scholarly journals Deep learning predicts postsurgical recurrence of hepatocellular carcinoma from digital histopathologic images

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Rikiya Yamashita ◽  
Jin Long ◽  
Atif Saleem ◽  
Daniel L. Rubin ◽  
Jeanne Shen
Keyword(s):  
Hepatocellular Carcinoma ◽  
Deep Learning ◽  
Surgical Resection ◽  
Treatment Strategies ◽  
Recurrence Risk ◽  
Disease Recurrence ◽  
Risk Scores ◽  
Post Surgery ◽  
Staging Systems ◽  
External Test

AbstractRecurrence risk stratification of patients undergoing primary surgical resection for hepatocellular carcinoma (HCC) is an area of active investigation, and several staging systems have been proposed to optimize treatment strategies. However, as many as 70% of patients still experience tumor recurrence at 5 years post-surgery. We developed and validated a deep learning-based system (HCC-SurvNet) that provides risk scores for disease recurrence after primary resection, directly from hematoxylin and eosin-stained digital whole-slide images of formalin-fixed, paraffin embedded liver resections. Our model achieved concordance indices of 0.724 and 0.683 on the internal and external test cohorts, respectively, exceeding the performance of the standard Tumor-Node-Metastasis classification system. The model’s risk score stratified patients into low- and high-risk subgroups with statistically significant differences in their survival distributions, and was an independent risk factor for post-surgical recurrence in both test cohorts. Our results suggest that deep learning-based models can provide recurrence risk scores which may augment current patient stratification methods and help refine the clinical management of patients undergoing primary surgical resection for HCC.

scholarly journals Deep learning predicts post-surgical recurrence of hepatocellular carcinoma from digital whole-slide images

2020 ◽  
Author(s):  
Rikiya Yamashita ◽  
Jin Long ◽  
Atif Saleem ◽  
Daniel L Rubin ◽  
Jeanne Shen
Keyword(s):  
Deep Learning ◽  
Surgical Resection ◽  
Tumor Recurrence ◽  
Recurrence Risk ◽  
Risk Scores ◽  
Test Set ◽  
Surgical Recurrence ◽  
Test Sets ◽  
External Test ◽  
Whole Slide Images

Recurrence risk stratification of patients undergoing primary surgical resection for hepatocellular carcinoma (HCC) is an area of active investigation, and several staging systems have been proposed to optimize treatment strategies. However, as many as 70% of patients still have tumor recurrence at 5 years post-surgery. Routine hematoxylin and eosin (H&E)-stained histopathology slides may contain morphologic features associated with tumor recurrence. In this study, we developed and independently validated a deep learning-based system (HCC-SurvNet) that provides risk scores for disease recurrence after primary surgical resection, directly from H&E-stained digital whole-slide images of formalin-fixed, paraffin embedded liver resections. Our model achieved a concordance index of 0.724 on a held-out internal test set of 53 patients, and 0.683 on an external test set of 198 patients, exceeding the performance of standard staging using the American Joint Committee on Cancer (AJCC)/International Union against Cancer (UICC) Tumor-Node-Metastasis (TNM) classification system, on both the internal and external test cohorts (p=0.018 and 0.025, respectively). We observed statistically significant differences in the survival distributions between low- and high-risk subgroups, as stratified by the risk scores predicted by HCC-SurvNet on both the internal and external test sets (log-rank p-value: 0.0013 and <0.0001, respectively). On multivariable Cox proportional hazards analysis, the risk score was an independent risk factor for post-surgical recurrence, on both the internal (hazard ratio (HR)=7.44 (95% CI: 1.60, 34.6), p=0.0105) and external (HR=2.37 (95% CI: 1.27, 4.43), p=0.0069) test sets. Our results suggest that deep learning-based models can provide recurrence risk scores which may augment current patient stratification methods, and help refine the clinical management of patients undergoing primary surgical resection for HCC.

scholarly journals 5-year Recurrence Prediction After Hepatocellular Carcinoma Resection Using Deep Learning and Cox Regression Models: A Large Prospective Study

2021 ◽  
Author(s):  
Hon-Yi Shi ◽  
King-The Lee ◽  
Chong-Chi Chiu ◽  
Jhi-Joung Wang ◽  
Ding-Ping Sun ◽  
...  
Keyword(s):  
Neural Network ◽  
Hepatocellular Carcinoma ◽  
Deep Learning ◽  
Surgical Resection ◽  
Scale Score ◽  
Cox Regression ◽  
Recurrence Prediction ◽  
Feature Importance ◽  
Importance Analysis ◽  
Hcc Recurrence

Abstract BackgroundRisk of hepatocellular carcinoma (HCC) recurrence after surgical resection is unknown. Therefore, the aim of this study was 5-year recurrence prediction after HCC resection using deep learning and Cox regression models.MethodsThis study recruited 520 HCC patients who had undergone surgical resection at three medical centers in southern Taiwan between April, 2011, and December, 2015. Two popular deep learning algorithms: a deep neural network (DNN) model and a recurrent neural network (RNN) model and a Cox proportional hazard (CPH) regression model were designed to solve both classification problems and regression problems in predicting HCC recurrence. A feature importance analysis was also performed to identify confounding factors in the prediction of HCC recurrence in patients who had undergone resection.ResultsAll performance indices for the DNN model were significantly higher than those for the RNN model and the traditional CPH model (p<0.001). The most important confounding factor in 5-year recurrence after HCC resection was surgeon volume followed by, in order of importance, hospital volume, preoperative Beck Depression Scale score, preoperative Beck Anxiety Scale score, co-residence with family, tumor stage, and tumor size. ConclusionsThe DNN model is useful for early baseline prediction of 5-year recurrence after HCC resection. Its prediction accuracy can be improved by further training with temporal data collected from treated patients. The feature importance analysis performed in this study to investigate model interpretability provided important insights into the potential use of deep learning models for predicting recurrence after HCC resection and for identifying predictors of recurrence.

Exploring prognostic indicators in the pathological images of hepatocellular carcinoma based on deep learning

Gut ◽  
2020 ◽  
pp. gutjnl-2020-320930 ◽  
Author(s):  
Jie-Yi Shi ◽  
Xiaodong Wang ◽  
Guang-Yu Ding ◽  
Zhou Dong ◽  
Jing Han ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Deep Learning ◽  
Genetic Alterations ◽  
The Cancer Genome Atlas ◽  
Clinical Staging ◽  
Patient Treatment ◽  
Prognostic Indicators ◽  
Omics Data ◽  
Learning Framework ◽  
Staging Systems

ObjectiveTumour pathology contains rich information, including tissue structure and cell morphology, that reflects disease progression and patient survival. However, phenotypic information is subtle and complex, making the discovery of prognostic indicators from pathological images challenging.DesignAn interpretable, weakly supervised deep learning framework incorporating prior knowledge was proposed to analyse hepatocellular carcinoma (HCC) and explore new prognostic phenotypes on pathological whole-slide images (WSIs) from the Zhongshan cohort of 1125 HCC patients (2451 WSIs) and TCGA cohort of 320 HCC patients (320 WSIs). A ‘tumour risk score (TRS)’ was established to evaluate patient outcomes, and then risk activation mapping (RAM) was applied to visualise the pathological phenotypes of TRS. The multi-omics data of The Cancer Genome Atlas(TCGA) HCC were used to assess the potential pathogenesis underlying TRS.ResultsSurvival analysis revealed that TRS was an independent prognosticator in both the Zhongshan cohort (p<0.0001) and TCGA cohort (p=0.0003). The predictive ability of TRS was superior to and independent of clinical staging systems, and TRS could evenly stratify patients into up to five groups with significantly different prognoses. Notably, sinusoidal capillarisation, prominent nucleoli and karyotheca, the nucleus/cytoplasm ratio and infiltrating inflammatory cells were identified as the main underlying features of TRS. The multi-omics data of TCGA HCC hint at the relevance of TRS to tumour immune infiltration and genetic alterations such as the FAT3 and RYR2 mutations.ConclusionOur deep learning framework is an effective and labour-saving method for decoding pathological images, providing a valuable means for HCC risk stratification and precise patient treatment.

