1
|
Zhang AD, Shi QL, Zhang HT, Duan WH, Li Y, Ruan L, Han YF, Liu ZK, Li HF, Xiao JS, Shi GF, Wan X, Wang RZ. Pairwise machine learning-based automatic diagnostic platform utilizing CT images and clinical information for predicting radiotherapy locoregional recurrence in elderly esophageal cancer patients. Abdom Radiol (NY) 2024:10.1007/s00261-024-04377-7. [PMID: 38831075 DOI: 10.1007/s00261-024-04377-7] [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: 03/17/2024] [Revised: 05/04/2024] [Accepted: 05/06/2024] [Indexed: 06/05/2024]
Abstract
OBJECTIVE To investigate the feasibility and accuracy of predicting locoregional recurrence (LR) in elderly patients with esophageal squamous cell cancer (ESCC) who underwent radical radiotherapy using a pairwise machine learning algorithm. METHODS The 130 datasets enrolled were randomly divided into a training set and a testing set in a 7:3 ratio. Clinical factors were included and radiomics features were extracted from pretreatment CT scans using pyradiomics-based software, and a pairwise naive Bayes (NB) model was developed. The performance of the model was evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA). To facilitate practical application, we attempted to construct an automated esophageal cancer diagnosis system based on trained models. RESULTS To the follow-up date, 64 patients (49.23%) had experienced LR. Ten radiomics features and two clinical factors were selected for modeling. The model demonstrated good prediction performance, with area under the ROC curve of 0.903 (0.829-0.958) for the training cohort and 0.944 (0.849-1.000) for the testing cohort. The corresponding accuracies were 0.852 and 0.914, respectively. Calibration curves showed good agreement, and DCA curve confirmed the clinical validity of the model. The model accurately predicted LR in elderly patients, with a positive predictive value of 85.71% for the testing cohort. CONCLUSIONS The pairwise NB model, based on pre-treatment enhanced chest CT-based radiomics and clinical factors, can accurately predict LR in elderly patients with ESCC. The esophageal cancer automated diagnostic system embedded with the pairwise NB model holds significant potential for application in clinical practice.
Collapse
Affiliation(s)
- An-du Zhang
- Department of Radiotherapy, Hebei Medical University Fourth Affiliated Hospital and Hebei Provincial Tumor Hospital, 12 Jiankang Road, Shijiazhuang, Hebei, 050011, People's Republic of China
| | - Qing-Lei Shi
- School of Medicine, Chinese University of Hong Kong (Shenzhen), No. 2001, Longxiang Avenue, Longgang District, Shenzhen, 518172, People's Republic of China
- Medical Big Data Laboratory, Shenzhen Research Institute of Big Data, Daoyuan Building, No. 2001, Longxiang Avenue, Longgang District, Shenzhen, 518172, People's Republic of China
| | - Hong-Tao Zhang
- Department of Oncology, Hebei General Hospital, NO. 348 Heping West Road, Xinhua District, Shijiazhuang, Hebei, 050051, People's Republic of China
| | - Wen-Han Duan
- School of Computer Science and Engineering, Beihang University, No. 37 Xueyuan Road, Haidian District, Beijing, 100191, People's Republic of China
| | - Yang Li
- Department of Radiotherapy, Hebei Medical University Fourth Affiliated Hospital and Hebei Provincial Tumor Hospital, 12 Jiankang Road, Shijiazhuang, Hebei, 050011, People's Republic of China
| | - Li Ruan
- School of Computer Science and Engineering, Beihang University, No. 37 Xueyuan Road, Haidian District, Beijing, 100191, People's Republic of China
| | - Yi-Fan Han
- School of Computer Science and Engineering, Beihang University, No. 37 Xueyuan Road, Haidian District, Beijing, 100191, People's Republic of China
| | - Zhi-Kun Liu
- Department of Radiotherapy, Hebei Medical University Fourth Affiliated Hospital and Hebei Provincial Tumor Hospital, 12 Jiankang Road, Shijiazhuang, Hebei, 050011, People's Republic of China
| | - Hao-Feng Li
- Medical Big Data Laboratory, Shenzhen Research Institute of Big Data, Daoyuan Building, No. 2001, Longxiang Avenue, Longgang District, Shenzhen, 518172, People's Republic of China
| | - Jia-Shun Xiao
- Medical Big Data Laboratory, Shenzhen Research Institute of Big Data, Daoyuan Building, No. 2001, Longxiang Avenue, Longgang District, Shenzhen, 518172, People's Republic of China
| | - Gao-Feng Shi
- Department of Radiotherapy, Hebei Medical University Fourth Affiliated Hospital and Hebei Provincial Tumor Hospital, 12 Jiankang Road, Shijiazhuang, Hebei, 050011, People's Republic of China.
| | - Xiang Wan
- Medical Big Data Laboratory, Shenzhen Research Institute of Big Data, Daoyuan Building, No. 2001, Longxiang Avenue, Longgang District, Shenzhen, 518172, People's Republic of China.
| | - Ren-Zhi Wang
- School of Medicine, Chinese University of Hong Kong (Shenzhen), No. 2001, Longxiang Avenue, Longgang District, Shenzhen, 518172, People's Republic of China.
| |
Collapse
|
2
|
Yan S, Shi YJ, Liu C, Li XT, Zhao B, Wei YY, Shen L, Lu ZH, Sun YS. Quantitative CT evaluation after two cycles of induction chemotherapy to predict prognosis of patients with locally advanced oesophageal squamous cell carcinoma before undergoing definitive chemoradiotherapy/radiotherapy. Eur Radiol 2023; 33:380-390. [PMID: 35927466 PMCID: PMC9755097 DOI: 10.1007/s00330-022-08994-y] [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: 01/16/2022] [Revised: 06/24/2022] [Accepted: 06/30/2022] [Indexed: 12/23/2022]
Abstract
OBJECTIVE To investigate the performance of quantitative CT analysis in predicting the prognosis of patients with locally advanced oesophageal squamous cell carcinoma (ESCC) after two cycles of induction chemotherapy before definitive chemoradiotherapy/radiotherapy. METHODS A total of 110 patients with locally advanced ESCC were retrospectively analysed. Baseline chest CT and CT after two cycles of induction chemotherapy were analysed. A multivariate Cox proportional-hazard regression model was used to identify independent prognostic markers for survival analysis. Then, a CT scoring system was established. Time-dependent receiver operating characteristic (ROC) curve analysis and the Kaplan-Meier method were employed for analysing the prognostic value of the CT scoring system. RESULTS Body mass index, treatment strategy, change ratios of thickness (ΔTHmax), CT value of the primary tumour (ΔCTVaxial) and the short diameter (ΔSD-LN), and the presence of an enlarged small lymph node (ESLN) after two cycles of chemotherapy were noted as independent factors for predicting overall survival (OS). The specificity of the presence of ESLN for death after 12 months was up to 100%. Areas under the curve value of the CT scoring system for predicting OS and progression-free survival (PFS) were higher than that of the RECIST (p < 0.05). Responders had significantly longer OS and PFS than non-responders. CONCLUSION Quantitative CT analysis after two cycles of induction chemotherapy could predict the outcome of locally advanced ESCC patients treated with definitive chemoradiotherapy/radiotherapy. The CT scoring system could contribute to the development of an appropriate strategy for patients with locally advanced ESCC. KEY POINTS • Quantitative CT evaluation after two cycles of induction chemotherapy can predict the long-term outcome of locally advanced oesophageal cancer treated with definitive chemoradiotherapy/radiotherapy. • A CT scoring system provides valuable imaging support for indicating the prognosis at the early stage of therapy. • Quantitative CT evaluation can assist clinicians in personalising treatment plans.
Collapse
Affiliation(s)
- Shuo Yan
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Department of Radiology, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Hai Dian District, Beijing, 100142 China
| | - Yan-Jie Shi
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Department of Radiology, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Hai Dian District, Beijing, 100142 China
| | - Chang Liu
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Early Drug Development Center, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Haidian District, Beijing, 100142 China
| | - Xiao-Ting Li
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Department of Radiology, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Hai Dian District, Beijing, 100142 China
| | - Bo Zhao
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Department of Radiology, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Hai Dian District, Beijing, 100142 China
| | - Yi-Yuan Wei
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Department of Radiology, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Hai Dian District, Beijing, 100142 China
| | - Lin Shen
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Early Drug Development Center, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Haidian District, Beijing, 100142 China ,Department of Gastrointestinal Oncology, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Peking University Cancer Hospital & Institute, No. 52 Fu-Cheng Road, Hai-Dian District, Beijing, 100142 China
| | - Zhi-Hao Lu
- Department of Gastrointestinal Oncology, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Peking University Cancer Hospital & Institute, No. 52 Fu-Cheng Road, Hai-Dian District, Beijing, 100142 China
| | - Ying-Shi Sun
- Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/ Beijing), Department of Radiology, Peking University Cancer Hospital & Institute, No. 52, Fucheng Road, Hai Dian District, Beijing, 100142 China
| |
Collapse
|
3
|
Change in Density Not Size of Esophageal Adenocarcinoma During Neoadjuvant Chemotherapy Is Associated with Improved Survival Outcomes. J Gastrointest Surg 2022; 26:2417-2425. [PMID: 36214951 DOI: 10.1007/s11605-022-05422-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/28/2022] [Accepted: 07/16/2022] [Indexed: 01/31/2023]
Abstract
BACKGROUND Changes in the size and density of esophageal malignancy during neoadjuvant chemotherapy (NCT) may be useful in predicting overall survival (OS). The aim of this study was to explore this relationship in patients with adenocarcinoma. METHODS A retrospective single-centre cohort study was performed. Consecutive patients with esophageal adenocarcinoma who received NCT followed by en bloc resection with curative intent were identified. Pre- and post-NCT computed tomography scans were reviewed. The percentage difference between the greatest tumor diameter, esophageal wall thickness and tumor density was calculated. Multivariate Cox regression analysis identified variables independently associated with OS. A ROC analysis was performed on radiological markers to identify optimal cut-off points with Kaplan-Meier plots subsequently created. RESULTS Of the 167 identified, 88 (51.5%) had disease of the gastro-esophageal junction and 149 (89.2%) were clinical T3. In total, 122 (73.1%) had node-positive disease. Increased tumor density (HR 1.01 per % change, 95% CI 1.00-1.02, p = 0.007), lymphovascular invasion (HR 3.23, 95% CI 1.34-7.52, p = 0.006) and perineural invasion (HR 2.51, 95% CI 1.03-6.08, p = 0.048) were independently associated with a decrease in OS. Patients who had a decrease in their tumor density during the time they received NCT of ≥ 20% in Hounsfield units had significantly longer OS than those who did not (75.5 months versus 34.4 months, 95% CI 38.83-105.13/18.63-35.07, p = 0.025). CONCLUSIONS Interval changes in the density, not size, of esophageal adenocarcinoma during the time that NCT are independently associated with OS.
Collapse
|
4
|
Xue L, E L, Wu Z, Guo D. Application Value of Gastroenterography Combined With CT in the Evaluation of Short-Term Efficacy and Prognosis in Patients With Esophageal Cancer Radiotherapy. Front Surg 2022; 9:898965. [PMID: 35756472 PMCID: PMC9218177 DOI: 10.3389/fsurg.2022.898965] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/18/2022] [Accepted: 04/04/2022] [Indexed: 11/16/2022] Open
Abstract
Purpose To observe the application value of gastroenterography combined with CT in the evaluation of short-term efficacy and prognosis in patients with esophageal cancer radiotherapy. Methods From January 2013 to December 2020, the clinical data of 207 patients with esophageal cancer treated by radiotherapy in our hospital were collected retrospectively. All patients received gastroenterography and CT examination before and after radiotherapy, and the patients were followed-up for 1 year, and the evaluation value of their short-term efficacy and prognosis was observed. Results After radiotherapy, the length diameter, short diameter, and volume of the lymph node were lower than those before radiotherapy (p < 0.05), but the maximum tube wall thickness had no significant difference (p > 0.05). The length diameter, short diameter, and volume of the lymph node, and the maximum tube wall thickness in the good efficacy group and the good prognosis group were lower, and the objective response rate in the good prognosis group was higher (p < 0.05). The area under the curve (AUC) of the length diameter, short diameter, and volume of the lymph node to evaluate the short-term efficacy of patients with esophageal cancer was 0.738, 0.705, and 0.748, respectively, and the AUC to evaluate the prognosis of patients with esophageal cancer was 0.751, 0.776, and 0.791, respectively. Conclusion Gastroenterography combined with CT has a good application value in the evaluation of short-term efficacy and prognosis in patients with esophageal cancer radiotherapy.
