Holmberg’s Randomized Controlled Study without Spin

The study below needs to be interpreted without spin for the benefit of prostate cancer patients. It’s a randomized controlled study, which is the best kind of study that can be done. Randomized controlled studies, when done correctly, provide the truth without bias. The question being asked by the study is: Does the radical prostatectomy provide survival benefit. The answer is no, it does not. As the authors say, “… there was no significant difference between surgery and watchful waiting in terms of overall survival.”

Lars Holmberg, M.D., Ph.D., Anna Bill-Axelson, M.D., Fred Helgesen, M.D., Jaakko O. Salo, M.D., Ph.D., Per Folmerz, M.D., Michael Häggman, M.D., Ph.D., Swen-Olof Andersson, M.D., Ph.D., Anders Spångberg, M.D., Christer Busch, M.D., Ph.D., Steg Nordling, M.D., Ph.D., Juni Palmgren, Ph.D., Hans-Olov Adami, M.D., Ph.D., Jan-Erik Johansson, M.D., Ph.D., Bo Johan Norlén, M.D., Ph.D: A Randomized Trial Comparing Radical Prostatectomy with Watchful Waiting in Early Prostate Cancer. New England Journal of Medicine. September 12, 2002, Number 11, Volume 347:781-789.

Instead of reporting the overall survival, many places focus on the prostate cancer-specific survival result from this study, which makes the radical prostatectomy look good to those who like to interpret data in a biased fashion in favor of surgery. Let’s be clear, the result of this study is that the radical prostatectomy does not extend survival.

Reporting prostate cancer-specific survival can make the ridiculous appear to be beneficial. Let’s say that 1 in 1,000 men get a slow-growing cancer of the right leg. Let’s do a study of 2,000 men after randomizing them. One thousand men will undergo watchful waiting and 1,000 will have their right legs amputated to prevent them from some day dying of a slow-growing cancer of the right lower leg. So what happens if we follow all the men until they all die? Only one man in the watchful waiting group gets slow-growing cancer. He dies of a heart attack, instead of his cancer. Since the two groups of men are essentially identical because they were randomized to be identical, all the men die at an average age of 74. The increase in overall survival for undergoing surgery is 0 percent. The cancer-specific survival for undergoing surgery is 100 percent as no man died of slow-growing cancer of the right leg. Now let’s run to the press and tout that we have 100 percent cancer-specific survival if we amputate everybody’s right leg for $8,000 a case, since we can make a lot of money.

In the Holmberg study, 695 men with localized prostate cancer were randomized: 348 to watchful waiting and 347 to radical prostatectomy.

Once statistics were run on the data from the study as in the table above, it was found that there was no statistically significant difference in overall survival between the radical prostatectomy and watchful waiting.

There were some small effects seen in the study, for example 15 fewer men in the radical prostatectomy group died of prostate cancer, and at the same time, 6 more men in the radical prostatectomy group died from other causes, for a net benefit in the radical prostatectomy group of 9 less deaths. Unfortunately for those wanting surgery to be successful, this was not a statistically significant finding – it could be easily explained by chance. That’s why the author’s concluded that: “…there was no significant difference between surgery and watchful waiting in terms of overall survival.” Even if it were true that 9 of 347 (2.59 percent) men benefited, could it ever be worth it? It took 6.2 years of living with the side effects of surgery to get this benefit, which is so small that statistics say it could simply be due to chance. Why not turn our attention to all the other, less harmful treatments for prostate cancer, which deserve their own randomized controlled studies. Should we do a harmful operation on 100 men so that 2.59 might live a little bit longer? Especially when the beneficial effect might all be due to chance? The answer is clearly no in my opinion. I will only recommend surgery, if and when an absolutely clear benefit is shown, and it is also shown that the benefit of surgery outweighs the harm of surgery. And if that scenario ever occurs, the benefit of surgery had better be far, far, greater than 2.59 percent!

------

© Bradley Hennenfent, MD November 24, 2004.
http://www.survivingprostatecancerwithoutsurgery.org