The authors have declared that no competing interests exist.
Conceived and designed the experiments: AH TR JW MS AS. Performed the experiments: AH CK BAP MM WW. Analyzed the data: AH AT HH GP. Contributed reagents/materials/analysis tools: AT HH TR GP AKW. Wrote the paper: AH WW TR GP MS AS. Data interpretation: HH JW AKW. Statistical advice: HH.
Calcineurin-inhibitors and hepatitis C virus (HCV) infection increase the risk of post-transplant diabetes mellitus. Chronic HCV infection promotes insulin resistance rather than beta-cell dysfunction. The objective was to elucidate whether a conversion from tacrolimus to cyclosporine A affects fasting and/or dynamic insulin sensitivity, insulin secretion or all in HCV-positive renal transplant recipients.
In this prospective, single-center study 10 HCV-positive renal transplant recipients underwent 2h-75g-oral glucose tolerance tests before and three months after the conversion of immunosuppression from tacrolimus to cyclosporine A. Established oral glucose tolerance test-based parameters of fasting and dynamic insulin sensitivity and insulin secretion were calculated. Data are expressed as median (IQR).
After conversion, both fasting and challenged glucose levels decreased significantly. This was mainly attributable to a significant amelioration of post-prandial dynamic glucose sensitivity as measured by the oral glucose sensitivity-index OGIS [422.17 (370.82–441.92) vs. 468.80 (414.27–488.57) mL/min/m2, p = 0.005), which also resulted in significant improvements of the disposition index (p = 0.017) and adaptation index (p = 0.017) as markers of overall glucose tolerance and beta-cell function. Fasting insulin sensitivity (p = 0.721), insulinogenic index as marker of first-phase insulin secretion [0.064 (0.032–0.106) vs. 0.083 (0.054–0.144) nmol/mmol, p = 0.093) and hepatic insulin extraction (p = 0.646) remained unaltered. No changes of plasma HCV-RNA levels (p = 0.285) or liver stiffness (hepatic fibrosis and necroinflammation, p = 0.463) were observed after the conversion of immunosuppression.
HCV-positive renal transplant recipients show significantly improved glucose-stimulated insulin sensitivity and overall glucose tolerance after conversion from tacrolimus to cyclosporine A. Considering the HCV-induced insulin resistance, HCV-positive renal transplant recipients may benefit from a cyclosporine A-based immunosuppressive regimen.
ClinicalTrials.gov
Post-transplant diabetes mellitus (PTDM) affects 5–35% of all renal transplant recipients (RTRs) and leads to an attenuated graft function, reduced graft and patient survival and increased cardiovascular mortality [
Although an increased diabetogenicity of TAC compared to CyA is generally acknowledged, the detailed pathophysiological mechanisms underlying the distinct glucometabolic effects of both CNIs, i.e. the differential regulation of insulin sensitivity and/or insulin secretion by TAC and CyA, remain to be determined. The vast majority of previous studies evaluating glucose tolerance in RTRs employed parameters of fasting insulin sensitivity, like the homeostasis model assessment of insulin resistance (HOMA-IR) or the quantitative insulin sensitivity check-index (QUICKI), but not of dynamic/glucose-stimulated insulin sensitivity and insulin secretion. This, however, is pivotal for the evaluation of glucose tolerance as fasting insulin sensitivity primarily reflects hepatic insulin resistance but not insulin resistance of two other major insulin-sensitive tissues skeletal muscle and adipose tissue [
Several studies provided strong evidence that chronic HCV infection promotes insulin resistance directly via interference with the intracellular insulin signalling cascade, increase of proinflammatory cytokines or downregulation of the glucose transporter-4 and -2 in skeletal muscle and liver. In addition, HCV induces hepatic fibrosis, which itself facilitates insulin resistance [
The objective of this study was to prospectively evaluate whether the conversion from TAC to CyA (i) alters glucose metabolism in HCV-positive RTRs and (ii) to assess potential underlying mechanisms of fasting and dynamic insulin sensitivity and insulin secretion.
