Help us to stop prostate diseases ruining lives
CONTINUATION OF THE PROJECT ON MOLECULAR GENETIC CHANGES IN PROSTATE CANCER:   GUIDES TO CLINICAL MANAGEMENT

Mark R Feneley, Senior Lecturer in Urological Oncology, UCL
Joe Nariculam, Research Registrar, UCL
Alex Freeman, Lecturer in Histopathology, UCL
Nicola Brookman-Amissah PhD, Prostate Cancer Research Centre, UCL and
Magali Williamson PhD, Prostate Cancer Research Centre, UCL

Mark Feneley

The aim of the project was to determine whether patients having radical prostatectomy with similar clinical and pathological features can be distinguished based on specific molecular alterations in the prostate tumour for predicting treatment failure.  Molecular changes in prostate cancer tissue were examined in patients with clinically localised cancer to distinguish those who may be cured by radical prostatectomy from those who are not.

The University College Hospital radical prostatectomy database was used to identify patients who have had radical prostatectomy for early stage prostate cancer.  41 patient pairs were mapped, characterised by distinct biochemical relapse status but matched for tumour stage and grade, and preoperative PSA.  Relapse status for comparison comprised PSA recurrence within 2 years of surgery or undetectable PSA at 3 years or more after surgery.  Using the matched pairs comparing samples from patients with and without biochemical progression, the specific research objectives are being achieved.

Prostate cancer tissue was embedded as a microarray to assess the prognostic value of various molecular biomarkers.  DNA and RNA has been extracted from the same tissues to assess the pattern of loss of heterozygosity (LOH) and pattern of mRNA expression.  Key alterations in tumour DNA, RNA and biomarker expression that may alter tumour behaviour were then examined.

The biomarkers investigated include p53, BCL-2 and Ki-67, E-cadherin, Plexin B1, Semaphorin 4D, PMSA, E2F3, EZH2, C-myc and MCM-2.  Previous published studies indicated these may be important after prostate cancer progression but others show quite variable prognostic value for predicting tumour behaviour.  Having demonstrated the reliability of our methodology, we showed that biomarker expression is abnormal in clinically localised prostate cancer for all but 2 of the biomarkers (E-cadherin and EZH2).  However, for all biomarkers except MCM-2, abnormal expression did not predict treatment failure after allowing for pathological stage, Gleason Grade and preoperative PSA level.  MCM-2 was studied in collaboration with Prof Gareth Williams at UCL and appears promising as an independent predictor of recurrence after treatment.

Nine sites of the genome were assessed for changes that may relate to cancer behaviour.  By extracting DNA from the prostate tumours, we found DNA alterations in 2 of the 9 sites that related to biochemical recurrence.  These 2 sites were then mapped for functional genes, and were shown to code for microRNA.  MicroRNAs are emerging as potentially important regulators of cell function, including regulation of Bcl-2.  We have developed a technique for extracting RNA from paraffin tissue, measured the expression of 5 microRNAs and hybridised microRNA from 5 tumour samples to microRNA microarrays.  With further analysis, we expect to identify a microRNA signature for recurrent prostate cancer.

Summary

Ongoing research will further characterise the molecular changes that are important in tumour behaviour and for selecting patients who will most benefit from radical prostatectomy.

The support of PRCUK is gratefully acknowledged.

Final summary report dated 01 December 2006
Project 2004/06