Leveraging machine learning for early recurrence prediction in hepatocellular carcinoma:A step towards precision medicine

The high rate of early recurrence in hepatocellular carcinoma(HCC)post curative surgical intervention poses a substantial clinical hurdle,impacting patient outcomes and complicating postoperative management.The advent of machine learning provides a unique opportunity to harness vast datasets,identifying subtle patterns and factors that elude conventional prognostic methods.Machine learning models,equipped with the ability to analyse intricate relationships within datasets,have shown promise in predicting outcomes in various medical disciplines.In the context of HCC,the application of machine learning to predict early recurrence holds potential for personalized postoperative care strategies.This editorial comments on the study carried out exploring the merits and efficacy of random survival forests(RSF)in identifying significant risk factors for recurrence,stratifying patients at low and high risk of HCC recurrence and comparing this to traditional COX proportional hazard models(CPH).In doing so,the study demonstrated that the RSF models are superior to traditional CPH models in predicting recurrence of HCC and represent a giant leap towards precision medicine.

Computed tomography-based radiomics to predict early recurrence of hepatocellular carcinoma post-hepatectomy in patients background on cirrhosis

BACKGROUND The prognosis for hepatocellular carcinoma(HCC)in the presence of cirrhosis is unfavourable,primarily attributable to the high incidence of recurrence.AIM To develop a machine learning model for predicting early recurrence(ER)of posthepatectomy HCC in patients with cirrhosis and to stratify patients’overall survival(OS)based on the predicted risk of recurrence.METHODS In this retrospective study,214 HCC patients with cirrhosis who underwent curative hepatectomy were examined.Radiomics feature selection was conducted using the least absolute shrinkage and selection operator and recursive feature elimination methods.Clinical-radiologic features were selected through univariate and multivariate logistic regression analyses.Five machine learning methods were used for model comparison,aiming to identify the optimal model.The model’s performance was evaluated using the receiver operating characteristic curve[area under the curve(AUC)],calibration,and decision curve analysis.Additionally,the Kaplan-Meier(K-M)curve was used to evaluate the stratification effect of the model on patient OS.RESULTS Within this study,the most effective predictive performance for ER of post-hepatectomy HCC in the background of cirrhosis was demonstrated by a model that integrated radiomics features and clinical-radiologic features.In the training cohort,this model attained an AUC of 0.844,while in the validation cohort,it achieved a value of 0.790.The K-M curves illustrated that the combined model not only facilitated risk stratification but also exhibited significant discriminatory ability concerning patients’OS.CONCLUSION The combined model,integrating both radiomics and clinical-radiologic characteristics,exhibited excellent performance in HCC with cirrhosis.The K-M curves assessing OS revealed statistically significant differences.

Postoperative serum tumor markers-based nomogram predicting early recurrence for patients undergoing radical resections of pancreatic ductal adenocarcinoma

BACKGROUND Early recurrence(ER)is associated with dismal outcomes in patients undergoing radical resection for pancreatic ductal adenocarcinoma(PDAC).Approaches for predicting ER will help clinicians in implementing individualized adjuvant therapies.Postoperative serum tumor markers(STMs)are indicators of tumor progression and may improve current systems for predicting ER.AIM To establish an improved nomogram based on postoperative STMs to predict ER in PDAC.METHODS We retrospectively enrolled 282 patients who underwent radical resection for PDAC at our institute between 2019 and 2021.Univariate and multivariate Cox regression analyses of variables with or without postoperative STMs,were performed to identify independent risk factors for ER.A nomogram was constructed based on the independent postoperative STMs.Receiver operating characteristic curve analysis was used to evaluate the area under the curve(AUC)of the nomogram.Survival analysis was performed using Kaplan-Meier survival plot and log-rank test.RESULTS Postoperative carbohydrate antigen 19-9 and carcinoembryonic antigen levels,preoperative carbohydrate antigen 125 levels,perineural invasion,and pTNM stage III were independent risk factors for ER in PDAC.The postoperative STMs-based nomogram(AUC:0.774,95%CI:0.713-0.835)had superior accuracy in predicting ER compared with the nomogram without postoperative STMs(AUC:0.688,95%CI:0.625-0.750)(P=0.016).Patients with a recurrence nomogram score(RNS)>1.56 were at high risk for ER,and had significantly poorer recurrence-free survival[median:3.08 months,interquartile range(IQR):1.80-8.15]than those with RNS≤1.56(14.00 months,IQR:6.67-24.80),P<0.001).CONCLUSION The postoperative STMs-based nomogram improves the predictive accuracy of ER in PDAC,stratifies the risk of ER,and identifies patients at high risk of ER for tailored adjuvant therapies.

Recurrence scoring system predicting early recurrence for patients with pancreatic ductal adenocarcinoma undergoing pancreatectomy and portomesenteric vein resection

BACKGROUND Pancreatectomy with concomitant portomesenteric vein resection(PVR)enables patients with portomesenteric vein(PV)involvement to achieve radical resection of pancreatic ductal adenocarcinoma,however,early recurrence(ER)is frequently observed.AIM To predict ER and identify patients at high risk of ER for individualized therapy.METHODS Totally 238 patients undergoing pancreatectomy and PVR were retrospectively enrolled and were allocated to the training or validating cohort.Univariate Cox and LASSO regression analyses were performed to construct serum recurrence score(SRS)based on 26 serum-derived parameters.Uni-and multivariate Cox regression analyses of SRS and 18 clinicopathological variables were performed to establish a Nomogram.Receiver operating characteristic curve analysis was used to evaluate the predictive accuracy.Survival analysis was performed using Kaplan-Meier method and log-rank test.RESULTS Independent serum-derived recurrence-relevant factors of LASSO regression model,including postoperative carbohydrate antigen 19-9,postoperative carcinoembryonic antigen,postoperative carbohydrate antigen 125,preoperative albumin(ALB),preoperative platelet to ALB ratio,and postoperative platelets to lymphocytes ratio,were used to construct SRS[area under the curve(AUC):0.855,95%CI:0.786–0.924].Independent risk factors of recurrence,including SRS[hazard ratio(HR):1.688,95%CI:1.075-2.652],pain(HR:1.653,95%CI:1.052-2.598),perineural invasion(HR:2.070,95%CI:0.827-5.182),and PV invasion(HR:1.603,95%CI:1.063-2.417),were used to establish the recurrence nomogram(AUC:0.869,95%CI:0.803-0.934).Patients with either SRS>0.53 or recurrence nomogram score>4.23 were considered at high risk for ER,and had poor long-term outcomes.CONCLUSION The recurrence scoring system unique for pancreatectomy and PVR,will help clinicians in predicting recurrence efficiently and identifying patients at high risk of ER for individualized therapy.

Identifying timing and risk factors for early recurrence of resectable rectal cancer: A single center retrospective study

BACKGROUND Colorectal cancer is a common malignancy and various methods have been introduced to decrease the possibility of recurrence.Early recurrence(ER)is related to worse prognosis.To date,few observational studies have reported on the analysis of rectal cancer.Hence,we reported on the timing and risk factors for the ER of resectable rectal cancer at our institute.AIM To analyze a cohort of patients with local and/or distant recurrence following the radical resection of the primary tumor.METHODS Data were retrospectively collected from the institutional database from March 2011 to January 2021.Clinicopathological data at diagnosis,perioperative and postoperative data,and first recurrence were collected and analyzed.ER was defined via receiver operating characteristic curve.Prognostic factors were evaluated using the Kaplan–Meier method and Cox proportional hazards modeling.RESULTS We included 131 patients.The optimal cut off value of recurrence-free survival(RFS)to differentiate between ER(n=55,41.9%)and late recurrence(LR)(n=76,58.1%)was 8 mo.The median post-recurrence survival(PRS)of ER and LR was 1.4 mo and 2.9 mo,respectively(P=0.008)but PRS was not strongly associated with RFS(R^(2)=0.04).Risk factors included age≥70 years[hazard ratio(HR)=1.752,P=0.047],preoperative concurrent chemoradiotherapy(HR=3.683,P<0.001),colostomy creation(HR=2.221,P=0.036),and length of stay>9 d(HR=0.441,P=0.006).CONCLUSION RFS of 8 mo was the optimal cut-off value.Although ER was not associated with PRS,it was still related to prognosis;thus,intense surveillance is recommended.

