Over the years, it has proved almost impossible to recruit patients to comparative trials of different types of first-line treatment in the management of localized prostate cancer — particularly in the USA. There have been some exceptions to this general rule, but to a great extent either the patients themselves (or their doctors) have been reluctant to let themselves be “randomized” to treatment A vs. treatment B because they have had strong feelings about what they wanted to do (or, in the case of their doctors, to recommend) under specific circumstances.
In 2009 a set of British researchers set out to see if they could determine whether it would be possible to enroll patients to a large, randomized, controlled clinical trial of brachytherapy vs. radical prostatectomy in men with low- to intermediate-risk, localized prostate cancer. They didn’t want to just try to do they trial. Their first goal was to try to see if it would even be possible (which seems like a smart idea).
The results of the so-called SABRE1 study have now been reported by Eccles et al. in BJU International. It is probable that they have finally confirmed that any attempt to conduct a large, randomized trial of two differing types of first-line treatment in men with localized prostate cancer is utterly unrealistic. If that is the case, it only adds weight to the need for us to develop a sophisticated registry database program that will allow us to compare data from large numbers of patients treated by a variety of methods but based on the consistent collection of highly specified data from diagnosis to long-term follow-up.
Eccles et al. used their parallel, two-group, multi-center feasibility trial to enroll men with histologically confirmed, localized, low-risk prostate cancer and good performance status from five UK hospitals. These men were — initially — randomly assigned to receive a DVD-based decision aid (DA group) or “standard” information from their diagnosing physician (control group). They were then meant to be re-randomzed to treatment by brachytherapy or by radical prostatectomy. However, there was no ‘blinding” of staff or patients, so everyone knew exactly what the patients’ options were along the pathway.
Here is what happened:
- 30 patients actually underwent the initial randomization between May 2009 and May 2011.
- 15/30 were randomized to the DA group.
- 15/30 were randomized to the control group.
- Only 4/30 patients agreed to be re-randomized to either brachytherapy or radical prostatectomy.
- 1/15 from the DA group
- 3/15 from the control group
- SABRE 1 closed early due to poor recruitment.
Subsequent analysis of patient logbooks and questionnaires completed during the course of the study showed the following:
- The primary reasons for declining re-randomization to brachytherapy or surgery were
- The patients’ own desires to choose their treatment
- The patients’ desire to elect active monitoring
- Among 10 men in the DA group who completed a post-DA questionnaire
- 4/10 men ‘”felt surgery and radiotherapy had been proven in a high quality trial”.
- 7/10 men felt that they “should make their treatment decision while knowing their doctors opinion”.
The “New” Prostate Cancer InfoLink is by no means surprised by this set of results, and is pleased to not that Eccles et al. decided to do this study rather than just embarking on yet another trial of first-line treatment options that would clearly have failed.
As the authors note, of course, this means that
The best treatment method for treating low-risk prostate cancer is still unproven in a head-to-head trial and the increasing number of options will make choices correspondingly more difficulty without good quality comparative research.
We see sophisticated, high-quality registry systems as a simple and relatively cost-efficient replacement for any further attempt to randomize patients to trials like this. As we have noted before, such registry systems come with a potential regulatory benefit as well, as newer forms of treatment get approved for the management of men with localized prostate cancer. As we have also noted, authorities in Australia have already started to pilot such a system in the state of Victoria.
Filed under: Diagnosis, Management, Treatment, Uncategorized | Tagged: first line, localized, randomized, registry, SABRE1, trial |
THE "NEW" PROSTATE CANCER INFOLINK 
Why would any man whose reading extended beyond the Daily Mail (UK newspaper not well-known for its editorial skills in the medical field) for his prostate cancer pre-treatment education want to volunteer for a radical prostatectomy (or brachytherapy come to that) if they were in the “low-risk” category? Were they exposed to any information on active surveillance?
Dear Richard:
Since several of the patients in this trial clearly did opt for monitoring as opposed to either brachytherapy or surgery, I think it is safe to say that they were almost certainly exposed to information about active monitoring if they were in the DA group.
On the other hand, I think you might be surprised to find out how many men in America are utterly unwilling to consider active monitoring as a first-line management strategy for low-risk prostate cancer … but still wouldn’t be willing to be randomized in a clinical trial like this.
Many (more than usual) typos in this post, Sitemaster? What is the stage of a low-risk “localized” prostate cancer? Are the vesicles or local lymph nodes involved? I understand “capsule confined” or “locally advanced” prostate cancer, but “localized” seems ambiguous.
Hard though it may be for American men to learn up their options, I agree with Richard. I find it scandalous that “low risk” men get mutilated and irradiated at the drop of a hat. My own case was first diagnosed as “capsule confined” (T2a/b), but Gleason 3 + 4 interpreted as “moderately aggressive” (so hardly “low risk”).
Knowing a bit about anatomy, I believed the pictures in received information sheets were misleading or wrong in showing a burst capsule. Lymphatic drainage is more subtle.
I therefore insisted (UK, NHS) on having lymphadenectomies done, which raised me to “locally advanced” (T3N1 …). Thereafter I was pointed to intermittent hormonal therapy which continues as a form of “active surveillance with appropriate interventions”. I also get bone and CT scans periodically — so far they provide evidence only of ageing. I am now 9.5 years out from the definitive surgery, and almost two-thirds through the top-end of the prognosed (“How long have I got, doc?”) range of 2 to 15 years,
I opine that I shall succumb to the side effects of drugs before the cancer itself gets me! Only in the last 2 years have I been able to wind down my fights with doctors over their desires to (as I see it) over-treat me. When the most recent fellow (February 2011) reviewed my full history and started to believe my account of it, he admitted — in writing! — that I had “done amazingly well.”
Dear Chedley:
(1) “Low risk” localized prostate cancer is very highly defined at the time of initial diagnosis. Based on the D’Amico classification, this refers exclusively to patients with a PSA level of less than 10 ng/ml; a Gleason score of 3 + 3 = 6 or less; and a clinical stage of T2a or lower. Other classifications have been developed that introduce minor modifications to the classic D’Amico classification.
(2) Other forms of “localized” prostate cancer can include any form of cancer that has not metastasized (i.e., cancer is still confined to the soft tissues within and close to the prostate such as the seminal vesicles and even the regional lymph nodes). However, a cancer that does not meet the definition given in (1) above is most certainly not “low risk”.
(3) You appear to have been diagnosed initially with “intermediate-risk” but still localized prostate cancer (based on your Gleason score of 3 + 4 = 7 and the incliusion of T2b in your clinical stage). However, if your PSA level at diagnosis was > 20 ng/ml you should have been classified as “high risk” according to the D’Amico criteria. It would be perfectly normal for a man with such a diagnosis to be given bone and CT scans at diagnosis; to have lymph node dissections as part of his initial surgery; and to continue to receive occasional bone and CT scans if there was any clear signal of a rising PSA while on intermittent ADT.
(4) The idea that intermittent ADT would ever be classified as “active surveillance” of any type makes no sense. There is a clear treatment plan being implemented. “Active surveillance” is a management stategy used exclusively in men who have never received any form of first-line treatment. Either you or your doctors are misusing this term in your case. All patients who receive treatment get regular check=ups after first-line therapy. Such testing is not “active surveillance”.
(5) It isn’t particularly difficult “for American men to learn up their options”. You are missing the point. Even when American men with low-risk disease are strongly advised by their physicians to go on active surveillance to avoid the risks of over-treatment, a very high percentage of them will insist on immediate active treatment because of their anxiety and fear of “cancer”. It is (at least in part) a cultural issue.
Dear Mike:
Thank you for your comments.
What I’ve always found difficult to understand, in almost discussion on criteria/protocols, and whenever D’Amico is mentioned, is when we quote PSA values as cut-off points, without taking into account prostate volume, or, to be more precise, PSA density.
In my opinion, a PSA of 10 is meaningless without further information on whether the majority of the reading is due to prostate volume or not. It is fairly routine, at least in the UK, to use PSA density as a major factor in deciding suitability for active surveillance.
Further information, such as PSA velocity and doubling time, DRE, symptoms, etc., are also useful, if not essential, in making decisions, of course.
Have I got it wrong when I go on about the almost negligible usefulness of a PSA figure without density information?
Dear Richard:
The degree of importance of PSA density data is (frankly) not really well defined by detailed studies. Does it make sense to consider this in decisions about treatment and active surveillance? Sure it does. Do we have cut-off data related to PSA density that have been validated in the way the D’Amico PSA levels were validated? Not that I am aware of.