Introduction

According to the latest data from the world health organization (WHO), in 2022, the human immunodeficiency virus (HIV) still affected 39 million people1. Hepatitis C (HCV) coinfection in patients living with HIV (PLHIV) is quite common, since the two viruses share routes of transmission.

HIV and HCV co-infection causes chronic activation of the immune system, leading to a persistent and deregulated response to the body’s inability to eradicate the virus. In the case of HIV infection, the constant replication of the virus and the destruction of CD4 + T cells induce generalized inflammation that is unable to resolve the HCV infection, causing a progressive exhaustion of immune cells, such as T cells and NK cells, and an increase in plasma soluble activation mediators such as IP-10, TNF-α, IL-6, IL-β, IL-7, INF-γ, as well as proinflammatory cytokines. The persistence of HCV infection together with these molecules, is associated with the development of liver-related and non-related co-morbidities.

Fortunately, direct active antivirals (DAAs) allow a successful elimination of HCV in majority of patients, including HIV/HCV coinfected individuals2. However, the presence of co-morbidities associated with HIV/HCV coinfection remains a public health concern because patients who eliminate the virus, either through treatment or spontaneously, do not always improve their health status and are at risk of developing co-morbidities associated with liver disease (cirrhosis, liver failure, hepatocellular carcinoma)3,4,5,6, cardiovascular diseases (ischemic heart disease, peripheral vascular disease, stroke)7,8,9, disorders of glycemic and lipid metabolism (dyslipidemia, diabetes)10,11,12,13,14, renal complications (glomerulonephritis, chronic kidney disease)15, and neurocognitive disorders (CNS involvement, cognitive impairment, neurological disorders)16, among others.

HIV and HVC infections are accompanied by changes in glucose and lipid metabolism that favour the alteration of metabolism-related markers that have been considered potential biomarkers of pathogenesis in HIV and HCV infection. For instance, the cytokine related to Neutrophil Gelatinase-Associated Lipocallin (NGAL), involved in the regulation of glycolipid and energy metabolism, is an indicator of acute-chronic liver failure17,18,19. Similarly, the retinol-Binding Protein-4 (RBP-4), involved in hepatocyte glycogenesis, has been associated with liver damage in HCV infection20. Likewise, elevated levels of vascular endothelial-related cell adhesion molecules, such as soluble intercellular adhesion molecule-1 (sICAM-1), which is a promoter of atherosclerosis processes, has been related to an increased risk of cardiovascular disease21,22,23. Similarly, alteration of cytokines involved in carbohydrate metabolism, such as Resistin24 and Peptide-C25, could imply the development of Insulin resistance, that can progress to type 2 diabetes. These molecules play a crucial role in glucose regulation, and the metabolic dysfunction resulting from their alteration not only affects glucose homeostasis, but also the appearance of hypertension and dyslipidemia.

On the other hand, there are protective markers such as adiponectin. Serum adiponectin concentrations are negatively associated with obesity, diabetes (type 2), and cardiovascular disease26 Adiponectin is also inversely associated with other traditional cardiovascular risk factors, such as blood pressure, low-density lipoprotein cholesterol and triglyceride levels27 and it is positively related to high-density lipoprotein cholesterol levels28. Recent research has indicated that adiponectin has anti-inflammatory properties, producing the anti-inflammatory mediator interleukin (IL)-10 in primary human monocytes, monocyte-derived macrophages and dendritic cells29. In addition, adiponectin significantly affected the production of the proinflammatory cytokine interferon-γ in human macrophages29.

Given the increased risk of developing the aforementioned comorbidities associated with HIV/HCV co-infection, the early detection of metabolic dysfunction is paramount to enable timely and effective intervention, thereby mitigating the progression to severe comorbidities. Consequently, this study aims to elucidate the impact of HCV clearance, achieved either through direct-acting antiviral (DAA) therapy or spontaneous resolution, on metabolic biomarkers in PLHIV, with the objective of potentially anticipating to the future onset of associated comorbidities.

