We included 150 ovarian cancer patients who presented to Cancer Genetics clinic of Oncology Institute in Istanbul University for BRCA1/2 investigation, and 100 healthy controls who were matched for age, sex, and ethnicity with patients were included in the study. The experimental procedures were continued using the RNA materials of a total of 127 ovarian cancer patients, and 98 healthy individuals after the quality control of biological material was performed. The mean age of the ovarian cancer patient group was 50±10(23-84y) years, and the mean age of the control group was 48±11 (23-84y) years.
The normality estimation of the investigated groups was identified using the Kolmogorov-Smirnov test. The normality distributions of the data were calculated for miR-423-5p (Kolmogorov-Smirnov Z=4,731; p<0.05); and for miR-664b-5p (Kolmogorov-Smirnov Z=5,619; p<0.05). We used the post-hoc ANOVA test in the resolution of the data as the groups showed normal distribution, and to perform calculation for subgroups, and there were a total of 5 subgroups.
The calculations based on 2-∆∆Ct values using the ANOVA test between the patient and control groups showed that miR-423-5p, and miR-664b-5p expression levels were found statistically significantly higher in patient group compared with the healthy controls (p:0.000).
Comparison of the Presence of BRCA mutation and miRNA expresion levels of patients: miR-423-5p expression was found to have increased in 52% (13/25), and decreased in 48% (12/25) of patient who were positive for BRCA mutation in our study (Table 1). However, miR-423-5p expression was found to have increased in 57.8% (59/102), and was found to have decreased in 42.2% (43/102) of patients who were negative for BRCA mutation. In general, miR-423-5p was found to have similar distribution in positive and negative group for BRCA mutation.
The comparison of the presence of BRCA mutation, and the miR-664b-5p expression showed that miR-664b-5p was detected to have decreased both in most negative, and positive cases. Accordingly, miR-664b-5p expression level was detected to have increased in 36%(9/25) of BRCA mutation positive patients, and to have decreased in 64% (16/25) of the patients. In addition, miR-664b-5p expression level was found to have increased in 39.2% (40/102), and decreased in 60.8% (62/102) in patients with negative BRCA mutation.
Table 1: The Association of the BRCA Mutation Status, and miRNA expression levels in patients with ovarian cancer.
BRCA Mutation Status
|
miR-423-5p
|
miR-664b-5p
|
|
Increased
n(%)
|
Decreased
n(%)
|
Increased
n(%)
|
Decreased
n(%)
|
Total
n(%)
|
Positive
|
13(52%)
|
12(48%)
|
9(36%)
|
16(64%)
|
25(19.6%)
|
Negative
|
59(57.8%)
|
43(42.2%)
|
40(39.2%)
|
62(60.8%)
|
102(80.3%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Comparison of the miRNA expression levels in the patient and control groups:
The evaluation of miR-423-5p, and miR-664b-5p expressions in all ovarian cancer patient group included in the study showed that miR-423-5p expression was decreased in 43.3% (55/127), and increased in 56.7%(72/127). miR-664b-5p expression was detected to have decreased in 61.5% (78/127), and increased in 38.5%(49/127) out of the total 127 patients (Table 2).
The evaluation of the miR-423-5p, and miR-664b-5p expressions in ovarian cancer and ovarian cancer subgroups showed that the number of patients diagnosed with only ovarian cancer with no other primary tumors was 103 out of the total 127 patients diagnosed with ovarian cancer. miR-423-5p expression was increased in 52.4% (54/103), and was decreased in 47.6%(49/103). miR-423-5p expression level was increased in 68.8%(11/16), and was decreased in 31.2% (5/16) in the group diagnosed with both ovarian and breast cancer. miR-423-5p expression was increased in 50% (1/2), and was decreased in 50% (1/2) in patients diagnosed with two different primary cancers as ovarian and endometrium cancer. miR-423-5p expression was found to have increased in all patients 100%(6/6) with a secondary disease with ovarian cancer except breast, and endometrium cancer (Table 2).
miR-664b-5p expression was increased in 35% (36/103), and was decreased in 65% (67/103) of 103 patients out of 127 ovarian cancer patients with only diagnosis of ovarian cancer, and no other primary tumors. miR-664b-5p expression level was found to have increased in 43.8% (7/16), and have decreased in 56.2% (9/16) of patient group diagnosed with both ovarian and breast cancers. miR-664b-5p expression was found to have increased in the half, and was decreased in the half of patients who were diagnosed with both ovarian, and endometrium cancers. miR-664b-5p expression was found to have increased in patients with a secondary primary disease with ovarian cancer except breast, and endometrium cancer (Table 2).
