Search

Objectives: To investigate the utility of deep learning (DL) automated segmentation-based MRI radiomic features and clinical-radiological characteristics in predicting early recurrence after curative resection of single hepatocellular carcinoma (HCC). Methods: This single-center, retrospective study included consecutive patients with surgically proven HCC who underwent contrast-enhanced MRI before curative hepatectomy from December 2009 to December 2021. Using 3D U-net-based DL algorithms, automated segmentation of the liver and HCC was performed on six MRI sequences. Radiomic features were extracted from the tumor, tumor border extensions (5 mm, 10 mm, and 20 mm), and the liver. A hybrid model incorporating the optimal radiomic signature and preoperative clinical-radiological characteristics was constructed via Cox regression analyses for early recurrence. Model discrimination was characterized with C-index and time-dependent area under the receiver operating curve (tdAUC) and compared with the widely-adopted BCLC and CNLC staging systems. Results: Four hundred and thirty-four patients (median age, 52.0 years; 376 men) were included. Among all radiomic signatures, HCC with5 mmtumorborderextensionandliver showed the optimal predictive performance (training set C-index, 0.696). By incorporating this radiomic signature, rim arterial phase hyperenhancement (APHE), and incomplete tumor "capsule," a hybrid model demonstrated a validation set C-index of 0.706 and superior 2-year tdAUC (0.743) than both the BCLC (0.550; p < 0.001) and CNLC (0.635; p = 0.032) systems. This model stratified patients into two prognostically distinct risk strata (both datasets p < 0.001). Conclusion: A preoperative imaging model incorporating the DL automated segmentation-based radiomic signature with rim APHE and incomplete tumor "capsule" accurately predicted early postsurgical recurrence of a single HCC. Critical relevance statement: The DL automated segmentation-based MRI radiomic model with rim APHE and incomplete tumor "capsule" hold the potential to facilitate individualized risk estimation of postsurgical early recurrence in a single HCC. Key Points: A hybrid model integrating MRI radiomic signature was constructed for early recurrence prediction of HCC. The hybrid model demonstrated superior 2-year AUC than the BCLC and CNLC systems. The model categorized the low-risk HCC group carried longer RFS. [ABSTRACT FROM AUTHOR]

Objectives: Contrast-enhanced MRI can provide individualized prognostic information for hepatocellular carcinoma (HCC). We aimed to investigate the value of MRI features to predict early (≤ 2 years)/late (> 2 years) recurrence-free survival (E-RFS and L-RFS, respectively) and overall survival (OS). Materials and methods: Consecutive adult patients at a tertiary academic center who received curative-intent liver resection for very early to intermediate stage HCC and underwent preoperative contrast-enhanced MRI were retrospectively enrolled from March 2011 to April 2021. Three masked radiologists independently assessed 54 MRI features. Uni- and multivariable Cox regression analyses were conducted to investigate the associations of imaging features with E-RFS, L-RFS, and OS. Results: This study included 600 patients (median age, 53 years; 526 men). During a median follow-up of 55.3 months, 51% of patients experienced recurrence (early recurrence: 66%; late recurrence: 34%), and 17% died. Tumor size, multiple tumors, rim arterial phase hyperenhancement, iron sparing in solid mass, tumor growth pattern, and gastroesophageal varices were associated with E-RFS and OS (largest p =.02). Nonperipheral washout (p =.006), markedly low apparent diffusion coefficient value (p =.02), intratumoral arteries (p =.01), and width of the main portal vein (p =.03) were associated with E-RFS but not with L-RFS or OS, while the VICT2 trait was specifically associated with OS (p =.02). Multiple tumors (p =.048) and radiologically-evident cirrhosis (p <.001) were the only predictors for L-RFS. Conclusion: Twelve visually-assessed MRI features predicted postoperative E-RFS (≤ 2 years), L-RFS (> 2 years), and OS for very early to intermediate-stage HCCs. Clinical relevance statement: The prognostic MRI features may help inform personalized surgical planning, neoadjuvant/adjuvant therapies, and postoperative surveillance, thus may be included in future prognostic models. Key Points: • Tumor size, multiple tumors, rim arterial phase hyperenhancement, iron sparing, tumor growth pattern, and gastroesophageal varices predicted both recurrence-free survival within 2 years and overall survival. • Nonperipheral washout, markedly low apparent diffusion coefficient value, intratumoral arteries, and width of the main portal vein specifically predicted recurrence-free survival within 2 years, while the VICT2 trait specifically predicted overall survival. • Multiple tumors and radiologically-evident cirrhosis were the only predictors for recurrence-free survival beyond 2 years. [ABSTRACT FROM AUTHOR]

