One of the major debates about the appropriate management of prostate cancer is whether deferral of treatment by low-risk patients is or is not associated with an increase in risk for disease progression and prostate cancer-specific mortality post-treatment.
To date there has been no prospective clinical trial completed that addresses this question. As a consequence, a number of research groups have carried out retrospective analyses of the outcomes data on their patients treated over the past decade or more, and not surprisingly they have come to a variety of very different conclusions. The obsessive reader might want to look at the data reported in studies by:
- Nam et al. (from 2003), based on 645 patients from the University of Toronto group
- Khan et al. (from 2004), based on 962 patients from the Johns Hopkins group
- Vickers et al. (from 2006), based on 3,149 patients from the Memorial Sloan-Kettering group
- Freedland et al. (from 2006), based on 895 patients reported in the SEARCH database
- van den Bergh et al. (from 2010), based on 227 patients reported in the Swedish data from the ERSPC trial
A new study just published by O’Brien et al. is based on Catalona’s series of 1,900 patients, all treated by standardized radical prostatectomy between 1989 and 2009. Of these 1,900 patients, 1,111 met the D’Amico criteria for low-risk prostate cancer (a biopsy Gleason score of ≤ 6, a clinical stage of ≤ T2a, and a preoperative PSA level of ≤ 10 ng/ml).
O’Brien et al. divided these 1,111 patients into two groups:
- Group A included all patients who were treated by radical surgery within 6 months of diagnosis (n = 1,052).
- Group B comprised only those patients whose surgery was delayed for at least 6 months from time of diagnosis (n = 59).
Careful analysis of the data from these two groups of men showed that:
- Men in Group B were older, and more likely to be African-American than men in Group A.
- Men with palpable disease (clinical stage T2) were significantly more likely to get treatment within 6 months of diagnosis (p = 0.02).
- There were no significant differences in the rates of organ-confined disease, extracapsular extension, positive surgical margins, seminal vesicle invasion, or positive lymph nodes between Group A and Group B.
- Men in Group B were significantly more likely to be upgraded to a Gleason score of 7 to 10 at surgery than men in Group A.
- 27 percent of men in Group A were upgraded at surgery.
- 47 percent of men in Group B were upgraded at surgery.
- Men in Group B had a slightly greater risk of biochemical recurrence (a PSA level of ≥ 0.2 ng/ml) within 6 months of surgery than men in Group A (p = 0.17).
- Biochemical recurrence was observed in 5 percent of men in Group B.
- Biochemical recurrence was observed in 12 percent of men in Group A.
- Men in Group B had a lower 5-year progression-free survival than men in Group A.
- After controlling for PSA and clinical stage, men in Group B had a 2.9-fold increase in risk of biochemical recurrence compared to men in Group A.
So let us be very clear up front, if you are initially diagnosed with low-risk prostate cancer and you do absolutely nothing about it at all for 6 months, it makes perfect sense that you have increased your risk for subsequently being found to have progressive disease if you then have surgery! It is for this very reason that most centers now carrying out carefully structured active surveillance studies consider it essential that you have repeat testing (including a repeat biopsy) before you are enrolled into an active surveillance protocol.
It is also worth noting that Catalona’s series of patients extends back into the late 1980s, when a standard biopsy was conducted by taking just six biopsy cores. This would very likely have increased the possibility of “missing” a small amount of Gleason 7 cancer or higher in patients so biopsied. The paper provides no information on the number or range of biopsy cores taken in this series of patients, so we have no idea what percentage of the 59 patients in Group B had an initial biopsy based on fewer than eight cores. However, we feel that it is disingenuous to refer to this series of patients as “contemporary.” Most surgerons today would reserve that term for a series of patients diagnosed and treated no earlier than about January 1, 2000.
We are going to need to wait for the results of the currently ongoing clinical trials like the START trial and the ProtecT study before we really can tell whether deferred therapy is associated with a significant increase in risk for progressive disease. It is worth noting that the START study isn’t expected to provide final results until 2023. By comparison, we can hope for data from the ProtecT study by some time in 2014.