Nomogram Incorporating PSA Level To Predict Cancer-Specific Survival For Men With Clinically Localized Prostate Cancer Managed Without Curative Intent

Main Category: Prostate / Prostate Cancer
Also Included In: Urology / Nephrology;  Men's health
Article Date: 08 Feb 2008 - 0:00 PDT

email to a friend   printer friendly   opinions  

UroToday.com- In the online version of Cancer, Dr. Kattan and collaborators report an internally validated nomogram for predicting the probability that a man will survive his prostate cancer (CaP) if he does not have it treated immediately. The research was a population based study using cases identified from six Cancer Registries in the United Kingdom.

Men included in the dataset had CaP diagnosed between 1990 and 1996 by transurethral resection of the prostate or prostate needle biopsy. Baseline PSA was <100ng/ml, metastatic evaluation negative, and excluded were men over age 76 years or with competing health risks. Men undergoing definitive therapy within 6 months of diagnosis were also excluded. Data was collected at the 51 hospitals by on site medical record reviews. Clinical stage and tumor grade were reviewed but when the Gleason score was not available it was statistically imputed. Baseline PSA values were required as they were incorporated into the prediction model.

Nomogram validation consisted of two steps. The discrimination was quantified with the concordance index, which provided the probability that in a randomly selected pair of patients in which one patient dies before the other, the patient who died first had the worse predicted outcome from the nomogram. This index was evaluated in the test set of patients using 5 different models. Secondly, calibration was assessed by grouping patients with respect to their nomogram-predicted probabilities and then comparing the mean of the group with the observed Kaplan-Meier estimate of disease-specific survival. This was also evaluated in the test set. A Cox regression model was fitted using clinical stage, biopsy Gleason grade, method of diagnosis, percent cancer, baseline PSA, age at diagnosis, and whether hormones were given at time of diagnosis. The nomogram discriminated with concordance indices of 0.73-0.74 based on the 5 different models.

In the patient dataset studied, 24% of men died of CaP within 10 years. The authors point out that this is similar to men treated with gold seeds. Biopsy Gleason grade had the largest effect on disease specific survival, followed by age, serum PSA, and clinical stage. They urge external validation using another dataset. Also, 27% of cases had no diagnostic tumor grading and 40% had no clinical staging.

In an accompanying editorial, Dr. Peter Albertsen points out that the model should be used with caution due to the patient cohort used. Patients included from the UK during the 1990's mostly did not have CaP screening and were diagnosed by TURP or nodules on examination. This does not reflect the contemporary group of men diagnosed in the US presently.

Kattan MW, Cuzick J, Fisher G, Berney DM, Oliver T, Foster CS, Møller H, Reuter V, Fearn P, Eastham J, Scardino PT, and the Transatlantic Prostate Group

Cancer. 2008 Jan 1;112(1):69-74
Doi: 10.1002/cncr.23106

Reported by UroToday.com Contributing Editor Christopher P. Evans, M.D

UroToday - the only urology website with original content written by global urology key opinion leaders actively engaged in clinical practice.

To access the latest urology news releases from UroToday, go to: www.urotoday.com

----------------------------
Copyright © 2007 - UroToday
Reproduced for Medical News Today with permission of UroToday.
----------------------------

• Additional
• References
• Citations