OBJECTIVE: To analyse the prognostic significance of pre- and post-treatment serum prostate-specific antigen (PSA) levels, together with a variety of other factors, in a multivariable analysis of survival in men with stage D2 prostate cancer. PATIENTS AND METHODS: Cox's proportional hazards model was used to compare survival in 134 men with stage D2 metastatic prostate cancer followed prospectively over a 4 year period, using both univariable and multivariable models. The influence of the following factors on survival was analysed: pre- and post-treatment PSA (both absolute and percentage values), age, treatment, testosterone, pre- and post-treatment alkaline phosphatase (absolute and percentage values), acid phosphatase, haemoglobin, symptom score and performance status and extent of disease on bone scan. RESULTS: Pre-treatment PSA levels did not significantly influence survival. Similarly, a low absolute post-treatment PSA level at 3 months after the start of treatment conferred no survival advantage relative to patients with a high PSA level at this time. A post-treatment percentage PSA of < 10% at 2, 3 and 6 months after commencement of treatment was associated with prolonged survival. Low pre-treatment alkaline phosphatase (less than the upper limit of normal) and high pre-treatment testosterone levels (> or = 10 nmol/L) were similarly associated with prolonged survival. CONCLUSIONS: The strong influence of post-treatment percentage PSA on survival in patients with stage D2 prostate cancer suggests that the percentage change in bulk of metastatic deposits is more important in determining survival than the absolute volume of tumour. Pre-treatment alkaline phosphatase seems to be a better indicator of tumour activity than pre-treatment PSA. These findings have important implications for the design and analysis of clinical trials of new therapies in men with stage D2 prostate cancer and for the future selection of alternative treatments for patients with this stage of the disease.
OBJECTIVE: To analyse the prognostic significance of pre- and post-treatment serum prostate-specific antigen (PSA) levels, together with a variety of other factors, in a multivariable analysis of survival in men with stage D2 prostate cancer. PATIENTS AND METHODS: Cox's proportional hazards model was used to compare survival in 134 men with stage D2 metastatic prostate cancer followed prospectively over a 4 year period, using both univariable and multivariable models. The influence of the following factors on survival was analysed: pre- and post-treatment PSA (both absolute and percentage values), age, treatment, testosterone, pre- and post-treatment alkaline phosphatase (absolute and percentage values), acid phosphatase, haemoglobin, symptom score and performance status and extent of disease on bone scan. RESULTS: Pre-treatment PSA levels did not significantly influence survival. Similarly, a low absolute post-treatment PSA level at 3 months after the start of treatment conferred no survival advantage relative to patients with a high PSA level at this time. A post-treatment percentage PSA of < 10% at 2, 3 and 6 months after commencement of treatment was associated with prolonged survival. Low pre-treatment alkaline phosphatase (less than the upper limit of normal) and high pre-treatment testosterone levels (> or = 10 nmol/L) were similarly associated with prolonged survival. CONCLUSIONS: The strong influence of post-treatment percentage PSA on survival in patients with stage D2 prostate cancer suggests that the percentage change in bulk of metastatic deposits is more important in determining survival than the absolute volume of tumour. Pre-treatment alkaline phosphatase seems to be a better indicator of tumour activity than pre-treatment PSA. These findings have important implications for the design and analysis of clinical trials of new therapies in men with stage D2 prostate cancer and for the future selection of alternative treatments for patients with this stage of the disease.
Authors: Evan Y Yu; Roman Gulati; Donatello Telesca; Peter Jiang; Stephen Tam; Kenneth J Russell; Peter S Nelson; Ruth D Etzioni; Celestia S Higano Journal: J Clin Oncol Date: 2010-04-26 Impact factor: 44.544