Radiomic predicts early response to CDK4/6 inhibitors in hormone receptor positive metastatic breast cancer

The combination of Cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) and endocrine therapy (ET) is the standard of care for hormone receptor-positive (HR + ), human epidermal growth factor receptor 2-negative (HER2-) metastatic breast cancer (MBC). Currently, there are no robust biomarkers that can predict response to CDK4/6i, and it is not clear which patients benefit from this therapy. Since MBC patients with liver metastases have a poorer prognosis, developing predictive biomarkers that could identify patients likely to respond to CDK4/6i is clinically important. Here we show the ability of imaging texture biomarkers before and a few cycles after CDK4/6i therapy, to predict early response and overall survival (OS) on 73 MBC patients with known liver metastases who received palbociclib plus ET from two sites. The delta radiomic model was associated with OS in validation set (HR: 2.4; 95% CI, 1.06-5.6; P = 0.035; C-index = 0.77). Compared to RECIST response, delta radiomic features predicted response with area under the curve (AUC) = 0.72, 95% confidence interval (CI) 0.67-0.88. Our study revealed that radiomics features can predict a lack of response earlier than standard anatomic/RECIST 1.1 assessment and warrants further study and clinical validation.

Abstract 5427: Improving non-small cell lung cancer segmentation on a challenging dataset

When applied to different datasets, performance of the same deep learning tumor segmentation model can greatly vary. In a non-small cell lung cancer CT scan segmentation study that consists of two datasets, we found that the SwinUNETR model achieves state-of-the-art DICE score on a public dataset NSCLC but performs badly on a private dataset of curated data collected clinically. This performance variation reduces the applicability of such models. To mitigate this gap, through experimentation, we identified a set of techniques and applied them in the following order: (1) normalize a dataset to reduce differences between images. (2) stratify a dataset according to tumor sizes to form a more diverse training set. (3) isolate the lung area before training to help the model focus on the right area. (4) before training, initialize models with self-supervised pre-training weights (5) use a new loss function to give more weights on the cancerous area (6) after a model is trained, perform 3-axis test time flipping augmentation and ensemble the final predictions. In our experiments, our set of techniques improved the test DICE score for both datasets we tested on, where the best improvement was a 53% improvement from 0.32 to 0.49 DICE score.

Citation Format: Yi Wei, Balaji Selvaraj, Mayank Patwari, Qin Li, Meng Xu, Konstantinos Sidiropoulos, Zhenning Zhang, Leon Fedden, Anant Madabhushi, Mohammadhadi Khorrami, Vidya Sankar Viswanathan, Amit Gupta. Improving non-small cell lung cancer segmentation on a challenging dataset. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5427.

A tumor vasculature-based imaging biomarker for predicting response and survival in patients with lung cancer treated with checkpoint inhibitors

Tumor vasculature is a key component of the tumor microenvironment that can influence tumor behavior and therapeutic resistance. We present a new imaging biomarker, quantitative vessel tortuosity (QVT), and evaluate its association with response and survival in patients with non-small cell lung cancer (NSCLC) treated with immune checkpoint inhibitor (ICI) therapies. A total of 507 cases were used to evaluate different aspects of the QVT biomarkers. QVT features were extracted from computed tomography imaging of patients before and after ICI therapy to capture the tortuosity, curvature, density, and branching statistics of the nodule vasculature. Our results showed that QVT features were prognostic of OS (HR = 3.14, 0.95% CI = 1.2 to 9.68, P = 0.0006, C-index = 0.61) and could predict ICI response with AUCs of 0.66, 0.61, and 0.67 on three validation sets. Our study shows that QVT imaging biomarker could potentially aid in predicting and monitoring response to ICI in patients with NSCLC.

Novel Radiomic Measurements of Tumor-Associated Vasculature Morphology on Clinical Imaging as a Biomarker of Treatment Response in Multiple Cancers

PURPOSE

The tumor-associated vasculature (TAV) differs from healthy blood vessels by its convolutedness, leakiness, and chaotic architecture, and these attributes facilitate the creation of a treatment-resistant tumor microenvironment. Measurable differences in these attributes might also help stratify patients by likely benefit of systemic therapy (e.g., chemotherapy). In this work, we present a new category of computational image-based biomarkers called quantitative tumor-associated vasculature (QuanTAV) features, and demonstrate their ability to predict response and survival across multiple cancer types, imaging modalities, and treatment regimens involving chemotherapy.

EXPERIMENTAL DESIGN

We isolated tumor vasculature and extracted mathematical measurements of twistedness and organization from routine pretreatment radiology (CT or contrast-enhanced MRI) of a total of 558 patients, who received one of four first-line chemotherapy-based therapeutic intervention strategies for breast (n = 371) or non-small cell lung cancer (NSCLC, n = 187).

