Prediction Tools

To provide more accurate outcome predictions for patients and their physicians, prognostic nomograms -- predictive devices -- gauge probabilities based on patient and treatment characteristics. For the prostate cancer patient, we have generated pre- and postoperative nomograms to be used before and after radical prostatectomy in order to predict the probability of recurrence of prostate cancer at 5 and 7 years, respectively. The goal of using the nomograms is to optimize predictive accuracy, critical in a setting in which treatment selection may depend on efficacy.

Few published studies have combined clinical prognostic factors into risk profiles that can be used to predict the likelihood of recurrence or metastatic progression after treatment of prostate cancer. We developed a nomogram that allows prediction of disease recurrence through use of preoperative clinical factors, such as PSA and Gleason grade, for patients with clinically localized prostate cancer who are candidates for treatment with a radical prostatectomy, i.e., surgical removal of the prostate.

Using statistical methods, we modeled the clinical data and disease follow-up for 983 men with clinically localized prostate cancer, whom we intended to treat with radical prostatectomy.

Clinical data included:

• pretreatment serum prostate specific antigen (PSA) levels,
• biopsy Gleason sum, and
• clinical stage.

Treatment failure was recorded when there was a rising serum PSA level after surgery (2 measurements of 0.4 ng/mL of greater and rising) or initiation of adjuvant therapy. Validation was performed on this set of men, as well as a separate sample of 168 men, also from our institution.

Treatment failure was noted in 196 of the 983 men; and the group without failure had a median follow-up of 30 months (range 1 to 146 months). The 5-year probability of freedom from treatment failure for the entire group was 73 percent (95 percent confidence interval [CI] = 69 percent to 76 percent).

The predictions from the nomogram appeared accurate and discriminating, with a validation sample area under the receiver operating characteristic curve (i.e., a comparison of the predicted probability with the actual outcome) of 0.79. Because of these results, we developed and now use the preoperative nomogram that can help predict the 5-year probability of treatment failure among men with clinically localized prostate cancer treated with radical prostatectomy.

[Enlarge Image] Preoperative Nomogram for Prostate Cancer Recurrence

Although models exist that place patients into discrete groups at various risk levels for disease recurrence following surgery for prostate cancer, we know of no published work that combines pathologic factors to predict an individual's probability of disease recurrence, a more precise endpoint.

Therefore, we developed a second or postoperative nomogram for this purpose.

	[Enlarge Image] Postoperative Nonogram for Prostate Cancer Recurrence

We modeled the clinical and pathologic data and disease follow-up for 996 men with clinical stage T1a to T3c NXM0 prostate cancer, who were treated with radical prostatectomy at our institution. Prognostic variables included:

• pretreatment serum PSA level,
• Gleason sum from biopsy specimens,
• prostatic capsular invasion,
• surgical margin status,
• seminal vesicle invasion, and
• lymph node status.

Treatment failure was recorded when there was a rising serum PSA level, clinical evidence of cancer recurrence, or initiation of adjuvant therapy. Validation was performed on this set of men, and a separate sample of 322 men from the practices of 5 other surgeons from our institution.

Cancer recurrence was noted in 189 of the 996 men. The recurrence-free group had a median follow-up period of 37 months (range 1 to 168 months). The 7-year recurrence-free probability for the entire group was 73 percent (95 percent CI = 68 percent to 76 percent). The predictions from the nomogram appeared to be accurate and discriminating, with a validation sample area under the receiver operating characteristic curve of 0.89. Our postoperative nomogram is useful for predicting the 7-year probability of disease recurrence among men treated with radical prostatectomy. (6/8/2000)