In this prospective, single-center, open study, all HCV-positive RTRs (n = 46, including 13 subjects with known PTDM, two with pre-transplant type 2 diabetes mellitus and 31 without known overt diabetes mellitus), who were admitted at the outpatient department of the Division of Nephrology and Dialysis, Department of Medicine III, Medical University of Vienna, between 01-July-2011 and 31-Aug 2012, were assessed for eligibility (
After an 12 hours’ overnight fast, 2h-75g-OGTTs with measurement of glucose, insulin and C-peptide levels at 0/30/60/90/120 minutes were performed in 10 HCV-positive RTRs before and three months after the conversion of immunosuppression from daily oral TAC [Prograf capsules (twice daily) or Advagraf capsules (once daily), Astellas Pharma GmBH] to twice daily oral CyA (Sandimmun capsules, Novartis GmBH). In an effort to decrease plasma HCV-RNA levels, CyA was initially titrated to the highest tolerated dose aiming at C2-levels of 416.4–833.0 nmoL/L and starting with 3–5 mg/kg/d (limited by an increase of serum creatinine and bilirubin by more than 20% from the pre-conversion level); three and ten days after the conversion of immunosuppression median CyA-trough levels were 89.64 (83.30–125.78) nmol/L and 167.85 (117.04–203.67) nmol/L, respectively, and median C2-levels were 506.46 (458.98–534.99) nmol/L and 619.34 (550.82–762.20) nmol/L, respectively. To avoid possible influences on glucose metabolism, concomitant immunosuppressants [mycophenolate acid (mean dose: 762.0±175.2 mg/d, range: 0–1500 mg/d), steroids (mean prednisone dose: 2.88±0.75 mg/d; range: 0–5 mg/d] and non-immunosuppressive medication known to influence glucose tolerance including DPP4-inhibitors, pre-mixed insulin, beta-blocking substances, renin-angiotensin-inhibitors, diuretics, statins, antidepressants/neuroleptics and levothyroxine-sodium, remained unaltered during the study period. Due to the potentially differential inter-individual resorption and the known alteration of glucose tolerance by CNIs the morning dose of the immunosuppressants was withheld until after the completion of the OGTT (trough level-conditions). Similarily, the morning dose of sitapliptin and pre-mixed insulin was withheld in the two subjects with known PTDM (last administration >24 hours and >12 hours, respectively).
Plasma concentrations of TAC and CyA were measured using CMIA and ACMIA, respectively, at certified laboratories of the Department of Laboratory Medicine, Medical University of Vienna.
Established OGTT-derived, model-based indices were employed to evaluate glucose tolerance. From glucose, insulin and C-peptide concentration measurements the following parameters of insulin kinetics were obtained: area under the curve (AUC) for glucose, insulin and C-peptide (by trapezoidal integration), total beta-cell insulin secretion rate (TIS), its suprabasal component (dynamicTIS) and hepatic insulin extraction (HIE) [
Estimated glomerular filtration rate (eGFR) was calculated according to the modification of diet in renal disease (MDRD)-formula.
Hepatic necroinflammatoin and function was estimated by measurement of standard biochemical liver parameters. For better characterization of our study collective, the Fib-4 index, a non-invasive score to assess liver fibrosis with was calculated as [age (years) x AST (U/L)] / [platelets (109/L) x ALT (U/L)½] [
In addition, liver stiffness was assessed by transient elastography (FibroScan, EchoSens, Paris, France) in eight subjects, as described previously [
The study protocol was performed in accordance to the declaration of Helsinki and approved by the Ethics Committee of the Medical University of Vienna (EK-No. 477/2011) and by the Austrian Federal Office for Safety in Health Care (AGES). The clinical and research activities being reported are consistent with the Principles of the Declaration of Istanbul as outlined in the “Declaration of Istanbul on Organ Trafficking and Transplant Tourism”. All study participants gave written informed consent. The study was registered at
The primary outcome variable was dynamic insulin sensitivity parameter OGIS; secondary outcome parameters included QUICKI, IGI, DI and AI. Wilcoxon signed-rank test was used to evaluate the intra-individual effects of the CNI-conversion on primary and secondary outcome variables as well as supportive parameters, including parameters of renal and hepatic function, before as compared to after conversion of immunosuppression. Spearman’s correlation coefficients were applied. Mann-Whitney U Test was used to assess differences in changes (Δ) of the primary and secondary outcome parameters between subjects with PTDM and those with normal glucose tolerance. Statistical analysis was performed using SAS Enterprise Guide 4.2 (SAS Institute Inc., Cary, NC) and IBM SPSS v23 statistical software (IBM Corp., Armonk, NY). Data are expressed as median (IQR). p<0.05 was considered statistically significant.
Considering insulin sensitivity as the primary endpoint, a post-hoc analysis found that a sample size of 10 had 99% power to detect a difference in OGIS of 47 ml min-1m-2, with a standard deviation of 31 using a paired
Compared to baseline values under the initial TAC-treatment, no significant differences in BMI, eGFR, urinary protein-creatinine-ratio, serum concentrations of creatinine, albumin, HDL-C, LDL-C, C-reactive protein, or thyroid hormones were observed three months after the conversion to CyA (
Although PTDM was known in only three study participants at time of inclusion, another subject featured overt diabetes at the initial OGTT.
