Unique patterns of allelic imbalance distinguish type 1 from type 2 sporadic papillary renal cell carcinoma

Abstract

The molecular genetic correlates of a recently proposed subclassification of papillary renal cell carcinoma (PRCC) that designates tumors as type 1 and type 2 based on histological features have not yet been established. Alterations of known genes in PRCC include missense mutations in the MET oncogene (7q31) and rare translocations fusing TFE3 at Xp11.2 with a variety of other loci. Previous cytogenetic and allelic loss studies of PRCC cases revealed gain of chromosome 3q, 7, 8, 12q, 16, 17, and 20q, and loss of 1p, 6q, 9p, 11p, 13q, 14q, 18, 21q, X, and Y. We analyzed a series of sporadic type 1 and type 2 PRCC cases for MET mutations, TFE3 rearrangements, and allelic imbalance (AI) on 3p, 6, 7q, 9p, 11, 13q, 14q, 17q, 18, 20q, and 21q and compared selected results with a series of conventional renal cell carcinomas. A somatic mutation M1149T was identified in MET exon 17 in 1 of 35 PRCC cases whereas TFE3 rearrangements were not detected in 22 PRCC cases examined. Significant differences in AI frequency between PRCCs and conventional renal cell carcinoma cases were seen on 3p (37.5% versus 77.8%, P = 0.01), 7q (42.9% versus 5.6%, P = 0.01), and 17q (54.5% versus 20.0%, P = 0.03). Significant differences in AI frequency between type 1 and type 2 PRCCs were noted on 17q (78.6% versus 12.5%, P = 0.006) and 9p (0% versus 37.5%, P = 0.02). Additional analyses suggested that the relationship between 17q AI and PRCC type may be independent of histological grade and stage. Our findings identify genetic differences between the recently proposed type 1 and type 2 PRCCs, and support the premise that these subtypes arise from distinct genetic pathways.

Figure 1.

Histopathology of representative PRCC. A: Type 1 PRCC case 153 characterized by papillae covered by small tumor cells with scant pale cytoplasm and small nuclei. B: Type 2 PRCC case 173 characterized by papillae covered by large tumor cells with abundant eosinophilic cytoplasm and large nuclei with prominent nucleoli.

Figure 2.

Results of AI testing in type 1 PRCC case 153. Microsatellite analysis of tumor and corresponding normal tissue demonstrated AI on chromosomes 3p (D3S 2409), 7q (not shown), and 17q (D17S 969) where T = tumor and N = normal tissue. The ratio of the two tumor alleles normalized for the corresponding ratio in normal tissue is shown below the autoradiograms.

Figure 3.

Results of microsatellite analysis in type 1 PRCC. X, Indeterminate; shaded box, AI; open boxes, retention of both alleles.

Figure 4.

Results of microsatellite analysis in type 2 PRCC. X, Indeterminate; shaded box, AI; open boxes, retention of both alleles.

Figure 5.

Association between AI on 9p or 17q and PRCC tumor type adjusting for Fuhrman grade or tumor stage. Cells in which AI was not observed do not contribute information to the test and thus were excluded from analysis. H, Maintenance of heterozygosity; +, Fisher’s exact test restricted to cells outlined in bold; *, Mantel-Haenszel test applied to examine the grade-adjusted association between AI on 17q and PRCC tumor type stratified by levels of Fuhrman grade; and #, Fisher’s exact test restricted to data pooled across cells outlined in bold.