CoChat
Sign Up
Cited by 391Hepatocellular Carcinoma Treatment and PrognosisCancer Immunotherapy and BiomarkersHepatitis B Virus Studies

Adjuvant and neoadjuvant immunotherapies in hepatocellular carcinoma

Open access
Llovet, Josep M., Pinyol, Roser, Yarchoan, Mark, Singal, Amit G., Marron, Thomas U., Schwartz, Myron, Pikarsky, Eli, Kudo, Masatoshi +1 more·2024·Nature Reviews Clinical Oncology
DOI 10.1038/s41571-024-00868-0PMC PMC11984461
Open this paper in CoChat — free accountOpen PDF

Abstract

Liver cancer, specifically hepatocellular carcinoma (HCC), is the sixth most common cancer and the third leading cause of cancer mortality worldwide. The development of effective systemic therapies, particularly those involving immune-checkpoint inhibitors (ICIs), has substantially improved the outcomes of patients with advanced-stage HCC. Approximately 30% of patients are diagnosed with early stage disease and currently receive potentially curative therapies, such as resection, liver transplantation or local ablation, which result in median overall survival durations beyond 60 months. Nonetheless, up to 70% of these patients will have disease recurrence within 5 years of resection or local ablation. To date, the results of randomized clinical trials testing adjuvant therapy in patients with HCC have been negative. This major unmet need has been addressed with the IMbrave 050 trial, demonstrating a recurrence-free survival benefit in patients with a high risk of relapse after resection or local ablation who received adjuvant atezolizumab plus bevacizumab. In parallel, studies testing neoadjuvant ICIs alone or in combination in patients with early stage disease have also reported efficacy. In this Review, we provide a comprehensive overview of the current approaches to manage patients with early stage HCC. We also describe the tumour immune microenvironment and the mechanisms of action of ICIs and cancer vaccines in this setting. Finally, we summarize the available evidence from phase II/III trials of neoadjuvant and adjuvant approaches and discuss emerging clinical trials, identification of biomarkers and clinical trial design considerations for future studies.

Integrity signals

Registry

Crossref · 10.1038/s41571-024-00868-0

Cites

157 of 157 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

110 of 167 cited sources checked · 44 supported · 3 flagged · accuracy not yet measured

  • Premisecitation

    Reported 5-year survival after resection varies widely with selection criteria, from about 50% in China to about 70% in Europe.

    “Consequently, the reported outcomes differ widely depending on the selection criteria applied, with 5-year survival rates ranging between 50% in China to 70% in Europe28.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [28] Reveron-Thornton, R. F. et al. Global and regional long-term survival following resection for HCC in the recent decade: a meta-analysis of 110 studies. Hepatol. Commun. 6, 1813–1826 (2022). source

      The Results (Overall survival; page unavailable) report 5-year survival of 48.30% in Europe, not about 70%, and 57.03% in Asia. The Discussion (page unavailable) suggests there may not be much difference in patient selection among regions, contrary to the claimed explanation for the regional differences. (LLM, conf=0.91)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
  • Interpretationcitation

    The study illustrates immunodominance, in which T cells targeting a dominant clone expand mainly at the expense of T cells reactive to subdominant clones.

    “This study highlights the concept of immunodominance, whereby T cells targeting a dominant clone are primarily expanded at the expense of T cells reactive to subdominant clones.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [110] Friedman, J. et al. Neoadjuvant PD-1 immune checkpoint blockade reverses functional immunodominance among tumor antigen-specific T cells. Clin. Cancer Res. 26, 679–689 (2020). source

      The Discussion supports the existence of immunodominance: dominant-antigen-specific T cells suppress responses to subdominant antigens. But it explicitly says the study did not demonstrate differential clonal expansion; instead, it found functional suppression of subdominant T cells associated with PD-1 expression. That contradicts the claim that dominant T cells primarily expand at their expense. (LLM, conf=0.89)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
  • Premisecitation

    The IMbrave 050 trial showed that adjuvant atezolizumab plus bevacizumab significantly improved recurrence-free survival compared with active surveillance in patients with hepatocellular carcinoma at high risk of recurrence after resection or local ablation.

    “In 2023, however, the treatment landscape for patients with early stage HCC evolved following positive results from IMbrave 050, in which adjuvant therapy with the ICI atezolizumab plus the anti-angiogenic agent bevacizumab significantly improved recurrence-free survival (RFS) over active surveillance in patients with a high risk of disease recurrence12.”
    • Also stated in: § IMbrave 050 · boosted
      “Results from the global, open-label, phase III IMbrave 050 trial12 have shown a significant RFS improvement in patients with HCC at high risk of recurrence after resection or local ablation who received adjuvant atezolizumab plus bevacizumab compared with those who underwent active surveillance in patients (Box 1).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [12] Qin, S. et al. Atezolizumab plus bevacizumab versus active surveillance in patients with resected or ablated high-risk hepatocellular carcinoma (IMbrave 050): a randomised, open-label, multicentre, phase 3 trial. Lancet 402, 1835–1847 (2023). source
    • Rests on an assumption: A recurrence-free survival improvement is sufficient to regard adjuvant atezolizumab plus bevacizumab as a major advance in hepatocellular carcinoma management, even before overall survival benefit is shown.
  • Premisecitation

    Improvements in systemic therapy for advanced hepatocellular carcinoma have not been matched by improved treatment of early-stage disease.

    “However, this remarkable progress in systemic therapy has not been paralleled by improvements in the treatment of early stage HCC, which typically involves administration of therapies with curative intent, such as resection, liver transplantation or local ablation1,2.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [1] Llovet, J. M. et al. Hepatocellular carcinoma. Nat. Rev. Dis. Prim. 7, 7 (2021). source
    • cites [2] Llovet, J. M. et al. Molecular pathogenesis and systemic therapies for hepatocellular carcinoma. Nat. Cancer 3, 386–401 (2022). source

      The Introduction describes major advances in systemic therapy, and 'Placing systemic therapies in the context of HCC management' identifies resection, transplantation and ablation as curative options for early-stage HCC. 'Future directions' calls neoadjuvant and adjuvant therapies an unmet need, but the paper does not establish the broader claim that early-stage treatment has not improved. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
  • Premisecitation

    Regimens containing immune-checkpoint inhibitors are currently recommended as first-line therapy in clinical guidelines for advanced hepatocellular carcinoma.

    “In particular, regimens including immune-checkpoint inhibitors (ICIs) are currently adopted as first-line therapies in clinical guidelines7.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [7] Singal, A. G. et al. AASLD practice guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology 78, 1922–1965 (2023). source
  • May combine two claims. 47
  • Classified as probably not a substantial claim. 12
  • The quote only partly states this claim. 1

  • May combine two claims.

    Premisereasoning

    In perioperative HCC, no surrogate endpoint has been validated, unlike pathological complete response in NSCLC and breast cancer.

    “Given the lack of successful trials in the perioperative space in HCC, no surrogate end point has been validated yet, as is the case for pCR in patients with NSCLC20,120 and breast cancer8.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [20] Forde, P. M. et al. Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer. N. Engl. J. Med. 386, 1973–1985 (2022). source

      The Discussion says that a trial-level association between pathological complete response and survival in NSCLC had not been shown prospectively. It describes pathological complete response as promising and says its association with survival benefit requires further evaluation, contrary to the claim that it is a validated surrogate endpoint in NSCLC. This paper does not assess surrogate-endpoint validation in HCC. (LLM, conf=0.88)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
    • cites [120] Wakelee, H. et al. Perioperative pembrolizumab for early-stage non-small-cell lung cancer. N. Engl. J. Med. 389, 491–503 (2023). source

      The Methods (Assessments and Endpoints; page not provided) identify pathological complete response as a secondary endpoint in the NSCLC trial, and the Results (Efficacy; page not provided) report that pembrolizumab improved it. Neither passage establishes that pathological complete response is a validated surrogate endpoint in NSCLC. The HCC and breast-cancer portions of the sentence are not attributed to this source. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [8] Schmid, P. et al. Event-free survival with pembrolizumab in early triple-negative breast cancer. N. Engl. J. Med. 386, 556–567 (2022). source

      The End Points section lists pathological complete response (pCR) as a primary endpoint in a breast-cancer trial, and the Efficacy section reports event-free survival by pCR status. Neither section says pCR is a validated surrogate endpoint in breast cancer; the paper measures event-free survival separately. The HCC and NSCLC portions of the citing sentence are not attributed to this source. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Surgical resection is recognized worldwide as the preferred treatment for early-stage hepatocellular carcinoma, while liver transplantation is indicated when resection is unsuitable.

    “Surgical resection is recognized worldwide as the preferred treatment for patients with early stage HCC, whereas liver transplantation is indicated in those with early stage disease who are not deemed suitable for resection7,22.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [7] Singal, A. G. et al. AASLD practice guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology 78, 1922–1965 (2023). source

      The Surgical resection section calls resection the treatment of choice for localized HCC without cirrhosis and recommends it for selected patients with compensated cirrhosis. The Liver transplantation section supports transplantation for early-stage patients ineligible for resection because of liver dysfunction or multifocality. The claim overgeneralizes resection as preferred for all early-stage patients worldwide and omits transplant eligibility; page numbers were not available in our copy. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites [22] Galle, P. R. et al. EASL Clinical Practice Guidelines: management of hepatocellular carcinoma. J. Hepatol. 69, 182–236 (2018). source

  • May combine two claims.

    Premisecitation

    After resection or local ablation, median overall survival exceeds 60 months, but 30-50% of patients have recurrence by 3 years.

    “Despite an impressive median overall survival (OS) beyond 60 months after resection or local ablation, up to 30–50% of patients have disease recurrence at 3 years, often resulting from intrahepatic metastases or de novo tumours arising in the underlying liver pathology1,2.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [1] Llovet, J. M. et al. Hepatocellular carcinoma. Nat. Rev. Dis. Prim. 7, 7 (2021). source
    • cites [2] Llovet, J. M. et al. Molecular pathogenesis and systemic therapies for hepatocellular carcinoma. Nat. Cancer 3, 386–401 (2022). source

      The ‘Placing systemic therapies in the context of HCC management’ section (page unavailable) identifies resection and ablation as curative options for early-stage HCC, but gives neither median survival after those treatments nor a three-year recurrence rate. Searches for 60-month survival, three-year recurrence, and recurrence after resection or ablation found no passage supporting those figures; the paper’s 30–50% figures refer to other populations or treatments. (LLM, conf=0.75)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Eligibility criteria for resection vary by location, and European and US guidelines recommend more restrictive criteria than Asian guidelines.

    “In general, European and US guidelines recommend more-restrictive criteria for resection than Asian guidelines7,22–27.”
    Quote in the paperNot a substantial claim3 of 7 sources checked
    • cites [7] Singal, A. G. et al. AASLD practice guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology 78, 1922–1965 (2023). source

      The Surgical resection section says resection is mainly limited to BCLC stage 0/A and advises that extended indications be considered only in high-volume centers after multidisciplinary discussion. It notes that most evidence for resection in patients with portal vein tumor thrombus comes from Asia, but it does not compare the eligibility criteria recommended by European, US, and Asian guidelines. The claimed regional ranking is therefore not established by this source. (LLM, conf=0.79)

      Experimental — accuracy not yet measured
    • cites [22] Galle, P. R. et al. EASL Clinical Practice Guidelines: management of hepatocellular carcinoma. J. Hepatol. 69, 182–236 (2018). source
    • cites [23] Kudo, M. et al. Management of hepatocellular carcinoma in Japan: JSH Consensus Statements and Recommendations 2021 Update. Liver Cancer 10, 181–223 (2021). source
    • cites [24] Xie, D., Shi, J., Zhou, J., Fan, J. & Gao, Q. Clinical practice guidelines and real-life practice in hepatocellular carcinoma: a Chinese perspective. Clin. Mol. Hepatol. 29, 206–216 (2023). source

      The Hepatectomy section says Chinese guidelines prefer resection for CNLC stages Ia–IIa and permit it for selected IIb and IIIa cases. This supports the account of Chinese eligibility criteria, but our copy does not state that European and US guidelines are more restrictive than Asian guidelines generally. (LLM, conf=0.80)

      Experimental — accuracy not yet measured
    • cites [25] Goh, M. J. et al. Clinical practice guideline and real-life practice in hepatocellular carcinoma: a Korean perspective. Clin. Mol. Hepatol. 29, 197–205 (2023). source

      The TREATMENT > Hepatic resection section (page unavailable) says the Korean guidelines selectively permit limited resection for some patients with mild portal hypertension or hyperbilirubinemia. That establishes a Korean eligibility criterion, but our copy does not compare it with European or US resection guidelines, so the claimed regional ranking is unsupported. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [26] Roayaie, S. et al. The role of hepatic resection in the treatment of hepatocellular cancer. Hepatology 62, 440–451 (2015). source
    • cites [27] Llovet, J. M., Schwartz, M. & Mazzaferro, V. Resection and liver transplantation for hepatocellular carcinoma. Semin. Liver Dis. 25, 181–200 (2005). source

  • May combine two claims.

    Premisecitation

    After local ablation, objective response and recurrence-free survival are inversely related to tumour size; for small hepatocellular carcinomas, 3-year recurrence-free survival is about 45%, 3-year overall survival about 75%, and 5-year overall survival about 60%.

    “ORRs and RFS after local ablation are inversely proportional to tumour size, with optimal outcomes observed for patients with small HCCs in whom 3-year RFS and OS are approximately 45% and 75%, respectively42,43, and 5-year OS is around 60% (refs. 44–47).”
    Quote in the paperNot a substantial claim2 of 6 sources checked
    • cites [42] Doyle, A. et al. Outcomes of radiofrequency ablation as first-line therapy for hepatocellular carcinoma less than 3 cm in potentially transplantable patients. J. Hepatol. 70, 866–873 (2019). source
    • cites [43] Shiina, S. et al. Radiofrequency ablation for hepatocellular carcinoma: 10-year outcome and prognostic factors. Am. J. Gastroenterol. 107, 569–577 (2012). source

      Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured
    • cites [44] Charalel, R. A. et al. Long-term survival after surgery versus ablation for early liver cancer in a large, nationally representative cohort. J. Am. Coll. Radiol. 19, 1213–1223 (2022). source
    • cites [45] Pompili, M. et al. Long-term effectiveness of resection and radiofrequency ablation for single hepatocellular carcinoma ≤ 3 cm. Results of a multicenter Italian survey. J. Hepatol. 59, 89–97 (2013). source
    • cites [46] Kudo, M. et al. Report of the 22nd nationwide follow-up Survey of Primary Liver Cancer in Japan (2012–2013). Hepatol. Res. 52, 5–66 (2022). source
    • cites [47] Yoon, J. S. et al. Hepatocellular carcinoma in Korea between 2008 and 2011: an analysis of Korean Nationwide Cancer Registry. J. Liver Cancer 20, 41–52 (2020). source

      Split into 2 atoms — 0 confirmed, 0 partially supported, 0 disputed, 0 misattributed, 2 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Radiofrequency ablation or microwave ablation provides better overall survival and objective response rates with fewer sessions than percutaneous ethanol injection.