Material and methods

Study design

We carried out a multicentre, prospective observational study from the Multidisciplinary Group of viral coinfection HIV/Hepatitis (COHIVEP) (Supplementary Data 1) from four Public Spanish Hospitals in Madrid Autonomous Community: Hospital Universitario La Paz, Hospital Doce de Octubre, Hospital Universitario Infanta Leonor and Hospital Universitario La Princesa. Samples were processed at the National Center for Microbiology, Institute of Health Carlos III, Madrid (Spain).

One hundred sixteen PLVIH with different status to HCV infection were enrolled: 1) CHC group (n = 45): PLHIV chronically infected with HCV (HIV + /HCV +) (positive PCR and positive HCV antibodies), naïve to any HCV treatment; 2) SC group (n = 36): PLHIV who had been exposed to HCV but experienced spontaneous viral clearance during the first 6 months after HCV infection (negative PCR and positive HCV antibodies, in the absence of anti-HCV treatment); 3) PLHIV control group (n = 35): HIV monoinfected patients that had never been in contact with HCV (negative PCR and negative HCV antibody). CHC patients were evaluated at baseline, before starting HCV antiviral therapy with DAAs, and at the end of the follow-up, 48 weeks after the achievement of sustained virological response (SVR) (n = 23). SC group was sampled with the same baseline and follow-up criteria (n = 12). (Fig. 1).

Fig. 1
figure 1

Flow chart of patients involved in this study. Patients were grouped according to their viral status: Chronic HCV (CHC) who cleared HCV with direct-acting antivirals agents (DAAs) and HIV + patients who spontaneously clarify HCV infection (SC). Samples were collected at baseline (CHCb and SCb) and 48 weeks after achieving a sustained virological response (CHCf) or one year of follow-up (SCf). A control group PLHIV (HIV +) without a previous HCV infection was included.

Full size image

Loss to follow-up was due to several factors, such as change of hospital, declining informed consent or death.

All patients were under ART during at least one year before baseline, undetectable for HIV and with CD4 + T-cells counts ≥ 500 cells/mm3.

Exclusion criteria were as follows: pregnancy, individuals below 18 years old, previously HCV treatment, no advance liver fibrosis (> F3), clinical evidence of hepatic decompensation, active drug or alcohol addiction, alcohol-induced liver injury, HBV active infection, opportunistic infections, and other concomitant diseases such as diabetes, neoplasia, autoimmune disease, among others.

Clinical and epidemiology data were obtained from medical records. No comorbidities were recorded as they were exclusion criteria.

Factors and outcome variables

The following metabolic biomarkers were studied as primary factors (n = 14): Adiponectin, C-Peptide, Cortisol, FGF-23, Ghrelin, GLP1, Insulin, Leptin, NGAL, PYY, RBP4, Resistin, sICAM-1 and ZAG (complete list available at Supplementary Data 2).

Their levels were analysed on each group as follow: i) Metabolic biomarkers at baseline, CHCb and SCb with respect to HIV control group; ii) Metabolic biomarkers evolution in CHC (CHCf vs CHCb) and SC (SCf vs SCb); iii) Metabolic biomarkers at follow-up, CHCf and SCf with respect to HIV control group.

Sample preparation

Peripheral venous blood samples were collected in EDTA tubes. Plasma was obtained after density gradient centrifugation, clarified and storage at −80 °C until use.

Metabolism biomarkers

We characterized 14 metabolism markers in plasma at the same time by multiplex immunoassay (Luminex xMAP technology) using the Bio-plex 200TM system (BioRad Laboratories, Hercules, California, USA) following the manufacturer’s specifications. Briefly, color-coded superparamagnetic beads coated with analyte-specific antibodies that recognize the different target analytes are mixed together and incubated with the sample. Captured analytes are subsequently detected using a cocktail of biotinylated detection antibodies and a streptavidin–phycoerythrin conjugate. We used the raw fluorescence intensity (FI) values as a relative quantification of the analyte abundances, as previously described30.

