Development and validation of a risk prediction algorithm for evaluating the efficacy of postoperative adjuvant TACE therapy for hepatocellular carcinoma: a retrospective study

doi:10.21203/rs.3.rs-1641318/v1

Download PDF

Research Article

Development and validation of a risk prediction algorithm for evaluating the efficacy of postoperative adjuvant TACE therapy for hepatocellular carcinoma: a retrospective study

https://doi.org/10.21203/rs.3.rs-1641318/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background and purpose

There is a lack of a reliable outcome prediction model for patients evaluating the feasibility of postoperative adjuvant transarterial chemoembolization (PA-TACE) therapy. Our goal was to develop an easy-to-use tool specifically for these patients.

Methods

From January 2013 to June 2017, patients with liver cancer from the Liver Center of the First Affiliated Hospital of Chongqing Medical University received postoperative adjuvant Transarterial chemoembolization (TACE) therapy after liver cancer resection. A Cox’s proportional hazards model was established for these patients, followed by internal validation (enhanced bootstrap resampling technique) to further evaluate the predictive performance and discriminance, and compare with other predictive models. The prognostic factors considered included tumour number, maximum tumor diameter, Edmondson-Steiner (ES) grade, Microvascular invasion (MVI) grade, Ki67, age, sex, hepatitis B surface antigen, cirrhosis, Alpha-fetoprotein(AFP), Albumin-bilirubin (ALBI) grade, Child-pugh grade, body mass inex (BMI), Neutrophil-lymphocyte ratio (NLR), Platelet-to-lymphocyte ratio (PLR).

Results

The end point of the study was overall survival. The median overall survival was 31.8 (95%CI: 30.0-33.6 ) months, with 1-year, 2-year and 3-year survival rates being 96.3%, 84.0% and 75.3%, respectively. Tumour number, MVI grade, and BMI was incorporated into the model, which had good differentiation and accuracy. Internal validation (enhanced bootstrap ) suggested that Harrell’s C statistic is 0.72 (a measure of discrimination, with higher values indicating better discrimination). The model consistently outperforms other currently available models.

Conclusions

This model may be an easy-to-use tool for screening patients suitable for PA-TACE treatment and guiding the selection of clinical protocols. But further research and external validation are required.

development and validation of a risk prediction algorithm

hepatocellular carcinoma

postoperative adjuvant TACE therapy

retrospective study

Liver cancer is currently the sixth most common cancer and the thirdly deadliest cancer¹. Surgery is the major part of its treatment, but the 5-year postoperative recurrence rate after surgery reaches 50%-70%², while the 5-year postoperative survival rate is comparatively low (25%-55%)³. Transarterial chemoembolization(TACE) is one of the main treatment methods for patients with advanced hepatocellular carcinoma(HCC). Apart from this, TACE has also been widely recommended to be used on patients before and after radical hepatectomy⁴. Many studies^5–7 have shown that PA-TACE is able to increase the RFS and OS of patients at high risk of recurrence, and significantly improve the prognosis of HCC patients. Nevertheless, there are too many indicators used to evaluate the benefit of PA-TACE in patients with HCC after surgery, and indications of PA-TACE vary from place to place. As a result, it is necessary to establish a model to screen for high-risk groupsand improve the survival rate through individualized treatment⁸. More importantly, this model should be able to estimate the probability of survival in patients experiencing postoperative TACE therapy and make comparison between the results produced by TACE and those by other treatment options, which helps better clinical decision making.

Relevent studies have found that factors affecting prognosis of radical hepatectomy include microvascular invasion^{5, 9}, tumor size, tumor number⁸, cirrhosis¹⁰, etc. At present, there is no existing prediction model for patients receiving adjuvant TACE therapy after hepatocellular carcinoma resection. Therefore, in this study, by screening for relevant prognostic factors and establishing a prognostic model, patients suitable for PA-TACE treatment were identified and so guidance can be provided to the selection of clinical protocols.

Patients and Methods

Patients

In this retrospective study, 180 patients with liver cancer underwent surgical resection in the Liver Center of the First Affiliated Hospital of Chongqing Medical University from January, 2013 to June, 2017. HCC was diagnosed by imaging or histological examination in accordance with AASLD or EASL guidelines, and patients with HCC who underwent surgical treatment were included.

Inclusion criteria

(1) A preoperative child-pugh score of A or B, with liver reserve function as well as residual liver volume that meet surgical requirements; (2) No tumor rupture or various primary/secondary infections; (3) The resection margin confirmed free of cancer cells by postoperative pathological examination; (4) Hepatocellular carcinoma confirmed by Postoperative pathological examination and immunohistochemistry ; (5) No invasion of large vessels or distant metastasis.

Exclusion criteria

(1)Tumor recurred, during TACE treatment, within 2 months after the operation; (2) Prior to the first diagnosis of tumor recurrence, patients received adjuvant therapy other than TACE; (3) Lost to follow-up within 1 year after surgery; (4)Patients was diagnosed with extraheptic diseases that cause abnormal numbers of blood cells such as blood or immune system diseases. Clinical data(age, sex, hepatitis B, cirrhosis, preoperative blood routine, liver function, and AFP, etc.) and tumor data (tumor size, number of tumors, degree of differentiation, MVI grade, etc.) of enrolled patients were collected, and the tumor data were issued by the Department of Clinicopathology, Molecular Medical Testing Center, Chongqing Medical University.

Treatment Procedures

Patients who met inclusion criteria were given segmental hepatectomy, hemihepatectomy or tumor resection according to the size, location and distribution of tumors. One month after the surgery, TACE therapy was applied to subjects with an MVI grade higher than or equal to M1 or with moderately or poorly differentiated carcinoma, as per their tolerance¹¹. Seldinger technique was used to puncture and catheterize femoral artery. Then transcatheter arterial chemoembolization was performed with lipiodil and chemotherapy drugs (oxaliplatin, irinotecan, epirubicin or pirubicin) after the assessment of dose based on body surface area and liver function. Before treatment, all patients signed the informed consent of relevant treatment and voluntarily chose TACE treatment.

Patients

Inclusion criteria: (1) A preoperative child-pugh score of A or B, with liver reserve function as well as residual liver volume that meet surgical requirements; (2) No tumor rupture or various primary/secondary infections; (3) The resection marginconfirmed free of cancer cells by postoperative pathological examination; (4) Hepatocellular carcinoma confirmed by Postoperative pathological examination and immunohistochemistry ; (5) No invasion of large vessels or distant metastasis.

Exclusion criteria: (1)Tumor recurred, during TACE treatment, within 2 months after the operation; (2) Prior to the first diagnosis of tumor recurrence, patients received adjuvant therapy other than TACE; (3) Lost to follow-up within 1 year after surgery; (4)Patients was diagnosed with extraheptic diseases that cause abnormal numbers of blood cells such as blood or immune system diseases. Clinical data(age, sex, hepatitis B, cirrhosis, preoperative blood routine, liver function, and AFP, etc.) and tumor data (tumor size, number of tumors, degree of differentiation, MVI grade, etc.) of enrolled patients were collected, and the tumor data were issued by the Department of Clinicopathology, Molecular Medical Testing Center, Chongqing Medical University.

Treatment Procedures

Follow-up and outcome indicators

The outpatient follow-up assessment included liver and kidney function tests, quantitative HBV-DNA, tumor markers, abdominal color Doppler ultrasound or enhanced computed tomography of the abdomen and chest. And the follow-up was set to end at June 20, 2020, with a median of 47 months (95%CI: 43.2–50.7). Overall Survival (OS) was defifined as the time interval between initial TACE and all-cause death.