    “Percutaneous ethanol injection was the original ablation technique, although this has been largely replaced by radiofrequency ablation (RFA) or microwave ablation (MWA), which provide both superior OS and objective response rates (ORRs) with fewer sessions41.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [41] Lin, S.-M., Lin, C.-J., Lin, C.-C., Hsu, C.-W. & Chen, Y.-C. Randomised controlled trial comparing percutaneous radiofrequency thermal ablation, percutaneous ethanol injection, and percutaneous acetic acid injection to treat hepatocellular carcinoma of 3 cm or less. Gut 54, 1151–1156 (2005). source

      The Discussion (page unavailable) reports higher overall survival with radiofrequency thermal ablation (RFTA) than with percutaneous ethanol injection (PEI) for hepatocellular carcinomas of 3 cm or less. The Complete tumour necrosis section (page unavailable) reports a higher necrosis rate and fewer treatment sessions with RFTA, but does not label the necrosis measure an objective response rate. The paper does not study microwave ablation or establish that either technique has largely replaced PEI. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Several gene signatures capturing inflamed tumour microenvironment components are associated with favourable response to immune-checkpoint inhibitors, but none has been clinically validated.

    “Several gene signatures capturing the inflamed components of the TME have been associated with a favourable response to ICIs69,72,78–81, but none has been clinically validated thus far.”
    Quote in the paperNot a substantial claim4 of 6 sources checked
    • cites [69] Zhu, A. X. et al. Molecular correlates of clinical response and resistance to atezolizumab in combination with bevacizumab in advanced hepatocellular carcinoma. Nat. Med. 28, 1599–1611 (2022). source
    • cites [72] Haber, P. K. et al. Molecular markers of response to anti-PD1 therapy in advanced hepatocellular carcinoma. Gastroenterology 164, 72–88.e18 (2023). source

      The Results section (‘Molecular Features of Patients With HCC Responding to Anti-PD1 in Front Line’; page unavailable) reports that several inflammatory gene signatures were enriched in HCC patients responding to first-line anti-PD1 therapy. The Discussion says the only FDA-approved anti-PD1 biomarkers are tumor mutational burden and microsatellite instability, and that the authors’ new signature still needs phase III evaluation. The paper therefore supports the association and the lack of established clinical use, but does not establish that no such signature has been clinically validated across all immune-checkpoint inhibitors and settings. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [78] Montironi, C. et al. Inflamed and non-inflamed classes of HCC: a revised immunogenomic classification. Gut 72, 129–140 (2023). source

      The Discussion (page unavailable) reports two inflammatory signatures associated with ICI response in advanced HCC, and the Results (page unavailable) find higher expression of the paper’s Inflamed signature in responders than non-responders. The paper describes a continuing need for clinically useful biomarkers, but it does not establish the categorical claim that none of these signatures has been clinically validated. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [79] Fehrenbacher, L. et al. Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial. Lancet 387, 1837–1846 (2016). source
    • cites [80] Sangro, B. et al. Association of inflammatory biomarkers with clinical outcomes in nivolumab-treated patients with advanced hepatocellular carcinoma. J. Hepatol. 73, 1460–1469 (2020). source
    • cites [81] Ayers, M. et al. IFN-γ-related mRNA profile predicts clinical response to PD-1 blockade. J. Clin. Invest. 127, 1–11 (2017). source

  • May combine two claims.

    Premisecitation

    In IMbrave 050, at a median follow-up of 17.4 months, median recurrence-free survival was not reached in either group, but the intervention was favoured over active surveillance (HR 0.72, 95% CI 0.56-0.93; P = 0.012).

    “At a median follow-up duration of 17.4 months, median RFS had not been reached in either group, although statistical analyses favoured intervention over active surveillance (HR 0.72, 95% CI 0.56–0.93; P = 0.012).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [12] Qin, S. et al. Atezolizumab plus bevacizumab versus active surveillance in patients with resected or ablated high-risk hepatocellular carcinoma (IMbrave 050): a randomised, open-label, multicentre, phase 3 trial. Lancet 402, 1835–1847 (2023). source

      The Results (Figure 2; page not supplied) report that median recurrence-free survival was not reached in either group and that atezolizumab plus bevacizumab was favoured at a median follow-up of 17.4 months (HR 0.72; p=0.012). The reported adjusted 95% CI is 0.53–0.98, not the claimed 0.56–0.93. (LLM, conf=0.94)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Most randomized trials of adjuvant therapy for hepatocellular carcinoma failed to show clinical benefit, and positive studies await validation in Western countries.

    “In summary, most of the studies failed to identify any clinical benefit and, among the positive studies, validation in trials conducted in Western countries is awaited.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [11] Haber, P. K. et al. Evidence-based management of hepatocellular carcinoma: systematic review and meta-analysis of randomized controlled trials (2002–2020). Gastroenterology 161, 879–898 (2021). source

      The Results (Adjuvant Treatments; page unavailable) report that only one of seven high-quality adjuvant trials met its primary endpoint, supporting the claim that most did not show benefit. The positive trial was conducted in Korea, and the Discussion notes important caveats, but our copy does not state that validation in Western-country trials is awaited. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    B cells have a dual role in hepatocellular carcinoma immunobiology, promoting tumour development while also enhancing immunotherapy response.

    “B cells have a dual role in HCC immunobiology, promoting tumour development but also enhancing the response to immunotherapy by producing antitumour antibodies and activating T cells65.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [65] Finkin, S. et al. Ectopic lymphoid structures function as microniches for tumor progenitor cells in hepatocellular carcinoma. Nat. Immunol. 16, 1235–1244 (2015). source

      The Results section, “ELSs express high amounts of growth promoting cytokines” (page unavailable), finds that B lymphocytes express LTβ within tumor-promoting ectopic lymphoid structures, supporting a role in HCC development. The paper does not report that B cells enhance immunotherapy response by producing antitumour antibodies or activating T cells. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    In IMbrave 050, grade 3-4 adverse events occurred in 41% of patients receiving atezolizumab plus bevacizumab versus 13% with active surveillance, and 8.7% discontinued both drugs because of adverse events.

    “The incidence of grade 3–4 adverse events was 41% and 13% with intervention and active surveillance, respectively, and 8.7% of patients discontinued both atezolizumab and bevacizumab owing to adverse events.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [12] Qin, S. et al. Atezolizumab plus bevacizumab versus active surveillance in patients with resected or ablated high-risk hepatocellular carcinoma (IMbrave 050): a randomised, open-label, multicentre, phase 3 trial. Lancet 402, 1835–1847 (2023). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    In some tumour types, tumour mutational burden or neoantigen number correlates with responsiveness to immune-checkpoint inhibitors.

    “In certain tumour types, the number of neoantigens in a tumour (or tumour mutational burden (TMB)) is correlated with responsiveness to ICIs67,68.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [67] Jardim, D. L., Goodman, A., de Melo Gagliato, D. & Kurzrock, R. The challenges of tumor mutational burden as an immunotherapy biomarker. Cancer Cell 39, 154–173 (2021). source
    • cites [68] Samstein, R. M. et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat. Genet. 51, 202–206 (2019). source

      The Abstract (page not specified in our copy) reports that higher TMB was associated with better survival among patients treated with immune-checkpoint inhibitors across many cancer types. The paper proposes that higher TMB may mean more tumour neoantigens, but it does not report a direct correlation between neoantigen number and treatment response. (LLM, conf=0.85)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    MYC overexpression leads to PD-L1 overexpression, while TP53 mutations promote recruitment of immunosuppressive cells.

