Haoxin Zheng1,2, Qi Miao3,4, Yongkai Liu1, Sohrab Afshari Mirak1, Melina Hosseiny1, Fabien Scalzo2,5, Steven S Raman1, Kyunghyun Sung1. 1. Radiological Sciences, University of California - Los Angeles, 757 Westwood Plaza, Los Angeles, CA, 90095, USA. 2. Computer Science, University of California - Los Angeles, Los Angeles, CA, 90095, USA. 3. Radiological Sciences, University of California - Los Angeles, 757 Westwood Plaza, Los Angeles, CA, 90095, USA. meganmiaoqi@126.com. 4. Department of Radiology, The First Affiliated Hospital of China Medical University, Shenyang City, 110001, Liaoning Province, China. meganmiaoqi@126.com. 5. Seaver College, Pepperdine University, Malibu, CA, 90263, USA.
Abstract
OBJECTIVE: To identify which patient with prostate cancer (PCa) could safely avoid extended pelvic lymph node dissection (ePLND) by predicting lymph node invasion (LNI), via a radiomics-based machine learning approach. METHODS: An integrative radiomics model (IRM) was proposed to predict LNI, confirmed by the histopathologic examination, integrating radiomics features, extracted from prostatic index lesion regions on MRI images, and clinical features via SVM. The study cohort comprised 244 PCa patients with MRI and followed by radical prostatectomy (RP) and ePLND within 6 months between 2010 and 2019. The proposed IRM was trained in training/validation set and evaluated in an internal independent testing set. The model's performance was measured by area under the curve (AUC), sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). AUCs were compared via Delong test with 95% confidence interval (CI), and the rest measurements were compared via chi-squared test or Fisher's exact test. RESULTS: Overall, 17 (10.6%) and 14 (16.7%) patients with LNI were included in training/validation set and testing set, respectively. Shape and first-order radiomics features showed usefulness in building the IRM. The proposed IRM achieved an AUC of 0.915 (95% CI: 0.846-0.984) in the testing set, superior to pre-existing nomograms whose AUCs were from 0.698 to 0.724 (p < 0.05). CONCLUSION: The proposed IRM could be potentially feasible to predict the risk of having LNI for patients with PCa. With the improved predictability, it could be utilized to assess which patients with PCa could safely avoid ePLND, thus reduce the number of unnecessary ePLND. KEY POINTS: • The combination of MRI-based radiomics features with clinical information improved the prediction of lymph node invasion, compared with the model using only radiomics features or clinical features. • With improved prediction performance on predicting lymph node invasion, the number of extended pelvic lymph node dissection (ePLND) could be reduced by the proposed integrative radiomics model (IRM), compared with the existing nomograms.
OBJECTIVE: To identify which patient with prostate cancer (PCa) could safely avoid extended pelvic lymph node dissection (ePLND) by predicting lymph node invasion (LNI), via a radiomics-based machine learning approach. METHODS: An integrative radiomics model (IRM) was proposed to predict LNI, confirmed by the histopathologic examination, integrating radiomics features, extracted from prostatic index lesion regions on MRI images, and clinical features via SVM. The study cohort comprised 244 PCa patients with MRI and followed by radical prostatectomy (RP) and ePLND within 6 months between 2010 and 2019. The proposed IRM was trained in training/validation set and evaluated in an internal independent testing set. The model's performance was measured by area under the curve (AUC), sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). AUCs were compared via Delong test with 95% confidence interval (CI), and the rest measurements were compared via chi-squared test or Fisher's exact test. RESULTS: Overall, 17 (10.6%) and 14 (16.7%) patients with LNI were included in training/validation set and testing set, respectively. Shape and first-order radiomics features showed usefulness in building the IRM. The proposed IRM achieved an AUC of 0.915 (95% CI: 0.846-0.984) in the testing set, superior to pre-existing nomograms whose AUCs were from 0.698 to 0.724 (p < 0.05). CONCLUSION: The proposed IRM could be potentially feasible to predict the risk of having LNI for patients with PCa. With the improved predictability, it could be utilized to assess which patients with PCa could safely avoid ePLND, thus reduce the number of unnecessary ePLND. KEY POINTS: • The combination of MRI-based radiomics features with clinical information improved the prediction of lymph node invasion, compared with the model using only radiomics features or clinical features. • With improved prediction performance on predicting lymph node invasion, the number of extended pelvic lymph node dissection (ePLND) could be reduced by the proposed integrative radiomics model (IRM), compared with the existing nomograms.
