Cite

Figure 1

Baseline radiomics features of primary tumours – Receiver operating characteristic curve (ROC) analysis. For each radiomics feature, the area under the ROC curve (AUC) with the corresponding 95% confidence interval (CI) is reported. AUC of 0.8 or above indicates a high level of predictive power, while an AUC of 0.6 or less indicates poor level of predictive power.
Baseline radiomics features of primary tumours – Receiver operating characteristic curve (ROC) analysis. For each radiomics feature, the area under the ROC curve (AUC) with the corresponding 95% confidence interval (CI) is reported. AUC of 0.8 or above indicates a high level of predictive power, while an AUC of 0.6 or less indicates poor level of predictive power.

Figure 2

Kaplan-Meier diagram – Small Run Emphasis (SRE). Blue: patients with SRE ≥ SREmedian, yellow: patients with SRE < SREmedian. The reported p-value is the result of log-rank test.
Kaplan-Meier diagram – Small Run Emphasis (SRE). Blue: patients with SRE ≥ SREmedian, yellow: patients with SRE < SREmedian. The reported p-value is the result of log-rank test.

Figure 3

Receiver operating characteristic (ROC) curve analysis. Blue: baseline iRADIOMICS multivariate logistic model (independent variables: Small Run Emphasis [SRE], Difference Entropy), yellow: baseline iRADIOMICS univariate logistic model (independent variable: SRE), grey: month 1 iRECIST univariate logistic model (independent variable: iRECIST response category), red: month 4 iRECIST univariate logistic model (independent variable: iRECIST response category). For each model, area under curve (AUC) and 95% confidence interval (CI) are reported.
Receiver operating characteristic (ROC) curve analysis. Blue: baseline iRADIOMICS multivariate logistic model (independent variables: Small Run Emphasis [SRE], Difference Entropy), yellow: baseline iRADIOMICS univariate logistic model (independent variable: SRE), grey: month 1 iRECIST univariate logistic model (independent variable: iRECIST response category), red: month 4 iRECIST univariate logistic model (independent variable: iRECIST response category). For each model, area under curve (AUC) and 95% confidence interval (CI) are reported.

Patient demographic and clinical data. The data is presented for all patients, responders (overall survival [OS] > 14.9 months), and non-responders (OS < 14.9 months). The reported p-value is the result of Mann-Whitney U-test (MWU) (continuous variables) and Fisher exact test (categorical variables) comparing differences between responders and non-responders

CharacteristicAll patients median (range)Responders (OS > 14.9 months) median (range)Non-responders (OS < 14.9 months) median (range)p-value
Number of patients301614
Age [years]65 (46–77)67 (48–76)61 (46–77)0.298
PD-L1 TPS [%]75 (3–100)77.5 (3–100)75 (10–100)0.933
Sex0.715
    Female1596
    Male1578
Histology0.532
    Adenocarcinoma1789
    Squamous cell carcinoma844
    Other541
Smoking status0.672
    Never101
    Former > 3 years ago1275
    Former < 3 years ago532
    Until current disease835
    Current smoker431
ECOG PS0.162
    0826
    118126
    2422
Line of treatment (immunotherapy)0.096
    1st15105
    2nd1349
    3rd220
Palliative RT during treatment0.657
    No241212
    Yes642

Baseline radiomics features of primary tumours – Mann-Whitney U-test (MWU) and receiver operating characteristic (ROC) curve analysis. Patients were dichotomized into 2 groups: responders (OS > 14.9 months) and non-responders (OS < 14.9 months). For each radiomics feature median value, range, p-value of MWU, and the area under the ROC curve (AUC) with the corresponding 95% confidence interval (CI), are reported. See also Figure 1

FeatureResponders (OS > 14.9 months) median (range)Non-responders (OS < 14.9 months) median (range)p-valueAUC (95% CI)
Volume [cm3]27.9 (2.64–351)44.4 (7.81–792)0.0980.69 (0.49–0.89)
SUVmax [g/ml]20.6 (5.21–32.1)15.6 (9.54–37.0)0.1850.65 (0.43–0.87)
Sum entropy3.69 (3.53–3.77)3.7 (3.54–3.76)0.3870.60 (0.38–0.82)
Entropy-GLCM4.07 (3.99–4.15)4.11 (4.03–4.14)0.0460.72 (0.52–0.92)
Difference entropy2.98 (2.74–3.07)2.89 (2.74–3.06)0.0800.70 (0.49–0.90)
Small Run Emphasis (SRE)0.0382 (0.00962–0.0615)0.0163 (0.00854–0.0303)0.0010.85 (0.69–1.00)

Baseline radiomics features of primary tumours – univariate and multivariate Cox proportional hazards regression analysis (Cox PH). For each radiomics feature, the hazard ratio (HR), corresponding 95% confidence interval (CI), and p-value of univariate analysis are reported. The 2-variable multivariate regression model was chosen based on the Akaike information criterion (AIC). In order to achieve comparable HRs, all radiomics features were normalized into z-scores

FeatureUnivariate HR (95% CI)Univariate p-valueMultivariate HR (95% CI)Multivariate p-value
Volume1.6 (1.1–2.4)0.015
SUVmax0.77 (0.46–1.3)0.320
Sum Entropy0.96 (0.60–1.5)0.860
Entropy-GLCM1.4 (0.82–2.3)0.230
Difference entropy0.62 (0.40–0.97)0.0370.54 (0.31–0.93)0.026
Small Run Emphasis (SRE)0.46 (0.26–0.81)0.0070.39 (0.20–0.76)0.006
eISSN:
1581-3207
Language:
English
Publication timeframe:
4 times per year
Journal Subjects:
Medicine, Clinical Medicine, Internal Medicine, Haematology, Oncology, Radiology