PURPOSE: Mismatch repair (MMR) deficiency and microsatellite instability (MSI) are associated with cisplatin resistance in human germ cell tumors (GCTs). BRAF mutation (V600E) is found in MSI colorectal cancers. The role of RAS/RAF pathway mutations in GCT treatment response is unknown. PATIENTS AND METHODS: Two patient cohorts were investigated: 100 control GCTs (50 seminomas and 50 nonseminomas) and 35 cisplatin-based chemotherapy-resistant GCTs. MMR proteins were analyzed by immunohistochemistry, and eight microsatellite loci were examined for MSI. Tumors were assessed for specific BRAF and KRAS mutations. RESULTS: Resistant tumors showed a higher incidence of MSI than controls: 26% versus 0% in two or more loci (P <.0001). All resistant tumors were wild-type KRAS, and two controls (2%) contained a KRAS mutation. There was a significantly higher incidence of BRAF V600E mutation in resistant tumors compared with controls: 26% versus 1% (P <.0001). BRAF mutations were highly correlated with MSI (P = .006), and MSI and mutated BRAF were correlated with weak or absent staining for hMLH1 (P = .017 and P = .008). Low or absent staining of hMLH1 was correlated with promoter hypermethylation (P <.001). Tumors lacking expression of hMLH1 or MSH6 were significantly more frequent in resistant GCTs than in controls (P = .001 and 0.0036, respectively). Within the subgroup of resistant tumors, patients with MSI showed a trend to longer progression-free survival (P = .068). CONCLUSION: We report for the first time a correlation between a gene mutation--BRAF V600E--and cisplatin resistance in nonseminomatous GCTs. Furthermore, a correlation between MMR deficiency, MSI, and treatment failure is confirmed.
PURPOSE: Mismatch repair (MMR) deficiency and microsatellite instability (MSI) are associated with cisplatin resistance in human germ cell tumors (GCTs). BRAF mutation (V600E) is found in MSI colorectal cancers. The role of RAS/RAF pathway mutations in GCT treatment response is unknown. PATIENTS AND METHODS: Two patient cohorts were investigated: 100 control GCTs (50 seminomas and 50 nonseminomas) and 35 cisplatin-based chemotherapy-resistant GCTs. MMR proteins were analyzed by immunohistochemistry, and eight microsatellite loci were examined for MSI. Tumors were assessed for specific BRAF and KRAS mutations. RESULTS: Resistant tumors showed a higher incidence of MSI than controls: 26% versus 0% in two or more loci (P <.0001). All resistant tumors were wild-type KRAS, and two controls (2%) contained a KRAS mutation. There was a significantly higher incidence of BRAF V600E mutation in resistant tumors compared with controls: 26% versus 1% (P <.0001). BRAF mutations were highly correlated with MSI (P = .006), and MSI and mutated BRAF were correlated with weak or absent staining for hMLH1 (P = .017 and P = .008). Low or absent staining of hMLH1 was correlated with promoter hypermethylation (P <.001). Tumors lacking expression of hMLH1 or MSH6 were significantly more frequent in resistant GCTs than in controls (P = .001 and 0.0036, respectively). Within the subgroup of resistant tumors, patients with MSI showed a trend to longer progression-free survival (P = .068). CONCLUSION: We report for the first time a correlation between a gene mutation--BRAF V600E--and cisplatin resistance in nonseminomatous GCTs. Furthermore, a correlation between MMR deficiency, MSI, and treatment failure is confirmed.