Performance of percutaneous core biopsy in predicting renal tumour pathology: a systematic review.

Abstract

To systematically assess the diagnostic accuracy of renal tumour core biopsy for detecting malignancy in adults with localised renal masses. A search of the Cochrane, Embase and Medline databases and the grey literature published up to September 2024 was performed using pre-published methods. Eligible studies were published in English, reported all true- and false-positive and -negative results for localised renal tumour core biopsies in unique adult cohorts, and used surgical histology as the reference standard. Studies were excluded if they exclusively reported patients with malignant biopsy or surgical histology. Primary outcomes were biopsy sensitivity and specificity for detecting malignancy. Twenty-one non-randomised cohorts were identified, totalling 1735 unique patients. In detecting malignancy, renal tumour core biopsy had a weighted mean sensitivity of 98% (95% confidence interval [CI] 0.96-1.00) and specificity of 94% (95% CI 0.84-1.00). Secondary outcomes were determined, including mean false-positive rate (6%), false-negative rate (2%), non-diagnostic rate on first (8%) and second biopsy attempt (9%), and concordance of tumour sub-type (92%) and Fuhrmann grade, both exact grade (60%) and simplified low/high grade (79%). Summary receiver-operating characteristic curve analysis demonstrated high test accuracy (area under the curve 0.97). Risk of bias was moderate to high for most studies. Core biopsy for renal masses in adults has high specificity and sensitivity. Excluding patients for whom biopsy will not change management, near-routine renal tumour biopsy may reduce rates of nephrectomy for ultimately benign tumours, and optimise operating theatre utilisation.