Table 2: The association of miR-423-5p, and miR-664b-5p expression status with the ovarian cancer, ovarian cancer patients subgroups
Diagnosis
|
miR-423-5p
|
miR-664b-5p
|
|
Increased
n(%)
|
Decreased
n(%)
|
Increased
n(%)
|
Decreased
n(%)
|
Total
n(%)
|
Only Ovarian Cancer
|
54(52.4%)
|
49(47.6%)
|
36(35%)
|
67(65%)
|
103(81.1%)
|
Ovarian Cancer + Breast Cancer
|
11(68.8%)
|
5(31.2%)
|
7(43.8%)
|
9(56.2%)
|
16(12.6%)
|
Ovarian Cancer + Endometrium Cancer
|
1(50%)
|
1(50%)
|
0(0%)
|
2(100%)
|
2(1.6%)
|
Ovarian Cancer +
Other Cancer
|
6(100%)
|
0(0%)
|
6(100%)
|
0(0%)
|
6(4.7%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
A statistical significance was detected between the miRAN expressions of both miRNAS in the statistical comparison between the healthy controls, and ovarian cancer patients groups(p:0.000). miR-423-5p expression was found to have increased 2.35 fold in ovarian cancer patients compared with the control group. Some patients with ovarian cancer were also diagnosed with an additional secondary cancer. Accordingly, some patients were diagnosed with ovarian and breast cancer, some patients were diagnosed with ovarian and endometrium cancer, and one group was diagnosed with another secondary cancer in addition to ovarian cancer. The comparison of the ovarian cancer subgroups with the control group showed that miR-423-5p expression level increased 3.65 fold in patients with both ovarian and breast cancer, and increased 4.69 fold in patients with both ovarian cancer and other cancer, however decreased 0.58 fold in patients diagnosed with ovarian and endometrium cancer (Figure 1).
The comparison of all ovarian cancer patients with the control group showed that the miR-664b-5p expression increased 2.47. The comparison of ovarian cancer patient subgroups with the control group for miR-664b-5p showed that the levels increased 1.95 fold in both ovarian and breast cancer patients, and increased in 2.95 fold in ovarian and other cancer patients, however the levels decreased 0.27 fold in patients with both ovarian and endometrium cancer (Figure 2).
Evaluation of miRNA expression status in accordance with the Family History of the patients: The evaluation of the miR-423-5p Expression Status with the family history of patients showed that (Table 3); the expression level decreased in 38.6% (17/44) of ovarian cancer patients with a family history of 2 or more than 2 ovarian and breast cancer, however the expression level increased in 61.4% (27/44). The expression level decreased in 45.8% (38/83), and levels increased in 54.2% (45/83) of ovarian cancer patients with a family history of ovarian and breast cancer less than 2. The expression level decreased in 45.7% (37/81), and increased in 54.3% (44/81) of ovarian cancer patients with a family history of 2 or more than 2 any type of cancer. The expression level decreased in 39% (18/46), and increased in 61% (28/46) in ovarian cancer patients with a history of less than 2 cancers in the family.
The evaluation of miR-664b-5p expression status in accordance with the family history of patients showed that the expression level decreased in 56.8%(25/44), and increased in 43.2%(19/44) of ovarian cancer patients with a history of 2 or more than 2 ovarian and breast cancers in the family. The expression level decreased in 64%(53/83), and increased in 36%(30/83) of ovarian cancer patients with a family history of less than 2 ovarian and breast cancer. The expression level decreased in 53%(43/81), and increased in 47%(38/81) of ovarian cancer patients with a family history of any type of cancer of 2 or more than 2 in the family. The expression level decreased in 76%(35/46), and increased in 24%(11/46) in ovarian cancer patients with a history of less than 2 cancer in the family.