Statistical analysis

The available sample size in both the experimental and control groups is sufficient for the proposed analysis, as calculated by granmo software (https://www.datarus.eu/aplicaciones/granmo/). This sample allow us to achieve a statistical power greater than 0.8 with a risk alpha of 0.05 in a two-sided test, as a minimum sample size of 22 participants at baseline and 11 at follow-up is required to detect a statistically significant difference of 1.2 units or greater. Assuming a common standard deviation of 1 and a correlation coefficient of 0.4 between baseline and follow-up measurements and anticipating a 10% attrition rate. Similarly, for the control group, a minimum of 12 participants is necessary to satisfy the same statistical criteria.

For the descriptive study of clinical and epidemiological data of the patients, continuous variables were summarized as median, and categorical as frequency and percentage. Significant differences between categorical data were calculated using the chi-squared test or Fisher´s exact test. Kruskal–Wallis and Mann–Whitney U tests were used to compare continuous variables among independent groups. We estimated differences in metabolism biomarkers levels by a generalized linear model (GLM) adjusted by sex for non-paired comparisons either at baseline (CHCb vs. HIV and SCb vs. HIV) or after 48 of follow-up (CHCf vs. HIV and SCf vs. HIV). No additional confounders were significant. A generalized linear mixed model (GLMM) was used for dependent groups in the longitudinal comparison (CHCf vs CHCb and SCf vs. SCb), both with gamma distribution (log-link). Differences were identified by a statistically significant p-value < 0.05 adjusted by false discovery rate (FDR) using Benjamin-Hochberg correction. IBM® SPSS Statistics (v.19) and statistical software R (v 3.2.0) (www.r-project.org) were used for all statistical analyses.

Ethics statement

The study was conducted in accordance with the Declaration of Helsinki; all patients gave their written consent before enrolment, and the Research Ethics Committee of the Institute of Heath Carlos III approved the study (CEI PI 81_2017-v3). Sample processing and clinical data handling was performed by a coding procedure (reversible dissociation) by unlinking the identifying information and replacing it with a code. Only the physician, and never the principal investigator, was able to identify the participants included in the study. The confidentiality of the information were always preserved, in accordance with Spanish and European legal requirements.

Results

Epidemiological and clinical characteristics of the patients

Overall, 55 percent were male with a median age of 50 years and the most used antiretroviral treatment was integrase strand transfer inhibitor (INSTI) (40.9%). Regarding lipid metabolism and liver function CHC group showed significantly lower glucose level than the SC subjects (p = 0.037) and higher values of liver function parameters such as AST, ALT and GGT (p < 0.001), and bilirubin values (p = 0.029). SC group showed higher values of albumin (p = 0.044). A complete characterization of the cohort can be found in Lara-Aguilar et al. 202331.

Biochemical parameters related to metabolic condition at baseline

First of all, CHC and SC groups were independently analyzed at baseline with respect to the HIV control group.

  • Overall, the CHCb group showed significantly higher levels of markers. Particularly in adiponectin (aAMR = 1.29), sICAM-1 (aAMR = 3.45) and NGAL (aAMR = 1.57), with respect to HIV group. On the contrary, lower levels of GLP1 (aAMR = 0.74) were observed (Fig. 2; Supplementary Data 3).

  • Regarding SCb group, only significantly lower levels of GLP1 were observed (aAMR = 0.70) with respect to the HIV group (Fig. 2; Supplementary Data 4).

  • We subsequently compared both CHCb and SCb groups. Lower expression of adiponectin (aAMR = 0.70) was observed in CHC (Fig. 2; Supplementary Data 5).

Fig. 2
figure 2

Comparison of plasma metabolism biomarkers at baseline between the three study groups. CHCb, chronic CHC group at baseline, SCb, spontaneous clearers at baseline.

Full size image

Evolution of metabolic biomarkers after HCV elimination (follow-up vs baseline)

Secondly, CHC and SC groups were independently analyzed at follow-up with respect to the HIV control group.

  • The HCV elimination with DAAs in CHC individuals leads to a significant reduction of adiponectin (aAMR = 0.54), NGAL (aAMR = 0.41), PYY (aAMR = 0.80], RBP4 (aAMR = 0.31) and sICAM-1 (aAMR = 0.17) whereas an increase in cortisol levels was observed after 48 weeks of SVR (aAMR = 1.71) (Fig. 3; Supplementary Data 6).

