Drs. Forman and Velasco provide a timely and thorough review of the maturing concept of applying radiation therapy to the prostatic fossa after radical prostatectomy. The guidelines for therapy continue to evolve because of the increasing reliance on blood prostate-specific antigen (PSA) level for both detecting a recurrence of disease and evaluating response to radiotherapy.
There is considerable evidence that moderate doses of external radiation to the prostatic fossa are extremely effective in preventing clinical (palpable) local failure in selected post-prostatectomy patients. The clinical local failure rate after treatment with a combination of surgery and postoperative radiotherapy is only about 5%. Verifying the complete destruction of all cancer cells in the prostatic fossa is much more difficult. PSA data do, however, support elimination of microscopic cancer within the radiation field, making it reasonable to conclude that postoperative radiotherapy works extraordinarily well. Accepting this, the next logical decision concerns who to treat and when.
Who Should Be Treated?
Management philosophy issues quickly become important in such a decision. If a patient has only a 20% chance of cure from a treatment modality that has relatively few side effects, should the treatment be applied to all patients, knowing that 80% may not benefit? If morbidity from the treatment were high or expense prohibitive, the impetus would be to treat only those who are most likely to benefit. However, since postoperative radiotherapy is safe and comparable in price to many other palliative treatments, should radiation be administered liberally? Who should answer this questionthe physician, the patient, the insurance company, or the payor? For example, a patient with seminal vesicle invasion in a radical prostatectomy specimen and a rising PSA less than 1 year after surgery is likely to have distant metastases. Should postoperative radiotherapy be offered if there is only a 20% chance for a durable normalization of PSA and cure? I dont know the answer.
Recently, in our Multidisciplinary Urologic Oncology Clinic comprised of physicians from the urology, radiation oncology, and medical oncology sections, we agreed on treatment guidelines for the patient with a rising PSA after radical prostactectomy. Basically, if the patient has positive nodes or a PSA that remained elevated postoperatively, he is encouraged to receive systemic therapy. All other patients are directed toward postoperative radiotherapy, ideally before the PSA rises above 1.0 ng/mL. However, when we encounter a patient who may have only one positive node or an immediate postoperative PSA of only 0.2 or 0.3 ng/mL that rises very slowly, we may deviate from own guidelines and treat the patient with postoperative radiotherapy.
The Forman-Velasco paper emphasizes the high success rate achieved with careful selection of patients. This is good information. Nevertheless, among the poor-risk group of patients with a PSA more than 2.0 ng/mL, 22% showed no evidence of disease after treatment. Similarly, in the poor-risk group with positive seminal vesicles or lymph node metastases, 38% had no evidence of disease. Thus, we still have the awkward dilemma of potentially withholding treatment when one-quarter or one-third of patients may benefit.
A Notable Observation
The authors go on to make an important and notable observation. If the PSA does not decline during treatment, logic would suggest a source of PSA production outside of the radiation field. It is conceivable that a decision could be made after the patient has received 40 or 50 Gy of radiation as to whether to continue or abort therapy based on a rising or declining PSA. Such a strategy, although a rather expensive therapeutic test, may obviate any treat- ment-related morbidity in those who would not benefit.
Are there any circumstances in which postoperative radiotherapy is advised based solely on pathologic characteristics when the postoperative PSA is undetectable? It is conceivable that a patient could have poorly differentiated cancer with a low postoperative PSA and grossly positive margins, making the risk for local recurrence high and the PSA level somewhat unreliable. However, no evidence exists to support the theory that postoperative radiotherapy delivered later in the disease course based on PSA status is equivalent to earlier postoperative radiotherapy based on pathologic stage. However, that uncertainty is offset by logic, in that the recurrences based on PSA are still very early, and local recurrences after postoperative radiation therapy based on PSA status are extremely rare. If patients are treated purely on the basis of pathologic stage, a large percentage of patients who are not destined to relapse will receive treatment unnecessarily.
Prostate-specific antigen as a guide to the use of radiotherapy after radical prostatectomy became widespread during a national trial that randomized patients to radiotherapy or observation based solely on pathologic stage. Reliance on PSA dramatically hindered accrual to this study and represents an illustration of how the rapid evolution of medical care can render a study somewhat outdated before it is completed.
Unfortunately, most diagnostic studies attempting to locate the site of disease in patients with a rising PSA after radical prostatectomy, (eg, digital rectal examination or imaging studies, such as CT, MRI, transrectal ultrasonography, and bone scan) are of negligible value. A radiolabeled monoclonal antibody imaging study (ProstaScint) seems to be able to detect both patients with recurrences limited to the prostatic fossa and those with disease elsewhere. However, experience with ProstaScint is limited in patients with extremely low levels of PSA, the very group we would like to target most aggressively. Yet, when a scan suggested disease outside of the radiation fields, 25% of these patients were judged free of disease by PSA. Thus, the ProstaScint test may be helpful but not conclusive.