Prostate Cancer increase life expectancy?

Interesting article relating to German researchers and highlighted on 'The "New" Prostate Cancer InfoLink' website:

Could having a diagnosis of localized prostate cancer increase life expectancy?

Posted on November 5, 2010

In what can only be described as an unexpected finding, a group of German researchers have documented a greater life expectancy for men diagnosed with early stage prostate cancer than for the male population in general, based on data from the Regensburg cancer registry!
Klotz et al. used epidemiological data from the Regensburg cancer registry to study the survival of patients diagnosed with organ-confined prostate cancer (T1-2N0M0) compared to the standardized age-adjusted survival of the normal male population.

Their analysis is based on data from 4,124 patients with prostate cancer diagnosed between 1998 and 2007, of whom 2,087 men had been diagnosed with localized disease. Here is what they claim to have been able to find:

•At 8 years after diagnosis, patients with T1-2N0Mo prostate cancer had a roughly 10 percent relative increase in survival compared with the normal male population.
•This relative increase in survival could be documented just 3 years after diagnosis.
What could possibly account for such a finding?

It can’t be explained by prostate cancer treatment, because any survival benefit from treatment takes at least 8 to 10 years and maybe longer (based on data from the ERSPC trial). The immediate implication is that (at least in Regensburg) men who have PSA tests (and therefore greater likelihood for a diagnosis of prostate cancer) may have better overall health status than men who do not.

This is not an entirely unreasonable suggestion. There is a well-understood phenomenon that associates positive health-related (“health-seeking”) behaviors with all sorts of other factors. Elsewhere on this site, we discuss the fact that although men who have vasectomies seem to be at greater risk for prostate cancer than men who do not have vasectomies, this is not because they have the vasectomies. Rather, it is because they exhibit such “health-seeking” behavior.

We suspect that in Regensburg (and potentially in many other places) having a PSA test is closely correlated with having regular health checks. Men who have regular health checks arguably may have a superior life expectancy to those who see doctor only when they absolutely have to (although such a benefit has never actually been documented, as far as we know). The other issue that then comes into play is whether men of higher income and/or better education are more likely to have regular health checks than those less fortunate. The authors refer to this in their paper as the relationship between PSA testing and a “socal gradient.”

Could having a diagnosis of localized prostate cancer increase life expectancy?