Radiomics model based on contrast-enhanced computed tomography to predict early recurrence in patients with hepatocellular carcinoma after radical resection

BACKGROUND Radical resection remains an effective strategy for patients with hepatocellular carcinoma(HCC).Unfortunately,the postoperative early recurrence(recurrence within 2 years)rate is still high.AIM To develop a radiomics model based on preoperative contrast-enhanced computed tomography(CECT)to evaluate early recurrence in HCC patients with a single tumour.METHODS We enrolled a total of 402 HCC patients from two centres who were diagnosed with a single tumour and underwent radical resection.First,the features from the portal venous and arterial phases of CECT were extracted based on the region of interest,and the early recurrence-related radiomics features were selected via the least absolute shrinkage and selection operator proportional hazards model(LASSO Cox)to determine radiomics scores for each patient.Then,the clinicopathologic data were combined to develop a model to predict early recurrence by Cox regression.Finally,we evaluated the prediction performance of this model by multiple methods.RESULTS A total of 1915 radiomics features were extracted from CECT images,and 31 of them were used to determine the radiomics scores,which showed a significant difference between the early recurrence and nonearly recurrence groups.Univariate and multivariate Cox regression analyses showed that radiomics scores and serum alphafetoprotein were independent indicators,and they were used to develop a combined model to predict early recurrence.The area under the receiver operating characteristic curve values for the training and validation cohorts were 0.77 and 0.74,respectively,while the C-indices were 0.712 and 0.674,respectively.The calibration curves and decision curve analysis showed satisfactory accuracy and clinical utilities.Kaplan-Meier curves based on recurrence-free survival and overall survival showed significant differences.CONCLUSION The preoperative radiomics model was shown to be effective for predicting early recurrence among HCC patients with a single tumour.

Patients with early recurrence of hepatocellular carcinoma have poor prognosis

BACKGROUND： Early recurrence （ER） after hepatic resection （HR） is a poor prognostic factor for patients with hepatocellular carcinoma （HCC）. This study aimed to identify the clinico- pathological features, outcomes, and risk factors for ER after HR for small HCC in order to clarify the reasons why ER is a worse recurrence pattern. METHODS： We retrospectively examined 130 patients who underwent HR for small HCC （___30 mm）. Recurrence was clas- sifted into ER （〈2 years） and late recurrence （LR） （_〉2 years）. The clinicopathological features, outcomes, and risk factors for ER were analyzed by multivariate analysis. RESULTS： ER was observed in 39 patients （30.0%）. The sur- vival rate of the ER group was significantly lower than that of the LR group （P〈0.005）, and ER was an independent prognos- tic factor for poor survival （P=0.0001）. The ER group had a significantly higher frequency （P=0.0039） and shorter interval （P=0.027） of development to carcinoma beyond the Milan criteria （DBMC） compared with the LR group, and ER was an independent risk factor for DBMC （P〈0.0001）. Multi-nodularity, non-simple nodular type, and microvascular invasion were independent predictors for ER （P=0.012, 0.010, and 0.019, respectively）.CONCLUSIONS： ER was a highly malignant recurrence pattern associated with DBMC and subsequent poor survival after HR for small HCC. Multi-nodularity, non-simple nodular type, and microvascular invasion predict ER, and taking these factors into consideration may be useful for the decision of the treatment strategy for small HCC after HR.

Preoperative serum carbohydrate antigen 19-9 levels predict early recurrence after the resection of early-stage pancreatic ductal adenocarcinoma

