How preconceptions impact prostate cancer management
Posted Mar 20 2010 12:00am
As Leah wrote in a comment on this blog just the other day, “Yes, the ‘D-word’ is still taboo. Death is UNAMERICAN.” The comment stimulated the thought that our preconceptions have significant impact on how we think about risk, diagnosis, and treatment of prostate cancer and yet there is really very little good research on this topic.
Anyone in America over the age of about 45 was brought up with the idea that “cancer” is a deadly disease. We are all “guilty” of lumping all cancers together in this way because back in the early 1960s and before (when we were born) most cancers still were deadly diseases. The American Society for Clinical Oncology wasn’t even founded until 1964. Now it has 27,000 members very nearly one for each American man that dies of prostate cancer each year.
Some weeks ago, we commented on an article by Beydoun et al. This article had attempted to identify the factual knowledge of prostate cancer patients at the time of diagnosis. The article has flaws, and the authors acknowledge this, but it is still an interesting article because it does several things:
It highlights the fact what what a patient actually knows or thinks he knows at the time he makes decisions about his prostate cancer management can profoundly impact his decision process (preconceptions impact management).
It gives some perspective on the opportunity to provide better information for primary care physicians, which would allow these traditional “family doctors” to take a stronger and better defined advisory role for patients newly diagnosed with prostate cancer.
It emphasizes a whole bunch of other things that we really don’t have good insights into about how our preconceptions affect the decisions we make about treatment for prostate cancer and other disorders too.
What we did not know until we had talked to one of the authors of the paper by Beydoun et al. is that this group has been investigating other aspects of what drives patient decision making in prostate cancer.
Specifically, they have two other published papers on this general topic to date. The first, by Mohan et al. , was published in March 2009, and deals with how the expectations of newly diagnosed patients with localized prostate cancer about their long-term survival with and without invasive treatment impact their decisions about that treatment.
Based on data from a group of 184 newly diagnosed patients with localized prostate cancer, the paper demonstrates that most of the patients ”grossly underestimated their life expectancy without treatment and grossly overestimated the survival benefit of treatment.” Clearly, if you start out by thinking that a diagnosis of localized prostate cancer is sure to lead to your death within 10 years (as believed by 64.0 percent of the participants in this study), then you are likely to decide you need treatment. And if you also believe that with treatment you will live for ≥ 20 years (i.e., treatment will double your life expectancy, as believed by 62.9 percent of the patients in this study), then immediate treatment becomes a near-certainty.
Now we do know with some degree of certainty that the vast majority of men diagnosed today with localized prostate cancer choose to be treated. Indeed, a study by Harlan et al. , published in 2004 and based on data from the CaPSURE initiative, suggested that 90 percent of men diagnosed with prostate cancer had an initial diagnosis of localized disease and that 94 percent of these patients elected to have some form of treatment. (Although we suspect that active surveillance may have increased in popularity since that time.)
By contrast, we do not actually know whether active early treatment of these men affects their survival at all. We assume it does; we hope it does; it “looks as though” it does; but we don’t know that it does. The randomized Scandinavian trial which compared radical prostatectomy to watchful waiting showed that at 10.8 years of follow-up the overall mortality for the surgery group was 39.5 percent compared to 44.8 percent for the watchful waiting group – but these were men who were diagnosed with clinically evident disease, and were therefore significantly more advanced at the time of treatment than the average man diagnosed with localized disease today. The difference is statistically significant but it is still small.
So where is the “man in the street” getting the idea that his overall survival will double if he is treated for early stage, localized prostate cancer? Hmmm … interesting question. The media? His doctors? The advocacy community? Or it is just some sort of societal assumption? We will need to wait for the results of the ProtecT trial in the UK before we can get any really good data to inform us whether early treatment actually impacts survival in a cohort of men diagnosed through early detection. The ProtecT study has been enrolling patients since 2001 and is still enrolling patients today. Exactly when it might be able to provide us with any meaningful data is a little hard to estimate.
So what should a physician tell a patient who is newly diagnosed with localized disease today? Hopefully, a good physician will tell that patient that his long-term outcome will depend on a multiplicity of individual factors: his age, his life expectancy, his PSA level, perhaps his PSA kinetics, his clinical stage, the amount of cancer in his prostate, his Gleason score, how he is treated (or not treated), his comorbidities, and other factors that may affect his health. Hopefully he will also tell him that his risk of actually dying from prostate cancer is probably no higher than about 3 percent (but we wouldn’t bet on it in most cases).
And exactly how much of this do we realistically think the newly diagnosed “man in the street” is taking in? What he has just heard is, “You have cancer.” The rest of the information went straight in one ear and straight out the other. Remember this guy was born in 1964. In his head, “Cancer kills people.” And we can’t talk about that because “Death is UNAMERICAN.” It is so unamerican that we can’t even talk about it not being true for 97 percent of the men diagnosed with prostate cancer today!
In their second paper , that addresses some more of these issues, published in February this year, Mohan et al. investigated whether men’s perceptions of their overall life expectancy had impact on their choice of treatment as presented according to a well-known set of national guidelines.
The authors provided the same 184 survey participants with a simple method to estimate their life expectancy and then the participants were asked to choose how they wanted to be treated based on the then-current NCCN guideline, inclusive of guidance based on their life expectancy. (It is worth noting that all these patients had been diagnosed between 2005 and 2007 at a single clinical practice, had met with their urologist to discuss their treatment, but had not actually been treated. A total of 356 patients were invited to participate in the survey; 184/356 or 51.7 percent agreed to do so.)
Among 88 patients at low risk for recurrence, 28/41 (68.3 percent) who met criteria for some form of expectant management elected active treatment (“over-treatment”) but only 4/47 patients (8.5 percent) who met criteria for active treatment elected expectant management (“under-treatment”). In total 58/88 patients (65.9 percent) at low risk for recurrence actually elected a form of management concordant with national guidelines. Among the 96 patients who were at intermediate or high risk for recurrence, 90/96 patients (93.8 percent) elected a form of management concordant with national guidelines.
Based on this information, it would certainly appear that, given a sound method to estimate their reasonable life expectancy, most patients (144/184 or 78.3 percent) are willing to follow national guidelines for management of localized prostate cancer. It will be interesting to see if this continues to hold true now that the recently revised NCCN guidelines recommend active surveillance for a much higher proportion of patients diagnosed with localized prostate cancer.
The research than Mohan and his colleagues have been conducting in our eyes represents a pilot project for a much more systematic opportunity to assess patient knowledge and related decision making. One of the constant complaints in the prostate cancer advocacy community is that patients get “rushed” into treatment. The model being offered by Mohan and his colleagues (albeit in need of refinement) appears to offer a first step toward a better way to help patients to come to good decisions about the form of management that is in their best interests. It would certainly help if we could improve their available knowledge by being able to give them definitive information about the survival benefits of early, localized therapy but as yet we just don’t have these data. And it would also help if we had tests that could differentiate between indolent and clinically significant forms of prostate cancer. Hopefully those aren’t too far away either.