Understanding why placebos can have therapeutic benefit


Regular readers will be familiar with the fact that new drugs for prostate cancer are often tested against a “placebo.” They may also have heard the term “the placebo effect,” which generally implies that some interventions with no known (or theoretically possible) therapeutic benefit can be demonstrated to have a very real therapeutic effect.

However, we are still only beginning to understand why placebos can have such clinical benefit under carefully defined circumstances. The human mind works in some very mysterious ways. But it is because of the recognition that the therapeutic impact of placebos is very definitely real that we evolved the process of randomized, double-blind, placebo-controlled clinical trials that is now the gold standard for demonstration of therapeutic benefit and safety of new drugs in particular.

In an excellent article in this month’s issue of The Saturday Evening Post, Sharon Begley (who is definitely one of the beter journalists who currently specializes in writing about complex scientific and medical matters for non-scientists) has addressed “Placebo power,” offering the lay reader an overview of the clinical power of placebos in medical practice — from the work of the (sometimes controversial) Henry Knowles Beecher and his colleagues to the very real randomized trial of true arthroscopic surgery for patients with arthritic knee joints (flushing out the joint and removing cartilage) versus sham surgery on others (see the abstract of the paper by Moseley et al.).

If you are one of those patients who is convinced that your specific type of treatment for prostate cancer really worked (or is still working) for you, even though there is no actual clinical evidence of its effectiveness from randomized, controlled clinical trials, Ms. Begley’s article may help you to understand why such trials really are important and why clinicians (and The “New” Prostate Cancer InfoLink) are often highly skeptical about whether such effects are anything more than a placebo effect.

Now this does not mean that placebo effects are bad. They aren’t. If some form of placebo is having a real therapeutic effect in individuals or groups of individuals, clearly this is a very good thing. Most placebos are extremely safe, for starters. In the world of prostate cancer, for example, placebo therapy could, possibly, be extremely effective and very safe in the treatment of men with low- and very low-risk forms of the disease, and may allow men to follow active surveillance protocols for years with limited need for follow-up biopsies and minimization of patient anxiety.

Now we are not actually recommending that such a trial be carried out, but it might be very interesting indeed to know what would happen if we were to conduct a randomized, double-blind clinical trial in men with low- or very low-risk prostate cancer in which the patients were divided into three groups:

  • Group A would consist of patients who were monitored by a standard form of active surveillance (to be defined).
  • Group B would include patients who were given a placebo that was formulated to look like an active, investigational drug (to be taken once a day) but was actually just a placebo (with a name printed on the tablet, etc.) and still monitored by the standard form of active surveillance.
  • Group C would include patients who were given a placebo that looked like a placebo (also to be taken once a day) but was actually exactly the same placebo as taken by patients in Group B.

The key clinical questions would be:

  • Whether men in Group B and Group C (and their physicians) suffered less anxiety about the likelihood of progression of their cancer than men in Group A, and were therefore less likely to seek (or recommend) active interventional treatment (surgery, radiation, etc.) in the absence of clear evidence of disease progression
  • Whether there was any real impact on progression-free survival of men in the three groups over time
  • Whether there was any real impact on overall survival of men in the three groups over time

The ethical appropriateness of such a trial would require extensive discussion before one could possibly consider its implementation, as would the appropriate form of active surveillance and the actual study endpoints, but it would certainly be informative to know whether either the apparent therapeutic intervention (despite being a placebo) or the obvious placebo had any statistically demonstrable therapeutic benefit.

7 Responses

  1. Re: “randomized, controlled clinical trials,” … It is my understanding that no such controlled studies have been done on the current treatments used to treat prostate cancer. Is that correct? Seems strange if true.

  2. The practice of psychiatry is based on this effect. If you think it will help you there is a good chance that it will. The question of actual slowing of tumor growth can only be a guesswork thing today. A trial would/might answer this.

  3. Dear Grant:

    This is not strictly true.

    Many years ago there was a small and much debated trial by Paulson et al. that claimed to show that radical prostatectomy was more effective that external beam radiation therapy in the treatment of localized prostate cancer. However, the structure of this trial — not to mention the types of patient, type of surgery, and the type of radiation therapy — have little relevance to men being treated for localized prostate cancer today.

    There have been two large, randomized trials comparing treatment with radical prostatectomy to “observation” of some type (including both active surveillance and simpler watchful waiting). Each of these trials demonstrated that men with low-risk prostate cancer and aged 65 years or above did just as well on “observation” as they did after surgery, and with fewer side effects.

    There was also a decent-sized trial in Canada not so long ago in which patients agreed to be randomized to treatment with either external beam radiation therapy or cryotherapy. This trial showed that there was no statistically significant benefit to either form of treatment, although there was a trend toward better outcomes among the men treated with radiation.

    However, it is true that there has never been a large, randomized, “modern” trial of any form of radiation therapy compared to any form of surgery.

    By comparision, almost every form of drug therapy approved for the treatment of prostate cancer (e.g., LHRH agonists, LHRH antagonists, antiandrogens, chemotherapy, and the most recent drugs like sipuleucel-T, abiraterone acetate and enzalutamide) have all been subjected to evaluation based on randomized, controlled clinical trials.

  4. Another fabulously compelling article. And that nice “placebo effect” is demonstrated best in the world of “herbal remedies” for obvious reasons.

  5. Has the “anti-placebo” effect been studied at all? In other words: To what extent, if any, is the clinical value of a medication diminished or vitiated by a patient’s (and/or doctor’s) disbelief in its efficacy, or ignorance of the fact that it is being administered?

    As an even greater possible example of anti-placebo effect: When a legal practice discovers a potential class-action cause like Agent Orange or powerline proximity, how many additional cases of occur for apparently no reason except the power of suggestion or publicity?

  6. Sorry Paul … I have no idea about research on any type of “anti-placebo” effect.

  7. There are the “nocebo” effects, which are the side effects or adverse reactions to a “placebo.” This can easily be demonstrated by the side effects suffered by patients in the placebo arms of randomized controlled trials.

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: