Wang QL, Santoni G, Ness-Jensen E, et al. Association Between Metformin Use and Risk of Esophageal Squamous Cell Carcinoma in a Population-Based Cohort Study. Am J Gastroenterol. 2020 Jan;115(1):73-78. doi: 10.14309/ajg.0000000000000478. (Original study)

OBJECTIVES: Esophageal cancer is a highly fatal malignant neoplasm, with 2 etiologically different histological types. A large prospective study is expected to elucidate the specific risk of the 90% subtype of esophageal cancer, esophageal squamous cell carcinoma (ESCC), with metformin therapy. This study aims to determine the association between metformin use and incident ESCC risk.

METHODS: This was a nationwide population-based prospective cohort study conducted in Sweden in 2005-2015. Among 8.4 million participants identified in the cohort, 411,603 (5%) were metformin users. The users were compared with 10 times as many frequency-matched nonusers of metformin (n = 4,116,030) by age and sex. Metformin use was treated as a time-varying variate, and multivariable cause-specific proportional hazards model was used to calculate hazard ratios (HR) with 95% confidence intervals (CI) for ESCC, adjusted for age, sex, calendar year, residence area, tobacco smoking, alcohol overconsumption, and use of nonsteroidal anti-inflammatory drugs or statins.

RESULTS: The incidence rates of ESCC were 3.5 per 100,000 person-years among the metformin users and 5.3 per 100,000 person-years in the nonusers. Metformin users overall were at a decreased risk of ESCC compared with nonusers (HR 0.68, 95% CI 0.54-0.85). The decrease in risk was more pronounced in new metformin users (HR 0.44, 95% CI 0.28-0.64) and participants aged 60-69 years (HR 0.45, 95% CI 0.31-0.66).

DISCUSSION: Metformin use decreases the risk of developing ESCC.

Discipline Area Score
Internal Medicine 7 / 7
Family Medicine (FM)/General Practice (GP) 6 / 7
General Internal Medicine-Primary Care(US) 6 / 7
Oncology - Gastrointestinal 6 / 7
Endocrine 5 / 7
Comments from MORE raters

Endocrine rater

This paper suggests that metformin may decrease cancer. Since there was no information that metformin increased esophageal cancer, this paper will have no impact on clinical practice, which is why I rated it as not very relevant. It`s a little disingenuous to emphasize a distinction for this paper because it focused on one type of esophageal cancer that comprises 97% of the cancers. Including both types would not have changed the results.

Endocrine rater

A HR of 0.66 is modest for an epidemiologic study to rule out confounding by unmeasured and unimagined confounders. This is a hypothesis-generating article.

Endocrine rater

There are important methodologic limitations in this study. Metformin users were compared with non-metformin users, regardless of whether they had diabetes and with no adjustment for diabetes status or other anti-hyperglycemic agents.

General Internal Medicine-Primary Care(US) rater

This information needs to be validated by further study, preferably prospective randomized trials to ensure effectiveness and safety.

General Internal Medicine-Primary Care(US) rater

Interesting study. I won't be prescribing metformin for prevention of esophageal cancer, though.

Internal Medicine rater

This is a novel finding of a protective effect for a commonly used medication. It would be interesting to see this replicated in other databases.

Oncology - Gastrointestinal rater

The authors showed a small reduction in squamous cell carcinomas of the esophagus with metformin use in this large observational study. As a practicing oncologist, this information is of little clinical value. The benefit seen is small and the association is not proof of a cause-and-effect. Large randomized trials are needed, which is not likely. One may be somewhat reassured that the risk is not higher with metformin use.
Comments from EvidenceAlerts subscribers

No subscriber has commented on this article yet.