Kent J Peterson1, Matthew T Simpson2, Melissa K Drezdzon2, Aniko Szabo3, Robin A Ausman4, Andrew S Nencka4, Paul M Knechtges5, Carrie Y Peterson2, Kirk A Ludwig2, Timothy J Ridolfi2. 1. Division of Colon and Rectal Surgery, Department of Surgery, Medical College of Wisconsin, 8701 Watertown Plank Rd, Wauwatosa, Milwaukee, WI, 53226, USA. kjpeterson@mcw.edu. 2. Division of Colon and Rectal Surgery, Department of Surgery, Medical College of Wisconsin, 8701 Watertown Plank Rd, Wauwatosa, Milwaukee, WI, 53226, USA. 3. Division of Biostatistics, Institute for Health & Equity, Medical College of Wisconsin, Milwaukee, WI, USA. 4. Center for Imaging Research, Medical College of Wisconsin, Milwaukee, WI, USA. 5. Department of Radiology, Medical College of Wisconsin, Milwaukee, WI, USA.
Abstract
BACKGROUND: Radiomics is an approach to medical imaging that quantifies the features normally translated into visual display. While both radiomic and clinical markers have shown promise in predicting response to neoadjuvant chemoradiation therapy (nCRT) for rectal cancer, the interrelationship is not yet clear. METHODS: A retrospective, single-institution study of patients treated with nCRT for locally advanced rectal cancer was performed. Clinical and radiomic features were extracted from electronic medical record and pre-treatment magnetic resonance imaging, respectively. Machine learning models were created and assessed for complete response and positive treatment effect using the area under the receiver operating curves. RESULTS: Of 131 rectal cancer patients evaluated, 68 (51.9%) were identified to have a positive treatment effect and 35 (26.7%) had a complete response. On univariate analysis, clinical T-stage (OR 0.46, p = 0.02), lymphovascular/perineural invasion (OR 0.11, p = 0.03), and statin use (OR 2.45, p = 0.049) were associated with a complete response. Clinical T-stage (OR 0.37, p = 0.01), lymphovascular/perineural invasion (OR 0.16, p = 0.001), and abnormal carcinoembryonic antigen level (OR 0.28, p = 0.002) were significantly associated with a positive treatment effect. The clinical model was the strongest individual predictor of both positive treatment effect (AUC = 0.64) and complete response (AUC = 0.69). The predictive ability of a positive treatment effect increased by adding tumor and mesorectal radiomic features to the clinical model (AUC = 0.73). CONCLUSIONS: The use of a combined model with both clinical and radiomic features resulted in the strongest predictive capability. With the eventual goal of tailoring treatment to the individual, both clinical and radiologic markers offer insight into identifying patients likely to respond favorably to nCRT.
BACKGROUND: Radiomics is an approach to medical imaging that quantifies the features normally translated into visual display. While both radiomic and clinical markers have shown promise in predicting response to neoadjuvant chemoradiation therapy (nCRT) for rectal cancer, the interrelationship is not yet clear. METHODS: A retrospective, single-institution study of patients treated with nCRT for locally advanced rectal cancer was performed. Clinical and radiomic features were extracted from electronic medical record and pre-treatment magnetic resonance imaging, respectively. Machine learning models were created and assessed for complete response and positive treatment effect using the area under the receiver operating curves. RESULTS: Of 131 rectal cancer patients evaluated, 68 (51.9%) were identified to have a positive treatment effect and 35 (26.7%) had a complete response. On univariate analysis, clinical T-stage (OR 0.46, p = 0.02), lymphovascular/perineural invasion (OR 0.11, p = 0.03), and statin use (OR 2.45, p = 0.049) were associated with a complete response. Clinical T-stage (OR 0.37, p = 0.01), lymphovascular/perineural invasion (OR 0.16, p = 0.001), and abnormal carcinoembryonic antigen level (OR 0.28, p = 0.002) were significantly associated with a positive treatment effect. The clinical model was the strongest individual predictor of both positive treatment effect (AUC = 0.64) and complete response (AUC = 0.69). The predictive ability of a positive treatment effect increased by adding tumor and mesorectal radiomic features to the clinical model (AUC = 0.73). CONCLUSIONS: The use of a combined model with both clinical and radiomic features resulted in the strongest predictive capability. With the eventual goal of tailoring treatment to the individual, both clinical and radiologic markers offer insight into identifying patients likely to respond favorably to nCRT.
Authors: Monique Maas; Regina G H Beets-Tan; Doenja M J Lambregts; Guido Lammering; Patty J Nelemans; Sanne M E Engelen; Ronald M van Dam; Rob L H Jansen; Meindert Sosef; Jeroen W A Leijtens; Karel W E Hulsewé; Jeroen Buijsen; Geerard L Beets Journal: J Clin Oncol Date: 2011-11-07 Impact factor: 44.544
Authors: In Ja Park; Y Nancy You; Atin Agarwal; John M Skibber; Miguel A Rodriguez-Bigas; Cathy Eng; Barry W Feig; Prajnan Das; Sunil Krishnan; Christopher H Crane; Chung-Yuan Hu; George J Chang Journal: J Clin Oncol Date: 2012-04-09 Impact factor: 44.544
Authors: Andrew G Renehan; Lee Malcomson; Richard Emsley; Simon Gollins; Andrew Maw; Arthur Sun Myint; Paul S Rooney; Shabbir Susnerwala; Anthony Blower; Mark P Saunders; Malcolm S Wilson; Nigel Scott; Sarah T O'Dwyer Journal: Lancet Oncol Date: 2015-12-17 Impact factor: 41.316
Authors: Danielle S Bitterman; Lucas Resende Salgado; Harvey G Moore; Nicholas J Sanfilippo; Ping Gu; Ioannis Hatzaras; Kevin L Du Journal: Front Oncol Date: 2015-12-22 Impact factor: 6.244
Authors: Francesca De Felice; Luciano Izzo; Daniela Musio; Anna Lisa Magnante; Nadia Bulzonetti; Federico Pugliese; Paolo Izzo; Pierfrancesco Di Cello; Pietro Lucchetti; Sara Izzo; Vincenzo Tombolini Journal: Oncotarget Date: 2016-05-31