Predictors and sample size

Relevant studies have shown that factors including tumor number, maximum diameter of tumors, Edmondson-Steiner grade⁸, MVI grade¹², Ki67¹³, age, gender, hepatitis B surface antigen⁵, cirrhosis¹⁰, alpha-fetoprotein, albumin-bilirubin grade¹⁴, Child-pugh grade, BMI¹⁵, NLR¹⁶, PLR¹⁷ all have an effect on the prognosis of adjuvant therapy after hepatectomy, we related these factors as predictors of prognosis in this study.

The number and maximum diameter of tumors were measured using enhanced CT or MRI before surgical resection. Edmondson-steiner grade, MVI grade, Ki67 were evaluated by the Clinicopathology Department of Molecular Medical Testing Center of Chongqing Medical University, according to corresponding criteria ^{18, 19}after surgical resection. Edmondson-steiner grade (≥3 or < 3), MVI grade (M0 or > M0), and Ki67 (≥20% or < 20%) along with the number of tumor (≥2 or < 2), and the maximum tumor diameter (≥5cm or< 5cm) were all considered as categorical variables.

Preoperative data inclueded age, sex, hepatitis B surface antigen, cirrhosis, alpha-fetoprotein, albumin-bilirubin grade, Child-pugh grade, BMI, NLR, PLR. Among them, age (≥55 years or < 55 years), hepatitis B surface antigen (Measured by blood drawing), cirrhosis(Diagnosis was made according to European guidelines for cirrhosis²⁰), alpha-fetoprotein (≥200mg/L or < 200mg/L), albumin-bilirubin grade (grade 1 or grade 2) and Child-pugh grade (A or B)" were categorical variables. And BMI, NLR and PLR were continuous variables.

Three predictive factors ((MVI (P=0.00643), tumor number (P=0.03056) and BMI (P=0.08092)) were further screened by least absolute shrinkage and selection sperator(Lasso) analysis for model development. Only complete data were analyzed in this study. The minimum sample size determined by EPV(Event per Variable) criterion²¹ is 30, so this thesis meets statistical requirements of sampling and has sufficient accuracy.

Model Derivation and Validation

NLR normally distributed after logarithmic transformation, and the rest continuous variables were basically of normal distribution as well. The converted values of non-normal distribution continuous variables were recorded. Continuous variables identified as predictors of survival were analyzed using spline functions to assess their correlation with outcome variables and test their nonlinearity.

Cox proportional risk model was used to screen variables, establish the model, and calculate the predicted values. Area under the curve(AUC) and Brier Score were calculated to evaluate the model. Then model deviration was performed, including linear relationship, strong influence point, multicollinearity, and proportional risk hypothesis. Harrell's C statistics and Brier score were obtained by enhanced bootstrap for internal validation to evaluate the differentiation and calibration of the model. In this study, Harrell's C statistic²² was equival to the overall C statistic of the model. On account of the lack of relevant prognostic models, the final model was compared with both ALBI and child-pugh grade by C statistic and likelihood ratio test. Corresponding P value < 0.05 indicates that the difference is statistically significant. All analyses are performed using R version 4.1.2. All authors have full access to the study data and have reviewed and approved this manuscript.

Ethics

This retrospective study, in compliance with the 1975 version of the Declaration of Helsinki, has been approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University, the approval number being 2019021.

Baseline characteristics

This retrospective study included 180 HCC patients who underwent surgical treatment. Among them, 86 patients met the inclusion criteria, while five patients withdrew from follow-up. In total, 81 patients were included. (Fig 1). The characteristics and related predictive factors of the research object are shown in Table 1.

Overall survival

The median follow-up time was 47 months(95%CI: 43.2-50.7 months). The median survival time was 31.8 months (95%CI: 30.0-33.6 months), and the 1-year, 2-year and 3-year survival rates were 96.3% 84.0% and 75.3%, respectively.

Table 1. Baseline characteristics and predictive factors

Baseline characteristics		Number
Gender
	male	70
	female	11
Age(years)	≥55	32
	＜55	49
BMI		22.79（16.14-29.14）
Maximum tumor diameter(cm)	≥5	42
	＜5	39
Tumour number	≥2	18
	1	63
MVI grade	＞M0	69
	M0	12
ES grade
	≥3	23
	＜3	58
Ki67	≥20%	38
	＜20%	43
Hepatitis B surface antigen	（+）	73
	（-）	8
Cirrhosis	yes	45
	no	36
Alpha-fetoprotein(mg/L)	≥200	33
	＜200	48
child-pugh grade	A	79
	B	2
Albumin-bilirubin grade	1	49
	2	32
NLR		3.1（0.65-10）
PLR		137.03（43.94-315.15）

BMI: body mass inex, MVI grade: Microvascular invasion grade, ES grade: Edmondson-Steiner(ES) grade, NLR: Neutrophil-lymphocyte ratio, PLR: Platelet-to-lymphocyte ratio.

Development of the prognostic model

The tumor number, maximum tumor diameter, Edmondson-Steiner grade, MVI grade, Ki67, age, gender, hepatitis B surface antigen, cirrhosis, alpha-fetoprotein, albumin-bilirubin grade, and Child-pugh grade were treated as classification variables, while BMI, NLR, and PLR were encoded as continuous variables. Restricted cubic spline functions showed that the three consecutive prognostic factors were linearly distributed and could be modeled as continuous linear variables. (Fig 2) Through Lasso regression analysis, three variables (tumor number, MVI grade, BMI) were screened out by us. The following model could be obtained after re-fitting and adjustment:

S(t)(Predict the probability of survival at a particular point in time)=S0(t)^exp(-1.40970*MVI+1.07147*Number-0.14921*BMI)

S0(24)=0.7096147

S0(28)=0.6629676

S0(32)=0.5946105

(AFP and MVI grade are binary variables with a value of 0 standing for AFP ≤200 ng/ml, and 0 for M0, respectively; 1 standing for AFP >200 ng/ml, 1 for > M0, respectively. BMI is a continuous variable). Table 2 presents the AUC value and Brier score of the model over time.

Table 2. the time-varying AUC (Area under the curve) and Brier score of the model

Time(months)	Area under the curve(95% CI)	Brier score(95% CI)
24	0.717(0.556-0.879)	0.121(0.070-0.172)
28	0.752(0.607-0.898)	0.131(0.081-0.180)
32	0.764(0.634-0.893)	0.148(0.100-0.196)

Model Diagnosis and Validation

Use the above model for model diagnosis, and it is found that BMI is still linearly related to outcome variables in the multivariate case(Fig 3). No strong influence point is observed in this analysis, which refers to the test point that has great influence on statistica resuslts. And no obvious multicollinearity is observed between variables, which means that there is no linear or nearly linear relationship between them. The proportional risk hypothesis suggests that the effect of variables on outcome does not change over time. Internal validation is performed by enhanced bootstrap method. According to the results(Table 3), thia model demonstrates good discrimination and validity.

Table 3. the result of enhanced bootstrap

Time(months)	Harrell's C statistics	C statistics	Brier score
24	0.719372	0.7205797	0.1202442
28	0.7214915	0.750996	0.1301134
32	0.7173631	0.7633653	0.1497409

Model comparison and nomogram

Due to the lack of relevant models, this model was compared with both albumin-bilirubin grade (C statistic test: P=0.08460646, likelihood ratio test: P=0.006066) and the child-pugh grade (C statistic test: P=0.001726161, likelihood ratio test: P=0.005525). The result suggests the model has good predictive and discriminative ability, and can be used to screen the target population. Based on the trained model, a nomogram was established, which was able to predict the 2-year survival probability of the target patients (Fig 4).

The presence of invisible intrahepatic metastasis before surgery is one of the main reasons for postoperative short-term recurrence of liver cancer (within 2 years after surgery) ²³. PA-TACE prevents possible micrometastasis by destroying potentially survived tumor with local chemotherapeutic drugs and selective vascular embolization, and so possible recurrence²⁴ is averted. On the other hand, the hypoxic environment after interventional embolization can trigger the expression of hypoxia-inducible factor 1α(HIF-1α) and vascular endothelial growth factor (VEGF), forming a tumor-friendly microenvironment that may promote tumor growth, metastasis, and recurrence²⁵. Therefore, it is necessary to establish a model to screen for high-risk groups. Based on 81 postoperative patients receiving TACE, a relevant model was established to predict the probability of survival through a specified period of time, thereby evaluating the prognosis of patients and so providing guidance for clinical decision making.

The median survival time of the subjects is 31.8 (95%CI: 30.0-33.6) months, and the 1-year, 2-year and 3-year survival rates are 96.3%, 84.0% and 75.3%, respectively. Compared with relevant studies^{26, 27}, the overall survival period of the subjects in this study is shorter, which may have resulted from insufficient follow-up time and the absence of other adjuvant therapies. In addition, most patients did not receive treatment after recurrence.

The novelties and advantages of this study lie in following facts: 1. This is the first model built for postoperative patients receiving TACE adjuvant therapy, with fair precision in prediction. 2. After most relevant factors in relation to prognosis were extracted from multiple predictors from various aspects and their relevance were examined, a concise and convenient model was ultimately established for clinical use. 3.The study was conducted and reported through TRIPO statements²¹, with minimized bias in the process. And patient information was comprehensively collected. The study provides individualized survival estimates.

The characteristics and heterogeneity of the target population determine the parameters and performance of the final model, and are of the highest importance to the model²⁸. This is the first study that focuses on postoperative patients receiving TACE adjuvant therapy, which is more oriented, so the prediction is closer to real situations.. This study involves miscellaneous factors, including the general condition and liver and kidney function of the patient, the general and pathological parameters of the tumor, immunohistochemistry, etc. MVI, the tumor number, and BMI were screened by Lasso analysis. Among them, MVI and the number of tumors may play a major role in prediction. MVI, also known as microhemangioma thrombus, mainly refers to the nests of cancer cells under the microscope in the vascular lumen lined with endothelial cells²⁹. Polynodular lesions may be the result of direct tissue invasion by malignant cells, or represent microvascular emboli that subsequently invade new areas of the liver. Both MVI and polynodular lesions are associated with tumor invasiveness (loss of intercellular adhesion, disruption of the surrounding extracellular matrix (ECM), and cell motility, etc.).²⁹

Therefore, the MVI grade and tumor number are important factors in the recurrence of HCC^{8, 30}, which also profoundly affect the treatment of the disease. Besides, BMI, which is used to roughly measure visceral fat density, is also incorporated into this model. A high visceral adipose tissue density may reflect poor nutritional status³¹ as well as chronic inflammation in patients with HCC and is also associated with poor prognosis of the disease after TACE¹⁵. Since there are few relevant models, our study only drew comparisons between this model and those based on albumin-bilirubin grading and child-pugh grading, suggesting good performance and precision, but further verification is required.

Our study still has limitations. First, the number of cases is relatively limited. Multi-centered large-sample studies need to be conducted. Secondly, the etiology, composition, and characteristics of HCC patients vary from place to place, and the unscreened prognostic factors mentioned in this study may be of vital importance. Thus, a large-scale external validation is necessary. In addition, selection bias is unavoidable in observation studies, even though we have included all patients on continuous postoperative TACE adjuvant therapy to minimize the associated risk. Given the above limitations, further research and validation under other conditions are needed for our model to be more effective as an important predictive tool.

