Association of Genomic Instability with HbA1c levels and Medication in Diabetic Patients

Abstract

Diabetes Mellitus type 2 (DM2) is associated with increased cancer risk. Instability of the genetic material plays a key role in the aetiology of human cancer. This study aimed to analyse genomic instability with the micronucleus cytome assay in exfoliated buccal cells depending on glycated haemoglobin (HbA1c) levels and medication in 146 female DM2 patients. The occurrence of micronuclei was significantly increased in DM2 patients compared to healthy controls. Furthermore, it was doubled in DM2 patients with HbA1c > 7.5% compared to subjects with HbA1c ≤ 7.5%. Positive correlations were found between micronuclei frequencies and HbA1c as well as fasting plasma glucose. Patients under insulin treatment showed a two-fold increase in micronuclei frequencies compared to subjects under first-line medication (no drugs or monotherapy with non-insulin medication). However, after separation of HbA1c (cut-off 7.5%) only patients with severe DM2 characterised by high HbA1c and insulin treatment showed higher micronuclei frequencies but not patients with insulin treatment and low HbA1c. We demonstrated that the severity of DM2 accompanied by elevated micronuclei frequencies predict a possible enhanced cancer risk among female DM2 patients. Therapy, therefore, should focus on a strict HbA1c control and personalised medical treatments.

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Figure 1. Differences of MN between HbA1c ≤ 7.5%, HbA1c > 7.5% and healthy controls.

Control group comprised 15 female healthy controls which were described previously. General characteristics of controls can be found in Supplementary Table 1. Bars show means and standard errors. White numbers in bars indicate the number of patients. Differences between the groups were analysed with Kruskal-Wallis test with pairwise comparisons. All groups were significantly different to each other with p 

Figure 2

Spearman correlation analyses were performed for total MN and HbA1c (a) and for total MN and FPG (b) for all DM2 subjects plus healthy controls (n = 161). Correlation analyses for controls (n = 15) and patients with either HbA1c ≤ 7.5% (n = 74) or HbA1c > 7.5% (n = 72) are presented for HbA1c with total MN (c). Control group comprised 15 female healthy controls which were described previously. r, spearman correlation coefficient; * indicates significance with p < 0.05 and *** for p < 0.001.

Figure 3. MN frequencies depending on Med groups.

Total MN frequencies were assessed in controls (n = 15) and patients of three medication groups: Med A: no medication or non-insulin monotherapy; Med B: non-insulin combination therapy; Med C: insulin medication (with or without other non-insulin medication).Control group comprised 15 female healthy controls which were described previously. Bars show means and standard errors. White numbers in bars indicate the number of patients. Differences between the groups were analysed with Kruskal-Wallis test with pairwise comparisons. Significance was assumed with p 

Figure 4. Distribution of MN frequencies depending on Med groups and HbA1c (cut-off 7.5%).

Med A: no medication or non-insulin monotherapy; Med B: non-insulin combination therapy; Med C: insulin medication (with or without other non-insulin medication). Each dot represents MN frequency of one patient. Med A within HbA1c > 7.5% was excluded from statistics due to only two cases. Differences between the groups were analysed with ANCOVA, diabetes duration and BMI as covariates. *represents significant difference to Med C of HbA1c > 7.5% with p < 0.01. #represents significant difference to Med B of HbA1c > 7.5% with p < 0.05. †indicates trend of difference to Med B of HbA1c > 7.5% with p < 0.1.

Figure 5. COVAIN results of PCA and bi-clustering of the 5 groups, divided by HbA1c and medication (Med A, B, C for HbA1c ≤ 7.5% and Med B, C for HbA1c > 7.5%).

Med A: no medication or non-insulin monotherapy; Med B: non-insulin combination therapy; Med C: insulin medication (with or without other non-insulin medication). Med A for HbA1c > 7.5% was excluded from statistics because of only 2 cases. (a) PCA plot shows two overlapping clusters representing the high HbA1c Med groups (red circle 1) and the low HbA1c Med groups (green circle 2). (b) Heat map shows bi-clustering of the 5 groups with contributing variables. A distinct grouping of Med B&C for either HbA1c ≤ 7.5% or HbA1c > 7.5% and a separate standing of Med A resulted. For detailed information about the differences between the 5 groups see Supplementary Table 2.