Dataset Information

Using machine learning to predict radiation pneumonitis in patients with stage I non-small cell lung cancer treated with stereotactic body radiation therapy.

ABSTRACT: To develop a patient-specific 'big data' clinical decision tool to predict pneumonitis in stage I non-small cell lung cancer (NSCLC) patients after stereotactic body radiation therapy (SBRT). 61 features were recorded for 201 consecutive patients with stage I NSCLC treated with SBRT, in whom 8 (4.0%) developed radiation pneumonitis. Pneumonitis thresholds were found for each feature individually using decision stumps. The performance of three different algorithms (Decision Trees, Random Forests, RUSBoost) was evaluated. Learning curves were developed and the training error analyzed and compared to the testing error in order to evaluate the factors needed to obtain a cross-validated error smaller than 0.1. These included the addition of new features, increasing the complexity of the algorithm and enlarging the sample size and number of events. In the univariate analysis, the most important feature selected was the diffusion capacity of the lung for carbon monoxide (DLCO adj%). On multivariate analysis, the three most important features selected were the dose to 15 cc of the heart, dose to 4 cc of the trachea or bronchus, and race. Higher accuracy could be achieved if the RUSBoost algorithm was used with regularization. To predict radiation pneumonitis within an error smaller than 10%, we estimate that a sample size of 800 patients is required. Clinically relevant thresholds that put patients at risk of developing radiation pneumonitis were determined in a cohort of 201 stage I NSCLC patients treated with SBRT. The consistency of these thresholds can provide radiation oncologists with an estimate of their reliability and may inform treatment planning and patient counseling. The accuracy of the classification is limited by the number of patients in the study and not by the features gathered or the complexity of the algorithm.

SUBMITTER: Valdes G

PROVIDER: S-EPMC5491385 | biostudies-other | 2016 Aug

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:BackgroundRadiation pneumonitis is a critical pulmonary toxicity after irradiation of the lung. Macrolides including clarithromycin (CAM) are antibiotics. They also have immunomodulatory properties and are used to treat respiratory inflammatory diseases. Radiation pneumonitis has similar pathology to them. Adverse reactions to macrolides are few and self-limited. We thus administered CAM to patients with high-risk factors for radiation pneumonitis, and retrospectively investigated whether CAM mitigated radiation pneumonitis following stereotactic body radiotherapy (SBRT).MethodsAmong consecutive patients treated with SBRT, we retrospectively examined lung cancer patients treated with a total dose of 40-60 Gy in 5-10 fractions and followed ≥6 months. Since January 2014, CAM has been administered in patients with pretreatment predictable radiation pneumonitis high-risk factors, including idiopathic interstitial pneumonias (IIPs), and elevated Krebs von den Lungen-6 (KL-6) and/or surfactant protein D (SP-D), and in patients developing early onset radiation pneumonitis.ResultsFive hundred and eighty eligible patients were identified and divided into 445 patients during the non-CAM-administration era (non-CAM-era) (before December 2013) and 136 patients during the CAM-administration era (CAM-era) (after January 2014). Median follow-up durations were 38.0 and 13.9 months, respectively. The rates of radiation pneumonitis ≥ grade 2 and ≥ grade 3 were significantly lower in CAM-era (grade ≥2, 16% vs. 9.6%, P=0.047; grade ≥3, 3.8% vs. 0.73%, P=0.037). For patients with the pretreatment predictable high-risk factors, the rate of radiation pneumonitis ≥ grade 3 was significantly lower, and that of grade ≥2 had a lower tendency (grade ≥3, 7.2% vs. 0%, P=0.011; grade ≥2, 21% vs. 9.6%, P=0.061). For patients developing early onset radiation pneumonitis, the rate of radiation pneumonitis ≥ grade 3 was also significantly lower (23% vs. 0%, P<0.05). Multivariate analysis revealed that dose-volumetric factor, the pretreatment predictable high-risk factors and non-CAM-administration era were significantly associated with or trended toward radiation pneumonitis ≥ grade 2 and ≥ grade 3.ConclusionsCAM mitigated radiation pneumonitis following SBRT. The efficacy of CAM should be conﬁrmed in prospective studies.

