Skip to main content

Table 5 Association between survival patterns predicted by the 19-miRNA prognostic classifier with clinico-pathologic characteristics or molecular subtypes in the TCGA cohort

From: Identification of common oncogenic and early developmental pathways in the ovarian carcinomas controlling by distinct prognostically significant microRNA subsets

Characteristic Subcategory Low risk Intermediate risk High risk Weighted Kappa
(n =231 ) (n =215 ) (n =28)
Number % Number % Number % Correlation coefficient P-value&
Age at initial pathologic diagnosis age < 50 36 19% 53 20% 4 17% 0.02 0.55
50 <= age <=70 110 57% 143 55% 8 35%   
age > 70 42 22% 58 22% 9 39%   
*others/no information 4 2% 5 2% 2 9%   
Stage Stage I-II 22 11% 13 5% 0 0% -0.06 0.026
Stage III 145 76% 203 78% 18 78%   
Stage IV 25 13% 42 16% 5 22%   
*others/no information 0 0% 1 0% 0 0%   
Grade Grade 1 30 16% 28 11% 3 13% -0.03 0.17
Grade 2 157 82% 228 88% 18 78%   
Grade 3 1 1% 0 0% 0 0%   
*others/no information 4 2% 3 1% 2 9%   
Tumor residual disease No_Macroscopic_disease 52 27% 43 17% 0 0% 0.15 5.58E-4
1-20 mm 83 43% 144 56% 11 48%   
>20_mm 27 14% 42 16% 7 30%   
*others/no information 30 16% 30 12% 5 22%   
Primary therapy outcome success Complete response 130 68% 133 51% 5 22% 0.15 1.12E-3
Partial response/stable disease 23 12% 49 19% 7 30%   
Progressive disease 11 6% 23 9% 1 4%   
*others/no information 28 15% 54 21% 10 43%   
^TCGA samples by molecular subtypes Proliferative 61 32% 60 23% 3 13% 0.23 4.23E-8
Immunoreactive/Differentiated 102 53% 118 46% 6 26%   
Mesenchymal 16 8% 71 27% 13 57%   
*others/no information 13 7% 10 4% 1 4%   
^TCGA samples by miRNA clustering C1 54 28% 59 23% 0 0% -0.07 0.034
C2 12 6% 89 34% 20 87%   
C3 113 59% 101 39% 2 9%   
*others/no information 13 7% 10 4% 1 4%   
let-7b miRNA Classifier Low Risk 119 62% 40 15% 0 0% 0.43 <1E-16
Intermediate Risk 72 38% 196 76% 9 39%   
High Risk 1 1% 23 9% 14 61%   
  1. The measure of agreement was calculated using weighted kappa correlation coefficient, and the significance of the agreement was estimated by kappa2* function in R package irr
  2. Note: *These subcategories are not included in the calculation of the Kappa correlation coefficient
  3. ^Sample groupings were provided by the authors of the TCGA study [14]
  4. &Significant associations are highlighted in boldface