Baseline characteristics and parameters of renal graft function and liver function before and three months after the conversion from tacrolimus to cyclosporine A. Data are expressed as median (IQR).
Parameter (n = 10) | before conversion | after conversion | p-value |
---|---|---|---|
age (years) | 51.32 (43.19–57.99) | ||
time since renal transplantation (years) | 8.31 (5.60–12.36) | ||
number of renal transplant (n) | 1.00 (1.00–2.00) | ||
CDC-PRA, latest (%, n = 9) | 4.00 (0–21.50) | ||
renal replacement therapy (years) | 4.211 (2.322–5.561) | ||
creatinine (μmoL/L) | 139.83 (107.09–164.61) | 147.80 (123.02–177.89) | 0.114 |
GFR (MDRD, mL/min) | 48.22 (40.80–57.51) | 42.56 (38.00–56.52) | 0.139 |
urinary protein-creatinine-ratio (mg/g) | 182.00 (133.00–1179.50) | 471.00 (134.50–1267.00) | 0.779 |
tacrolimus trough level (nmoL/L) | 6.283 (5.429–7.351) | ||
cyclosporine A trough level (nmoL/L) | 93.30 (71.22–155.56) | ||
C2 level (nmoL/L) | 344.03 (329.04–645.58) | ||
body mass index (kg/m²) | 26.25 (20.76–29.42) | 27.17 (20.59–29.64) | 0.779 |
C-reactive protein (mg/L) | 0.950 (0.500–3.025) | 1.200 (0.375–3.250) | 0.779 |
TSH (mIU/L) | 2.330 (1.750–3.710) | 2,620 (1.610–3.840) | 0.445 |
albumin (g/L) | 404.0 (390.3–438.3) | 421.5 (399.3–436.3) | 0.153 |
total cholesterol (mmoL/L) | 4.102 (3.496–4.599) | 5.070 (3.818–5.966) | 0.028 |
LDL-C (mmoL/L) | 2.198 (1.731–2.655) | 2.828 (2.091–3.786) | 0.051 |
HDL-C (mmoL/L) | 1.032 (0.890–1.367) | 1.019 (0.935–1.393) | 0.674 |
triglycerides (mmoL/L) | 1.174 (0.909–1.875) | 1.682 (1.211–2.890) | 0.047 |
total bilirubin (μmoL/L) | 11.97 (7.49–21.00) | 17.10 (9.03–23.22) | 0.066 |
aspartate aminotransferase (U/L) | 29.50 (22.75–39.50) | 23.00 (20.00–39.50) | 0.047 |
alanine aminotransferase (U/L) | 30.00 (24.75–44.75) | 22.00 (17.50–25.75) | 0.028 |
gamma-glutamyl transferase (U/L) | 54.50 (19.50–141.75) | 36.50 (14.50–147.50) | 0.139 |
normotest (%) | 114.00 (96.50–134.50) | 122.00 (107.50–141.50) | 0.123 |
HCV-PCR (copies/mL) | 546000 (351750–2965000) | 2560000 (798000–3370000) | 0.285 |
liver stiffness (kPA, n = 8) | 6.80 (5.60–13.65) | 10.25 (6.15–20.43) | 0.463 |
FIB-4 | 2.00 (1-60-4.03) | 2.25 (1.58–4.20) | 0.337 |
Following the conversion of CNIs, fasting and challenged plasma glucose levels significantly decreased (
Determination of plasma glucose (A), serum insulin (B) and C-peptide (C) levels by OGTTs performed prior to (full symbols) and three months after (blank symbols) the conversion from taxrolimus to cyclosporine A. Data are expressed as median (IQR). + indicates p <0.05 (in detail for plasma glucose levels: p = 0.012 at 0min, p = 0.025 at 30 min, p = 0.059 at 60 min, p = 0.022 at 90 min, p = 0.028 at 120 min).
Dynamic insulin sensitivity parameter OGIS increased highly significantly post-conversion (p = 0.005,
Insulin sensitivity, insulin secretion and other parameters of glucose tolerance before and three months after the conversion from tacrolimus to cyclosporine A. Data are expressed as median (IQR).
