You are here

Article Text

Article menu
Immunotherapy biomarkers
Original research
Prognostic model for unresectable hepatocellular carcinoma treated with dual PD-1 and angiogenesis blockade therapy
  1. Zhiqiang Mo1,
  2. Ling Lv1,2,
  3. Qicong Mai1,
  4. Qiao Li3,
  5. Jian He4,
  6. Tao Zhang5,
  7. Jingwu Xu6,
  8. Jiayan Fang7,
  9. Ning Shi3,
  10. Qing Gou1,
  11. Xiaoming Chen1,
  12. Jing Zhang1,
  13. Wenhang Zhuang1 and
  14. Haosheng Jin3
  1. 1Department of Minimally Invasive Intervention, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
  2. 2Department of Radiology, Guangdong Provincial Key Laboratory of Major Obstetric Discases; Guangdong Provincial Clinical Research Center for 0bstetricsc and Gynecology;The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou, China
  3. 3Department of General Surgery, Guangdong Provincial People's Hospital(Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
  4. 4Department of Interventional Radiology, Shenzhen Traditional Chinese Medicine Hospital, Shenzhen, China
  5. 5Department of Radiology, Guangzhou Medical University Affiliated Cancer Hospital, Guangzhou, Guangdong, China
  6. 6Department of Oncology and Peripheral Interventional Radiology, People’s Hospital of Huazhou, Maoming, China
  7. 7Department of Internal Medicine-Oncology, Dongguan Songshan Lake Central Hospital, Dongguan, China
  1. Correspondence to Professor Haosheng Jin; kinghaos{at}126.com; Dr Wenhang Zhuang; zhuangwenxing{at}gdph.org.cn

Abstract

Background and aims Dual programmed death 1 (PD-1) and angiogenesis blockade therapy is a frontline treatment for hepatocellular carcinoma (HCC). An accepted model for survival prediction and risk stratification in individual patients receiving this treatment is lacking. Aimed to develop a simple prognostic model specific to these patients.

Approach and results Patients with unresectable HCC undergoing dual PD-1 and angiogenesis blockade therapy were included in training cohort (n=168) and validation cohort (n=72). We investigated the prognostic value of clinical variables on overall survival using a Cox model in the training set. A prognostic score model was then developed and validated. Predictive performance and discrimination were also evaluated. Largest tumor size and Alpha-fetoprotein concentration at baseline and Neutrophil count and Spleen volume change after 6 weeks of treatment were identified as independent predictors of overall survival in multivariable analysis and used to develop LANS score. Time-dependent receiver operating characteristic analysis, calibration curves, and C-index showed LANS score had favorable performance in survival prediction. Patients were divided into three risk categories based on LANS score. Median survival for patients with low, intermediate, and high LANS scores was 31.7, 23.5, and 11.5 months, respectively (p<0.0001). The disease control rates were 96.4%, 64.3%, and 32.1%, respectively (p<0.0001). The predictive performance and risk stratification ability of the LANS score were confirmed in validation and entire cohorts.

Conclusion The LANS score model can provide individualized survival prediction and risk stratification in patients with unresectable HCC undergoing dual PD-1 and angiogenesis blockade therapy.

  • Immune Checkpoint Inhibitors
  • Immunotherapy
  • Liver Neoplasms

Data availability statement

Data are available upon reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/jitc-2023-008191

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • As dual programmed death 1 (PD-1) and angiogenesis blockade therapy emerges as the new benchmark for systemic first-line treatment of unresectable hepatocellular carcinoma (uHCC), the significance of accurately identifying patients likely to benefit from this combinatory approach has been increasingly emphasized.

    WHAT THIS STUDY ADDS

    • A prognostic score model entitled as LANS score specifically developed to recommend candidates of dual PD-1 and angiogenesis blockade therapy.

    • We had integrated available and non-invasive clinical variables to refine LANS score in consideration of the dynamic nature of uHCC disease and its management.

    • LANS score displaying continuous data could easily provide individualized survival prediction and risk stratification.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • We anticipate that the LANS score could facilitate clinical decision-making for patients with uHCC undergoing dual PD-1 and angiogenesis blockade therapy.