Adjuvant lenvatinib in combination with TACE for hepatocellular carcinoma patients with high risk of postoperative relapse (LANCE): Interim results from a muticenter prospective cohort study.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. 4580-4580
Author(s):  
Jin-hong Chen ◽  
Lu Lu ◽  
Tian-Fu Wen ◽  
Zhi-Yong Huang ◽  
Ti Zhang ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Cohort Study ◽  
Bile Duct ◽  
Adjuvant Therapy ◽  
Surgical Resection ◽  
Prospective Cohort Study ◽  
Prospective Cohort ◽  
Recurrence Risk ◽  
Tace Group ◽  
The Media

4580 Background: Surgical resection was the main treatment for hepatocellular carcinoma (HCC) in China. Multiple clinical studies had demonstrated that the overall survival (OS) of the surgical resection group was significantly better than the transcatheter arterial chemoembolization (TACE) or radiotherapy group even for HCC patients with BCLC stage B or C. There was no standard adjuvant therapy for HCC patients to decrease the post-operative tumor relapse. For HCC patients with high recurrence risk, TACE significantly reduced tumor recurrence, prolonged the disease free survival (DFS) and OS, and was recommended as the adjuvant therapy. However, its effect is not very satisfactory. The purpose of this study was to assess the efficacy and safety of lenvatinib in combination with TACE versus TACE alone as adjuvant therapy in HCC patients with high recurrence risk after resection. Methods: This is a muti-center prospective cohort study. The criteria of HCC patients with high postoperative recurrence risk included: accompanied with gross vascular or bile duct invasion (tumor thrombi in portal vein, hepatic vein or bile duct); or tumor rupture or invasion of adjacent organs; or grade 2 of microvascular invasion (MVI) (M2) along with the tumor number more than 3 or the maximum diameter of tumor larger than 8cm or tumor showed invasive growth with unclear boundaries and imcomplete capsules. The patients were divided into two groups, the lenvatinb (8mg qd for weights < 60kg and 12mg qd for weights≥60kg) in combination with TACE (Len+TACE) group and the TACE group. Results: A total of 90 patients were enrolled into the study, while 45 patients in the Len+TACE group and 45 in TACE group. The media age was 52 years (range from 23 to 73 years). Most patients were males (82.2%) and 66 patients had HBV background (73.3%). There were no significant differences between the two groups in the baseline clinicopathological characteristics including gender, age, HBV background, liver cirrhosis, liver function, tumor characteristic and AFP level. The media DFS was 12.0 months (95% CI 8.0-NA) in the Len+ TACE group, which was longer than that of TACE group (8.0 months, 95% CI 6.0-12.0, P = 0.0359; HR 0.5, 95% CI 0.3-1.0). The most common grade 3 or 4 adverse events were hypertension (11.1%) and diarrhea (7.7%) in the Len+TACE group. Conclusions: Lenvatinib in combination with TACE was effective and safe as adjuvant therapy, which can prolong the DFS of HCC patients with high recurrence risk after resection. Clinical trial information: NCT03838796 .

