This issue represents a rather detailed look at the current state of blood and tumor tissue testing in prostate cancer. The PSA is the classic example against which new tests can be compared. The PSA is largely useful in so far as the measured value is proportional to the total mass of prostate tissue, normal plus malignant. As a screening tool, PSA elevation can signal the presence of cancer and the PSA rate of increase can reflect the pace of cancer growth.
Unfortunately, other processes, such as prostatitis, can also cause PSA elevation and aggressive cancers can make less PSA per gram of tissue. These factors have limited the value of the PSA for screening.
For patients who are post prostatectomy or radiation therapy, there are fewer issues and the PSA doubling time is a widely used prognostic indicator for patients with recurrent disease.
How do the other tests discussed in this issue act to provide additional or more accurate information about prostate cancer?
Genomic tests can provide information on important functional differences between normal tissue and cancer. Perhaps the best example of this are mutations in the DNA repair genes, BRCA2 and ATM. Germline mutations in these genes increase the risk of prostate cancer and the resultant cancers tend to be more aggressive. These mutations can also develop as prostate cancer progresses and have been observed in 20-30% of hormone resistant cases.
Furthermore, cancer cells bearing these mutations are more likely to respond to drugs that inhibit PARP, a different DNA repair protein. Thus, testing for BRCA2 or ATM tells us about changes in the cancer’s ability to repair DNA damage, shed light on the risk of aggressive disease, and help select effective treatment. Tests like these that reveal functional differences between normal and cancer cells have great promise to improve treatment of prostate cancer. This approach has already been successfully applied to other cancers, such as non-small cell lung cancer, where it has revolutionized treatment.
Genomic changes need not be assessed at such a focused fashion as is done with BRCA2. Instead, multigene DNA, RNA, or protein patterns can be measured and tested for their ability to predict cancer aggression or response to standard therapy. Examples of this approach include the Decipher, Oncotype Dx, and Prolaris tests.
These tests can be viewed as supplementary to or even competitive with standard histologic evaluation by a pathologist. This has naturally led to some controversy about their role. In this issue, Dr Epstein does an effective job critically comparing histology versus the genomic approach. In my view, the relative role of the two ways of assessing future cancer behavior is still very much an open issue.
We hope you enjoy this issue as much as we have enjoyed putting it together.
Charles E. Myers, Jr., MD