Negative lymph nodes status has been attached more attention as a prognostic indicator for nonmetastatic penile cancer. We aimed to identify the appropriate number of negative lymph nodes dissection for nonmetastatic penile cancer using the Surveillance, Epidemiology and End Results database. A total of 1,470 nonmetastatic patients with penile squamous cell carcinoma were identified during 2004 and 2013. All patients were categorised according to different risk levels and lymphadenectomy. Univariate and multivariate Cox regression analyses were performed to evaluate the relationship between prognostic risk factors and cancer-specific survival. The optimal cut-off value of negative lymph nodes dissection was determined using the X-Tile program. A total of 1,470 patients were categorised into low- (pT1G1), intermediate- (pT1G2) or high-risk (pT1G3 and all higher stages) groups. In multivariate Cox analysis, lymphadenectomy improved the cancer-specific survival for patients in high-risk group (p = 0.014). Further, the optimal cut-off value of negative lymph nodes dissection for high-risk patients was 5 and patients with >5 negative lymph nodes had a higher cancer-specific survival (chi(2) = 9.3676, p < 0.05). Therefore, lymphadenectomy improved survival for high-risk penile cancer and the removal of more than five negative lymph nodes was correlated with higher cancer-specific survival for high-risk patients who underwent lymphadenectomy.