Patient Selection, Resection, and Outcomes for Hepatocellular Carcinoma

2012 ◽  
pp. 265-269
Author(s):  
Claudius Conrad ◽  
Kenneth K. Tanabe
Keyword(s):  
Hepatocellular Carcinoma ◽  
Surgical Resection ◽  
Patient Selection ◽  
Laboratory Data ◽  
Intraoperative Ultrasound ◽  
Definitive Treatment ◽  
Transfusion Requirements ◽  
Staging Systems ◽  
Selection Of Patients ◽  
Selection Of

Overview: Hepatocellular carcinoma (HCC) is an aggressive malignancy of the liver that most often arises in patients with cirrhosis and other chronic liver diseases. Worldwide, it is the sixth most common cancer and the third most common cause of cancer-related death. Median survival is poor, ranging from 6 to 20 months. Definitive treatment options for HCC are surgical resection, ablation, or transplantation. The selection of patients for surgical resection is based on clinical findings, laboratory data, and imaging. Although a number of staging systems exist, all have their limitations. A multidisciplinary approach to patient selection for surgery that includes the input of an experienced liver surgeon assures optimal outcomes. Sound understanding of liver segmentation, modern surgical techniques, and the use of intraoperative ultrasound have led to a reported perioperative mortality rate below 3%, blood transfusion requirements of less than 10%, and 5-year survival rates of at least 50%. Advances in laparoscopic technique and technology have expanded the indications for a safe and oncologically appropriate minimally invasive resection. Deciding which treatment option to employ depends on tumor resectability and the degree of underlying liver disease, which is present in 80% to 85% of patients with HCC; however, despite these surgical advances, a high recurrence rate of 70% in patients with cirrhosis and a survival rate of 65% to 80% in well-selected transplant patients are expected. This article will focus on the evaluation and selection of patients for surgical intervention, considerations in selecting the appropriate type of resection, and expected outcomes following liver resection.

nm23-H1 Expression and disease recurrence after surgical resection of small hepatocellular carcinoma

1994 ◽  
Vol 107 (2) ◽  
pp. 486-491 ◽  
Author(s):  
Loreto Boix ◽  
Jordi Bruix ◽  
Elías Campo ◽  
Manel Sole ◽  
Antoni Castells ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Surgical Resection ◽  
Disease Recurrence ◽  
Small Hepatocellular Carcinoma

scholarly journals Hepatic resection for recurrent hepatocellular carcinoma during pregnancy: a case report

2020 ◽  
Vol 6 (1) ◽  
Author(s):  
Takashi Maeda ◽  
Daisuke Imai ◽  
Huanlin Wang ◽  
Kyohei Yugawa ◽  
Nao Kinjo ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Informed Consent ◽  
Hepatic Resection ◽  
Retention Rate ◽  
Laboratory Data ◽  
Treatment Strategies ◽  
Normal Range ◽  
Her Family ◽  
Post Surgery

Abstract Background Hepatocellular carcinoma (HCC) during pregnancy is extremely rare. Treatment strategies for cancers detected during pregnancy have been controversial. We herein report a case of recurrent HCC detected at 20 weeks of pregnancy, which subsequently prompted hepatic resection after abortion. Case presentation A 36-year-old woman underwent laparoscopic partial hepatectomy for HCC (20 mm in diameter) in segment 5 of the liver during follow-up after being determined as a hepatitis B virus carrier two and a half years ago. Post-surgery follow-up abdominal ultrasonography revealed a 36-mm tumor in segment 7 of the liver. Abdominal contrast-enhanced computed tomography revealed a well-enhanced tumor with a 40-mm diameter in segment 7 adjacent to the inferior vena cava and right hepatic vein, suggesting HCC recurrence. Laboratory data revealed total bilirubin (0.4 mg/dL), aspartate aminotransferase (28 IU/L), alanine aminotransferase (30 IU/L), glutamyltransferase (16 IU/L), prothrombin time (115.3%), and indocyanine green retention rate at 15 min (7.0%). α-Fetoprotein (AFP) (12,371.5 ng/mL; normal range < 10 ng/mL) and PIVKA-II (208 mAU/mL; normal range < 40 mAU/mL) were both significantly elevated. After discussions with a cancer board consisting of experts from the departments of gastroenterology, obstetrics and gynecology, and surgery, as well as obtaining appropriate informed consent from the patient and her family, we decided to perform a hepatic resection after abortion. Subsequently, abortion surgery was performed at 21 weeks and 2 days of pregnancy. After 6 days, subsegmentectomy of liver segment 7 was performed under general and epidural anesthesia, with a pathological diagnosis which was moderately differentiated HCC being established. Given the good postoperative course, without particular complications, the patient was subsequently discharged 10 days after the operation. Approximately 2 years after the surgery, the patient remains alive without recurrence, while both AFP and PIVKA-II were within normal limits. Conclusions Treatment strategies for HCC detected during pregnancy remain controversial. As such, decisions should be made based on HCC growth and fetal maturity after thorough multidisciplinary team discussions and obtaining appropriate informed consent from the patient and her family.

