/14/17)
IAS: Slowing the premature aging process: the benefits of viral control and smoking cessation on telomere length - (07/27/17)
CROI: Impact of antiretroviral treatment containing tenofovir difumarate on the telomere length of aviremic HIV-infected patients / NRTIs, HIV & Telomerase, Aging, Comorbidities - (03/02/17)
Objective: To evaluate the in vivo relevance of the inhibitory effect of tenofovir on telomerase activity observed in vitro.
Design: Cross-sectional study of HIV-infected patients with suppressed virological replication (HIV RNA <50 copies/mL for more than 1 year).
Methods: Telomere length in whole blood was measured by quantitative real-time polymerase chain reaction. We performed a multivariate analysis to elucidate variables associated with telomere length and also evaluated the association between telomere length and use of tenofovir difumarate (TDF) adjusted by significant confounders.
Results: 200 patients included, 72% men, median age 49 (IQR 45-54.5), 103 with exposure to a TDF containing antiretroviral treatment (ART) regimen (69.9% for more than 5 years) and 97 never exposed to a TDF containing ART regimen. In the multivariate analysis, significant predictors of shorter telomere length were older age (P = 0.008), parental age at birth (P = 0.038), white race (P = 0.048), and longer time of known HIV infection (10-20 and ≥20 years compared with <10 years, P = 0.003 and P = 0.056, respectively). There was no association between TDF exposure and telomere length after adjusting for possible confounding factors (age, parental age at birth, race, and time of HIV infection). Total time receiving ART and duration of treatment with nucleoside reverse transcriptase inhibitors were associated with shorter telomere length, but these associations were explained by time of known HIV infection.
Conclusions: Our data do not suggest that telomerase activity inhibition caused by TDF in vitro leads to telomere shortening in peripheral blood of HIV-infected patients.
"In our multivariate analysis, longer time of HIV infection was associated with shorter telomere length. Compared with patients with less than a decade of known HIV infection, patients with longer durations of known HIV infection had telomeres that were 9%-13% shorter. Time with HIV infection had substantial colinearity with time receiving ART and time receiving NRTIs. When we included these 3 variables in our model, the effect of total time on ART and time receiving NRTI on telomere shortening was explained by total time with known HIV infection. Our study adds further evidence that HIV by itself or by causing persistent inflammation, immune activation, or immune senescence seem to be the predominant factors causing telomere shortening and not the use of specific antiretrovirals such as TDF.7-9,11,12,26
In summary, we have found that despite its confirmed ability to inhibit telomerase in vitro, ART including TDF does not seem to lead to telomere shortening in the peripheral blood of HIV-infected patients.
Impact of Time Receiving N(t)RTIs on Telomere Length
Telomere length decreased with time on N(t)RTI, with 4% attrition for every 5 year of treatment with N(t)RTIs [exp(b) = 0.96; 95% CI: 0.93 to 0.99, P = 0.01]. However, after adjusting for age, the relationship between shorter telomere length and longer time on NRTI decreased to 3% and did not reach statistical significance (95% CI: 0.95 to 1.00, P = 0.076).
Impact of Time Receiving PIs on Telomere Length
At the crude level, in patients with previous exposure to PIs, telomere length was 7.6% shorter than in those not previously exposed [exp(b) = 0.924; CI 95%: 0.857 to 0.996; P = 0.039]. However, further adjustment by age and total time on ART made this association disappear [exp(b) = 0.953; CI 95%: 0.884 to 1.027; P = 0.204]. In the group of patients exposed to PI, no association was evident between length of exposure and telomere length, and no confounders were identified.
Impact of Treatment Containing TDF on Telomere Length
Patients exposed to TDF had telomeres that were 4% shorter than those of patients who have never received TDF, but this difference did not reach statistical significance [exp(b) = 0.96; 95% confidence interval (CI): 0.90 to 1.03, P = 0.238], nor any other independent variable was identified as a confounder of this effect.
In the group of patients exposed to TDF, in the crude analysis, compared with those with less than 5 years of exposure, those with 5-10 years had 8.7% longer telomere length [exp(b) = 1.087; 95% CI: 0.997 to 1.185, P = 0.060] and those with over 10 years had 6.4% longer telomere length [exp(b) = 1.064; 95% CI: 0.970 to 1.167, P = 0.189], although the global significance of this association was far from statistical significance (P = 0.163), no confounders were identified.
Our study is limited by its cross-sectional nature that leads to unmeasured bias in the distribution of different ART regimens. However, we think that our results are inconsistent with a large effect of TDF on telomere shortening at least in peripheral blood in HIV-infected patients. Despite the fact that TDF is the strongest inhibitor of telomerase activity in vitro, this effect seems to be compensated by an unknown mechanism in vivo. One possible explanation is that because HIV-Tat protein by itself can downregulate telomerase expression and activity,24 the negative effect of TDF seen in vitro is compensated by its antiviral activity in vivo. If the inhibition of telomerase caused directly by HIV is substantially higher than the inhibition caused by TDF, then the net effect of TDF on telomere shortening could be positive and similar to other antiretrovirals. We explored also if treatment with PIs could antagonize the negative effect of TDF because an in vitro study has shown that saquinavir can upregulate telomerase activity.18 However, results of our analysis do not support this hypothesis. Additional mechanisms by which telomerase inhibition may lead to abnormal aging in the absence of shorter telomeres include DNA damage induced by deprotection of telomeres that may finally result in premature senescence.29
INTRODUCTION
HIV-infected patients have an increased risk for several "non-AIDS" complications such as cardiovascular disease, cerebrovascular events, malignancy, liver disease, kidney disease, bone disease, and neurocognitive decline that are classically associated with the normal aging process.1
There is a continuous debate about if the higher risk of these complications in HIV-infected patients is the expression of an "accelerated" aging process-complications occurring prematurely-, or an "accentuated" aging process-higher prevalence of complications at every age strata. The Danish HIV-cohort study has not found evidence to suggest accelerated aging in the HIV-infected population.2 In contrast, 2 recent studies using epigenetic biomarkers of aging have found that HIV-infected patients have an age advancement of approximately 5 years compared with HIV uninfected controls.3,4
Proposed mechanisms for the abnormal aging of HIV-infected patients are the proinflammatory state and immune activation associated to even well-controlled HIV infection,5 traditional risk factors (such as smoking) that are more prevalent among HIV-infected people, or other still unknown causes.
Another potential cause of accelerated or accentuated aging in HIV-infected patients could be telomere attrition. There is a close association between shortened telomere length in peripheral blood mononuclear cells and diseases of aging, including cardiovascular diseases, dementia, and cancer.6
Interestingly, multiple studies have reported shorter telomeres in-HIV-infected patients compared with HIV negative controls.7-12 In HIV-infected patients, telomere attrition could be caused by inhibition of human telomerase by antiretroviral drugs, more specifically nucleos(t)ide reverse transcriptase inhibitors [N(t)RTIs].13-15
Two recent studies have reported that tenofovir (TFV) at therapeutic concentrations is a potent inhibitor of telomerase activity16,17 causing telomere shortening in vitro. Of the currently recommended N(t)RTIs, TFV is a more potent inhibitor of telomerase than abacavir, lamivudine, or emtricitabine. In contrast, certain protease inhibitors (PIs) such as saquinavir can upregulate telomerase activity in vitro.18Although the inhibition of telomerase caused by N(t)RTIs has been repeatedly demonstrated in vitro, there are very limited data about the in vivo impact of different antiretroviral treatment (ART) regimens on telomere shortening. Indeed, to the best of our knowledge, there is no study that has explored the impact of TFV containing regimens on telomere length of HIV-infected patients receiving ART. This is a relevant issue because TFV administered as tenofovir difumarate (TDF) or tenofovir alafenamide is recommended as a preferred treatment option for initial treatment of HIV infection in all expert guidelines.
To try to determine the impact of TFV on telomere length of HIV-infected patients receiving ART, we have compared telomere length in a cohort of virologically suppressed, HIV-infected patients who were receiving antiretroviral regimens including and not including TDF. Our research hypothesis was that exposure to TDF would be associated with shorter telomere lengths.
DISCUSSION
In our study, we have shown that ART including TDF does not seem to have an intrinsic negative impact on telomere length in peripheral blood of HIV-infected patients with virological suppression. Compared with patients who have never received TDF, patients who have received TDF for a prolonged period of time-70% for more than 5 years-did not have shorter telomeres in a multivariate analysis looking specifically at the impact of TDF on telomere length.
Ours is the first study that has looked specifically for an in vivo effect of TDF on telomere length in HIV-infected patients. For this reason, we ought to compare a group of patients who have received long-term treatment with TDF with a group of patients who have never been treated with TDF. Both groups were highly comparable in terms of factors that in previous studies have been associated with telomere length: sex distribution, age, income, and smoking status.6 Besides, it has been shown that HIV by itself can downregulate telomerase activity.22-24 Importantly, all patients in both groups have prolonged virological suppression. Consequently, our results are not affected by differences between groups in virological control. Finally, both groups were comparable with regard to total time receiving ART or time receiving N(t)RTIs, 2 factors that theoretically could affect telomere length.16,17 It is important to highlight that our participants treated with TDF had a longer duration of N(t)RTI exposure and despite this fact, TDF exposure was not associated with shorter telomeres.
