#3880 REDUCED EGFR IS ASSOCIATED WITH IMPROVED SURVIVAL IN PATIENTS WITH RENAL CANCER WHO RECEIVED TARGETED SYSTEMIC ANTI-CANCER THERAPY

Abstract

Abstract Background and Aims Treatment with vascular endothelial growth factor signalling pathway inhibitors (VSPI) and immune checkpoint inhibitors (ICI) has transformed outcomes in advanced renal cancer. A significant proportion of people with renal cancer have co-existing chronic kidney disease (CKD) and concerns persist about the usage of these agents in patients with CKD. We sought to analyse the effect of reduced kidney function on the survival of patients with renal cancer treated with VSPI or ICI. Methods The ChemoCare and NHS West of Scotland SafeHaven databases were linked (data collection spanning 2008–2020), to identify adults from the Greater Glasgow and Clyde Health board who had received either a VSPI or ICI as an anti-cancer therapy. We included participants with two available serum creatinine values (at least 3 months apart) before the date of initiation of treatment. The estimated glomerular filtration rate (eGFR) was calculated using CKD-EPI (2009), the average of the two eGFR results (at least 3 months apart) was included in the analyses. Proteinuria was defined as any positive urine albumin: creatinine (>3 mg/mmol) or protein:creatinine (>15 mg/mmol) ratio before treatment. Factors associated with all-cause mortality were analysed using Cox proportional hazards models with R Software. Results We identified 349 patients with renal cancer who received at least one cycle of ICI and/or VSPI. Sufficient serum creatinine results were available for analysis in 337 of these patients. The average age at first treatment was 63.0 (IQR 55-71) years, 62.5% were male and the median BMI was 28.2kg/m2 (IQR 23.9-31.6). Proteinuria results were recorded in 125 patients and 149 patients had nephrectomy prior to treatment. The majority of patients received VSPI (88.7%). Over a median follow-up of 335 days (IQR 131 days – 840 days), 281 patients died (Table 1). On univariable analysis, lower baseline eGFR (per 10mL/min/1.73 m2 decline in eGFR: HR 0.89, CI 0.84-0.95, p = <0.001) and prior nephrectomy (HR 0.65, CI 0.52-0.83, p= <0.001) were associated with a higher hazards of death. Age (HR 0.99, CI 0.98-1.00, p = 0.245), sex (HR 0.93, CI 0.74-1.18, p = 0.570), BMI (HR 0.99, CI 0.96-1.03, P = .775) and treatment class (HR 0.76, CI 0.51-1.14, p = 0.253) were not associated with higher hazards of death. After adjustment for age, sex and prior nephrectomy, lower eGFR was associated with lower hazards of death (per 10mL/min/1.73 m2 decline in eGFR: HR 0.91, CI 0.84-0.97, p= 0.007). In a sensitivity analysis in people who had complete eGFR and proteinuria data available, the presence of proteinuria was associated with greater hazards of death (HR 1.63, CI 1.05-2.12, p= 0.029) after adjustment for age, sex, eGFR and nephrectomy (Figure 1). Conclusion Lower baseline eGFR before treatment was associated with reduced hazards of death. This finding was not fully explained by the association of prior nephrectomy with better cancer outcomes. This suggests that other factors may contribute to these discrepancies, such as underlying selection bias of patients for treatment, or bias from the marker used to estimate GFR in this group of patients. The presence of proteinuria was associated with an increased hazards of death and maybe a better marker of renal-associated risk in this group than eGFR and warrants further investigation.