European mitochondrial haplogroups predict liver-related outcomes in patients coinfected with HIV and HCV: a retrospective study

Hepatitis C virus (HCV) infection is the leading cause of end-stage liver disease, hepatocellular carcinoma (HCC), and liver-related death in developed countries [1]. It is estimated that about 10–20% individuals with chronic hepatitis C develop cirrhosis around 20 to 30 years after acquiring HCV and that those who develop cirrhosis have a 1% to 5% annual risk of developing HCC and a 3% to 6% annual risk of liver decompensation [2, 3]. However, chronic hepatitis C is highly variable among individuals, ranging from minimal histological changes to extensive fibrosis and cirrhosis [3].

Although we lack predictive models to estimate the risk of fibrosis and clinical progression in individuals early on in their HCV infection, several factors are associated with increased risk of progression of liver fibrosis, including age, male gender, high alcohol intake, obesity, insulin resistance, type 2 diabetes, coinfection with human immunodeficiency virus (HIV) hepatitis B virus, and immunosuppressive therapy [3]. Host nuclear genetic factors also influence the natural history of HCV infection and include viral clearance, progression of fibrosis, and development of cirrhosis and HCC [3, 4].

In the last decade, the role of mitochondrial genetics in human disease has been increasingly recognized. Mitochondria provide energy to eukaryotic cells via oxidative phosphorylation and regulate cellular survival via control of apoptosis [5]. Mutations in mitochondrial DNA (mtDNA) have been acquired throughout history by natural selection owing to adaptation to environmental conditions [6]. Consequently, the human population can be subdivided into several mitochondrial clades or haplogroups, which are defined based on specific mtDNA polymorphisms [7]. In Europe, macro-haplogroups HV, U, and JT are about 90% of the population [8]. Of them, 50% of the Europeans belongs to the macro-haplogroup HV and 45% are haplogroup H. The macro-haplogroups U is divided into many sub-haplogroups that comprise approximately 20% of the Caucasian population. The macro-haplogroup JT is subdivided in haplogroups J (8% of the population) and T (9% of Europeans).

MtDNA haplogroups have been increasingly recognized as contributors to diseases such as cancer, sepsis, diabetes, and degenerative diseases [9, 10]. However, there is now clear evidence that mtDNA variants within haplogroups may be the trigger of the large number of diseases with which mtDNA haplogroups have been linked. Functional studies are scarce and technically complicated because the likely biochemical effect of mtDNA polymorphisms may be subtle. In addition, many of these mtDNA polymorphisms are found in genomes that contain other polymorphisms, which may be interacting [11]. Cybrid technology is widely used for the study of phenotypical effect of mutations in the mtDNA. In this “in vitro” model, it has been observed that haplogroup H cybrids contain higher levels of mtDNA and mRNA, growing faster, have a higher membrane potential, and consume more oxygen than haplogroup Uk and T cybrids. Other studies have reported that haplogroup J cybrids have slower rate of assembly of the mitochondrial complexes and lower ATP and ROS production than haplogroup H cybrids [11]. In addition, some of these findings have also been described in patients, although the article number is smaller.

Preliminary studies have also shown associations between mtDNA haplogroups and clinical outcomes in patients with HIV infection [12, 13], including clinical progression [14‐17], CD4+ T cell recovery after combination antiretroviral therapy (cART) [18], metabolic disorders [19, 20], and toxicities due to nucleoside reverse-transcriptase inhibitors (e.g., peripheral neuropathy and lipoatrophy) [21, 22]. In previous reports, an association between major European mtDNA haplogroups and liver fibrosis in HIV/HCV-coinfected patients were also found [23, 24], but the design of this studies was cross-sectional.

