Open Access

Quantitative dynamic contrast-enhanced parameters and intravoxel incoherent motion facilitate the prediction of TP53 status and risk stratification of early-stage endometrial carcinoma


Cite

FIGURE 1.

Flowchart of the present study.EC = endometrial carcinoma
Flowchart of the present study.EC = endometrial carcinoma

FIGURE 2.

(A–I) A 53-year-old woman with low-risk endometrial carcinoma (EC) (arrowheads, endometrioid type, grade 2, stage IA, lymphovascular space invasion (LVSI) negative, and TP53-wild). (J–R) A 56-year-old woman with non-low-risk (intermediate) EC (arrowheads, endometrioid type, grade 1, stage IB, LVSI negative, and TP53-mutant). (A, J) Sagittal T2-weighted imaging maps; (B, K) Oblique axial pseudo colored maps of volume transfer constant (Ktrans); (C, L) Oblique axial pseudo colored maps of rate transfer constant (Kep); (D, M) Oblique axial pseudo colored maps of the volume of extravascular extracellular space per unit volume of tissue (Ve); (E, N) Oblique axial colored maps of true diffusion coefficient (D); (F, O) Oblique axial colored maps of pseudo-diffusion coefficient (D*); (G, P) Oblique axial colored maps of microvascular volume fraction (f), and (H, Q) Histopathological images (magnification = 100), and (I, R) Immunohistochemical image (magnification = 200).
(A–I) A 53-year-old woman with low-risk endometrial carcinoma (EC) (arrowheads, endometrioid type, grade 2, stage IA, lymphovascular space invasion (LVSI) negative, and TP53-wild). (J–R) A 56-year-old woman with non-low-risk (intermediate) EC (arrowheads, endometrioid type, grade 1, stage IB, LVSI negative, and TP53-mutant). (A, J) Sagittal T2-weighted imaging maps; (B, K) Oblique axial pseudo colored maps of volume transfer constant (Ktrans); (C, L) Oblique axial pseudo colored maps of rate transfer constant (Kep); (D, M) Oblique axial pseudo colored maps of the volume of extravascular extracellular space per unit volume of tissue (Ve); (E, N) Oblique axial colored maps of true diffusion coefficient (D); (F, O) Oblique axial colored maps of pseudo-diffusion coefficient (D*); (G, P) Oblique axial colored maps of microvascular volume fraction (f), and (H, Q) Histopathological images (magnification = 100), and (I, R) Immunohistochemical image (magnification = 200).

FIGURE 3.

Plots show individual data points, averages, and standard deviations of true diffusion coefficient (D) (A, G), pseudo-diffusion coefficient (D*) (B, H), microvascular volume fraction (f) (C, I), volume transfer constant (Ktrans) (D, J), the volume of extravascular extracellular space per unit volume of tissue (Ve) (E, K), and rate transfer constant (Kep) (F, L) in low-risk and non-low-risk groups (A–F), TP53-mutant and TP53-wild groups (G–L). Individual points are averages of values calculated by 2 readers. *P < 0.05, **P < 0.01, ***P < 0.001, and ● P > 0.005.
Plots show individual data points, averages, and standard deviations of true diffusion coefficient (D) (A, G), pseudo-diffusion coefficient (D*) (B, H), microvascular volume fraction (f) (C, I), volume transfer constant (Ktrans) (D, J), the volume of extravascular extracellular space per unit volume of tissue (Ve) (E, K), and rate transfer constant (Kep) (F, L) in low-risk and non-low-risk groups (A–F), TP53-mutant and TP53-wild groups (G–L). Individual points are averages of values calculated by 2 readers. *P < 0.05, **P < 0.01, ***P < 0.001, and ● P > 0.005.

FIGURE 4.

Receiver operating characteristic (ROC) curves, (A) shows each parameter and the combination of independent predictors for differentiation of TP53-mutant and TP53-wild early-stage endometrial carcinoma (EC); (B) shows each parameter and the combination of independent predictors for differentiation of low-risk and non-low-risk early-stage EC.
Receiver operating characteristic (ROC) curves, (A) shows each parameter and the combination of independent predictors for differentiation of TP53-mutant and TP53-wild early-stage endometrial carcinoma (EC); (B) shows each parameter and the combination of independent predictors for differentiation of low-risk and non-low-risk early-stage EC.

FIGURE 5.

In the prediction of TP53 status, receiver operating characteristic curves (A), calibration curves (B), and decision curve analysis (C) of the validation model.
In the prediction of TP53 status, receiver operating characteristic curves (A), calibration curves (B), and decision curve analysis (C) of the validation model.

FIGURE 6.