    “MYC overexpression leads to PD-L1 overexpression, whereas TP53 mutations promote the recruitment of immunosuppressive cells75.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [75] Xu, Y. et al. Translation control of the immune checkpoint in cancer and its therapeutic targeting. Nat. Med. 25, 301–311 (2019). source

      The Results section ('MYC reprograms PD-L1 expression at the translation level'; page unavailable) reports increased PD-L1 protein in MYC-expressing, KRAS-mutant liver tumors. The paper does not support the claim that TP53 mutations recruit immunosuppressive cells: its immune-cell findings concern MYC and KRAS, not TP53. (LLM, conf=0.90)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Patients with inflamed hepatocellular carcinomas tend to have a favourable prognosis and are the most likely to benefit from immune-checkpoint inhibitors.

    “Overall, patients with inflamed HCCs tend to have a favourable prognosis and are the most likely to have improved outcomes when receiving ICIs owing to the presence of responsive immune cells72,78.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [72] Haber, P. K. et al. Molecular markers of response to anti-PD1 therapy in advanced hepatocellular carcinoma. Gastroenterology 164, 72–88.e18 (2023). source

      The Results section, “Molecular Features of Patients With HCC Responding to Anti-PD1 in Front Line” (page unavailable), reports a higher objective-response rate for the Inflamed class than for the Intermediate and Excluded classes (7/9 versus 5/19). It does not establish a generally favourable prognosis for patients with inflamed HCC; the class’s progression-free-survival difference was a nonsignificant trend. The Discussion links immune-cell composition to response but does not establish that responsive cells cause the improved outcomes. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [78] Montironi, C. et al. Inflamed and non-inflamed classes of HCC: a revised immunogenomic classification. Gut 72, 129–140 (2023). source

      In the Results (Inflamed class; page unavailable), the Inflamed signature was higher in ICI responders than non-responders, supporting the suggestion that these tumors may benefit from ICIs. The Results (Defining inflamed and non-inflamed HCCs; page unavailable) report better survival only for the Immune Active subclass; the Immune-like subclass showed no survival difference from non-inflamed profiles, so a favourable prognosis is not established for the Inflamed class as a whole. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Randomized trials conducted in China have reported clinical benefits with neoadjuvant or adjuvant therapies, including hepatic intra-arterial FOLFOX and adjuvant TACE.

    “Conversely, RCTs conducted in China have reported clinical benefits with (neo)adjuvant therapies.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [92] Li, S.-H. et al. Postoperative adjuvant hepatic arterial infusion chemotherapy with FOLFOX in hepatocellular carcinoma with microvascular invasion: a multicenter, phase III randomized study. J. Clin. Oncol. 41, 1898–1908 (2023). source
    • cites [93] Wang, Z. et al. Adjuvant transarterial chemoembolization for HBV-related hepatocellular carcinoma after resection: a randomized controlled study. Clin. Cancer Res. 24, 2074–2081 (2018). source

  • May combine two claims.

    Premisecitation

    Late hepatocellular carcinoma recurrence after 12 months probably reflects new primary tumours and is related to the underlying liver disease.

    “By contrast, late HCC recurrence (>12 months) probably reflects new primary tumours (also known as de novo HCCs) and is related to the underlying liver disease.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [58] Xu, X.-F. et al. Risk factors, patterns, and outcomes of late recurrence after liver resection for hepatocellular carcinoma: a multicenter study from China. JAMA Surg. 154, 209–217 (2019). source

      The Abstract and Discussion associate late recurrence with de novo tumours and underlying hepatitis or cirrhosis. But the paper defines late recurrence as more than 2 years after surgery, not more than 12 months; its findings therefore do not establish the claim for recurrences between 12 and 24 months. Page numbers are not available in the supplied copy. (LLM, conf=0.90)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    In hepatocellular carcinoma, large-scale copy-number alterations cause loss of antigen-presentation genes, suggesting that these alterations shape the tumour microenvironment.

    “Indeed, in HCC, the presence of large-scale copy-number alterations results in the loss of genes involved in antigen presentation, suggesting that copy-number alterations contribute shaping of the TME70.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [70] Bassaganyas, L. et al. Copy-number alteration burden differentially impacts immune profiles and molecular features of hepatocellular carcinoma. Clin. Cancer Res. 26, 6350–6361 (2020). source

      The Results report that losses of antigen-presentation genes, including HLA-DQB1, were enriched in HCC tumors with high burdens of broad copy-number alterations. The Discussion proposes that these losses could affect the tumor immune environment, but says further studies are needed to establish causality. Thus, “results in” overstates the evidence. Page numbers were not available in our copy. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    In IMbrave 050, the most common immune-related adverse events were hepatitis, rash, and hypothyroidism, mostly grade 1-2.

    “The most common immune-related AEs (irAEs) were hepatitis (32% versus 15%), rash (20% versus 2%) and hypothyroidism (20% versus <1%), mostly of grade 1–2 (ref. 12).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [12] Qin, S. et al. Atezolizumab plus bevacizumab versus active surveillance in patients with resected or ablated high-risk hepatocellular carcinoma (IMbrave 050): a randomised, open-label, multicentre, phase 3 trial. Lancet 402, 1835–1847 (2023). source

      The Results (page not available in our copy) identify hepatitis-related events and hypothyroidism as the most common immune-mediated adverse events and say most were grade 1–2. They do not identify rash among those events or give the cited immune-related percentages. Table 2 reports rash and hypothyroidism under adverse events regardless of causality, with different percentages, so it does not verify those figures. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Inflamed tumours make up about 30% of hepatocellular carcinomas and show extensive immune infiltration, increased immune checkpoint expression, interferon signalling activation, and a low burden of large chromosomal alterations.

    “Inflamed tumours constitute ~30% of HCCs and exhibit extensive immune cell infiltration and immune activity, detected as increased expression of immune checkpoints (such as PD-1 or its ligand PD-L1), activation of IFN signalling and a low burden of large chromosomal alterations70,71,78.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [70] Bassaganyas, L. et al. Copy-number alteration burden differentially impacts immune profiles and molecular features of hepatocellular carcinoma. Clin. Cancer Res. 26, 6350–6361 (2020). source

      The Introduction (page unavailable) describes an HCC Immune Class comprising about one third of cases, with high immune infiltration and PD-1/PD-L1 expression. The Results (page unavailable) link low broad copy-number alteration burden to interferon signalling, immune infiltration, and higher checkpoint expression. But low-burden tumours account for 17% of HCCs, and the paper reports an association with the Immune Class—not that all inflamed tumours have low alteration burden. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [71] Sia, D. et al. Identification of an immune-specific class of hepatocellular carcinoma, based on molecular features. Gastroenterology 153, 812–826 (2017). source
    • cites [78] Montironi, C. et al. Inflamed and non-inflamed classes of HCC: a revised immunogenomic classification. Gut 72, 129–140 (2023). source

  • May combine two claims.

    Interpretationcitation

    The S1801 findings come from a tumour type generally responsive to immune checkpoint inhibitors, so their relevance to less-responsive tumours such as HCC remains to be demonstrated.

    “These findings are from a study in a tumour type that tends to be responsive to ICIs and thus, their relevance to less-responsive tumour types, such as HCC, remains to be demonstrated.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [9] Patel, S. P. et al. Neoadjuvant–adjuvant or adjuvant-only pembrolizumab in advanced melanoma. N. Engl. J. Med. 388, 813–823 (2023). source

  • May combine two claims.

    Interpretationcitation

    Neoadjuvant immune checkpoint inhibitors can induce a more robust immune response than adjuvant immune checkpoint inhibitors, owing to the immunosuppressive systemic effects of invasive surgery.

    “neoadjuvant ICIs can induce a more robust immune response than adjuvant ICIs, given the immunosuppressive systemic effects of invasive surgery125–127.”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [125] Lissoni, P. et al. Effects of the conventional antitumor therapies surgery, chemotherapy, radiotherapy and immunotherapy on regulatory T lymphocytes in cancer patients. Anticancer Res. 29, 1847–1852 (2009).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [126] Tang, F., Tie, Y., Tu, C. & Wei, X. Surgical trauma-induced immunosuppression in cancer: recent advances and the potential therapies. Clin. Transl. Med. 10, 199–223 (2020). source

      The Background and Conclusion and Future Prospects sections describe systemic immunosuppression after surgery and link its severity to the invasiveness of surgical trauma. The Combination therapy of immunotherapy and surgery section discusses preoperative immune stimulation and checkpoint blockade, but does not compare immune responses to neoadjuvant versus adjuvant checkpoint inhibitors. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [127] Bakos, O., Lawson, C., Rouleau, S. & Tai, L.-H. Combining surgery and immunotherapy: turning an immunosuppressive effect into a therapeutic opportunity. J. Immunother. Cancer 6, 86 (2018). source

      The Checkpoint inhibitors section reports better survival with neoadjuvant than adjuvant checkpoint treatment in two mouse models, alongside markers of antitumor immunity. The Background describes surgery-induced immunosuppression, and the authors suggest that treatment before surgery might help preserve immune activation. The paper does not establish that the neoadjuvant treatment produced a more robust immune response than the adjuvant treatment; its direct comparison is of survival. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Neoadjuvant immune checkpoint inhibitor efficacy and safety are firmly established in multiple tumour types, with FDA approvals for NSCLC and triple-negative breast cancer.

    “Since results from one of the first clinical trials of such a therapeutic approach in patients with NSCLC were reported in 2018 (ref. 19), the efficacy and safety of neoadjuvant ICIs have been firmly established in multiple tumour types, with FDA approvals for two indications (NSCLC20,120 and triple-negative breast cancer8).”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites [19] Yang, X. et al. Neoadjuvant PD-1 blockade in resectable lung cancer. N. Engl. J. Med. 379, e14 (2018). source
    • cites [20] Forde, P. M. et al. Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer. N. Engl. J. Med. 386, 1973–1985 (2022). source
    • cites [120] Wakelee, H. et al. Perioperative pembrolizumab for early-stage non-small-cell lung cancer. N. Engl. J. Med. 389, 491–503 (2023). source

  • May combine two claims.

    Interpretationcitation

    The three initial HCC neoadjuvant studies demonstrate the utility of the window-of-opportunity setting and provide mechanistic insight because resected tumour tissue allows in-depth immune analysis instead of relying on scant on-treatment biopsies.

    “These three initial studies highlight the utility of the neoadjuvant ‘window-of-opportunity’ setting and provide insight on the mechanisms of action of ICIs, given the potential to use resected tumour tissue for in-depth immune analysis rather than relying on scant on-treatment biopsy samples128.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [128] Marron, T. U. et al. Neoadjuvant clinical trials provide a window of opportunity for cancer drug discovery. Nat. Med. 28, 626–629 (2022). source

      Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured

  • May combine two claims. The quote only partly states this claim.

    Premisecitation

    In a trial of neoadjuvant cemiplimab in HCC resection candidates, patients received two doses and underwent resection as early as 22 days after starting therapy; 20% of the 20 resected patients had significant tumour necrosis.

    “Out of the 20 patients who underwent resection, 20% had significant tumour necrosis.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [13] Marron, T. U. et al. Neoadjuvant cemiplimab for resectable hepatocellular carcinoma: a single-arm, open-label, phase 2 trial. Lancet Gastroenterol. Hepatol. 7, 219–229 (2022). source

  • May combine two claims.

    Premisecitation

    In the neoadjuvant cemiplimab trial, responsive tumours had high T cell infiltration, but some non-responsive HCCs also had T cell enrichment.

    “In the trial of neoadjuvant cemiplimab82, the analysis of samples taken at the time of resection (after treatment) revealed that responsive tumours had high levels of T cell infiltration, although some non-responsive HCCs also had T cell enrichment (Fig. 2).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [82] Magen, A. et al. Intratumoral dendritic cell–CD4+ T helper cell niches enable CD8+ T cell differentiation following PD-1 blockade in hepatocellular carcinoma. Nat. Med. 29, 1389–1399 (2023). source

      The Results report that all treatment-responsive lesions and 40% of nonresponsive lesions were highly infiltrated with T cells after resection. However, this analysis included 20 patients treated with cemiplimab in the trial and nine treated with nivolumab off trial, so the paper does not establish the stated finding for cemiplimab-trial samples alone. The indexed copy provides no page number for this passage. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Cancer-prevention vaccines against human papillomavirus and HBV have greatly reduced the incidence of certain virally driven cancers, including HBV-related HCC.

    “Cancer-prevention vaccines, such as those for human papillomavirus and HBV, have greatly reduced the incidence or certain virally driven cancers, including HBV-related HCC130.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [130] Schiller, J. T. & Lowy, D. R. Vaccines to prevent infections by oncoviruses. Annu. Rev. Microbiol. 64, 23–41 (2010). source

      The Hepatitis B Virus section documents a reduction in childhood HCC incidence in Taiwan following universal HBV vaccination. It also says HBV was the only oncovirus vaccine shown to prevent human cancer; the cited paper does not establish that HPV vaccination had already reduced cancer incidence. (LLM, conf=0.90)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    In mouse models of colon and prostate cancer, combined anti-CTLA4 and anti-PD-1 antibodies in the low tumour burden state after primary tumour resection improved control of established tumours but compromised antitumour immunity.

    “in mouse models of colon and prostate cancer, combined administration of anti-CTLA4 and anti-PD-1 antibodies in the low tumour burden state (following resection of the primary tumour) provided improved control of established tumours but compromised antitumour immunity109.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [109] Pai, C.-C. S. et al. Clonal deletion of tumor-specific T cells by interferon-γ confers therapeutic resistance to combination immune checkpoint blockade. Immunity 50, 477–492.e8 (2019). source

      The Results section on low tumour burden reports that adding anti-PD-1 to anti-CTLA-4 compromised tumour control, and the section on memory responses reports weaker protection on tumour re-challenge. The cited experiments treated mice after tumour implantation, not after primary-tumour resection. Improved control with the combination is therefore not supported in the low-burden setting described by the claim. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Long-term PD-1 blockade predisposes patients to a higher risk of immune-related adverse events and increased financial burden without a proven effect on outcomes.

    “long-term PD-1 blockade predisposes patients to a higher risk of irAEs and increased financial burden without a proven effect on outcomes129.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [129] Marron, T. U. et al. Considerations for treatment duration in responders to immune checkpoint inhibitors. J. Immunother. Cancer 9, e001901 (2021). source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    HCC trial design guidelines recommend mRECIST for assessing responses in early and intermediate-stage HCC, and both RECIST and mRECIST for systemic therapies in advanced-stage disease.

    “guidelines on trial design for patients with HCC119 recommend the use of modified Response Evaluation Criteria In Solid Tumors (mRECIST) for assessing responses to therapies in patients with early stage and intermediate-stage HCC, and both RECIST and mRECIST to evaluate responses to systemic therapies in the advanced-stage setting.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [119] Llovet, J. M. et al. Trial design and endpoints in hepatocellular carcinoma: AASLD consensus conference. Hepatology 73, 158–191 (2021). source

      The paper’s 'Overall assessment of tumor response by RECIST' section lists RECIST 1.1 or mRECIST for assessing tumor response. The passages found do not establish the claimed stage-specific recommendations: mRECIST for early- and intermediate-stage HCC, and both criteria for advanced-stage systemic therapies. (LLM, conf=0.72)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Atezolizumab plus bevacizumab, durvalumab plus tremelimumab, lenvatinib, and sorafenib are globally approved for first-line advanced HCC, but patients in the approval trials had not received prior adjuvant systemic therapy.

