Zumsteg et al. searched the database at Memorial Sloan-Kettering Cancer Center (MSKCC) to determine the risk factors associated with cancer progression after primary radiation treatment, and the timing of progression. They report their results in European Urology, along with editorial comment by Mack Roach of the University of California, San Francisco.
This retrospective analysis looked at records of 2,694 patients treated at MSKCC with radiation doses between 76 and 86 Gy. The median follow-up was 83 months for all patients and 122 months for those who experienced biochemical failure (defined by the nadir PSA + 2 ng/ml). The authors did not report what, if any, salvage treatment was used.
The researchers found that:
- 23 percent of patients experienced biochemical failure.
- The median time from biochemical failure to detection of distant metastases was 5.4 years.
- The median time from biochemical failure to prostate cancer-specific mortality was 10.5 years, 5.1 years after metastases were detected.
- Risk of clinical progression following biochemical failure were independently associated with:
- Shorter PSA doubling time
- Higher clinical stage
- Higher Gleason score
- Shorter time to biochemical failure
Based on the John Hopkins surgical series, Freedland et al. reported (in 2005) that, for men treated with surgery, 19 percent experienced biochemical failure.
Some of the difference may be attributable to the inadequate dose of radiation (76 Gy) used on some patients in the MSKCC series, or that those patients were diagnosed with more aggressive disease. The median time from biochemical recurrence to detection of distant metastases in the Johns Hopkins series was 8 years, 3 years among those who did not have salvage radiation after biochemical recurrence (see Antonarakis et al.), The shorter time in the radiation study may reflect the fact that patients choosing radiation have historically been older and further progressed at time of diagnosis. The median time to death after metastases were detected was 5 years – identical in both studies. They all report the same risk factors for clinical progression.
The numbers reported for initial radiation therapy are similar, at first blush, to those reported for initial prostatectomy. Because there will probably never be a randomized clinical trial of surgery vs. radiation, it is tempting for the patient faced with the choice of initial therapy after diagnosis to compare these data sets, both from top institutions in their respective specialty. While I would very much like to see the patient characteristics and the data stratified by risk group and salvage treatment, if any, there does seem to be a similar overall pattern. Some patients will have already experienced undetected micrometastases before treatment, and they will not be cured by either therapy using current methods. Other patients — most in fact — will be cured by either therapy.
Editorial comment: This commentary was written by Allen Edel for The “New” Prostate Cancer InfoLink.
Filed under: Diagnosis, Management, Risk, Treatment | Tagged: failure, first line, outcome, radiation, surgery |
THE "NEW" PROSTATE CANCER INFOLINK 
Although Allen Edel is correct that (with 99.9% certainly) there is never going to be a true head-to-head, randomized trial of first-line radiation therapy vs. first-line surgery in a directly comparable series of patients — here in the USA or anywhere else in the world, it is worth noting the ongoing ProtecT trial in the UK.
ProtecT is a very highly structured registry trial in which some of the patients have been randomized to active treatment (with surgery or radiation therapy) or active surveillance. I remain optimistic that this trial is going to give us some much clearer data than anything we have to date on the relative merits of surgery vs. radiation therapy as first-line treatment in a large number of well-defined patients (as well as real data on the relative benefits of active surveillance compared to active treatment).
A paper by Lane et al. has already provided preliminary data showing that > 1,600 patients have been randomized in this study (out of the 100,000+ men who were asked to come for screening as possible candidates and the nearly 3,000 men who were diagnosed with prostate cancer as a consequence).
Actual data giving first-line results from the ProtecT trial, may be available by the time of the ASCO meeting in 2016 or some time in 2017.