    Introduction

    Treatment of unresectable hepatocellular carcinoma (uHCC) has undergone a major paradigm shift in recent decades. After sorafenib was shown to prolong survival in patients with uHCC,1 agents targeting the vascular endothelial growth factor axis have found widespread clinical application.2 In addition, immune checkpoint inhibitors have continued to evolve. Treatment strategies based on molecular targeted agents and immune checkpoint inhibitors have been extensively investigated in patients with HCC. In a phase III trial, the combination of atezolizumab and bevacizumab was superior to sorafenib in terms of overall and progression-free survival in patients with uHCC, which resulted in dual programmed death 1 (PD-1) and angiogenesis blockade therapy becoming a frontline uHCC treatment.3

    The efficacy of dual PD-1 and angiogenesis blockade therapy is rather variable in these patients, however, owing to significant interpatient variability in tumor load, liver function, macrovascular invasion, and extrahepatic metastasis.4 5 This treatment remains inadequate because of a lack of evidence in identifying appropriate candidate patients.6 Identifying individuals who would clearly benefit from combination therapy would assist with clinical decision-making in patients with uHCC.

    The Response Evaluation Criteria in Solid Tumors (RECIST) are widely used to assess clinical treatment response.7 8 However, they are unable to predict survival when assessed before treatment or in the early stages of treatment. A prognostic model specific to patients with HCC is needed. Although several biomarkers have been approved for predicting the response to immunotherapy in certain advanced solid tumors, none have been recognized or validated in HCC.6 Clinical parameters such as tumor burden, complete blood count, biochemical testing of liver function and tumor markers, macrovascular invasion and extrahepatic metastasis have been related to survival in patients with HCC and can be used to develop a model for prognosticating patient outcomes. This study aimed to develop and validate such a model that can predict survival and provide risk stratification in patients with uHCC undergoing dual PD-1 and angiogenesis blockade therapy.

    Materials and methods

    Study population

    Patients diagnosed with HCC who received combined immunotherapy and angiogenesis blockade therapy for advanced disease were retrospectively reviewed. The training cohort included patients from Guangdong Provincial People’s Hospital and Shenzhen Traditional Chinese Medicine Hospital between March, 2016 and December, 2022. The validation cohort included patients from Affiliated Cancer Hospital and Institute of Guangzhou Medical University, People’s Hospital of Huazhou and Dongguan Songshan Lake Central Hospital between June 2017 and September 2022. Patients with the following criteria were eligible for study inclusion: (1) diagnosis of HCC according to the guidelines of the American Association for the Study of Liver Disease9; (2) no treatment before initiation of combined immunotherapy and angiogenesis blockade therapy; (3) age 18–80 years; (4) Eastern Cooperative Oncology Group performance status score <2; (5) Child-Pugh score <9; (6) no other malignant tumor; and (7) at least one measurable lesion as defined by the modified RECIST.10 We excluded patients who received immunotherapy plus angiogenesis blockade therapy in combination with locoregional therapy and those who did not undergo the appropriate contrast-enhanced CT before or after treatment. We also excluded patients with incomplete medical or follow-up data.

    Treatment procedures

    All patients received bevacizumab (15 mg/kg of body weight) plus atezolizumab (1,200 mg) intravenously every 3 weeks, or bevacizumab biosimilar (IBI305, 15 mg/kg of body weight) plus sintilimab (200 mg) intravenously every 3 weeks, or oral apatinib (250 mg daily) in conjunction with intravenous camrelizumab (200 mg for body weight ≥50 kg or 3 mg/kg for body weight <50 kg) every 2 weeks. The treatment was continued until disease progression, patient withdrawal of consent, or the occurrence of unacceptable toxicity or death.

    Follow-up and evaluation criteria

    All patients underwent contrast-enhanced CT before and 6 weeks after treatment initiation (after the second cycle). Therapeutic effect was evaluated using the HCC-specific modified RECIST by two independent radiologists with 20 years of experience.

    Peripheral blood samples were collected for complete blood count and biochemical testing of liver function and tumor markers at baseline and 6 weeks after treatment initiation. Chest radiography was performed every 6 months in patients diagnosed with extrahepatic metastases. The final follow-up was on April, 2023.

    Spleen volume estimation

    Spleen volume was measured and calculated on CT at the times specified above. Change in spleen volume was also determined. The maximal length (Lmax) of the spleen was defined as the longest dimension between the poles of the spleen in the transverse section. Hilum thickness (Thhilum) was measured in the central part of the hilum perpendicularly to the long axis of the spleen. Vertical height (Hvert) was obtained from the longest vertical dimension between the cranial and caudal borders of the spleen in the coronal section. The three measurements were made in centimeters. Spleen volume was calculated as 30+0.58 (Lmax×Thhilum ×Hvert).11 12 Spleen measurements and volume calculations were performed by one radiologist who was blinded to other patient data.