Parameter (n = 10, respectively) | before conversion | after conversion | p-value |
---|---|---|---|
OGIS (mL/min/m2) | 422.17 (370.82–441.92) | 468.80 (414.27–488.57) | 0.005 |
QUICKI | 0.416 (0.402–0.451) | 0.435 (0.406–0.458) | 0.721 |
Insulinogenic Index (nmoL/mmoL) | 0.064 (0.032–0.106) | 0.083 (0.054–0.144) | 0.093 |
Disposition Index | 1.303 (0.991–1.574) | 1.533 (1.351–1.971) | 0.017 |
Adaptation Index | 17.81 (15.04–23.13) | 22.78 (18.97–29.22) | 0.017 |
hepatic insulin extraction (%) | 78.09 (68.44–82.26) | 77.60 (67.01–81.67) | 0.646 |
HbA1c (%) | 5.70 (5.05–6.18) | 5.30 (5.08–5.80) | 0.385 |
No significant correlations were found between ΔOGIS, ΔDI or ΔAI and changes (Δ) in parameters of renal or hepatic function, lipids, thyroid hormones or BMI (all p>0.05).
TAC-trough levels correlated negatively with fasting insulin (r = -0.709, p = 0.022,
Scatter plots showing the correlation between tacrolimus trough levels and fasting insulin levels before conversion from tacrolimus to cyclosporine A.
Additionally, no differences were observed in ΔOGIS (p = 0.352), ΔQUICKI (p = 0.762), ΔDI (p = 0.114), ΔAI (p = 0.610) or ΔIGI (p = 0.610) between subjects with PTDM and those with normal glucose tolerance.
Here we demonstrate that the conversion from TAC to CyA results in a profound reduction of both fasting and challenged plasma glucose concentrations without increasing the pancreatic effort in terms of an additionally aggravated hyperinsulinemia. Of note, the observed beneficial effects were mainly attributable to a highly significant amelioration of dynamic (post-challenge) insulin sensitivity and to a lesser extent to an improvement of first-phase insulin secretion, which together resulted in enhanced overall glucose tolerance in HCV-positive RTRs. It is worth noting also that post-challenge insulin sensitivity mimics the process in post-prandial dynamic conditions, the depiction of which is fundamental for characterizing the metabolic state of a subject.
Most previous studies comparing the diabetogenicity of TAC versus CyA were cross-sectional in design and reported on lower PTDM-rates under CyA- as compared to TAC-treatment [
As mentioned above, only few studies assessed dynamic/glucose-stimulated parameters of insulin sensitivity and insulin secretion in RTRs. In contrast to our results, recent data from Bulanowski et al. [
The association between HCV positivity and insulin resistance is well recognized and may involve an HCV-related impairment of insulin signaling, production of proinflammatory cytokines and induction of liver fibrosis resulting in a reduced hepatic and systemic insulin sensitivity and responsiveness [
However, impaired insulin secretion rather than insulin resistance was suggested to represent the predominant mechanism underlying PTDM [
HCV is an independent risk factor for the development of PTDM [
Despite the marked amelioration of glucose tolerance, the conversion from TAC to CyA also elicited adverse side effects including a well-acknowledged hyperlipidemia and a statistically non-significant increase of serum creatinine levels. This, however, is one of the main limitations of our study as the sample size, although adequately powered to observe changes in OGIS, was too small to evaluate safety issues. In clinical practice, numbers of HCV-positive RTRs declined in the last years, affecting only 3.3% (46 of 1394) of all RTRs at our center during the study period. This is in line with other epidemiologic data showing a decreasing number of HCV-positive subjects in European ESRD-population–likely related to improved hygienic precautions during hemodialysis, use of erythropoietin-stimulating agents (rather than RBC transfusions), screening of kidney donors and the development of more efficient anti-HCV therapies including interferon-free regimens of directly acting antivirals [
Nonetheless, our findings entail significant clinical implications for the post-transplant management of HCV-positive subjects. Diabetes remains a major cause for ESRD and HCV infection represents a major issue in many parts of the world with a prevalence of up to 68% in subjects with ESRD [
(PDF)
(DOC)
adaptation index
area-under-the-curve
complement-dependent cytotoxicity—panel reactive antibodies
calcineurin-inhibitor
cyclosporine A
disposition index
estimated glomerular filtration rate
Efficacy Limiting Toxicity Elimination-Symphony
end-stage renal disease
hepatitis C-virus
hepatic insulin extraction
homeostasis model assessment of insulin resistance
insulinogenic index
oral glucose insulin sensitivity-index
oral glucose tolerance test
post-transplant diabetes mellitus
quantitative insulin sensitivity check-index
renal transplant recipients
tacrolimus
total insulin secretion