  • Regarding SC dynamics, only sICAM-1 showed a significant reduction during the follow-up (aAMR = 0.68) (Fig. 3; Supplementary Data 7).

Fig. 3
figure 3

Evolution of plasma metabolism biomarkers in HCV patients at baseline and after 1 year of follow-up after HCV clearance with DAAs. SC, spontaneous cleares; CHC, chronic HCV group.

Full size image

Biochemical parameters related to metabolic condition after HCV elimination (follow-up vs control)

Thirdly, CHC and SC groups were analyzed at follow-up with respect to their baseline groups.

  • At the end of the follow-up (48 weeks after achievement of SVR), CHCf group showed lower levels of RBP4 (aAMR = 0.38) compared to HIV + individuals (Fig. 4; Supplementary Data 8) whereas the adiponectin, NGAL and sICAM levels, significantly increased at baseline, were like the HIV group after HCV elimination (at 48 weeks after SVR).

  • By contrast, the SCf group did not show any significant differences compared to the HIV group (Fig. 4; Supplementary Data 9).

  • Finally, no significant differences were identified between CHCf and SCf groups, suggesting a similar normalization of metabolism plasma biomarkers after either chronic or acute infection (Fig. 4; Supplementary Data 10).

Fig. 4
figure 4

Comparison of plasma metabolism biomarkers at the end of the follow-up between the three study groups. CHCf, chronic CHC group at the end of follow-up, SCf, spontaneous clearers at the end of follow-up.

Full size image

Discussion

Our study characterizes the plasma profile evolution of biomarkers related to lipid and glucidic metabolism in PLHIV with different histories of HCV infection, an acute infection followed by spontaneous HCV clarification and a chronic infection followed by HCV elimination with DAAs.

Our results indicate that CHC infection in PLHIV leads to an increase in the levels of sICAM, adiponectin and NGAL, cytokines involved into the lipid metabolism. sICAM levels, were 3 times higher in CHC patients than in the control group, which indicates a greater inflammatory state in PLHIV with an active CHC infection. The results of our study are in line with previous reports in which HIV/HCV co-infection induces alterations in serum sICAM levels32. There is an association between increased serum levels of sICAM and an elevated inflammatory state, leading to an increased risk of cardiovascular disease32,33. Similarly, increased levels of siCAM in HIV/HCV co-infected individuals, led to the development of advanced liver fibrosis34. Therefore, siCAM could be considered as a marker of liver progression.

Higher adiponectin levels during HIV/HCV coinfection are observed despite lower BMI values. HCV infection promotes higher adiponectin levels, regardless of liver fibrosis or HCV viral load35 and Ndombi et al. also described higher adiponectin levels in HCV-monoinfected individuals compared to HIV-monoinfected subjects who had not received antiretroviral treatment36. Altered adiponectin levels could suggest a state of altered adiposity, although depending on the location of the adiposity (central or peripheral), the relationship between BMI and adiponectin values may differ36,37 indicating a modification of the anti-inflammatory response38, since adiponectin is an adipokine with marked anti-inflammatory properties39. The elevated levels of adiponectin in CHC group may suggest the existence of a compensatory mechanism in response to the generalized inflammation caused by coinfection. In line with these results, the higher levels of adiponectin has been related to insulin sensitivity and the greater protection against inflammation40.

NGAL is linked to a negative regulation of inflammatory activity41 and it is associated with acute kidney injury42. Thus, the higher levels of NGAL observed in CHC could suggest greater renal dysfunction risk than HIV + monoinfected individuals.

Otherwise, HCV infection promotes the reduction of the glucose metabolism-related marker, glucagon-like peptide-1 (GLP-1)43, an effect observed in the SC group. GLP-1, primarily associated with carbohydrate metabolism, is an intestinal hormone synthesized in the L cells of the intestine, and its physiology is based on controlling blood glucose levels. However, it also plays multiple roles in metabolic homeostasis after nutrient absorption44,45. Lower GLP-1 levels in HCV individuals who have clear HCV spontaneously, could suggest that the mentioned regulatory characteristic of glucose homeostasis might be reduced in these patients. On the other hand, the fact that there are significant differences only in the plasma levels of GLP1 in the SC group compared to the HIV-monoinfected group could suggest that spontaneous clearance does not cause the same degree of metabolic dysfunction as active HCV coinfection.