There are very few previous studies that have explored in vivo the impact of different types of ART regimens on telomere length and none that have specifically focused on the impact of TDF. Leeansyah et al17 in a cross-sectional study with a small sample of just 53 patients found in a univariate analysis that duration of N(t)RTI-containing ART was inversely associated with telomere length and that there was no association with telomerase activity. However, in a multivariate analysis, duration of N(t)RTI-containing ART was no longer significantly related to telomere length. In a substudy of the MONET clinical trial, comparing darunavir/ritonavir monotherapy versus darunavir/ritonavir and 2 N(t)RTIs for maintenance of virological suppression, there was no significant association between telomere length and the duration of previous N(t)RTI treatment.25 Besides, in MONET there were no significant differences between the 2 arms after 3 years of follow-up in telomerase activity or mean change per year of telomere length. Finally, in a cohort of 229 HIV-infected patients, Zanet et al7 found no evidence of a relationship between telomere length and antiretroviral therapy exposure, or current type of antiretroviral therapy, although in this study the prevalence of treatment with TDF was not reported.
In our multivariate analysis, longer time of HIV infection was associated with shorter telomere length. Compared with patients with less than a decade of known HIV infection, patients with longer durations of known HIV infection had telomeres that were 9%-13% shorter. Time with HIV infection had substantial colinearity with time receiving ART and time receiving NRTIs. When we included these 3 variables in our model, the effect of total time on ART and time receiving NRTI on telomere shortening was explained by total time with known HIV infection. Our study adds further evidence that HIV by itself or by causing persistent inflammation, immune activation, or immune senescence seem to be the predominant factors causing telomere shortening and not the use of specific antiretrovirals such as TDF.7-9,11,12,26
The other variables associated with telomere shortening were as expected, older age20 and parental age at birth.27 In our study white race was also associated with shorter telomere length. This finding has to be considered with caution because the impact of race on telomere length is controversial with conflicting findings depending on the study and the sample analyzed.28
Our study is limited by its cross-sectional nature that leads to unmeasured bias in the distribution of different ART regimens. However, we think that our results are inconsistent with a large effect of TDF on telomere shortening at least in peripheral blood in HIV-infected patients. Despite the fact that TDF is the strongest inhibitor of telomerase activity in vitro, this effect seems to be compensated by an unknown mechanism in vivo. One possible explanation is that because HIV-Tat protein by itself can downregulate telomerase expression and activity,24 the negative effect of TDF seen in vitro is compensated by its antiviral activity in vivo. If the inhibition of telomerase caused directly by HIV is substantially higher than the inhibition caused by TDF, then the net effect of TDF on telomere shortening could be positive and similar to other antiretrovirals. We explored also if treatment with PIs could antagonize the negative effect of TDF because an in vitro study has shown that saquinavir can upregulate telomerase activity.18 However, results of our analysis do not support this hypothesis. Additional mechanisms by which telomerase inhibition may lead to abnormal aging in the absence of shorter telomeres include DNA damage induced by deprotection of telomeres that may finally result in premature senescence.29
Another limitation of our study is that we did not determine telomere length on CD4+ or CD8+ T cells, specific subsets of T cells or non-T-cell population or in high replicating tissues. It remains possible that the effect of TDF on the telomerase of these cell subsets could be different from its overall effect on whole blood. Beside, TDF levels could be different depending on the body compartment, and its impact in the different tissues should be studied. In addition, time of HIV infection was no precisely measured in our study because this variable was calculated from the time of diagnosis and not from the actual time of seroconversion. Finally, because ours was a retrospective study using medical records, and for certain variables such as complete antiretroviral history, not all the data were available.
Our results do not completely rule out that there are in vivo differences among antiretrovirals in its ability to produce telomere shortening in vivo. We recognize that a better control group would be HIV-infected patients never exposed to nucleosides, because other nucleosides such as abacavir have an impact on telomerase activity.16,17 The issue of a different impact of various types of ART on telomere length would not be definitively answered if this endpoint were not measured in a randomized clinical trial comparing different ART regimens. In particular, it would be very interesting to compare in a prospective randomized study the impact on telomere length in different cells and tissues of N(t)RTI-containing and N(t)RTI-sparing regimens in naive patients who start ART.
In summary, we have found that despite its confirmed ability to inhibit telomerase in vitro, ART including TDF does not seem to lead to telomere shortening in the peripheral blood of HIV-infected patients.
|
|
| |
| |
|
|
|
