1
|
Ye G, Wu G, Qi Y, Li K, Wang M, Zhang C, Li F, Wee L, Dekker A, Han C, Liu Z, Liao Y, Shi Z. Non-invasive multimodal CT deep learning biomarker to predict pathological complete response of non-small cell lung cancer following neoadjuvant immunochemotherapy: a multicenter study. J Immunother Cancer 2024; 12:e009348. [PMID: 39231545 PMCID: PMC11409329 DOI: 10.1136/jitc-2024-009348] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 08/04/2024] [Indexed: 09/06/2024] Open
Abstract
OBJECTIVES Although neoadjuvant immunochemotherapy has been widely applied in non-small cell lung cancer (NSCLC), predicting treatment response remains a challenge. We used pretreatment multimodal CT to explore deep learning-based immunochemotherapy response image biomarkers. METHODS This study retrospectively obtained non-contrast enhanced and contrast enhancedbubu CT scans of patients with NSCLC who underwent surgery after receiving neoadjuvant immunochemotherapy at multiple centers between August 2019 and February 2023. Deep learning features were extracted from both non-contrast enhanced and contrast enhanced CT scans to construct the predictive models (LUNAI-uCT model and LUNAI-eCT model), respectively. After the feature fusion of these two types of features, a fused model (LUNAI-fCT model) was constructed. The performance of the model was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. SHapley Additive exPlanations analysis was used to quantify the impact of CT imaging features on model prediction. To gain insights into how our model makes predictions, we employed Gradient-weighted Class Activation Mapping to generate saliency heatmaps. RESULTS The training and validation datasets included 113 patients from Center A at the 8:2 ratio, and the test dataset included 112 patients (Center B n=73, Center C n=20, Center D n=19). In the test dataset, the LUNAI-uCT, LUNAI-eCT, and LUNAI-fCT models achieved AUCs of 0.762 (95% CI 0.654 to 0.791), 0.797 (95% CI 0.724 to 0.844), and 0.866 (95% CI 0.821 to 0.883), respectively. CONCLUSIONS By extracting deep learning features from contrast enhanced and non-contrast enhanced CT, we constructed the LUNAI-fCT model as an imaging biomarker, which can non-invasively predict pathological complete response in neoadjuvant immunochemotherapy for NSCLC.
Collapse
Affiliation(s)
- Guanchao Ye
- Department of Thoracic Surgery, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China
- Department of Thoracic Surgery, Huazhong University of Science and Technology Tongji Medical College Union Hospital, Wuhan, Hubei, China
| | - Guangyao Wu
- Department of Radiology, Huazhong University of Science and Technology Tongji Medical College Union Hospital, Wuhan, Hubei, China
| | - Yu Qi
- Department of Thoracic Surgery, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China
| | - Kuo Li
- Department of Thoracic Surgery, Huazhong University of Science and Technology Tongji Medical College Union Hospital, Wuhan, Hubei, China
| | - Mingliang Wang
- Department of Thoracic Surgery, Henan Provincial People's Hospital, Zhengzhou, Henan, China
| | - Chunyang Zhang
- Department of Thoracic Surgery, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China
| | - Feng Li
- Department of Thoracic Surgery, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China
| | - Leonard Wee
- Clinical Data Science, Faculty of Health Medicine and Life Sciences, Maastricht University, Maastricht, The Netherlands
- Department of Radiation Oncology (Maastro), GROW School of Oncology and Reproduction, Maastricht University Medical Centre+, Maastricht, The Netherlands
| | - Andre Dekker
- Department of Radiation Oncology (Maastro), GROW School of Oncology and Reproduction, Maastricht University Medical Centre+, Maastricht, The Netherlands
| | - Chu Han
- Department of Radiology, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
- Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, China
- Medical Research Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Zaiyi Liu
- Department of Radiology, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
- Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, China
| | - Yongde Liao
- Department of Thoracic Surgery, Huazhong University of Science and Technology Tongji Medical College Union Hospital, Wuhan, Hubei, China
| | - Zhenwei Shi
- Department of Radiology, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
- Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, China
- Medical Research Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| |
Collapse
|
2
|
Shen YC, Liu TH, Nicholas A, Soyama A, Yuan CT, Chen TC, Eguchi S, Yoshizumi T, Itoh S, Nakamura N, Kosaka H, Kaibori M, Ishii T, Hatano E, Ogawa C, Naganuma A, Kakizaki S, Cheng CH, Lin PT, Su YY, Chuang CH, Lu LC, Wu CJ, Wang HW, Rau KM, Hsu CH, Lin SM, Huang YH, Hernandez S, Finn RS, Kudo M, Cheng AL. Clinical Outcomes and Histologic Findings of Patients With Hepatocellular Carcinoma With Durable Partial Response or Durable Stable Disease After Receiving Atezolizumab Plus Bevacizumab. J Clin Oncol 2024:JCO2400645. [PMID: 39197119 DOI: 10.1200/jco.24.00645] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/28/2024] [Revised: 05/30/2024] [Accepted: 06/13/2024] [Indexed: 08/30/2024] Open
Abstract
PURPOSE Durable partial response (PR) and durable stable disease (SD) are often seen in patients with hepatocellular carcinoma (HCC) receiving atezolizumab plus bevacizumab (atezo-bev). This study investigates the outcome of these patients and the histopathology of the residual tumors. PATIENTS AND METHODS The IMbrave150 study's atezo-bev group was analyzed. PR or SD per RECIST v1.1 lasting more than 6 months was defined as durable. For histologic analysis, a comparable real-world group of patients from Japan and Taiwan who had undergone resection of residual tumors after atezo-bev was investigated. RESULTS In the IMbrave150 study, 56 (77.8%) of the 72 PRs and 41 (28.5%) of the 144 SDs were considered durable. The median overall survival was not estimable for patients with durable PR and 23.7 months for those with durable SD. The median progression-free survival was 23.2 months for patients with durable PR and 13.2 months for those with durable SD. In the real-world setting, a total of 38 tumors were resected from 32 patients (23 PRs and nine SDs) receiving atezo-bev. Pathologic complete responses (PCRs) were more frequent in PR tumors than SD tumors (57.7% v 16.7%, P = .034). PCR rate correlated with time from atezo-bev initiation to resection and was 55.6% (5 of 9) for PR tumors resected beyond 8 months after starting atezo-bev, a time practically corresponding to the durable PR definition used for IMbrave150. We found no reliable radiologic features to predict PCR of the residual tumors. CONCLUSION Durable PR patients from the atezo-bev group showed a favorable outcome, which may be partly explained by the high rate of PCR lesions. Early recognition of PCR lesions may help subsequent treatment decision.
