DOI QR코드

DOI QR Code

Development and Validation of a Model Using Radiomics Features from an Apparent Diffusion Coefficient Map to Diagnose Local Tumor Recurrence in Patients Treated for Head and Neck Squamous Cell Carcinoma

  • Minjae Kim (Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Jeong Hyun Lee (Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Leehi Joo (Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Boryeong Jeong (Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Seonok Kim (Department of Clinical Epidemiology and Biostatistics, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Sungwon Ham (Department of Convergence Medicine, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Jihye Yun (Department of Convergence Medicine, University of Ulsan College of Medicine, Asan Medical Center) ;
  • NamKug Kim (Department of Convergence Medicine, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Sae Rom Chung (Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Young Jun Choi (Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Jung Hwan Baek (Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center) ;
  • Ji Ye Lee (Department of Radiology, Seoul National University Hospital) ;
  • Ji-hoon Kim (Department of Radiology, Seoul National University Hospital)
  • Received : 2022.05.10
  • Accepted : 2022.08.17
  • Published : 2022.11.01

Abstract

Objective: To develop and validate a model using radiomics features from apparent diffusion coefficient (ADC) map to diagnose local tumor recurrence in head and neck squamous cell carcinoma (HNSCC). Materials and Methods: This retrospective study included 285 patients (mean age ± standard deviation, 62 ± 12 years; 220 male, 77.2%), including 215 for training (n = 161) and internal validation (n = 54) and 70 others for external validation, with newly developed contrast-enhancing lesions at the primary cancer site on the surveillance MRI following definitive treatment of HNSCC between January 2014 and October 2019. Of the 215 and 70 patients, 127 and 34, respectively, had local tumor recurrence. Radiomics models using radiomics scores were created separately for T2-weighted imaging (T2WI), contrast-enhanced T1-weighted imaging (CE-T1WI), and ADC maps using non-zero coefficients from the least absolute shrinkage and selection operator in the training set. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance of each radiomics score and known clinical parameter (age, sex, and clinical stage) in the internal and external validation sets. Results: Five radiomics features from T2WI, six from CE-T1WI, and nine from ADC maps were selected and used to develop the respective radiomics models. The area under ROC curve (AUROC) of ADC radiomics score was 0.76 (95% confidence interval [CI], 0.62-0.89) and 0.77 (95% CI, 0.65-0.88) in the internal and external validation sets, respectively. These were significantly higher than the AUROC values of T2WI (0.53 [95% CI, 0.40-0.67], p = 0.006), CE-T1WI (0.53 [95% CI, 0.40-0.67], p = 0.012), and clinical parameters (0.53 [95% CI, 0.39-0.67], p = 0.021) in the external validation set. Conclusion: The radiomics model using ADC maps exhibited higher diagnostic performance than those of the radiomics models using T2WI or CE-T1WI and clinical parameters in the diagnosis of local tumor recurrence in HNSCC following definitive treatment.

