ProstateCalculator - Overview of PSA Recurrance Calculator

This calculator has been developed to predict whether a man undergoing radical prostatectomy as primary treatment for clinically localized prostate cancer will have a PSA recurrence of his prostate cancer. Dr. Ash Tewari of Cornell Department of Urology. Dr. Tewari's research on prostate cancer also includes robotic surgery to remove the prostate with faster healing and fewer complications.

The rationale:

Approximately 40,000 men undergo radical prostatectomy for the treatment of clinically localized prostate cancer in USA every year. Thirty to forty percent develop recurrence of serum prostate specific antigen (PSA) in 5-10 years. Physicians and patients need guidance in estimating the risk for recurrence in an individual patient. We have developed a model for calculating such risk, which accounts for various tumor characteristics.

The objective of this study was to estimate risk for PSA recurrence at 7 years following radical prostatectomy based on preoperative PSA level, clinical stage, biopsy Gleason score, and pathological stage in men with localized prostate cancer.

Where the data comes from:

This calculator was developed as part of a study conducted by Dr. Ashutosh Tewari and his colleagues at the Josephine Ford Cancer Center. The study design was a multi-institutional retrospective cohort study. Data from 2,065 men with localized prostate cancer treated with radical prostatectomy at the participating medical institutions between 1992-1999 were analyzed. These patients did not receive adjunct hormonal therapy or radiation treatment prior to developing PSA recurrence. Proportional hazards regression analysis was used to model recurrence as a function of age, race, baseline PSA, pathologic tumor stage, seminal vesicle involvement, margin status, Gleason score and perineural invasion in the final pathology.

The mean patient age was 68.4 years and mean PSA before surgery was 11.6 ng/mL. Mean length of follow-up was 41.5 months and 67% patients had Gleason scores 5-7. Eighty percent had clinical stage T2 and 5% T3. In our series 64% of patients had organ-confined cancer, 33% positive margin and 14% had seminal vesicle invasion. Lymph node positive patients were not included in this series. PSA progression was noted in 30.6%.

Methods:

The data set was divided into a training dataset with 1195 patients and a test dataset with 624 patients. Forward variable selection was used with the training dataset to identify a multivariable model for recurrence-free survival. A p-value of 0.05 or less was required for a variable to be added to the model. Once no further variables met the entry criterion, first order interactions among the final set of model variables were computed and tested for forward stepwise inclusion.

By entry of few routinely used parameters, the model correctly identified 76% of the patients who developed a recurrence in the validation set.

Acknowledgements:

In addition to the lead investigator, Ashutosh Tewari, MD, the following individuals contributed to the development of the PSA Recurrence Model (version 2.0): Christine Cole Johnson, PhD; Mutta Issa, MD; Rizk El-Galley, MD; Hans Stricker, MD; James Peabody, MD; Julio Pow-Sang, MD; Aseem Shukla, MD; Zev Wajsman, MD; Mark Rubin, MD, John Wei, MD, James Montie, MD, Chris Porter, MD, Michael Harris, MD, Raymond Demers, MD, MPH, Dawn E. Thomas, MPH, George W. Divine, PhD; E. David Crawford, MD; Eduard J. Gamito, and Mani Menon, MD.

Special thanks to Dr. Steven Tucker for his valuable input in the review and testing of this calculator.