BACKGROUND Pancreatic ductal adenocarcinoma(PDAC)is a serious disease with a poor prognosis.Only a minority of patients undergo surgery due to the advanced stage of the disease,and patients with early-stage disease,who are expected to have a better prognosis,often experience recurrence.Thus,it is important to identify the risk factors for early recurrence and to develop an adequate treatment plan.AIM To evaluate the predictive factors associated with the early recurrence of earlystage PDAC.METHODS This study enrolled 407 patients with stage I PDAC undergoing upfront surgical resection between January 2000 and April 2016.Early recurrence was defined as a diagnosis of recurrence within 6 mo of surgery.The optimal cutoff values were determined by receiver operating characteristic(ROC)analyses.Univariate and multivariate analyses were performed to identify the risk factors for early recurrence.RESULTS Of the 407 patients,98 patients(24.1%)experienced early disease recurrence:26(26.5%)local and 72(73.5%)distant sites.In total,253(62.2%)patients received adjuvant chemotherapy.On ROC curve analysis,the optimal cutoff values for early recurrence were 70 U/mL and 2.85 cm for carbohydrate antigen 19-9(CA 19-9)levels and tumor size,respectively.Of the 181 patients with CA 19-9 level>70 U/mL,59(32.6%)had early recurrence,compared to 39(17.4%)of 226 patients with CA 19-9 level≤70 U/mL(P<0.001).Multivariate analysis revealed that CA 19-9 level>70 U/mL(P=0.006),tumor size>2.85 cm(P=0.004),poor differentiation(P=0.008),and non-adjuvant chemotherapy(P=0.025)were significant risk factors for early recurrence in early-stage PDAC.CONCLUSION Elevated CA 19-9 level(cutoff value>70 U/mL)can be a reliable predictive factor for early recurrence in early-stage PDAC.As adjuvant chemotherapy can prevent early recurrence,it should be recommended for patients susceptible to early recurrence.

Prediction of early recurrence of hepatocellular carcinoma after liver transplantation based on computed tomography radiomics nomogram

Background:Early recurrence results in poor prognosis of patients with hepatocellular carcinoma(HCC)after liver transplantation(LT).This study aimed to explore the value of computed tomography(CT)-based radiomics nomogram in predicting early recurrence of patients with HCC after LT.Methods:A cohort of 151 patients with HCC who underwent LT between December 2013 and July 2019 were retrospectively enrolled.A total of 1218 features were extracted from enhanced CT images.The least absolute shrinkage and selection operator algorithm(LASSO)logistic regression was used for dimension reduction and radiomics signature building.The clinical model was constructed after the analysis of clin-ical factors,and the nomogram was constructed by introducing the radiomics signature into the clinical model.The predictive performance and clinical usefulness of the three models were evaluated using re-ceiver operating characteristic(ROC)curve analysis and decision curve analysis(DCA),respectively.Cali-bration curves were plotted to assess the calibration of the nomogram.Results:There were significant differences in radiomics signature among early recurrence patients and non-early recurrence patients in the training cohort(P<0.001)and validation cohort(P<0.001).The nomogram showed the best predictive performance,with the largest area under the ROC curve in the training(0.882)and validation(0.917)cohorts.Hosmer-Lemeshow testing confirmed that the nomogram showed good calibration in the training(P=0.138)and validation(P=0.396)cohorts.DCA showed if the threshold probability is within 0.06-1,the nomogram had better clinical usefulness than the clinical model.Conclusions:Our CT-based radiomics nomogram can preoperatively predict the risk of early recurrence in patients with HCC after LT.

α-fetoprotein,vascular endothelial growth factor receptor-1 and early recurrence of hepatoma