This is the first prognostic model established for patients receiving TACE adjuvant therapy after surgical resection, which is convenient for clinical use. It provides sufficient predictive factors for individualized prediction and may be useful for evaluating outcomes in both clinical decision making and practice. But further validation and large sample studies are still necessary.

Author contributions

Jie Tao, Xiaoli Shi, and Zhengrong Shi contributed to the research concept and design. Jie Tao, Xiaoli Shi, Xu Feng, Xinhua Wu, Shiguai Qi, Guoying Feng, Xu Yang, Yufei Zhao participated in literature retrieval, data collection, and data analysis. Jie Tao, Xiaoli Shi, Hangjia Zuo and Zhengrong Shi contributed to the drafting and review of the final manuscript. All authors read and approved the final manuscript.

Conflict of Interest

The authors have no conflict of interest.

Funding

This work was supported by Young and Middle aged High-end Talents Program of Medical Research Program of Chongqing Health and Family Planning Commission (No. 2019GDRC002), and Chongqing Science and Technology Bureau (cstc2019jcyj-msxmX0837).

Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021; 71: 209–49.
Vogel A, Cervantes A, Chau I, et al. Hepatocellular carcinoma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. ANN ONCOL. 2018; 29: v238-55.
Couri T, Pillai A. Goals and targets for personalized therapy for HCC. HEPATOL INT. 2019; 13: 125–37.
Zhang Z, Liu Q, He J, Yang J, Yang G, Wu M. The effect of preoperative transcatheter hepatic arterial chemoembolization on disease-free survival after hepatectomy for hepatocellular carcinoma. CANCER-AM CANCER SOC. 2000; 89: 2606–12.
Chen MY, Juengpanich S, Hu JH, et al. Prognostic factors and predictors of postoperative adjuvant transcatheter arterial chemoembolization benefit in patients with resected hepatocellular carcinoma. World J Gastroenterol. 2020; 26: 1042–55.
Feng AL, Zhu JK, Yang Y, et al. Repeated postoperative adjuvant TACE after curative hepatectomy improves outcomes of patients with HCC. Minim Invasive Ther Allied Technol. 2021; 30: 163–8.
Wang Z, Ren Z, Chen Y, et al. Adjuvant Transarterial Chemoembolization for HBV-Related Hepatocellular Carcinoma After Resection: A Randomized Controlled Study. CLIN CANCER RES. 2018; 24: 2074–81.
EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J HEPATOL. 2018; 69: 182–236.
Fan ST. Hepatocellular carcinoma–resection or transplant? Nat Rev Gastroenterol Hepatol. 2012; 9: 732–7.
Marasco G, Colecchia A, Colli A, et al. Role of liver and spleen stiffness in predicting the recurrence of hepatocellular carcinoma after resection. J HEPATOL. 2019; 70: 440–8.
EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. J HEPATOL. 2012; 56: 908–43.
Ye JZ, Chen JZ, Li ZH, et al. Efficacy of postoperative adjuvant transcatheter arterial chemoembolization in hepatocellular carcinoma patients with microvascular invasion. World J Gastroenterol. 2017; 23: 7415–24.
Zhao YF, Xiong X, Chen K, Tang W, Yang X, Shi ZR. Evaluation of the Therapeutic Effect of Adjuvant Transcatheter Arterial Chemoembolization Based on Ki67 After Hepatocellular Carcinoma Surgery. FRONT ONCOL. 2021; 11: 605234.
Pinato DJ, Sharma R, Allara E, et al. The ALBI grade provides objective hepatic reserve estimation across each BCLC stage of hepatocellular carcinoma. J HEPATOL. 2017; 66: 338–46.
Li Q, Zhang L, Hou ZH, et al. High Visceral Adipose Tissue Density Correlates With Unfavorable Outcomes in Patients With Intermediate-Stage Hepatocellular Carcinoma Undergoing Transarterial Chemoembolization. Front Cell Dev Biol. 2021; 9: 710104.