RESULTS

Across four chemotherapy-based treatment strategies, classifiers of QuanTAV measurements significantly (P < 0.05) predicted response in held out testing cohorts alone (AUC = 0.63-0.71) and increased AUC by 0.06-0.12 when added to models of significant clinical variables alone. Similarly, we derived QuanTAV risk scores that were prognostic of recurrence-free survival in treatment cohorts who received surgery following chemotherapy for breast cancer [P = 0.0022; HR = 1.25; 95% confidence interval (CI), 1.08-1.44; concordance index (C-index) = 0.66] and chemoradiation for NSCLC (P = 0.039; HR = 1.28; 95% CI, 1.01-1.62; C-index = 0.66). From vessel-based risk scores, we further derived categorical QuanTAV high/low risk groups that were independently prognostic among all treatment groups, including patients with NSCLC who received chemotherapy only (P = 0.034; HR = 2.29; 95% CI, 1.07-4.94; C-index = 0.62). QuanTAV response and risk scores were independent of clinicopathologic risk factors and matched or exceeded models of clinical variables including posttreatment response.

CONCLUSIONS

Across these domains, we observed an association of vascular morphology on CT and MRI-as captured by metrics of vessel curvature, torsion, and organizational heterogeneity-and treatment outcome. Our findings suggest the potential of shape and structure of the TAV in developing prognostic and predictive biomarkers for multiple cancers and different treatment strategies.

Texture-based CT radiomics distinguishes radiation and immunotherapy induced pneumonitis in stage III NSCLC

8555

Background: Recent changes to the standard of care for unresectable stage III NSCLC include chemoradiation followed by consolidative immunotherapy (IO). Pneumonitis is a well-known complication of radiotherapy (RT) and has been increasingly reported in association with IO. Although rare, pneumonitis can cause severe morbidity and possibly death in extreme cases. Differentiating RT and IO-induced pneumonitis (RTP vs IOP) is crucial for acute management and future considerations of individualized treatment. However, the clinical and radiological features of RTP and IOP may be similar and often indistinguishable on computed tomography (CT). Texture-based CT radiomics has previously been used to distinguish benign and malignant nodules on lung CT. In this study, we explore if radiomic features extracted from lung CT can distinguish between RTP and IOP. Methods: From 236 patients with stage III NSCLC who underwent chemoradiation followed by consolidative durvalumab, we identified 110 cases of treatment-related pneumonitis. IOP cases were identified through a retrospective review of electronic medical records and independently verified by a thoracic oncologist using features such as bilateral lung involvement, inflammatory changes outside the field of RT, temporal relationship to IO, and response to treatment. Inflammatory lesions were manually annotated using Slicer 3D. After excluding cases without discernible cause and non-identifiable lung lesions (n = 61), we included 49 cases in the study (RTP n = 20; IOP n = 29). A total of 555 features from Gabor, Laws, Laplace, and Haralick feature families were extracted on a pixel level from post-treatment CT images. A support vector machine (SVM) classifier was trained with the most discriminating features identified by Wilcoxon rank-sum test feature selection method. The classifier performance for distinguishing RTP vs. IOP was assessed by averaging the area under the receiver operating characteristic curve (AUC) values computed over 100 iterations of threefold cross-validation. Results: We identified the top 5 radiomic texture features distinguishing RTP from IOP including Haralick entropy, Haralick info, Laws median, and high- and low-frequency Gabor. Using 3-fold cross-validation, the SVM classifier model built on the radiomic features achieved an AUC of 0.83 (95% confidence interval, 0.78 - 0.86). Conclusions: Pneumonitis is a severe complication of both RT and IO that must be taken into consideration when evaluating future risks of IO-based therapies. The distinction between RTP and IOP remains challenging based on CT findings alone. Radiomic texture features analysis of post-treatment CT images can potentially differentiate RTP from IOP in stage III NSCLC patients who received RT followed by consolidative durvalumab. Additional multi-site independent validation of these quantitative image-based biomarkers is warranted.

Novel imaging biomarkers predict outcomes in stage III unresectable non-small cell lung cancer treated with chemoradiation and durvalumab

Background

The landmark study of durvalumab as consolidation therapy in NSCLC patients (PACIFIC trial) demonstrated significantly longer progression-free survival (PFS) in patients with locally advanced, unresectable non-small cell lung cancer (NSCLC) treated with durvalumab (immunotherapy, IO) therapy after chemoradiotherapy (CRT). In clinical practice in the USA, durvalumab continues to be used in patients across all levels of programmed cell death ligand-1 (PD-L1) expression. While immune therapies have shown promise in several cancers, some patients either do not respond to the therapy or have cancer recurrence after an initial response. It is not clear so far who will benefit of this therapy or what the mechanisms behind treatment failure are.

Methods

A total of 133 patients with unresectable stage III NSCLC who underwent durvalumab after CRT or CRT alone were included. Patients treated with durvalumab IO after CRT were randomly split into training (D1=59) and test (D2=59) sets and the remaining 15 patients treated with CRT alone were grouped in D3. Radiomic textural patterns from within and around the target nodules were extracted. A radiomic risk score (RRS) was built and was used to predict PFS and overall survival (OS). Patients were divided into high-risk and low-risk groups based on median RRS.