    Statistical analysis

    A restricted cubic splines model fitted for Cox proportional hazards models with four knots was used to determine the cut-off value for the pretreatment and post-treatment spleen volumes and volume change during treatment, translating continuous variables into categorical data. Continuous data are expressed as means with SE and were compared using the independent samples t-test. Categorical data are expressed as numbers with percentage and were compared using the χ2 test or Fisher’s exact test as appropriate. Overall survival (OS) was defined as the time from treatment initiation to the time of death. Patients who were still alive were censored at the date of last contact. Survival curves were constructed using the Kaplan-Meier method and compared with the log-rank test. The training and validation cohorts were generated by computer-generated random numbers. Univariate and multivariate analyses were conducted using Cox proportional hazards regression to calculate HRs with 95% CIs in the training cohort; the result was validated in the validation cohort and the entire cohort. Variables with p<0.05 in the univariable analyses were included in the multivariate analyses to identify independent predictors of OS. A prognostic nomogram was formulated based on the independent predictors. The score was calculated in the training, validation, and entire cohorts based on the coefficient in the multivariate Cox proportional hazards model. The C-index for three cohorts were applied. The area under the time-dependent receiver operating characteristic curve was used to evaluate the predictive ability of the model. Bootstrapping with 1,000 resamples was used to validate the nomogram. A calibration curve was constructed to quantify overfitting of the modeling. All testing was two-tailed. P value<0.05 was considered significant. Statistical analyses were performed using R V.4.0.2 (R Foundation, Vienna, Austria).

    Results

    Patient characteristics

    Two hundred and forty patients were included in the training (n=168) and validation (n=72) cohorts. Patient characteristics are described in table 1. Median estimated follow-up in the entire, training, and validation cohorts was 26.3 months (95% CI, 14.9 to 30.2), 29.5 months (95% CI, 15.6 to 32.7), and 23.2 months (95% CI, 14.7 to 29.5), respectively.

    View this table:
    Table 1

    Patient characteristics

    Survival analysis

    Median OS in the entire cohort was 24.5 months (95% CI, 12.7 to 27.4). Six-month, 1-year, and 2-year OS was 82.5%, 59.6%, and 34.6%, respectively (online supplemental table S1). Median OS did not significantly differ between the training cohort (25.4 months (95% CI, 13.4 to 28.5]) and validation cohort (21.0 months (95% CI, 12.1 to 26.9); online supplemental table S1).

    Supplemental material

    Univariate and multivariate Cox regression analysis for risk factors related to survival (training cohort)

    In univariate analysis, 33 factors related to survival were evaluated for the 168 patients in the training cohort. Among these, 19 were confirmed to be important predictors and included in the multivariable analysis.

    The multivariate analysis suggested that largest tumor size (HR, 1.03; p=0.01), baseline alpha-fetoprotein (AFP) concentration 20–400 ng/mL (HR, 4.45; p=0.08), baseline AFP concentration >400 ng/mL (HR, 19.1; p=0.01), neutrophil count 6 weeks after treatment initiation (HR, 1.38; p=0.01), and spleen volume change >20.05 cm3 (HR, 4.2; p=0.01) were significantly associated with OS and considered for use in prognostic model development (table 2).

    View this table:
    Table 2

    Univariable and multivariable Cox regression analyses of factors potentially associated with overall survival in the training cohort

    Development and assessment of LANS score

    Given that Largest tumor size, AFP concentration, peripheral Neutrophil count, and Spleen volume change were prognostic factors, a nomogram entitled the LANS score was created to predict 6-month, 1-year, and 2-year OS for individual patients during treatment in the training cohort (figure 1).

    Figure 1
    Figure 1

    Nomogram of the LANS score for individual survival prediction. AFP, alpha-fetoprotein; OS, overall survival.

    In time-dependent receiver operating characteristic curve analysis, the area under the curve (AUC) values for LANS score prediction of 6-month, 1-year, and 2-year survival in the training cohort were 0.81 (95% CI, 0.68 to 0.94), 0.85 (95% CI, 0.77 to 0.94), and 0.87 (95% CI, 0.80 to 0.95), respectively. Corresponding values in the validation cohort were 0.84 (95% CI, 0.65 to 1.00), 0.67 (95% CI, 0.52 to 0.82), and 0.73 (95% CI, 0.61 to 0.84), respectively. In the entire cohort, the corresponding values were 0.83 (95% CI, 0.72 to 0.94), 0.78 (95% CI, 0.70 to 0.86), and 0.80 (95% CI, 0.71 to 0.83), respectively (figure 2). When comparing the Cox model fit with the Kaplan-Meier plots, the calibration curve reflected a high consistency between predictions and observed survival probabilities (figure 3). The C-index of the LANS score model for predicting OS in the training, validation, and entire cohorts was 0.81 (95% CI, 0.75 to 0.87), 0.72 (95% CI, 0.61 to 0.83), and 0.77 (95% CI, 0.71 to 0.83), respectively.

    Figure 2
    Figure 2

    Time-dependent areas under the receiver operating characteristic curve for LANS score in the training cohort (A) validation cohort (B) and entire cohort (C).AUC,area under the curve

    Figure 3
    Figure 3

    Calibration curves of predicted overall survival probability and actual overall survival at 6 months, 1 year, and 2 years survival probability in the training cohort (A) validation cohort. (B) And entire cohorts (C). OS, overall survival.AUC,Area Under the Curve

    The performance and discrimination of LANS score in survival prediction

    Patients were classified according to LANS score into low, intermediate, and high score groups using two cut-off values (4.42 and 8.52). Kaplan-Meier OS curves significantly differed between the three groups in the training, validation, and entire cohorts (figure 4).

    Figure 4
    Figure 4

    Kaplan-Meier survival curves according to LANS score in the training cohort (A) validation cohort (B) and entire cohort (C).

    In the training cohort, median OS in the low, intermediate, and high LANS score groups was 31.7 months (95% CI, 27.5 to 39.3), 23.5 months (95% CI, 17.4 to 29.1), and 11.5 months (95% CI, 7.1 to 15.0), respectively (p<0.0001). The corresponding median OS rates in the validation cohort were 28.5 months (95% CI, 21.5 to 35.2), 21.3 months (95% CI, 16.3 to 27.2), and 10.5 months (95% CI, 5.6 to 13.9), respectively (p<0.0001). In the entire cohort, the corresponding rates were 29.93 months (95% CI, 23.5 to 36.5), 21.9 months (95% CI, 17.1 to 28.5), and 11.2 months (95% CI, 6.1 to 14.5), respectively (p<0.0001; online supplemental table S2).

    Association between LANS score and treatment response

    In the training cohort, the proportion of patients with complete response or partial response in the low, intermediate, and high LANS score groups was 41.1%, 25.0%, and 14.2%, respectively. Corresponding proportions of patients with stable disease were 55.3%, 39.3%, and 17.9%, respectively, and corresponding proportions with progressive disease were 3.6%, 35.7%, and 67.9%, respectively. The disease control rates (DCRs) in the low, intermediate, and high LANS score groups were 96.4%, 64.3%, and 32.1%, respectively (p<0.0001; online supplemental table S2). In the validation and entire cohorts, the differences of DCR in the low, intermediate, and high LANS score groups remained unchanged (online supplemental table S2).

    Discussion

    Several clinical trials in patients with uHCC have demonstrated that dual PD-1 and angiogenesis blockade therapy can achieve a higher tumor response rate than sorafenib with greater tolerability.13–15 However, the prognoses of individual patients vary and are related to patient and tumor characteristics such as liver function, performance status, tumor burden, and macrovascular invasion.16 17 According to recent guidelines, risk stratification and survival prediction in these patients is important for treatment decision-making.

    In our study of patients with uHCC undergoing dual PD-1 and angiogenesis blockade therapy, largest tumor size, baseline AFP concentration, and neutrophil count and spleen volume change after 6 weeks of treatment were independent predictors of OS. Accordingly, we developed the LANS score, which was based on these variables using multivariate Cox proportional hazards regression. In the training cohort, patients were stratified into three groups based on LANS score: low, intermediate, and high. Median OS significantly differed between the groups: 31.7, 23.5, and 11.5 months, respectively. Crucially, the LANS score retained favorable performance and discrimination in an independent validation cohort. Using this nomogram can provide an estimate of survival probability early in treatment and be used as a reference to compare with expected survival using treatments other than dual PD-1 and angiogenesis blockade therapy. Notably, the three LANS score strata were approximately consistent with radiological response in our study. Achieving radiological control (response or stabilization) can be considered a treatment success since it is associated with better OS. These findings are clinically relevant for several reasons. First, the LANS score is easily applied in a real-world clinical setting as it is derived from readily available clinical parameters and easily modeled. Second, the LANS score can be used to identify patients more likely to respond to treatment. Third, it can assist physicians and patients with treatment decision-making and identify patients who may benefit from a treatment other than dual PD-1 and angiogenesis blockade therapy.

    Previous studies have identified that performance status, liver function, laboratory parameters, AFP concentration, and tumor profiles are prognostic indicators in patients with HCC.16–21 In recent studies, there are existing models including CRAFITY score and MAPS-CRAFITY score targeting patients with HCC treated with immunotherapy.20 22 These former models mainly derived from pretreatment clinical variables, which may not reflect the biological and clinical characteristics that change over time, as well as the patient’s response to treatment. Our results suggested that largest tumor size, AFP concentration, neutrophil count, and spleen volume change can be modeled as continuous linear variables. In considering the dynamic nature of disease and its management, we were working on integrating such data to refine our prognostic models. Moreover, we adopted evidence-based LANS score cut-off values to provide a basis for patient stratification.

    Typically, tumor burden seems to be a crucial determinant of prognosis and was primarily considered for model development.23–25 Even though the maximum tumor diameter had a smaller HR in our study, it could contribute to the granularity of the model. Its impact on prognosis might interact with other clinical parameters in ways not fully captured by the stronger predictors in patient with heterogeneous HCC population, owing to the fact that larger tumors are more likely to invade blood vessels, metastasize, or cause liver failure.

    AFP concentration has been identified as a risk factor for development of HCC.19 26 Emerging evidence has suggested it is not just a tumor marker.27–29 Recent studies have reported that AFP suppresses the antitumor immune response by inhibiting phagocytic activity in macrophages, proliferation of T lymphocytes, activity of natural killer cells, differentiation of dendritic cells, and increasing T suppressor cell activity.30 In addition, AFP has been implicated in enrichment of microvessel density in HCC. Although there is a paucity of research regarding the mechanistic link between AFP and angiogenesis, co-dysregulation of AFP expression and vascular endothelial growth factor has been observed in HCC cells in vitro.18 31 32 All these data imply that AFP may hamper antitumor therapy directly or indirectly via modulation of angiogenesis and immunosuppression.33 34 Consistent with previous studies, it makes sense to consider baseline AFP concentration as a valuable prognostic parameter.

    Multiple immune cell types in the tumor microenvironment play an important role in cancer biology and treatment response.6 In our study, neutrophil count after 6 weeks of treatment was a predictor of poor prognosis and recognized as a prognostic parameter. The current understanding of the negative effect of neutrophils on treatment is limited. Neutrophils constitute a substantial proportion of the immune infiltrate in many types of cancer. Some studies have suggested that cancer-related neutrophils may switch on angiogenic and immunosuppressive functions when exposed to environmental cues in the tumor microenvironment.35–37 However, it is still a disputed issue to use circulating neutrophils to elucidate intratumoral and peritumoral environment features. Clinical detection of tumor-infiltrating immune cells mainly relies on biopsy, which is not practical to perform periodically. Previous studies have shown that computational medical images contain cellular and molecular tumor information that is relevant to biological behavior. Imaging features have been explored as a surrogate method for evaluating the tumor microenvironment to predict prognosis and assist with treatment decision-making. In our study, spleen volume change between baseline and 6 weeks after treatment initiation was associated with outcome and considered a prognostic parameter. The observed change in spleen volume is due to accumulation of myeloid-derived suppressor cells (MDSCs) during immunotherapy treatment. MDSCs are a population of immature myeloid cells which can modulate immunosuppression and angiogenesis to support tumor progression.38–42 In metastatic non-small cell lung, advanced pancreatic, and metastatic colorectal cancer, spleen volume change might be a surrogate marker of MDSC-dependent treatment inhibition.11 40 43 These results may explain why neutrophil count and spleen volume change were included in the prognostic model.

    Our study has several limitations. First, it was non-randomized, retrospective, and had a small sample size. Selection bias may have been present. However, we minimized this risk by including consecutive patients from five medical centers. Second, the LANS score was developed in Chinese patients with mainly hepatitis B-related HCC. Our model may not be applicable in patients from other regions where HCC is mainly caused by other etiologies such as hepatitis C or alcohol. Third, all patients in the study were treatment naive. However, in clinical practice, some patients receive systemic or locoregional treatment before dual PD-1 and angiogenesis blockade therapy. Given the above limitations, further model validation in other clinical settings is necessary.

    In conclusion, we have developed a novel prognostic score model which showed favorable performance and risk stratification ability in patients with uHCC undergoing dual PD-1 and angiogenesis blockade therapy. Considering the model is based on four readily available clinical parameters, the LANS score should assist physicians and patients with clinical decision-making.

    Data availability statement

    Data are available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    The study was conducted in accordance with the principles of the Helsinki Declaration and approved by the ethics committee of Guangdong Provincial People’s Hospital (KYQ202206601). Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    We are grateful to the staff and patients who supported this clinical study. We also thank Liwen Bianji (Edanz) (https://www.liwenbianji.cn) for editing the English text of a draft of this manuscript.

    References

    1. ↵
      1. Llovet JM,
      2. Ricci S,
      3. Mazzaferro V, et al
      . Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008;359:378–90. doi:10.1056/NEJMoa0708857
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Kudo M,
      2. Finn RS,
      3. Qin S, et al
      . Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet 2018;391:1163–73. doi:10.1016/S0140-6736(18)30207-1
      OpenUrlCrossRefPubMed
    3. ↵
      1. Finn RS,
      2. Qin S,
      3. Ikeda M, et al
      . Atezolizumab plus Bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med 2020;382:1894–905. doi:10.1056/NEJMoa1915745
      OpenUrlCrossRefPubMed
    4. ↵
      1. Llovet JM,
      2. Kelley RK,
      3. Villanueva A, et al
      . Hepatocellular carcinoma. Nat Rev Dis Primers 2021;7:6. doi:10.1038/s41572-020-00240-3
      OpenUrl
    5. ↵
      1. Qin S,
      2. Ren Z,
      3. Feng Y-H, et al
      . Atezolizumab plus Bevacizumab versus sorafenib in the Chinese subpopulation with unresectable hepatocellular carcinoma: phase 3 randomized, open-label Imbrave150 study. Liver Cancer 2021;10:296–308. doi:10.1159/000513486
      OpenUrlCrossRefPubMed
    6. ↵
      1. Llovet JM,
      2. Castet F,
      3. Heikenwalder M, et al
      . Immunotherapies for hepatocellular carcinoma. Nat Rev Clin Oncol 2022;19:151–72. doi:10.1038/s41571-021-00573-2
      OpenUrlCrossRefPubMed
    7. ↵
      1. Therasse P,
      2. Arbuck SG,
      3. Eisenhauer EA, et al
      . New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States. J Natl Cancer Inst 2000;92:205–16. doi:10.1093/jnci/92.3.205
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Eisenhauer EA,
      2. Therasse P,
      3. Bogaerts J, et al
      . New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228–47. doi:10.1016/j.ejca.2008.10.026
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Heimbach JK,
      2. Kulik LM,
      3. Finn RS, et al
      . AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology 2018;67:358–80. doi:10.1002/hep.29086
      OpenUrlCrossRefPubMed
    10. ↵
      1. Lencioni R,
      2. Llovet JM
      . Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis 2010;30:52–60. doi:10.1055/s-0030-1247132
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Galland L,
      2. Lecuelle J,
      3. Favier L, et al
      . Splenic volume as a surrogate marker of immune checkpoint inhibitor efficacy in metastatic non small cell lung cancer. Cancers (Basel) 2021;13:3020. doi:10.3390/cancers13123020
    12. ↵
      1. Kucybała I,
      2. Ciuk S,
      3. Tęczar J
      . Spleen enlargement assessment using computed tomography: which coefficient correlates the strongest with the real volume of the spleen? Abdom Radiol (NY) 2018;43:2455–61. doi:10.1007/s00261-018-1500-9
      OpenUrl
    13. ↵
      1. Ren Z,
      2. Xu J,
      3. Bai Y, et al
      . Sintilimab plus a Bevacizumab Biosimilar (Ibi305) versus sorafenib in unresectable hepatocellular carcinoma (ORIENT-32): a randomised, open-label, phase 2-3 study. Lancet Oncol 2021;22:977–90. doi:10.1016/S1470-2045(21)00252-7
      OpenUrlCrossRefPubMed
    14. ↵
      1. Galle PR,
      2. Finn RS,
      3. Qin S, et al
      . Patient-reported outcomes with Atezolizumab plus Bevacizumab versus sorafenib in patients with Unresectable hepatocellular carcinoma (Imbrave150): an open-label, randomised, phase 3 trial. Lancet Oncol 2021;22:991–1001. doi:10.1016/S1470-2045(21)00151-0
      OpenUrlCrossRef
    15. ↵
      1. Kelley RK,
      2. Rimassa L,
      3. Cheng A-L, et al
      . Cabozantinib plus Atezolizumab versus sorafenib for advanced hepatocellular carcinoma (COSMIC-312): a multicentre, open-label, randomised, phase 3 trial. Lancet Oncol 2022;23:995–1008. doi:10.1016/S1470-2045(22)00326-6
      OpenUrlCrossRefPubMed
    16. ↵
      1. Llovet JM,
      2. Brú C,
      3. Bruix J
      . Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999;19:329–38. doi:10.1055/s-2007-1007122
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Reig M,
      2. Forner A,
      3. Rimola J, et al
      . BCLC strategy for prognosis prediction and treatment recommendation: the 2022 update. J Hepatol 2022;76:681–93. doi:10.1016/j.jhep.2021.11.018
      OpenUrlCrossRefPubMed
    18. ↵
      1. Shan Y,
      2. Huang Y,
      3. Xie Y, et al
      . Angiogenesis and Clinicopathologic characteristics in different hepatocellular carcinoma subtypes defined by Epcam and Α-Fetoprotein expression status. Med Oncol 2011;28:1012–6. doi:10.1007/s12032-010-9600-6
      OpenUrlPubMed
    19. ↵
      1. Galle PR,
      2. Foerster F,
      3. Kudo M, et al
      . Biology and significance of alpha-Fetoprotein in hepatocellular carcinoma. Liver Int 2019;39:2214–29. doi:10.1111/liv.14223
      OpenUrlPubMed
    20. ↵
      1. Scheiner B,
      2. Pomej K,
      3. Kirstein MM, et al
      . Prognosis of patients with hepatocellular carcinoma treated with immunotherapy - development and validation of the CRAFITY score. J Hepatol 2022;76:353–63. doi:10.1016/j.jhep.2021.09.035
      OpenUrlCrossRef
    21. ↵
      1. Kudo M
      . Newly developed modified ALBI grade shows better Prognostic and predictive value for hepatocellular carcinoma. Liver Cancer 2022;11:1–8. doi:10.1159/000521374
      OpenUrl
    22. ↵
      1. Ouyang J,
      2. Yang Y,
      3. Zhou Y, et al
      . The MAPS-CRAFITY score: a novel efficacy predictive tool for Unresectable hepatocellular carcinoma treated with targeted therapy plus immunotherapy. Hepatol Int 2023;17:1519–31. doi:10.1007/s12072-023-10580-3
      OpenUrl
    23. ↵
      1. Mazzaferro V,
      2. Llovet JM,
      3. Miceli R, et al
      . Predicting survival after liver transplantation in patients with hepatocellular carcinoma beyond the Milan criteria: a retrospective, exploratory analysis. Lancet Oncol 2009;10:35–43. doi:10.1016/S1470-2045(08)70284-5
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Wang Q,
      2. Xia D,
      3. 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. doi:10.1016/j.jhep.2019.01.013
      OpenUrl
    25. ↵
      1. Yamakado K,
      2. Miyayama S,
      3. Hirota S, et al
      . Subgrouping of intermediate-stage (BCLC stage B) hepatocellular carcinoma based on tumor number and size and child-pugh grade correlated with prognosis after transarterial chemoembolization. Jpn J Radiol 2014;32:260–5. doi:10.1007/s11604-014-0298-9
      OpenUrlCrossRefPubMed
    26. ↵
      1. Chen T,
      2. Dai X,
      3. Dai J, et al
      . AFP promotes HCC progression by suppressing the Hur-mediated Fas/FADD apoptotic pathway. Cell Death Dis 2020;11:822. doi:10.1038/s41419-020-03030-7
    27. ↵
      1. Memon K,
      2. Kulik L,
      3. Lewandowski RJ, et al
      . Alpha-fetoprotein response correlates with EASL response and survival in solitary hepatocellular carcinoma treated with transarterial therapies: a subgroup analysis. J Hepatol 2012;56:1112–20. doi:10.1016/j.jhep.2011.11.020
      OpenUrlCrossRefPubMed
    28. ↵
      1. Kim BK,
      2. Ahn SH,
      3. Seong JS, et al
      . Early Α-Fetoprotein response as a predictor for clinical outcome after localized concurrent chemoradiotherapy for advanced hepatocellular carcinoma. Liver Int 2011;31:369–76. doi:10.1111/j.1478-3231.2010.02368.x
      OpenUrlCrossRefPubMed
    29. ↵
      1. Zhu AX,
      2. Kang Y-K,
      3. Yen C-J, et al
