**2.3 Clinical studies of CKD regression**

The clearest clinical demonstration of glomerulosclerosis regression was provided by Fioretto et al., who showed that pancreas transplantation in patient with type 1 diabetes caused regression of established lesions of glomerulosclerosis in patients with type 1 diabetes (Fioretto et al., 1998).

There are also several studies which examined the effect of RAS inhibition on structural changes in diabetic and non-diabetic CKD. In the study on type 1 diabetic patients with microalbuminuria, treatment with enalapril, perindopril, or metoprolol resulted in a decrease in glomerular basement membrane thickness after 3-4 years of follow-up (Nankervis et al., 1998) (Rudberg et al., 1999). Other studies have suggested that glomerular volumes may be reduced by RAS inhibition, however the contribution of changes in blood pressure is unclear (Perrin et al., 2008). On the other hand, a recent study by Mauer et al. did not detect a statistical difference in mesangial fractional volume in patients treated with placebo, ARB, or ACEI (Mauer et al., 2009). In the ESPRIT study, 3-year treatment with enalapril or nifedipine did not cause a significant change in renal structural abnormalities (2001).

In the case of type 2 diabetes, the study by the Diabiopsies group suggested that treatment with perindopril resulted in stabilization of the percentage of sclerosed glomeruli, but this could not be confirmed by electron microscopy (Cordonnier et al., 1999). In the case of nondiabetic CKD, Ohtake et al. reported that treatment of 15 patients with mild to moderate IgA and non-IgA mesangial proliferative glomerulonephritis with an ARB for an average of 28 months caused a decrease in mesangial matrix expansion and interstitial fibrosis (Ohtake et al., 2008). In summary, although there is encouraging evidence that RAS inhibition can cause regression of glomerular structural changes in humans, the clinical data are not as clear as the data from animal experiments, possibly because the human studies have not focused on the use of high-dose RAS inhibitors.

#### **2.4 The search for clinical biomarkers of disease regression**

One of the reasons that there are relative few large-scale studies on CKD regression is that demonstration of resolution of glomerular lesions requires repeat kidney biopsies, which may not be feasible in large populations. One potential way to overcome this problem is to find surrogate biomarkers of disease regression in the serum and urine of patients with early (stage 1-2) CKD, using the new science of metabolomics (Hayashi et al., 2011). Metabolomics is a discipline dedicated to the global study of metabolites, their dynamics, composition, interactions, and responses to interventions or to changes in their environment (Oresic, 2009), and the recent development of metabolome analysis technology allows the global 'metabolome' to be assessed comprehensively in individual patients. An important advantage of metabolome analysis is the potential to identify new and unidentified metabolites which could have important pathophysiological functions. In a recent study, we obtained serum and urine samples from 15 patients and 7 healthy volunteers, and compared the metabolome profiles of the two groups. Serum or urine samples (100 ul) were added to methanol (900 ul) containing internal standards, deproteinised, and subjected to anionic and cationic capillary electrophoresis time-of-flight mass spectrometry (CE-TOFMS) analysis

The results of our metabolome analysis suggested that serum and urine levels of several amino acid, nucleic acid, and carbohydrate metabolites were altered in patients from an early stage of CKD (Hayashi et al., 2011). We also found evidence for the presence of several novel metabolites which were markedly increased or decreased in the patients with CKD compared to controls. We are performing further studies to examine the structure of these unidentified products, with the final aim to find new biomarkers of disease regression which may be utilized in clinical studies.