Results

RRS was found to be significantly associated with PFS in D1 (HR=2.67, 95% CI 1.85 to 4.13, p<0.05, C-index=0.78) and D2 (HR=2.56, 95% CI 1.63 to 4, p<0.05, C-index=0.73). Similarly, RRS was associated with OS in D1 (HR=1.89, 95% CI 1.3 to 2.75, p<0.05, C-index=0.67) and D2 (HR=2.14, 95% CI 1.28 to 3.6, p<0.05, C-index=0.69), respectively. RRS was found to be significantly associated with PFS in high PD-L1 (HR=3.01, 95% CI 1.41 to 6.45, p=0.0044) and low PD-L1 (HR=2.74, 95% CI 1.8 to 4.14, p=1.77e-06) groups. Moreover, RRS was not significantly associated with OS in the high PD-L1 group (HR=2.08, 95% CI 0.98 to 4.4, p=0.054) but was significantly associated with OS in the low PD-L1 group (HR=1.61, 95% CI 1.14 to 2.28, p=0.0062). In addition, RRS was significantly associated with PFS (HR=2.77, 95% CI 1.17 to 6.52, p=0.019, C-index=0.77) and OS (HR=2.62, 95% CI 1.25 to 5.51, p=0.01, C-index=0.77) in D3, respectively.

Conclusions

Tumor radiomics of pretreatment CT images from patients with stage III unresectable NSCLC were prognostic of PFS and OS to CRT followed by durvalumab IO and CRT alone.

Effects of excessive water extraction on groundwater ecosystem services: Vulnerability assessments using biophysical approaches

In this research, a systematic, integrated framework was developed to evaluate the biophysical state and vulnerability intensity of groundwater supply ecosystem service (GSES) regarding excessive groundwater withdrawal in the Mashhad plain located in the northeastern of Khorasan Razavi province in Iran. At first, following an indexing approach, the biophysical aspects of the ecosystem, including capacity, flow, and the benefits of GSES, were analyzed. Afterward, the relationship between the capacity and flow ecosystem service (ES) was assessed to identify ecosystem's sustainability status. Furthermore, GSES stability was spatially shown. Finally, GSES vulnerability and its associated ESs was assessed based on 3D model of vulnerability via indexing three components of exposure, sensitivity, and adaptive capacity. The final map spatially indicated the zoning of groundwater ecosystem services' vulnerability intensity in Mashhad Plain. The outcomes indicate a high and very high vulnerability in more than 35% of studying area. The results indicate that about 18%, 30%, and 15% of studying land show moderate, low, and no vulnerability, respectively. Finally, it was observed that due to groundwater's over-extraction, supplying the aquifer ecosystem services was disrupted. This method can be used as a solution for the sustainable management of groundwater resources, especially in the arid and semi-arid countries facing the depletion of water resources.

Novel Non-Invasive Radiomic Signature on CT Scans Predicts Response to Platinum-Based Chemotherapy and Is Prognostic of Overall Survival in Small Cell Lung Cancer

BACKGROUND

Small cell lung cancer (SCLC) is an aggressive malignancy characterized by initial chemosensitivity followed by resistance and rapid progression. Presently, there are no predictive biomarkers that can accurately guide the use of systemic therapy in SCLC patients. This study explores the role of radiomic features from both within and around the tumor lesion on pretreatment CT scans to a) prognosticate overall survival (OS) and b) predict response to chemotherapy.

METHODS

One hundred fifty-three SCLC patients who had received chemotherapy were included. Lung tumors were contoured by an expert reader. The patients were divided randomly into approximately equally sized training (Str = 77) and test sets (Ste = 76). Textural descriptors were extracted from the nodule (intratumoral) and parenchymal regions surrounding the nodule (peritumoral). The clinical endpoints of this study were OS, progression-free survival (PFS), and best objective response to chemotherapy. Patients with complete or partial response were defined as "responders," and those with stable or progression of disease were defined as "non-responders." The radiomic risk score (RRS) was generated by using the least absolute shrinkage and selection operator (LASSO) with the Cox regression model. Patients were classified into the high-risk or low-risk groups based on the median of RRS. Association of the radiomic signature with OS was evaluated on Str and then tested on Ste. The features identified by LASSO were then used to train a linear discriminant analysis (LDA) classifier (MRad) to predict response to chemotherapy. A prognostic nomogram (NRad+Clin) was also developed on Str by combining clinical and prognostic radiomic features and validated on Ste. The Kaplan-Meier survival analysis and log-rank statistical tests were performed to assess the discriminative ability of the features. The discrimination performance of the NRad+Clin was assessed by Harrell's C-index. To estimate the clinical utility of the nomogram, decision curve analysis (DCA) was performed by calculating the net benefits for a range of threshold probabilities in predicting which high-risk patients should receive more aggressive treatment as compared with the low-risk patients.