In the prediction of risk stratification, receiver operating characteristic curves (A), calibration curves (B), and decision curve analysis (C) of the validation model.
In the prediction of risk stratification, receiver operating characteristic curves (A), calibration curves (B), and decision curve analysis (C) of the validation model.

Clinicopathologic features of the patients

Variable Data
Age (mean ± SD) (years) 54.00 ± 7.91
Maximum diameter (mean ± SD) (mm) 25.10 (13.76, 42.58)
FIGO stage n (%)
  IA 44 (59.46)
  IB 30 (40.54)
Histologic subtype n (%)
  Adenocarcinoma 67 (90.54)
  Non-adenocarcinoma 7 (9.46)
    Clear-cell 3 (4.06)
    Undifferentiated carcinoma 2 (2.70)
    Carcinosarcoma 2 (2.70)
Lymphovascular space invasion n (%)
  Positive 10 (6.76)
  Negative 64 (93.24)
Histologic grade n (%)
  Grade 1 54 (72.98)
  Grade 2 10 (13.51)
  Grade 3 10 (13.51)
Risk stratification n (%)
  Low 44 (59.46)
  Intermediate 20 (27.03)
  High-intermediate 0 (0.00)
  High 10 (13.51)
TP53 expression
  Mutant 21 (28.38)
  Wild 25 (33.78)
  No result 28 (37.84)

Comparison of different parameters

Parameters D (×10−3mm2/s) D* (×10−3mm2/s) f (%) Ktrans (min−1) Ve Kep(min−1)
Risk stratification
High-risk (n = 10) 0.63 (0.40, 0.73) 58.40 (40.10, 88.73) 1.64 ± 0.60 0.35 (0.15, 0.43) 0.30 ± 0.07 1.23 (0.48, 1.54)
High-intermediate-risk (n = 0) / / / / / /
Intermediate-risk (n = 20) 0.55 (0.40, 0.81) 52.00 (26.88, 74.33) 1.74 ± 0.96 0.37 (0.29, 0.47) 0.33 ± 0.14 1.24 (0.82, 1.98)
Low-risk (n = 44) 0.86 (0.64, 1.16) 44.35 (21.93, 95.33) 2.43 ± 1.08 0.61 (0.43, 1.14) 0.58 ± 0.25 1.53 (0.79, 2.21)
P-value 0.033 a 0.464 a 0.012 a < 0.001 a < 0.001 a 0.191 a
P-value (High vs Intermediate) 0.880 b 0.248 b 0.735 c 0.397 b 0.532 c 0.307 b
P-value (High vs Low) 0.009 b 0.238 b 0.004 c < 0.001 b 0.001 c 0.099 b
P-value (Intermediate vs Low) 0.001 b 0.937 b 0.014 c < 0.001 b < 0.001 c 0.582
Low-risk (n = 44) 0.86 (0.64, 1.16) 44.35 (21.93, 95.33) 2.43 ± 1.08 0.61 (0.43, 1.14) 0.58 ± 0.25 1.53 (0.79, 2.21)
Non-low-risk (High + Intermediate, n = 30) 0.58 (0.40, 0.77) 55.25 (34.63, 72.78) 1.71 ± 0.84 0.37 (0.28, 0.45) 0.32 ± 0.12 1.23 (0.82, 1.87)
z/t value − 3.793 −0.523 3.234 −5.109 5.304 −1.233
P-value < 0.001 b 0.601 b 0.002 c < 0.001 b < 0.001 c 0.218 b
TP53 expression
  Mutant (n = 21) 0.72 ± 0.31 43.70 (16.30, 90.75) 2.30 ± 1.09 0.67 (0.41, 1.14) 0.32 (0.25, 0.91) 1.67 (1.17, 2.09)
  Wild (n = 25) 0.91 ± 0.29 50.60 (26.90, 82.75) 2.20 ± 1.02 0.43 (0.37, 0.49) 0.49 (0.36, 0.76) 0.90 (0.58, 1.55)
Z/t value −2.155 −0.518 0.321 −2.073 −0.783 −3.165
P-value 0.037 c 0.604 b 0.750 c 0.038 b 0.434 b 0.002 b