AIM:To investigate whether α-fetoprotein (AFP) and vascular endothelial growth factor receptor (VEGFR)-1 correlate with early recurrence of hepatoma/hepatocel-lular carcinoma (HCC).METHODS:From 2000 to 2005,114 consecutive pa-tients with HCC underwent primary curative hepatecto-my.The mean age was 60.7 (8.7) years and 94 patients were male.The median follow-up period was 71.2 mo (range:43-100 mo).Immediately prior to commencing laparotomy,5 mL bone marrow was aspirated from thesternum and collected in citrate-coated test tubes.The initial 2 mL of bone marrow aspirate was discarded in each case.AFP mRNA and VEGFR-1 mRNA in the bone marrow and peripheral blood (BM-and PH-AFP mRNA and BM-and PH-VEGFR-1 mRNA,respectively) were measured by real-time quantitative reverse transcription polymerase chain reaction.As normal controls,VEGFR-1 mRNA in the bone marrow and peripheral blood was also measured in 11 living liver donors.These data were evaluated for any correlation with early recurrence,comparing clinical and pathological outcomes.RESULTS:The cut-off value of the BM-AFP mRNA and PH-AFP mRNA level in patients with HCC was set at 1.92 × 10-7 and zero,respectively,based on data from the controls.A total of 34 (29.8%) and six (5.4%) patients were positive for BM-AFP mRNA and PH-AFP mRNA,respectively.The BM-VEGFR-1 mRNA levels in all HCC patients were higher than those in the normal con-trols,and this was the case also for PH-VEGFR-1mRNA.The 25-percentile values for the BM-and PH-VEGFR-1 mRNA in HCC patients were used as the cut-off values for assigning the patients into two groups based on these transcript levels.The High group for BM-VEG-FR-1 mRNA contained 81 (71.1%) HCC cases and the Low group was assigned 33 (28.9%) patients.These numbers for PH-VEGFR-1mRNA were 78 (75.0%) and 26 (25.0%),respectively.HCC recurred in 80 patients;in the remnant liver in 48 cases,in the remnant liver and remote tissue in 20,and in the remote tissue alone in 12.BM-AFP mRNA-positive cases showed a signifi-cantly higher rate of early recurrence (within 1 year of surgical treatment) compared with BM-AFP mRNA-negative patients (P=0.0091).Patients were classified into four groups according to the level/status of their BM-VEGFR-1 and BM-AFP mRNA as follows:group A (n=23),BM-VEGFR-1/BM-AFP mRNA=low/negative;group B (n=57) high/negative;group C (n=10) low/positive;group D (n=24),high/positive.This classifi-cation was found to correlate with a recurrence of thisdisease within 1 year (P=0.0228).The disease-free survival curve of group A was significantly better than that of groups B,C or D (P=0.0437,P=0.0325,P=0.0225).No other classification (i.e.,PH-VEGF-R1/BM-AFP,BM-VEGF-R1/PH-AFP,and PH-VEGF-R1/PH-AFP mRNA) showed such a correlation.CONCLUSION:The evaluation of BM-AFP and BM-VEG-FR-1 mRNA in patients with HCC may be a valuable pre-dictor of disease recurrence following curative resection.

Deep learning-based radiomics based on contrast-enhanced ultrasound predicts early recurrence and survival outcome in hepatocellular carcinoma

BACKGROUND Hepatocellular carcinoma(HCC)is the most common primary liver malignancy.AIM To predict early recurrence(ER)and overall survival(OS)in patients with HCC after radical resection using deep learning-based radiomics(DLR).METHODS A total of 414 consecutive patients with HCC who underwent surgical resection with available preoperative grayscale and contrast-enhanced ultrasound images were enrolled.The clinical,DLR,and clinical+DLR models were then designed to predict ER and OS.RESULTS The DLR model for predicting ER showed satisfactory clinical benefits[area under the curve(AUC=0.819 and 0.568 in the training and testing cohorts,respectively)],similar to the clinical model(AUC=0.580 and 0.520 in the training and testing cohorts,respectively;P>0.05).The C-index of the clinical+DLR model in the prediction of OS in the training and testing cohorts was 0.800 and 0.759,respectively.The clinical+DLR model and the DLR model outperformed the clinical model in the training and testing cohorts(P<0.001 for all).We divided patients into four categories by dichotomizing predicted ER and OS.For patients in class 1(high ER rate and low risk of OS),retreatment(microwave ablation)after recurrence was associated with improved survival(hazard ratio=7.895,P=0.005).CONCLUSION Compared to the clinical model,the clinical+DLR model significantly improves the accuracy of predicting OS in HCC patients after radical resection.

Preoperative risk factors for early recurrence in patients with resectable pancreatic ductal adenocarcinoma after curative intent surgical resection

