Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.823235
Title: The relationship between type 2 diabetes mellitus and cancer-related mortality and survival using linked CPRD data
Author: Alam, Nasra
ISNI:       0000 0005 0290 3585
Awarding Body: University of Manchester
Current Institution: University of Manchester
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Background: Type 2 diabetes (T2D) and cancer commonly co-occur. Excess body fatness, commonly expressed as Body Mass Index (BMI), is a risk factor for T2D and up to 13 cancer types, here referred to as obesity-related cancers (ORCs). Individuals with T2D have higher cancer mortality than individuals without T2D and it is proposed that individuals with cancer and T2D have a poorer survival than those without T2D. Excess body weight may be important in understanding this observation and may be a mechanism for driving increased cancer risk, cancer-related mortality and cancer survival. The aims of this thesis were threefold: i) to determine whether individuals with T2D have a higher cancer-related mortality compared with individuals without T2D, and if this excess cancer death was mainly from ORCs (DIAMOND 1A); ii) to evaluate the relationship between excess body weight and cancer mortality in individuals with T2D (DIAMOND 1B); and iii) to evaluate the impact of T2D on survival in patients with cancer, independent of its effect on survival in individuals without cancer (DIAMOND 2). Methods: DIAMOND 1A and B were performed in a matched cohort study in the Clinical Practice Research Datalink (CPRD), linked with mortality statistics (1998-2015), to derive cohorts with incident T2D (176,886) versus never diabetes (control: 852,946), aged 30 to 85 years. The primary outcome was total cancer mortality, categorised into deaths from ORC and non-ORC. Several sources of bias were explored. To understand the role of BMI, the cohort with incident T2D was used, comprising 144,802 individuals with T2D with BMI recorded at the time of diagnosis. Hazard ratios (HRs) were estimated using Cox proportional hazard models. DIAMOND 2 was performed in a matched cohort study in the CPRD linked with the National Cancer Registration and Analysis Service to derive a cohort of patients with incident cancer (167,360) (1998-2015) and a cohort without cancer (771,795). The primary outcome was overall survival (OS); the secondary outcome was cancer-specific survival. 10-year OS rates were estimated using the Kaplan-Meier survival function. Stratified Cox regression models incorporating the interaction between cancer and prevalent T2D were carried out to investigate the impact of T2D on survival in patients, independent of its effect on survival in individuals without cancer. Results: In DIAMOND 1A, with a mean follow-up of 7.1 years, there were 47,459 cancer deaths (T2D: 9,601; non-diabetes: 37,850). Compared with the control population, T2D was associated with higher ORC mortality in men (HR: 1.71, 95% CIs: 1.56-2.87) and women (HR: 1.34 95% CIs: 1.24-1.44). In DIAMOND 1B, among individuals with T2D, associations with BMI were generally null or inverse, except for positive associations in specific cancers, such as endometrial (HR per 5 kg/m2: 1.42 95% CIs: 1.25-1.60) and ovarian (HR per 5 kg/m2: 1.14 95% CIs: 1.02-1.29) cancers. Evidence suggesting reverse causality was found. In DIAMOND 2, in men, the 10-year OS rates were: cancer never T2D, 32%; cancer prevalent T2D, 28%; never cancer never T2D 68%; never cancer prevalent T2D, 57%. In women, the respective 10-year OS were: 45%, 36%, 78%, 62%. A significant interaction between cancer and prevalent T2D was observed in women (HR 0.85 CI: 0.78-0.92) but not in men (HR: 1.05, CI: 0.99-1.12). Conclusions: Individuals with T2D have higher ORC mortality than those without, supporting the notion that obesity is a likely mechanism contributing to excess cancer mortality in T2D. However, factors beyond body fatness may also contribute to this excess. The co-occurrence of diabetes at the time of cancer diagnosis is associated with a poorer 10-year survival compared with those who never have diabetes. But there was evidence that the disadvantage attributed to diabetes in patients with cancer is less than that expected, in women but not in men.
Supervisor: Renehan, Andrew ; Ashcroft, Darren ; Buchan, Iain ; Wright, Alison Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.823235  DOI: Not available
Share: