Association Between Lifestyle Factors, Vitamin and Garlic Supplementation, and Gastric Cancer Outcomes: A Secondary Analysis of a Randomized Clinical Trial

Abstract

Importance: The associations of lifestyle factors with gastric cancer (GC) are still underexplored in populations in China. Long-term nutritional supplementation may prevent GC in high-risk populations, but the possible effect modification by lifestyle factors remains unknown.

Objective: To evaluate how lifestyle factors, including smoking, alcohol intake, and diet, may change the risk of GC incidence and mortality and whether the effects of vitamin and garlic supplementation on GC are associated with major lifestyle factors.

Design, setting, and participants: This is a secondary analysis of the Shandong Intervention Trial, a masked, randomized, placebo-controlled trial that aimed to assess the effect of vitamin and garlic supplementations and Helicobacter pylori treatment on GC in a factorial design with 22.3 years of follow-up. The study took place in Linqu County, Shandong province, China, a high-risk area for GC. Data were collected from Jully 1995 to December 2017. Overall, 3365 participants aged 35 to 64 years identified in 13 randomly selected villages who agreed to undergo gastroscopy were invited to participate in the trial and were included in the analysis. Data analysis was conducted from March to May 2019.

Interventions: Participants received vitamin and garlic supplementation for 7.3 years, H pylori treatment for 2 weeks (among participants with H pylori ), or placebo.

Main outcomes and measures: The primary outcomes were GC incidence and GC mortality (1995-2017). We also examined the progression of gastric lesions (1995-2003) as a secondary outcome.

Results: Of the 3365 participants (mean [SD] age, 47.1 [9.2] years; 1639 [48.7%] women), 1677 (49.8%) were randomized to receive active vitamin supplementation, with 1688 (50.2%) receiving placebo, and 1678 (49.9%) receiving active garlic supplementation, with 1687 (50.1%) receiving placebo. Overall, 151 GC cases (4.5%) and 94 GC deaths (2.8%) were identified. Smoking was associated with increased risk of GC incidence (odds ratio, 1.72; 95% CI, 1.003-2.93) and mortality (hazard ratio [HR], 2.01; 95% CI, 1.01-3.98). Smoking was not associated with changes to the effects of vitamin or garlic supplementation. The protective effect on GC mortality associated with garlic supplementation was observed only among those not drinking alcohol (never drank alcohol: HR, 0.33; 95% CI, 0.15-0.75; ever drank alcohol: HR, 0.92; 95% CI, 0.55-1.54; P for interaction = .03), and significant interactions were only seen among participants with H pylori (never drank alcohol: HR, 0.31; 95% CI, 0.12-0.78; ever drank alcohol: HR, 0.91; 95% CI, 0.52-1.60; P for interaction = .04). No significant interactions between vitamin supplementation and lifestyle factors were found.

Conclusions and relevance: In this secondary analysis of a randomized clinical trial, smoking was associated with an increased risk of GC incidence and mortality. Not drinking alcohol was associated with a stronger beneficial effect of garlic supplementation on GC prevention. Our findings provide new insights into lifestyle intervention for GC prevention, suggesting that mass GC prevention strategies may need to be tailored to specific population subgroups to maximize the potential beneficial effect.

Trial registration: ClinicalTrials.gov Identifier: NCT00339768.

Conflict of interest statement

Conflict of Interest Disclosures: None reported.

Figures

Figure 1.. Study Flow Diagram

Figure 2.. Associations Between Lifestyle Factors and Gastric Cancer Incidence and Mortality

HR indicates hazard ratio; OR, odds ratio. aLogistic regression adjusted for baseline histology, age, sex, history of ever using alcohol, history of ever smoking, vitamin supplementation (treatment or placebo), garlic supplementation (treatment or placebo), and H pylori treatment (H pylori treatment, placebo, or no H pylori). bEver smoking vs never smoking. cEver using alcohol vs never using alcohol. dConsuming at least 225 kg/y vs less than 225 kg/y (levels correspond to median distribution of intake between the 2 groups). eConsuming at least 8 kg/y vs less than 8 kg/y (levels correspond to median distribution of intake between the 2 groups). fConsuming at least 92 kg/y vs less than 92 kg/y (levels correspond to median distribution of intake between the 2 groups). gConsuming at least 81 kg/y vs less than 81 kg/y (levels correspond to median distribution of intake between the 2 groups). hCox regression adjusted for baseline histology, age, sex, history of ever using alcohol, history of ever smoking, vitamin supplementation (treatment or placebo), garlic supplementation (treatment or placebo), and H pylori treatment (H pylori treatment, placebo, or no H pylori).