RESULTS

A univariable Cox regression analysis indicated that RRS was significantly associated with OS in Str (HR: 1.53; 95% CI, [1.1-2.2; p = 0.021]; C-index = 0.72) and Ste (HR: 1.4, [1.1-1.82], p = 0.0127; C-index = 0.69). The RRS was also significantly associated with PFS in Str (HR: 1.89, [1.4-4.61], p = 0.047; C-index = 0.7) and Ste (HR: 1.641, [1.1-2.77], p = 0.04; C-index = 0.67). MRad was able to predict response to chemotherapy with an area under the receiver operating characteristic curve (AUC) of 0.76 ± 0.03 within Str and 0.72 within Ste. Predictors, including the RRS, gender, age, stage, and smoking status, were used in the prognostic nomogram. The discrimination ability of the NRad+Clin model on Str and Ste was C-index [95% CI]: 0.68 [0.66-0.71] and 0.67 [0.63-0.69], respectively. DCA indicated that the NRad+Clin model was clinically useful.

CONCLUSIONS

Radiomic features extracted within and around the lung tumor on CT images were both prognostic of OS and predictive of response to chemotherapy in SCLC patients.

Combination of quantitative features from H&E biopsies and CT scans predicts response to chemotherapy and overall survival in small cell lung cancer (SCLC)

8572

Background: Small Cell Lung Cancer (SCLC) is an aggressive malignancy with a rapid growth, and Chemotherapy remains mainstay of treatment. Identifying therapeutic targets in SCLC presents a challenge, partially due to a lack of accurate and consistently predictive biomarkers. In this study we sought to evaluate the utility of a combination of computer-extracted radiographic and pathology features from pretreatment baseline CT and H&E biopsy images to predict sensitivity to platinum-based chemotherapy and overall survival (OS) in SCLC. Methods: Seventy-eight patients with extensive and limited-stage SCLC who received platinum-doublet chemotherapy were selected. Objective response to chemotherapy (RECIST criteria) and overall survival (OS) as clinical endpoints were available for 51 and 78 patients respectively. The patients were divided randomly into two sets (Training (Sd), Validation (Sv)) with a constraint (equal number of responders and nonresponders in Sd)—Sd comprised twenty-one patients with SCLC. Sv included thirty patients. CT scans and digitized Hematoxylin Eosin-stained (H&E) biopsy images were acquired for each patient. A set of CT derived (46%) and tissue derived (53%) image features were captured. These included shape and textural patterns of the tumoral and peritumoral regions from CT scans and of tumor regions on H&E images. A random forest feature selection and linear regression model were used to identify the most predictive CT and H&E derived image features associated with chemotherapy response from Sd. A Cox proportional hazard regression model was used with these features to compute a risk score for each patients in Sd. Patients in Sv were stratified into high and low-risk groups based on the median risk score. Kaplan-Meier survival analysis was used to assess the prognostic ability of the risk score on Sv. Results: The risk score comprised nine CT (intra and peri-tumoral texture) and six H&E derived (cancer cell texture and shape) features. A linear regression model in conjunction with these 15 features was significantly associated with chemo-sensitivity in Sv (AUC = 0.76, PRC = 0.81). A multivariable model with these 15 features was significantly associated with OS in Sv (HR = 2.5, 95% CI: 1.3-4.9, P = 0.0043). Kaplan-Meier survival analysis revealed a significantly reduced OS in the high-risk group compared to the low-risk group. Conclusions: A combined CT and H&E tissue derived image signature model predicted response to chemotherapy and improved OS in SCLC patients. Image features from baseline CT scans and H&E tissue slide images may help in better risk stratification of SCLC patients. Additional independent validation of these quantitative image-based biomarkers is warranted.

Novel imaging biomarkers predict progression-free survival in stage 3 NSCLC treated with chemoradiation and durvalumab

3054

Background: The current management of stage III non-small cell lung cancer (NSCLC) is chemoradiation followed by durvalumab consolidation. There are no robust biomarkers that predict benefit from this regimen. We evaluated the utility of novel imaging biomarkers (radiomics) to distinguish patients with stage III NSCLC who will benefit from treatment from those likely to progress despite therapy. Methods: Patients with stage III NSCLC treated at our center with chemoradiation and durvalumab from July 2017 - July 2019 were identified. We collected patient clinical outcomes, subtype of NSCLC, and PD-L1 expression as well as pre-treatment CT images. Images were split into training and test sets. Lung tumors were contoured on 3D-Slicer software and 1542 radiomic features capturing both intra- and peritumoral texture patterns were extracted. The primary endpoint of this study was progression-free survival (PFS), and the secondary objective was difference in PFS within high PD-L1 (≥50%) and low PDL1 (<50%) groups. We used the least absolute shrinkage and selection operator (LASSO) Cox regression model to build the radiomic signature for PFS. A risk score was computed according to a linear combination of selected features and their corresponding coefficients. High- and low-risk groups were defined based on median radiomics risk score. Multivariable Cox regression analysis was performed to evaluate the effect of each factor on PFS. We performed Kaplan–Meier survival analysis and log-rank tests to assess prognostic ability of the features. Results: We identified 118 patients who fit our criteria with available CT images and randomly divided them into a training (n=59) and a test set (n=59). The radiomic risk score was calculated using a linear combination of the top six selected features with corresponding coefficients. In a multivariable analysis using clinicopathologic and radiomic signatures, the radiomics risk-score and PD-L1 expression were found to be significantly associated with PFS in training (risk-score: HR = 2.3, 95% CI: [1.46 – 3.63], P = 0.0003; PD-L1: HR = 0.31, 95% CI: [0.081 – 0.96], P = 0.038) and test sets (risk-score: HR= 2.56, 95% CI: [1.75 – 4], P = 8.7e-05; PD-L1: HR = 0.27, 95% CI: [0.048 – 0.58], P = 0.005). Kaplan-Meier analyses showed a significantly shorter PFS in the high-risk radiomics group versus the low-risk group (P < 0.0001). The radiomics risk scores were also predictive of significant differences in PFS within both the low (p=0.0005) and high (p=0.0007) PD-L1 groups. Conclusions: Radiomic biomarkers from pre-treatment CT images in stage III NSCLC patients were predictive of PFS to chemoradiation followed by durvalumab and could predict outcomes regardless of PD-L1 level. Pre-treatment radiomics may allow early prediction of benefit and expedite more aggressive treatment for high-risk patients. Additional validation of these imaging biomarkers is warranted.

Distinguishing granulomas from adenocarcinomas by integrating stable and discriminating radiomic features on non-contrast computed tomography scans

OBJECTIVE

To identify stable and discriminating radiomic features on non-contrast CT scans to develop more generalisable radiomic classifiers for distinguishing granulomas from adenocarcinomas.

METHODS

In total, 412 patients with adenocarcinomas and granulomas from three institutions were retrospectively included. Segmentations of the lung nodules were performed manually by an expert radiologist in a 2D axial view. Radiomic features were extracted from intra- and perinodular regions. A total of 145 patients were used as part of the training set (S_tr), whereas 205 patients were used as part of test set I (S_te¹) and 62 patients were used as part of independent test set II (S_te²). To mitigate the variation of CT acquisition parameters, we defined 'stable' radiomic features as those for which the feature expression remains relatively unchanged between different sites, as assessed using a Wilcoxon rank-sum test. These stable features were used to develop more generalisable radiomic classifiers that were more resilient to variations in lung CT scans. Features were ranked based on two criteria, firstly based on discriminability (i.e. maximising AUC) alone and subsequently based on maximising both feature stability and discriminability. Different machine-learning classifiers (Linear discriminant analysis, Quadratic discriminant analysis, Support vector machines and random forest) were trained with features selected using the two different criteria and then compared on the two independent test sets for distinguishing granulomas from adenocarcinomas, in terms of area under the receiver operating characteristic curve.

RESULTS

In the test sets, classifiers constructed using the criteria involving maximising feature stability and discriminability simultaneously achieved higher AUC compared with the discriminating alone criteria (S_te¹ [n = 205]: maximum AUCs of 0.85versus . 0.80; p-value = 0.047 and S_te² [n = 62]: maximum AUCs of 0.87 versus. 0.79; p-value = 0.021). These differences held for features extracted from scans with <3 mm slice thickness (AUC = 0.88 versus. 0.80; p-value = 0.039, n = 100) and for the ≥3 mm cases (AUC = 0.81 versus. 0.76; p-value = 0.034, n = 105). In both experiments, shape and peritumoural texture features had a higher stability compared with intratumoural texture features.

CONCLUSIONS

Our study suggests that explicitly accounting for both stability and discriminability results in more generalisable radiomic classifiers to distinguish adenocarcinomas from granulomas on non-contrast CT scans. Our results also showed that peritumoural texture and shape features were less affected by the scanner parameters compared with intratumoural texture features; however, they were also less discriminating compared with intratumoural features.

Novel, non-invasive imaging approach to identify patients with advanced non-small cell lung cancer at risk of hyperprogressive disease with immune checkpoint blockade

Purpose

Hyperprogression is an atypical response pattern to immune checkpoint inhibition that has been described within non-small cell lung cancer (NSCLC). The paradoxical acceleration of tumor growth after immunotherapy has been associated with significantly shortened survival, and currently, there are no clinically validated biomarkers to identify patients at risk of hyperprogression.

Experimental design

A total of 109 patients with advanced NSCLC who underwent monotherapy with Programmed cell death protein-1 (PD1)/Programmed death-ligand-1 (PD-L1) inhibitors were included in the study. Using RECIST measurements, we divided the patients into responders (n=50) (complete/partial response or stable disease) and non-responders (n=59) (progressive disease). Tumor growth kinetics were used to further identify hyperprogressors (HPs, n=19) among non-responders. Patients were randomized into a training set (D₁=30) and a test set (D₂=79) with the essential caveat that HPs were evenly distributed among the two sets. A total of 198 radiomic textural patterns from within and around the target nodules and features relating to tortuosity of the nodule associated vasculature were extracted from the pretreatment CT scans.

