Objective Through studying a diabetic patient accompanied with pancreatic cancer by means of evidence-based clinical practice, to find out the relationship between diabetes mellitus and cancer and whether the long-acting insulin glargine increases the risk of cancer or not, which is regarded as a disputable hot issue at present. Methods Such databases as The Cochrane Library (Issue 3, 2010), OVID-EBM Reviews (1991 to Sept. 2010), MEDLINE (1950 to Sept. 2010) and CNKI (2000 to Sept. 2010) were retrieved to collect high quality clinical evidence, and the best therapy was formulated in accordance with the willingness of patients themselves. Results Eight randomized controlled trials (RCTs), four meta-analyses and one RCT meta-analysis were included. The evidence indicated that: a) Diabetes mellitus was kind of related to the occurrence of malignancies; b) There was no evidence at present showing the relationship between long-acting insulin glargine and cancer; c) Strictly controlling of blood sugar did not increase the risk of tumorigenesis, but hyperglycemia causing cancer was proofless; and d) Whether the diabetic patient with cancer should stop taking long-acting insulin glargine or not should require suggestions from specialists rather than patients themselves. Conclusion No evidence at present shows that tumorigenesis is related to diabetes mellitus, long-acting insulin glargine and strict controlling of blood sugar. It is necessary to require more evidence to decide whether the therapy should be adjusted or not for the diabetic patient with cancer who is in the process of glargine therapy.