*Institute of Clinical Infectious Diseases, Catholic University, Rome, Italy
†Institute of Infectious Diseases 2, University Hospital of Siena, Siena, Italy
‡Department of Biomedical Sciences, University of Modena and Reggio Emilia, Modena, Italy
§Department of Primary Care & Population Sciences, Royal Free and University College Medical School, London, UK; and
‖Infectious Diseases Clinic, Azienda Ospedaliero-Universitaria, Modena, Italy
Correspondence to: Andrea De Luca, MD, Institute of Infectious Diseases 2, University Hospital of Siena, Siena, Italy (e-mail: firstname.lastname@example.org).
Supported by the VII Programma Nazionale AIDS, Istituto Superiore di Sanità, Italy, grant number 30G.8 (to A.D.L.) and 30G.62 (to A.C.), and by the Fondi Ateneo 2007, Università Cattolica S Cuore, Rome, Italy (to A.D.L.).
Presented in part at the 11th European AIDS Conference, October 24–27, 2007, Madrid, Italy (poster P4.4/05).
A.D.L. has been a member of advisory boards or has received speaker honoraria from ViiV, Abbott Virology, Janssen-Cilag, Siemens Diagnostics, and Monogram Biosciences; R.C. has been a member of advisory boards or has received speaker honoraria from GlaxoSmithKline, Gilead, Abbott Virology, Boehringer Ingelheim, and Merck; C.M. has been a member of advisory boards or has received speakers honoraria from ViiV, Bristol-Myers-Squibb, Merck-Sharp and Dohme, Gilead Sciences, and Janssen-Cilag.
A. De Luca, R. Cauda, and A. Cossarizza designed the study; A. Cozzi-Lepri did the statistical analysis; A. Cossarizza and C. Mussini designed and coordinated the laboratory examinations; M. Nasi and M. Pinti did the laboratory examinations; S. Di Giambenedetto coordinated the clinical data collection; A. Marzocchetti, M. Fabbiani, and L. Bracciale collected the clinical data, prepared the database, and the clinical samples; A. De Luca wrote the first draft of the manuscript; all the authors contributed to the interpretation of the results and gave relevant input to the intellectual content of the manuscript and approved its final version.
Received June 27, 2011
Accepted October 18, 2011
Potent combination antiretroviral therapy (cART) is responsible for long-term morphologic and metabolic side effects, characterized by lipodystrophy, with lipoatrophy and central fat accumulation.1 The pathogenesis of lipodystrophy is multifactorial.2–6 Nucleoside reverse transcriptase inhibitors are believed to be at least partially responsible of these outcomes through a mechanism of mitochondrial toxicity.7,8
Variations of mitochondrial DNA in the form of single nucleotide polymorphisms and of their combination define different mitochondrial haplogroups, which are differently represented in populations and ethnic groups.9
Human mitochondrial DNA variation has been associated with a number of metabolic, cardiovascular, neurodegenerative, neoplastic, and autoimmune disorders10–16 and with disease progression in HIV-1–infected patients.17 Several candidate genes have been analyzed for their association with the incidence of lipoatrophy and lipoaccumulation during cART.18,19
The aim of the present study was to analyze the association of 9 major European mitochondrial haplogroups with the incidence of any lipodystrophy, of fat accumulation and of lipoatrophy during long-term cART administration.
PATIENTS AND METHODS
Patients and Study Procedures
We assessed the 9 major European mitochondrial DNA haplogroups in peripheral blood mononuclear cells of HIV-1–infected patients observed at the Clinic of Infectious Diseases of the Catholic University in Rome, Italy, initiating cART from 1996 onward, as previously described.20 Mitochondrial DNA belonging to haplogroups different from the major ones were classified as “others.”
Patients peripheral blood mononuclear cells were collected between January 2001 and December 2005. Beginning from January 2000, patients were cross-sectionally evaluated at the outpatient unit of the Clinic using a standardized structured interview and objective evaluation and subsequently prospectively assessed by periodical (6 ± 2 months intervals) interview and physical examination for the development of lipodystrophy distinguishing lipoaccumulation (any abnormal fat accumulation occurring between the neck and the pelvis) and lipoatrophy (any abnormal fat wasting on face, upper and/or lower limbs, and buttocks). Each new patient-reported sign of lipodystrophy had to be confirmed by physical examination or, in the prospective phase, by anthropometric measurements indicative of >5% variation in any circumference. All these signs had to be confirmed at the subsequent visit.
Other patients' variables were extracted from the local observational database of HIV patients. All patients signed a written informed consent for this study, including consent for genetic analysis.
The incidence of lipoatrophy, fat accumulation, and any lipodystrophy was analyzed using a time-to-event approach. Person-years at risk were calculated from the time of cART initiation until the last lipodystrophy assessment or until the development of the event of interest, whichever occurred first.
At the time of cART initiation, patients were assumed to be free of any lipodystrophy. Poisson regression multivariable model was employed to study whether the mitochondrial haplogroups were associated with the risk of developing lipoatrophy, fat accumulation, or any lipodystrophy.
Because patients at risk for lipoatrophy were concomitantly exposed to the competing risk of developing fat accumulation (and vice versa), a competing-risk approach to the analysis was adopted (main analysis).
Alternatively, a cause-specific analysis approach was also used, and results were interpreted accordingly. The main difference is that in the competing-risk approach, patients' follow-up are censored at the date of their last follow-up rather than at the date of occurrence of the competing event.
All analyses were performed using the “genmod” procedure in SAS version 9.1 (SAS, Cary, NC).
Mitochondrial haplogroups were studied in 240 patients (238 white, 2 Black African), but 187 had a prospective assessment of lipodystrophy from the date of cART initiation and were used in this analysis; their baseline characteristics and mitochondrial haplogroup distribution are summarized in Tables 1 and 2.
Over follow-up and after the date of starting cART, the antiretroviral drugs received by the patients were lamivudine (72.7% of the follow-up time), stavudine (43.0%), zidovudine (41.7%), didanosine (19.4%), abacavir (11.7%), and tenofovir (8.5%); nevirapine (14.0%) and efavirenz (13.3%); and unboosted protease inhibitor (38.3%) and boosted protease inhibitor (21% of the follow-up).
Incidence and Predictors of Any Lipodystrophy
Among the 187 patients cumulatively contributing 947 person-years of follow-up (PYFU) (median of 4.67 years of follow-up per patient, range, 0.17–11.08), lipodystrophy was observed in 102 [incidence rate (IR) 10.8 per 100 PYFU, 95% confidence interval (CI): 8.8 to 13.1]. The estimates of the association between mitochondrial haplogroups and other analyzed factors with this outcome are shown in Table 3. Patients carrying mitochondrial haplogroup K, as compared with those carrying the most prevalent haplogroup H, showed a higher crude incidence of any lipodystrophy (IR 27.2 versus 8.8 per 100 PYFU, Fisher exact P = 0.007), whereas no difference was observed comparing H with any of the other haplogroups (P > 0.12).
In a multivariable model adjusting for a number of patients-related factors and antiretroviral variables, patients with haplogroup K showed a 4-fold higher risk of developing any lipodystrophy as compared with haplogroup H carriers. Other variables independently associated with a higher risk of lipodystrophy were female gender, older age, and higher current CD4 values, whereas a longer duration of exposure to tenofovir was associated with a reduced incidence of any lipodystrophy.
Incidence and Predictors of Lipoatrophy
Lipoatrophy developed in 84 subjects during 1022 person-years (IR 8.22, 95% CI: 6.6 to 10.1 per 100 PYFU). The association between the mitochondrial haplogroups and other factors with the incidence of lipoatrophy was analyzed using a competing-risk approach and summarized in Table 3. Patients carrying mitochondrial haplogroup K showed a similar crude incidence of lipoatrophy to that of haplogroup H carriers (IR 6.3 versus 7.2 per 100 PYFU, Fisher exact P = 0.82).
However, after adjusting for potential confounders, patients with haplogroup K showed >2-fold higher risk of developing lipoatrophy as compared with haplogroup H carriers, although not significant (P = 0.09).
A similar but not significantly higher risk for lipoatrophy was observed in haplogroup T carriers as compared with patients with haplogroup H.
Other variables independently associated with a higher incidence of lipoatrophy were higher current CD4 values and the specific nucleoside pair currently in use, whereas a longer duration exposure to tenofovir was, again, associated with a reduced incidence. The same analyses were conducted using the cause-specific approach. The multivariable analysis showed that haplogroup K carriers had an independently higher risk of lipoatrophy, which was 3-fold higher as compared with haplogroup H carriers (hazard ratio 2.99, 95% CI: 1.10 to 8.11, P = 0.03).
Moreover, the independent associations of the other variables with the risk of lipoatrophy were confirmed by this approach. Furthermore, female sex and the older age category were also associated with an independently higher risk of this event.
To verify the hypothesis that the age-associated fat loss may have been different according to gender, we formally tested the interaction between gender and age. We found no evidence for such an interaction: the relative risk of lipoatrophy associated with an age of >35 versus ≤35 years was 1.39 (95% CI: 0.68 to 2.84) in female subjects and 1.12 (95% CI: 0.62 to 2.04) in male subjects (interaction P = 0.66).
Incidence and Predictors of Fat Accumulation
Fat accumulation was observed in 57 patients during 1175 person-years (IR 4.85, 95% CI: 3.7 to 6.2 per 100 PYFU). The association between mitochondrial haplogroups and other factors with the risk of fat accumulation are summarized in Table 3. There was no evidence that the crude incidence of fat accumulation was higher in a certain haplogroup as compared with another. Only in univariate analysis the risk for patients with haplogroup J tended to be higher than that of patients carrying the haplogroup H (IR 9.2 versus 4.1 per 100 PYFU, Fisher exact P = 0.09).
In the multivariable analysis, patients with haplogroup K showed a higher risk of developing fat accumulation as compared with those with haplogroup H carriers, although statistical significance was obtained only when the cause-specific approach to the analysis was employed (hazard ratio 5.27, 95% CI: 1.31 to 21.13, P = 0.02).
Other variables independently associated with a higher risk of fat accumulation in the main competing-risk analysis were a higher current CD4 values and the specific nucleoside pair currently in use, whereas the longer exposure to tenofovir was, again, an independent predictor of a reduced risk of fat accumulation. The same variables showed similar associations with this outcome also when using the cause-specific approach (not shown).
In this largely prospective study, we observed an association between the presence of mitochondrial haplogroup K and the risk of lipodystrophy of any type. When we evaluated fat loss and fat accumulation separately using a competing-risk approach, our analysis showed a similar trend; however, the association was not statistically significant, possibly because of reduced power.
In an alternative cause-specific analysis approach, the associations were stronger and, especially in the analysis of lipoaccumulation, a large difference in the magnitude of risks between patients carrying H and K haplogroups was shown. This association was independent of the other studied patients and treatment-related characteristics. Female patients with higher current levels of CD4 and the type of currently used nucleoside reverse transcriptase inhibitor pair also showed independently higher risks of all lipodystrophy outcomes, whereas older age predicted a higher risk of lipoatrophy only, and longer exposure to tenofovir was associated with a reduced risk of lipodystrophy.
In contrast, current use of tenofovir seemed to be associated with an increased risk of fat distribution. Although this may sound contradictory, this finding may simply be explained by a prescription bias of tenofovir to patients perceived at risk.
We have chosen the competing-risk approach as our main analysis to make the interpretation of results easier as the risk of developing different aspects of the lipodystrophy syndrome (fat loss versus accumulation) are unlikely to be independent. In a smaller prospective study, with a shorter follow-up, haplogroup J carriers tended to be associated with a lower risk of limb fat loss as measured by dual-energy x ray absorptiometry, as compared with patients carrying other mitochondrial haplogroups.21
In our series we could not confirm this association, although there was a tendency for a lower risk when comparing J and H, especially in patients who were currently exposed to either didanosine or stavudine (data not shown). The discrepancy with the previously published report may partially depend on the different outcomes definition and on a different treatment exposure in the 2 study populations.
