In this retrospective study the relationship between serum prostate-specific antigen (PSA) and tumour volume (assessed by the planimetric method on whole mount section) was analysed in 56 patients subjected to radical prostatectomy, of whom 28 received 3 months of neoadjuvant GnRH analogue (triptorelin) treatment. Serum PSA in the control group was strongly correlated to the tumour volume while no such correlation was found after hormonal pretreatment (r = 0.84 vs. 0.18), indicating that serum PSA is unreliable as a tumour marker after hormonal deprivation. When the pretreatment PSA (before hormonal deprivation) per tumour volume ratio was calculated, a group of 10 patients (36%) showed considerably higher values, suggesting true tumour volume reduction in those patients as a result of the neoadjuvant hormonal treatment.