4S Extension Data and What Happened After the Trial Ended

At a glance

Parameter	Detail
N	4,444 (2,221 simvastatin, 2,223 placebo)
Intervention	Simvastatin 20 to 40 mg/day
Comparator	Placebo
Median in-trial follow-up	5.4 years
Post-trial observation	Extended to ~10 years total
Primary endpoint	All-cause mortality
Key result	30% relative risk reduction in all-cause mortality (p = 0.0003)

Why Extension Data Mattered More Than the Primary Result

When the 4S results were published in November 1994, the 30% reduction in all-cause mortality among patients with established coronary heart disease was unprecedented for a lipid-lowering agent. But the trial's median 5.4-year follow-up left open several questions that only longer observation could answer.

Critics raised three specific concerns. First, earlier cholesterol-lowering trials (WHO clofibrate, for example) had shown reductions in cardiac events offset by increases in non-cardiovascular death. Could the same pattern emerge with simvastatin over a longer horizon? Second, low cholesterol had been epidemiologically linked to cancer incidence in observational studies, raising the possibility that statin-induced LDL reduction might promote malignancy with a latency period beyond the trial window. Third, because the original 4S protocol allowed dose titration from 20 mg to 40 mg to hit target cholesterol levels, it was unclear whether the benefit would hold once participants were unblinded and treatment patterns changed.

These were not theoretical objections. They shaped FDA labeling language, influenced European prescribing guidelines through the late 1990s, and kept many physicians cautious about prescribing statins to elderly patients for years after the primary results were available.

The Post-Trial Follow-Up Design

The 4S investigators arranged an extended follow-up through the Scandinavian national registries and direct patient contact. The key methodological points:

• Unblinding occurred in 1994 after the Data Safety Monitoring Board stopped the trial early due to clear mortality benefit.
• Crossover was immediate and substantial. Within months of unblinding, the majority of placebo-group patients began statin therapy. By 2 years post-trial, statin use in the former placebo group exceeded 80%.
• Follow-up extended to approximately 10 years from randomization. Cause-specific mortality was ascertained through national death registries, which in Scandinavian countries carry near-complete capture.
• No new randomization occurred. This was observational follow-up of the original randomized cohorts, meaning the groups progressively lost their treatment contrast as placebo patients adopted statins.

The follow-up results were reported by Strandberg et al. in The Lancet in 2004, covering outcomes through 2001.

Results From the Extended Observation

Mortality Outcomes at 10 Years

The attenuation of the mortality difference at 10 years was expected and, in fact, confirms the treatment effect rather than undermining it. Once placebo patients received statins, their risk trajectories improved. The between-group gap stopped widening not because simvastatin failed, but because the control group caught up. This is a textbook example of crossover contamination diluting an intention-to-treat signal in long-term follow-up.

Cancer Incidence: The Question That Took a Decade to Answer

The original 4S publication reported no excess cancer in the simvastatin group during the trial period, but 5 years was considered insufficient to rule out a latent carcinogenic effect tied to sustained LDL lowering. The extended follow-up tracked cancer diagnoses and cancer-specific mortality for an additional 5 years.

Results from the extension were reassuring:

No signal emerged for any individual cancer type. This finding was consistent with later meta-analyses from the Cholesterol Treatment Trialists' (CTT) Collaboration, which pooled over 170,000 participants across 26 statin trials and found no excess cancer risk at any site. The 4S extension data was among the earliest large-scale evidence that put the statin-cancer hypothesis to rest.

Non-Cardiovascular Mortality

The clofibrate-era concern about non-cardiovascular death replacing cardiovascular death did not materialize. Extended follow-up showed:

• No excess in deaths from violence, suicide, or accident
• No excess in hepatobiliary or gastrointestinal mortality
• No excess in deaths from respiratory disease

This mattered because some earlier lipid-lowering agents had shown exactly these patterns. The absence of non-cardiovascular safety signals in a 10-year window gave regulators and guideline writers confidence that the simvastatin benefit was net positive, not a trade of one cause of death for another.

Regression to the Mean and the Durability Question

One methodological criticism of the original 4S trial centered on regression to the mean. Patients were enrolled during a period of clinical stability (at least 6 months post-MI or with stable angina), and cholesterol was measured at a dietary run-in phase. Critics suggested the apparent LDL reductions might partly reflect natural fluctuation.

The extension data addressed this indirectly. The lipid changes observed in-trial (35% LDL reduction, 8% HDL increase) were consistent with the pharmacological profile of simvastatin at 20 to 40 mg, not with regression artifacts. Patients who continued simvastatin post-trial maintained similar lipid profiles. And the clinical benefit tracked with the lipid change in a dose-response pattern that regression to the mean cannot explain.

The Heart Protection Study (HPS), which enrolled over 20,000 patients and was reported in 2002, later provided definitive confirmation: simvastatin 40 mg reduced major vascular events by about one-quarter regardless of baseline cholesterol, eliminating any plausible regression-to-the-mean explanation.

What the Extension Could Not Show

Long-term follow-up has limits, and the 4S extension is honest about several:

Crossover contamination. The most important limitation. Once placebo patients started statins, the randomized comparison degraded. The 10-year mortality data does not measure "statin vs. no statin for 10 years." It measures "statin for 10 years vs. statin for 5 years." This actually underestimates the true long-term benefit.

Single-drug, single-dose era. The 4S trial used simvastatin 20 to 40 mg. Modern practice uses a range of statins at varying intensities. The extension cannot speak to the durability of high-intensity atorvastatin or rosuvastatin regimens.