Table 3: The association of miR-423-5p, and miR-664b-5p expression status with family history
Family History
|
miR-423-5p
|
miR-664b-5p
|
|
miRNA Expression Status
|
Increased
|
Decreased
|
Increased
|
Decreased
|
Total
|
History of ovarian and breast cancer in the family
|
n(%)
|
n(%)
|
n(%)
|
n(%)
|
n(%)
|
<2 individuals
|
45(62.5%)
|
38(69%)
|
30(61.2%)
|
53(68%)
|
83(65.3%)
|
>=2 individuals
|
27(37.5%)
|
17(31%)
|
19(38.8%)
|
25(32%)
|
44(34.7%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
History of other cancer in the family
|
|
|
|
|
|
<2 individuals
|
28(38.9%)
|
18(32.7%)
|
11(22.4%)
|
35(44.8%)
|
46(36.2%)
|
>=2 individuals
|
44(61.1%)
|
37(67.3%)
|
38(77.6%)
|
43(55.2%)
|
81(63.8%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Comparison of the miRNA expression levels with the clinical features of patients
The clinical and pathologic stage features of patients, and their association with miR-423-5p and miR-664b-5p expression levels, and their distributions are given in Table 5 (Table 4). The comparison of the clinical stage and miRNA expression levels of the patients showed that miR-423-5p expression levels were decreased in 43.3%(55/127) of patients, and expression levels increased in 56.7%(72/120) of patients (p>0.005). miR-664b-5p expression was found to have decreased in 61.5% (78/127) of the patient, and increased in 38.5%(49/120) (p>0.005).
Table 4: Comparison of the expression status of miR-423-5p, and miR-664b-5p with the clinical and pathologic features
Clinical Characteristics
|
miR-423-5p
|
miR-664b-5p
|
|
miRNA Expression Status
|
Increased
|
Decreased
|
Increased
|
Decreased
|
Total
|
Age [mean 50±10 interval (25-84)]
|
n(%)
|
n(%)
|
n(%)
|
n(%)
|
n(%)
|
<40 age
|
10(13.9%)
|
6(11%)
|
6(12.2%)
|
10(12.8%)
|
16(12.5%)
|
>=40 age
|
62(86.1%)
|
49(89%)
|
43(87.8%)
|
68(87.1%)
|
111(87.5%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Clinical Stage
|
|
|
|
|
|
Stage 1
|
12(17%)
|
10(18%)
|
6(12.2%)
|
16(20.5%)
|
22(17.3%)
|
Stage 2
|
8(11.1%)
|
6(11%)
|
10(20.4%)
|
4(5.2%)
|
14(11%)
|
Stage 3
|
48(66.7%)
|
28(51%)
|
28(57.2%)
|
48(61.5%)
|
76(59.9%)
|
Stage 4
|
4(5.5%)
|
11(20%)
|
5(10.2%)
|
10(12.8%)
|
15(11.8%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Pathological Stage
|
|
|
|
|
|
Stage 1
|
11(15.3%)
|
11(20%)
|
7(14.2%)
|
15(19.2%)
|
22(17.3%)
|
Stage 2
|
11(15.3%)
|
8(14.5%)
|
9(12.5%)
|
10(12.8%)
|
19(15%)
|
Stage 3
|
50(69.4%)
|
36(65.5%)
|
33(67.3%)
|
53(68%)
|
86(67.7%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Histological Gradee
|
|
|
|
|
|
Grade 1
|
11(15.3%)
|
14(25.5%)
|
7(14.3%)
|
18(23%)
|
25(19.7%)
|
Grade 2
|
14(19.4%)
|
10(18.2%)
|
10(20.4%)
|
14(18%)
|
24(18.9%)
|
Grade 3
|
47(65.3%)
|
31(56.3%)
|
32(65.3%)
|
46(59%)
|
78(61.4%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Tumor Size
|
|
|
|
|
|
<2 cm
|
15(21%)
|
13(23.6%)
|
12(24.4%)
|
16(20.5%)
|
28(22%)
|
>=2 cm
|
57(79%)
|
42(76.4%)
|
37(75.6%)
|
62(79.5%)
|
99(78%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Treatment
|
|
|
|
|
|
Surgery
|
|
Yes
|
68(94.4%)
|
54(98.2%)
|
46(93.9%)
|
76(97.4%)
|
122(96%)
|
No
|
