General outline:

On 01.12.05 we started the EU-funded project PREDICTIONS (PREvention of DIabetic ComplicaTIONS). The full title of the project is “The Identification of Risk Factors for the Development of Diabetic Nephropathy: The PREDICTIONS Project” (FP6 Proposal/Contract no.: 018733).

The main objective of the project is to identify biomarkers associated with the risk to develop diabetic nephropathy (DN). DN is one of the most severe and life-threatening complications of diabetes mellitus. About 30% of patients with type 2 diabetes will eventually develop DN. Duration of diabetes as well as glycaemic and blood pressure control do not sufficiently explain the risk of developing DN. About 31 genes have been reported to possibly contribute to DN susceptibility. A major susceptibility gene was mapped to 18q and a sequence variant in the CNDP1 gene was identified to cause a 3-fold elevated risk of DN. On the proteomic level, protein modifications by advanced glycation end-products (AGEs) are known to be sufficient to cause DN.

Different variables promise to have the potential of serving as biomarkers with high predictive value for DN risk assessment. Although there is a focus on DN, the effects of DN-related markers on retinopathy will also be studied. The PREDICTIONS project focuses on prospective research and case-control studies to:

The latter will be realized both on the genomic (expression profiling in kidney biopsies) and proteomic level (mass-spectometry). In 3 multivariate analyses these markers will be correlated with the success of DN-treatment with ACE inhibitors, irbesartan, and benfotiamine. A risk model will be developed for use by clinicians, who will evaluate the model prospectively in the future. We hope, our work will have significant impact on patient care. It aims to provide scientific evidence for novel individualized therapeutic approaches.

An independent confirmation of the identification of CNDP1 as risk factor.

Genetic association studies on CNDP1 will be performed and carnosine levels and serum carnosinase activity will be determined in blood samples of all study patients. This activity will open the way to (a) early identification of patients at risk, (b) evaluation of the efficiency of current therapies for DN against the background of different CNDP1 variants and (c) initiate the development of novel therapies aimed at increasing carnosine levels. By the end of the first year it will be known whether confirmation of the predicted role of CNDP1 in DN has been obtained or not.

Re-evaluation of previously proposed genetic variants.

Because CNDP1 is not expected to be the only DN-causing factor, some additional factors may play an important role in ‘CNDP1-independent’ DN. Of special interest is the SLC12A3 gene.  SLC12A3 and CNDP1 are the only candidate genes resulting from systematic genome studies known so far. A total of 31 markers will be examined in at least 200 patients. Potentially associated markers will be examined in over 1000 patients. 

Expression profiling for the identification of genes involved in the pathogenesis of DN.

With the emergence of new methods for quantification of thousands of mRNA levels at the same time, it is now possible to create expression profiles for different kidney diseases. We have recently established methodology to identify differently expressed genes in kidneys with DN. This technique will be employed to find new genes and thus new genetic markers involved in the pathogenesis of DN. It is foreseen to study kidney material from at least 20 independent patients with DN.

Analysis of the urinary proteome for the identification of a secretion pattern associated both with a specific genetic trait and a high-risk vascular profile of diabetic patients.

Previously used techniques for the determination of the associated secretion pattern (two-dimensional polyacrylamide gel electrophoresis and mass spectrometry) were too cumbersome for routine use. With the availability of capillary electrophoresis and mass spectrometry, fast and accurate analysis of hundreds of polypeptides in small volumes has become feasible and will be employed in urine and serum samples of the study patients (WP 7). It is planned to test 300-360 patients, some with DN, some with microalbuminuria and some with diabetes without kidney disease or with non-diabetic kidney disease.

Biomarkers of protein glycation and oxidative stress as prognostic factors.

Various AGEs and molecules involved in AGE formation and oxidative stress play an important role in the pathogenesis of DN. To this end, AGEs and several markers of oxidative stress will be assessed in the study patients. It will be evaluated, which of the selected biomarkers may serve as prognostic factor.

Exploration of the association between biomarkers and response to treatment.

Until now, several therapeutic modalities are being applied in patients with DN. It is known that diabetes complications may “escape” after these therapies have been started. The Consortium has access to materials and data of therapeutic trials on ACE-inhibitors and ATR-1 blockers (Irbesartan) conducted previously and thus will be able to investigate and identify “escapers”. A Benfotiamine trial with the same objective will complement this study. 

Development of a risk model.

Of the variety of possibly relevant biomarkers, those with a significant predictive value will be identified and used to develop a risk model to be used in clinical practice.

Ethical issues

We will strictly comply with widely recognized international guidelines for conducting studies in humans. The projects will respect all ethical requirements in objectives and methodology, which will be approved by the institutional Ethics Committees at the respective partner universities. Collection of patient tissue or blood samples will demand written consent. All specimens will be coded to ensure confidentiality and to protect the identity of patients. In the case-control study and clinical trials, patients will be recruited in a consecutive manner without regard to race, sex, social status, religion or other considerations. Confidentiality laws will be strictly observed when processing human clinical information. To this end, codes for patient identification will be employed, thereby guaranteeing confidentiality of patient data. A professional secured database will be used. Names, addresses and other identifying information will remain with the recruiting institute, together with the ID-codes. The ID-codes will identify the patients in the database. No patient-identifying information will be made available to the general public or be contained in publications.

Figure: Schematic representation of the flow of person-sensitive data. The local clinicians are responsible for the codification of patients and all materials. Each code consists of letters and numbers. The first part of the code identifies the recruiting centre, the second part identifies the patient. The system safeguards, that sensitive person-identifying data will not be accessible to other members of the consortium or to the outside world.