Do statins reduce cancer-related mortality?
Published in NPS Direct
Date published: About this date
A study investigating the association between statin use and cancer-related mortality has been conducted in patients over the age of 40 diagnosed with cancer between 1995 and 2007, drawn from a network of national databases in Denmark. The incidence of death from cancer among statin users was reduced by an average of 15% compared with patients who had never used statins but the study had some limitations. It was a retrospective observational cohort study, more than 70% of patients were excluded from some analyses due to missing baseline data and there may have been a “healthy user” bias among patients on statins. Future prospective randomised controlled trials are required to evaluate the hypothesis that statin use starting before a diagnosis of cancer is associated with reduced cancer-related mortality.
- Both a full cohort analysis and a nested 1:3 matched analysis showed that regular statin use was associated with reduced death from cancer-related causes and all-cause mortality by around 15% in an uncontrolled retrospective observational cohort study (people over 40 later diagnosed with cancer).
- The study has several limitations, including large numbers of patients with missing data regarding tumour size, rates of metastasis and incidence of radiotherapy and chemotherapy treatments for both the statin-user and non-statin-user groups, impacting on the mortality analyses.
- Incidence of smoking was not taken into account.
- The study population was predominantly Caucasian and of Danish descent, which may limit the application of the results to the wider population.
- There was no correlation between statin dose and mortality reduction.
- Further trials are required to determine whether statins reduce cancer-related mortality.
An observational cohort study suggests benefits of statins in prolonging survival in people with cancer
A recent Danish retrospective observational cohort study using data from the Danish Civil Registration System, the Danish Cancer Register and the Danish Register of Medicinal Products Statistics investigated whether the use of statins before and after a diagnosis of cancer was associated with reduced cancer-related mortality when compared with outcomes in patients with no statin use before their cancer diagnosis.1
This nationwide cohort study was based on patient records of 295,925 people aged ≥ 40 years diagnosed with cancer. People who regularly used statins (N = 18,721) within 2 years before their cancer diagnosis were significantly less likely to die of cancer (hazard ratio [HR] 0.85, 95% confidence interval [CI] 0.82 to 0.87) or of all-cause mortality (HR 0.85, 95% CI 0.83 to 0.87) than people who had never used statins (N = 277,204).1 People diagnosed with cancer at < 40 years of age were excluded from the study, as they would have been unlikely to have received statins.1 The population in this nationwide cohort study was not matched for age, physical characteristics or comorbidities.1 The observed reduction in mortality was about the same as previously observed for reduction in mortality from cardiovascular disease attributed to statins.2
The authors also performed a nested 1:3 matched study in which each statin user (N = 15,247) was matched with three people who had never used statins (N = 45,741) for factors associated with an increased risk of death from cancer, including sex, age at cancer diagnosis, year of diagnosis and cancer type. People with diabetes or cardiovascular disease were excluded to improve matching of patient characteristics. Cancer-related and all-cause mortality were also significantly reduced in the nested 1:3 matched analysis (reported as a sub-analysis for different daily doses of a statin) (p > 0.001).1
No dose–response relationship was found for statin dose and cancer-related mortality or all-cause mortality.
Death from cardiovascular causes not reduced
In both the nationwide cohort and the nested 1:3 matched studies death from cardiovascular causes was not significantly reduced by statin use, and 0.76–1.5 mg daily doses significantly worsened mortality outcomes.1 When data from the nested 1:3 matched study were adjusted using propensity scoring to attempt to adjust for differences in the medical history of people with cancer, this significant worsening was no longer apparent, highlighting the effect of uncontrolled variables.
Effects of statins may be cancer-type specific
Pre-specified subgroup analysis of the data from both the nationwide cohort study and the nested 1:3 matched study included 27 different cancer types and 9 patient characteristics. Of the 27 different cancer types investigated, statin use was shown to reduce mortality associated with 13 types, including lung, colon, prostate and breast cancer.1 The study also used data recorded in the Danish Cancer Registry over the period 1995–2003 related to treatment by radiotherapy and chemotherapy initiated within 4 months of diagnosis.1 No significant difference in cancer-related deaths were shown in either study for people who underwent chemotherapy treatment (cohort p = 0.34, nested 1:3 matched study p = 0.91).1
A retrospective review of a large number of patient records, but with limitations
This study was not a randomised controlled trial but a retrospective review of a large number of comprehensive patient records, which reduces the power of the evidence. Selection bias was avoided by the inclusion of all eligible persons from the Danish population and there was no loss to follow-up.1
The authors did not indicate the incidence of various cancers in the Danish population or take into account whether any specific cancers were surgically removed.
Some cancer-related baseline patient characteristic data were missing in more than 70% of patients. Characteristics for which data were missing included tumour size, lymphatic spread of cancer, metastases and treatments such as chemotherapy or radiotherapy. The potential for bias as a result of excluding these patients’ data was partly overcome by the nested 1:3 matched study.
The study cohort was predominantly Caucasian and of Danish descent, which limits the generalisability of these finding to the wider population.1
Did statin use introduce a ‘healthy user’ bias in this study?
The results of the study could have been biased if statin users had ‘increased health awareness’, reflected in more medication usage, which may have improved cancer outcomes, or increased cancer screening leading to earlier diagnosis. Adjustment for tumour size in the analysis was introduced in an attempt to correct for this latter bias.
The frequency of smoking was also not taken into account. Since it is possible that people taking statins to reduce cardiovascular risk may also have reduced their smoking frequency, this may have biased the results with respect to mortality rates from smoking-related cancers.
In addition, to adjust for differences in the medical history of patients with cancer in order to avoid a ‘healthy-user’ bias, the nested 1:3 matched analysis was repeated with propensity-score matching, and the results were similar to those of the full cohort analysis. The nested 1:3 propensity-score-matched study also showed a reduction in all-cause mortality and death from cancer-related causes (reported as sub-analyses for different daily doses of statin); however, statin use was no longer associated with increased cardiovascular mortality. Similar to the previous analysis, a dose-dependent relationship between statin used and reduced cancer-related mortality was not shown.1
This large retrospective observational study in a predominantly Caucasian Danish population showed that people who regularly took statins before and after they were diagnosed with cancer had reduced all-cause mortality and cancer-related mortality.1 The extent to which these results reflect a potentially ‘true’ statin effect in reducing cancer-related mortality remains uncertain. Confounding factors not addressed in the study, such as general health and lifestyle factors (including smoking and alcohol consumption) may have differentiated the people taking statins from those who did not, and the influence of these factors on mortality requires further investigation.
Since the effect of statin use on cancer-related mortality may influence clinical decisions about prescribing statins, especially in people diagnosed with cancer, prospective controlled clinical trials are required to further investigate this potential effect.
- Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancer-related mortality. N Engl J Med 2012;367:1792–802. [PubMed]
- Baigent C, Blackwell L, Emberson J, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010;376:1670–81. [PubMed]