Pharmacokinetics of connective tissue growth factor : implications for its use as biomarker

Fibrosis represents the common final pathway of many chronic diseases including inflammatory, ischemic and metabolic disorders. There is great need for non-invasive biomarkers that reflect activity of the fibrogenic process and can help predict progression. Connective tissue growth factor (CTGF) seems an ideal biomarker for monitoring ongoing fibrosis, since it is an important fibrogenic cytokine that is readily quantifiable in body fluids. Plasma and urine CTGF levels are increased in various chronic diseases. However, there is lack of knowledge of the processes contributing to elevated CTGF levels, which hampers correct interpretation. Thus far, CTGF levels were expected to reflect CTGF tissue expression. Aim of this thesis was to determine whether elimination and distribution processes may also influence CTGF levels. To investigate this, the pharmacokinetic profiles of CTGF and its N-terminal fragment (N-CTGF), the predominant form of CTGF in plasma and urine, were studied. Rodent and human studies showed that plasma N-CTGF is highly dependent on glomerular filtration rate (GFR). In healthy rodents, intravenously administered N-CTGF was primarily eliminated by the kidney. Plasma N-CTGF accumulated in chronic kidney disease (CKD) patients with the highest levels in end stage kidney disease. Successful kidney transplantation resulted in a decrease of plasma CTGF proportional to the increase of GFR. In contrast, full length CTGF was primarily eliminated by the liver in healthy rodents, presumably via a low density lipoprotein receptor-related protein1-mediated pathway. Full length CTGF had a large distribution volume due to a protamine-sensitive process, presumably extensive binding to heparan sulphate proteoglycans. Rodent and human studies demonstrated that urinary CTGF is highly dependent on tubular function. In healthy subjects hardly any CTGF was excreted into the urine, due to almost complete proximal tubular reabsorption of filtered CTGF via megalin-mediated endocytosis. Impairment of tubular protein reabsorption, either by Gelofusine or in megalin-deficient mice, resulted in substantially increased urinary CTGF excretion. Urinary CTGF was tightly associated with β2-microglobulin in CKD patients, indicating that elevated urinary CTGF may reflect proximal tubular dysfunction. In diabetes mellitus, a disorder characterized by high renal CTGF expression, urinary CTGF was determined by both decreased tubular reabsorption of filtered CTGF and increased CTGF expression in glomeruli and medullary tubules. Plasma CTGF was further explored in two large patient populations. In hemodialysis patients, plasma CTGF was associated with residual kidney function and presence of cardiovascular disease. Elevated plasma CTGF increased the risk of all-cause mortality. Plasma CTGF was not removed by low flux hemodialysis and its concentration increased over time while hemodiafiltration substantially lowered plasma CTGF resulting in a sustained decrease. In patients with manifest vascular disease, plasma CTGF was related to GFR and total cholesterol and showed a robust association with the risk of new cardiovascular events and mortality. In conclusion, this thesis shows that impaired elimination and altered distribution may influence CTGF levels in chronic disorders, in addition to increased CTGF production. This knowledge is essential for correct interpretation of CTGF levels. Furthermore, CTGF may be regarded as risk marker for mortality in hemodialysis patients and patients with vascular disease