Diagnosis of Renal Adverse Drug Reactions

DIAGNOSIS OF RENAL ADVERSE DRUG REACTIONS

None of the described functional or morphologic alterations to the kidney are pathognomonic to ADR. So, general principles of renal diagnostic procedures apply to the evaluation of adverse renal drug reactions.

Although glomerular and tubular processes coop-erate in renal excretory function, renal function is routinely expressed as GFR or creatinine clear-ance. Measurement of creatinine clearance requires a 24-hour urine collection, which is cumbersome and prone to error. Therefore, it is now generally accepted to calculate creatinine clearance using nomo-grams like the Cockroft–Gault formula (Cockroft and Gault, 1976; Gault et al., 1992) or the MDRD formula (Levey et al., 1999, 2000) (Table 38.3). Care must be taken always to compare the result of the creatinine clearance calculation to an age- and gender-matched population (Elseviers et al., 1987).

The determination of renal function by means of the creatinine clearance, however, remains a poorly sensitive method of monitoring the kidney func-tion. Therefore, in experimental settings, a more accurate way of assessing changes in GFR is to measure the clearance of a compound that is freely filtered by the glomerulus but is neither secreted nor absorbed by the tubules. Radiolabeled sodium iodothalamate and ethylenediaminetetraacetic acid (EDTA) are substances commercially available for this purpose.

The most common urinary biomarker used in renal diagnosis is proteinuria. Under normal conditions, the glomerular filtration barrier restricts the transfer of high molecular weight proteins from the plasma to the lumen of the tubule. High molecular weight proteins appearing in the urine points to a pathological condi-tion of the glomerulus, changing the permselectiv-ity of the filter. Under normal conditions, a minute amount of low molecular weight proteins are filtered, which then undergo endocytic reabsorption by prox-imal tubular cells. When the reabsorptive capacity of the proximal tubule is compromised, low molecu-lar weight proteins appear in the urine in measurable amounts. Determination of the quantity and the quality of urinary proteins allows for the distinction between ‘glomerular’ and ‘tubular’ proteinuria.

Enzymuria has been extensively used by toxi-cologists to detect early renal damage. Urinary enzymes bear the potential of determining the site of damage because different enzymes are localised in specific segments of the nephron. For exam-ple, alanine aminopeptidase, alkaline phosphatase and glutamyltransferase are enzymes bound to the brush border of proximal tubular cells. Their appearance in the urine should be indicative for turnover of brush border. The general acceptance of urinary enzyme excretion as a measure of tubular dysfunction in human safety studies has been limited for several reasons. First, it has been impossible to link the pres-ence of the different enzymes appearing in the urine to specific tubular disease states. Secondly, a relation-ship between the magnitude of the enzymuria and the severity of tubular injury has not been established. Furthermore, enzymuria may occur in normal situa-tions due to increased brush border turnover, altered membrane permeability or increased synthesis.

In general a renal biopsy is not needed to estab-lish the diagnosis of a renal adverse event. When a glomerulopathy is suspected, only a biopsy allows to distinguish between the different histopathologic types. Ideally, the diagnosis of acute interstitial nephritis is also confirmed by histopathologic examination.

Presently, during drug development, preclinical toxicity tests involve the use of animal models. However, advances in cell and tissue culture will permit the development of in vitro toxicity assays. The aim of the development of in vitro tests is not only to replace in vivo animal testing but also to study the mechanisms of cell modulation by toxic compounds. Recently, for example, in vitro studies involving renal cells in culture suggested that the underlying mechanism of the proteinuria associated with the use of rosuvastatin was inhibition by the statin of the endocytotic uptake of proteins by the proximal tubular cell (Verhulst, D’Haese and De Broe, 2004). Several permanent and immortalised cell lines of human and non-human origin are available, offer-ing several advantages over primary cultures such as an unlimited life span and the lack of time-consuming isolation procedures. The most widely used renal epithelial cell lines of animal origin are the LLC-PK1 (Hampshire pig) and OK (American opossum) cell lines, exhibiting characteristics suggestive of proxi-mal tubular origin, and the MDCK (Cocker Spaniel) cell line, exhibiting characteristics suggestive of distal origin.