In another report, mitochondrial haplogroups, which were characterized using an identical technique as in this study,18 were not associated with differences in quantitative body fat measures in patients referred for lipodystrophy. Discrepancies can be explained by study design, because the published report is a prevalent cohort with a cross-sectional design, whereas our study prospectively analyzes incidence.
In a different study, haplogroup H was strongly associated with HIV lipoatrophy, whereas the haplogroup T had a slightly protective role. Potential differences that may well explain the discordances observed with our study are the population that was only composed of male subjects, and the study design that was based on a single cross-sectional assessment of the atrophic component of lipodystrophy.22
Haplogroup K carriers are relatively rare among whites: our 6% prevalence is in line with the existing data (4.1%–13.5%).9,23,24 This haplogroup is defined by nucleotide substitution G9055A in the ATP6 gene of mitochondrial DNA, which leads to an alanine to threonine substitution.
The pathogenic mechanism underlying the association between this specific mitochondrial haplogroup K and the higher incidence of lipodystrophy during cART remains unexplained and requires further investigation. It might be hypothesized that variations of mitochondrial DNA leading to haplogroup classification are associated with a reduced mitochondrial function, which may exacerbate the mitochondrial toxicity of antiretroviral treatment by modulating ATP production. Moreover, mitochondrial DNA variation may be associated with increased apoptosis of adipocytes, which may explain the association with the lipoatrophy component but not with the fat accumulation. Alternatively, haplogroup K may be in linkage disequilibrium with other gene variants involved in the development of fat accumulation or lipoatrophy.20
The association of female gender with lipoatrophy and fat accumulation and of older age with lipoatrophy is not unexpected and confirms previous observations.20,25 The association of the different types of lipodystrophy with higher current CD4 counts is likely to be explained by a longer exposure to cART and consequently to its long-term toxicity. Finally, the inverse correlation between a longer duration exposure to tenofovir and the risk of lipoatrophy and fat accumulation suggests a lower potential for this drug to cause mitochondrial toxicity and abnormal fat accumulation as shown in other studies.3,26
Among the limitations of this study there is the phenotype definition that may have resulted in an underestimate of the true incidence of the lipodystrophy outcomes as compared with studies using imaging techniques and dual-energy x ray absorptiometry for the outcomes definition.26 On the other hand, the alterations perceived by the patient and observed by the clinician could be more relevant in terms of patients' quality of life. Moreover, the validity of the assessment of lipodystrophy using an objective physical examination for analyzing the association with other genetic variants has been previously established.20
Another limitation is represented by limited numbers of observations for each of the haplogroups. The major strength of our analysis is represented by the long duration of the prospective follow-up (5.0–6.1 years, on average, depending on the specific end point).
Should our results be repeated by other studies including patients from different ethnic backgrounds, mitochondrial DNA variation could be investigated to identify genetic predisposition to abnormalities leading to fat redistribution observed during prolonged combination antiretroviral treatment.
1. Carr A, Samaras K, Burton S, et al.. A syndrome of peripheral lipodystrophy, hyperlipidemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS. 1998;12:F51–F58.
2. Haubrich RH, Riddler SA, DiRienzo AG, et al.. Metabolic outcomes in a randomized trial of nucleoside, nonnucleoside and protease inhibitor-sparing regimens for initial HIV treatment. AIDS. 2009;23:1109–1118.
3. Noor MA, Seneviratne T, Aweeka FT, et al.. Indinavir acutely inhibits insulin-stimulated glucose disposal in humans: a randomized, placebo-controlled study. AIDS. 2002;16:F1–F8.
4. Brown TT, Li X, Cole SR, et al.. Cumulative exposure to nucleoside analogue reverse transcriptase inhibitors is associated with insulin resistance markers in the multicenter AIDS cohort study. AIDS. 2005;19:1375–1383.
5. Brinkman K, Smeitink JA, Romijn JA, et al.. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet. 1999;354:1112–1115.
6. Cossarizza A, Moyle G. Antiretroviral nucleoside and nucleotide analogues and mitochondria. AIDS. 2004;18:137–151.
7. Cossarizza A, Mussini C, Viganò A. Mitochondria in the pathogenesis of lipodystrophy induced by anti-HIV antiretroviral drugs: actors or bystanders? Bioessays. 2001;23:1070–1080.
8. Ferraresi R, Troiano L, Roat E, et al.. Protective effect of acetyl-L-carnitine against oxidative stress induced by antiretroviral drugs. FEBS Lett. 2006;580:6612–6616.
9. Torroni A, Huoponen K, Francalacci P, et al.. Classification of European mtDNAs from an analysis of three European populations. Genetics. 1996;144:1835–1850.
10. Rego-Pérez I, Fernández-Moreno M, Fernández-López C, et al.. Mitochondrial DNA haplogroups: role in the prevalence and severity of knee osteoarthritis. Arthritis Rheum. 2008;58:2387–2396.
11. Rosa A, Fonseca BV, Krug T, et al.. Mitochondrial haplogroup H1 is protective for ischemic stroke in Portuguese patients. BMC Med Genet. 2008;9:57.
12. Datta S, Majumder M, Biswas NK, et al.. Increased risk of oral cancer in relation to common Indian mitochondrial polymorphisms and autosomal GSTP1 locus. Cancer. 2007;110:1991–1999.
13. Li XY, Su M, Huang HH, Li H, et al.. mtDNA evidence: genetic background associated with related populations at high risk for esophageal cancer between Chaoshan and Taihang Mountain areas in China. Genomics. 2007;90:474–481.
14. Bai RK, Leal SM, Covarrubias D, et al.. Mitochondrial genetic background modifies breast cancer risk. Cancer Res. 2007;67:4687–4694.
15. Wallace DC. A mitochondrial paradigm of metabolic and degenerative diseases, aging, and cancer: a dawn for evolutionary medicine. Annu Rev Genet. 2005;39:359–407.
16. van der Walt JM, Nicodemus KK, et al.. Mitochondrial polymorphisms significantly reduce the risk of Parkinson disease. Am J Hum Genet. 2003;72:804–811.
17. Hendrickson SL, Hutcheson HB, Ruiz-Pesini E, et al.. Mitochondrial DNA haplogroups influence AIDS progression. AIDS. 2008;22:2429–2439.
18. Bonnet E, Genoux A, Bernard J, et al.. Impact of genetic polymorphisms on the risk of lipid disorders in patients on anti-HIV therapy. Clin Chem Lab Med. 2007;45:815–821.
19. Zanone Poma B, Riva A, Nasi M, et al.. Genetic polymorphisms differently influencing the emergence of atrophy and fat accumulation in HIV-related lipodystrophy. AIDS. 2008;22:1769–1778.
20. Nasi M, Guaraldi G, Orlando G, et al.. Mitochondrial DNA haplogroups and highly active antiretroviral therapy-related lipodystrophy. Clin Infect Dis. 2008;47:962–968.
21. Hulgan T, Tebas P, Canter JA, et al.. Mitochondrial haplogroups, and peripheral lipoatrophy during antiretroviral therapy. J Infect Dis. 2008;193:858–866.
22. Hendrickson SL, Kingsley LA, Ruiz-Pesini E, et al.. Mitochondrial DNA haplogroups influence lipoatrophy after highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2009;51:111–116.
23. Wiesbauer M, Meierhofer D, Mayr JA, et al.. Multiplex primer extension analysis for rapid detection of major European mitochondrial haplogroups. Electrophoresis. 2006;27:3864–3868.
24. Manwaring N, Jones MM, Wang JJ, et al.. Prevalence of mitochondrial DNA haplogroups in an Australian population. Intern Med J. 2006;36:530–533.
25. Galli M, Cozzi-Lepri A, Ridolfo AL, et al.. Incidence of adipose tissue alterations in first-line antiretroviral therapy: the LipoICoNa study. Arch Intern Med. 2002;162:2621–2628.
26. Law M, Puls R, Cheng AK, et al.. Evaluation of the HIV lipodystrophy case definition in a placebo-controlled, 144-week study in antiretroviral-naive adults. Antivir Ther. 2007;11:179–186.
© 2012 Lippincott Williams & Wilkins, Inc.