In this issue of ONCOLOGY, Dr. Hanks further establishes his legacy by leading the charge for radiotherapy as the treatment of choice in men with clinically localized prostate cancer. Most urologists and some radiation oncologists tend to consider radiotherapy an alternative to radical prostatectomy. They tolerate radiotherapy for "older" patients but continue to argue that radiotherapy is not the treatment of choice for "young men," ie, those £ 65 years old. In contrast, Dr. Hanks and colleagues eloquently summarize the merits of state-of-the-art three-dimensional conformal radiotherapy (3D CRT). The authors argue that young men are cured at rates comparable to surgery, but with fewer side effects. The latter point is supported by recent studies that defend the notion that sexual function and continence fare more favorably with radiotherapy than surgery. Conversely, rectal irritation tends to be worse after radiation. To what degree these outcome differences affect overall quality of life remains to be defined. That said, the focus of my editorial is on the issue of cure.
What Does ‘Cured’ Mean?
Legally, a patient who has no clinical or biochemical evidence of disease at 3 years can be considered cured at last follow-up. This does not mean that he will not relapse later. At the other extreme, a man who dies 20 years after radiotherapy treatment, with no clinical or biochemical evidence of disease, would require an autopsy with a histopathologic step-section evaluation of the prostate to prove conclusively that every last cancer cell was eliminated. Obviously, neither of these examples provides a practical standard for defining what most people mean when they use the term "cure." Dr. Hanks and coauthors provide an argument for cure being placed somewhere in the spectrum between the two extreme examples described, but to which extreme does their concept come closest? Put another way, how confident can we be that these patients are really cured?
Before attempting to address this question, I must mention that I don’t believe cure is the most important issue when treating the patient with prostate cancer. As an oncologist who has treated many patients with cancer, I frequently consider it an acceptable outcome if I can help someone live longer and have a good quality of life, even if they are not cured. Don’t get me wrongI would like to cure every patient I treat; unfortunately, I don’t. If the therapy I administer is associated with as good a chance of cure as an alternative therapy (such as surgery), I am comfortable that I am doing "the right thing."
Although it remains unclear to me what percentage of patients are really cured with radiation, I concur that it appears to be a number similar to those who are cured by surgery. That raises another question: What percentage of patients are cured with radical prostatectomy?
Cure After Prostatectomy and Radiotherapy
Essentially all of the large institutional series that have reported failure rates after radical prostatectomy have shown continued failures.[2-4] For example, in the series by Stein et al, only 40% of patients with T1/2 disease were disease-free at 10 years, and additional failures were noted at 15 years. In this report, there appeared to be a constant failure rate of approximately 7% per year through the entire timeline that was provided.
Similarly, there was no plateau in the disease-specific survival at 20 years in a more contemporary series from Duke University Medical Center. Of note, when matched for stage and Gleason score, the disease-specific survival rates at 5, 10, and 15 years appeared very similar to rates of patients treated with radiotherapy alone in prospective, randomized trials by the Radiation Therapy Oncology Group (RTOG). Thus, I agree with Dr. Hanks; survival rates or cure appear to be comparable between surgery and radiation. However, I’m still not sure how many men are truly cured.
What about those curves that demonstrate an apparent cure after 5 years? Using the same criteria for failure (the American Society for Therapeutic Radiology and Oncology consensus definition), Vicini et al noted the criticality of adequate follow-up for reaching firm conclusions about prostate-specific antigen (PSA) failure rates. They concluded: "Depending upon the length of time after treatment, significantly different rates [of PSA failure] (varying by 15% to 30%) can be calculated for the same time interval chosen for analysis." These authors suggested that to accurately estimate PSA failure rates, the median follow-up needed to be on the order of 2 years beyond the date of interest. Therefore, their data would imply that only the authors’ 3-year estimates are stable, and that the 5-year control rates are likely to further decline with longer follow-up.
Thus, the following assertions by Dr. Hanks et al must be viewed with caution: "Hazard functions show that the risk of failure is very low after 5 years [and] we have not seen any after 6 years." In the data presented, only 20 patients are at risk at 8 years and only 6 are at risk at 10 years. I am uncomfortable about estimating cure rates with this limited follow-up.
This criticism is consistent with observations that were made in a recent surgical series. This series demonstrated that (at least) 25% of PSA failures occurred 5 years or more after prostatectomy. Thus, it would be quite remarkable if a similar percentage of patients did not ultimately fail radiotherapy. Otherwise, one would be obligated to argue that long-term results with radiotherapy would be expected to be better than surgery, since 5-year PSA failure rates appear to be comparable between surgery and radiation. Let’s also consider the fact that in the surgical series just mentioned, the median time to metastatic disease after biochemical failure following prostatectomy was reported to be 8 years, and in time to death, an additional 5 years later. These data suggest that death after biochemical failure following prostatectomy can be expected to occur on the order of 15 years after PSA failure.
If patients continue to die from prostate cancer more than 20 years after undergoing radiotherapy, it is likely that at least some of them will have experienced failure beyond 5 years. Data reported by Crook et al and the RTOG support this assertion; however, these patients were treated to relatively low doses.[5,9]
Yes, survival rates following radiotherapy are comparable to those of surgery. Some men are considered cured, but I’m not sure how many of them truly are. The authors suggest that there is light at the end of the tunnel, implying that with a median follow-up of 5 years, the number cured is fairly high. Dr. Hanks has been right before, and he might be again. I hope he is, but only time (and longer follow-up) will tell.