Collapse
Affiliation(s)
- Ying-Chun Shen
- Department of Medical Oncology, National Taiwan University Cancer Center, Taipei, Taiwan
- Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan
- Graduate Institute of Oncology, College of Medicine, National Taiwan University, Taipei, Taiwan
| | - Tsung-Hao Liu
- Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan
- Graduate Institute of Oncology, College of Medicine, National Taiwan University, Taipei, Taiwan
| | | | - Akihiko Soyama
- Department of Surgery, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan
| | - Chang-Tsu Yuan
- Department of Pathology, National Taiwan University Cancer Center, Taipei, Taiwan
| | - Tse-Ching Chen
- Department of Pathology, Chang Gung Memorial Hospital Linkou Branch, Taoyuan, Taiwan
| | - Susumu Eguchi
- Department of Surgery, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan
| | - Tomoharu Yoshizumi
- Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
| | - Shinji Itoh
- Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
| | | | - Hisashi Kosaka
- Department of Surgery, Kansai Medical University, Hirakata, Japan
| | - Masaki Kaibori
- Department of Surgery, Kansai Medical University, Hirakata, Japan
| | - Takamichi Ishii
- Division of Hepato-Biliary-Pancreatic Surgery and Transplantation, Department of Surgery, Graduate School of Medicine, Kyoto University, Kyoto, Japan
| | - Etsuro Hatano
- Division of Hepato-Biliary-Pancreatic Surgery and Transplantation, Department of Surgery, Graduate School of Medicine, Kyoto University, Kyoto, Japan
| | - Chikara Ogawa
- Department of Gastroenterology and Hepatology, Takamatsu Red Cross Hospital, Takamatsu, Japan
| | - Atsushi Naganuma
- Department of Gastroenterology, NHO Takasaki General Medical Center, Takasaki, Japan
| | - Satoru Kakizaki
- Department of Clinical Research, NHO Takasaki General Medical Center, Takasaki, Japan
| | - Chih-Hsien Cheng
- Department of Liver and Transplantation Surgery, Chang-Gung Transplantation Institute, Chang Gung Memorial Hospital Linkou Branch, Taoyuan, Taiwan
| | - Po-Ting Lin
- Department of Gastroenterology and Hepatology, Chang Gung Memorial Hospital Linkou Branch, Taoyuan, Taiwan
- College of Medicine, Chang Gung University, Taoyuan, Taiwan
| | - Yung-Yeh Su
- National Institute of Cancer Research, National Health Research Institutes, Taipei, Taiwan
| | - Chien-Huai Chuang
- Department of Medical Oncology, National Taiwan University Cancer Center, Taipei, Taiwan
| | - Li-Chun Lu
- Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan
- Graduate Institute of Oncology, College of Medicine, National Taiwan University, Taipei, Taiwan
| | - Chi-Jung Wu
- Division of Gastroenterology and Hepatology, Department of Internal Medicine, Taipei Veterans General Hospital, Taipei, Taiwan
| | - Hung-Wei Wang
- Center for Digestive Medicine, Department of Internal Medicine, China Medical University Hospital, Taichung, Taiwan
| | - Kun-Ming Rau
- Department of Hematology and Oncology, E-Da Cancer Hospital, Kaohsiung, Taiwan
| | - Chih-Hung Hsu
- Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan
- Graduate Institute of Oncology, College of Medicine, National Taiwan University, Taipei, Taiwan
| | - Shi-Ming Lin
- Department of Gastroenterology and Hepatology, Chang Gung Memorial Hospital Linkou Branch, Taoyuan, Taiwan
- College of Medicine, Chang Gung University, Taoyuan, Taiwan
| | - Yi-Hsiang Huang
- Division of Gastroenterology and Hepatology, Healthcare and Service Center, Taipei Veterans General Hospital, Taipei, Taiwan
- Institute of Clinical Medicine, College of Medicine, National Yang Ming Chiao Tung University, Hsinchu, Taiwan
| | | | - Richard S Finn
- Division of Hematology/Oncology, Department of Medicine, Geffen School of Medicine at UCLA, Los Angeles, CA
| | - Masatoshi Kudo
- Department of Gastroenterology and Hepatology, Kindai University Faculty of Medicine, Osaka-sayama, Japan
| | - Ann-Lii Cheng
- Department of Medical Oncology, National Taiwan University Cancer Center, Taipei, Taiwan
- Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan
- Graduate Institute of Oncology, College of Medicine, National Taiwan University, Taipei, Taiwan
| |
Collapse
|
3
|
Zhang R, Zhu H, Chen M, Sang W, Lu K, Li Z, Wang C, Zhang L, Yin FF, Yang Z. A dual-radiomics model for overall survival prediction in early-stage NSCLC patient using pre-treatment CT images. Front Oncol 2024; 14:1419621. [PMID: 39206157 PMCID: PMC11349529 DOI: 10.3389/fonc.2024.1419621] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/18/2024] [Accepted: 07/26/2024] [Indexed: 09/04/2024] Open
Abstract
Introduction Radiation therapy (RT) is one of the primary treatment options for early-stage non-small cell lung cancer (ES-NSCLC). Therefore, accurately predicting the overall survival (OS) rate following radiotherapy is crucial for implementing personalized treatment strategies. This work aims to develop a dual-radiomics (DR) model to (1) predict 3-year OS in ES-NSCLC patients receiving RT using pre-treatment CT images, and (2) provide explanations between feature importanceand model prediction performance. Methods The publicly available TCIA Lung1 dataset with 132 ES-NSCLC patients received RT were studied: 89/43 patients in the under/over 3-year OS group. For each patient, two types of radiomic features were examined: 56 handcrafted radiomic features (HRFs) extracted within gross tumor volume, and 512 image deep features (IDFs) extracted using a pre-trained U-Net encoder. They were combined as inputs to an explainable boosting machine (EBM) model for OS prediction. The EBM's mean absolute scores for HRFs and IDFs were used as feature importance explanations. To evaluate identified feature importance, the DR model was compared with EBM using either (1) key or (2) non-key feature type only. Comparison studies with other models, including supporting vector machine (SVM) and random forest (RF), were also included. The performance was evaluated by the area under the receiver operating characteristic curve (AUCROC), accuracy, sensitivity, and specificity with a 100-fold Monte Carlo cross-validation. Results The DR model showed highestperformance in predicting 3-year OS (AUCROC=0.81 ± 0.04), and EBM scores suggested that IDFs showed significantly greater importance (normalized mean score=0.0019) than HRFs (score=0.0008). The comparison studies showed that EBM with key feature type (IDFs-only demonstrated comparable AUCROC results (0.81 ± 0.04), while EBM with non-key feature type (HRFs-only) showed limited AUCROC (0.64 ± 0.10). The results suggested that feature importance score identified by EBM is highly correlated with OS prediction performance. Both SVM and RF models were unable to explain key feature type while showing limited overall AUCROC=0.66 ± 0.07 and 0.77 ± 0.06, respectively. Accuracy, sensitivity, and specificity showed a similar trend. Discussion In conclusion, a DR model was successfully developed to predict ES-NSCLC OS based on pre-treatment CT images. The results suggested that the feature importance from DR model is highly correlated to the model prediction power.
Collapse
Affiliation(s)
- Rihui Zhang
- Medical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, China
| | - Haiming Zhu
- Medical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, China
| | - Minbin Chen
- Department of Radiotherapy & Oncology, The First People’s Hospital of Kunshan, Kunshan, Jiangsu, China
| | - Weiwei Sang
- Medical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, China
| | - Ke Lu
- Deparment of Radiation Oncology, Duke University, Durham, NC, United States
| | - Zhen Li
- Radiation Oncology Department, Shanghai Sixth People’s Hospital, Shanghai, China
| | - Chunhao Wang
- Deparment of Radiation Oncology, Duke University, Durham, NC, United States
| | - Lei Zhang
- Medical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, China
| | - Fang-Fang Yin
- Medical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, China
| | - Zhenyu Yang
- Medical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, China
| |
Collapse
|
4
|
Liu J, Sui C, Bian H, Li Y, Wang Z, Fu J, Qi L, Chen K, Xu W, Li X. Radiomics based on 18F-FDG PET/CT for prediction of pathological complete response to neoadjuvant therapy in non-small cell lung cancer. Front Oncol 2024; 14:1425837. [PMID: 39132503 PMCID: PMC11310012 DOI: 10.3389/fonc.2024.1425837] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/30/2024] [Accepted: 07/09/2024] [Indexed: 08/13/2024] Open
Abstract
Purpose This study aimed to establish and evaluate the value of integrated models involving 18F-FDG PET/CT-based radiomics and clinicopathological information in the prediction of pathological complete response (pCR) to neoadjuvant therapy (NAT) for non-small cell lung cancer (NSCLC). Methods A total of 106 eligible NSCLC patients were included in the study. After volume of interest (VOI) segmentation, 2,016 PET-based and 2,016 CT-based radiomic features were extracted. To select an optimal machine learning model, a total of 25 models were constructed based on five sets of machine learning classifiers combined with five sets of predictive feature resources, including PET-based alone radiomics, CT-based alone radiomics, PET/CT-based radiomics, clinicopathological features, and PET/CT-based radiomics integrated with clinicopathological features. Area under the curves (AUCs) of receiver operator characteristic (ROC) curves were used as the main outcome to assess the model performance. Results The hybrid PET/CT-derived radiomic model outperformed PET-alone and CT-alone radiomic models in the prediction of pCR to NAT. Moreover, addition of clinicopathological information further enhanced the predictive performance of PET/CT-derived radiomic model. Ultimately, the support vector machine (SVM)-based PET/CT radiomics combined clinicopathological information presented an optimal predictive efficacy with an AUC of 0.925 (95% CI 0.869-0.981) in the training cohort and an AUC of 0.863 (95% CI 0.740-0.985) in the test cohort. The developed nomogram involving radiomics and pathological type was suggested as a convenient tool to enable clinical application. Conclusions The 18F-FDG PET/CT-based SVM radiomics integrated with clinicopathological information was an optimal model to non-invasively predict pCR to NAC for NSCLC.
