RETROPERITONEAL LYMPH NODE DISSECTION AFTER INDUCTION CHEMOTHERAPY IN METASTATIC TESTICULAR NON-SEMINOMA

Objective: to evaluate the outcome of retroperitoneal lymph node dissection (RLND) in disseminated testicular non-seminoma patients with residual metastases after induction chemotherapy. Material and methods. The RLND performed in 1983 to 2007 were analyzed in 367 testicular non-seminoma patients with residual retroperitoneal masses after ineffective induction chemotherapy. The median age was 26.06.9 years. Orchidectomy was performed in all patients. Category N1 was regarded in 12 (3.3%) patients, N2 in 79 (21.5%), N3 in 238 (64.9%), Nx in 38 (10.4%). Distant metastases were present in 133 (36.2%) cases. The baseline tumor marker level was elevated in 328 (89.4%) patients (S1 in 169 (46.0%), S2 in 108 (29.4%), S3 in 51 (13.9%), Sx in 39 (10.6%)). According to the IGCCCG prognostic model, 149 (40.6%) patients were classified as good prognostic group, 100 (27.2%) as moderate, 77 (21.0%) as poor ones; the prognostic group was not defined in 41 (11.2%) cases who had started treatment at another facility due to data unavailability. After orchifuniculectomy, all patients received induction cisplatin-based chemotherapy which resulted in tumor shrinkage <50% in 70 (19.1%), 51-90% in 166 (45.2%), and >90% - in 29 (7.9%) cases. The response was not properly assessed in 102 (27.8%) cases. CT scan revealed residual retroperitoneal masses after chemotherapy in all patients (<2 cm - 52 (14.2%), 2-5 cm - 166 (45.2%), >5 cm - 149 (40.6%)). The tumor markers level remained elevated following chemotherapy in 70 (19.1%) cases. All patients underwent RLND (complete in 295 (80.4%) cases). Radical RLND demanded resection of adjacent organs in 22 (5.9%) cases. Extraretroperitoneal metastases were removed simultaneously with retroperitoneal tumor in 22 (5.9%) patients. Postoperative chemotherapy was administered in 100 (27.2%) cases. The median followup was 82.1 (3-188) months. Results. Complications developed in 31 (8.5%) of the 367 of patients. Mortality rate was 0.6% (2/367 cases). Resection of the adjacent organs did not influence mortality rates. Histology revealed necrosis in 149 (40.6%), teratoma in 141 (38.4%), cancer in 77 (21.0%) specimens. The significant predictive factors for necrosis were normal levels of markers following chemotherapy, a residual mass size of < 2 cm, tumor shrinkage >90% (the accuracy of the logistic model for probability of necrosis in the removed specimen was 78%). Discordant pathologic findings between the retroperitoneum and other metastatic sites were in 3 (13.6%) of 22 cases. Ten-year overall, specific and progression- free survival (PFS) was 92.1, 92.4, and 46%, respectively. A poor and moderate prognostic group IGCCCG (p<0.0001), incomplete resection of residual mass (p<0.0001) and presence of cancer in the removed specimens (p<0.0001), initial retroperitoneal masses >5 cm (p=0.042), presence of extraretroperitoneal metastases (p<0.0001), category S>S1 (p<0.0001), positive marker levels after induction (p=0.048) were found to have an adverse impact on PFS. Removal of residual extraretroperitoneal metastases after chemotherapy improved progression-free survival (p=0.022). Postoperative chemotherapy did not influence survival significantly. Multivariate analysis confirmed the predictive value of the radicality of RLND (p=0.036). Conclusion. Radical RLND improves the results of combined treatment in metastatic testicular non-seminoma. It is expedient to make resection of the adjacent organs and extraretroperitoneal metastasectomy in order to achieve a complete removal of residual masses. Whether adjuvant chemotherapy should be used in cases with cancer in residual mass is under discussion.