Pathologic concordance rate and outcomes by histologic subtype in advanced papillary renal cell (pRCC) carcinoma: An analysis from the SWOG S1500 (PAPMET) trial.

Abstract

4562 Background: The PAPMET trial demonstrated significantly improved progression-free survival (PFS) with cabozantinib vs. sunitinib in advanced pRCC (median: 9.0 vs. 5.6 mos; hazard ratio (HR): 0.60; 95% CI 0.37–0.97). Until recently, pRCC was subclassified as type 1 vs. 2 based on histologic features. During conception of the PAPMET study, we hypothesized that patients with type 1 pRCC will derive more benefit from cabozantinib compared to those with type 2 disease, given enrichment of MET alterations in the former. In this secondary analysis, we report the outcomes stratified by histologic subtype per central review. Methods: Patients with advanced pRCC who had received up to 1 line of therapy were randomized to receive either sunitinib, or cabozantinib, crizotinib or savolitinib stratified by pRCC subtype (type 1, type 2 or not otherwise specified [NOS]). Pathologic data from local pathology review was catalogued. Central pathology review was conducted by 3 expert genitourinary pathologists who independently reviewed tumor specimens. Discordant opinions among these pathologists were individually arbitrated and a unified diagnosis was assigned. Central pathology review results were considered the true positive and the sensitivity, and the positive predictive value (PPV) of local review was estimated for histologic subtypes. Clinical outcomes (objective response rate (ORR) and PFS) were compared between histologic subtypes (by central review) in the cabozantinib and sunitinib arms. Results: Amongst the 147 patients with available local and central pathology review, 17.7% (n=26), 53% (n=78) and 29.3% (n=43) were characterized as having type 1, 2 or NOS/mixed pRCC by local review compared to 29.3% (n=43), 45.6% (n=67) and 25.2% (n=37) respectively by central review. Individual cases were frequently reclassified and local pathology review demonstrated low sensitivity (type 1: 49%, type 2: 67% and NOS: 43%) and PPV (type 1: 80%, type 2: 57% and NOS/mixed: 37%) across subtypes. During central review, complete agreement amongst the 3 pathologists was seen in only 33.3% (n=49) of cases and was highest in type 2 tumors (40%; n=27). Compared to sunitinib, cabozantinib demonstrated numerically higher ORR and PFS in both type 1 and type 2 pRCC subgroups. Conclusions: Designation of pRCC subtype by central review did not identify a subset of patients with greater clinical benefit from cabozantinib. Further, classification of subtype was challenged by discordance in both local and central review. Supporting the removal of type 1 and 2 designations from the 2022 WHO guidelines, these findings highlight the limited clinical value and significant challenges associated with subtyping of pRCC. Clinical trial information: NCT02761057 . [Table: see text]