4(5.6%)
|
1(1.8%)
|
3(6.1%)
|
2(2.6%)
|
5(4%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Chemotherapy
|
|
Yes
|
66(91.7%)
|
51(92.7%)
|
49(100%)
|
68(87.1%)
|
117(92%)
|
No
|
6(8.3%)
|
4(7.3%)
|
0(0%)
|
10(12.9%)
|
10(8%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Radiotherapy
|
|
Yes
|
10(13.9%)
|
5(9%)
|
6(12.2%)
|
9(11.5%)
|
15(12%)
|
No
|
62(86.1%)
|
50(91%)
|
43(87.8%)
|
69(88.5%)
|
112(88%)
|
Total
|
72(56.7%)
|
55(43.3%)
|
49(38.5%)
|
78(61.5%)
|
127
|
Survival Analysis
The mean survival time of ovarian cancer patients that we investigated in the study was 103 (SD ± 5.7 months) months, and we found that 45(35.4%) patients out of 127 patients that we screened between 2010-2019 died, and 78 (61.4%) survived after the diagnosis, and the condition of 4(3.2%) were unknown.
We performed the Kaplan-Meier analysis for investigating the possible variables that might affect the survival time of miR-423-5p, and miR-664b-5p expression levels . We found no correlation between the gene expression levels and survival time (p>0.005) (Table 5, and Figure 3-4).
Table 5: The association of gene expression levels and survival time (Kaplan-Meier analysis)
miR-423-5p expression
|
Total Number (n)
|
n of Events
|
Censored
|
n
|
Percentage
|
Decreased
|
55
|
32
|
23
|
41.8%
|
Increased
|
72
|
46
|
26
|
36.1%
|
Total
|
127
|
77
|
49
|
38.6%
|
miR-664b-5p expression
|
|
|
|
|
Decreased
|
78
|
44
|
34
|
43.6%
|
Increased
|
49
|
34
|
15
|
30.6%
|
Total
|
127
|
78
|
49
|
38.6%
|
ROC Analysis
We performed the Receiver Operator Characteristics (ROC) analysis for demonstrating the differentiation strenght of ovarian cancer patient, and healthy control group and also the diagnostic performance of miR-423-5p, and miR-664b-5p molecules as the candidate bioindicators (Figure 5). Table 6, and Figure 5 demonstrate the identified ROC-AUC values for each miRNA, and 95% CI (confidence interval). These results showed that the diagnostic strenght of miR-664b-5p miRNA variables were statistically significant in the diagnosis of ovarian cancer patients (p<0.005), however the diagnostic strenght of miR-423-5p was found statistically unsignificant (p>0.005).
Table 6: ROC-AUC values of miRNAs in differentiation of ovarian cancer patients, and healthy groups
miRNAs
|
ROC-AUC
|
95% CI
|
p value
|
miR-423-5p
|
0.531
|
0.465 - 0.606
|
.429
|
miR-664b-5p
|
0.374
|
0.301 - 0.448
|
.001
|
The analysis of the change of CA-125 value in ovarian cancer patients
We performed the“Spearman’s correlation analysis” for identifying the possible linear association between the numerical measurements of miRNA expression levels and the Ca-125 values obtained from the clinical data of the ovarian cancer patients. We found no correlation and statistical significance between the expression levels of 423-5p, and miR-664b-5p with the Ca-125 numerical variables in the beginning of treatment, and during treatment (p>0.05) (Table 7).
Table 7: The correlation statistics of miRNA expression levels and CA-125 variables
p values
|
Initial CA-125
|
Final CA-125
|
miR-423-5p
|
0.278
|
0.742
|
miR-664b-5p
|
0.313
|
0.362
|