Results

The random forest classifier using the top features associated with hyperprogression was able to distinguish between HP and other radiographical response patterns with an area under receiver operating curve of 0.85±0.06 in the training set (D₁=30) and 0.96 in the validation set (D₂=79). These features included one peritumoral texture feature from 5 to 10 mm outside the tumor and two nodule vessel-related tortuosity features. Kaplan-Meier survival curves showed a clear stratification between classifier predicted HPs versus non-HPs for overall survival (D₂: HR=2.66, 95% CI 1.27 to 5.55; p=0.009).

Conclusions

Our study suggests that image-based radiomics markers extracted from baseline CTs of advanced NSCLC treated with PD-1/PD-L1 inhibitors may help identify patients at risk of hyperprogressions.

Stable and discriminating radiomic predictor of recurrence in early stage non-small cell lung cancer: Multi-site study

OBJECTIVES

To evaluate whether combining stability and discriminability criteria in building radiomic classifiers will improve the prognosis of cancer recurrence in early stage non-small cell lung cancer on non-contrast computer tomography (CT).

MATERIALS AND METHODS

CT scans of 610 patients with early stage (IA, IB, IIA) NSCLC from four independent cohorts were evaluated. A total of 350 patients from Cleveland Clinic Foundation and University of Pennsylvania were divided into two equal sets for training (D₁) and validation set (D₂). 80 patients from The Cancer Genome Atlas Lung Adenocarcinoma and Squamous Cell Carcinoma and 195 patients from The Cancer Imaging Archive, were used as independent second (D₃) and third (D₄) validation sets. A linear discriminant analysis (LDA) classifier was built based on the most stable and discriminate features. In addition, a radiomic risk score (RRS) was generated by using least absolute shrinkage and selection operator, Cox regression model to predict time to progression (TTP) following surgery.

RESULTS

A feature selection strategy focusing on both feature discriminability and stability resulted in the classifier having a higher discriminability on validation datasets compared to the discriminability alone criteria in discriminating cancer recurrence (D₂, AUC of 0.75 vs. 0.65; D₃, 0.74 vs. 0.62; D₄, 0.76 vs. 0.63). The RRS generated by most stable-discriminating features was significantly associated with TTP compared to discriminating alone criteria (HR = 1.66, C-index of 0.72 vs. HR = 1.04, C-index of 0.62).

CONCLUSION

Accounting for both stability and discriminability yielded a more generalizable classifier for predicting cancer recurrence and TTP in early stage NSCLC.

Changes in CT Radiomic Features Associated with Lymphocyte Distribution Predict Overall Survival and Response to Immunotherapy in Non-Small Cell Lung Cancer

No predictive biomarkers can robustly identify patients with non-small cell lung cancer (NSCLC) who will benefit from immune checkpoint inhibitor (ICI) therapies. Here, in a machine learning setting, we compared changes ("delta") in the radiomic texture (DelRADx) of CT patterns both within and outside tumor nodules before and after two to three cycles of ICI therapy. We found that DelRADx patterns could predict response to ICI therapy and overall survival (OS) for patients with NSCLC. We retrospectively analyzed data acquired from 139 patients with NSCLC at two institutions, who were divided into a discovery set (D1 = 50) and two independent validation sets (D2 = 62, D3 = 27). Intranodular and perinodular texture descriptors were extracted, and the relative differences were computed. A linear discriminant analysis (LDA) classifier was trained with 8 DelRADx features to predict RECIST-derived response. Association of delta-radiomic risk score (DRS) with OS was determined. The association of DelRADx features with tumor-infiltrating lymphocyte (TIL) density on the diagnostic biopsies (n = 36) was also evaluated. The LDA classifier yielded an AUC of 0.88 ± 0.08 in distinguishing responders from nonresponders in D1, and 0.85 and 0.81 in D2 and D3 DRS was associated with OS [HR: 1.64; 95% confidence interval (CI), 1.22-2.21; P = 0.0011; C-index = 0.72). Peritumoral Gabor features were associated with the density of TILs on diagnostic biopsy samples. Our results show that DelRADx could be used to identify early functional responses in patients with NSCLC.

Predicting pathologic response to neoadjuvant chemoradiation in resectable stage III non-small cell lung cancer patients using computed tomography radiomic features

OBJECTIVE

The use of a neoadjuvant chemoradiation followed by surgery in patients with stage IIIA NSCLC is controversial and the benefit of surgery is limited. There are currently no clinically validated biomarkers to select patients for such an approach. In this study we evaluate computed tomography (CT) derived intratumoral and peritumoral texture and nodule shape features in their ability to predict major pathological response (MPR). MPR being defined as ≤10% of residual viable tumor, assessed at the time of surgery.

MATERIAL AND METHODS

Ninety patients with stage III NSCLC treated with chemoradiation prior to surgical resection were selected. The patients were divided randomly into two equal sets, one for training and one for independent testing. The radiomic texture and shape features were extracted from within the nodule (intra) and from the parenchymal regions immediately surrounding the nodule (peritumoral). A univariate regression analysis was performed on the image and clinicopathologic variables and then included into a multivariable logistic regression (MLR) for binary outcome prediction of MPR. The radiomic signature risk-score was generated by using a multivariate Cox regression model and association of the signature with OS and DFS was also evaluated.