Background: Hepatocellular carcinoma (HCC) expressing cytokeratin (CK) 7 or CK19 has a cholangiocyte phenotype that stimulates HCC proliferation, metastasis, and sorafenib therapy resistance This study aims to noninvasively predict cholangiocyte phenotype-positive HCC and assess its prognosis after hepatectomy. Methods: Between January 2010 and May 2022, preoperative contrast-enhanced MRI was performed on consecutive patients who underwent hepatectomy and had pathologically confirmed solitary HCC. Two abdominal radiologists separately assessed the MRI features. A predictive model for cholangiocyte phenotype HCC was created using logistic regression analysis and five-fold cross-validation. A receiver operating characteristic curve was used to calculate the model performance. Kaplan–Meier and log-rank methods were used to evaluate survival outcomes. Results: In total, 334 patients were included in this retrospective study. Four contrast-enhanced MRI features, including "rim arterial phase hyperenhancement" (OR = 5.9, 95% confidence interval [CI]: 2.9–12.0, 10 points), "nodule in nodule architecture" (OR = 3.5, 95% CI: 2.1–5.9, 7 points), "non-smooth tumor margin" (OR = 1.6, 95% CI: 0.8–2.9, 3 points), and "non-peripheral washout" (OR = 0.6, 95% CI: 0.3–1.0, − 3 points), were assigned to the cholangiocyte phenotype HCC prediction model. The area under the curves for the training and independent validation set were 0.76 and 0.73, respectively. Patients with model-predicted cholangiocyte phenotype HCC demonstrated lower rates of recurrence-free survival (RFS) and overall survival (OS) after hepatectomy, with an estimated median RFS and OS of 926 vs. 1565 days (p < 0.001) and 1504 vs. 2960 days (p < 0.001), respectively. Conclusions: Contrast-enhanced MRI features can be used to predict cholangiocyte phenotype-positive HCC. Patients with pathologically confirmed or MRI model-predicted cholangiocyte phenotype HCC have a worse prognosis after hepatectomy. Critical relevance statement: Four contrast-enhanced MRI features were significantly associated with cholangiocyte phenotype HCC and a worse prognosis following hepatectomy; these features may assist in predicting prognosis after surgery and improve personalized treatment decision-making. Key points: • Four contrast-enhanced MRI features were significantly associated with cholangiocyte phenotype HCC. • A noninvasive cholangiocyte phenotype HCC predictive model was established based on MRI features. • Patients with cholangiocyte phenotype HCC demonstrated a worse prognosis following hepatic resection. [ABSTRACT FROM AUTHOR]

Objective: To develop and validate a risk score based on preoperative clinical-radiological parameters for predicting overall survival (OS) in patients undergoing surgical resection for hepatocellular carcinoma (HCC). Methods: From July 2010 to December 2021, consecutive patients with surgically-proven HCC who underwent preoperative contrast-enhanced MRI were retrospectively enrolled. A preoperative OS risk score was constructed in the training cohort using a Cox regression model and validated in a propensity score-matched internal validation cohort and an external validation cohort. Results: A total of 520 patients were enrolled, among whom 210, 210, and 100 patients were from the training, internal validation, and external validation cohorts, respectively. Independent predictors for OS included incomplete tumor "capsule," mosaic architecture, tumor multiplicity, and serum alpha-fetoprotein, which were incorporated into the "OSASH score." The C-index the OSASH score was 0.85, 0.81, and 0.62 in the training, internal, and external validation cohorts, respectively. Using 32 as the cutoff point, the OSASH score stratified patients into prognostically distinct low- and high-risk groups among all study cohorts and six subgroups (all p < 0.05). Furthermore, patients with BCLC stage B-C HCC and OSASH-low risk achieved comparable OS to that of patients with BCLC stage 0-A HCC and OSASH-high risk in the internal validation cohort (5-year OS rates, 74.7 vs. 77.8%; p = 0.964). Conclusion: The OSASH score may help predict OS in HCC patients undergoing hepatectomy and identify potential surgical candidates among those with BCLC stage B-C HCC. Clinical relevance statement: By incorporating three preoperative MRI features and serum AFP, the OSASH score may help predict postsurgical overall survival in patients with hepatocellular carcinoma and identify potential surgical candidates among those with BCLC stage B and C HCC. Key Points: • The OSASH score incorporating three MRI features and serum AFP can be used to predict OS in HCC patients who received curative-intent hepatectomy. • The score stratified patients into prognostically distinct low- and high-risk strata in all study cohorts and six subgroups. • Among patients with BCLC stage B and C HCC, the score identified a subgroup of low-risk patients who achieved favorable outcomes after surgery. [ABSTRACT FROM AUTHOR]