Project description:Objective The study aims to establish and validate an effective CT-based radiation pneumonitis (RP) prediction model using the multiomics method of radiomics and EQD2-based dosiomics. Materials and Methods The study performed a retrospective analysis on 91 nonsmall cell lung cancer patients who received radiotherapy from 2019 to 2021 in our hospital. The patients with RP grade ≥1 were labeled as 1, and those with RP grade < 1 were labeled as 0. The whole lung excluding clinical target volume (lung-CTV) was used as the region of interest (ROI). The radiomic and dosiomic features were extracted from the lung-CTV area's image and dose distribution. Besides, the equivalent dose of the 2 Gy fractionated radiation (EQD2) model was used to convert the physical dose to the isoeffect dose, and then, the EQD2-based dosiomic (eqd-dosiomic) features were extracted from the isoeffect dose distribution. Four machine learning (ML) models, including DVH, radiomics combined with DVH (radio + DVH), radiomics combined with dosiomics (radio + dose), and radiomics combined with eqd-dosiomics (radio + eqdose), were established to construct the prediction model via eleven different classifiers. The fivefold cross-validation was used to complete the classification experiment. The area under the curve (AUC) of the receiver operating characteristics (ROC), accuracy, precision, recall, and F1-score were calculated to assess the performance level of the prediction models. Results Compared with the DVH, radio + DVH, and radio + dose model, the value of the training AUC, accuracy, and F1-score of radio + eqdose was higher, and the difference was statistically significant (p < 0.05). Besides, the average value of the precision and recall of radio + eqdose was higher, but the difference was not statistically significant (p > 0.05). Conclusion The performance of using the ML-based multiomics method of radiomics and eqd-dosiomics to predict RP is more efficient and effective.

Project description:Background: To investigate the role of stereotactic body RT (SBRT) in decreased total peripheral lymphocyte count (TLC) in patients with early-stage lung cancer and to explore possible risk factors for RT-induced lymphopenia. Materials and Methods: We analyzed the TLCs and lymphocyte subsets of 76 patients in our prospective clinical database who received SBRT for early-stage lung cancer treatment. Relationships between clinical factors or dosimetric parameters and TLC were evaluated using Spearman's correlation analysis and Chi-square tests for continuous and categorical variables, respectively. Multivariate linear regression analysis was used to control for confounding factors. Kaplan-Meier analysis with a log-rank test and a multivariate Cox regression model were used for survival analysis. Results: Most patients (64/76, 84.2%) experienced decreased absolute lymphocyte counts following SBRT, as well as shifts in lymphocyte subset distributions. Spearman's correlation coefficients between post-SBRT TLC and the percentage of the lung and heart receiving 5 to 50 Gy (in 5 Gy increments) shown that most lung DVH parameters [V(10)-V(50)] were significantly negatively correlated with post-SBRT TLC, while only heart V(5), V(20), V(25), V(30), and V(45) were significant. Univariate analyses revealed that a lower Pre-SBRT TLC level, higher mean lung dose, longer treatment duration, and longer TBT were significantly associated with a lower Post-SBRT TLC level (all P < 0.05). Stepwise multivariate linear regression, which incorporated all of the significantly clinical variables and SBRT-related parameters in univariate analysis, revealed that lower pre -SBRT TLC (P < 0.001), higher heart V5 (P = 0.002), and longer total beam-on time (TBT) (P = 0.001) were the independent risk factors for decrease in post-SBRT TLC. Patients with lower post-SBRT TLC and longer TBT exhibited significantly inferior progression-free survival (PFS) (P < 0.001 and P = 0.013) and overall survival (P = 0.006 and P = 0.043). Conclusions: G2 and more severe lymphopenia after SBRT might be an independent prognostic factor for poorer outcome in early-stage lung cancer. Lowering heart V5 and TBT when designing SBRT plans may spare circulating lymphocytes and have the potential to further improve survival outcomes.

Dataset Information

Using machine learning to predict radiation pneumonitis in patients with stage I non-small cell lung cancer treated with stereotactic body radiation therapy.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets