Ryo Kakino1,2,3, Mitsuhiro Nakamura1,2, Takamasa Mitsuyoshi2,4, Takashi Shintani2,5, Masaki Kokubo4, Yoshiharu Negoro6, Masato Fushiki7, Masakazu Ogura8, Satoshi Itasaka9, Chikako Yamauchi10, Shuji Otsu11, Takashi Sakamoto12, Masato Sakamoto5, Norio Araki13, Hideaki Hirashima2,3, Takanori Adachi1,2, Yukinori Matsuo2, Takashi Mizowaki2. 1. Division of Medical Physics, Department of Information Technology and Medical Engineering, Human Health Sciences, Graduate School of Medicine, Kyoto University, 53 Shogoin-Kawahara-cho, Sakyo-ku, Kyoto, 606-8507, Japan. 2. Department of Radiation Oncology and Image-applied Therapy, Kyoto University Hospital, 54 Shogoin-Kawahara-cho, Sakyo-ku, Kyoto, 606-8507, Japan. 3. Japan Society for the Promotion of Science, 5-3-1 Kojimachi, Chiyoda-ku, Tokyo, 102-0083, Japan. 4. Department of Radiation Oncology, Kobe City Medical Center General Hospital, 2-1-1, Minatojimaminamimachi, Chuo-ku, Kobe, Hyogo, 650-0047, Japan. 5. Department of Radiology, Japanese Red Cross Fukui Hospital, 2-4-1 Tsukimi, Fukui, 918-8501, Japan. 6. Department of Radiology, Tenri Hospital, 200 Mishima-cho, Tenri, Nara, 632-8552, Japan. 7. Department of Radiation Oncology, Nagahama City Hospital, 313 Oinui-cho, Nagahama, Shiga, 526-0043, Japan. 8. Department of Radiation Oncology, Kishiwada City Hospital, 1001 Gakuhara-cho, Kishiwada, Osaka, 596-8501, Japan. 9. Department of Radiation Oncology, Kurashiki Central Hospital, 1-1-1 Miwa, Kurashiki, Okayama, 710-8602, Japan. 10. Department of Radiation Oncology, Shiga General Hospital, 5-4-30 Moriyama, Shiga, 524-8524, Japan. 11. Department of Radiation Oncology, Kyoto City Hospital, 1-2 Mibuhigashitakada-cho, Nakagyo-ku, Kyoto, 604-8845, Japan. 12. Department of Radiation Oncology, Kyoto-Katsura Hospital, 17 Yamadahirao-cho, Nishikyo-ku, Kyoto, 615-8256, Japan. 13. Department of Radiation Oncology, National Hospital Organization Kyoto Medical Center, 1-1 Fukakusamukaihata-cho, Fushimi-ku, Kyoto, 612-8555, Japan.
Abstract
PURPOSE: To predict local recurrence (LR) and distant metastasis (DM) in early stage non-small cell lung cancer (NSCLC) patients after stereotactic body radiotherapy (SBRT) in multiple institutions using breath-hold computed tomography (CT)-based radiomic features with random survival forest. METHODS: A total of 573 primary early stage NSCLC patients who underwent SBRT between January 2006 and March 2016 and met the eligibility criteria were included in this study. Patients were divided into two datasets: training (464 patients in 10 institutions) and test (109 patients in one institution) datasets. A total of 944 radiomic features were extracted from manually segmented gross tumor volumes (GTVs). Feature selection was performed by analyzing inter-segmentation reproducibility, GTV correlation, and inter-feature redundancy. Nine clinical factors, including histology and GTV size, were also used. Three prognostic models (clinical, radiomic, and combined) for LR and DM were constructed using random survival forest (RSF) to deal with total death as a competing risk in the training dataset. Robust models with optimal hyper-parameters were determined using fivefold cross-validation. The patients were dichotomized into two groups based on the median value of the patient-specific risk scores (high- and low-risk score groups). Gray's test was used to evaluate the statistical significance between the two risk score groups. The prognostic power was evaluated by the concordance index with the 95% confidence intervals (CI) via bootstrapping (2000 iterations). RESULTS: The concordance indices at 3 yr of clinical, radiomic, and combined models for LR were 0.57 [CI: 0.39-0.75], 0.55 [CI: 0.38-0.73], and 0.61 [CI: 0.43-0.78], respectively, whereas those for DM were 0.59 [CI: 0.54-0.79], 0.67 [CI: 0.54-0.79], and 0.68 [CI: 0.55-0.81], respectively, in the test dataset. The combined DM model significantly discriminated its cumulative incidence between high- and low-risk score groups (P < 0.05). The variable importance of RSF in the combined model for DM indicated that two radiomic features were more important than other clinical factors. The feature maps generated on the basis of the most important radiomic feature had visual difference between high- and low-risk score groups. CONCLUSIONS: The radiomics approach with RSF for competing risks using breath-hold CT-based radiomic features might predict DM in early stage NSCLC patients who underwent SBRT although that may not have potential to predict LR.
PURPOSE: To predict local recurrence (LR) and distant metastasis (DM) in early stage non-small cell lung cancer (NSCLC) patients after stereotactic body radiotherapy (SBRT) in multiple institutions using breath-hold computed tomography (CT)-based radiomic features with random survival forest. METHODS: A total of 573 primary early stage NSCLCpatients who underwent SBRT between January 2006 and March 2016 and met the eligibility criteria were included in this study. Patients were divided into two datasets: training (464 patients in 10 institutions) and test (109 patients in one institution) datasets. A total of 944 radiomic features were extracted from manually segmented gross tumor volumes (GTVs). Feature selection was performed by analyzing inter-segmentation reproducibility, GTV correlation, and inter-feature redundancy. Nine clinical factors, including histology and GTV size, were also used. Three prognostic models (clinical, radiomic, and combined) for LR and DM were constructed using random survival forest (RSF) to deal with total death as a competing risk in the training dataset. Robust models with optimal hyper-parameters were determined using fivefold cross-validation. The patients were dichotomized into two groups based on the median value of the patient-specific risk scores (high- and low-risk score groups). Gray's test was used to evaluate the statistical significance between the two risk score groups. The prognostic power was evaluated by the concordance index with the 95% confidence intervals (CI) via bootstrapping (2000 iterations). RESULTS: The concordance indices at 3 yr of clinical, radiomic, and combined models for LR were 0.57 [CI: 0.39-0.75], 0.55 [CI: 0.38-0.73], and 0.61 [CI: 0.43-0.78], respectively, whereas those for DM were 0.59 [CI: 0.54-0.79], 0.67 [CI: 0.54-0.79], and 0.68 [CI: 0.55-0.81], respectively, in the test dataset. The combined DM model significantly discriminated its cumulative incidence between high- and low-risk score groups (P < 0.05). The variable importance of RSF in the combined model for DM indicated that two radiomic features were more important than other clinical factors. The feature maps generated on the basis of the most important radiomic feature had visual difference between high- and low-risk score groups. CONCLUSIONS: The radiomics approach with RSF for competing risks using breath-hold CT-based radiomic features might predict DM in early stage NSCLCpatients who underwent SBRT although that may not have potential to predict LR.