Background: To establish a preoperative score based on gadoxetate disodium-enhanced magnetic resonance imaging (EOB-MRI) and clinical indicators for predicting histologic differentiation of solitary HCC up to 5 cm. Methods: From July 2015 to January 2022, consecutive patients with surgically proven solitary HCC measuring ≤ 5 cm at preoperative EOB-MRI were retrospectively enrolled. All MR images were independently evaluated by two radiologists who were blinded to all clinical and pathologic information. Univariate and multivariate logistic regression analyses were performed to identify significant clinicoradiological features associated with poorly differentiated (PD) HCC, which were then incorporated into the predictive score. The predictive score was validated in an independent validation set by area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. Results: A total of 182 patients were included, 42 (23%) with PD HCC. According to the multivariate analysis, marked hepatobiliary phase hypointensity (odds ratio [OR], 9.98), LR-M category (OR, 5.60), and serum alpha-fetoprotein (AFP) level > 400 ng/mL (OR, 3.58) were incorporated into the predictive model; the predictive score achieved an AUC of 0.802 and 0.830 on the training and validation sets, respectively. The sensitivity, specificity, and accuracy of the predictive score were 66.7%, 85.7%, and 81.3%, respectively, on the training set and 66.7%, 81.0%, and 77.8%, respectively, on the validation set. Conclusion: The proposed score integrating two EOB-MRI features and AFP level can accurately predict PD HCC in the preoperative setting. Key points: EOB-MRI features help capture the characteristics of tumor biology and heterogeneity. EOB-MRI-based HCC differentiation score allowed accurate assessment of poor tumor differentiation preoperatively. This scoring system might be useful for prompting tailored treatment selection. [ABSTRACT FROM AUTHOR]

Background: The Liver Imaging Reporting and Data System (LI‐RADS) is widely used for diagnosing hepatocellular carcinoma (HCC), however, with unsatisfactory sensitivity, complex ancillary features, and inadequate integration with gadoxetate disodium (EOB)‐enhanced MRI. Purpose: To modify LI‐RADS (mLI‐RADS) on EOB‐MRI. Study Type: Secondary analysis of a prospective observational study. Population: Between July 2015 and September 2018, 224 consecutive high‐risk patients (median age, 51 years; range, 26–83; 180 men; training/testing sets: 169/55 patients) with 742 (median size, 13 mm; interquartile range, 7–27; 498 HCCs) LR‐3/4/5 observations. Field Strength/Sequence: 3.0 T T2‐weighted fast spin‐echo, diffusion‐weighted spin‐echo based echo‐planar, and 3D T1‐weighted gradient echo sequences. Assessment: Three radiologists (with 5, 5, and 10 years of experience in liver MR imaging, respectively) blinded to the reference standard (histopathology or imaging follow‐up) reviewed all MR images independently. In the training set, the optimal LI‐RADS version 2018 (v2018) features selected by Random Forest analysis were used to develop mLI‐RADS via decision tree analysis. Statistical Tests: In an independent testing set, diagnostic performances of mLI‐RADS, LI‐RADS v2018, and the Korean Liver Cancer Association (KLCA) guidelines were computed using a generalized estimating equation model and compared with McNemar's test. A two‐tailed P < 0.05 was statistically significant. Results: Five features (nonperipheral "washout," restricted diffusion, nonrim arterial phase hyperenhancement [APHE], mild–moderate T2 hyperintensity, and transitional phase hypointensity) constituted mLI‐RADS, and mLR‐5 was nonperipheral washout coupled with either nonrim APHE or restricted diffusion. In the testing set, mLI‐RADS was significantly more sensitive (72%) and accurate (80%) than LI‐RADS v2018 (sensitivity, 61%; accuracy 74%; both P < 0.001) and the KLCA guidelines (sensitivity, 64%; accuracy 74%; both P < 0.001), without sacrificing positive predictive value (mLI‐RADS, 94%; LI‐RADS v2018, 94%; KLCA guidelines, 92%). Data Conclusion: In high‐risk patients, the EOB‐MRI‐based mLI‐RADS was simpler and more sensitive for HCC than LI‐RADS v2018 while maintaining high positive predictive value. Level of Evidence: 2 Technical Efficacy: Stage 2 [ABSTRACT FROM AUTHOR]