Logistic regression analyses

Parameters Univariate Analyses P-value Multivariate Analyses P-value


OR for 1 SD (95% CI) OR for 1 SD (95% CI)
Low vs non-low risk
Age (year) 1.462 (0.894–2.388) 0.130 / /
Tumor size (mm) 1.055 (1.003–1.110) 0.038 1.083 (0.979–1.197) 0.123
TP53 mutant 1.506 (0.407–5.578) 0.540 / /
D (×10−3mm2/s) 0.089 (0.021–0.373) 0.001 0.144 (0.015–1.334) 0.088
D* (×10−3mm2/s) 0.867 (0.533–1.412) 0.567 / /
f (%) 0.419 (0.226–0.776) 0.006 0.292 (0.093–0.921) 0.036
Ktrans (min−1) 0.009 (0.001–0.153) 0.001 0.001 (0.000–0.089) 0.003
Ve 0.173 (0.069–0.432) < 0.001 0.130 (0.022–0.766) 0.024
Kep (min−1) 0.642 (0.367–1.126) 0.122 / /
TP53 mutant vs wild
Age (year) 0.855 (0.465–1.548) 0.605 / /
Tumor size (mm) 1.175 (0.649–2.127) 0.594 / /
Subtype 77.708 (0.001–100.5) 0.999 / /
Grade 2.099 (0.957–4.602) 0.064 1.961 (0.816–4.717) 0.132
Risk stratification 1.506 (0.407–5.578) 0.540 / /
FIGO stage 1.360 (0.739–2.505) 0.323 / /
LVSI 802.578 (0.001–1150.5) 0.999 / /
D (×10−3mm2/s) 2.063 (1.016–4.191) 0.045 8.274 (2.066–33.136) 0.003
D* (×10−3mm2/s) 1.020 (0.567–1.835) 0.948 / /
f (%) 0.906 (0.504–1.629) 0.742 / /
Ktrans (min−1) 0.487 (0.236–1.003) 0.051 0.155 (0.034–0.710) 0.016
Ve 1.008 (0.560–1.812) 0.979 / /
Kep (min−1) 0.501 (0.244–1.032) 0.061 1.172 (0.425–3.234) 0.759

Imaging protocol parameters

Parameters T1WI T2WI DWI IVIM DCE-MRI
Sequence 2D-FSE 2D-FSE 2D-SS-EPI 2D-SS-EPI 3D-LAVA
Orientation Oblique Axial Oblique Axial Oblique Axial Oblique Axial Oblique Axial
TR/TE (ms) 659/12.3 6000/95 3708/74.3 2000/80.7 3.5/1.7
FOV (cm2) 40 × 40 40 × 40 40 × 40 40 × 40 36 × 36
Matrix 288 × 192 320 × 320 96 × 128 128 × 192 288 × 192
Flip angle (°) 160 160 90 90 15
Slice thickness (mm) 6 6 6 6 6
No. of sections 20 20 20 Based on lesion's size 26
NEX 1 1 1, 4 1, 1, 1, 1, 1, 1, 2, 4, 4, 6 0.73
Fat suppression / STIR STIR STIR FLEX
b-values (s/mm2) / / 0, 800 0, 20, 40, 80, 160, 200, 400, 600, 800, 1000 /
Respiratory compensation Free Free Free Free Free
Scan time 1 min 56 s 48 s 1 min 04 s 3~6min 6 min 08 s (40 phases)

Predictive performance of different parameters

Parameters AUC (95% CI) P-value Cutoff Sensitivity Specificity Comparison with combined diagnosis
Low vs non-low risk
D (×10−3mm2/s) 0.761 (0.648–0.853) < 0.001 0.691 73.33% 72.73% Z = 3.113, P = 0.002
D* (×10−3mm2/s) 0.536 (0.416–0.653) 0.598 / / / /
f (%) 0.688 (0.569–0.790) 0.003 1.240 36.67% 93.18% Z = 4.317, P < 0.001
Ktrans (min−1) 0.852 (0.750–0.924) < 0.001 0.487 90.00% 68.18% Z = 2.713, P = 0.007
Ve 0.808 (0.700–0.890) < 0.001 0.401 83.33% 70.45% Z = 3.175, P = 0.002
Kep (min−1) 0.585 (0.652–0.849) 0.204 / / / /
Combined diagnosis 1 0.947 (0.869–0.986) < 0.001 / 83.33% 93.18% /
TP53 mutant vs wild
D (×10−3mm2/s) 0.694 (0.541–0.821) 0.019 0.605 92.00% 47.62% Z = 2.169, P = 0.030
D* (×10−3mm2/s) 0.545 (0.391–0.692) 0.498 / / / /
f (%) 0.535 (0.382–0.648) 0.388 / / / /
Ktrans (min−1) 0.679 (0.525–0.809) 0.036 0.499 80.00% 61.90% Z = 2.572, P = 0.010
Ve 0.568 (0.413–0.713) 0.675 / / / /
Kep (min−1) 0.773 (0.626–0.884) < 0.001 1.557 80.00% 66.67% Z = 1.272, P = 0.203
Combined diagnosis 2 0.867 (0.734–0.949) < 0.001 / 92.00% 80.95% /
eISSN:
1581-3207
Language:
English
Publication timeframe:
4 times per year
Journal Subjects:
Medicine, Clinical Medicine, Internal Medicine, Haematology, Oncology, Radiology