Journal of the National Cancer Institute Advance Access originally published online on June 24, 2008
JNCI Journal of the National Cancer Institute 2008 100(13):972-973; doi:10.1093/jnci/djn160
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2008. Published by Oxford University Press.
CORRESPONDENCE |
Re: The Association Between Statins and Cancer Incidence in a Veterans Population
Correspondence to: Uffe Ravnskov, MD, PhD, Magle Stora Kyrkogata 9, 22350 Lund, Sweden (e-mail: ravnskov{at}tele2.se).
In their cohort study of cancer and statin therapy, Farwell et al. (1) may have introduced a serious bias by excluding nonmelanoma skin cancer. In 4S, the Scandinavian Simvastatin Survival Study (2), and in HPS, the Heart Protection Study (3), the two first simvastatin trials, this cancer type was observed more often in the treatment groups. The difference was statistically significant if the results from both studies are combined (in simvastatin groups, 256 of the 12 490 participants; and in control groups, 208 of the 12 490 participants; P = .028). For unknown reasons, nonmelanoma skin cancers have been excluded in all reports from subsequent statin trials, but I hope that the relevant data can be made available on request.
The lag time between carcinogenic exposure and the clinical appearance of a cancer depends on its location. Lung cancer, for instance, is not diagnosed until after decades of smoking, whereas superficial nonmelanoma cancers may be observed much earlier. An increased number of patients with skin cancer in a trial is therefore alarming because this is the first cancer type that we should expect to find under conditions of general carcinogenicity.
A statistically significant number of other easily detectable cancer types associated with statin treatment compared with nontreatment have been reported. Iwata et al. (4) found that 29 (13.3%) of 221 patients with lymphoid cancers had been treated with statins (mean treatment time = 48 months) but that only 64 (7.3%) of the 879 orthopedic and otorhinolaryngological control individuals without cancer had used statins (multivariable odds ratio = 2.24; 95% confidence interval [CI] = 1.37 to 3.66; P = .001). Furthermore, in CARE, the Cholesterol and Recurrent Events Trial (5), 12 of the 286 women in the pravastatin group but only one of the 290 in the placebo group had breast cancer at follow-up (P = .002).
Several of these breast cancers were recurrences, again a disquieting finding because recurrences may appear earlier than primary cancers. The hypothesis that statin treatment may provoke recurrences is not possible to test, however, because after the publication of the CARE report, previous cancer has become an exclusion criterion in all trials.
Farwell et al. (1) also excluded patients who were diagnosed with cancer within 2 years after entry. Dormant cancer is a common finding in elderly people, and a carcinogenic effect should therefore appear earlier in that patient group. Indeed, in the PROSPER trial (6), which included elderly people only, 245 of the 2891 participants in the pravastatin group but only 199 of the 2913 in the placebo group had new cancer (the number of recurrences was not given). The difference was already obvious after 1 year, and it increased steadily during the trial period to become statistically significant (P = .02) after 4 years.
Another bias may have been introduced by comparing patients on statin treatment with untreated individuals. The first group is a selection of people with high cholesterol and the second is dominated by people with normal and low cholesterol, and numerous cohort studies have found an inverse association between cholesterol and cancer.
A further warning comes from a cohort study by Matsuzaki et al. (7). In a cohort study of 47 294 Japanese patients treated with low-dose simvastatin and followed for 6 years, the authors found that the number of cancer deaths was statistically significantly higher in patients whose total cholesterol at follow-up was less than 160 mg/dL than in those whose cholesterol was 200–219 mg/dL (relative risk = 3.16; 95% CI = 1.72 to 5.81; P = .001).
Statin treatment is prescribed to millions of people for the rest of their life. More information in this area is therefore urgently required. For this reason, meta-analyses of the statin trials should include subgroup analyses of obvious risk groups (for instance, older people and smokers) and of early detectable cancers and recurrences.
REFERENCES
1. Scandinavian Simvastatin Survival Study GroupThe association between statins and cancer incidence in a veterans population. J Natl Cancer Inst. (2008) 100(2):134–139.
2. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease the Scandinavian Simvastatin Survival Study (4S). Lancet (1994) 344(8934):1383–1389.[CrossRef][Web of Science][Medline]
3. Heart Protection Study Collaborative Group. MRC/BHF heart protection study of cholesterol lowering in 20536 high-risk individuals: a randomised placebo-controlled trial. Lancet (2002) 360(9326):7–22.[CrossRef][Web of Science][Medline]
4. Iwata H, Matsuo K, Hara S, et al. Use of hydroxy-methyl-glutaryl coenzyme A reductase inhibitors is associated with risk of lymphoid malignancies. Cancer Sci. (2006) 97(2):133–138.[CrossRef][Medline]
5. Sacks FM, Pfeffer MA, Moye LA, et al. for the Cholesterol and Recurrent Events Trial investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med (1996) 335(14):1001–1009.
6. Shepherd J, Blauw GJ, Murphy MB, et al. Pravastatin in elderly individuals at risk of vascular disease (PROSPER): a randomised controlled trial. Lancet (2002) 360(946):1623–1630.[CrossRef][Web of Science][Medline]
7. Matsuzaki M, Kita T, Mabuchi H, et al. Large scale cohort study of the relationship between serum cholesterol concentration and coronary events with low-dose simvastatin therapy in Japanese patients with hypercholesterolemia. Circ J (2002(12)) 66:1087–1095.[CrossRef][Web of Science]
Response to this Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 973-974.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||