A prognostic model for early recurrence prediction in hepatocellular carcinoma after liver resection.

2013 ◽  
Vol 31 (15_suppl) ◽  
pp. e15021-e15021
Author(s):  
Dan-Yun Ruan ◽  
Yang Li ◽  
Nan Jiang ◽  
Ze-Xiao Lin ◽  
XiangBo Wan ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Liver Resection ◽  
Prognostic Model ◽  
Risk Model ◽  
Recurrence Risk ◽  
Early Recurrence ◽  
Tumor Burden ◽  
High Morbidity ◽  
Staging Systems ◽  
Early Tumor

e15021 Background: Recurrence resulting in high morbidity of hepatocellular carcinoma (HCC) patients received radical liver resection, the majority of which occur within 1 to 3 years. The study aims to find risk factors and set up a prognostic model predicting early recurrence of HCC. Methods: From 2003 to 2010, 196 HCC patients underwent liver resection were retrospectively analyzed. Univariate and multivariate analyses were used to assess variables. A recurrence risk model was developed with independent prognostic factors. Area under the ROC curve (AUC) was carried out to evaluate its predictive value. Results: The median follow-up time was 33 months (1-103M), median RFS was 22 months. Total tumor volume (TTV), hepatitis B, Child-Pugh score, and portal vein tumor thrombus were independent factors of recurrence. Patients with TTV>115cm3 had worse RFS (28.9% vs 51.7%, p<0.000) and OS (64.4% vs 80.1%, p=0.032) than those TTV≤115cm3. We established a risk model consisted of the 4 parameters, and classified patients into four stages. There were significant differences between each stage, especially for the 1st year. (1-yr RFS: 80.1% vs 47.8% vs 28.6% vs 0%, p=0.000, AUC 0.682). Compared to four currently used staging systems (BCLC, TNM, CLIP, and Okuda), the new model could well predict patient’s survival with the largest AUC in both RFS and OS (Table). Conclusions: TTV has been found a preferable description of tumor burden and a prognostic factor in HCC. This prognostic model predicts early tumor recurrence and survival of patients received radical liver resection and might contributes to the selection of patients who may benefit from surgery. [Table: see text]

scholarly journals Yttrium-90 Radioembolization of Isolated Hepatic Adrenocortical Carcinoma Metastases With Negative Surgical Pathology

2020 ◽  
Author(s):  
Sen Lu ◽  
Jasreman Dhillon ◽  
Julie Hallanger Johnson ◽  
Ghassan El-Haddad
Keyword(s):  
Surgical Resection ◽  
External Beam Radiotherapy ◽  
Hepatic Metastases ◽  
Subsequent Development ◽  
Disease Recurrence ◽  
Adrenal Crisis ◽  
Inferior Phrenic Artery ◽  
Post Surgery ◽  
Phrenic Artery ◽  
The Right