    “At present, atezolizumab plus bevacizumab4, durvalumab plus tremelimumab148, lenvatinib149 and sorafenib3 are approved globally for the first-line treatment of advanced-stage HCC, although clearly no patients in the studies that led to these approvals received previous adjuvant systemic therapy1,2,6.”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [4] Finn, R. S. et al. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N. Engl. J. Med. 382, 1894–1905 (2020). source

      The Patients section (p. not available) says participants in the atezolizumab–bevacizumab trial had not previously received systemic therapy for liver cancer, supporting the statement about prior adjuvant systemic therapy for this trial. The paper establishes that this combination was studied as first-line treatment, but our copy does not establish its global approval. The other regimens have separate citations in the citing sentence and are not attributed to this paper. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [148] Abou-Alfa, G. K. et al. Phase 3 randomized, open-label, multicenter study of tremelimumab (T) and durvalumab (D) as first-line therapy in patients (pts) with unresectable hepatocellular carcinoma (uHCC): HIMALAYA. J. Clin. Oncol. 40, 379 (2022). source
    • cites [149] Kudo, M. et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet 391, 1163–1173 (2018). source

      The Background (p. 5) identifies this as a first-line lenvatinib trial, and the Discussion (p. 16) reports noninferior overall survival versus sorafenib. Study Eligibility (p. 8) says patients with prior systemic therapy for HCC were excluded. The paper does not establish the citing sentence’s later claim that lenvatinib was approved globally; citation 149 is attached to lenvatinib, not to the other named regimens. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    mRNA-based cancer vaccines are typically given alone or with immune checkpoint inhibitors after resection, though they have also been used in neoadjuvant-adjuvant approaches.

    “Cancer mRNA-based vaccines tend to be administered alone or in combination with ICIs, and after resection, although they have also been used in neoadjuvant–adjuvant treatment approaches136,138,139 (Fig. 3c,d).”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites [136] Rojas, L. A. et al. Personalized RNA neoantigen vaccines stimulate T cells in pancreatic cancer. Nature 618, 144–150 (2023). source

      The Main and Discussion describe an mRNA vaccine given after surgery for pancreatic cancer, in combination with the checkpoint inhibitor atezolizumab. This trial supports that treatment approach as an example, but it does not establish that cancer mRNA vaccines are typically administered this way or document a neoadjuvant–adjuvant approach. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [138] Ott, P. A. et al. An immunogenic personal neoantigen vaccine for patients with melanoma. Nature 547, 217–221 (2017). source

      The Abstract (page unavailable) describes vaccination after surgical resection and reports that two patients later received anti-PD-1 therapy. But this was a long-peptide vaccine, not an mRNA vaccine; the paper does not establish typical practice for mRNA vaccines or describe a neoadjuvant–adjuvant approach. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [139] mRNA vaccine slows melanoma recurrence. Cancer Discov. 13, 1278 (2023). source

  • May combine two claims.

    Premisecitation

    Large intrahepatic tumour burden is associated with a lower likelihood of objective response and a higher risk of liver injury from embolic therapies.

    “those with a large intrahepatic tumour burden (such as bilobar multifocal disease) might be considered unsuitable for TACE and therefore as candidates for systemic therapies, given a lower likelihood of objective responses and a higher risk of liver injury with embolic therapies147.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [147] Kudo, M. et al. Lenvatinib as an initial treatment in patients with intermediate-stage hepatocellular carcinoma beyond up-to-seven criteria and Child–Pugh a liver function: a proof-of-concept study. Cancers 11, 1084 (2019). source

  • May combine two claims.

    Interpretationcitation

    The estimated duration of neoadjuvant approaches is around 6-8 weeks, though longer time frames might be justified based on a regimen's clinical activity.

    “The estimated duration of neoadjuvant approaches — aimed to achieve a balance between exposure to ICIs and prevention of tumour progression — is estimated to be around 6–8 weeks, but longer time frames might be justified based on clinical activity of a regimen13,15.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [13] Marron, T. U. et al. Neoadjuvant cemiplimab for resectable hepatocellular carcinoma: a single-arm, open-label, phase 2 trial. Lancet Gastroenterol. Hepatol. 7, 219–229 (2022). source

      The Discussion (page unavailable) supports balancing treatment exposure against the risks of delaying surgery: it says longer preoperative therapy can increase toxicity and the risk of disease progression. The Methods describe two cemiplimab doses three weeks apart, but the paper does not estimate a general 6–8-week duration or say that a regimen’s clinical activity justifies longer treatment; it says the optimal duration remains to be identified. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [15] Kaseb, A. O. et al. Perioperative nivolumab monotherapy versus nivolumab plus ipilimumab in resectable hepatocellular carcinoma: a randomised, open-label, phase 2 trial. Lancet Gastroenterol. Hepatol. 7, 208–218 (2022). source

      The Methods (Procedures; page not available) schedule surgery at week 6, and the Discussion (page not available) suggests that some patients might have developed a response with continued treatment. The paper does not estimate a general 6–8-week duration for neoadjuvant approaches or establish that this duration balances ICI exposure against tumour progression. (LLM, conf=0.85)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    In a salvage liver transplantation study of 110 patients, 5-year overall survival in the intention-to-treat population was 69%, and 55% achieved cure after resection or successful salvage transplantation.

    “In a salvage liver transplantation study involving 110 patients, the 5-year OS in the intention-to-treat population was 69%, with 55% of patients achieving cure after resection or successful salvage liver transplantation144.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [144] de Haas, R. J. et al. Curative salvage liver transplantation in patients with cirrhosis and hepatocellular carcinoma: an intention-to-treat analysis. Hepatology 67, 204–215 (2018). source

      The Study Population section reports 110 enrolled patients, and Long-Term Outcomes reports 69% five-year overall survival from initial resection. But Success of the SLT Strategy reports cure in 60 of 108 patients (56%) after excluding two rescue transplants, rather than the cited 55% of 110. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Recurrence after resection or local ablation remains substantial, at 50-70% by 5 years, with the highest risk in the first 12 months.

    “The likelihood of disease recurrence after resection or local ablation remains substantial, ranging from 50% to 70% at 5 years (Table 1); the risk of recurrence is highest during the first 12 months following curative treatment54.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [54] Tabrizian, P., Jibara, G., Shrager, B., Schwartz, M. & Roayaie, S. Recurrence of hepatocellular cancer after resection: patterns, treatments, and prognosis. Ann. Surg. 261, 947–955 (2015). source

  • May combine two claims.

    Interpretationreasoning

    Interactions among T cells, antigen-presenting cells, and tumour cells are more likely when a large primary tumour burden is present, providing a mechanistic rationale for why neoadjuvant immunotherapies might be preferable to adjuvant immunotherapies.

    “Such interactions are more likely to occur when a large burden of primary tumour (containing the antigens targeted by the immune system) is still present, providing a potential mechanistic rationale for why neoadjuvant immunotherapies might be preferable to adjuvant immunotherapies (Fig. 3).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Antitumour immune responses with immunotherapy depend on interactions among T cells, antigen-presenting cells, and tumour cells.

  • May combine two claims.

    Premisecitation

    Analyses of samples from IMbrave 150 identified molecular predictors of improved outcomes, including CD8+ T cell density, effector T cell signature, and high PD-L1 expression, and of inferior outcomes, including a high Treg:Teff cell ratio, with atezolizumab plus bevacizumab.

