RESUMEN
BACKGROUND AND OBJECTIVES: Deep learning models (DLMs) are applied across domains of health sciences to generate meaningful predictions. DLMs make use of neural networks to generate predictions from discrete data inputs. This study employs DLM on prechemotherapy cross-sectional imaging to predict patients' response to neoadjuvant chemotherapy. METHODS: Adult patients with colorectal liver metastasis who underwent surgery after neoadjuvant chemotherapy were included. A DLM was trained on computed tomography images using attention-based multiple-instance learning. A logistic regression model incorporating clinical parameters of the Fong clinical risk score was used for comparison. Both model performances were benchmarked against the Response Evaluation Criteria in Solid Tumors criteria. A receiver operating curve was created and resulting area under the curve (AUC) was determined. RESULTS: Ninety-five patients were included, with 33,619 images available for study inclusion. Ninety-five percent of patients underwent 5-fluorouracil-based chemotherapy with oxaliplatin and/or irinotecan. Sixty percent of the patients were categorized as chemotherapy responders (30% reduction in tumor diameter). The DLM had an AUC of 0.77. The AUC for the clinical model was 0.41. CONCLUSIONS: Image-based DLM for prediction of response to neoadjuvant chemotherapy in patients with colorectal cancer liver metastases was superior to a clinical-based model. These results demonstrate potential to identify nonresponders to chemotherapy and guide select patients toward earlier curative resection.
RESUMEN
INTRODUCTION: Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality and often arises in the setting of cirrhosis. The present series reviews outcomes following 791 operations. METHODS: Retrospective review surgical MWA for HCC from March 2007 through December 2022 at a high-volume institution was performed using a prospective database. Primary outcome was overall survival. RESULTS: A total of 791 operations in 623 patients and 1156 HCC tumors were treated with surgical MWA. Median tumor size was 2 cm (range 0.25-10 cm) with an average of 1 tumor ablated per operation (range 1-7 tumors). Nearly 90 % of patients had cirrhosis with a median MELD score of 8 (IQR = 6-11). Mortality within 30 days occurred in 13 patients (1.6 %). Per tumor, the rate of incomplete ablation was 2.25 % and local recurrence was 2.95 %. Previous ablation and tumor size were risk factors for recurrence. One-year overall survival was 82.0 % with a median overall survival of 36.5 months (95 % CI 15.7-93.7) and median disease-free survival of 15.9 months (range 5.7-37.3 months). CONCLUSION: Surgical MWA offers a low-morbidity approach for treatment of HCC, affording low rates of incomplete ablation and local recurrence.