RESUMEN
Background: With increasingly prevalent coexistence of chronic hepatitis B (CHB) and hepatic steatosis (HS), simple, non-invasive diagnostic methods to accurately assess the severity of hepatic inflammation are needed. We aimed to build a machine learning (ML) based model to detect hepatic inflammation in patients with CHB and concurrent HS. Methods: We conducted a multicenter, retrospective cohort study in China. Treatment-naive CHB patients with biopsy-proven HS between April 2004 and September 2022 were included. The optimal features for model development were selected by SHapley Additive explanations, and an ML algorithm with the best accuracy to diagnose moderate to severe hepatic inflammation (Scheuer's system ≥ G3) was determined and assessed by decision curve analysis (DCA) and calibration curve. This study is registered with ClinicalTrials.gov (NCT05766449). Findings: From a pool of 1,787 treatment-naive patients with CHB and HS across eleven hospitals, 689 patients from nine of these hospitals were chosen for the development of the diagnostic model. The remaining two hospitals contributed to two independent external validation cohorts, comprising 509 patients in validation cohort 1 and 589 in validation cohort 2. Eleven features regarding inflammation, hepatic and metabolic functions were identified. The gradient boosting classifier (GBC) model showed the best performance in predicting moderate to severe hepatic inflammation, with an area under the receiver operating characteristic curve (AUROC) of 0.86 (95% CI 0.83-0.88) in the training cohort, and 0.89 (95% CI 0.86-0.92), 0.76 (95% CI 0.73-0.80) in the first and second external validation cohorts, respectively. A publicly accessible web tool was generated for the model. Interpretation: Using simple parameters, the GBC model predicted hepatic inflammation in CHB patients with concurrent HS. It holds promise for guiding clinical management and improving patient outcomes. Funding: This research was supported by the National Natural Science Foundation of China (No. 82170609, 81970545), Natural Science Foundation of Shandong Province (Major Project) (No. ZR2020KH006), Natural Science Foundation of Jiangsu Province (No.BK20231118), Tianjin Key Medical Discipline (Specialty), Construction Project, TJYXZDXK-059B, Tianjin Health Science and Technology Project key discipline special, TJWJ2022XK034, and Research project of Chinese traditional medicine and Chinese traditional medicine combined with Western medicine of Tianjin municipal health and Family Planning Commission (2021022).
RESUMEN
BACKGROUND & AIMS: Hepatocellular carcinoma (HCC) arises in a cirrhotic, pro-angiogenic microenvironment. Inhibiting angiogenesis is a key mode of action of multikinase inhibitors and current non-cirrhotic models are unable to predict treatment response. We present a novel mouse cirrhotic model of xenotransplant that predicts the natural biology of HCC and allows personalized therapy. METHODS: Cirrhosis was induced in NOD Scid gamma mice with 4 months of thioacetamide administration. Patient derived xenografts (PDXs) were created by transplant of human HCC subcutaneously into non-cirrhotic mice and intra-hepatically into both cirrhotic and non-cirrhotic mice. The applicability of cirrhotic PDXs for drug testing was tested with 16 days of either sorafenib or lenvatinib. Treatment response was evaluated by MRI. RESULTS: 8 out of 19 (42%) human HCC engrafted in the cirrhotic model compared with only 3 out of 19 (16%) that engrafted in the subcutaneous non-cirrhotic model. Tumor vasculature was preserved in the cirrhotic model but was diminished in the non-cirrhotic models. Metastasis developed in 3 cirrhotic PDX lines and was associated with early HCC recurrence in all 3 corresponding patients (100%), compared with only 5 out of 16 (31%) of the other PDX lines, P = .027. The cirrhotic model was able to predict response and non-response to lenvatinib and sorafenib respectively in the corresponding patients. Response to lenvatinib in the cirrhotic PDX was associated with reduction in CD34, VEGFR2 and CLEC4G immunofluorescence area and intensity (all P ≤ .03). CONCLUSIONS: A clinically relevant cirrhotic PDX model preserves tumor angiogenesis and allows prediction of response to multikinase inhibitors for personalized therapy.