Identification of CT radiomic features robust to acquisition and segmentation variations for improved prediction of radiotherapy-treated lung cancer patient recurrence.

The primary objective of the present study was to identify a subset of radiomic features extracted from primary tumor imaged by computed tomography of early-stage non-small cell lung cancer patients, which remain unaffected by variations in segmentation quality and in computed tomography image acquisition protocol. The robustness of these features to segmentation variations was assessed by analyzing the correlation of feature values extracted from lesion volumes delineated by two annotators. The robustness to variations in acquisition protocol was evaluated by examining the correlation of features extracted from high-dose and low-dose computed tomography scans, both of which were acquired for each patient as part of the stereotactic body radiotherapy planning process. Among 106 radiomic features considered, 21 were identified as robust. An analysis including univariate and multivariate assessments was subsequently conducted to estimate the predictive performance of these robust features on the outcome of early-stage non-small cell lung cancer patients treated with stereotactic body radiation therapy. The univariate predictive analysis revealed that robust features demonstrated superior predictive potential compared to non-robust features. The multivariate analysis indicated that linear regression models built with robust features displayed greater generalization capabilities by outperforming other models in predicting the outcomes of an external validation dataset.

Predictive clinical and dosimetric parameters for risk of relapse in early-stage non-small cell lung cancer treated by SBRT: A large single institution experience.

Purpose: To evaluate the impact of dosimetric parameters on efficacy of stereotactic body radiation therapy (SBRT) in early-stage non-small cell lung cancer (ES-NSCLC), using Hypofractionated Treatment Effects in the Clinic (HyTEC) reporting standards. Methods: From April 2010 to December 2020, 497 patients who received SBRT for ES-NSCLC at the University Hospital of Liège were retrospectively enrolled. A total dose of 40 to 60 Gy in 3-5 fractions (72-180 Gy biologically effective dose with an α/ß ratio of 10 (BED10)) was prescribed to the 80 % isodose line of the PTV. Potential clinical and dosimetric predictors of recurrence, overall survival (OS) and disease specific survival (DSS) were evaluated using univariate and multivariate analyses. Results: After a median follow-up of 32 months (range 3-143 months), the local control and disease-free survival (DFS) rates at 3 years were 91 % (95 % CI: 90 %-93 %) and 75 % (95 % CI: 73 %-77 %), respectively. The median OS was 41.6 months and the median DSS was not reached. On multivariate analysis, a higher gross tumor volume (GTV) Dmax (BED10) (cut-off 198 Gy) and a larger percent of the GTV receiving ≥110 % of the prescribed dose were predictive of a better local control, only GTV volume was correlated with DSS and no parameter was correlated with OS and regional or distant recurrences. Conclusion: Lung SBRT for ES-NSCLC in 3 to 5 fractions resulted in high local control rates. A higher percent of GTV receiving ≥110 % of the prescribed dose and a higher GTV Dmax (BED10) seem to allow a better local control.

Multicentric development and evaluation of [18F]FDG PET/CT and CT radiomic models to predict regional and/or distant recurrence in early-stage non-small cell lung cancer treated by stereotactic body radiation therapy.

PURPOSE: To develop machine learning models to predict regional and/or distant recurrence in patients with early-stage non-small cell lung cancer (ES-NSCLC) after stereotactic body radiation therapy (SBRT) using [18F]FDG PET/CT and CT radiomics combined with clinical and dosimetric parameters. METHODS: We retrospectively collected 464 patients (60% for training and 40% for testing) from University Hospital of Liège and 63 patients from University Hospital of Brest (external testing set) with ES-NSCLC treated with SBRT between 2010 and 2020 and who had undergone pretreatment [18F]FDG PET/CT and planning CT. Radiomic features were extracted using the PyRadiomics toolbox®. The ComBat harmonization method was applied to reduce the batch effect between centers. Clinical, radiomic, and combined models were trained and tested using a neural network approach to predict regional and/or distant recurrence. RESULTS: In the training (n = 273) and testing sets (n = 191 and n = 63), the clinical model achieved moderate performances to predict regional and/or distant recurrence with C-statistics from 0.53 to 0.59 (95% CI, 0.41, 0.67). The radiomic (original_firstorder_Entropy, original_gldm_LowGrayLevelEmphasis and original_glcm_DifferenceAverage) model achieved higher predictive ability in the training set and kept the same performance in the testing sets, with C-statistics from 0.70 to 0.78 (95% CI, 0.63, 0.88) while the combined model performs moderately well with C-statistics from 0.50 to 0.62 (95% CI, 0.37, 0.69). CONCLUSION: Radiomic features extracted from pre-SBRT analog and digital [18F]FDG PET/CT outperform clinical parameters in the prediction of regional and/or distant recurrence and to discuss an adjuvant systemic treatment in ES-NSCLC. Prospective validation of our models should now be carried out.

