Introduction During recent years, several new life-prolonging therapeutic options have been introduced for patients with metastatic prostate cancer (mPCa). The aim of the present study was to evaluate the changes in the survival of patients diagnosed with mPCa prior to and in the early period of the implementation of these new agents. Patients and methods The study population consisted of 207 men diagnosed in 1997 and 316 men diagnosed in the period 2007–2013 with de novo mPCa and managed with initial endocrine therapy. Men were followed for overall survival and PCa-specific survival. Results At the time of diagnosis, men diagnosed in the period 2007–2013 had less co-morbidity, lower prostrate-specific antigen levels and lower clinical tumour categories than men diagnosed in 1997. A significantly higher proportion of men diagnosed in 1997 were managed with surgical castration (57% versus 9%). Only one patient diagnosed in 1997 received second-line therapy compared with 81 men (26%) diagnosed in the period 2007–2013. The median overall survival was significantly longer for men diagnosed between 2007 and 2013 compared with men diagnosed in 1997 (39.4 months versus 24.2 months, p < 0.0001). Likewise, the cumulative incidence of PCa-specific death was higher among men diagnosed in 1997 compared with men diagnosed between 2007 and 2013, with 5-year cumulative incidences of 72% and 47%, respectively (p < 0.0001). Conclusion Survival in men diagnosed with metastatic PCa has improved significantly over time. The improved survival can in part be explained by lead-time bias, but also by the introduction of new life-prolonging treatments.