Abstract BackgroundAdrenocortical carcinoma (ACC) is an uncommon malignancy with an estimated 15,400 new cases annually across the globe. The prognosis is generally poor as the disease is often already advanced at initial diagnosis due to non-specific symptoms. Even for local disease, recurrence after surgical resection is high. Treatment choices for advanced disease include Mitotane, chemotherapy, ablation, chemoembolization, radioembolization, and external beam radiotherapy, with varying degrees of efficacy. To the best of our knowledge, there have only been two prior case studies of complete clinical and radiological response of stage 4 disease at 1 year and 2 years after [90Y]yttrium (Y-90) microsphere selective internal radiation therapy (SIRT) of isolated hepatic metastases post-surgery and chemotherapy.Case presentationWe present a case of stage 1 ACC at time of diagnosis with subsequent development of hepatic metastases 6 years later, treated with Y-90 SIRT along with surgical resection and chemotherapy, with a surgically-proven negative pathology after partial hepatectomy 7 months after Y-90 SIRT. Bland embolization was used to successfully redistribute flow from a feeding right inferior phrenic artery to the right hepatic artery for effective selective radioembolization. Due to long term use of hydrocortisone causing adrenal insufficiency, the patient developed a post SIRT adrenal crisis, which was successfully controlled with steroids, highlighting the need for pre SIRT stress dose steroids. ConclusionsThis case continues to add to the literature supporting Y-90 radioembolization as an effective treatment for isolated hepatic ACC metastases. Our case is the first to demonstrate surgically proven negative pathology after radioembolization. Further prospective study is warranted to better establish efficacy as well as safety of SIRT for ACC liver metastases.

P-L12 Exploring Pathological Signatures for Predicting Recurrence of Early-stage Hepatocellular Carcinoma Based on Deep Learning

2021 ◽  
Vol 108 (Supplement_9) ◽  
Author(s):  
Yinghong Shi ◽  
Weifeng Qu ◽  
Mengxin Tian ◽  
Jingtao Qiu ◽  
Kun Qian ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Deep Learning ◽  
Validation Cohort ◽  
Early Stage ◽  
External Validation ◽  
Microvascular Invasion ◽  
Risk Scores ◽  
Forecast Performance ◽  
Histological Score ◽  
The Relationship

Abstract Background Early-stage hepatocellular carcinoma (HCC) is the ideal indication for liver resection. High recurrence rate limits the radical possibility. Current clinicopathological determinants are insufficient to reliably evaluate the recurrence risk after surgery. To address this global issue, we aimed to use deep learning to explore novel pathological signatures based on histological slides for predicting early-stage HCC recurrence and to evaluate the relationship between histological features and multi-omics information. Methods 576 pathological images collected from 547 patients with BCLC stage 0-A HCC who underwent hepatectomy from 2006 to 2015 were randomly divided into the training cohort and validation cohort. The external validation cohort was composed of 147 TNM I patients from TCGA database. Weakly supervised convolutional neural networks were used to identify six classes of HCC tissues. Pathological signatures were extracted and two novel risk scores were constructed by LASSO Cox to predict recurrence. The forecast performance of the scores and patients' prognosis were evaluated. The relationship between histological score (HS) and immune infiltrating cells was estimated by clustering analysis. Results The classification accuracy of HCC tissue was 94.17%. The C-indexes of histological score in the training, validation and TCGA cohorts were 0.804, 0.739 and 0.708, respectively. Multivariate analysis showed that microvascular invasion (HR = 1.46, 95% CI: 1.09-1.95) and HS (HR = 4.05, 95% CI: 3.40-4.84) were independent risk factors for recurrence-free survival. Patients in HS high-risk group had elevated  alpha fetoprotein, worse tumor differentiation and higher proportion of microvascular invasion. HS was positively correlated with the expression of CD14 in adjacent normal liver tissue (P = 0.013), and negatively correlated with the expression of CD8 in tumor (P &lt; 0.001). Conclusions This study established and validated two novel risk scores based on the histological slides using deep learning. HS performed well in recurrence prediction for early-stage HCC patients and indication of important clinicopathological features.

Sign in / Sign up

Export Citation Format

Share Document