Radiomic phenotype features predict pathological response in non-small cell lung cancer

Abstract

Background and purpose Radiomics can quantify tumor phenotype characteristics non-invasively by applying advanced imaging feature algorithms. In this study we assessed if pre-treatment radiomics data are able to predict pathological response after neoadjuvant chemoradiation in patients with locally advanced non-small cell lung cancer (NSCLC). Materials and Methods 127 NSCLC patients were included in this study. Fifteen radiomic features selected based on stability and variance were evaluated for its power to predict pathological response. Predictive power was evaluated using area under the curve (AUC). Conventional imaging features (tumor volume and diameter) were used for comparison. Results Seven features were predictive for pathologic gross residual disease (AUC > 0.6, p-value < 0.05), and one for pathologic complete response (AUC = 0.63, p-value = 0.01). No conventional imaging features were predictive (range AUC = 0.51–0.59, p-value > 0.05). Tumors that did not respond well to neoadjuvant chemoradiation were more likely to present rounder shape (spherical disproportionality, AUC = 0.63, p-value = 0.009) and heterogeneous texture (LoG 5 mm 3D - GLCM entropy, AUC = 0.61, p-value = 0.03). Conclusion We identified predictive radiomic features for pathological response, although no conventional features were significantly predictive. This study demonstrates that radiomics can provide valuable clinical information, and performed better than conventional imaging features.

Texture analysis of medical images for radiotherapy applications

“…81 Other works focused on prediction of lung recurrence. For example, from PET images, it was found that heterogeneity measures, such as entropy, can predict diseasespecific survival, 82,83 whereas CT images were used for the assessment of pathologic response, 84 overall survival and distant metastases, 85 finding that texture analysis can outperform conventional indices (as tumour volume and diameter).…”

CT-based radiomic analysis of stereotactic body radiation therapy patients with lung cancer

“…tumor diameter as a predictor of response). Radiomic features have been associated with tumor characteristics, such as genotype and protein expression [19][20][21], and have been prognostic of clinical outcomes [22][23][24][25][26].…”

Defining the biological and clinical basis of radiomics

“…Image acquisition has experienced substantial advances over the last decade, especially in terms of hardware and image reconstruction algorithms [36,37]. In addition, standardized imaging protocols have contributed in harmonizing images across hospitals [37,65]. From those images, segmentations are of the tumors are obtained.…”

“…Therefore, we presented additional research in Chapters 7-9 in which we for the first time described and applied an extensive range of machine learning algorithms that are applicable to predicting overall survival of patients with non-small cell lung cancer (NSCLC) [63,64]. Furthermore, we used these algorithms to train both univariate and multivariate models to predict the pathological response to neoadjuvant chemotherapy [65] and to predict the future development of distant metastases in patients with NSCLC [12]. In particular, we discovered novel radiomic signature that improved clinical prognostic models, results some of which were also validated in independent data.…”

“…Predicting pathological response prior to surgery would allow omitting surgery and potentially intensify chemoradiation for patients who would otherwise not benefit from added surgery. In Chapter 8, we investigated whether radiomics can be used to predict pathological response for neoadjuvant chemotherapy in NSCLC by developing an unsupervised feature selection algorithm [65]. We applied this algorithm to a comprehensive radiomic feature set of 1,603 features to identify a canonical subset of features.…”