Keywords

References

  1. Agra IM, Carvalho AL, Ulbrich FS, de Campos OD, Martins EP, Magrin J, et al. Prognostic factors in salvage surgery for recurrent oral and oropharyngeal cancer. Head Neck 2006;28:107-113  https://doi.org/10.1002/hed.20309
  2. Goodwin WJ Jr. Salvage surgery for patients with recurrent squamous cell carcinoma of the upper aerodigestive tract: when do the ends justify the means? Laryngoscope 2000;110(3 Pt 2 Suppl 93):1-18  https://doi.org/10.1097/00005537-200003001-00001
  3. Kowalski LP, Bagietto R, Lara JR, Santos RL, Silva JF Jr, Magrin J. Prognostic significance of the distribution of neck node metastasis from oral carcinoma. Head Neck 2000;22:207-214  https://doi.org/10.1002/(SICI)1097-0347(200005)22:3<207::AID-HED1>3.0.CO;2-9
  4. Carvalho AL, Magrin J, Kowalski LP. Sites of recurrence in oral and oropharyngeal cancers according to the treatment approach. Oral Dis 2003;9:112-118  https://doi.org/10.1034/j.1601-0825.2003.01750.x
  5. Bahadur S, Amatya RC, Kacker SK. The enigma of postradiation oedema and residual or recurrent carcinoma of the larynx and pyriform fossa. J Laryngol Otol 1985;99:763-765  https://doi.org/10.1017/S0022215100097620
  6. Aerts HJ, Velazquez ER, Leijenaar RT, Parmar C, Grossmann P, Carvalho S, et al. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun 2014;5:4006 
  7. Kumar V, Gu Y, Basu S, Berglund A, Eschrich SA, Schabath MB, et al. Radiomics: the process and the challenges. Magn Reson Imaging 2012;30:1234-1248  https://doi.org/10.1016/j.mri.2012.06.010
  8. Wong AJ, Kanwar A, Mohamed AS, Fuller CD. Radiomics in head and neck cancer: from exploration to application. Transl Cancer Res 2016;5:371-382  https://doi.org/10.21037/tcr.2016.07.18
  9. Giraud P, Giraud P, Gasnier A, El Ayachy R, Kreps S, Foy JP, et al. Radiomics and machine learning for radiotherapy in head and neck cancers. Front Oncol 2019;9:174 
  10. Jajodia A, Aggarwal D, Chaturvedi AK, Rao A, Mahawar V, Gairola M, et al. Value of diffusion MR imaging in differentiation of recurrent head and neck malignancies from post treatment changes. Oral Oncol 2019;96:89-96  https://doi.org/10.1016/j.oraloncology.2019.06.037
  11. Driessen JP, Caldas-Magalhaes J, Janssen LM, Pameijer FA, Kooij N, Terhaard CH, et al. Diffusion-weighted MR imaging in laryngeal and hypopharyngeal carcinoma: association between apparent diffusion coefficient and histologic findings. Radiology 2014;272:456-463  https://doi.org/10.1148/radiol.14131173
  12. Vaid S, Chandorkar A, Atre A, Shah D, Vaid N. Differentiating recurrent tumours from post-treatment changes in head and neck cancers: does diffusion-weighted MRI solve the eternal dilemma? Clin Radiol 2017;72:74-83  https://doi.org/10.1016/j.crad.2016.09.019
  13. Desouky S, AboSeif S, Shama S, Gaafar A, Gamaleldin O. Role of dynamic contrast enhanced and diffusion weighted MRI in the differentiation between post treatment changes and recurrent laryngeal cancers. Egypt J Radiol Nucl Med 2015;46:379-389  https://doi.org/10.1016/j.ejrnm.2015.01.012
  14. Vandecaveye V, De Keyzer F, Nuyts S, Deraedt K, Dirix P, Hamaekers P, et al. Detection of head and neck squamous cell carcinoma with diffusion weighted MRI after (chemo) radiotherapy: correlation between radiologic and histopathologic findings. Int J Radiat Oncol Biol Phys 2007;67:960-971  https://doi.org/10.1016/j.ijrobp.2006.09.020
  15. Park JE, Park SY, Kim HJ, Kim HS. Reproducibility and generalizability in radiomics modeling: possible strategies in radiologic and statistical perspectives. Korean J Radiol 2019;20:1124-1137  https://doi.org/10.3348/kjr.2018.0070
  16. Kang D, Park JE, Kim YH, Kim JH, Oh JY, Kim J, et al. Diffusion radiomics as a diagnostic model for atypical manifestation of primary central nervous system lymphoma: development and multicenter external validation. Neuro Oncol 2018;20:1251-1261  https://doi.org/10.1093/neuonc/noy021
  17. Kim JY, Park JE, Jo Y, Shim WH, Nam SJ, Kim JH, et al. Incorporating diffusion- and perfusion-weighted MRI into a radiomics model improves diagnostic performance for pseudoprogression in glioblastoma patients. Neuro Oncol 2019;21:404-414  https://doi.org/10.1093/neuonc/noy133
  18. Amin MB, Greene FL, Edge SB, Compton CC, Gershenwald JE, Brookland RK, et al. The eighth edition AJCC cancer staging manual: continuing to build a bridge from a population-based to a more "personalized" approach to cancer staging. CA Cancer J Clin 2017;67:93-99  https://doi.org/10.3322/caac.21388
  19. Nolden M, Zelzer S, Seitel A, Wald D, Muller M, Franz AM, et al. The medical imaging interaction toolkit: challenges and advances. Int J Comput Assist Radiol Surg 2013;8:607-620  https://doi.org/10.1007/s11548-013-0840-8
  20. Maes F, Collignon A, Vandermeulen D, Marchal G, Suetens P. Multimodality image registration by maximization of mutual information. IEEE Trans Med Imaging 1997;16:187-198  https://doi.org/10.1109/42.563664
  21. Avants BB, Tustison NJ, Song G, Cook PA, Klein A, Gee JC. A reproducible evaluation of ANTs similarity metric performance in brain image registration. Neuroimage 2011;54:2033-2044  https://doi.org/10.1016/j.neuroimage.2010.09.025
  22. Shinohara RT, Sweeney EM, Goldsmith J, Shiee N, Mateen FJ, Calabresi PA, et al. Statistical normalization techniques for magnetic resonance imaging. Neuroimage Clin 2014;6:9-19  https://doi.org/10.1016/j.nicl.2014.08.008
  23. Zwanenburg A, Vallieres M, Abdalah MA, Aerts HJWL, Andrearczyk V, Apte A, et al. The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology 2020;295:328-338  https://doi.org/10.1148/radiol.2020191145
  24. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989;45:255-268  https://doi.org/10.2307/2532051
  25. Hepp T, Schmid M, Gefeller O, Waldmann E, Mayr A. Approaches to regularized regression-a comparison between gradient boosting and the lasso. Methods Inf Med 2016;55:422-430  https://doi.org/10.3414/ME16-01-0033
  26. Gui J, Li H. Penalized cox regression analysis in the high-dimensional and low-sample size settings, with applications to microarray gene expression data. Bioinformatics 2005;21:3001-3008  https://doi.org/10.1093/bioinformatics/bti422
  27. Wu TT, Chen YF, Hastie T, Sobel E, Lange K. Genome-wide association analysis by lasso penalized logistic regression. Bioinformatics 2009;25:714-721  https://doi.org/10.1093/bioinformatics/btp041
  28. Tibshirani R. Regression shrinkage and selection via the lasso. J R Stat Soc Ser B Methodol 1996;58:267-288  https://doi.org/10.1111/j.2517-6161.1996.tb02080.x
  29. Youden WJ. Index for rating diagnostic tests. Cancer 1950;3:32-35  https://doi.org/10.1002/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3
  30. Surov A, Stumpp P, Meyer HJ, Gawlitza M, Hohn AK, Boehm A, et al. Simultaneous (18)F-FDG-PET/MRI: associations between diffusion, glucose metabolism and histopathological parameters in patients with head and neck squamous cell carcinoma. Oral Oncol 2016;58:14-20  https://doi.org/10.1016/j.oraloncology.2016.04.009
  31. Swartz JE, Driessen JP, van Kempen PMW, de Bree R, Janssen LM, Pameijer FA, et al. Influence of tumor and microenvironment characteristics on diffusion-weighted imaging in oropharyngeal carcinoma: a pilot study. Oral Oncol 2018;77:9-15  https://doi.org/10.1016/j.oraloncology.2017.12.001
  32. Ren J, Tian J, Yuan Y, Dong D, Li X, Shi Y, et al. Magnetic resonance imaging based radiomics signature for the preoperative discrimination of stage I-II and III-IV head and neck squamous cell carcinoma. Eur J Radiol 2018;106:1-6  https://doi.org/10.1016/j.ejrad.2018.07.002
  33. Suh CH, Lee KH, Choi YJ, Chung SR, Baek JH, Lee JH, et al. Oropharyngeal squamous cell carcinoma: radiomic machine-learning classifiers from multiparametric MR images for determination of HPV infection status. Sci Rep 2020;10:17525  https://doi.org/10.1038/s41598-020-74479-x
  34. Fujima N, Shimizu Y, Yoshida D, Kano S, Mizumachi T, Homma A, et al. Machine-learning-based prediction of treatment outcomes using MR imaging-derived quantitative tumor information in patients with sinonasal squamous cell carcinomas: a preliminary study. Cancers (Basel) 2019;11:800 
  35. Zhang L, Zhou H, Gu D, Tian J, Zhang B, Dong D, et al. Radiomic nomogram: pretreatment evaluation of local recurrence in nasopharyngeal carcinoma based on MR imaging. J Cancer 2019;10:4217-4225  https://doi.org/10.7150/jca.33345
  36. Haralick RM, Shanmugam K, Dinstein IH. Textural features for image classification. IEEE Trans Syst Man Cybern 1973;SMC-3:610-621  https://doi.org/10.1109/TSMC.1973.4309314