RESULTS

Thirteen stable and predictive intratumoral and peritumoral radiomic texture features were found to be predictive of MPR. The MLR classifier yielded an AUC of 0.90 ± 0.025 within the training set and a corresponding AUC = 0.86 in prediction of MPR within the test set. The radiomic signature was also significantly associated with OS (HR = 11.18, 95% CI = 3.17, 44.1; p-value = 0.008) and DFS (HR = 2.78, 95% CI = 1.11, 4.12; p-value = 0.0042) in the testing set.

CONCLUSION

Texture features extracted within and around the lung tumor on CT images appears to be associated with the likelihood of MPR, OS and DFS to chemoradiation.

Intra and perinodular CT delta radiomic features associated with early response to predict overall survival (OS) in immunotherapy-treated non-small cell lung cancer (NSCLC): A multisite multi-agent study

2588

Background: None of the current biomarkers for predicting response to checkpoint inhibitors (ICIT) for advanced NSCLC are associated with long-term benefits, such as improved OS. In this multi-agent (nivolumab, pembrolizumab, or atezolizumab) multi-site study (Cleveland Clinic, Univ. of Pennsylvania), we demonstrate that changes in computer-extracted textural patterns, from within and 30mm outside the nodules, between baseline and post-treatment CT following ICIT correlate with RECIST-derived responses, and are prognostic of OS. Methods: CT scans from 139 NSCLC patients both pre-, and post 2-3 cycles of ICIT were acquired from 2 sites. Patients with objective response/stable disease per RECIST v1.1 were defined as ‘responders’, and those with progressive disease were ‘non-responders’. The cohort was divided into a discovery (D1 = 50) and two validation sets (D2 = 62, D3 = 27). 454 intranodular texture (IT) features, and 7426 perinodular features (PT) were extracted from the temporalscans, Relative differences were computed to yield a set of ‘delta-radiomic’ descriptors. In D1, 8 features that evolved the most between baseline and post-treatment CT, and performed the best in identifying responders, were determined. These were then used with a Linear Discriminant Analysis classifier to identify the responders from the non-responders. We then computed a radiomic risk score (RRS) system and tested its prognostic ability in assessing differences in OS. Results: A combination of 5 IT, 3 PT delta radiomic features yielded an AUC of 0.88 ± 0.08 in D1 and a corresponding AUC = 0.85 and 0.81 in D2 and D3, respectively. Multivariate survival metrics are shown in Table. Conclusions: Delta-radiomic features, both from inside and outside the nodules, could be used to identify patients likely to derive clinical benefit from ICIT (eg: OS) beyond anatomic response. [Table: see text]

Combination of Peri- and Intratumoral Radiomic Features on Baseline CT Scans Predicts Response to Chemotherapy in Lung Adenocarcinoma

Purpose

To identify the role of radiomics texture features both within and outside the nodule in predicting (a) time to progression (TTP) and overall survival (OS) as well as (b) response to chemotherapy in patients with non-small cell lung cancer (NSCLC).

Materials and Methods

Data in a total of 125 patients who had been treated with pemetrexed-based platinum doublet chemotherapy at Cleveland Clinic were retrospectively analyzed. The patients were divided randomly into two sets with the constraint that there were an equal number of responders and nonresponders in the training set. The training set comprised 53 patients with NSCLC, and the validation set comprised 72 patients. A machine learning classifier trained with radiomic texture features extracted from intra- and peritumoral regions of non-contrast-enhanced CT images was used to predict response to chemotherapy. The radiomic risk-score signature was generated by using least absolute shrinkage and selection operator with the Cox regression model; association of the radiomic signature with TTP and OS was also evaluated.

Results

A combination of radiomic features in conjunction with a quadratic discriminant analysis classifier yielded a mean maximum area under the receiver operating characteristic curve (AUC) of 0.82 ± 0.09 (standard deviation) in the training set and a corresponding AUC of 0.77 in the independent testing set. The radiomics signature was also significantly associated with TTP (hazard ratio [HR], 2.8; 95% confidence interval [CI]: 1.95, 4.00; P < .0001) and OS (HR, 2.35; 95% CI: 1.41, 3.94; P = .0011). Additionally, decision curve analysis demonstrated that in terms of clinical usefulness, the radiomics signature had a higher overall net benefit in prediction of high-risk patients to receive treatment than the clinicopathologic measurements.

Conclusion

This study suggests that radiomic texture features extracted from within and around the nodule on baseline CT scans are (a) predictive of response to chemotherapy and (b) associated with TTP and OS for patients with NSCLC.© RSNA, 2019Supplemental material is available for this article.

Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas

Purpose To evaluate ability of radiomic (computer-extracted imaging) features to distinguish non-small cell lung cancer adenocarcinomas from granulomas at noncontrast CT. Materials and Methods For this retrospective study, screening or standard diagnostic noncontrast CT images were collected for 290 patients (mean age, 68 years; range, 18-92 years; 125 men [mean age, 67 years; range, 18-90 years] and 165 women [mean age, 68 years; range, 33-92 years]) from two institutions between 2007 and 2013. Histopathologic analysis was available for one nodule per patient. Corresponding nodule of interest was identified on axial CT images by a radiologist with manual annotation. Nodule shape, wavelet (Gabor), and texture-based (Haralick and Laws energy) features were extracted from intra- and perinodular regions. Features were pruned to train machine learning classifiers with 145 patients. In a test set of 145 patients, classifier results were compared against a convolutional neural network (CNN) and diagnostic readings of two radiologists. Results Support vector machine classifier with intranodular radiomic features achieved an area under the receiver operating characteristic curve (AUC) of 0.75 on the test set. Combining radiomics of intranodular with perinodular regions improved the AUC to 0.80. On the same test set, CNN resulted in an AUC of 0.76. Radiologist readers achieved AUCs of 0.61 and 0.60, respectively. Conclusion Radiomic features from intranodular and perinodular regions of nodules can distinguish non-small cell lung cancer adenocarcinomas from benign granulomas at noncontrast CT. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Nishino in this issue.

Static and dynamic phase transitions in multidimensional voting models on continua

A voting model (or a generalization of the Glauber model at zero temperature) on a multidimensional lattice is defined as a system composed of a lattice, each site of which is either empty or occupied by a single particle. The reactions of the system are such that two adjacent sites, one empty, the other occupied, may evolve to a state where both of these sites are either empty or occupied. The continuum version of this model in a D-dimensional region with a boundary is studied, and two general behaviors of such systems are investigated, the stationary behavior of the system, and the dominant way of relaxation of the system toward its stationary state. Based on the first behavior, a static phase transition (discontinuous changes in the stationary profiles of the system) is studied. Based on the second behavior, a dynamical phase transition (discontinuous changes in the relaxation times of the system) is studied. It is shown that the static phase transition is induced by the bulk reactions only, while the dynamical phase transition is a result of both bulk reactions and boundary conditions.

Autonomous multispecies reaction-diffusion systems with more-than-two-site interactions

Autonomous multispecies systems with more-than-two-neighbor interactions are studied. Conditions necessary and sufficient for the closedness of the evolution equations of the n-point functions are obtained. The average numbers of the particles at each site for one species and three-site interactions, and its generalization to the more-than-three-site interactions, are explicitly obtained. Generalizations of the Glauber model in different directions, using generalized rates, generalized numbers of states at each site, and generalized numbers of interacting sites, are also investigated.

Exactly solvable models through the empty-interval method

The most general one dimensional reaction-diffusion model with nearest-neighbor interactions that can be solved exactly through empty-interval method has been introduced. Assuming translationally invariant initial conditions, the probability that n consecutive sites are empty, E(n), has been exactly obtained. Here, however, we do not consider reactions changing two empty neighboring sites. In the thermodynamic limit, the large-time behavior of the system has also been investigated. Releasing translationally invariance, the evolution equation for the probability that n consecutive sites, starting from the site k, are empty, E(k,n), is obtained. In the thermodynamic limit, the large time behavior of the system is also considered. Finally, the continuum limit of the model is considered and the empty-interval probability function is obtained.

Phase transition in an asymmetric generalization of the zero-temperature q-state Potts model

An asymmetric generalization of the zero-temperature q-state Potts model on a one-dimensional lattice, with and without boundaries, has been studied. The dynamics of the particle number, and especially the large time behavior of the system, has been analyzed. In the thermodynamic limit, the system exhibits two kinds of phase transitions, a static and a dynamic phase transition.

Solvable multispecies reaction-diffusion processes

A family of one-dimensional multispecies reaction-diffusion processes on a lattice is introduced. It is shown that these processes are exactly solvable, provided a nonspectral matrix equation is satisfied. Some general remarks on the solutions to this equation, and some special solutions are given. The large-time behavior of the conditional probabilities of such systems is also investigated.

Phase transition in an asymmetric generalization of the zero-temperature Glauber model

An asymmetric generalization of the zero-temperature Glauber model on a lattice is introduced. The dynamics of the particle-density and especially the large-time behavior of the system is studied. It is shown that the system exhibits two kinds of phase transitions, a static one and a dynamic one.

Multispecies reaction-diffusion systems

Multispecies reaction-diffusion systems, for which the time evolution equations of correlation functions become a closed set, are considered. A formal solution for the average densities is found. Some special interactions and the exact time dependence of the average densities in these cases are also studied. For the general case, the large-time behavior of the average densities has also been obtained.

Asymmetric one-dimensional exclusion processes: a two-parameter exactly solvable example

We consider a two-parameter family of asymmetric exclusion processes for particles living on a continuous one-dimensional space. Using the Bethe ansatz, the exact solution to the master equation, and from that the drift and the diffusion rate in the two particle sector, are obtained.