PURPOSE - Current European Association of Urology guidelines state that a PSA≥100 ng/ml at diagnosis indicates metastatic disease. The aim of the present study was to examine the association between PSA values ≥100 ng/ml at diagnosis, distant metastasis and prostate cancer-specific survival.
MATERIALS AND METHODS - 15,635 men with prostate cancer diagnosed between 1998 and 2009 identified in the Prostate Cancer Data Base Sweden (PCBaSe) were included in a population-based register study. Prostate cancer-specific survival was compared between three groups: 1,879 men with PSA ≥100 ng/ml and negative imaging investigations (M0); 5,642 men with distant metastases on imaging (M1) and PSA ≥ 100 ng/ml, and 3,828 men with M1 and PSA < 100 ng/ml. A fourth group consisted of 4,286 men with PSA ≥ 100 ng/ml who had not undergone imaging (Mx), but were not included in the assessment of survival.
RESULTS - Of 7,521 men with PSA≥100 ng/ml undergoing imaging for staging, 26% were classified as M1. Only 59% of 3,527 men with PSA 100-300mg/ml had distant metastases on imaging. The 5-year prostate cancer-specific survival was 72% (95% confidence interval [CI] 70-74%) in men with PSA≥100 ng/ml and M0, 24% (95% CI 23-25%) in men with PSA≥100 ng/ml and M1, and 39% (95% CI 37-40%) in men with PSA
CONCLUSIONS - One fourth of men with PSA values ≥100 ng/ml did not have distant metastases and had a 2-3 fold higher 5-year survival than men with distant metastases on imaging. Our findings strongly suggest that the use of PSA≥100 ng/ml as indicator of metastatic disease should be reconsidered.
J Urol. 2015 Jul 17. pii: S0022-5347(15)04415-8. doi: 10.1016/j.juro.2015.07.082. [Epub ahead of print]
Stattin K1, Sandin F2, Bratt O3, Lambe M4.
1 Department of Surgical Sciences, Uppsala University, Sweden.
2 Regional Cancer Center, Uppsala University Hospital, Uppsala, Sweden.
3 Urological Cancer Research Unit, Department of Translational Medicine, Lund University , Sweden; Department of Urology, Cambridge Hospitals, United Kingdom.
4 Regional Cancer Center, Uppsala University Hospital, Uppsala, Sweden; Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.