TBS-Based Preoperative Score to Predict Non-transplantable Recurrence and Identify Candidates for Upfront Resection Versus Transplantation for Hepatocellular Carcinoma

New predictors of microvascular invasion for small hepatocellular carcinoma ≤ 3 cm

BACKGROUND

Microvascular invasion (MVI) is a risk factor for postoperative recurrence of hepatocellular carcinoma (HCC), even in early-stage HCC. In small HCC ≤ 3 cm, treatment options include anatomical resection or non-anatomical resection, and MVI has a major effect on treatment decisions. We aimed to identify the predictors of MVI in small HCC ≤ 3 cm.

METHODS

We retrospectively studied 129 patients with very early or early-stage HCC ≤ 3 cm who had undergone 18F-fluorodeoxyglucose positron emission tomography/computed tomography and subsequent hepatic resection from January 2016 to August 2023. These patients were divided into the derivation cohort (n = 86) and validation cohort (n = 43). We examined the risk factors for MVI using logistic regression analysis, and established a predictive scoring system in the derivation cohort. We evaluated the accuracy of our scoring system in the validation cohort.

RESULTS

In the derivation cohort, a Lens culinaris agglutinin-reactive fraction of alpha-fetoprotein (AFP-L3), prothrombin induced by vitamin K deficiency or antagonist-II (PIVKA-II), and metabolic tumor volume (MTV) were independent predictors of MVI. We established the scoring system using these three factors. In the validation test, there were no MVI-positive cases with a score of 0 and 1, and all cases were MVI-positive with a score of 4. Moreover, with a score ≥ 2, the sensitivity, specificity, and accuracy of our scoring system were 100%, 71.4%, and 81.4%, respectively.

CONCLUSIONS

Our scoring system can accurately predict MVI in small HCC ≤ 3 cm, and could contribute to establishing an appropriate treatment strategy.

Salvage liver transplantation versus curative treatment for patients with recurrent hepatocellular carcinoma: A systematic review and meta-analysis

BACKGROUND

Salvage liver transplantation (SLT) is an effective treatment option for recurrent hepatocellular carcinoma (rHCC) following primary curative treatment (CUR). However, its efficacy remains controversial compared to that of CURs, including repeat liver resection (RLR) and local ablation. This meta-analysis compared the efficacy and safety of these procedures.

METHODS

A systematic literature search of the PubMed, Embase, Web of Science, and Cochrane Library databases for studies investigating SLT and CUR was performed. Outcome data, including overall and disease-free survival, tumor response, and operative and postoperative outcomes, were independently extracted and analyzed by two authors using a standardized protocol.

RESULTS

Fifteen cohort studies comprising 508 and 2050 patients with rHCC, who underwent SLT or CUR, respectively, were included. SLT achieved significantly longer overall survival than both CUR (hazard ratio [HR]: 0.56, 95 % confidence interval [CI]: 0.45-0.68; I2 = 34.6 %, p = 0.105) and RLR (HR: 0.64, 95 % CI: 0.49-0.84; I2 = 0.0 %, p = 0.639). Similar significantly better survival benefits were observed compared with CUR (HR: 0.30, 95 % CI: 0.20-0.45; I2 = 51.1 %, p = 0.038) or RLR (HR: 0.31, 95 % CI: 0.18-0.56; I2 = 65.7 %, p = 0.005) regarding disease-free survival. However, SLT resulted in a longer operative duration and hospital stay, larger amount of blood loss, higher rate of transfusion and postoperative morbidity, and slightly higher postoperative mortality than CUR.

CONCLUSION

SLT was associated with better long-term survival than CUR or RLR in patients with rHCC after primary curative treatment.

Impact of an artificial intelligence based model to predict non-transplantable recurrence among patients with hepatocellular carcinoma

OBJECTIVE

We sought to develop Artificial Intelligence (AI) based models to predict non-transplantable recurrence (NTR) of hepatocellular carcinoma (HCC) following hepatic resection (HR).

METHODS

HCC patients who underwent HR between 2000-2020 were identified from a multi-institutional database. NTR was defined as recurrence beyond Milan Criteria. Different machine learning (ML) and deep learning (DL) techniques were used to develop and validate two prediction models for NTR, one using only preoperative factors and a second using both preoperative and postoperative factors.

RESULTS

Overall, 1763 HCC patients were included. Among 877 patients with recurrence, 364 (41.5%) patients developed NTR. An ensemble AI model demonstrated the highest area under ROC curves (AUC) of 0.751 (95% CI: 0.719-0.782) and 0.717 (95% CI:0.653-0.782) in the training and testing cohorts, respectively which improved to 0.858 (95% CI: 0.835-0.884) and 0.764 (95% CI: 0.704-0.826), respectively after incorporation of postoperative pathologic factors. Radiologic tumor burden score and pathological microvascular invasion were the most important preoperative and postoperative factors, respectively to predict NTR. Patients predicted to develop NTR had overall 1- and 5-year survival of 75.6% and 28.2%, versus 93.4% and 55.9%, respectively, among patients predicted to not develop NTR (p < 0.0001).

