A commonly raised question is whether men with benign prostatic hyperplasia (BPH) are at greater risk for prostate cancer than men who have no sign of BPH. A new, nation-wide study of the male population of Denmark over the past 27 years offers us some insights.
First, we need to be clear that there is no definitive evidence whatsoever that BPH is a precursor to or a pre-malignant biologic marker for prostate cancer. What Orsted et al. set out to study was only whether the presence of clinically evident BPH was associated, in Danish men, with an increase in risk for a diagnosis of prostate cancer and/or prostate cancer-specific mortality. This type of study is possible in Denmark (as in some other Nordic nations) because of the sophisticated, population-based databases that have been constructed there over the past 30 years.
Osted et al. were able to evaluate data from the entire Danish male population from 1980 through 2006. Their study database included:
- 3,009,258 Danish men
- 53,315 cases of prostate cancer diagnosis (histologically proven)
- 25,459 cases of prostate cancer-specific mortality
- 187,591 cases of clinically diagnosed (but not histologically proven) BPH among hospitalized patients (between 1980 and 2006)
- 77,698 cases of surgical treatment for clinically diagnosed BPH (between 1980 and 2006)
- 143,365 cases of treatment with α-adrenergic receptor antagonists like tamulosin (between 1995 and 2006)
- 47,465 cases of treatment with 5α-reductase inhibitors (5-ARIs) like dutasteride (between 1995 and 2006)
The authors were able to use these data to derive the following:
- Among all men hospitalized for BPH (compared to general population controls)
- The multivariate, adjusted hazard ratio (HR) for incidence of prostate cancer was 2.22.
- The multivariate, adjusted HR for prostate cancer-specific mortality was 2.00.
- Among all men treated surgically for BPH (again compared to general population controls)
- The multivariate, adjusted hazard ratio (HR) for incidence of prostate cancer was 3.26.
- The multivariate, adjusted HR for prostate cancer-specific mortality was 7.85.
- For age-matched groups of men (compared to age-matched population controls)
- The multivariate, adjusted HR for incidence of prostate cancer among men hospitalized for BPH was 3.04.
- The multivariate, adjusted HR for incidence of prostate cancer among men treated surgically for BPH was 2.60.
- The multivariate, adjusted HR for incidence of prostate cancer among men treated with α-adrenergic receptor antagonists was 4.49.
- The multivariate, adjusted HR for incidence of prostate cancer among men treated with 5-ARIs was 2.54.
In commenting on these data, the authors make the following key points:
- Although each of the clinical categories of BPH defined in this study has limitations, these limitations differ from category to category. It is therefore unlikely that the limitations explain the results.
- Among this national cohort of > 3 million Danish males, followed for up to 27 years, clinical evidence of BPH is associated with
- A two- to threefold increase in risk for a diagnosis of prostate cancer
- A two- to eightfold increase in risk for prostate cancer-specific mortality
- These data can not be used to infer that BPH is a cause of prostate cancer.
This final point made by the authors is extremely important. What they are effectively saying is that BPH and prostate cancer well may have a common causative factor (possibly associated with levels of testosterone and/or other factors in prostate cells over time) and that therefore it is not surprising to find that men with BPH are at greater risk for prostate cancer than men who have no sign of BPH. However, there is still no evidence at all to suggest that having BPH is in itself a cause of prostate cancer or that before you get prostate cancer you are bound to have some degree of BPH.
What this study does show us, however, at a national level, is that (at least in Denmark) men with clinical symptoms of BPH are at increased risk of also having a diagnosis of prostate cancer and potentially dying from prostate cancer over time compared to the BPH-free male population. It should be recognized, however, that since these data go back to 1980 (long before the advent of the PSA era) the risk of prostate cancer-specific mortality associated with a finding of BPH for a man diagnosed with BPH today is likely to be significantly lower than the risk calculatred in this study. On the other hand, this study does suggest that any man with a diagnosis of BPH is at significantly increased risk for a diagnosis of prostate cancer, and therefore needs to be carefully monitored with this possibility in mind.