The accuracy of renal tumor biopsy: analysis from a national prospective study.

ABSTRACT

OBJECTIVE: To assess preoperative renal tumor biopsy (RTB) accuracy. MATERIALS AND METHODS: As part of the prospective NEPHRON study, data from 1,237 renal tumors were collected, including the use and results of RTB and final histology following nephrectomy. During the 6 months period of inclusion, 130 preoperative biopsies were performed. We used the kappa coefficient of the McNemar test to determine the concordance between the biopsy and the nephrectomy specimen (NS) regarding four parameters malignant/benign status, histological subtype, Fuhrman grade and microscopic necrosis. RESULTS: Preoperative biopsies were performed in 9.7 and 11.4 % of the 667 radical and 570 partial nephrectomies, respectively. Tumor biopsy was inconclusive in 7.7 % of the cases. In 117 cases, a comparison between RTB and NS was available. Benign tumors accounted for three (2.6 %) and five (4.3 %) of the RTB and NS, respectively (κ = 0.769, good). With seven (6 %) discordant results in terms of histological subtype characterization between RTB and final pathology, RTB accuracy was considered excellent (κ = 0.882). In 33 cases (31.7 %), Fuhrman grade was underestimated at biopsy resulting in an intermediate concordance level (κ = 0.498). Tumor microscopic necrosis was identified in 12 RTB (10.4 %) versus 33 NS (28.4 %) (κ = 0.357, poor). CONCLUSIONS: RTB provides good to excellent diagnostic performance for discriminating malignancy and tumor histological subtype. However, its performance is intermediate or even poor when considering prognostic criteria like Fuhrman grade or microscopic necrosis. Thus, this possible inaccuracy should be taken into consideration when using RTB for accurate guidance of treatment strategy.