We studied 93 female participants. Demographic and clinical variables in case and control groups are presented in Table 1.
In one-way ANOVA analysis, there were no significant differences among groups regarding SBP, DBP, total cholesterol, and menopausal age (p-Values > 0.05). Age, BMI, TG, HDL-C, LDL-C, ox-LDL, and LCAT activity levels differed significantly among groups (all p-values < 0.05) (Table 1).
As the duration of diabetes was not distributed normally, a nonparametric Mann-Whitney U test was performed. The median value (IQR) of durations of diabetes were significant between patients with diabetes with endometrial carcinoma (2 ± 2 year) and without endometrial carcinoma (4 ± 3 year), Mann-Whitney U = 106.50; p-value = 0.001.
The age of endometrial carcinoma in diabetes patients (58.37 ± 8.77) was significantly higher than the other groups, p-value ≤ 0.05. No significant differences among the three other groups were detected.
Table 1
Comparison of demographic and clinical variables in case and control groups
Variable | Endometrial carcinoma with diabetes (n = 19) | Endometrial carcinoma without diabetes (n = 17) | Diabetes (n = 31) | HC (n = 26) | F (p-value) |
Age (years) | 58.37 ± 8.77#$ | 51.47 ± 9.30 | 49.68 ± 6.46 | 49.62 ± 6.28 | 6.43 (0.001) |
Menopausal age (years) | 48.94 ± 3.54 | 46.07 ± 7.16 | 46.00 ± 6.00 | 47.46 ± 5.51 | 1.18 (0.321) |
Post menopause (%) | 89.5% | 47.1% | 48.4% | 42.3% | 11.78 (0.008) |
BMI (kg/m2) | 30.41 ± 6.39 | 31.83 ± 5.78# | 28.14 ± 4.27 | 27.32 ± 3.33 | 3.63 (0.016) |
SBP (mmHg) | 125.19 ± 13.04 | 120.73 ± 13.59 | 124.35 ± 11.37 | 124.42 ± 10.42 | 0.49 (0.68) |
DBP (mmHg) | 76.63 ± 10.33 | 78.20 ± 8.05 | 78.84 ± 9.69 | 77.88 ± 8.62 | 0.20 (0.89) |
Duration of Diabetes (years)* | 2.58 ± 2.34 | - | 5.23 ± 3.56 | - | 106.50** (0.001) |
FBS (mg/dl) ▪ | 138.53 ± 41.22 | 121.00 ± 27.45 | 196.81 ± 67.23 | 88.50 ± 6.16 | 3.80▪▪ (< 0.001) |
HbA1C (%) ▪ | 7.24 ± 1.12 | 5.46 ± 0.74 | 9.77 ± 2.25 | 4.96 ± 0.29 | 4.31▪▪ (< 0.001) |
Total cholesterol (mg/dl) | 191.89 ± 49.71 | 209.70 ± 54.63 | 208.16 ± 47.31 | 207.11 ± 25.16 | 0.68 (0.56) |
TG (mg/dl) | 204.21 ± 85.35# | 191.06 ± 61.77# | 175.87 ± 69.87# | 107.31 ± 36.27 | 10.45 (< 0.001) |
HDL-C (mg/dl) | 34.05 ± 8.34#$ | 35.82 ± 8.01#$ | 44.70 ± 9.39# | 54.69 ± 11.27 | 21.94 (< 0.001) |
LDL-C (mg/dl) | 77.63 ± 28.12#$ | 89.17 ± 31.33 | 111.20 ± 36.36 | 110.19 ± 12.64 | 7.289 (< 0.001) |
ox-LDL (mU/L) | 17.47 ± 0.84#$α | 12.36 ± 0.91#$ | 13.80 ± 1.35# | 7.57 ± 1.36 | 218.69 (< 0.001) |
ox-LDL/LDL-C | 0.24 ± 0.06#$α | 0.15 ± 0.06# | 0.14 ± 0.07# | 0.06 ± 0.01 | 30.73 (< 0.001) |
ox-LDL/HDL-C | 0.54 ± 0.14#$α | 0.36 ± 0.08# | 0.32 ± 0.07# | 0.14 ± 0.04 | 77.53 (< 0.001) |
LCAT activity (nmol/ml/h) | 28.31 ± 2.78#$ | 34.00 ± 4.97#$ | 46.58 ± 9.47# | 59.92 ± 2.46 | 236.73 (< 0.001) |
Data was presented as mean ± SD BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; FBS: fasting blood sugar; HbA1c: hemoglobin A1c; LCAT: Lecithin total cholesterol acyltransferase |
p-value < 0.05 was considered statistically significant.
αSignificant difference to endometrial carcinoma without diabetes
$Significant difference to diabetes
#Significant difference to Healthy
*Based on Mann-Whitney U test **Statistic Value: Mann-Whitney U
▪Based on Student’s t-test ▪▪Statistic Value: t
A comparison of menopausal status among groups showed different frequencies of premenopausal and postmenopausal states, and post-menopause status was higher in the endometrial carcinoma with diabetes group (Chi-square 11.78, p-value 0.008).
The BMI in the endometrial carcinoma without diabetes group (31.83 ± 5.78) was significantly higher than the healthy group (27.32 ± 3.33), p-value ≤ 0.05, but there were no significant differences among other groups.
The FBS and HbA1c values in the diabetes group were significantly higher than in the endometrial carcinoma with diabetes group (t = 3.80 and p-value < 0.001, t = 4.31 and p-value < 0.001, respectively).
Diabetes and Healthy groups had significantly higher levels of plasma LDL-C compared to endometrial carcinoma in diabetes patients, p-value ≤ 0.05, but there was no significant between the EC group compared to diabetes (DM) and healthy one. The level of ox-LDL was significantly higher in endometrial carcinoma with diabetes (17.47 ± 0.84) compared to the other three groups (all p-values ≤ 0.05). EC without diabetes (12.36 ± 0.91) and diabetes (13.80 ± 1.35) had a significantly higher level of ox-LDL compared to DM and healthy groups(p-value ≤ 0.05). In the DM group, ox-LDL was significantly higher than the healthy one(p-value ≤ 0.05). The level of plasma HDL-C in healthy controls (54.69 ± 11.27) and patients with diabetes (44.70 ± 9.39) were significantly higher than endometrial carcinoma with diabetes (34.05 ± 8.34) and endometrial carcinoma without diabetes (35.82 ± 8.01) groups, p-value ≤ 0.05, but there was no significant difference between cancer groups. Compared to healthy controls, the level of plasma TG in both cancer groups was significantly higher, p-value ≤ 0.05. The ox-LDL/LDL-C ratio was significantly higher in endometrial carcinoma with diabetes than in other groups (all p-values ≤ 0.05). There was a significantly higher ox-LDL/LDL-C ratio in the EC group and DM group compared to the healthy one, but no significant difference was seen between DM and EC groups. The ox-LDL/HDL-C ratio was significantly higher in all three patient groups compared to the healthy group (all p-values ≤ 0.05). Also, there was a significantly higher ox-LDL/HDL-C ratio in EC with DM patients compared to EC and DM groups.
The highest LCAT activity was seen in healthy controls, and the lowest activity was observed in patients with endometrial carcinoma and diabetes. According to pairwise comparisons, LCAT activity was significantly decreased in patients with endometrial carcinoma with and without diabetes than in patients with diabetes alone and controls (p-value < 0.001). However, there were no significant differences between the two groups of endometrial carcinoma regarding levels of LCAT (28.31 ± 2.78 VS. 34.00 ± 4.97, p-value = 0.77). Patients with diabetes had significantly lower LCAT activity levels than the control group (46.58 ± 9.47 VS. 59.92 ± 2.46, p-value < 0.001).
After adjustment for BMI, age, LDL-C, ox-LDL, HDL-C, and TG, the LCAT activity was compared among the groups, which yielded a significant difference (F = 42.43, p-value < 0.001, Fig. 1, A). Likewise, after adjustment for FBS, HbA1c, and diabetes duration, LCAT activity in endometrial carcinoma with the diabetes group was compared to the diabetes group. There was a significantly higher LCAT activity in the diabetes group (F = 136.41, p-value < 0.001).
According to the histopathological type of endometrial carcinoma, we divided patients with endometrial carcinoma into two groups, including 25 patients with type I (endometrioid) endometrial carcinoma and 12 patients with type II (non-endometrioid) endometrial carcinoma. After controlling for age, BMI, FBS, HbA1C, Chol., TG, HDL-C, LDL-C, and ox-LDL, there were no significant differences between the data of these two groups (all p-values were > 0.05).
There are two different histopathologic types of endometrial carcinoma. Type I lesions are more common, mostly endometrioid adenocarcinomas, sensitive to hormones, usually in lower stages, related to endometrial hyperplasia, hyperestrogenism, hyperlipidemia, obesity, and usually better prognosis. Type II lesions are predominantly non-endometrioid serous carcinoma, with lower differentiation, prone to deep invasion and recurrence, and hormone-independent. They arise from an atrophic endometrium and have a worse prognosis and higher mortality[3, 4, 30].
We classified each histopathological type group according to the presence of diabetes (number of patients and the amount of mean ± SD in each group is presented): Endometrioid endometrial carcinoma with diabetes (N = 13, 27.84 ± 2.11), endometrioid endometrial carcinoma without diabetes (N = 12, 35.36 ± 5.25), non-endometrioid endometrial carcinoma with diabetes (N = 6, 29.33 ± 3.93), and non-endometrioid endometrial carcinoma without diabetes (N = 6, 31.16 ± 3.67). Tukey’s test showed that LCAT level in endometrioid endometrial carcinoma without diabetes group is significantly higher than in endometrioid endometrial carcinoma with diabetes group (p-value = 0.001). However, there was no similar finding between non-endometrioid endometrial carcinoma patients (p-value = 0.854).
We also performed an interaction analysis between endometrial carcinoma and diabetes on serum LCAT activity. Three types of effects were obtained by regression analysis (Beta ± SE): the effect of diabetes alone, the effect of endometrial cancer alone, and the joint effect of diabetes and cancer on measured marker levels. The main effect of diabetes was − 13.345 ± 1.6, p = 0.001, and the main effect of endometrial carcinoma was − 25.92 ± 1.9, p = 0.001. Our results revealed a significant interaction effect between endometrial carcinoma and diabetes on LCAT activity 7.66 ± 2.63, p = 0.005. A company effect of both diseases was lower than simple quantitative addition of the effects of each disease, which means the association between endometrial carcinoma and diabetes represented sub additive interaction effect on LCAT activity (Fig. 1, B).
We independently studied the correlations between LCAT level and other variables in each group. In patients with concomitant endometrial carcinoma and diabetes, LCAT activity was not correlated to any other variables. In the endometrial carcinoma without diabetes group, LCAT activity was positively correlated to BMI (r = 0.612, p-value = 0.015). In diabetes group LCAT activity had a negative correlation to HbA1c (r= -0.686, p-value < 0.001) and ox-LDL / LDL-C ratio (r= -0.508, p-value = 0.004) and a week positive correlation to total cholesterol (r = 0.369, p-value = 0.041). In healthy group LCAT activity had a negative correlation to SBP (r= -0.549, p-value = 0.004), DBP (r= -0.559, p-value = 0.003), TG (r= -0.448, p-value = 0.022) and total cholesterol (r= -0.454, p-value = 0.020).