Collapse
Affiliation(s)
- Jianjing Liu
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Cancer Hospital Airport Hospital, Tianjin, China
| | - Chunxiao Sui
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
| | - Haiman Bian
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
- Department of Radiology, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
| | - Yue Li
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
- Department of Lung Cancer, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
| | - Ziyang Wang
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Cancer Hospital Airport Hospital, Tianjin, China
| | - Jie Fu
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Cancer Hospital Airport Hospital, Tianjin, China
| | - Lisha Qi
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
- Department of Pathology, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
| | - Kun Chen
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
| | - Wengui Xu
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
| | - Xiaofeng Li
- Department of Molecular Imaging and Nuclear Medicine, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
- National Clinical Research Center for Cancer, Tianjin’s Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
| |
Collapse
|
5
|
Xing X, Li L, Sun M, Yang J, Zhu X, Peng F, Du J, Feng Y. Deep-learning-based 3D super-resolution CT radiomics model: Predict the possibility of the micropapillary/solid component of lung adenocarcinoma. Heliyon 2024; 10:e34163. [PMID: 39071606 PMCID: PMC11279278 DOI: 10.1016/j.heliyon.2024.e34163] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/30/2024] [Revised: 07/03/2024] [Accepted: 07/04/2024] [Indexed: 07/30/2024] Open
Abstract
Objective Invasive lung adenocarcinoma(ILA) with micropapillary (MPP)/solid (SOL) components has a poor prognosis. Preoperative identification is essential for decision-making for subsequent treatment. This study aims to construct and evaluate a super-resolution(SR) enhanced radiomics model designed to predict the presence of MPP/SOL components preoperatively to provide more accurate and individualized treatment planning. Methods Between March 2018 and November 2023, patients who underwent curative intent ILA resection were included in the study. We implemented a deep transfer learning network on CT images to improve their resolution, resulting in the acquisition of preoperative super-resolution CT (SR-CT) images. Models were developed using radiomic features extracted from CT and SR-CT images. These models employed a range of classifiers, including Logistic Regression (LR), Support Vector Machines (SVM), k-Nearest Neighbors (KNN), Random Forest, Extra Trees, Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Multilayer Perceptron (MLP). The diagnostic performance of the models was assessed by measuring the area under the curve (AUC). Result A total of 245 patients were recruited, of which 109 (44.5 %) were diagnosed with ILA with MPP/SOL components. In the analysis of CT images, the SVM model exhibited outstanding effectiveness, recording AUC scores of 0.864 in the training group and 0.761 in the testing group. When this SVM approach was used to develop a radiomics model with SR-CT images, it recorded AUCs of 0.904 in the training and 0.819 in the test cohorts. The calibration curves indicated a high goodness of fit, while decision curve analysis (DCA) highlighted the model's clinical utility. Conclusion The study successfully constructed and evaluated a deep learning(DL)-enhanced SR-CT radiomics model. This model outperformed conventional CT radiomics models in predicting MPP/SOL patterns in ILA. Continued research and broader validation are necessary to fully harness and refine the clinical potential of radiomics when combined with SR reconstruction technology.
Collapse
Affiliation(s)
- Xiaowei Xing
- Cancer Center, Department of Radiology, Zhejiang Provincial People's Hospital, (Affiliated People's Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China
| | - Liangping Li
- Department of Radiology, Zhejiang Hospital, Hangzhou, Zhejiang, China
| | - Mingxia Sun
- Department of Radiology, Zhejiang Hospital, Hangzhou, Zhejiang, China
| | - Jiahu Yang
- Department of Radiology, Zhejiang Hospital, Hangzhou, Zhejiang, China
| | - Xinhai Zhu
- Department of Thoracic Surgery, Zhejiang Hospital, Hangzhou, Zhejiang, China
| | - Fang Peng
- Department of Pathology, Zhejiang Hospital, Hangzhou, Zhejiang, China
| | - Jianzong Du
- Department of Respiratory Medicine, Zhejiang Hospital, Hangzhou, Zhejiang, China
| | - Yue Feng
- Cancer Center, Department of Radiology, Zhejiang Provincial People's Hospital, (Affiliated People's Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China
| |
Collapse
|
6
|
Yang M, Li X, Cai C, Liu C, Ma M, Qu W, Zhong S, Zheng E, Zhu H, Jin F, Shi H. [ 18F]FDG PET-CT radiomics signature to predict pathological complete response to neoadjuvant chemoimmunotherapy in non-small cell lung cancer: a multicenter study. Eur Radiol 2024; 34:4352-4363. [PMID: 38127071 DOI: 10.1007/s00330-023-10503-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/04/2023] [Revised: 11/07/2023] [Accepted: 11/29/2023] [Indexed: 12/23/2023]
Abstract
OBJECTIVES This study aims to develop and validate a radiomics model based on 18F-fluorodeoxyglucose positron emission tomography-computed tomography ([18F]FDG PET-CT) images to predict pathological complete response (pCR) to neoadjuvant chemoimmunotherapy in non-small cell lung cancer (NSCLC). MATERIALS AND METHODS One hundred eighty-five patients receiving neoadjuvant chemoimmunotherapy for NSCLC at 5 centers from January 2019 to December 2022 were included and divided into a training cohort and a validation cohort. Radiomics models were constructed via the least absolute shrinkage and selection operator (LASSO) method. The performances of models were evaluated by the area under the receiver operating characteristic curve (AUC). In addition, genetic analyses were conducted to reveal the underlying biological basis of the radiomics score. RESULTS After the LASSO process, 9 PET-CT radiomics features were selected for pCR prediction. In the validation cohort, the ability of PET-CT radiomics model to predict pCR was shown to have an AUC of 0.818 (95% confidence interval [CI], 0.711, 0.925), which was better than the PET radiomics model (0.728 [95% CI, 0.610, 0.846]), CT radiomics model (0.732 [95% CI, 0.607, 0.857]), and maximum standard uptake value (0.603 [95% CI, 0.473, 0.733]) (p < 0.05). Moreover, a high radiomics score was related to the upregulation of pathways suppressing tumor proliferation and the infiltration of antitumor immune cell. CONCLUSION The proposed PET-CT radiomics model was capable of predicting pCR to neoadjuvant chemoimmunotherapy in NSCLC patients. CLINICAL RELEVANCE STATEMENT This study indicated that the generated 18F-fluorodeoxyglucose positron emission tomography-computed tomography radiomics model could predict pathological complete response to neoadjuvant chemoimmunotherapy, implying the potential of our radiomics model to personalize the neoadjuvant chemoimmunotherapy in lung cancer patients. KEY POINTS • Recognizing patients potentially benefiting neoadjuvant chemoimmunotherapy is critical for individualized therapy of lung cancer. • [18F]FDG PET-CT radiomics could predict pathological complete response to neoadjuvant immunotherapy in non-small cell lung cancer. • [18F]FDG PET-CT radiomics model could personalize neoadjuvant chemoimmunotherapy in lung cancer patients.