    “Similarly, analyses of samples from patients enrolled in IMbrave 150 led to the identification of molecular predictors of improved outcomes (CD8+ T cell density, effector T (Teff) cell signature and high expression of PD-L1) and also inferior outcomes (high Treg:Teff cell ratio) with atezolizumab plus bevacizumab69.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [69] Zhu, A. X. et al. Molecular correlates of clinical response and resistance to atezolizumab in combination with bevacizumab in advanced hepatocellular carcinoma. Nat. Med. 28, 1599–1611 (2022). source

  • May combine two claims.

    Interpretationreasoning

    Adjuvant immune checkpoint inhibitors stimulate antitumour immunity against micrometastases after primary tumour removal, whereas neoadjuvant immunotherapies use the primary tumour as an antigen source; in both settings, micrometastases can eventually cause recurrence.

    “Adjuvant ICIs stimulate antitumour immunity against micrometastases after the primary tumour is removed, whereas neoadjuvant immunotherapies use the primary tumour as a source of antigens to stimulate such responses; in both situations, those micrometastases can eventually lead to disease recurrence.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Antitumour immune responses with immunotherapy depend on interactions among T cells, antigen-presenting cells, and tumour cells.

  • May combine two claims.

    Premisecitation

    The phase III STORM trial compared sorafenib with placebo after resection or local ablation in 1,114 patients with hepatocellular carcinoma and did not improve recurrence-free survival (33.3 versus 33.7 months).

    “Similarly, the phase III STORM trial compared sorafenib versus placebo after resection or local ablation in 1,114 patients with HCC and did not show an improvement in RFS (33.3 versus 33.7 months)90.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [90] Bruix, J. et al. Adjuvant sorafenib for hepatocellular carcinoma after resection or ablation (STORM): a phase 3, randomised, double-blind, placebo-controlled trial. Lancet Oncol. 16, 1344–1354 (2015). source

  • May combine two claims.

    Premisecitation

    In the neoadjuvant setting, with the primary tumour present, immune checkpoint inhibitors more efficiently promote de novo T cell-mediated immunity, expansion of pre-existing antitumour T cells, and a more diverse tumour-specific T cell repertoire than after tumour removal.

    “ICIs can promote de novo induction of T cell-mediated immunity, expansion of pre-existing antitumour T cells and development of a more diverse tumour-specific T cell repertoire more efficiently than after tumour removal (adjuvant setting)105 (Fig. 3a,b).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [105] Lucas, M. W., Versluis, J. M., Rozeman, E. A. & Blank, C. U. Personalizing neoadjuvant immune-checkpoint inhibition in patients with melanoma. Nat. Rev. Clin. Oncol. 20, 408–422 (2023). source

  • May combine two claims.

    Premisecitation

    In a phase I study of neoadjuvant nivolumab plus cabozantinib in locally advanced HCC, 12 of 15 enrolled patients underwent resection and 5 had major pathological response defined as ≥90% tumour necrosis.

    “Out of the 15 patients enrolled, 12 successfully underwent resection and 5 had a major pathological response defined as ≥90% tumour necrosis.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [14] Ho, W. J. et al. Neoadjuvant cabozantinib and nivolumab converts locally advanced HCC into resectable disease with enhanced antitumor immunity. Nat. Cancer 2, 891–903 (2021).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    In melanoma and breast cancer, pathological response is superior to radiological assessment and is an accepted surrogate for recurrence-free survival.

    “In certain tumour types, such as melanoma and breast cancer, pathological response has been proven to be superior to radiological assessment115,116 and is currently an accepted surrogate for RFS.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [115] Rozeman, E. A. et al. Identification of the optimal combination dosing schedule of neoadjuvant ipilimumab plus nivolumab in macroscopic stage III melanoma (OpACIN-neo): a multicentre, phase 2, randomised, controlled trial. Lancet Oncol. 20, 948–960 (2019). source
    • cites [116] von Minckwitz, G. et al. Definition and impact of pathologic complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J. Clin. Oncol. 30, 1796–1804 (2012). source

  • May combine two claims.

    Premisecitation

    In that breast cancer model, giving immune checkpoint inhibitors after surgery enabled T cell expansion and produced the best antitumour activity while reducing toxicity.

    “Delivering ICIs after surgery enabled T cell expansion, and yielded the best antitumour activity while also decreasing toxicity106.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [106] Liu, J. et al. Improved efficacy of neoadjuvant compared to adjuvant immunotherapy to eradicate metastatic disease. Cancer Discov. 6, 1382–1399 (2016). source

  • May combine two claims.

    Interpretationreasoning

    Endpoints in neoadjuvant HCC trials were defined arbitrarily because of the paucity of data, underscoring the need for more comprehensive studies to establish appropriate surrogate endpoints.

    “All of these end points were defined arbitrarily owing to the paucity of data from studies testing neoadjuvant approaches in HCC, underscoring the need for more comprehensive studies to establish appropriate surrogate end points.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    The T cell-rich infiltration pattern is the most favourable for response to immune checkpoint inhibitors, but it is not a definitive predictor of response.

    “Although the T cell-rich infiltration pattern is the most favourable in terms of response to ICIs, this pattern is not a definitive predictor of response.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    In NSCLC, histopathological features of response to neoadjuvant immune checkpoint inhibitors include increased lymphocyte and macrophage infiltration, tertiary lymphoid structures, proliferative fibrosis, and neovascularization.

    “The histopathological features of response to neoadjuvant ICIs in patients with NSCLC have also been described111. These features include increased infiltration of lymphocytes and macrophages in tumours, presence of tertiary lymphoid structures, proliferative fibrosis and neovascularization.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [111] Cottrell, T. R. et al. Pathologic features of response to neoadjuvant anti-PD-1 in resected non-small-cell lung carcinoma: a proposal for quantitative immune-related pathologic response criteria (irPRC). Ann. Oncol. 29, 1853–1860 (2018). source

  • May combine two claims.

    Premisecitation

    In melanoma patients treated with neoadjuvant BRAF-MEK inhibitors, complete responders had significantly higher 1-year and 2-year recurrence-free survival than those without pathological complete response (88% vs 63% and 79% vs 13%), with no significant difference between pathological partial response and non-response.

    “Complete responders had a significantly higher 1-year and 2-year RFS than those without a pCR (88% versus 63% and 79% versus 13%, respectively, although no significant difference in RFS was observed between patients with a pathological partial response and those with a pathological non-response)105.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [105] Lucas, M. W., Versluis, J. M., Rozeman, E. A. & Blank, C. U. Personalizing neoadjuvant immune-checkpoint inhibition in patients with melanoma. Nat. Rev. Clin. Oncol. 20, 408–422 (2023). source

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    HCC recurrence after resection has a bimodal distribution.

    “Given that HCC recurrence after resection has a bimodal distribution”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: HCC recurrence after resection has a bimodal distribution.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    An alternative pathway for biomarker-based approval of immunotherapies has been proposed.

    “an alternative pathway for biomarker-based approval of immunotherapies has been proposed”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [102] Llovet, J. M. Exploring a new pathway for biomarker-based approval of immunotherapies. Nat. Rev. Clin. Oncol. 20, 279–280 (2023). source

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    Demonstration of a role for an ICI-based regimen in the adjuvant setting opens the door to studying these regimens in the neoadjuvant setting before curative locoregional approaches.

    “Now the demonstration of a role for an ICI-based regimen in the adjuvant setting opens the door to studies evaluating these regimens in the neoadjuvant setting before locoregional curative approaches.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Whether adjuvant systemic therapy improves overall survival in HCC remains to be established.