Hong YM, Cho M, Yoon KT, et al. Neutrophil-lymphocyte ratio predicts the therapeutic benefit of neoadjuvant transarterial chemoembolization in patients with resectable hepatocellular carcinoma. Eur J Gastroenterol Hepatol. 2020; 32: 1186–91.
Chen L, Ke Z, Xiong F, et al. Platelet-to-lymphocyte ratio predicts therapy outcomes of transarterial chemoembolization plus apatinib in the treatment of advanced hepatocellular carcinoma. Anticancer Drugs. 2020; 31: 966–72.
Zhou L, Rui JA, Ye DX, Wang SB, Chen SG, Qu Q. Edmondson-Steiner grading increases the predictive efficiency of TNM staging for long-term survival of patients with hepatocellular carcinoma after curative resection. WORLD J SURG. 2008; 32: 1748–56.
Zhang X, Li J, Shen F, Lau WY. Significance of presence of microvascular invasion in specimens obtained after surgical treatment of hepatocellular carcinoma. J Gastroenterol Hepatol. 2018; 33: 347–54.
EASL Clinical Practice Guidelines for the management of patients with decompensated cirrhosis. J HEPATOL. 2018; 69: 406–60.
Moons KG, Altman DG, Reitsma JB, et al. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration. ANN INTERN MED. 2015; 162: W1-73.
Harrell FJ, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. STAT MED. 1996; 15: 361–87.
Choi J, Jo C, Lim YS. Tenofovir Versus Entecavir on Recurrence of Hepatitis B Virus-Related Hepatocellular Carcinoma After Surgical Resection. HEPATOLOGY. 2021; 73: 661–73.
Yang J, Liang H, Hu K, et al. The effects of several postoperative adjuvant therapies for hepatocellular carcinoma patients with microvascular invasion after curative resection: a systematic review and meta-analysis. CANCER CELL INT. 2021; 21: 92.
Ader I, Brizuela L, Bouquerel P, Malavaud B, Cuvillier O. Sphingosine kinase 1: a new modulator of hypoxia inducible factor 1alpha during hypoxia in human cancer cells. CANCER RES. 2008; 68: 8635–42.
Dong ZR, Zhang PF, Wang CH, et al. Postoperative adjuvant transcatheter arterial chemoembolization for resectable multiple hepatocellular carcinoma beyond the Milan criteria: a retrospective analysis. AM J CANCER RES. 2015; 5: 450–7.
Liang L, Li C, Diao YK, et al. Survival benefits from adjuvant transcatheter arterial chemoembolization in patients undergoing liver resection for hepatocellular carcinoma: a systematic review and meta-analysis. Therap Adv Gastroenterol. 2020; 13: 320800739.
Wang Q, Xia D, Bai W, et al. Development of a prognostic score for recommended TACE candidates with hepatocellular carcinoma: A multicentre observational study. J HEPATOL. 2019; 70: 893–903.
Erstad DJ, Tanabe KK. Prognostic and Therapeutic Implications of Microvascular Invasion in Hepatocellular Carcinoma. ANN SURG ONCOL. 2019; 26: 1474–93.
Lee S, Kang TW, Song KD, et al. Effect of Microvascular Invasion Risk on Early Recurrence of Hepatocellular Carcinoma After Surgery and Radiofrequency Ablation. ANN SURG. 2021; 273: 564–71.
Charette N, Vandeputte C, Ameye L, et al. Prognostic value of adipose tissue and muscle mass in advanced colorectal cancer: a post hoc analysis of two non-randomized phase II trials. BMC CANCER. 2019; 19: 134.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Development and validation of a risk prediction algorithm for evaluating the efficacy of postoperative adjuvant TACE therapy for hepatocellular carcinoma: a retrospective study

Status:

Version 1

Abstract

Background and purpose

Methods

Results

Conclusions

Figures

Introduction

Patients and Methods

Patients

Treatment Procedures

Patients And Methods

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1