Incidence, risk factors, and CT characteristics of radiation recall pneumonitis induced by immune checkpoint inhibitor in lung cancer.

BACKGROUND AND PURPOSE: Radiation recall pneumonitis (RRP) is a delayed radiation-induced lung toxicity triggered by systemic agents, typically anticancer drugs. Immune checkpoint inhibitors (ICIs) have recently been identified as potential causal agents of RRP but its real incidence and potential risk factors remain unknown. MATERIALS AND METHODS: Medical records and CTs of patients treated with programmed death 1 (PD-1) or programmed death ligand 1 (PD-L1) inhibitors for advanced lung cancer between 2014 and 2019 at our tertiary center, and who had a previous history of lung irradiation were retrospectively analyzed. We identified RRP as lung CT modifications occurring in the irradiation field >6 months after conventionally fractionated radiotherapy completion and >1 year after stereotactic body radiation therapy. Clinical and dosimetric data were analyzed to identify potential risk factors for RRP. RESULTS: Among 348 patients treated with ICIs, data from 80 eligible patients were analyzed (median age, 69 years [interquartile range, 11]; 45 men). Fifteen patients (18.8%) presented with RRP. Median time between end of radiotherapy and RRP was 450 days (range, 231-1859). No risk factor was significantly associated with RRP. ICI-related pneumonitis was associated with RRP in 33.3% of cases (p = 0.0021), developing either concomitantly or after RRP. Incidence of grade ≥ 3 pneumonitis in the RRP population was 13.3 %. CONCLUSION: We demonstrated a high incidence of RRP (18.8%) in our population of previously irradiated patients treated with ICIs for lung cancer. We identified no risk factors for RRP, but an association was noted between RRP and ICI-related pneumonitis.

Stereotactic robotic body radiotherapy for patients with oligorecurrent pulmonary metastases.

BACKGROUND: Our aim is to report treatment efficacy and toxicity of patients treated by robotic (Cyberknife®) stereotactic body radiotherapy (SBRT) for oligorecurrent lung metastases (ORLM). Additionally we wanted to evaluate influence of tumor, patient and treatment related parameters on local control (LC), lung and distant progression free- (lung PFS/Di-PFS) and overall survival (OS). METHODS: Consecutive patients with up to 5 ORLM (confirmed by FDG PET/CT) were included in this study. Intended dose was 60Gy in 3 fractions (prescribed to the 80% isodose volume). Patients were followed at regular intervals and tumor control and toxicity was prospectively scored. Tumor, patient and treatment data were analysed using competing risk- and Cox regression. RESULTS: Between May 2010 and March 2016, 104 patients with 132 lesions were irradiated from primary lung carcinoma (47%), gastro-intestinal (34%) and mixed primary histologies (19%). The mean tumor volume was 7.9 cc. After a median follow up of 22 months, the 1, 2 and 3 year LC rate (per lesion) was 89.3, 80.0 and 77.8% respectively. The corresponding (per patient) 1, 2 and 3 years lung PFS were 66.3, 50.0, 42.6%, Di-PFS were 80.5, 64.4, 60.6% and OS rates were 92.2, 80.9 and 72.0% respectively. On univariable analysis, gastro-intestinal (GI) as primary tumor site showed a significant superior local control versus the other primary tumor sites. For OS, significant variables were primary histology and primary tumor site with a superior OS for patients with metastases of primary GI origin. LC was significantly affected by the tumor volume, physical and biologically effective dose coverage. Significant variables in multivariable analysis were BED prescription dose for LC and GI as primary site for OS. The vast majority of patients developed no toxicity or grade 1 acute and late toxicity. Acute and late grade 3 radiation pneumonitis (RP) was observed in 1 and 2 patients respectively. One patient with a centrally located lesion developed grade 4 RP and died due to possible RT-induced pulmonary hemorrhage. CONCLUSIONS: SBRT is a highly effective local therapy for oligorecurrent lung metastases and could achieve long term survival in patients with favourable prognostic features.