Background: The Liver Imaging Reporting and Data System (LI‐RADS) is widely accepted as a reliable diagnostic scheme for hepatocellular carcinoma (HCC) in at‐risk patients. However, its application is hampered by substantial complexity and suboptimal diagnostic sensitivity. Purpose: To propose data‐driven modifications to the LI‐RADS version 2018 (v2018) major feature system (rLI‐RADS) on gadoxetate disodium (EOB)‐enhanced magnetic resonance imaging (MRI) to improve sensitivity and simplicity while maintaining high positive predictive value (PPV) for detecting HCC. Study Type: Retrospective. Population: Two hundred and twenty‐four consecutive at‐risk patients (training dataset: 169, independent testing dataset: 55) with 742 LR‐3 to LR‐5 liver observations (HCC: N = 498 [67%]) were analyzed from a prospective observational registry collected between July 2015 and September 2018. Field Strength/Sequence: 3.0 T/T2‐weighted fast spin‐echo, diffusion‐weighted spin‐echo based echo‐planar and three‐dimensional (3D) T1‐weighted gradient echo sequences. Assessment: All images were evaluated by three independent abdominal radiologists who were blinded to all clinical, pathological, and follow‐up information. Composite reference standards of either histopathology or imaging follow‐up were used. Statistical Tests: In the training dataset, LI‐RADS v2018 major features were used to develop rLI‐RADS based on their associated PPV for HCC. In an independent testing set, diagnostic performances of LI‐RADS v2018 and rLI‐RADS were computed using a generalized estimating equation model and compared with McNemar's test. A P value <0.05 was considered statistically significant. Results: The median (interquartile range) size of liver observations was 13 mm (7–27 mm). The diagnostic table for rLI‐RADS encompassed 9 cells, as opposed to 16 cells for LI‐RADS v2018. In the testing set, compared to LI‐RADS v2018, rLI‐RADS category 5 demonstrated a significantly superior sensitivity (76% vs. 61%) while maintaining comparably high PPV (92.5% vs. 94.1%, P = 0.126). Data Conclusion: Compared with LI‐RADS v2018, rLI‐RADS demonstrated improved simplicity and significantly superior diagnostic sensitivity for HCC in at‐risk patients. Level of Evidence: 3 Technical Efficacy Stage: 2 [ABSTRACT FROM AUTHOR]

To develop a predictive model integrating clinical and MRI features for postoperative survival in patients with hepatocellular carcinoma (HCC) and portal vein tumor thrombus (PVTT). Between January 2008 and May 2021, consecutive HCC patients with PVTT who underwent preoperative contrast-enhanced MRI and surgical resection at a tertiary hospital were retrospectively enrolled. The MR images were independently reviewed by two blinded radiologists. Univariate and multivariate Cox regression analyses were performed to construct a prognostic score for overall survival (OS). Ninety-four patients were included (mean age, 50.1 years; 84 men). During a median follow-up period of 15.3 months, 72 (76.6%) patients died (median OS, 15.4 months; median disease-free survival [DFS], 4.6 months). The sum size of the two largest tumors (hazard ratio [HR], 3.050; p < 0.001) and tumor growth subtype (HR, 1.928; p = 0.006) on MRI, serum albumin (HR, 0.948; p = 0.02), and age (HR, 0.978; p = 0.04) were associated with OS and incorporated in the prognostic score. Accordingly, patients were stratified into a high-risk or low-risk group, and the OS in the high-risk group was shorter than that in the low-risk group for the entire cohort (11.7 vs. 25.0 months, p < 0.001) and for patients with Cheng's type I (12.1 vs. 25.9 months, p = 0.002) and type II PVTT (11.7 vs. 25.0 months, p = 0.004). The DFS in the high-risk group was shorter than that in the low-risk group for the entire cohort (4.5 vs. 6.1 months, p = 0.001). Based on the sum size of the two largest tumors, tumor growth subtype, albumin, and age, the prognostic score allowed accurate preoperative risk stratification in HCC patients with PVTT, independent of Cheng's PVTT classification. [ABSTRACT FROM AUTHOR]