Background: Postoperative early recurrence(ER) in patients with pancreatic ductal adenocarcinoma(PDAC) is frequently encountered after curative intent surgery. Nonetheless, clinical significance and risk factors of ER after surgery for PDAC have not been extensively investigated. The aim of this study was to determine preoperative risk predictors for ER in patients with PDAC after upfront surgery. Methods: Eighty-one consecutive patients with PDAC who underwent curative intent surgical resection at Kangbuk Samsung Hospital between January 2004 and May 2015 were enrolled. ER was defined as tumor relapse within 6 months after surgery. Results: ER occurred in 26 patients(32.1%), whereas 49 patients(60.5%) had late recurrence( ≥ 6 months after surgery), and 6 patients had no recurrence(7.4%). Univariate analysis showed that C-reactive protein(CRP) > 3.0 mg/dL, modified Glasgow prognostic score(mGPS) = 2, decrease of total lymphocyte count by > 50% of baseline value in the preoperative period, prognostic nutritional index(PNI) < 45, neutrophilto-lymphocyte ratio(NLR) ≥ 3, and preoperative maximum standardized uptake value(SUVmax) were significantly associated with ER. Multivariate logistic regression analysis revealed that CRP > 3.0 mg/dL, decrease of total lymphocyte count by > 50% of baseline value, and preoperative SUVmax were significant and independent contributors of ER in patients with resectable PDAC who underwent curative intent surgery. Conclusions: Postoperative ER for resectable PDAC was frequent with poor prognosis after curative intent upfront surgery. It is reasonable to suggest that there is a subgroup of resectable PDAC patients at highrisk of ER and neoadjuvant therapy should be considered in these patients in a clinical trial setting.

Early Recurrence of a Cardiac Malignant Fibrous Histiocytoma

Malignant fibrous histiocytoma (MFH) is a rare tumor of the heart and the patients with these tumors usually have a poor prognosis. We report a case of MFH with an origin from the left superior pulmonary vein, involving the left atrium and protruding through the mitral valve, which needed urgent surgery. Complete resection was performed but local recurrence was detected one month later.

Predictors of early and late hepatocellular carcinoma recurrence

Hepatocellular carcinoma(HCC)is the most frequent liver neoplasm,and its incidence rates are constantly increasing.Despite the availability of potentially curative treatments(liver transplantation,surgical resection,thermal ablation),long-term outcomes are affected by a high recurrence rate(up to 70%of cases 5 years after treatment).HCC recurrence within 2 years of treatment is defined as“early”and is generally caused by the occult intrahepatic spread of the primary neoplasm and related to the tumor burden.A recurrence that occurs after 2 years of treatment is defined as“late”and is related to de novo HCC,independent of the primary neoplasm.Early HCC recurrence has a significantly poorer prognosis and outcome than late recurrence.Different pathogenesis corresponds to different predictors of the risk of early or late recurrence.An adequate knowledge of predictive factors and recurrence risk stratification guides the therapeutic strategy and post-treatment surveillance.Patients at high risk of HCC recurrence should be referred to treatments with the lowest recurrence rate and when standardized to combined or adjuvant therapy regimens.This review aimed to expose the recurrence predictors and examine the differences between predictors of early and late recurrence.

Development of a machine learning-based model for predicting risk of early postoperative recurrence of hepatocellular carcinoma

BACKGROUND Surgical resection is the primary treatment for hepatocellular carcinoma(HCC).However,studies indicate that nearly 70%of patients experience HCC recurrence within five years following hepatectomy.The earlier the recurrence,the worse the prognosis.Current studies on postoperative recurrence primarily rely on postoperative pathology and patient clinical data,which are lagging.Hence,developing a new pre-operative prediction model for postoperative recurrence is crucial for guiding individualized treatment of HCC patients and enhancing their prognosis.AIM To identify key variables in pre-operative clinical and imaging data using machine learning algorithms to construct multiple risk prediction models for early postoperative recurrence of HCC.METHODS The demographic and clinical data of 371 HCC patients were collected for this retrospective study.These data were randomly divided into training and test sets at a ratio of 8:2.The training set was analyzed,and key feature variables with predictive value for early HCC recurrence were selected to construct six different machine learning prediction models.Each model was evaluated,and the bestperforming model was selected for interpreting the importance of each variable.Finally,an online calculator based on the model was generated for daily clinical practice.RESULTS Following machine learning analysis,eight key feature variables(age,intratumoral arteries,alpha-fetoprotein,preoperative blood glucose,number of tumors,glucose-to-lymphocyte ratio,liver cirrhosis,and pre-operative platelets)were selected to construct six different prediction models.The XGBoost model outperformed other models,with the area under the receiver operating characteristic curve in the training,validation,and test datasets being 0.993(95%confidence interval:0.982-1.000),0.734(0.601-0.867),and 0.706(0.585-0.827),respectively.Calibration curve and decision curve analysis indicated that the XGBoost model also had good predictive performance and clinical application value.CONCLUSION The XGBoost model exhibits superior performance and is a reliable tool for predicting early postoperative HCC recurrence.This model may guide surgical strategies and postoperative individualized medicine.

