Diabetes, metformin and gastric adenocarcinoma

Sammanfattning: Diabetes may increase the risk and mortality of certain cancers, but its association with gastric adenocarcinoma is not clear. On the other hand, metformin, a first-line treatment for diabetes, may reduce cancer risks and improve cancer-related survival, but these associations have not been confirmed in gastric adenocarcinoma as well. The aim of this thesis is to evaluate if and how diabetes or its biomarkers and the use of metformin influence the risk and prognosis of gastric adenocarcinoma. Study I investigated if diabetes or prediabetes influenced the risk of gastric adenocarcinoma in a population-based Swedish cohort. Participants of the Northern Swedish Health and Disease Study were included and followed up from 1985 to 2017. Participants exposed to diabetes or prediabetes, as confirmed by self-reports or oral glucose tolerance tests, were compared with those of normoglycaemia for the incidence of gastric adenocarcinoma, identified from the Swedish Cancer Registry. Cox proportional hazard regression analyses with adjustment for sex, age, calendar year, body mass index, tobacco smoking, and education level showed no associations between prediabetes or diabetes and the risk of gastric adenocarcinoma (hazard ratio [HR] 1.07, 95% confidence interval [CI] 0.79-1.44 for prediabetes; HR 0.77, 95% CI 0.46-1.29 for diabetes). Study II was a systematic review and meta-analysis assessing associations between serum Haemoglobin A1c (HbA1c) or glucose and the risk of developing gastric cancer, based on studies identified from three databases: MEDLINE, Embase and Cochrane library. The random-effects model showed that elevated levels of serum HbA1c were associated with an increased risk of gastric cancer (pooled HR 1.36, 95% CI 1.06-1.74), but not so for elevated levels of serum glucose (pooled HR 1.11, 95% CI 0.98-1.26). Study III was a population-based cohort study evaluating whether diabetes worsened the prognosis in gastric adenocarcinoma, including all patients diagnosed with gastric adenocarcinoma between 1990 and 2014 in Sweden. Co-existing diabetes at diagnosis of gastric adenocarcinoma was analysed by the Cox proportional hazard regression with the risk of mortality due to gastric adenocarcinoma (disease-specific mortality) as well as all-cause mortality. The HRs were adjusted for sex, age, calendar year, and co-morbidity. Patients with diabetes at diagnosis had a moderately increased risk of disease-specific mortality compared with those without diabetes (HR 1.17, 95% CI 1.11-1.22). Besides, diabetes was also associated with an increased risk of all-cause mortality among patients who underwent gastrectomy for gastric adenocarcinoma (HR 1.23, 95% CI 1.10-1.38). Study IV was a population-based cohort study assessing whether metformin use decreased the risk of gastric adenocarcinoma. All data were retrieved from four national health data registries and participants were selected from users of certain commonly prescribed medications. Two sub-cohorts were established, a diabetes cohort and a matched cohort. Cox proportional hazard regressions with adjustment for sex, age, calendar year, comorbidity, Helicobacter pylori eradication treatment, use of non-steroidal anti-inflammatory drugs, and use of statins were used to compare users and non-users of metformin in relation to the risk of gastric non-cardia and cardia adenocarcinoma. The risk of gastric non-cardia adenocarcinoma was not decreased among metformin users compared with non-users in either sub-cohorts (HR 0.93, 95% CI 0.78-1.12 in the diabetes cohort; HR 1.30, 95% CI 1.18-1.42 in the matched cohort). Besides, metformin use did not decrease, but rather increased the risk of gastric cardia adenocarcinoma in either sub-cohorts (HR 1.49, 95% CI 1.09-2.02 in the diabetes cohort; HR 1.58, 95% CI 1.38-1.81 in the matched cohort). Study V was a population-based cohort study aiming to test if pre-diagnosis use of metformin improved the prognosis in gastric adenocarcinoma. The study included all diabetes patients who were diagnosed with gastric adenocarcinoma between 2005 and 2018. Associations between metformin use within two years before the diagnosis of gastric adenocarcinoma and the risk of disease-specific and all-cause mortality were analysed with Cox proportional hazard regressions, with adjustment for sex, age, calendar year, comorbidity, use of non-steroidal anti-inflammatory drugs, and use of statins. Metformin use was associated with a decreased risk of disease-specific mortality (HR 0.79, 95% CI 0.67- 0.93) and all-cause mortality (HR 0.78, 95% CI 0.68-0.90) among diabetes patients with gastric adenocarcinoma. To conclude, diabetes or prediabetes did not increase the risk of gastric adenocarcinoma in the Swedish population of Study I, but aggregated evidence indicated that long-term hyperglycaemia may increase the risk of gastric cancer (Study II). Gastric adenocarcinoma patients with co-existing diabetes had a higher risk of disease-specific mortality compared with those without diabetes (Study III). Besides, although metformin use might not prevent gastric adenocarcinoma (Study IV), it may improve the prognosis of this cancer among diabetes patients (Study V).