EvidenceHealthVitaminsSupplementsCancer

β‑Carotene Supplements & Lung Cancer in Male Smokers (ATBC)

Study type:: Large randomised, double‑blind, placebo‑controlled 2×2 factorial cancer‑prevention trial in male smokers
Topic:: β‑carotene supplementation and lung cancer risk in male smokers
Added:: 2026-01-12
Status:: published

Strength of evidence

Very low – strong randomisation and blinding with clinically important endpoints, but no true Null‑control arm and cancer defined entirely via diagnostic procedures and pathology; under WorthDay's strict evidence criteria this provides only very low‑quality causal evidence, while showing a robust harm signal for β‑carotene in smokers.

Source

The Alpha‑Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and β carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med. 1994;330(15):1029–1035.

Open original study

Science Logbook All studies

The ATBC trial enrolled over 29,000 Finnish male smokers and randomised them to β‑carotene (20 mg/day), vitamin E (50 mg/day), both, or placebo. Instead of preventing lung cancer, β‑carotene supplementation increased lung cancer incidence and overall mortality compared with no β‑carotene. Under WorthDay's criteria, the study cannot deliver a hard causal proof, but it offers a strong warning signal: high‑dose β‑carotene supplements are not protective in smokers and may be harmful.

Key findings

29,000+ Finnish men aged 50–69 years, all active smokers, were randomised to vitamin E, β‑carotene, both, or placebo in a 2×2 factorial design.
Men receiving β‑carotene had more lung cancers than those not receiving β‑carotene; relative risks were around 15–30 % higher, depending on subgroup.
Overall mortality and lung‑cancer‑specific mortality were also higher in the β‑carotene groups compared with no β‑carotene.
Vitamin E alone did not clearly protect against lung cancer or other major cancers.
According to WorthDay's evidence criteria, the trial relies on diagnostically defined cancer endpoints and lacks a true Null‑control arm, so it offers very low‑quality causal evidence – but it strongly challenges the idea that β‑carotene supplements are safe or beneficial for smokers.

Background

β‑carotene, a precursor of vitamin A and a plant pigment, has long been framed as a powerful antioxidant. Observational data once suggested that people with higher β‑carotene intakes had lower rates of lung cancer, leading to the hypothesis that supplements might protect smokers from smoke‑related oxidative damage. The Alpha‑Tocopherol, Beta Carotene Cancer Prevention (ATBC) Study was designed as a large, randomised trial to test whether vitamin E and β‑carotene could reduce lung cancer and other cancers in male smokers.

Methods

Species / population: 29,133 Finnish men aged 50–69 years, all of whom were active smokers at baseline. Design: Randomised, double‑blind, placebo‑controlled 2×2 factorial trial. Interventions: Vitamin E: 50 mg/day of dl‑α‑tocopheryl acetate vs matching placebo. β‑carotene: 20 mg/day of synthetic β‑carotene vs matching placebo. Participants were assigned to one of four groups: 1. Vitamin E only 2. β‑carotene only 3. Vitamin E + β‑carotene 4. Double placebo. Follow‑up: Median follow‑up in the original analysis was about 6 years, with extended follow‑up available in later publications. Endpoints (as defined by the investigators): Primary endpoint: incidence of lung cancer, confirmed by diagnostic work‑up and pathology. Secondary endpoints: incidence of other cancers and overall mortality. All cancer diagnoses were based on clinical detection, imaging and histopathological classification – i.e. through diagnostic systems rather than direct, purely mechanical observation.

Results

β‑carotene increased lung cancer incidence: Men assigned to β‑carotene (alone or with vitamin E) developed significantly more lung cancers than those not receiving β‑carotene. Relative risks for lung cancer were typically in the range of about 1.15–1.3 for β‑carotene groups vs non‑β‑carotene groups, indicating a 15–30 % higher incidence. β‑carotene increased overall mortality: All‑cause mortality was higher among β‑carotene recipients, with a meaningful contribution from excess lung‑cancer deaths. Vitamin E alone showed no strong protective effect: Vitamin E supplementation did not clearly reduce lung cancer or overall cancer incidence and did not reverse the harmful effect of β‑carotene. These findings directly contradicted the initial hypothesis that β‑carotene supplementation would protect smokers from lung cancer.

Interpretation

From a WorthDay perspective, S0035 is a key example of how supplement‑based prevention strategies can fail – or even backfire: In a large, well‑conducted, double‑blind trial in male smokers, high‑dose β‑carotene supplements did not prevent lung cancer. Instead, they were associated with statistically significant increases in lung cancer incidence and overall mortality. This overturns the simple narrative that "more antioxidants" must be beneficial, especially under high oxidative stress such as heavy smoking. The trial suggests that high‑dose β‑carotene may interact with tobacco smoke in ways that promote, rather than prevent, carcinogenesis. Under WorthDay's stricter evidence criteria, important methodological reservations are also highlighted: No true Null‑control arm: All participants swallowed capsules (active or placebo). Under WorthDay's logic, this means potential capsule‑ or expectation‑related effects cannot be fully separated from specific β‑carotene effects. Diagnostic cancer endpoints: Lung cancer is defined via imaging, biopsies and histopathology, all of which rely on diagnostic protocols and classification systems. WorthDay treats such endpoints as "soft" if used for strong causal claims about a specific supplement, because they embed assumptions and interpretive steps. Emphasis on relative risks: The harmful effect is mainly reported as a relative increase in risk. While here the effect is directionally clear and substantial, WorthDay still stresses the importance of absolute risk differences and confidence intervals when judging practical impact. Even with these caveats, the overall pattern is consistent and unfavourable to β‑carotene supplementation in smokers: There is no credible signal of protection; instead, there is a reproducible pattern of more lung cancer and more deaths in the β‑carotene groups.

Limitations

Population restricted to Finnish male smokers aged 50–69 years; results may not fully generalise to women, non‑smokers, other ethnic groups or different baseline diets.
The trial did not focus on individuals with clinically proven carotenoid deficiency; effects in genuinely deficient populations remain unknown.
Cancer detection depends on screening and diagnostic practices, which can influence incidence figures and may change over time.
Lack of a Null‑control group without any capsule intake means behavioural and expectation effects related to taking capsules cannot be entirely ruled out in WorthDay's framework.

WorthDay perspective

S0035 provides a strong warning signal against using high‑dose β‑carotene supplements as a "protective" strategy in smokers.
For WorthDay, this study illustrates that isolated, high‑dose antioxidant supplements can behave very differently from naturally occurring compounds in whole foods and can increase, rather than decrease, major disease risks.
The platform can use ATBC to highlight the principle: supplement‑based shortcuts to "detox" or "protection" are not automatically safe – robust trials can reveal real harms, especially when high doses are combined with heavy toxic exposures such as smoking.

Permalink: https://www.worth-day.com/science-compass/s0035-carotene-supplements-lung-cancer-in-male-smokers-atbc

Related studies

Science meets practice

Turn proven habits into your daily 100 points.

Start your free WorthDay now