Authors: Timo F W Soeterik; Tom A Hueting; Bas Israel; Harm H E van Melick; Lea M Dijksman; Saskia Stomps; Douwe H Biesma; Hendrik Koffijberg; Michiel Sedelaar; J Alfred Witjes; Jean-Paul A van Basten Journal: BJU Int Date: 2021-05-16 Impact factor: 5.588
Authors: Nicola Fossati; Peter-Paul M Willemse; Thomas Van den Broeck; Roderick C N van den Bergh; Cathy Yuhong Yuan; Erik Briers; Joaquim Bellmunt; Michel Bolla; Philip Cornford; Maria De Santis; Ekelechi MacPepple; Ann M Henry; Malcolm D Mason; Vsevolod B Matveev; Henk G van der Poel; Theo H van der Kwast; Olivier Rouvière; Ivo G Schoots; Thomas Wiegel; Thomas B Lam; Nicolas Mottet; Steven Joniau Journal: Eur Urol Date: 2017-01-24 Impact factor: 20.096
Authors: Waldemar Wilczak; Corinna Wittmer; Till Clauditz; Sarah Minner; Stefan Steurer; Franziska Büscheck; Till Krech; Maximilian Lennartz; Luisa Harms; Diane Leleu; Marc Ahrens; Sebastian Ingwerth; Christian T Günther; Christina Koop; Ronald Simon; Frank Jacobsen; Maria C Tsourlakis; Viktoria Chirico; Doris Höflmayer; Eik Vettorazzi; Alexander Haese; Thomas Steuber; Georg Salomon; Uwe Michl; Lars Budäus; Derya Tilki; Imke Thederan; Christoph Fraune; Cosima Göbel; Marie-Christine Henrich; Manuela Juhnke; Katharina Möller; Ahmed Abdullah Bawahab; Ria Uhlig; Meike Adam; Sören Weidemann; Burkhard Beyer; Hartwig Huland; Markus Graefen; Guido Sauter; Thorsten Schlomm Journal: Eur Urol Date: 2018-06-13 Impact factor: 20.096
Authors: Constantinos Zamboglou; Montserrat Carles; Tobias Fechter; Selina Kiefer; Kathrin Reichel; Thomas F Fassbender; Peter Bronsert; Goeran Koeber; Oliver Schilling; Juri Ruf; Martin Werner; Cordula A Jilg; Dimos Baltas; Michael Mix; Anca L Grosu Journal: Theranostics Date: 2019-04-13 Impact factor: 11.556
Authors: Matthijs C F Cysouw; Bernard H E Jansen; Tim van de Brug; Daniela E Oprea-Lager; Elisabeth Pfaehler; Bart M de Vries; Reindert J A van Moorselaar; Otto S Hoekstra; André N Vis; Ronald Boellaard Journal: Eur J Nucl Med Mol Imaging Date: 2020-07-31 Impact factor: 9.236
Authors: Katharina Sprute; Vasko Kramer; Stefan A Koerber; Manuel Meneses; Rene Fernandez; Cristian Soza-Ried; Mathias Eiber; Wolfgang A Weber; Isabel Rauscher; Kambiz Rahbar; Michael Schaefers; Tadashi Watabe; Motohide Uemura; Sadahiro Naka; Norio Nonomura; Jun Hatazawa; Constantin Schwab; Viktoria Schütz; Markus Hohenfellner; Tim Holland-Letz; Juergen Debus; Clemens Kratochwil; Horacio Amaral; Pete L Choyke; Uwe Haberkorn; Camilo Sandoval; Frederik L Giesel Journal: J Nucl Med Date: 2020-08-17 Impact factor: 10.057