No one seems to be quite sure what to make of this, but a new study has suggested that men who are diagnosed with clinically detectable prostate cancer — but do not actually die from their prostate cancer — are at higher risk for all other cause mortality than men who show no signs or symptoms of prostate cancer.
Let’s see if we can make this absolutely clear … The study suggests that if you are diagnosed with prostate cancer based on evident symptoms of a prostate disorder (a positive DRE; other lower urinary tract symptoms; etc.) but you don’t actually die of your prostate cancer, then you are at higher risk of all-cause mortality than men who show no symptoms of prostate cancer (even if they are diagnosed with it), so long as they also do not die of their prostate cancer.
van Leeuwen et al. set out to assess the cause-specific mortality rate unrelated to prostate cancer itself in men with screening-related and clinically diagnosed prostate cancer, based on data from participants in the European Randomized Study of Screening for Prostate Cancer (ERSPC). All study participants were aged between 55 and 74 years of age. Participants were initially enrolled into the screening or non-screening arms of the study between 1993 and 2001, and all participants were followed through December 31, 2007.
To obtain the actual study group for this analysis, the investigators began by excluding from their database all patients who actually died of prostate cancer (as determined by the causes of death committee). This gave them a study group that included:
- 372 men who were diagnosed with prostate cancer as a consequence of screening (Group A)
- 1,488 men who were screened but found to be cancer-free
- 744 men who were at possible risk for prostate cancer based on PSA data but cancer-free on biopsy (Group B1)
- 744 men who were considered to be at no risk for prostate cancer based on their PSA data and swere never biopsied (Group B2)
- 221 men who were not screened but were diagnosed with prostate cancer after developing symptoms and subsequently given “usual care” (Group C)
- 884 men who were not screened and also not diagnosed with prostate cancer (Group D)
The 372 men in Group A were each matched to two men from Group B1 and two men from Group B2. The 221 men in Group C were matched to four men in Group D. All patient matching was carried out with respect to date of birth and the screening and/or diagnosis criteria described above
The subsequent data analysis revealed the following results:
- For men in Group A
- There was no statistically significant difference in overall mortality compared to men in Group B1 or Group B2.
- Relative risk (RR) for non-prostate cancer mortality among men in Group B1 was 1.26.
- RR for non-prostate cancer mortality among men in Group B2 was 1.13.
- For men in Group C
- Overall mortality was statistically significantly higher compared to men in Group D (RR 1.43).
- This difference was because of an increased risk of dying from neoplasms and from diseases of the circulatory or respiratory system among the men in Group C (RR 1.61).
- 27 percent of men in this group were treated with androgen deprivation therapy (ADT) at some point in time.
- Among the men in Groups A and C
- Just under 12 percent died of cancers other than prostate tumors.
- 5 percent died of heart disease or stroke.
- Among the men in Groups B1, B2, and D
- 7 percent died of cancers other than prostate tumors.
- Just over 3 percent died of heart disease or stroke.
The authors are careful to note that interpretation of their study results should recognize the relatively small sample size of the participants in this study (although the sample sizes are not available in the abstract of the paper).
The implications of this study have been discussed in additional detail in a Reuters commentary. In that commentary, Dr. Anthony D’Amico (who had no involvement in the study itself) notes the following:
- Nothing in the current study is sufficient to suggest changes to current clinical practice.
- It is certainly possible that the use of ADT among men in Group C might help to explain the increased risk rate of death from cardiovascular problems in this group — but there is no specific evidence for this.
- It is also possible that men with more aggressive forms of prostate cancer “may be genetically vulnerable to other cancers as well.”
- “This is a study of associations, and does not prove cause and effect. It’s really [just] hypothesis-generating.”
Filed under: Diagnosis, Living with Prostate Cancer, Risk Tagged: | all other cause, mortality, non-prostate-cancer
Search for new and
ongoing trials on the
CTAG PCa web site
Having had prostate cancer and had 37 radiation therapy sessions I thought all was well. Now I read this, and makes me feel a lot better (NOT!!!!!)
Dear Tony:
(1) Sooner or later we are all going to die! (2) If you had no actual signs or symptoms of prostate cancer prior to your treatment (i.e., your cancer was only found because of an elevated PSA level), this study shows your risk of all-other-cause mortality is now the same as that of a man who never had prostate cancer in the first place.
Why would this be a mystery? I expect that people in poor health are more likely to get cancer and treatment will degrade their health even further. Thus this group is more likely to die of any cause than the group that has not had these health-degrading events.
Dear John:
That is certainly one of several possible interpretations. Whether it can be justified by the available data is a completely different issue. You are making multiple assumptions.
This is no surprise to me at all. African American men have the highest rate of prostate cancer in the U.S., but they also have a terrible diet, consisting of too many fat products. So heart problems and adult-onset, type 2 diabetes are two factors that can cause death other than prostate cancer.
What amazes me is the lack of a good, in-depth study of all men, all races, here in the U.S. Diets, smoking, alcohol use, drug use, genetics, type of work, drinking water contamination (i.e., high levels of arsenic, etc.) are just some of the factors that should be gathered. Certainly now, in 2012, there must be a considerable amount of data that can be gathered, which could ultimately point doctors in a direction to find a cause and effect hypothesis. Small numbers in these studies are not statistically significant enough to warrant any eurekas (in my humble opinion)
Does this not really go back to the definition of death by prostate cancer? Already mentioned were possibly prostate cancer-related causes: cardiovascular problems due to ADT, development of other cancers due to prostate cancer, developing cancer because of poor health conditions.
There are, possibly, others:
– Development of diabetes for ADT patients
– Death associated with chemotherapy (I have a friend who died of “kidney failures” which were related to the chemotherapy he received)
– Increase of stress due to having cancer
– Increased suicide rate
– Simply giving up (“I am going to die because I have cancer”)
– Increased cholesterol levels; lack of exercise leading to heart problems.
I am sure there are others, and I am sure that you can create a similar list for all of the other cancers.
I am having a big problem with the narrow definition of “death by prostate cancer” when so many deaths that are not diagnosed as deaths from prostate cancer would not occur if there were not any cancer in the first place. Therefore, I have to question the stats on prostate cancer-specific and non-prostate cancer-specific mortalities as I feel that more deaths are attributable to prostate cancer.