      . Ramucirumab after sorafenib in patients with advanced hepatocellular carcinoma and increased Α-Fetoprotein concentrations (REACH-2): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol 2019;20:282–96. doi:10.1016/S1470-2045(18)30937-9
      OpenUrlCrossRefPubMed
    30. ↵
      1. Pardee AD,
      2. Shi J,
      3. Butterfield LH
      . Tumor-derived Α-Fetoprotein impairs the differentiation and T cell stimulatory activity of human dendritic cells. J Immunol 2014;193:5723–32. doi:10.4049/jimmunol.1400725
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Terris B,
      2. Cavard C,
      3. Perret C
      . Epcam, a new marker for cancer stem cells in hepatocellular carcinoma. J Hepatol 2010;52:280–1. doi:10.1016/j.jhep.2009.10.026
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Terentiev AA,
      2. Moldogazieva NT
      . Alpha-Fetoprotein: a renaissance. Tumour Biol 2013;34:2075–91. doi:10.1007/s13277-013-0904-y
      OpenUrl
    33. ↵
      1. Pinter M,
      2. Jain RK,
      3. Duda DG
      . The current landscape of immune checkpoint blockade in hepatocellular carcinoma: a review. JAMA Oncol 2021;7:113–23. doi:10.1001/jamaoncol.2020.3381
      OpenUrl
    34. ↵
      1. Fukumura D,
      2. Kloepper J,
      3. Amoozgar Z, et al
      . Enhancing cancer immunotherapy using antiangiogenics: opportunities and challenges. Nat Rev Clin Oncol 2018;15:325–40. doi:10.1038/nrclinonc.2018.29
      OpenUrlCrossRefPubMed
    35. ↵
      1. Geh D,
      2. Leslie J,
      3. Rumney R, et al
      . Neutrophils as potential therapeutic targets in hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol 2022;19:257–73. doi:10.1038/s41575-021-00568-5
      OpenUrl
    36. ↵
      1. Wang Y,
      2. Zhao Q,
      3. Zhao B, et al
      . Remodeling tumor-associated neutrophils to enhance Dendritic cell-based HCC Neoantigen Nano-vaccine efficiency. Adv Sci (Weinh) 2022;9:2105631. doi:10.1002/advs.202105631
    37. ↵
      1. Shaul ME,
      2. Fridlender ZG
      . Tumour-associated neutrophils in patients with cancer. Nat Rev Clin Oncol 2019;16:601–20. doi:10.1038/s41571-019-0222-4
      OpenUrlCrossRefPubMed
    38. ↵
      1. Gabrilovich DI,
      2. Nagaraj S
      . Myeloid-derived Suppressor cells as regulators of the immune system. Nat Rev Immunol 2009;9:162–74. doi:10.1038/nri2506
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. Gabrilovich DI
      . Myeloid-derived suppressor cells. Cancer Immunol Res 2017;5:3–8. doi:10.1158/2326-6066.CIR-16-0297
      OpenUrlAbstract/FREE Full Text
    40. ↵
      1. Niogret J,
      2. Limagne E,
      3. Thibaudin M, et al
      . Baseline splenic volume as a prognostic biomarker of FOLFIRI efficacy and a surrogate marker of MDSC accumulation in metastatic colorectal carcinoma. Cancers (Basel) 2020;12:1429. doi:10.3390/cancers12061429
    41. ↵
      1. Limagne E,
      2. Euvrard R,
      3. Thibaudin M, et al
      . Accumulation of MDSC and Th17 cells in patients with metastatic colorectal cancer predicts the efficacy of a FOLFOX-Bevacizumab drug treatment regimen. Cancer Res 2016;76:5241–52. doi:10.1158/0008-5472.CAN-15-3164
      OpenUrlAbstract/FREE Full Text
    42. ↵
      1. Oura K,
      2. Morishita A,
      3. Tani J, et al
      . Tumor immune microenvironment and immunosuppressive therapy in hepatocellular carcinoma: a review. Int J Mol Sci 2021;22:11. doi:10.3390/ijms22115801
      OpenUrlCrossRef
    43. ↵
      1. Aarnink A,
      2. Richard C,
      3. Truntzer C, et al
      . Baseline splenic volume as a surrogate marker of FOLFIRINOX efficacy in advanced pancreatic carcinoma. Oncotarget 2018;9:25617–29. doi:10.18632/oncotarget.25424
      OpenUrl

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • ZM, LL and QM contributed equally.

    • Contributors HJ, WZ, and ZM conceived and designed the study. ZM, LL, and QM analyzed the data and drafted the manuscript. QL, JH, and NS were the study coordinators. QG, TZ, XC, and JZ conducted follow-up and data collection. Guarantor: HJ. All authors read and approved the final manuscript.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.