Narrow population. 4S enrolled Scandinavian patients aged 35, 70 with total cholesterol 5.5 to 8.0 mmol/L and established CHD. The extension follow-up applied to this same cohort. Generalizability to primary prevention, to patients with lower baseline cholesterol, or to more ethnically diverse populations requires data from other trials.

No myopathy tracking post-trial. While the in-trial safety profile showed low rates of myopathy, the post-trial observational phase did not systematically collect muscle-related adverse events. Later pharmacovigilance and the FDA simvastatin label revision in 2011, which restricted the 80 mg dose due to myopathy risk, drew on other data sources entirely.

Clinical Translation: Why This Still Matters

The 4S extension data shaped three durable clinical principles:

1. Early initiation pays off. Patients randomized to simvastatin from the start accumulated a mortality advantage that persisted even after the placebo group began treatment. The current ACC/AHA cholesterol guidelines cite this principle when recommending statin therapy be started promptly after an acute coronary event rather than deferred.

2. Statins do not trade cardiovascular benefit for non-cardiovascular harm. The 10-year safety data closed the door on the hypothesis that had haunted lipid-lowering therapy since the 1970s.

3. Long-term adherence is the real bottleneck. The extension showed that once patients were given statins (even the former placebo group), outcomes improved. The clinical challenge is not whether statins work over decades but whether patients keep taking them. Real-world adherence data consistently shows that 40 to 50% of patients discontinue statins within 1 year of initiation.

Frequently asked questions

›How long were 4S participants followed after the trial ended?

The original randomized period lasted a median of 5.4 years. Post-trial follow-up extended observation to approximately 10 years from the date of randomization. Outcomes were tracked through Scandinavian national death and hospital registries.

›Did the mortality benefit of simvastatin persist in long-term follow-up?

The absolute mortality gap narrowed over the extended follow-up period, but this was driven by placebo-group crossover to statin therapy. The attenuation reflects treatment of the control group, not failure of the intervention. Patients who received simvastatin from the start retained an early-accrued survival advantage.

›Was there any increase in cancer with long-term simvastatin use?

No. At 10 years, cancer incidence was 12.2% in the simvastatin group versus 12.0% in the placebo group. No individual cancer type showed excess risk. These findings were later confirmed by the CTT meta-analysis of over 170,000 statin-treated patients.

›Did non-cardiovascular deaths increase in the simvastatin group?

No. There was no excess in deaths from cancer, trauma, suicide, liver disease, or any non-cardiovascular category. This was a direct rebuttal to the hypothesis, raised by earlier clofibrate trials, that cholesterol lowering might increase non-cardiac mortality.

›Why did the between-group mortality difference shrink after the trial ended?

Because most placebo patients crossed over to statin therapy within months of unblinding. By 2 years post-trial, over 80% of the former placebo group was taking a statin. This crossover contamination diluted the randomized comparison, which is a known limitation of post-trial observational follow-up.

›What dose of simvastatin was used in the 4S trial?

Patients started at 20 mg daily. The dose was titrated to 40 mg if total cholesterol remained above 5.2 mmol/L at the first scheduled visit. About 37% of patients required the higher dose. The 80 mg dose, later restricted by the FDA for myopathy risk, was not used.

›Did the 4S extension data influence current statin guidelines?

Yes. The 4S extension contributed to the evidence base for early statin initiation in secondary prevention and helped establish long-term safety. Both the 2018 ACC/AHA cholesterol guidelines and EAS/ESC guidelines cite the 4S trial as foundational evidence for statin use in established coronary disease.

›How does the 4S extension compare to other long-term statin follow-up studies?

The WOSCOPS 20-year follow-up (primary prevention with pravastatin) and the LIPID trial extension are the most comparable. All three showed durable benefit with no late-emerging safety signals. The 4S extension is unique in being the first to demonstrate this pattern and in covering the highest-risk population (established CHD with elevated cholesterol).

›Were there any muscle-related safety signals in long-term follow-up?

The post-trial observational period did not systematically track myopathy. In-trial myopathy rates were low (under 0.1%). The more clinically relevant myopathy data came from later pharmacovigilance, which led the FDA to restrict the simvastatin 80 mg dose in 2011. The 4S trial used only the 20 mg and 40 mg doses.

›Is the 4S trial still clinically relevant today?

The specific drug and doses have been largely superseded by higher-intensity regimens (atorvastatin 80 mg, rosuvastatin 20 to 40 mg). But the 4S trial and its extension remain clinically relevant as the first proof that lowering LDL with a statin reduces all-cause mortality, and that this benefit is durable and safe over a decade of observation.

References

1. Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344(8934):1383-1389. PubMed

2. Strandberg TE, Pyörälä K, Cook TJ, et al. Mortality and incidence of cancer during 10-year follow-up of the Scandinavian Simvastatin Survival Study (4S). Lancet. 2004;364(9436):771-777. PubMed

3. Cholesterol Treatment Trialists' (CTT) Collaboration. Lack of effect of lowering LDL cholesterol on cancer: meta-analysis of individual data from 175,000 people in 27 randomised trials of statin therapy. PLoS One. 2012;7(1):e29849. PubMed

4. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360(9326):7-22. PubMed

5. Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol. J Am Coll Cardiol. 2019;73(24):e285-e350. PubMed

6. FDA. Zocor (simvastatin) prescribing information. Revised 2012. FDA Label