Background: With the induction of targeted agents and immune checkpoint inhibitors, treatment strategies for systemic treatment of metastatic renal cell carcinoma (mRCC) have been changed with improved survival. However, those agents were approved based on phase 3 study results designed for clear cell mRCC. For patients with non-clear cell mRCC, those drugs designed for clear cell mRCC have also been used because no validated systemic therapy exists. This retrospective study aimed to identify patients with non-clear cell mRCC who benefit from systemic therapy. Methods: A total of 255 patients with mRCC were reviewed (Institutional review board approval No 2013-0425). Among them, 41 patients who pathologically diagnosed as non-clear cell mRCC and received systemic therapy were retrospectively analyzed. Clinical and pathological data were retrieved and analyzed retrospectively. The prognostic effect of each variable on progression free survival (PFS) and overall survival (OS) were investigated. OS for patients with non-clear cell mRCC was further investigated with univariate and multivariate Cox’s proportional hazards regression models. Results: After a median follow-up of 33.9 months after first line treatment initiation, the median PFS was 3.3 and 19.7 months for non-clear cell and clear cell RCC, respectively (p<0.0001). The median OS was 20.9 and 50.1 months for non-clear cell and clear cell RCC, respectively (p<0.0001). For patients with non-clear cell RCC, univariate analyses revealed duration of first line treatment, baseline C-reactive protein, and International Metastatic RCC Database Consortium (IMDC) risk group (favorable vs intermediate/poor) were significantly correlated with OS, respectively. Multivariate analysis revealed that the duration of first line treatment (HR: 0.997, p=0.027) and the baseline C-reactive protein (HR:1.128, p=0.004) were independent predictors for longer OS in non-clear cell mRCC patients. Conclusions: Systemic therapy for non-clear cell mRCC patients resulted in relatively shorter PFS and OS. A shorter duration of first line treatment and elevated baseline C-reactive protein are correlated with longer OS.