Elsevier

The Lancet

Volume 364, Issue 9436, 28 August–3 September 2004, Pages 771-777
The Lancet

Articles
Mortality and incidence of cancer during 10-year follow-up of the Scandinavian Simvastatin Survival Study (4S)

https://doi.org/10.1016/S0140-6736(04)16936-5Get rights and content

Summary

Background

The effects of cholesterol-lowering treatment with statins on mortality and risk of cancer beyond the usual 5–6-year trial periods are unknown. We extended post-trial follow-up of participants in the Scandinavian Simvastatin Survival Study (4S) to investigate cause-specific mortality and incidence of cancer 5 years after closure of the trial.

Methods

4S was a randomised double-blind trial of simvastatin or placebo in patients with coronary heart disease, serum total cholesterol 5·5–8·0 mmol/L, and serum triglycerides 2·5 mmol/L or lower. The double-blind period lasted for a median of 5·4 years (range for survivors 4·9–6·3) and ended in 1994. After the trial, most patients in both groups received open-label lipid-lowering treatment. National registers were used to assess mortality and causes of death and cancer incidence in the original treatment groups for a median total follow-up time of 10·4 years (range for survivors 9·9–11·3). Analysis was by intention to treat.

Findings

414 patients originally allocated simvastatin and 468 assigned placebo died during the 10·4-year follow-up (relative risk 0·85 [95% CI 0·74–0·97], p=0·02), a difference largely attributable to lower coronary mortality in the simvastatin group (238 vs 300 deaths; 0·76 [0·64–0.90], p=0·0018). 85 cancer deaths arose in the simvastatin group versus 100 in the placebo group (0·81 [0·60–1·08], p=0·14), and 227 incident cancers were reported in the simvastin group versus 248 in the placebo group (0·88 [0·73–1·05], p=0·15). Incidence of any specific type of cancer did not rise in the simvastatin group.

Interpretation

Simvastatin treatment for 5 years in a placebo-controlled trial, followed by open-label statin therapy, was associated with survival benefit over 10 years of follow-up compared with open-label statin therapy for the past 5 years only. No difference was noted in mortality from and incidence of cancer between the original simvastatin group and placebo group.

Introduction

Publication of the Scandinavian Simvastatin Survival Study (4S) in The Lancet in 19941 was early trial evidence of the benefits of 3-hydroxy-3-methylglutaryl-CoA reductase inhibitors (statins) in the prevention of coronary events and in the reduction of all-cause mortality. The benefits of statin treatment both in primary and secondary prevention of coronary heart disease and other forms of atherosclerotic disease have since then become well established.2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12

In general, the safety profile of statins has been good, but not all concerns have been silenced, particularly with respect to the possible increased risk of cancer associated with pronounced cholesterol lowering.13, 14 This concern originated in the early 1990s from findings of prospective epidemiological studies showing some rise in non-cardiovascular mortality, particularly cancer deaths, in people with low cholesterol concentrations,15 and from results of early trials of cholesterol lowering.16 Furthermore, some researchers showed that lipid-lowering drugs, including statins, increase the occurrence of several types of cancer in rodents.17 Most statin trials, which generally last 5–6 years, have not shown any rise in cancer incidence in statin-treated participants, but in two studies some excess of cancer was reported. In the CARE trial,3 incidence of female breast cancer rose, and in the PROSPER trial in elderly people,9 incidence of all cancers increased in patients given pravastatin. However, meta-analyses of data from pravastatin9 and statin13 studies, and the large Heart Protection Study with simvastatin,7 did not show any significant excess of cancers. The average duration of randomised double-blind trials of statin treatment is, however, a fairly short period to study incidence of cancer. Therefore, longer follow-up of participants in statin trials has been called for.14

After completion of 4S in 1994, we decided that follow-up of the randomised patients would be extended by 5 years after the end of the double-blind period. Our aim was to investigate cause-specific mortality and cancer incidence in the original simvastatin and placebo groups on the basis of data from national registers in the five participating countries. Here, we report our 10-year follow-up results.

Section snippets

Methods

4S was a randomised, double-blind, placebo-controlled trial of simvastatin treatment in patients with clinically established coronary heart disease undertaken in five Nordic countries (Denmark, Finland, Iceland, Norway, and Sweden). The design, organisation, and practical aspects of the trial have been described in detail.1

Between May 19, 1988, and Aug 16, 1989, we recruited patients age 35–70 years with previous myocardial infarction or angina pectoris, serum total cholesterol 5·5–8·0 mmol/L,

Results

Figure 1 shows an outline of the 4S double-blind trial and the 5-year extension of follow-up. At the end of the double-blind period, 1967 survivors in the placebo group and 2039 in the simvastatin group did not differ with respect to sex and age distribution, smoking status, or baseline diagnoses (previous myocardial infarction, coronary-artery bypass graft or percutaneous coronary intervention, claudication, hypertension, diabetes), but as expected, serum lipid concentrations between patients

Discussion

The main finding of this 10-year follow-up study of the participants of 4S was that the survival benefit of patients allocated simvastatin compared with those allocated placebo that accrued during the double-blind trial period persisted during follow-up. The reduction in the relative risk between the two original treatment groups was not unexpected, because open-label treatment with lipid-lowering drugs (mostly statins) was given to most patients when the trial ended. After 3 years, more than

References (20)

There are more references available in the full text version of this article.

Cited by (0)

Collaborators and centres are listed in a previous report.

View full text