Overexpression of COX-2 Found in Prostate Cancer Metastases

by Ed Susman | Doctor's Guide | 11.09.2006

PHILADELPHIA, PA -- A retrospective review of tissue samples from prostate cancer patients found an association between overexpression of cyclooxygenase-2 (COX-2) and the presence of distant metastases, researchers said here at the American Society of Therapeutic Radiology and Oncology (ASTRO) 48th annual meeting.

The researchers looked at the genetic makeup of the tumors of patients in the Radiation Treatment Oncology Group protocol 92-02 study that considered the duration of androgen deprivation therapy in men with prostate cancer and particularly the possible role of COX-2, a known marker of inflammation and cell proliferation.

"This is the first study to establish the association of COX-2 expression to prostate cancer outcome in radiation-treated patients," said Li-Yan Khor, MD, research associate in radiation oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania.

In her presentation on November 8th, Dr. Khor reported that higher levels of COX-2 expression were associated with biological failure -- an increasing prostate specific antigen (PSA) level -- and distant metastases.

The study used a COX-2 Intensity Score, an automated scale of the depth of the color related to COX-2 staining on pathological samples.

Dr. Khor said that about 10% of patients with scores under 134 on the scale had metastatic disease compared with 14% of patients with scores greater than 134, the level she used to compare outcomes.

Those patients whose tumors had staining scores 134 or greater had a 49.9% greater risk (95% CI, 1.010, 2.227, P = .0448) of developing distant metastases on the basis of a multivariate analysis, Dr. Khor said.

Other risk factors for metastatic disease as developed through the multivariate analysis were a Gleason score of 7-10 -- associated with a 2.107-fold increase in the likelihood of distant metastases (P = .0019), and T3 or T4 clinical stage, associated with a 67.4% increased risk of distant metastases (P = .0150).

Dr. Khor also said the analysis showed that using the 134 COX-2 Intensity Score as a cutoff resulted in a 48.9% increased risk of biological failure (P = .01) in patients were assigned to a short-course of androgen deprivation therapy in addition to radiation treatment. There did not appear to be any significant risk of biological failure based on COX-2 staining intensity of patients who were assigned to long-term androgen deprivation treatment, she reported.

"These results indicate COX-2 may be more useful as a tool for selecting patients for short-term androgen deprivation rather than for predicting treatment outcome," Dr. Khor said.

However, she said the work holds out the possibility that in high-risk prostate cancer patients the use of COX-2 inhibitors might be potentially beneficial and should be considered for use in future clinical trials.

[Presentation title: Cox-2 Overexpression Predicts Prostate Cancer Outcome: An analysis of RTOG 92-02. Abstract 1124]

Copyright 2006 Doctor's Guide