From prediction to prevention:Machine learning revolutionizes hepatocellular carcinoma recurrence monitoring

In this editorial,we comment on the article by Zhang et al entitled Development of a machine learning-based model for predicting the risk of early postoperative recurrence of hepatocellular carcinoma.Hepatocellular carcinoma(HCC),which is characterized by high incidence and mortality rates,remains a major global health challenge primarily due to the critical issue of postoperative recurrence.Early recurrence,defined as recurrence that occurs within 2 years posttreatment,is linked to the hidden spread of the primary tumor and significantly impacts patient survival.Traditional predictive factors,including both patient-and treatment-related factors,have limited predictive ability with respect to HCC recurrence.The integration of machine learning algorithms is fueled by the exponential growth of computational power and has revolutionized HCC research.The study by Zhang et al demonstrated the use of a groundbreaking preoperative prediction model for early postoperative HCC recurrence.Challenges persist,including sample size constraints,issues with handling data,and the need for further validation and interpretability.This study emphasizes the need for collaborative efforts,multicenter studies and comparative analyses to validate and refine the model.Overcoming these challenges and exploring innovative approaches,such as multi-omics integration,will enhance personalized oncology care.This study marks a significant stride toward precise,efficient,and personalized oncology practices,thus offering hope for improved patient outcomes in the field of HCC treatment.

Development of a machine learning model to predict early recurrence for hepatocellular carcinoma after curative resection

Background:Early recurrence is common for hepatocellular carcinoma(HCC)after surgical resection,being the leading cause of death.Traditionally,the COX proportional hazard(CPH)models based on linearity assumption have been used to predict early recurrence,but predictive performance is limited.Machine learning models offer a novel methodology and have several advantages over CPH models.Hence,the purpose of this study was to compare random survival forests(RSF)model with CPH models in prediction of early recurrence for HCC patients after curative resection.Methods:A total of 4,758 patients undergoing curative resection from two medical centers were included.Fifteen features including age,gender,etiology,platelet count,albumin,total bilirubin,AFP,tumor size,tumor number,microvascular invasion,macrovascular invasion,Edmondson-Steiner grade,tumor capsular,satellite nodules and liver cirrhosis were used to construct the RSF model in training cohort.Discrimination,calibration,clinical usefulness and overall performance were assessed and compared with other models.Results:Five hundred survival trees were used to generate the RFS model.The five highest Variable Importance(VIMP)were tumor size,macrovascular invasion,microvascular invasion,tumor number and AFP.In training,internal and external validation cohort,the C-index of RSF model were 0.725[standard errors(SE)=0.005],0.762(SE=0.011)and 0.747(SE=0.016),respectively;the Gönen&Heller’s K of RSF model were 0.684(SE=0.005),0.711(SE=0.008)and 0.697(SE=0.014),respectively;the time-dependent AUC(2 years)of RSF model were 0.818(SE=0.008),0.823(SE=0.014)and 0.785(SE=0.025),respectively.The RSF model outperformed early recurrence after surgery for liver tumor(ERASL)model,Korean model,American Joint Committee on Cancer tumor-node-metastasis(AJCC TNM)stage,Barcelona Clinic Liver Cancer(BCLC)stage and Chinese stage.The RSF model is capable of stratifying patients into three different risk groups(low-risk,intermediate-risk,high-risk groups)in the training and two validation cohorts(all P<0.0001).A web-based prediction tool was built to facilitate clinical application(https://recurrenceprediction.shinyapps.io/surgery_predict/).Conclusions:The RSF model is a reliable tool to predict early recurrence for patients with HCC after curative resection because it exhibited superior performance compared with other models.This novel model will be helpful to guide postoperative follow-up and adjuvant therapy.