In this article, with a cohort of HIV/HCV-coinfected patients who were followed up for a median of approximately 8 years, we found that the presence of mtDNA haplogroup T was related to an increased hazard of LREs. We also found that the presence of mtDNA haplogroup H was related to a reduced hazard of LREs in the competing-risks regression analysis, which did not achieve statistical significance in the adjusted regression analysis. To our knowledge, this is the first time mtDNA haplogroups have been found to be associated with LREs in HIV/HCV coinfected patients with chronic hepatitis C. Moreover, since there have been recently reported that the incidence of HCC has increased even after treatment of chronic hepatitis C [33‐36], the results of this study also highlight the importance of mitochondrial genetics in patients with chronic hepatitis C in the future.

Several aspects of mitochondrial function are affected during HCV infection, such as the alteration of mitochondrial membrane potential, excessive production of reactive oxygen species (ROS), the significant fall in the level of adenosine triphosphate (ATP), and bioenergetic failure [37, 38]. Also, HCV-induced mitochondrial dysfunction may contribute to viral persistence by attenuating apoptosis of infected cells [39]. The mtDNA haplogroups tend to be associated with subtle differences in oxidative phosphorylation capacity and the generation of ROS [9]. Therefore, it is conceivable that the chronic oxidative stress due to HCV infection can be modulated by the DNA genetic. Of note, mtDNA haplogroup H has been associated with higher activity in the electron transport chain, leading to more significant quantities of ATP and ROS than other haplogroups, such as mtDNA haplogroup J and T, which exhibits lower energy efficiency [40, 41].

In a previous biopsy-based cross-sectional study of HIV/HCV-coinfected patients using the same cohort of patients [24], we found that mtDNA haplogroup H was strongly associated with reduced likelihood of having advanced fibrosis, cirrhosis, and fibrosis progression rates. In the present study, we have found haplogroup T was related to the development of LREs, whereas that haplogroup H was protective against LREs in Fine and Gray competing-risks regression analysis, but it did not achieve significant p-value in multivariate regression analysis. As discussed above, haplogroup T has a biochemical and energetic effect opposite to haplogroup H [40, 41]; and in our experience, we found opposing associations for other outcome variables, in other studies, when we compared haplogroup T versus haplogroup H in HIV infected subjects [18, 19].

The relevance of our findings for clinical practice can be limited by the fact that the data are from an era in which anti-HCV therapy was interferon-based rather that based on all oral direct-acting antivirals. However, our findings expand the knowledge about the pathogenesis of chronic hepatitis C in HIV-infected patients. Viewed from a broader perspective, our results suggest that mtDNA haplogroups may influence the natural history of chronic liver diseases of various etiologies; a notion that is supported by two recent studies. In the first study, the presence of mtDNA haplogroup L was found to exercise a protective effect against the development of non-alcoholic steatohepatitis and pericellular fibrosis in patients with non-alcoholic fatty liver disease [42]; in the second study, carried out in Chinese population with HCC, patients with haplogroup M8 had a lower survival rate than patients with haplogroup D4 [43]. The fact that a large proportion of the participants were treated with interferon could also limit the study’s implications on the natural history of HCV infection. However, we analyzed the SVR as a time-dependent covariate, and we also performed a sub-analysis discarding patients who achieved SVR, finding equivalent results in both cases. Non-responders to interferon therapy were kept in the study because this fact does not protect against the progression of chronic hepatitis C in the long term [44]. Another limitation is the small sample size, which may have impaired the ability to detect less robust associations, as we did not perform detailed analyses on some of the less common mtDNA haplogroups. Finally, our study was retrospective and consequently cannot prove causality. However, it must be considered that patients included in the study were highly selected and met a set of restrictive criteria for starting HCV treatment. Besides, they were followed in our institution by the same physicians throughout their disease, with clinical and laboratory assessment every 3 to 6 months according to standard practice with HIV-infected patients in Spain [45]. Furthermore, complications of cirrhosis were prevented or managed following protocols based on current clinical practice guidelines.

In conclusion, our data suggest that mitochondrial haplogroups could influence the natural history of hepatitis C and may warrant further confirmatory longitudinal studies.