Commentary (Keane): Prostate-Specific Antigen as a Marker of Disease Activity in Prostate Cancer

This is a well-written and timely review of a topic that has recently become both complex to urologists and confusing to nonurologists. The authors discuss the physiology of prostate-specific antigen (PSA) and its role in a variety of clinical situations, highlighting the areas of proven utility and identifying areas of controversy.

The issue of PSA screening is not examined in this article, but the various sources of PSA are detailed, and the overlap of benign and malignant disease, the appropriate use of free- and bound-PSA values, and the benefits and limitations of ultrasensitive assays are emphasized. Further discussion of current methods to increase PSA sensitivity and specificity-such as the use of age-specific ranges, PSA velocity, and PSA density-and their role in cancer diagnosis, would also have been helpful.

Uses and Limits of PSA Measurement

The authors correctly point out that PSA is an excellent marker of disease activity following definitive local therapy, but there is still considerable debate as to the most appropriate value at which to initiate further investigation and therapy once an elevation has developed. PSA is clearly useful in distinguishing local from distant recurrence, and the authors describe a variety of techniques to improve accuracy, including ultrasensitive assays, formulas, rate of rise, timing of elevation, and correlation with Gleason grade. What is apparent is the stark contrast between the effectiveness of PSA as a marker of disease activity and the lack of useful imaging techniques to further define this clinical situation.

Moreover, the role of PSA as a surrogate marker of survival in advanced disease following hormonal therapy or secondary radiation therapy for local recurrence after surgery is unproven. No randomized study has yet confirmed a survival advantage for patients who achieve a PSA nadir post-radiation therapy following surgical failure. If one looks at the experience in metastatic disease, the use of PSA as a survival marker is even more questionable. In the largest randomized study of complete androgen blockade to date, a significant number of patients with metastatic disease normalized their PSA at 3 months following an orchiectomy plus antiandrogen therapy vs orchiectomy alone. However, this did not translate into a significant survival advantage.[1]

Although there is no doubt that PSA is very effective in identifying disease progression at any stage, should our focus be on the development of more sensitive markers of progression, methods to prevent progression, or at least more effective therapy for such progressions? Earlier identification has not yet translated into improved survival, and in view of the documented long natural history of this disease, does every recurrence require treatment? In order to answer these questions, we will either have to wait a long time or develop better surrogate markers of survival.

PSA Monitoring Postradiotherapy

The role of PSA in monitoring patients after surgery or radiation therapy is discussed in detail and well summarized. But the reader is struck by the obvious disparity between defining a recurrence following surgery and one following irradiation. The identification of a radiation failure appears to be after the fact, requiring at least three rises in PSA 3 to 4 months apart, and then claiming the time to recurrence as the midpoint between the nadir and the first of three consecutive rises.[2] This may be acceptable today, given that there is no definitive curative therapy in such circumstances. If an effective therapy is developed, this definition will clearly have to be revised, because the resultant delay in declaring a radiation failure could have a negative impact on outcome.

Once again, considerable effort has gone into identifying patients who are likely to fail radiation therapy both before and after therapy. As in surgical series, PSA, PSA doubling time, nadir, Gleason grade, and clinical stage are used to determine low- and high-risk groups.

Hormones and Radiation

Finally, the role of hormonal therapy in high-risk patients undergoing radiation therapy is discussed. [Editor’s note: Part 2 of the article by Partin et al, to appear in the September issue of ONCOLOGY, will more fully explore the use of PSA monitoring in prostate cancer patients receiving hormonal therapy.] The first trial, RTOG 8610, examined the use of neoadjuvant hormonal therapy during radiation therapy (4 months total) vs radiation alone. Although this study showed an improvement in overall survival in the combined-therapy arm, it was only seen in patients with well-differentiated (Gleason 2-6) tumors-a finding that is confusing and may demonstrate the problems of using subset analysis.[3]

The second study mentioned is RTOG 9202, which compared neoadjuvant and adjuvant hormonal therapy plus radiation therapy vs neoadjuvant hormonal therapy plus irradiation alone. With a median follow-up of 5 years, a benefit in overall survival has not been shown, but subset analysis and comparison with other studies does suggest a benefit in patients with higher Gleason grades. Once again, these results must be interpreted with great caution, because the comparison of the overall groups within the study failed to show an overall survival benefit. Longer follow-up will hopefully settle the issue.

Overall, this is a very informative article highlighting the benefits and pitfalls of PSA assessment in the management of prostate cancer; it clearly defines the state of the art.

References:

1. Eisenburger MA, Blumenstein BA, Crawford ED, at al: Bilateral orchiectomywith or without flutamide for metastatic prostate cancer. N Engl J Med339:1036-1042, 1998.

2. ASTRO Consensus panel: Consensus statement: Guidelines for PSA followingradiation therapy. Int J Radiat Oncol Biol Phys 37:1035-1041, 1997.

3. Androgen ablation adjuvant to definitive radiotherapy in carcinoma of theprostate: Year 2000 update of RTOG Phase111 studies 8610 and 8531. AmericanSociety of Therapeutic Radiation Oncology, 2000.