A predictive model for early recurrence of colorectal-cancer liver metastases based on clinical parameters

Background:The prognosis for patients with colorectal-cancer liver metastases(CRLM)after curative surgery remains poor and shows great heterogeneity.Early recurrence,defined as tumor recurrence within 6 months of curative surgery,is associated with poor survival,requiring earlier detection and intervention.This study aimed to develop and validate a bedside model based on clinical parameters to predict early recurrence in CRLM patients and provide insight into post-operative surveillance strategies.Material and methods:A total of 202 consecutive CRLM patients undergoing curative surgeries between 2012 and 2019 were retrospectively enrolled and randomly assigned to the training(n=150)and validation(n=52)sets.Baseline information and radiological,pathological,and laboratory findings were extracted from medical records.Predictive factors for early recurrence were identified via a multivariate logistic-regression model to develop a predictive nomogram,which was validated for discrimination,calibration,and clinical application.Results:Liver-metastases number,lymph-node suspicion,neurovascular invasion,colon/rectum location,albumin and post-operative carcinoembryonic antigen,and carbohydrate antigen 19–9 levels(CA19–9)were independent predictive factors and were used to construct the nomogramfor early recurrence after curative surgery.The area under the curve was 0.866 and 0.792 for internal and external validation,respectively.The model significantly outperformed the clinical risk score and Beppu’s model in our data set.In the lift curve,the nomogram boosted the detection rate in post-operative surveillance by two-fold in the top 30%high-risk patients.Conclusion:Our model for early recurrence in CRLM patients after curative surgeries showed superior performance and could aid in the decision-making for selective follow-up strategies.

A simple scoring system to predict early recurrence of Bismuth–Corlette type IV perihilar cholangiocarcinoma

Background:Early recurrence has been reported to be predictive of a poor prognosis for patients with perihilar cholangiocarcinoma(pCCA)after resection.The objective of our study was to construct a useful scoring system to predict early recurrence for Bismuth–Corlette type IV pCCA patients in clinic and to investigate the value of early recurrence in directing post-operative surveillance and adjuvant therapy.Methods:In total,244 patients who underwent radical resection for type IV pCCA were included.Data on clinicopathological characteristics,perioperative details and survival outcomes were analyzed.Survival curves were generated using the Kaplan–Meier method.Univariate and multivariate logistic-regression models were used to identify factors associated with early recurrence.Results:Twenty-one months was defined as the cutoff point to distinguish between early and late recurrence.Univariate and multivariate analysis revealed that CA19-9 level>200 U/mL,R1 resection margin,higher N category and positive lymphovascular invasion were independent predictors of early recurrence.The scoring system was constructed accordingly.The early-recurrence rates of patients with scores of 0,1,2,3,4,and 5 were 23.9%,38.7%,60.0%,78.6%,83.4%,and 100%,respectively.Adjuvant therapy was significantly associated with higher overall survival rate for patients with early recurrence,but not for those with late recurrence.Patients in the early-recurrence group with scores2 had better prognoses after adjuvant therapy.Conclusions:A simple scoring system using CA19-9 level,N category,resection margin and lymphovascular invasion status could predict early recurrence,and thus might direct post-operative surveillance and adjuvant therapy for patients with type IV pCCA.

Metastatic bone cancer as a recurrence of early gastric cancer - characteristics and possible mechanisms

The surgical outcome of most early gastric cancer （EGC） is usually satisfactory. Some cases show bone metastasis even though the depth of cancer invasion is confined to the mucosa. The most frequent site for recurrence of EGC is the liver. Cases of EGC with bone metastasis are reviewed to clarify the clinicopathological characteristics of EGC giving rise to bone metastasis. Possible mechanisms and risk factors underlying this rare condition are proposed. Forty-six cases of bone metastasis from EGC are reviewed from published reports and meeting proceedings in Japan. This investigation suggests that risk factors for bone metastasis from EGC include depressed-type signet-ring cell carcinoma, poorly differentiated carcinoma, and/or the likely involvement of lymph node metastasis, even though the cancer is confined to the gastric mucosa. The risk factors do not include recurrence of EGC in the liver. We speculate that the mechanism of bone metastasis from EGC is via lymphatic channels and systemic circulation. Postoperative follow-up of cases should consider the development of bone metastasis from EGC. We propose the use of elevated alkaline phosphatase levels for the detection of bone metastasis and recommend bone scintigraphy in positive cases. 2005 The WJG Press and Elsevier Inc. All rights reserved