One of the most complicated questions in dealing with prostate cancer is correctly assessing whether men who are known to have had prostate cancer actually died of their prostate cancer or simply with it, especially among men of more advanced age or with co-morbid conditions. In other words, “What do prostate cancer patients die of?” A new study having this precise title, and conducted by a group of German and Swedish researchers, has been able to throw a little more light on this question.
Riihimäki et al. used data from the Swedish Family Cancer Database to investigate the assigned causes of death of all men listed in the database who were born before 1957 and were living in Sweden between 1987 and 2006. This gave them a total study population of 2.27 men, of whom 117,000 had been diagnosed with prostate cancer as their initial invasive form of cancer. (The total number of individuals listed in the Swedish Family Cancer Database is about 11.7 million.)
The causes of death in the Swedish Family Cancer Database for the period 1987 to 2006 include a primary cause and up to 10 additional, multiple causes, all coded using the ICD-9 and ICD-10 versions of the International Classification of Diseases, so it was relatively easy to compare data across the database.
Here is what they were able to show:
There were a total of 686,500 observed deaths in the database in the 20-year period from 1987 to 2006.
62,500/686,500 deaths (9.1 percent) were among men with prostate cancer as their primary cancer diagnosis.
42,719/116,945 men with prostate cancer as their primary cancer diagnosis (36.5 percent) had prostate cancer listed as either the only or one of their multiple causes of death during this 20-year timeframe.
Another 8,316 men who were not diagnosed with prostate cancer as their primary cancer diagnosis also had prostate cancer listed as either the only or one of their multiple causes of death during this 20-year timeframe.
For the men who had prostate cancer listed as their primary cancer, the three most probable underlying (primary) causes of death other than prostate cancer compared to men in whom prostate cancer was not listed as a primary cancer were:
For the men who had prostate cancer listed as their primary cancer but had multiple possible listed causes of their death including prostate cancer, the three most probable causes of death other than prostate cancer compared to men in whom prostate cancer was not listed as a primary cancer were:
Now there are still all sorts of questions about whether causes of death were properly assigned in all cases, although other studies have shown that in Sweden assignment of prostate cancer as a primary cause of death is generally reliable. We also need to recognize that PSA testing as a means to identify risk of prostate cancer was not widely used in Sweden duing the 20-year timeframe of this study, which means that Swedish men being diagnosed with prostate cancer during this timeframe were commonly being diagnosed clinically, with more advanced forms of prostate cancer than would be customary in America (and probably Sweden) today.
What this study does make clear, however, is that:
We need to pay greater attention to the co-morbid conditions of men with a diagnosis of prostate cancer to make sure we are paying attention to all the problems that are likely to affect their mortality.
The non-use of PSA testing as a means to help identify men early in the course of their disease (while there is a high possibility for appropriate curative care) is associated with a high risk from prostate cancer-specific mortality (> 36.5 percent in this study).