Statin Use and Incident Prostate Cancer Risk
Statin Use and Incident Prostate Cancer Risk
In our study of 66 741 male subjects, statin use was associated with a decreased incidence of prostate cancer; the negative association was stronger with increasing total dose and longer periods of treatment, a reduction of over 70% was found in those treated for over 5 years from the start of the study (HR 0.26 95% CI 0.22–0.31). We observed both a time-related and a cumulative dose relationship between statin use and reduction in risk of prostate cancer. It suggests that this relationship is related to the biological activity of statins. This association is consistent with experimental data demonstrating statins' ability to kill prostate cancer cells.Lovastatin, fluvastatin and simvastatin have been shown to inhibit proliferation of prostate cancer cells via the downregulation of the insulin-like growth factor 1 receptor.
The limitations of our study include the lack of information for the use of aspirin and other NSAIDs, which may affect prostate cancer; these are non-prescription drugs and are therefore not all recorded on the health maintenance organization's database. Clearly, the population taking statins is very different from those not taking statins; although we have used models to try and adjust for some of these differences, we cannot rule out residual confounding. In previous studies, it has been suggested that statins reduce the risk mainly of higher-grade prostate cancer. In our study, information on the grade and severity of cancer at the time of diagnosis was unavailable. As population screening for prostate cancer has not been recommended in Clalit Health Services, it is possible that most of the patients had more advanced disease. This may explain why statin use was protective in our study, consistent with other studies of advanced disease.
Our results, indicating that statin use is associated with a reduction in incident prostate cancer, concur with the findings from several earlier observational studies; this effect was strengthened both with increasing duration of treatment and increasing cumulative dose. However, in our study this association was not consistent for all types of statin or even when grouping statins by hydrophilic or hydrophobic nature. It has previously been suggested that lipophilicity might influence the ability of various statins to suppress the growth of prostate cancers, as the increased lipid solubility of the lipophilic statins could improve their ability to pass through cell membranes. In our study, pravastatin showed no clear association with incident prostate cancer (HR 0.93, CI 0.72–1.16); this is in contrast to results reported in the long-term follow-up of the West of Scotland Coronary Prevention Study, the only randomized study to report prostate cancer rates specifically for pravastatin (HR 1.46, CI 1.05–2.02).
One previous epidemiological study may not have been powered to find any association between statin use and incident prostate cancer, with only 78 cases of prostate cancer in the follow-up period. In most other epidemiological studies to date, analysis was not performed for individual statin groups; if the effects of statins on incident prostate cancer varies between statin type, variations in statin usage may explain why some studies have not found an overall association between statin use and incident prostate cancer, although others have found a protective effect.Similar inconsistencies between statin type and association with prostate cancer were demonstrated in the Finnish prostate-screening trial. The overall relative risk of prostate cancer was lower among current statin users HR 0.62, 95% CI 0.53–0.72; however, a dose-dependent decreasing trend in the risk was observed only for atorvastatin, simvastatin and fluvastatin. Although we saw no reduction in incident prostate cancer among fluvastatin users, this may be because of the small number of fluvastatin users in our cohort.
It would appear that the mechanism of action in reducing cancer-cell proliferation and apoptosis is not only dependent on the statins' ability to reduce cholesterol, as this would cause a fairly uniform reduction in cancer incidence independent of the statin type. It appears likely that a number of mechanisms are working simultaneously and that not all types of statin trigger these different mechanisms.
The results of our large population-based study suggest that the use of statins reduces the risk of prostate cancer in a dose-related and time-dependent manner, although this appears not to be equally true for all types of statins. It would appear that inconsistencies between studies to date may be partly explained by differing the use of statin types. Further studies on the mechanisms of the preventive effect of statins on prostate cancer are required to explain these apparent inconsistencies.
Discussion
In our study of 66 741 male subjects, statin use was associated with a decreased incidence of prostate cancer; the negative association was stronger with increasing total dose and longer periods of treatment, a reduction of over 70% was found in those treated for over 5 years from the start of the study (HR 0.26 95% CI 0.22–0.31). We observed both a time-related and a cumulative dose relationship between statin use and reduction in risk of prostate cancer. It suggests that this relationship is related to the biological activity of statins. This association is consistent with experimental data demonstrating statins' ability to kill prostate cancer cells.Lovastatin, fluvastatin and simvastatin have been shown to inhibit proliferation of prostate cancer cells via the downregulation of the insulin-like growth factor 1 receptor.
The limitations of our study include the lack of information for the use of aspirin and other NSAIDs, which may affect prostate cancer; these are non-prescription drugs and are therefore not all recorded on the health maintenance organization's database. Clearly, the population taking statins is very different from those not taking statins; although we have used models to try and adjust for some of these differences, we cannot rule out residual confounding. In previous studies, it has been suggested that statins reduce the risk mainly of higher-grade prostate cancer. In our study, information on the grade and severity of cancer at the time of diagnosis was unavailable. As population screening for prostate cancer has not been recommended in Clalit Health Services, it is possible that most of the patients had more advanced disease. This may explain why statin use was protective in our study, consistent with other studies of advanced disease.
Our results, indicating that statin use is associated with a reduction in incident prostate cancer, concur with the findings from several earlier observational studies; this effect was strengthened both with increasing duration of treatment and increasing cumulative dose. However, in our study this association was not consistent for all types of statin or even when grouping statins by hydrophilic or hydrophobic nature. It has previously been suggested that lipophilicity might influence the ability of various statins to suppress the growth of prostate cancers, as the increased lipid solubility of the lipophilic statins could improve their ability to pass through cell membranes. In our study, pravastatin showed no clear association with incident prostate cancer (HR 0.93, CI 0.72–1.16); this is in contrast to results reported in the long-term follow-up of the West of Scotland Coronary Prevention Study, the only randomized study to report prostate cancer rates specifically for pravastatin (HR 1.46, CI 1.05–2.02).
One previous epidemiological study may not have been powered to find any association between statin use and incident prostate cancer, with only 78 cases of prostate cancer in the follow-up period. In most other epidemiological studies to date, analysis was not performed for individual statin groups; if the effects of statins on incident prostate cancer varies between statin type, variations in statin usage may explain why some studies have not found an overall association between statin use and incident prostate cancer, although others have found a protective effect.Similar inconsistencies between statin type and association with prostate cancer were demonstrated in the Finnish prostate-screening trial. The overall relative risk of prostate cancer was lower among current statin users HR 0.62, 95% CI 0.53–0.72; however, a dose-dependent decreasing trend in the risk was observed only for atorvastatin, simvastatin and fluvastatin. Although we saw no reduction in incident prostate cancer among fluvastatin users, this may be because of the small number of fluvastatin users in our cohort.
It would appear that the mechanism of action in reducing cancer-cell proliferation and apoptosis is not only dependent on the statins' ability to reduce cholesterol, as this would cause a fairly uniform reduction in cancer incidence independent of the statin type. It appears likely that a number of mechanisms are working simultaneously and that not all types of statin trigger these different mechanisms.
The results of our large population-based study suggest that the use of statins reduces the risk of prostate cancer in a dose-related and time-dependent manner, although this appears not to be equally true for all types of statins. It would appear that inconsistencies between studies to date may be partly explained by differing the use of statin types. Further studies on the mechanisms of the preventive effect of statins on prostate cancer are required to explain these apparent inconsistencies.
Source...