CONCLUSION

The AI preoperative model may help inform decision of HR versus LT for HCC, while the combined AI model can frame individualized postoperative care (https://altaf-pawlik-hcc-ntr-calculator.streamlit.app/).

Construction of a prognostic model for hepatocellular carcinoma patients receiving transarterial chemoembolization treatment based on the Tumor Burden Score

BACKGROUND

Patients with hepatocellular carcinoma (HCC) who undergo transarterial chemoembolization (TACE) may have varied outcomes based on their liver function and tumor burden diversity. This study aims to assess the prognostic significance of the tumor burden score (TBS) in these patients and develop a prognostic model for their overall survival.

METHODS

The study involved a retrospective analysis of 644 newly diagnosed HCC patients undergoing TACE treatment. The individuals were assigned randomly to a training cohort (n = 452) and a validation cohort (n = 192). We utilized a multivariate Cox proportional risk model to identify independent preoperative predictive factors. We then evaluated model performance using the area under the curve (AUC), consistency index (c-index), calibration curve, and decision curve analysis (DCA) methods.

RESULTS

The multivariate analysis revealed four prognostic factors associated with overall survival: Tumor Burden Score, Tumor Extent, Types of portal vein invasion (PVI), and Child-Pugh score. The total score was calculated based on these factors. The model demonstrated strong discriminative ability with high AUC values and c-index, providing high net clinical benefits for patients. Based on the model's scoring results, patients were categorized into high, medium, and low-risk groups. These results were validated in the validation cohort.

CONCLUSIONS

The tumor burden score shows promise as a viable alternative prognostic indicator for assessing tumor burden in cases of HCC. The new prognostic model can place patients in one of three groups, which will estimate their individual outcomes. For high-risk patients, it is suggested to consider alternative treatment options or provide the best supportive care, as they may not benefit significantly from TACE treatment.

Optimal policy tree to assist in adjuvant therapy decision-making after resection of colorectal liver metastases

BACKGROUND

Although systemic postoperative therapy after surgery for colorectal liver metastases is generally recommended, the benefit of adjuvant chemotherapy has been debated. We used machine learning to develop a decision tree and define which patients may benefit from adjuvant chemotherapy after hepatectomy for colorectal liver metastases.

METHODS

Patients who underwent curative-intent resection for colorectal liver metastases between 2000 and 2020 were identified from an international multi-institutional database. An optimal policy tree analysis was used to determine the optimal assignment of the adjuvant chemotherapy to subgroups of patients for overall survival and recurrence-free survival.

RESULTS

Among 1,358 patients who underwent curative-intent resection of colorectal liver metastases, 1,032 (76.0%) received adjuvant chemotherapy. After a median follow-up of 28.7 months (interquartile range 13.7-52.0), 5-year overall survival was 67.5%, and 3-year recurrence-free survival was 52.6%, respectively. Adjuvant chemotherapy was associated with better recurrence-free survival (3-year recurrence-free survival: adjuvant chemotherapy, 54.4% vs no adjuvant chemotherapy, 46.8%; P < .001) but no overall survival significant improvement (5-year overall survival: adjuvant chemotherapy, 68.1% vs no adjuvant chemotherapy, 65.7%; P = .15). Patients were randomly allocated into 2 cohorts (training data set, n = 679, testing data set, n = 679). The random forest model demonstrated good performance in predicting counterfactual probabilities of death and recurrence relative to receipt of adjuvant chemotherapy. According to the optimal policy tree, patient demographics, secondary tumor characteristics, and primary tumor characteristics defined the subpopulation that would benefit from adjuvant chemotherapy.

CONCLUSION

A novel artificial intelligence methodology based on patient, primary tumor, and treatment characteristics may help clinicians tailor adjuvant chemotherapy recommendations after colorectal liver metastases resection.

Prediction of preoperative microvascular invasion by dynamic radiomic analysis based on contrast-enhanced computed tomography

PURPOSE

Microvascular invasion (MVI) is a common complication of hepatocellular carcinoma (HCC) surgery, which is an important predictor of reduced surgical prognosis. This study aimed to develop a fully automated diagnostic model to predict pre-surgical MVI based on four-phase dynamic CT images.

METHODS

A total of 140 patients with HCC from two centers were retrospectively included (training set, n = 98; testing set, n = 42). All CT phases were aligned to the portal venous phase, and were then used to train a deep-learning model for liver tumor segmentation. Radiomics features were extracted from the tumor areas of original CT phases and pairwise subtraction images, as well as peritumoral features. Lastly, linear discriminant analysis (LDA) models were trained based on clinical features, radiomics features, and hybrid features, respectively. Models were evaluated by area under curve (AUC), accuracy, sensitivity, specificity, positive and negative predictive values (PPV and NPV).

RESULTS

Overall, 86 and 54 patients with MVI- (age, 55.92 ± 9.62 years; 68 men) and MVI+ (age, 53.59 ± 11.47 years; 43 men) were included. Average dice coefficients of liver tumor segmentation were 0.89 and 0.82 in training and testing sets, respectively. The model based on radiomics (AUC = 0.865, 95% CI: 0.725-0.951) showed slightly better performance than that based on clinical features (AUC = 0.841, 95% CI: 0.696-0.936). The classification model based on hybrid features achieved better performance in both training (AUC = 0.955, 95% CI: 0.893-0.987) and testing sets (AUC = 0.913, 95% CI: 0.785-0.978), compared with models based on clinical and radiomics features (p-value < 0.05). Moreover, the hybrid model also provided the best accuracy (0.857), sensitivity (0.875), and NPV (0.917).

CONCLUSION

The classification model based on multimodal intra- and peri-tumoral radiomics features can well predict HCC patients with MVI.

PIVKA-II combined with tumor burden score to predict long-term outcomes of AFP-negative hepatocellular carcinoma patients after liver resection

BACKGROUND

This study aimed to establish a simple prognostic scoring model based on tumor burden score (TBS) and PIVKA-II to predict long-term outcomes of α-fetoprotein (AFP)-negative hepatocellular carcinoma (HCC) patients.

METHODS

511 patients were divided into the training cohort (n = 305) and the validation cohort (n = 206) at a ratio of 6:4. Receiver operating characteristic curves (ROC) were established to identify cutoff values of TBS and PIVKA-II. Kaplan-Meier curves were used to analyze survival outcomes. The multivariable Cox regression was used to identify variables independently associated with survival outcomes. The predictive performance of the TBS-PIVKA II score (TPS) model was compared with Barcelona clinic liver cancer (BCLC) stage and American Joint Committee on Cancer (AJCC TNM) stage.

RESULTS

The present study established the TPS model using a simple scoring system (0, 1 for low/high TBS [cutoff value: 4.1]; 0, 1 for low/high PIVKA-II [cutoff value: 239 mAU/mL]). The TPS scoring model was divided into three levels according to the summation of TBS score and PIVKA-II score: TPS 0, TPS 1, and TPS 2. The TPS scoring model was able to stratify OS (training: p < 0.001, validation: p < 0.001) and early recurrence (training: p < 0.001; validation: p = 0.001) in the training cohort and the validation cohort. The TPS score was independently associated with OS (TPS 1 vs. 0, HR: 2.28, 95% CI: 1.01-5.17; TPS 2 vs. 0, HR: 4.21, 95% CI: 2.01-8.84) and early recurrence (TPS 1 vs. 0, HR: 3.50, 95% CI: 1.71-7.16; TPS 2 vs. 0, HR: 3.79, 95% CI: 1.86-7.75) in the training cohort. The TPS scoring model outperformed BCLC stage and AJCC TNM stage in predicting OS and early recurrence in the training cohort and the validation cohort. But the TPS scoring model was unable to stratify the late recurrence in the training cohort (p = 0.872) and the validation cohort (p = 0.458).

CONCLUSIONS

The TPS model outperformed the BCLC stage and AJCC TNM stage in predicting OS and early recurrence of AFP-negative HCC patients after liver resection, which might better assist surgeons in screening AFP-negative HCC patients who may benefit from liver resection.

Impact of Surgical Margin Width on Prognosis Following Resection of Hepatocellular Carcinoma Varies on the Basis of Preoperative Alpha-Feto Protein and Tumor Burden Score

BACKGROUND

We sought to examine the prognostic impact of margin width at time of hepatocellular carcinoma (HCC) resection relative to the alpha-feto protein tumor burden score (ATS).

PATIENTS AND METHODS

Patients who underwent curative-intent hepatectomy for HCC between 2000 and 2020 were identified from a multi-institutional database. The impact of margin width on overall survival and recurrence-free survival was examined relative to ATS using univariable and multivariable analyses.

RESULTS

Among 782 patients with HCC who underwent resection, median ATS was 6.5 [interquartile range (IQR) 4.3-10.2]. Most patients underwent R0 resection (n = 613, 78.4%); among patients who had an R0 resection, 325 (41.6%) had a margin width > 5 mm while 288 (36.8%) had a 0-5 mm margin width. Among patients with high ATS, an increasing margin width was associated with incrementally better overall and recurrence-free survival. In contrast, among patients with low ATS, margin width was not associated with long-term outcomes. On multivariable Cox regression analysis, each unit increase in ATS was independently associated with a 7% higher risk of death [hazard ratio (HR) 1.07; 95% confidence interval (CI) 1.03-1.11, p < 0.001]. While the incidence of early recurrence was not associated with margin width among patients with low ATS, wider margin width was associated with an incrementally lower incidence of early recurrence among patients with high ATS.

CONCLUSION

ATS, an easy-to-use composite tumor-related metric, was able to risk stratify patients following resection of HCC relative to overall survival and recurrence-free survival. The therapeutic impact of resection margin width had a variable impact on long-term outcomes relative to ATS.