Comparative effectiveness research and localized prostate cancer


Randomized clinical trials will be carried out to determine the most effective forms of treatment for localized prostate cancer — but most of them probably aren’t going to be done in the good old USA. According to an article in The Gray Sheet, this was the take-away from a discussion on comparative effectiveness research in prostate cancer held in early December as part of a meeting coordinated by the National Institutes of Health and the Agency for Healthcare Research and Quality (AHRQ).

Although the research community is generally agreed that randomized clinical trials are the “best” way to get good answers to such questions as whether brachytherapy is a better option than robot-assisted laparoscopic prostatectomy (RALP) for the treatment of early stage, localized prostate cancer, things like device marketing, patient preferences, physician bias, and financial incentives make it near to impossible to randomize patients to appropriate clinical trials in the U.S.

“The key comparative research questions in prostate cancer will be answered, but they won’t be answered in this country,” according to Anthony Zietman, professor of radiation oncology at Harvard Medical School. “They’re going to be answered in Britain,” he said, and perhaps other countries like Sweden and Canada, where physicians’ incomes aren’t as dependent on the technology they use and patients aren’t bombarded with advertising for the latest gizmo and gadget.

David F. Penson, professor of urologic surgery at Vanderbilt Medical Center, appears to have been in nearly complete agreement with Zeitman, suggesting that the use of newer technologies proposed as methods for the treatment of prostate cancer should be covered in the U.S. only when given to patients enrolled in randomized clinical trials.

The difficulties associated with randomizing patients to trials that compare things like surgery and radiation, or radiation and high-intensity focused ultrasound, in the U.S., are well understood. Indeed, many American researchers were astonished to learn that Canadian researchers had been able to randomize over 240 patients to treatment with external beam radiation therapy (EBRT) and cryotherapy at a single institution some years ago (although the original goal had actually been 480 patients).

The ProtecT trial, which has randomized > 1,500 newly diagnosed, British patients with localized prostate cancer to surgery or EBRT or active surveillance (out of a projected 2,050), used special research nurses rather physicians to recruit patients rather than physicians. The nurses had been trained to explain the uncertainties of prostate cancer management, and were very successful in convincing patients to accept randomization. By contrast, in the SPIRIT trial, which was conducted in Canada, patients were asked to meet with a urologist and a radiation oncologist who each presented their preferred treatment options to the patient in the same room, together, immediately following a biopsy. The trial failed to accrue anything like the necessary number of patients.

The crux of the problem was nicely summarized by Steven Pearson, MD, who is the president of the Institute for Clinical & Economic Review at Harvard Medical School. He pointed out carefully that there is a real difficulty with asking patients to become informed participants in the management of their health care and simultaneously asking them to be participants in randomized clinical trials.

“We like to talk about the importance of patient preferences and decision-making or individualized health care,” Pearson stated, “And yet we want patients to be willing to be randomized between surgery, radiation and active surveillance.” He continued, “If you feed these patients information with the goal of their making preference-driven choices, it’s a very hard thing to then say, ‘Let’s flip a coin between your going into the hospital and having major surgery versus going home.’ ”

It may be the case that some questions can be resolved without the necessity for randomized clinical trials — at least in Dr. Zeitman’s opinion. However, The “New” Prostate Cancer InfoLink thinks that we either need trials of this type or we are going to need  data from a very sophisticated series of national, registry-based studies. Such registry studies may be as difficult to conduct as the randomized clinical trials themselves.

5 Responses

  1. Get back to me in 20 years on this. Although I am pretty certain we will probably have 20 new ways to treat localized prostate cancer ~ all claiming to be the best.

  2. This same discussion occurred over the last 20 years with everyone saying it will take too long. The time it takes to get a very important question answered should be the last reason not to undertake such a study. Unless something changes, another 20 years will go by with patients continuing to be uninformed about the real benefits and risks of each option.

  3. Several issues. Money is one of the primary ones. What would be the cost of the trials? Can we afford that now? What would we give up? What would we know at the end of the trial? As Tony says, would treatment even be similar when the trials are completed? Are we talking about one big trial or many little trials? Can we get good information through the cancer registries? SEER data base? Are there changes in trials that can speed up the process that can be used? If the CER discussion is being driven by cost savings then maybe we need good information about total cost for treatments that include treatment of side effects. Will there be decisions about treatments that will be made based on a lack of information? How will that impact on the individual? How do we make sure that as this process moves forward we don’t give up what is best for the individual for the good of the group? Maybe we need to begin to include ethicists in the discussions?

  4. Kathy asked a number of questions in her prior post. I have offered (in bold italic type) answers to some of these questions below.

    • What would be the cost of the trials? Enormous and as yet unknown if the data is to be worth having.
    • Can we afford that now? Frankly, I’d be surprised.
    • What would we give up? Choice and autonomy, which is why it will be almost impossible to do this using randomized trials in the USA.
    • What would we know at the end of the trial? We might know a lot; it depends on the trial structure. We might, for example, know the difference in clinical effectiveness of active surveillance, watchful waiting, and at least one form of invasive therapy in well-defined groups of patients with certain types of co-morbidities, and life expectancies.
    • Are we talking about one big trial or many little trials? Probably neither. My suspicion is that we are talking about several large trials over many years.
    • Can we get good information through the cancer registries? Not using the databases that exist at present; they don’t collect sufficient, relevant information.
    • Are there changes in trials that can speed up the process that can be used? That depends on what is being studied.
    • Will there be decisions about treatments that will be made based on a lack of information? We already make most decisions that way. If information is lacking, it is inevitable.
    • How do we make sure that as this process moves forward we don’t give up what is best for the individual for the good of the group? By thinking about two things at once. The harder question we are going to be faced with is (for example): How do we justify giving Person A Treatment X and Person B Treatment Y when there is only (say) a 5% difference in the probabilities of “better” outcomes and Treatment X costs $50 whereas Treatment Y costs $50,000.
    • Maybe we need to begin to include ethicists in the discussions? Ethicists are already highly engaged in this discussion.
  5. This article may add to the discussion about this topic.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: