Less aggressive treatment and worse overall survival in cancer patients with diabetes: A large population based analysis

Abstract

The purpose of this study was to document the prevalence of diabetes among newly diagnosed cancer patients and to evaluate the influence of diabetes on stage at diagnosis, treatment and overall survival. We performed a population‐based analyses of all 58,498 cancer patients newly diagnosed between 1995 and 2002 in the registration area of the Eindhoven Cancer Registry. Stage of cancer, cancer treatment and comorbidities were actively collected by hospital medical records review. Follow‐up of all patients was completed until January 1, 2005. Nine percent of all cancer patients had diabetes at the time of cancer diagnosis. The prevalence of diabetes was highest among patients with cancer of the pancreas (19%), uterus (14%) and among young men with kidney cancer (8%). Colon, breast and ovarian cancer patients with diabetes were more often diagnosed with a higher tumour stage (p < 0.05). Patients with diabetes and cancer of the oesophagus, colon, breast and ovary were treated less aggressively compared to those without diabetes (p < 0.05). During the follow‐up period 3,902 of 5,555 cancer patients with diabetes died and 29,909 of 52,943 cancer patients without diabetes died. For all cancers combined, in a multivariate cox‐regression model, adjusting for age, gender, stage, treatment and cardiovascular disease, patients with diabetes experienced a significant increase in overall mortality (HR = 1.44, 95% CI 1.40–1.49), ranging however from 0 to 40% for different types of cancer, compared to those without diabetes. In conclusion, diabetic cancer patients frequently were treated less aggressively and had a worse prognosis compared to those without diabetes. © 2007 Wiley‐Liss, Inc.

Impact of cancer on adherence to glucose-lowering drug treatment in individuals with diabetes

“…Another study showed that individuals with diabetes and recent HbA 1c values >9.0% (75 mmol/mol; OR 1.5; 95% CI 1.3, 1.7) had higher mortality than those with recent 'normal' HbA 1c values between 6.5% (48 mmol/mol) and 9% (75 mmol/mol) [18]. Thus, the drop in MPR observed among patients with cancer in our study might have negatively influenced survival via the aforementioned mechanisms and may (partly) explain the established association between diabetes, cancer and survival [1][2][3].…”

“…Although the impact of cancer was more pronounced among cancers with a worse prognosis and among those with more advanced TNM stages, the difference in prognosis associated with these cancers seemed to only partly explain the impact of cancer on medication adherence. The decline in adherence seen among users of GLDs with cancer might negatively affect survival and (partly) explain the established association between diabetes, cancer and survival [1][2][3]. In future studies, the reason for the decline in MPR needs to be further elucidated among the different cancer types-is it the patient who prioritises the fight against cancer or the advice of the physician to stop the treatment?…”

“…Cancer patients with diabetes have a significantly higher overall mortality risk than those without diabetes [1][2][3]. To understand the association between diabetes and cancer, the American Diabetes Association and American Cancer Society reviewed the state of science regarding this in 2010 [4].…”

Is Diabetes Mellitus a Prognostic Factor for Survival in Patients with Small Cell Lung Cancer?

“…It is limited and conflicting. Although the few prior studies (Park et al, 2006;Van de Poll et al, 2007;Win et al, 2008;Varlotto et al, 2011) showed that pre-existing DM that at the time of diagnosis strongly associated increase OS. Contrary to this, the meta-analysis analysis by Barone et al (2008) and the study of Tammemagi et al (2003) no observed this relationship between diabetes and lung cancer.…”

“…It was demonstrated previously parameters that good PS, disease stage, age and weight loss associated as strong prognostic factors whereas DM was not clearly or consistently identified (Park et al, 2006;Van de Poll et al, 2007;Barone et al, 2008;Win et al, 2008;Varlotto et al, 2011).…”