Is the Evidence Sufficient to Recommend Statins for All Men with Prostate Cancer?

In the United States, an estimated 19 million men receive the class of lipid-lowering medications known as statins.1 These drugs have been shown in randomized trials to lower risk of and mortality from cardiovascular disease2,3, likely via the inhibition of the key cholesterol biosynthesis enzyme, 3-hydroxy-3-methylglutaryl-coenzyme A reductase. In contrast to the clear benefits for cardiovascular health, early experimental studies in rodent models raised public health concerns that statins could initiate or promote cancer in humans.4

This concern was soon debunked by results from epidemiologic studies of statins and the risk of cancer, including prostate cancer. Most studies of prostate cancer have focused on risk of cancer development, finding either no association with statins5,6 or a small inverse association7; however, studies of total prostate cancer risk may have missed important findings, given the marked heterogeneity in the disease’s biologic potential and the impact of prostate-specific antigen (PSA) screening on the natural history of prostate cancer in epidemiologic studies.8 In 2006, a landmark epidemiologic study focused on the risk of advanced and fatal prostate cancer, and demonstrated a strong beneficial effect of statins on these clinical prostate cancer end points that were indicative of an aggressive phenotype.9

Epidemiologic studies are quite consistent with regard to a beneficial role of statin use after diagnosis across the continuum of prostate cancer.9-13 In a meta-analysis of six studies, the use of statins was associated with a 24% lower risk of cancer mortality among patients with prostate cancer (relative risk, 0.76; 95% CI, 0.64 to 0.89).14 A subsequent study by Yu et al11 of 11,000 men with nonmetastatic prostate cancer supported the findings of the meta-analysis and made the important observation that the effect of postdiagnostic statin use on prostate cancer mortality was stronger among men who had been taking a statin before diagnosis (hazard ratio [HR], 0.55; 95% CI, 0.41 to 0.74) compared with those who were only taking statins postdiagnosis (HR, 0.82; 95% CI, 0.71 to 0.96), which suggests that the duration and/or timing of exposure might be important. In the Health Professionals Follow-up Study, there was a strong protective effect of postdiagnostic statin use and risk of lethal prostate cancer among men with clinical stage T2 to T3 disease (HR, 0.65; 95% CI, 0.43 to 0.97), but no association among those who were diagnosed with lower-risk stage T1 cancers.15 Finally, in studies of men with advanced prostate cancer, those who were taking a statin at the time of the initiation of androgen-deprivation therapy had a longer time to progression compared with nonusers of statins.10,16

In the article that accompanies this editorial, Larsen et al17 provides additional evidence on the benefits of statins among men with prostate cancer. Their study included 31,790 men with prostate cancer, of whom 7,365 died of the disease, and reported an HR for postdiagnostic statin use and prostate cancer mortality of 0.83 (95% CI, 0.77 to 0.89). The association between statin use and prostate cancer mortality was similar among men who were diagnosed with either localized or advanced prostate cancer, and results were in line with earlier observations that found reduced progression among men who underwent androgen-deprivation therapy10. The strength of the study lies in its considerable statistical power, which is afforded by the use of the nationwide Danish health care registries. The large sample size also provided the opportunity to explore key subgroup analyses by disease characteristics with ample statistical power. One important and somewhat surprising feature of the study population was the high rate of prostate cancer mortality, a result, perhaps, of a lower prevalence of PSA screening, but also possibly because of the types of treatment used (not described). Nonetheless, the high mortality allowed the authors to assess the clinical utility of statins in a more advanced disease setting.

Many biologic mechanisms that underlie this association have been postulated.18 Statins can act via cholesterol-, geranylgeranyl-, or farnesyl-mediated signaling pathways in prostate tumors. Cholesterol is a precursor of androgens, and, as such, statins can act by reducing androgen bioavailability, thereby limiting tumor growth. Cholesterol accumulates in prostatic lipid rafts that regulate the activation of signaling pathways, including phosphatidylinositol 3-kinase/Akt. There is suggestive evidence that cholesterol-independent pathways that mediate statin action in prostate cancer may also be operative.19 Androgen precursors use the solute carrier organic anionic transporters for uptake into prostate cells. Statins similarly use these transporters and interfere with androgen precursor cellular uptake by blocking solute carrier organic anionic transporters.10 Given the multiplicity of possible mechanisms by which statins might work in this context, it would be of clinical interest to know whether nonstatin lipid-lowering drugs have the same effect as statins on prostate cancer mortality.

Taken together, the data from Larsen et al17 and other epidemiologic studies point toward a substantial salutary effect associated with statins, with HRs comparable to many of the more toxic and more expensive agents that are now used for advanced prostate cancer. Whereas epidemiologic studies represent a powerful approach to address clinically relevant questions, there is a risk that systematic error across studies may explain the observed associations. Thus, the key next question is whether we believe these findings on postdiagnostic statins and prostate cancer mortality to be causal, or if chance, bias, or confounding could underlie the inverse association.

Given the consistency of the association across multiple populations, chance as an explanation of the findings can likely be ruled out. Information on statin use has been collected in these studies from multiple data sources, including prescription registers and self-report, written questionnaires. Whereas both sources of data may have their limitations, any misclassification of this exposure would tend to bias the results to the null; therefore, misclassification is unlikely to be a major concern. An alternate explanation is that the beneficial effect is not statins, but some unmeasured factor that is correlated with statin use and is itself inversely associated with prostate cancer mortality—that is, confounding. Several epidemiologic studies have adjusted for multiple lifestyle factors, medications, and clinical factors in multivariable models, and, of interest, these have largely had only minor effects on relative risk estimates. Given the size of the effect estimates for statin use and prostate cancer mortality, the relationship between the unmeasured confounder with both statin use and, independently, prostate cancer mortality would have to be large to explain the observed associations. As such, it is challenging to imagine what unmeasured confounder(s) could explain the protective effect of statins.

If we believe the finding is causal, the next question is whether there is sufficient evidence to recommend prescribing a statin for men who are diagnosed with prostate cancer and, if not, what additional evidence is needed to translate the findings clinically. At the least, we believe that clinicians can assure patients who are taking statins that they can continue their use during prostate cancer care, as there is no evidence of harm reported by epidemiologic studies.

Despite the weight of evidence in favor of postdiagnosis statin use, we believe it is timely to perform a prospective proof-of-concept trial before broadly prescribing statin use among men with prostate cancer. Such a trial, of course, is challenging in light of the high prevalence of statin use in the population, the long natural history of prostate cancer and challenges with using prostate cancer mortality as an end point, and the rapidly evolving landscape of treatment in advanced disease.

A key consideration is whether postdiagnostic statin use among men who were not previously on a statin will be beneficial. Here, the evidence seems to be mixed, with the study by Yu et al11 suggesting a weaker effect of postdiagnostic statin use among men who were not previously taking a statin, whereas Larsen et al17 state in their discussion that there was no difference in the effect of postdiagnostic statin use on the basis of prediagnostic use. If the former is true, a prospective trial that is undertaken among men who are not taking a statin before diagnosis may not demonstrate much benefit.

Another critical issue is on what disease state a prospective trial should focus. For example, an active surveillance population is an appealing target group for secondary prevention studies, where statins could have a potentially beneficial effect on prostate cancer–specific and cardiovascular outcomes; however, the Health Professionals Follow-up Study15 found that among men with low-risk prostate cancer on the basis of stage (T1), there was no association between postdiagnostic statin use and cancer mortality. Although the study by Larsen et al17 did examine associations stratified by treatment type, it is unclear if their group of other/no initial therapy reflects an active surveillance population.

We suggest that one possible trial of postdiagnostic statins could target men with nonmetastatic prostate cancer (M0) who have high-risk features, such as a rapid PSA doubling time. Men would be randomly assigned in a 1:1 fashion to androgen-deprivation therapy with or without a statin. The primary end point for this study would be metastasis-free survival, which has been demonstrated by the ICECaP (Intermediate Clinical Endpoint for Cancer of the Prostate) study as a robust surrogate for prostate cancer mortality20. Ideally, a racially diverse population would be recruited, given the marked disparities in prostate cancer mortality in African-American men, as well as the potentially lower prevalence of statin use in this group.1,21 Finally, we would encourage prospective trials to include in their protocols a prospective biorepository that could be leveraged to identify biologic markers—inherited genetic variants and tumor mutations—that may enhance the likelihood that a patient will benefit from statins.

Journal of Clinical Oncology  DOI: 10.1200/JCO.2017.74.7915 [Epub ahead of print]

Lorelei A. Mucci, MPH, ScD  Philip W. Kantoff, MD

Harvard T.H. Chan School of Public Health, Harvard Medical School, Dana-Farber Cancer Institute, Sanofi, Janssen Pharmaceuticals
Bellicum Pharmaceuticals, Placon, Druggablity Technologies, Tarveda Therapeutics, Bavarian Nordic, Janssen Pharmaceuticals, Astellas Pharma, Bellicum Pharmaceuticals, BIND Biosciences, Metamark Genetics, Merck, MTG Therapeutics, OncoCellMDX, Oncogenex, Sanofi, Bayer, Genentech, Ipsen, Omnitura, MorphoSys, GTx, Tarveda Therapeutics, Druggablity Technologies, Progenity, Thermo Fisher Scientific Life Sciences, Medivation (Inst), Sanofi (Inst), Oncogenex (Inst), Aragon Pharmaceuticals (Inst), Amgen (Inst), Astellas Pharma (Inst), Bayer (Inst), Bavarian Nordic (Inst), Dendreon (Inst), Exelixis (Inst), Janssen Pharmaceuticals (Inst)


1. Adedinsewo D, Taka N, Agasthi P, et al: Prevalence and factors associated with statin use among a nationally representative sample of US adults: National Health and Nutrition Examination Survey, 2011-2012. Clin Cardiol 39:491-496, 2016
2. Tonelli M, Lloyd A, Clement F, et al: Efficacy of statins for primary prevention in people at low cardiovascular risk: A meta-analysis. CMAJ 183:E1189-E1202, 2011
3. Chou R, Dana T, Blazina I, et al: Statins for prevention of cardiovascular disease in adults: Evidence report and systematic review for the US Preventive Services Task Force. JAMA 316:2008-2024, 2016
4. Newman TB, Hulley SB: Carcinogenicity of lipid-lowering drugs. JAMA 275:55-60, 1996
5. Chan JM, Litwack-Harrison S, Bauer SR, et al: Statin use and risk of prostate cancer in the prospective Osteoporotic Fractures in Men (MrOS) Study. Cancer Epidemiol Biomarkers Prev 21:1886-1888, 2012  
6. Jacobs EJ, Newton CC, Thun MJ, et al: Long-term use of cholesterol-lowering drugs and cancer incidence in a large United States cohort. Cancer Res 71:1763-1771, 2011
7. Farwell WR, D’Avolio LW, Scranton RE, et al: Statins and prostate cancer diagnosis and grade in a veterans population. J Natl Cancer Inst 103:885-892, 2011
8. Wilson KM, Giovannucci EL, Mucci LA: Lifestyle and dietary factors in the prevention of lethal prostate cancer. Asian J Androl 14:365-374, 2012  
9. Platz EA, Leitzmann MF, Visvanathan K, et al: Statin drugs and risk of advanced prostate cancer. J Natl Cancer Inst 98:1819-1825, 2006
10. Harshman LC, Wang X, Nakabayashi M, et al: Statin use at the time of initiation of androgen deprivation therapy and time to progression in patients with hormone-sensitive prostate cancer. JAMA Oncol 1:495-504, 2015
11. Yu O, Eberg M, Benayoun S, et al: Use of statins and the risk of death in patients with prostate cancer. J Clin Oncol 32:5-11, 2014  
12. Stopsack KH, Greenberg AJ, Mucci LA: Common medications and prostate cancer mortality: A review. World J Urol 35:875-882, 2017
13. Mucci LA, Stampfer MJ: Mounting evidence for prediagnostic use of statins in reducing risk of lethal prostate cancer. J Clin Oncol 32:1-2, 2014
14. Raval AD, Thakker D, Negi H, et al: Association between statins and clinical outcomes among men with prostate cancer: a systematic review and meta-analysis. Prostate Cancer Prostatic Dis 19:151-162, 2016 [Erratum: Prostate Cancer Prostatic Dis 19:222, 2016]
15. Chan JM, Kenfield SA, Paciorek A, et al: Postdiagnostic statin use and the risk of lethal prostate cancer in the Health Professionals Follow-up Study. Cancer Epidemiol Biomarkers Prev 24:1638-1640, 2015
16. Jung J, Lee C, Lee C, et al: Effects of statin use on the response duration to androgen deprivation therapy in metastatic prostate cancer. Korean J Urol 56:630-636, 2015
17. Larsen SB, Dehlendorff C, Skriver C, et al: Postdiagnosis statin use and mortality in Danish patients with prostate cancer. J Clin Oncol doi:10.1200/JCO.2016.71.8981 Abstract
18. Alfaqih MA, Allott EH, Hamilton RJ, et al: The current evidence on statin use and prostate cancer prevention: Are we there yet? Nat Rev Urol 14:107-119, 2017  
19. Roy M, Kung HJ, Ghosh PM: Statins and prostate cancer: role of cholesterol inhibition vs. prevention of small GTP-binding proteins. Am J Cancer Res 1:542-561, 2011 Medline
20. Sweeney C: ICECaP (Intermediate Clinical Endpoint for Cancer of the Prostate) study. American Society for Clinical Oncology Annual Meeting, Chicago, IL, June 3-7, 2016
21. Kantor ED, Lipworth L, Fowke JH, et al: Statin use and risk of prostate cancer: Results from the Southern Community Cohort Study. Prostate 75:1384-1393, 2015  

Journal of Clinical Oncology