Discrepancy in calculated and measured glomerular filtration rates in patients treated with PARP inhibitors

To describe discrepancies in calculated and measured glomerular filtration rate in patients using PARP (poly ADP ribose polymerase) inhibitors who had an elevation in serum creatinine levels.

Retrospective cohort, single center study. Patients included were those with ovarian or endometrial cancer taking olaparib, rucaparib or niraparib, and in in whom an increased serum creatinine was identified. The study cohort included those who also underwent technetium-99m radioisotope renography (glomerular filtration rate (GFR) scan). The main objective is to describe the discrepancies in calculated glomerular filtration rate using the Cockcroft-Gault method and measured glomerular filtration rate using a GFR scan.

211 patients were included in the study; 64 (30%) had on-treatment elevated serum creatinine, and 23 (36%) underwent a GFR scan. 32 GFR scans were performed (six patients had more than one scan). Using a clinical cut-off ≥50 mL/min as normal renal function, both calculated and estimated glomerular filtration rates were below normal in 6 of 32 GFR scans. In those patients undergoing a GFR scan, serum creatinine had risen a median 49% (IQR 20-66%, range 0-144%) above baseline. Discordance between a calculated low glomerular filtration rate and an estimated normal glomerular filtration rate occurred in 63% (range of glomerular filtration rate discrepancy: -46% to +237%). Despite increases in serum creatinine on therapy and a subsequent significant decline in the per patient calculated creatinine clearance (mean 65.6 mL/min vs 43.4 mL/min; p<0.0001), the estimated glomerular filtration rate from the renal scan was nearly identical to the patient's baseline (65.6 mL/min vs 66.1 mL/min; p=0.89).

Serum creatinine elevation in patients taking PARP inhibitors may not be associated with a true decrease in glomerular filtration rate. A high index of suspicion should be maintained for alternative causes of elevated serum creatinine in patients treated with PARP inhibitors who lack other sources of renal injury.