Collapse
Affiliation(s)
- Minglei Yang
- Department of Thoracic Surgery, Ningbo No. 2 Hospital, Ningbo, China
| | - Xiaoxiao Li
- Shanghai Universal Cloud Medical Imaging Diagnostic Center, Shanghai, China
| | - Chuang Cai
- School of Computer Science and Communication Engineering, Jiangsu University, Zhenjiang, Jiangsu, China
| | - Chunli Liu
- Shanghai Universal Cloud Medical Imaging Diagnostic Center, Shanghai, China
| | - Minjie Ma
- Department of Thoracic Surgery, The First Hospital of Lanzhou University, Lanzhou, Gansu, China
| | - Wendong Qu
- Department of Thoracic Surgery, The Affiliated Hospital of Zunyi Medical University, Zunyi, Guizhou, China.
| | | | - Enkuo Zheng
- Department of Thoracic Surgery, Ningbo No. 2 Hospital, Ningbo, China
| | - Huangkai Zhu
- Department of Thoracic Surgery, Ningbo No. 2 Hospital, Ningbo, China
| | - Feng Jin
- Shandong Key Laboratory of Infectious Respiratory Diseases, Shandong Public Health Clinical Center, Shandong University, Shandong, China.
| | - Huazheng Shi
- Shanghai Universal Cloud Medical Imaging Diagnostic Center, Shanghai, China.
| |
Collapse
|
7
|
Zhang M, Zhu L, Liang S, Mao Z, Li X, Yang L, Yang Y, Wang K, Wang P, Chen W. Pulmonary function test-related prognostic models in non-small cell lung cancer patients receiving neoadjuvant chemoimmunotherapy. Front Oncol 2024; 14:1411436. [PMID: 38983930 PMCID: PMC11231186 DOI: 10.3389/fonc.2024.1411436] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/03/2024] [Accepted: 06/11/2024] [Indexed: 07/11/2024] Open
Abstract
Background This study aimed to establish a comprehensive clinical prognostic risk model based on pulmonary function tests. This model was intended to guide the evaluation and predictive management of patients with resectable stage I-III non-small cell lung cancer (NSCLC) receiving neoadjuvant chemoimmunotherapy. Methods Clinical pathological characteristics and prognostic survival data for 175 patients were collected. Univariate and multivariate Cox regression analyses, and least absolute shrinkage and selection operator (LASSO) regression analysis were employed to identify variables and construct corresponding models. These variables were integrated to develop a ridge regression model. The models' discrimination and calibration were evaluated, and the optimal model was chosen following internal validation. Comparative analyses between the risk scores or groups of the optimal model and clinical factors were conducted to explore the potential clinical application value. Results Univariate regression analysis identified smoking, complete pathologic response (CPR), and major pathologic response (MPR) as protective factors. Conversely, T staging, D-dimer/white blood cell ratio (DWBCR), D-dimer/fibrinogen ratio (DFR), and D-dimer/minute ventilation volume actual ratio (DMVAR) emerged as risk factors. Evaluation of the models confirmed their capability to accurately predict patient prognosis, exhibiting ideal discrimination and calibration, with the ridge regression model being optimal. Survival analysis demonstrated that the disease-free survival (DFS) in the high-risk group (HRG) was significantly shorter than in the low-risk group (LRG) (P=2.57×10-13). The time-dependent receiver operating characteristic (ROC) curve indicated that the area under the curve (AUC) values at 1 year, 2 years, and 3 years were 0.74, 0.81, and 0.79, respectively. Clinical correlation analysis revealed that men with lung squamous cell carcinoma or comorbid chronic obstructive pulmonary disease (COPD) were predominantly in the LRG, suggesting a better prognosis and potentially identifying a beneficiary population for this treatment combination. Conclusion The prognostic model developed in this study effectively predicts the prognosis of patients with NSCLC receiving neoadjuvant chemoimmunotherapy. It offers valuable predictive insights for clinicians, aiding in developing treatment plans and monitoring disease progression.
Collapse
Affiliation(s)
- Min Zhang
- Department of Respiratory and Critical Care Medicine, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, China
| | - Liang Zhu
- Department of Rheumatology, the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
| | - Sibei Liang
- Department of Respiratory and Critical Care Medicine, Center for Oncology Medicine, the Fourth Affiliated Hospital of School of Medicine, and International School of Medicine, International Institutes of Medicine, Zhejiang University, Yiwu, China
- Zhejiang Key Laboratory of Precision Diagnosis and Treatment for Lung Cancer, Yiwu, China
| | - Zhirong Mao
- Department of Respiratory and Critical Care Medicine, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, China
| | - Xiaolin Li
- Department of Nutrition, the Fourth Affiliated Hospital of School of Medicine, and International School of Medicine, International Institutes of Medicine, Zhejiang University, Yiwu, China
| | - Lingge Yang
- Department of Respiratory and Critical Care Medicine, Center for Oncology Medicine, the Fourth Affiliated Hospital of School of Medicine, and International School of Medicine, International Institutes of Medicine, Zhejiang University, Yiwu, China
- Zhejiang Key Laboratory of Precision Diagnosis and Treatment for Lung Cancer, Yiwu, China
| | - Yan Yang
- Department of Respiratory and Critical Care Medicine, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, China
| | - Kai Wang
- Department of Respiratory and Critical Care Medicine, Center for Oncology Medicine, the Fourth Affiliated Hospital of School of Medicine, and International School of Medicine, International Institutes of Medicine, Zhejiang University, Yiwu, China
- Zhejiang Key Laboratory of Precision Diagnosis and Treatment for Lung Cancer, Yiwu, China
| | - Pingli Wang
- Department of Respiratory and Critical Care Medicine, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, China
| | - Weiyu Chen
- Department of Respiratory and Critical Care Medicine, Center for Oncology Medicine, the Fourth Affiliated Hospital of School of Medicine, and International School of Medicine, International Institutes of Medicine, Zhejiang University, Yiwu, China
- Zhejiang Key Laboratory of Precision Diagnosis and Treatment for Lung Cancer, Yiwu, China
| |
Collapse
|
8
|
Ye G, Wu G, Zhang C, Wang M, Liu H, Song E, Zhuang Y, Li K, Qi Y, Liao Y. CT-based quantification of intratumoral heterogeneity for predicting pathologic complete response to neoadjuvant immunochemotherapy in non-small cell lung cancer. Front Immunol 2024; 15:1414954. [PMID: 38933281 PMCID: PMC11199789 DOI: 10.3389/fimmu.2024.1414954] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/09/2024] [Accepted: 05/29/2024] [Indexed: 06/28/2024] Open
Abstract
Objectives To investigate the prediction of pathologic complete response (pCR) in patients with non-small cell lung cancer (NSCLC) undergoing neoadjuvant immunochemotherapy (NAIC) using quantification of intratumoral heterogeneity from pre-treatment CT image. Methods This retrospective study included 178 patients with NSCLC who underwent NAIC at 4 different centers. The training set comprised 108 patients from center A, while the external validation set consisted of 70 patients from center B, center C, and center D. The traditional radiomics model was contrasted using radiomics features. The radiomics features of each pixel within the tumor region of interest (ROI) were extracted. The optimal division of tumor subregions was determined using the K-means unsupervised clustering method. The internal tumor heterogeneity habitat model was developed using the habitats features from each tumor sub-region. The LR algorithm was employed in this study to construct a machine learning prediction model. The diagnostic performance of the model was evaluated using criteria such as area under the receiver operating characteristic curve (AUC), accuracy, specificity, sensitivity, positive predictive value (PPV), and negative predictive value (NPV). Results In the training cohort, the traditional radiomics model achieved an AUC of 0.778 [95% confidence interval (CI): 0.688-0.868], while the tumor internal heterogeneity habitat model achieved an AUC of 0.861 (95% CI: 0.789-0.932). The tumor internal heterogeneity habitat model exhibits a higher AUC value. It demonstrates an accuracy of 0.815, surpassing the accuracy of 0.685 achieved by traditional radiomics models. In the external validation cohort, the AUC values of the two models were 0.723 (CI: 0.591-0.855) and 0.781 (95% CI: 0.673-0.889), respectively. The habitat model continues to exhibit higher AUC values. In terms of accuracy evaluation, the tumor heterogeneity habitat model outperforms the traditional radiomics model, achieving a score of 0.743 compared to 0.686. Conclusion The quantitative analysis of intratumoral heterogeneity using CT to predict pCR in NSCLC patients undergoing NAIC holds the potential to inform clinical decision-making for resectable NSCLC patients, prevent overtreatment, and enable personalized and precise cancer management.
Collapse
Affiliation(s)
- Guanchao Ye
- Department of Thoracic Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
- Department of Thoracic Surgery, the First Affiliated Hospital of Zhengzhou University, Zhengzhou University, Zhengzhou, China
| | - Guangyao Wu
- Department of Radiology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
| | - Chunyang Zhang
- Department of Thoracic Surgery, the First Affiliated Hospital of Zhengzhou University, Zhengzhou University, Zhengzhou, China
| | - Mingliang Wang
- Department of Thoracic Surgery, Henan Provincial People’s Hospital, Zhengzhou University, Zhengzhou, China
| | - Hong Liu
- School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China
| | - Enmin Song
- School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China
| | - Yuzhou Zhuang
- School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China
| | - Kuo Li
- Department of Thoracic Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
| | - Yu Qi
- Department of Thoracic Surgery, the First Affiliated Hospital of Zhengzhou University, Zhengzhou University, Zhengzhou, China
| | - Yongde Liao
- Department of Thoracic Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
| |
Collapse
|
9
|
Mei T, Wang T, Zhou Q. Multi-omics and artificial intelligence predict clinical outcomes of immunotherapy in non-small cell lung cancer patients. Clin Exp Med 2024; 24:60. [PMID: 38554212 PMCID: PMC10981593 DOI: 10.1007/s10238-024-01324-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2023] [Accepted: 03/05/2024] [Indexed: 04/01/2024]
Abstract
In recent years, various types of immunotherapy, particularly the use of immune checkpoint inhibitors targeting programmed cell death 1 or programmed death ligand 1 (PD-L1), have revolutionized the management and prognosis of non-small cell lung cancer. PD-L1 is frequently used as a biomarker for predicting the likely benefit of immunotherapy for patients. However, some patients receiving immunotherapy have high response rates despite having low levels of PD-L1. Therefore, the identification of this group of patients is extremely important to improve prognosis. The tumor microenvironment contains tumor, stromal, and infiltrating immune cells with its composition differing significantly within tumors, between tumors, and between individuals. The omics approach aims to provide a comprehensive assessment of each patient through high-throughput extracted features, promising a more comprehensive characterization of this complex ecosystem. However, features identified by high-throughput methods are complex and present analytical challenges to clinicians and data scientists. It is thus feasible that artificial intelligence could assist in the identification of features that are beyond human discernment as well as in the performance of repetitive tasks. In this paper, we review the prediction of immunotherapy efficacy by different biomarkers (genomic, transcriptomic, proteomic, microbiomic, and radiomic), together with the use of artificial intelligence and the challenges and future directions of these fields.
Collapse
Affiliation(s)
- Ting Mei
- Department of Medical Oncology, Cancer Center, West China Hospital, Sichuan University, Chengdu, Sichuan, China
- Lung Cancer Center, West China Hospital, Sichuan University, Chengdu, 610000, China
| | - Ting Wang
- Department of Medical Oncology, Cancer Center, West China Hospital, Sichuan University, Chengdu, Sichuan, China
- Lung Cancer Center, West China Hospital, Sichuan University, Chengdu, 610000, China
| | - Qinghua Zhou
- Department of Medical Oncology, Cancer Center, West China Hospital, Sichuan University, Chengdu, Sichuan, China.
- Lung Cancer Center, West China Hospital, Sichuan University, Chengdu, 610000, China.
| |
Collapse
|
10
|
Schöneck M, Lennartz S, Zopfs D, Sonnabend K, Wawer Matos Reimer R, Rinneburger M, Graffe J, Persigehl T, Hentschke C, Baeßler B, Lourenco Caldeira L, Große Hokamp N. Robustness of radiomic features in healthy abdominal parenchyma of patients with repeated examinations on dual-layer dual-energy CT. Eur J Radiol 2024; 175:111447. [PMID: 38677039 DOI: 10.1016/j.ejrad.2024.111447] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/06/2023] [Revised: 03/19/2024] [Accepted: 03/25/2024] [Indexed: 04/29/2024]
Abstract
OBJECTIVES Robustness of radiomic features in physiological tissue is an important prerequisite for quantitative analysis of tumor biology and response assessment. In contrast to previous studies which focused on different tumors with mostly short scan-re-scan intervals, this study aimed to evaluate the robustness of radiomic features in cancer-free patients and over a clinically encountered inter-scan interval. MATERIALS AND METHODS Patients without visible tumor burden who underwent at least two portal-venous phase dual energy CT examinations of the abdomen between May 2016 and January 2020 were included, while macroscopic tumor burden was excluded based upon follow-up imaging for all patients (≥3 months). Further, patients were excluded if no follow-up imaging was available, or if the CT protocol showed deviations between repeated examinations. Circular regions of interest were placed and proofread by two board-certified radiologists (4 years and 5 years experience) within the liver (segments 3 and 6), the psoas muscle (left and right), the pancreatic head, and the spleen to obtain radiomic features from normal-appearing organ parenchyma using PyRadiomics. Radiomic feature robustness was tested using the concordance correlation coefficient with a threshold of 0.75 considered indicative for deeming a feature robust. RESULTS In total, 160 patients with 480 repeated abdominal CT examinations (range: 2-4 per patient) were retrospectively included in this single-center, IRB-approved study. Considering all organs and feature categories, only 4.58 % (25/546) of all features were robust with the highest rate being found in the first order feature category (20.37 %, 22/108). Other feature categories (grey level co-occurrence matrix, grey level dependence matrix, grey level run length matrix, grey level size zone matrix, and neighborhood gray-tone difference matrix) yielded an overall low percentage of robust features (range: 0.00 %-1.19 %). A subgroup analysis revealed the reconstructed field of view and the X-ray tube current as determinants of feature robustness (significant differences in subgroups for all organs, p < 0.001) as well as the size of the region of interest (no significant difference for the pancreatic head with p = 0.135, significant difference with p < 0.001 for all other organs). CONCLUSION Radiomic feature robustness obtained from cancer-free subjects with repeated examinations using a consistent protocol and CT scanner was limited, with first order features yielding the highest proportion of robust features.
Collapse
Affiliation(s)
- Mirjam Schöneck
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany.
| | - Simon Lennartz
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| | - David Zopfs
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| | - Kristina Sonnabend
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany; Philips Healthcare Market DACH, Röntgenstraße 22, 22335 Hamburg, Germany
| | - Robert Wawer Matos Reimer
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| | - Miriam Rinneburger
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| | - Josefine Graffe
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| | - Thorsten Persigehl
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| | | | - Bettina Baeßler
- University Hospital Würzburg, Department of Diagnostic and Interventional Radiology, Oberdürrbacher Straße 6, 97080 Würzburg, Germany
| | - Liliana Lourenco Caldeira
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| | - Nils Große Hokamp
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Institute for Diagnostic and Interventional Radiology, Kerpener Straße 62, 50937 Cologne, Germany
| |
Collapse
|
11
|
Choi W, Jia Y, Kwak J, Werner-Wasik M, Dicker AP, Simone NL, Storozynsky E, Jain V, Vinogradskiy Y. Novel Functional Radiomics for Prediction of Cardiac Positron Emission Tomography Avidity in Lung Cancer Radiotherapy. JCO Clin Cancer Inform 2024; 8:e2300241. [PMID: 38452302 PMCID: PMC10939651 DOI: 10.1200/cci.23.00241] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/16/2023] [Revised: 12/22/2023] [Accepted: 01/26/2024] [Indexed: 03/09/2024] Open
Abstract
PURPOSE Traditional methods of evaluating cardiotoxicity focus on radiation doses to the heart. Functional imaging has the potential to provide improved prediction for cardiotoxicity for patients with lung cancer. Fluorine-18 (18F) fluorodeoxyglucose (FDG)-positron emission tomography (PET)/computed tomography (CT) imaging is routinely obtained in a standard cancer staging workup. This work aimed to develop a radiomics model predicting clinical cardiac assessment using 18F-FDG PET/CT scans before thoracic radiation therapy. METHODS Pretreatment 18F-FDG PET/CT scans from three study populations (N = 100, N = 39, N = 70) were used, comprising two single-institutional protocols and one publicly available data set. A clinician (V.J.) classified the PET/CT scans per clinical cardiac guidelines as no uptake, diffuse uptake, or focal uptake. The heart was delineated, and 210 novel functional radiomics features were selected to classify cardiac FDG uptake patterns. Training data were divided into training (80%)/validation (20%) sets. Feature reduction was performed using the Wilcoxon test, hierarchical clustering, and recursive feature elimination. Ten-fold cross-validation was carried out for training, and the accuracy of the models to predict clinical cardiac assessment was reported. RESULTS From 202 of 209 scans, cardiac FDG uptake was scored as no uptake (39.6%), diffuse uptake (25.3%), and focal uptake (35.1%), respectively. Sixty-two independent radiomics features were reduced to nine clinically pertinent features. The best model showed 93% predictive accuracy in the training data set and 80% and 92% predictive accuracy in two external validation data sets. CONCLUSION This work used an extensive patient data set to develop a functional cardiac radiomic model from standard-of-care 18F-FDG PET/CT scans, showing good predictive accuracy. The radiomics model has the potential to provide an automated method to predict existing cardiac conditions and provide an early functional biomarker to identify patients at risk of developing cardiac complications after radiotherapy.
Collapse
Affiliation(s)
- Wookjin Choi
- Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| | - Yingcui Jia
- Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| | - Jennifer Kwak
- Department of Radiology, University of Colorado School of Medicine, Aurora, CO
| | - Maria Werner-Wasik
- Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| | - Adam P. Dicker
- Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| | - Nicole L. Simone
- Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| | - Eugene Storozynsky
- Department of Cardiology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| | - Varsha Jain
- Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| | - Yevgeniy Vinogradskiy
- Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA
| |
Collapse
|
12
|
Yang N, Yue H, Zhang B, Chen J, Chu Q, Wang J, Yu X, Jian L, Bin Y, Liu S, Liu J, Zeng L, Yang H, Zhou C, Jiang W, Liu L, Zhang Y, Xiong Y, Wang Z. Predicting pathological response to neoadjuvant or conversion chemoimmunotherapy in stage IB-III non-small cell lung cancer patients using radiomic features. Thorac Cancer 2023; 14:2869-2876. [PMID: 37596822 PMCID: PMC10542462 DOI: 10.1111/1759-7714.15052] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/19/2023] [Revised: 07/12/2023] [Accepted: 07/14/2023] [Indexed: 08/20/2023] Open
Abstract
BACKGROUND To develop a radiomics model based on chest computed tomography (CT) for the prediction of a pathological complete response (pCR) after neoadjuvant or conversion chemoimmunotherapy (CIT) in patients with non-small cell lung cancer (NSCLC). METHODS Patients with stage IB-III NSCLC who received neoadjuvant or conversion CIT between September 2019 and July 2021 at Hunan Cancer Hospital, Xiangya Hospital, and Union Hospital were retrospectively collected. The least absolute shrinkage and selection operator (LASSO) were used to screen features. Then, model 1 (five radiomics features before CIT), model 2 (four radiomics features after CIT and before surgery) and model 3 were constructed for the prediction of pCR. Model 3 included all nine features of model 1 and 2 and was later named the neoadjuvant chemoimmunotherapy-related pathological response prediction model (NACIP). RESULTS This study included 110 patients: 77 in the training set and 33 in the validation set. Thirty-nine (35.5%) patients achieved a pCR. Model 1 showed area under the curve (AUC) = 0.65, 64% accuracy, 71% specificity, and 50% sensitivity, while model 2 displayed AUC = 0.81, 73% accuracy, 62% specificity, and 92% sensitivity. In comparison, NACIP yielded a good predictive value, with an AUC of 0.85, 81% accuracy, 81% specificity, and 83% sensitivity in the validation set. CONCLUSION NACIP may be a potential model for the early prediction of pCR in patients with NSCLC treated with neoadjuvant/conversion CIT.
Collapse
Affiliation(s)
- Nong Yang
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
- Lung Cancer and Gastrointestinal Unit, Department of Medical Oncology, Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of MedicineCentral South UniversityChangshaChina
| | - Hai‐Lin Yue
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Bai‐Hua Zhang
- Department of Thoracic SurgeryHunan Cancer HospitalChangshaChina
| | - Juan Chen
- Department of Pharmacy, Xiangya HospitalCentral South UniversityChangshaChina
| | - Qian Chu
- Department of Oncology, Tongji Hospital, Tongji Medical CollegeHuazhong University of Science and TechnologyWuhanChina
| | - Jian‐Xin Wang
- Lung Cancer and Gastrointestinal Unit, Department of Medical Oncology, Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of MedicineCentral South UniversityChangshaChina
| | - Xiao‐Ping Yu
- Department of Diagnostic Radiology, Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of MedicineCentral South UniversityChangshaChina
| | - Lian Jian
- Department of Diagnostic Radiology, Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of MedicineCentral South UniversityChangshaChina
| | - Ya‐Wen Bin
- Cancer Center, Union Hospital, Tongji Medical CollegeHuazhong University of Science and TechnologyWuhanChina
| | - Si‐Ye Liu
- Department of Diagnostic Radiology, Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of MedicineCentral South UniversityChangshaChina
| | - Jin Liu
- Lung Cancer and Gastrointestinal Unit, Department of Medical Oncology, Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of MedicineCentral South UniversityChangshaChina
| | - Liang Zeng
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Hai‐Yan Yang
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Chun‐Hua Zhou
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Wen‐Juan Jiang
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Li Liu
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Yong‐Chang Zhang
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Yi Xiong
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| | - Zhan Wang
- Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and EngineeringCentral South UniversityChangshaChina
| |
Collapse
|
13
|
Kong X, Mao Y, Xi F, Li Y, Luo Y, Ma J. Development of a nomogram based on radiomics and semantic features for predicting chromosome 7 gain/chromosome 10 loss in IDH wild-type histologically low-grade gliomas. Front Oncol 2023; 13:1196614. [PMID: 37781185 PMCID: PMC10541227 DOI: 10.3389/fonc.2023.1196614] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/30/2023] [Accepted: 08/29/2023] [Indexed: 10/03/2023] Open
Abstract
Purpose To predict chromosome 7 gain and chromosome 10 loss (+7/-10) in IDH wild-type (IDH-wt) histologically low-grade gliomas (LGG) by machine learning models based on MRI radiomics and semantic features. Methods A total of 122 patients diagnosed as IDH-wt histologically LGG were retrospectively included in this study. The patients were randomly divided into a training group and a test group in a ratio of 7:3. The radiomics features were extracted from axial T1WI, T2WI, FLAIR and CET1 sequences, respectively. The distance correlation (DC) and least absolute shrinkage and selection operator (LASSO) were used to select the radiomics signatures. Three machine learning algorithms including neural network (NN), support vector machine (SVM), and linear discriminant analysis (LDA) were used to construct radiomics models. In addition, a nomogram was developed by combining the optimal radiomics signature with clinical risk factors, and the potential clinical utility of the nomogram was evaluated using decision curve analysis. Results The LDA+DC model was identified as the optimal classifier among the six radiomics models. Necrosis was determined as a risk factor for +7/-10 in IDH-wt histologically LGG. The nomogram achieved the best performance, with an AUC of 0.854 and an accuracy of 0.778 in the independent test group. The decision curve of the nomogram confirmed its clinical usefulness in a wide range of thresholds. Conclusion The nomogram combining radiomics and semantic features can predict the +7/-10 status effectively, which may contribute to the risk stratification and individualized treatment planning of patients with IDH-wt histologically LGG.
Collapse
Affiliation(s)
- Xin Kong
- Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Yu Mao
- Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Fengjun Xi
- Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Yan Li
- Department of Radiology, Beijing Fengtai Hospital, Beijing, China
| | - Yuqi Luo
- Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Jun Ma
- Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| |
Collapse
|
14
|
Martin FL, Dickinson AW, Saba T, Bongers T, Singh MN, Bury D. ATR-FTIR Spectroscopy with Chemometrics for Analysis of Saliva Samples Obtained in a Lung-Cancer-Screening Programme: Application of Swabs as a Paradigm for High Throughput in a Clinical Setting. J Pers Med 2023; 13:1039. [PMID: 37511652 PMCID: PMC10381591 DOI: 10.3390/jpm13071039] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/12/2023] [Revised: 06/18/2023] [Accepted: 06/19/2023] [Indexed: 07/30/2023] Open
Abstract
There is an increasing need for inexpensive and rapid screening tests in point-of-care clinical oncology settings. Herein, we develop a swab "dip" test in saliva obtained from consenting patients participating in a lung-cancer-screening programme being undertaken in North West England. In a pilot study, a total of 211 saliva samples (n = 170 benign, 41 designated cancer-positive) were randomly taken during the course of this prospective lung-cancer-screening programme. The samples (sterile Copan blue rayon swabs dipped in saliva) were analysed using attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy. An exploratory analysis using principal component analysis (PCA,) with or without linear discriminant analysis (LDA), was then undertaken. Three pairwise comparisons were undertaken including: (1) benign vs. cancer following swab analysis; (2) benign vs. cancer following swab analysis with the subtraction of dry swab spectra; and (3) benign vs. cancer following swab analysis with the subtraction of wet swab spectra. Consistent and remarkably similar patterns of clustering for the benign control vs. cancer categories, irrespective of whether the swab plus saliva sample was analysed or whether there was a subtraction of wet or dry swab spectra, was observed. In each case, MANOVA demonstrated that this segregation of categories is highly significant. A k-NN (using three nearest neighbours) machine-learning algorithm also showed that the specificity (90%) and sensitivity (75%) are consistent for each pairwise comparison. In detailed analyses, the swab as a substrate did not alter the level of spectral discrimination between benign control vs. cancer saliva samples. These results demonstrate a novel swab "dip" test using saliva as a biofluid that is highly applicable to be rolled out into a larger lung-cancer-screening programme.
Collapse
Affiliation(s)
- Francis L Martin
- Biocel UK Ltd., Hull HU10 6TS, UK
- Department of Cellular Pathology, Blackpool Teaching Hospitals NHS Foundation Trust, Whinney Heys Road, Blackpool FY3 8NR, UK
| | - Andrew W Dickinson
- Department of Cellular Pathology, Blackpool Teaching Hospitals NHS Foundation Trust, Whinney Heys Road, Blackpool FY3 8NR, UK
| | - Tarek Saba
- Department of Cellular Pathology, Blackpool Teaching Hospitals NHS Foundation Trust, Whinney Heys Road, Blackpool FY3 8NR, UK
| | - Thomas Bongers
- Department of Cellular Pathology, Blackpool Teaching Hospitals NHS Foundation Trust, Whinney Heys Road, Blackpool FY3 8NR, UK
| | - Maneesh N Singh
- Biocel UK Ltd., Hull HU10 6TS, UK
- Chesterfield Royal Hospital, Chesterfield Road, Calow, Chesterfield S44 5BL, UK
| | | |
Collapse
|
15
|
Schroeder KE, Acharya L, Mani H, Furqan M, Sieren JC. Radiomic biomarkers from chest computed tomography are assistive in immunotherapy response prediction for non-small cell lung cancer. Transl Lung Cancer Res 2023; 12:1023-1033. [PMID: 37323179 PMCID: PMC10261870 DOI: 10.21037/tlcr-22-763] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/24/2022] [Accepted: 04/12/2023] [Indexed: 06/17/2023]
Abstract
Background Immunotherapies, such as programmed death 1/programmed death ligand 1 (PD-1/PD-L1) antibodies have been shown to improve overall and progression-free survival (PFS) in patients with locally advanced or metastatic non-small cell lung cancer (NSCLC). However, not all patients derive a meaningful clinical benefit. Additionally, patients receiving anti-PD-1/PD-L1 therapy can experience immune-related adverse events (irAEs). Clinically significant irAEs may require temporary pause or discontinuation of treatment. Having a tool to identify patients who may not benefit and/or are at risk for developing severe irAEs from immunotherapy will aid in an informed decision-making process for the patients and their physicians. Methods Computed tomography (CT) scans and clinical data were retrospectively collected for this study to develop three prediction models using (I) radiomic features, (II) clinical features, and (III) radiomic and clinical features combined. Each subject had 6 clinical features and 849 radiomic features extracted. Selected features were run through an artificial neural network (NN) trained on 70% of the cohort, maintaining the case and control ratio. The NN was assessed by calculating the area-under-the-receiver-operating-characteristic curve (AUC-ROC), area-under-the-precision-recall curve (AUC-PR), sensitivity, and specificity. Results A cohort of 132 subjects, of which 43 (33%) had a PFS ≤90 days and 89 (67%) of which had a PFS >90 days was used to develop the prediction models. The radiomic model was able to predict progression-free survival with a training AUC-ROC of 87% and testing AUC-ROC, sensitivity, and specificity of 83%, 75%, and 81%, respectively. In this cohort, the clinical and radiomic combined features did add a slight increase in the specificity (85%) but with a decrease in sensitivity (75%) and AUC-ROC (81%). Conclusions Whole lung segmentation and feature extraction can identify those that would see a benefit from anti-PD-1/PD-L1 therapy.
Collapse
Affiliation(s)
| | - Luna Acharya
- Department of Internal Medicine, Hematology, Oncology and Blood and Marrow Transplantation, University of Iowa, Iowa City, IA, USA
| | - Hariharasudan Mani
- Department of Internal Medicine, Hematology, Oncology and Blood and Marrow Transplantation, University of Iowa, Iowa City, IA, USA
| | - Muhammad Furqan
- Department of Internal Medicine, Hematology, Oncology and Blood and Marrow Transplantation, University of Iowa, Iowa City, IA, USA
- Holden Comprehensive Cancer Center, University of Iowa, Iowa City, IA, USA
| | - Jessica C. Sieren
- Department of Radiology, University of Iowa, Iowa City, IA, USA
- Holden Comprehensive Cancer Center, University of Iowa, Iowa City, IA, USA
- Roy J. Carver Department of Biomedical Engineering, University of Iowa, Iowa City, IA, USA
| |
Collapse
|
16
|
Xu M, Yang H, Yang Q, Teng P, Hao H, Liu C, Yu S, Liu G. Radiomics nomogram based on digital breast tomosynthesis: preoperative evaluation of axillary lymph node metastasis in breast carcinoma. J Cancer Res Clin Oncol 2023:10.1007/s00432-023-04859-z. [PMID: 37208454 DOI: 10.1007/s00432-023-04859-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/25/2023] [Accepted: 05/13/2023] [Indexed: 05/21/2023]
Abstract
PURPOSE This study aimed to establish a radiomics nomogram model based on digital breast tomosynthesis (DBT) images, to predict the status of axillary lymph nodes (ALN) in patients with breast carcinoma. METHODS The data of 120 patients with confirmed breast carcinoma, including 49 cases with axillary lymph node metastasis (ALNM), were retrospectively analyzed in this study. The dataset was randomly divided into a training group consisting of 84 patients (37 with ALNM) and a validation group comprising 36 patients (12 with ALNM). Clinical information was collected for all cases, and radiomics features were extracted from DBT images. Feature selection was performed to develop the Radscore model. Univariate and multivariate logistic regression analysis were employed to identify independent risk factors for constructing both the clinical model and nomogram model. To evaluate the performance of these models, receiver operating characteristic (ROC) curve analysis, calibration curve, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discriminatory improvement (IDI) were conducted. RESULTS The clinical model identified tumor margin and DBT_reported_LNM as independent risk factors, while the Radscore model was constructed using 9 selected radiomics features. Incorporating tumor margin, DBT_reported_LNM, and Radscore, the radiomics nomogram model exhibited superior performance with AUC values of 0.933 and 0.920 in both datasets, respectively. The NRI and IDI showed a significant improvement, suggesting that the Radscore may serve as a useful biomarker for predicting ALN status. CONCLUSION The radiomics nomogram based on DBT demonstrated effective preoperative prediction performance for ALNM in patients with breast cancer.
Collapse
Affiliation(s)
- Maolin Xu
- Department of Radiology, China-Japan Union Hospital of Jilin University, Xiantai Street, Changchun, 130033, China
| | - Huimin Yang
- Department of Radiology, China-Japan Union Hospital of Jilin University, Xiantai Street, Changchun, 130033, China
| | - Qi Yang
- Department of Radiology, The First Hospital of Jilin University, No.71 Xinmin Street, Changchun, 130012, China.
| | - Peihong Teng
- Department of Radiology, China-Japan Union Hospital of Jilin University, Xiantai Street, Changchun, 130033, China
| | - Haifeng Hao
- Department of Radiology, China-Japan Union Hospital of Jilin University, Xiantai Street, Changchun, 130033, China
| | - Chang Liu
- Department of Radiology, China-Japan Union Hospital of Jilin University, Xiantai Street, Changchun, 130033, China
| | - Shaonan Yu
- Department of Radiology, China-Japan Union Hospital of Jilin University, Xiantai Street, Changchun, 130033, China.
| | - Guifeng Liu
- Department of Radiology, China-Japan Union Hospital of Jilin University, Xiantai Street, Changchun, 130033, China.
| |
Collapse
|
17
|
Wang Y, Huang S, Feng X, Xu W, Luo R, Zhu Z, Zeng Q, He Z. Advances in efficacy prediction and monitoring of neoadjuvant immunotherapy for non-small cell lung cancer. Front Oncol 2023; 13:1145128. [PMID: 37265800 PMCID: PMC10229830 DOI: 10.3389/fonc.2023.1145128] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/15/2023] [Accepted: 05/03/2023] [Indexed: 06/03/2023] Open
Abstract
The use of immune checkpoint inhibitors (ICIs) has become mainstream in the treatment of non-small cell lung cancer (NSCLC). The idea of harnessing the immune system to fight cancer is fast developing. Neoadjuvant treatment in NSCLC is undergoing unprecedented change. Chemo-immunotherapy combinations not only seem to achieve population-wide treating coverage irrespective of PD-L1 expression but also enable achieving a pathological complete response (pCR). Despite these recent advancements in neoadjuvant chemo-immunotherapy, not all patients respond favorably to treatment with ICIs plus chemo and may even suffer from severe immune-related adverse effects (irAEs). Similar to selection for target therapy, identifying patients most likely to benefit from chemo-immunotherapy may be valuable. Recently, several prognostic and predictive factors associated with the efficacy of neoadjuvant immunotherapy in NSCLC, such as tumor-intrinsic biomarkers, tumor microenvironment biomarkers, liquid biopsies, microbiota, metabolic profiles, and clinical characteristics, have been described. However, a specific and sensitive biomarker remains to be identified. Recently, the construction of prediction models for ICI therapy using novel tools, such as multi-omics factors, proteomic tests, host immune classifiers, and machine learning algorithms, has gained attention. In this review, we provide a comprehensive overview of the different positive prognostic and predictive factors in treating preoperative patients with ICIs, highlight the recent advances made in the efficacy prediction of neoadjuvant immunotherapy, and provide an outlook for joint predictors.
Collapse
Affiliation(s)
- Yunzhen Wang
- Department of Thoracic Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China
| | - Sha Huang
- Department of Thoracic Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China
| | - Xiangwei Feng
- Department of Thoracic Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China
| | - Wangjue Xu
- Department of Thoracic Surgery, Longyou County People’s Hospital, Longyou, China
| | - Raojun Luo
- Department of Thoracic Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China
| | - Ziyi Zhu
- Department of Thoracic Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China
| | - Qingxin Zeng
- Department of Thoracic Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China
| | - Zhengfu He
- Department of Thoracic Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China
| |
Collapse
|
18
|
Wang MY, Jia CG, Xu HQ, Xu CS, Li X, Wei W, Chen JC. Development and Validation of a Deep Learning Predictive Model Combining Clinical and Radiomic Features for Short-Term Postoperative Facial Nerve Function in Acoustic Neuroma Patients. Curr Med Sci 2023; 43:336-343. [PMID: 37059936 PMCID: PMC10103675 DOI: 10.1007/s11596-023-2713-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/15/2022] [Accepted: 01/14/2023] [Indexed: 04/16/2023]
Abstract
OBJECTIVE This study aims to construct and validate a predictable deep learning model associated with clinical data and multi-sequence magnetic resonance imaging (MRI) for short-term postoperative facial nerve function in patients with acoustic neuroma. METHODS A total of 110 patients with acoustic neuroma who underwent surgery through the retrosigmoid sinus approach were included. Clinical data and raw features from four MRI sequences (T1-weighted, T2-weighted, T1-weighted contrast enhancement, and T2-weighted-Flair images) were analyzed. Spearman correlation analysis along with least absolute shrinkage and selection operator regression were used to screen combined clinical and radiomic features. Nomogram, machine learning, and convolutional neural network (CNN) models were constructed to predict the prognosis of facial nerve function on the seventh day after surgery. Receiver operating characteristic (ROC) curve and decision curve analysis (DCA) were used to evaluate model performance. A total of 1050 radiomic parameters were extracted, from which 13 radiomic and 3 clinical features were selected. RESULTS The CNN model performed best among all prediction models in the test set with an area under the curve (AUC) of 0.89 (95% CI, 0.84-0.91). CONCLUSION CNN modeling that combines clinical and multi-sequence MRI radiomic features provides excellent performance for predicting short-term facial nerve function after surgery in patients with acoustic neuroma. As such, CNN modeling may serve as a potential decision-making tool for neurosurgery.
Collapse
Affiliation(s)
- Meng-Yang Wang
- Department of Neurosurgery, Zhongnan Hospital of Wuhan University, Wuhan, 430071, China
| | - Chen-Guang Jia
- Department of Neurosurgery, Zhongnan Hospital of Wuhan University, Wuhan, 430071, China
| | - Huan-Qing Xu
- School of Medical Information Engineering, Anhui University of Chinese Medicine, Hefei, 230012, China
| | - Cheng-Shi Xu
- Department of Neurosurgery, Zhongnan Hospital of Wuhan University, Wuhan, 430071, China
| | - Xiang Li
- Department of Neurosurgery, Zhongnan Hospital of Wuhan University, Wuhan, 430071, China
| | - Wei Wei
- Department of Neurosurgery, Zhongnan Hospital of Wuhan University, Wuhan, 430071, China.
| | - Jin-Cao Chen
- Department of Neurosurgery, Zhongnan Hospital of Wuhan University, Wuhan, 430071, China.
| |
Collapse
|