The terminal stage of prostate cancer is metastatic castration-resistant prostate cancer (mCRPC) with poor prognosis and high mortality. Abiraterone acetate (AA), as a new generation drug for endocrine therapy, has been demonstrated to prolonged overall survival signicantly in mCRPC patients. In addition, a corticosteroid drug must be administered during treatment to avoid side effects of abiraterone. It has been found that a corticosteroid switch from prednisone to dexamethasone could delay the development of resistance, significantly prolong the progression-free survival rate of patients, with well tolerance. But the specific mechanism and long-term clinical benefit still need further study. This emerging treatment paradigm provides new ideas for treatment options for patients with mCRPC, however, caution is still needed in clinical practice, and it is recommended to determine the treatment plan after considering all aspects of the patients.

Objective:To evaluate the impact of metastatic site on the prognosis in patients with metastatic renal cell carcinoma (mRCC), and it′s value for modifying the International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) criteria.Methods:The data of 218 patients pathologically diagnosed with mRCC were analyzed retrospectively in West China Hospital from Jan. 2009 to Dec. 2019. Among all patients, 71.6%(156/218) were male, and 89.0% (194/218) underwent nephrectomy. Most of the patients were pathologically diagnosed with renal clear cell carcinoma (176 patients, 80.7%). Lung (137/218, 62.8%) was the most observed metastatic site, following by bone (47/218, 26.1%), lymph node (37/218, 17.0%) and liver (23/218, 10.6%). All patients were classified into favorable (26 patients, 11.9%), intermediate (126 patients, 57.8%) or poor (37 patients, 17.0%) risk group according to IMDC criteria. Endpoints of this study were progression-free survival (PFS), overall survival (OS) and tumor response. The impact of metastatic sites on patients’ prognosis was analyzed, and those that had significant relationship with prognosis were then added into IMDC criteria and a modifying IMDC model was established. Predictive value of this model was further evaluated by calculating concordance index (C-index).Results:In the whole cohort, median PFS and OS were 13.0 and 33.0 months. Survival analysis suggested that patients with bone ( P=0.004), brain ( P=0.042) and liver ( P=0.046) had significantly shorter OS. Thus, patients were divided into two groups: patients with bone/brain/liver metastasis (82 patients, 37.6%) and patients with other metastatic sites (136 patients, 62.4%). Compared with patients with other metastatic sites, those who with bone/brain/liver metastasis had inferior tumor response by TKIs treatment (disease control rate: 51.2% vs. 73.5%, P=0.004). Multivariate analysis suggested that bone/brain/liver metastasis had negative impact on OS (25.0 vs. 47.0 mo, P=0.039). Furthermore, bone/brain/liver metastasis also showed significant relationship with shorter OS in IMDC low (30.0 vs. 62.0 months, P=0.036), intermediate (31.0 vs. 48.0 months, P=0.048) or high (7.0 vs. 18.0 months, P=0.037) risk group, indicating that metastatic site had predictive value for prognosis of mRCC patients. Based on that, bone/brain/liver metastasis were added into the IMDC criteria, and weighting each parameter was weighted according to its coefficient to patients’ OS. Finally, a modified IMDC scoring system were established. C-index of this modified model was 0.669 (0.599 for current IMDC criteria). Conclusions:Bone/brain/liver metastasis in mRCC patients indicated a shorter OS duration. When adding bone/brain/liver metastasis as a predictive parameter for prognosis of mRCC patients into IMDC criteria, the modified IMDC criteria could offer more accurate prediction for patients’ survival.