HCV clearance by DAA treatment resulted in a significant reduction of the levels of adiponectin, NGAL and sICAM biomarkers one year after SVR, reaching similar levels to the HIV control group. In line, previous studies have shown that in coinfected individuals, levels of proinflammatory cytokines such as IP-10, ICAMs, IL-8, IFN-gamma, and CD23a significantly decrease after HCV elimination, after being treated with DAA therapy in addition to the antiretroviral treatment, reaching similar levels to those observed prior to HCV infection46. Likewise, the HCV elimination with DAAs in HIV/HCV-coinfected individuals with advanced cirrhosis also promoted an improvement in plasma biomarkers (inflammation, coagulopathy, and angiogenesis) as well as the severity of advanced cirrhosis47. The reduction in systemic inflammation due to HCV clearance leads to a reduction in plasma levels of proinflammatory cytokines in PHIV, which could lead to a decreased risk of accelerated comorbidities associated with HIV/HCV co-infection.

In contrast, cortisol levels increased 48 weeks after DAA treatment, despite SVR having been achieved in the CHC group. The alteration in cortisol levels after HCV clearance could be due, at least in part, to a readjustment of the hypothalamic–pituitary–adrenal (HPA) axis. During active HCV co-infection, as at baseline, a state of generalized inflammation was observed, characterized by elevated levels of pro-inflammatory cytokines such as IL1, IL6 or TNF α, which modulate the activity of the HPA axis by stimulating the release of corticotropin-releasing hormone (CRH) in the hypothalamus, which activates the secretion of adrenocorticotrophic hormone (ACTH) in the pituitary gland and, consequently, increasing cortisol production in the adrenal glands48. However, the activity of the HPA axis during active HCV infection could be inhibited by negative feedback mechanisms, and when the cause of inflammation, i.e. HCV, is removed, the system may react with a compensatory increase in cortisol, which would justify the results obtained49.

On the other hand, cortisol is related to stress and depression states50,51 commons in diseases like HIV and HCV. The presence of such infections in the organism damages the central nervous system52 and leads to cognitive and depressive disorders53. Moreover, HIV/HCV coinfection increases the prevalence of depression, which is more severe than in cases of single infection54. HIV/HCV coinfected individuals on ART, especially those with depressive symptoms, have a higher risk of non-adherence to treatment, possibly due to the lack of adequate antidepressant treatment55, which constitutes a problem of enormous magnitude for the control of the disease caused by HIV.

To accurately interpret our findings, it is essential to acknowledge certain limitations. First of all, we have to consider that the available sample size could reduce our statistical power to detect smaller differences. Secondly, the lack of potential unmeasured confounders (e.g., diet, physical activity, ART regimens) could potentially affect our results. Additionally, we have used appropriate statistical models adjusted for clinical and epidemiological relevant covariates, which improves the robustness of our results. It has not been possible to record long-term development of central nervous system alterations, so at this stage we cannot determine whether the reduced cortisol levels observed after HCV clearance with DAA could accelerate the onset of cognitive and depressive disorders in these patients.

Further investigations with longer follow-up would be necessary to confirm a persistent cortisol disturbance, which could lead to the emergence of more severe consequences associated with metabolic and mental disorders in PHIV after HCV elimination.

Conclusion

Significant increased values of adiponectin, NGAL, sICAM were observed in CHC in HIV/HCV coinfected individuals, unlike the absence of elevated baseline levels of the cytokines assessed in patients with spontaneous clearance. However, 48 weeks after elimination of HCV with DAAs, CHC group experience a generalized reduction of the metabolic cytokine values, reaching similar levels of the control group. However, cortisol levels increased after 48 weeks of follow up since HCV elimination with DAAs in CHC group which could accelerate premature metabolic and mental dysfunctions and associated complications.