Clinical pathological characteristics and prognostic analysis of diabetic women with luminal subtype breast cancer.

This study selected luminal-type breast cancer patients as the study subjects. The patients were divided into groups according to the presence of diabetes and the types of medication used, and the patients' clinicopathological characteristics and prognostic indicators were explored. A total of 5,785 patients with luminal-type breast cancer admitted to Tianjin Medical University Cancer Institute and Hospital between January 2002 and December 2006 were selected as the study subjects. The subjects included 680 breast cancer patients with diabetes and 5,105 breast cancer patients without diabetes. The patients were divided into Luminal A, Luminal B (high ki67), and Luminal B (her-2/neu+) subtypes. Each subtype was further divided into a metformin group, a non-metformin group, and a nondiabetic group. The research indicators included breast cancer mortality, age, body mass index (BMI), amenorrhea, the presence of cardiovascular and cerebrovascular disease, pathological stage, pathological type, lymph node involvement, vessel carcinoma embolus, and the chemotherapy and endocrine regimen. A Kaplan-Meier analysis was conducted to analyze the differences in breast cancer mortality rates among the groups. The Cox proportional hazard model was adopted to detect independent factors related to prognosis. Kaplan-Meier univariate analysis showed that for the Luminal A, Luminal B (high ki67), and Luminal B (her-2/neu+) subtypes, the cancer-specific mortality rates differed significantly among the metformin, non-metformin, and nondiabetic groups. The 5-year survival rates were 94%, 82%, and 91% (P = 0.002); 93.5%, 81%, and 89% (P < 0.001); and 84%, 77%, and 83% (P = 0.035) for the subtypes within each group, respectively. Cox regression multivariate analysis showed that compared with the metformin group, all three subtypes of the, the non-metformin group showed poorer prognosis (hazard ratio [HR], 3.579; 95% confidence interval [CI], 1.506-8.506 [P = 0.004]; HR, 3.232; 95% CI, 1.839-5.678 [P < 0.001]; HR, 2.034; 95% CI,1.019-4.059 [P = 0.044] for Luminal A, Luminal B (high ki67), and Luminal B (her-2/neu+, respectively). Compared with the metformin group, the diabetic group showed poorer prognosis only for the Luminal B (high ki67) subtype (HR, 1.762; 95% CI, 1.033-3.005 [P = 0.038]). In addition, for the Luminal A, Luminal B (high ki67), and Luminal B (her-2/neu+) subgroups, there was a higher proportion of elderly patients (P < 0.001) and postmenopausal patients (P < 0.001) in the metformin and non-metformin groups than in the nondiabetic group. Moreover, the probability of having cardiovascular and cerebrovascular disease was also higher (P < 0.001) in the metformin and non-metformin groups. For the Luminal B (high ki67) and Luminal B (her-2/neu +) subgroups, there was a higher proportion of obese patients in the metformin and non-metformin groups (P < 0.001). In terms of clinical characteristics, for the Luminal B (high ki67) subtype, the proportion of patients with invasive ductal carcinoma was lower in the non-metformin group than in the other two groups (P = 0.001). In both the metformin and non-metformin groups, the proportion of T3/4 patients was higher (P < 0.001), the proportion of patients with lymph node metastasis was higher (P = 0.001), and the proportion of patients with vessel carcinoma embolus was higher (P = 0.001) compared with the nondiabetic group. In conclusion, compared with the metformin group, the non-metformin group had a poorer prognosis for all subtypes of luminal breast cancer. In the diabetic group, only patients with the Luminal B (high ki67) subtype exhibited a poorer prognosis. Therefore, different diabetes medication may have a different impact on the prognosis of different subtypes of luminal breast cancer.