Collapse
|
5
|
Li Y, Su H, Yang L, Yue M, Wang M, Gu X, Dai L, Wang X, Su X, Zhang A, Ren J, Shi G. Can lymphovascular invasion be predicted by contrast-enhanced CT imaging features in patients with esophageal squamous cell carcinoma? A preliminary retrospective study. BMC Med Imaging 2022; 22:93. [PMID: 35581563 PMCID: PMC9116049 DOI: 10.1186/s12880-022-00804-7] [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: 12/07/2021] [Accepted: 04/20/2022] [Indexed: 11/10/2022] Open
Abstract
Background To investigate the value of contrast-enhanced CT (CECT)-derived imaging features in predicting lymphovascular invasion (LVI) status in esophageal squamous cell carcinoma (ESCC) patients. Methods One hundred and ninety-seven patients with postoperative pathologically confirmed esophageal squamous cell carcinoma treated in our hospital between January 2017 and January 2019 were enrolled in our study, including fifty-nine patients with LVI and one hundred and thirty-eight patients without LVI. The CECT-derived imaging features of all patients were analyzed. The CECT-derived imaging features were divided into quantitative features and qualitative features. The quantitative features consisted of the CT attenuation value of the tumor (CTVTumor), the CT attenuation value of the normal esophageal wall (CTVNormal), the CT attenuation value ratio of the tumor-to-normal esophageal wall (TNR), the CT attenuation value difference between the tumor and normal esophageal wall (ΔTN), the maximum thickness of the tumor measured by CECT (Thickness), the maximum length of the tumor measured by CECT (Length), and the gross tumor volume measured by CECT (GTV). The qualitative features consisted of an enhancement pattern, tumor margin, enlarged blood supply or drainage vessels to the tumor (EVFDT), and tumor necrosis. For the clinicopathological characteristics and CECT-derived imaging feature analysis, the chi-squared test was used for categorical variables, the Mann–Whitney U test was used for continuous variables with a nonnormal distribution, and the independent sample t-test was used for the continuous variables with a normal distribution. The trend test was used for ordinal variables. The association between LVI status and CECT-derived imaging features was analyzed by univariable logistic analysis, followed by multivariable logistic regression and receiver operating characteristic (ROC) curve analysis. Results The CTVTumor, TNR, ΔTN, Thickness, Length, and GTV in the group with LVI were higher than those in the group without LVI (P < 0.05). A higher proportion of patients with heterogeneous enhancement pattern, irregular tumor margin, EVFDT, and tumor necrosis were present in the group with LVI (P < 0.05). As revealed by the univariable logistic analysis, the CECT-derived imaging features, including CTVTumor, TNR, ΔTN and enhancement pattern, Thickness, Length, GTV, tumor margin, EVFDT, and tumor necrosis were associated with LVI status (P < 0.05). Only the TNR (OR 8.655; 95% CI 2.125–37.776), Thickness (OR 6.531; 95% CI 2.410–20.608), and tumor margin (OR 4.384; 95% CI 2.004–9.717) were independent risk factors for LVI in the multivariable logistic regression analysis. The ROC curve analysis incorporating the above three CECT-derived imaging features showed that the area under the curve obtained by the multivariable logistic regression model was 0.820 (95% CI 0.754–0.885). Conclusion The CECT-derived imaging features, including TNR, Thickness, tumor margin, and their combination, can be used as predictors of LVI status for patients with ESCC. Supplementary Information The online version contains supplementary material available at 10.1186/s12880-022-00804-7.
Collapse
Affiliation(s)
- Yang Li
- Department of Computed Tomography and Magnetic Resonance, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Haiyan Su
- Department of Computed Tomography and Magnetic Resonance, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Li Yang
- Department of Computed Tomography and Magnetic Resonance, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Meng Yue
- Department of Pathology, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Mingbo Wang
- Department of Thoracic Surgery, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Xiaolong Gu
- Department of Computed Tomography and Magnetic Resonance, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Lijuan Dai
- Department of Computed Tomography and Magnetic Resonance, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Xiangming Wang
- Department of Computed Tomography and Magnetic Resonance, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | - Xiaohua Su
- Department of Oncology, Hebei General Hospital, Shijiazhuang, 050051, China
| | - Andu Zhang
- Department of Radiotherapy, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China
| | | | - Gaofeng Shi
- Department of Computed Tomography and Magnetic Resonance, Fourth Hospital of Hebei Medical University, Shijiazhuang, 050011, China.
| |
Collapse
|
6
|
Wu Y, Li J. Change in Maximal Esophageal Wall Thickness Provides Prediction of Survival and Recurrence in Patients with Esophageal Squamous Cell Carcinoma After Neoadjuvant Chemoradiotherapy and Surgery. Cancer Manag Res 2021; 13:2433-2445. [PMID: 33758542 PMCID: PMC7979351 DOI: 10.2147/cmar.s295646] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/04/2020] [Accepted: 02/16/2021] [Indexed: 01/03/2023] Open
Abstract
Purpose This study aimed to evaluate the relationship of the percentage decrease of maximal esophageal wall thickness with pathological complete response (pCR) and recurrence in esophageal squamous cell carcinoma (ESCC). Patients and Methods A total of 146 ESCC patients treated with neoadjuvant chemoradiotherapy (NCRT) and surgery were included. The prognostic factors for overall survival (OS) and disease-free survival (DFS) were analyzed. The recurrence site, time, and frequency were included in the analysis. The percentage decrease of maximal esophageal wall thickness after NCRT was determined with the formula: [(pre-post)/pre] × 100. Results Overall, only 42 patients achieved pCR. Multivariable logistic analyses showed that the percentage decrease of maximal esophageal wall thickness (HR: 2.504; 95% CI: 1.112–5.638, P=0.027) was independently correlated with pCR. In multivariable Cox analyses, a ≤40% percentage decrease of maximal esophageal wall thickness was an independent adverse factor for both OS (HR: 1.907, 95% CI: 1.149–3.165; P=0.012) and DFS (HR: 2.054, 95% CI: 1.288–3.277; P=0.003). Compared with patients with a ≤40% percentage decrease, those with a >40% percentage decrease had better 5-year OS (29.0% vs 60.1%, P<0.05) and DFS (27.8% vs 54.4%, P<0.05). Perineural invasion (PNI) was also an unfavorable factor for OS (HR: 2.138, 95% CI: 0.094–4.178; P=0.026). Lymph vessel invasion (HR: 2.874, 95% CI: 1.574–5.248; P=0.001) and PNI (HR: 2.050; 95% CI: 1.044–4.023; P=0.037) were independent prognosticators for DFS. The rates of local and distant recurrence were also significantly difference between those with a percentage decrease of ≤40% and of >40% (P<0.05). Conclusion The percentage decrease of maximal esophageal wall thickness is associated with pCR and recurrence in ESCC patients who undergo NCRT and surgery and can thus be used to independently predict prognosis.
Collapse
Affiliation(s)
- Yahua Wu
- Department of Radiation Oncology, Fujian Medical University Cancer Hospital, Fujian Cancer Hospital, Fuzhou, Fujian, 350014, People's Republic of China
| | - Jiancheng Li
- Department of Radiation Oncology, Fujian Medical University Cancer Hospital, Fujian Cancer Hospital, Fuzhou, Fujian, 350014, People's Republic of China
| |
Collapse
|