    “whether or not the use of systemic therapy in the adjuvant setting improves OS remains to be established”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    Limited data guide the best choice after recurrence on adjuvant ICI-based therapy, but many clinicians would probably favour other approved first-line therapies.

    “limited data are available to guide the ‘best choice’ in such a situation but many clinicians would probably favour other approved first-line therapies.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationnone given

    Treatments after progression on adjuvant atezolizumab plus bevacizumab are debated and have not yet been tested in clinical trials.

    “subsequent treatments for patients with disease progression during or after adjuvant atezolizumab plus bevacizumab are under debate and have not yet been tested in clinical trials.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    HCC management has improved markedly since the first BCLC classification was proposed in 1999.

    “The management of patients with HCC has improved markedly since the first BCLC classification was proposed in 1999143.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [143] Llovet, J. M., Brú, C. & Bruix, J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin. Liver Dis. 19, 329–338 (1999). source

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Disease recurrence in this setting is incurable.

    “Given that disease recurrence in this setting is incurable, local ablative therapies can be considered only in certain situations such as patients with oligometastatic disease and recurrence after a very long disease-free interval (years).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    In KEYNOTE-942, resected high-risk stage III-IV melanoma patients receiving pembrolizumab plus mRNA-4157 had improved recurrence-free survival versus pembrolizumab alone (78.6% vs 62.2% at 18 months; HR 0.56, 95% CI 0.31-1.02), with no major toxicity increase.

    “This approach resulted in improved RFS compared with pembrolizumab alone (78.6% versus 62.2% at 18 months; HR 0.56, 95% CI 0.31–1.02), with no major increases in toxicity140.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [140] Weber et al. Individualised neoantigen therapy mRNA-4157 (V940) plus pembrolizumab versus pembrolizumab monotherapy in resected melanoma (KEYNOTE-942): a randomised, phase 2b study. Lancet 403, 632–644 (2024). source

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    Some patients with HCC are cured by resection or local ablation alone.

    “some patients will be cured with resection or local ablation alone”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Some patients with HCC are cured by resection or local ablation alone.
57 of 57

  • The paper relies on this without support.

    A recurrence-free survival improvement is sufficient to regard adjuvant atezolizumab plus bevacizumab as a major advance in hepatocellular carcinoma management, even before overall survival benefit is shown.

    “This result is a major advance in the management of HCC7.”
    “Whether this RFS benefit translates into an improvement in OS (a key secondary end point of the trial) remains to be determined.”

  • The paper relies on this without support.

    The immune microenvironment features and immunotherapy principles described for hepatocellular carcinoma apply across tumour stages, including the early-stage setting.

    “Currently, the main immunotherapeutic approach for patients with HCC — regardless of tumour stage — involves restoring antitumour immunity with ICIs, which are monoclonal antibodies that block inhibitory checkpoints or their ligands5.”

  • The authors state this as an assumption.

    Early recurrences after resection are caused by occult micrometastases that were already present at the time of resection.

    “which are presumed to be related to occult micrometastases already present at the time of resection.”

  • The paper relies on this without support.

    Therapeutic cancer vaccines have largely failed to improve outcomes despite decades of research.

    “despite decades of intensive research efforts, vaccines designed to treat cancer have largely failed to improve outcomes for patients with cancer.”

  • The paper relies on this without support.

    Micrometastases present during adjuvant therapy are less immunogenic than macroscopically detectable lesions.

    “micrometastases, which can be present during adjuvant therapy, are believed to be less immunogenic than macroscopically detectable lesions.”

  • The paper relies on this without support.

    Antitumour immune responses with immunotherapy depend on interactions among T cells, antigen-presenting cells, and tumour cells.

    “Antitumour immune responses with immunotherapy depend on interactions among T cells, antigen-presenting cells and tumour cells.”

  • The paper relies on this without support.

    Most anti-PD-(L)1 antibodies have a half-life longer than 3 weeks and remain in circulation well into postoperative recovery.

    “most anti-PD-(L)1 antibodies have a half-life of >3 weeks and remain in circulation well into postoperative recuperation.”

  • The paper relies on this without support.

    Some patients with HCC are cured by resection or local ablation alone.

    “some patients will be cured with resection or local ablation alone”

  • The paper relies on this without support.

    HCC recurrence after resection has a bimodal distribution.

    “HCC recurrence after resection has a bimodal distribution”

  • The paper relies on this without support.

    No companion biomarker identifying HCC subgroups most likely to benefit from immune checkpoint inhibitors has been approved.

    “no companion biomarker that enables identifying subgroups of patients with HCC who are most likely to benefit from these therapies has been approved.”
10 of 10

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Opposite claims about when ICIs worked best in the same mouse study

    Minor

    The text says a breast-cancer model showed that giving ICIs before surgery might outperform giving them afterward. Its next account of that study says administration after surgery yielded the best antitumour activity.

  • HBV antivirals are assigned opposite recurrence effects

    Minor

    The paper credits effective antiviral therapy with reducing recurrence, then says HBV antivirals after resection decrease recurrence-free survival relative to no antiviral therapy. A decrease in recurrence-free survival points in the opposite direction.

  • Major pathological response uses two different 90% cut-offs

    Minor

    The cabozantinib-nivolumab trial's response threshold is given once as more than 90% tumour necrosis and later as at least 90%. The definitions differ for a tumour with exactly 90% necrosis.

  • A later claim overlooks a positive phase III adjuvant trial

    Minor

    The paper says no other treatment has improved recurrence-free survival in a phase III study, but earlier reports a statistically significant improvement in a phase III trial of adjuvant cytokine-induced killer cell therapy.

4 of 4

Cites

Flagged first, then by citations

8 of the 165 references couldn’t be matched to a known paper.

Research works printed but not linked: 1

  1. Ho, W. J. et al. Neoadjuvant cabozantinib and nivolumab converts locally advanced HCC into resectable disease with enhanced antitumor immunity. Nat. Cancer 2, 891–903 (2021).

Cited by

Sorted by citations

Related

Authors
Llovet, J. M., Pinyol, R., Yarchoan, M., Singal, A. G., Marron, T. U., Schwartz, M., Pikarsky, E., Kudo, M., & Finn, R. S.
Affiliations
  • Institució Catalana de Recerca i Estudis Avançats (ES)
  • Hospital Clínic de Barcelona (ES)
  • Johns Hopkins University (US)
  • The University of Texas Southwestern Medical Center (US)
  • Tisch Cancer Institute (US)
  • Icahn School of Medicine at Mount Sinai (US)
  • Hebrew University of Jerusalem (IL)
  • Kindai University Hospital (JP)
  • University of California, Los Angeles (US)
Corresponding author
Llovet, Josep M.
ORCID
Venue
Nature Reviews Clinical Oncology
Publisher
Springer Science and Business Media LLC
Published
2024-02-29
Volume
21
Issue
4
Pages
294-311
ISSN
  • 1759-4774 (ISSN-L)
  • 1759-4782
DOI
10.1038/s41571-024-00868-0
PMC
PMC11984461
OpenAlex
W4392298264
Cited by
391
References
165
Citations by year
2026: 141 · 2025: 187 · 2024: 63
FWCI
28.16
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Reviews Clinical Oncology (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion, cc-by)
  • Dipòsit Digital de la Universitat de Barcelona (Universitat de Barcelona) (acceptedVersion)
  • Dipòsit Digital de la Universitat de Barcelona (Universitat de Barcelona) (submittedVersion, public-domain)
Funders
  • Cancer Research UK (C9380/A268-13, A26813, 26813)
  • European Commission (101136622)
  • Fundación Científica Asociación Española Contra el Cáncer (C9380/A26813, PRYGN223117LLOV)
  • Generalitat de Catalunya
  • Ministerio de Ciencia e Innovación (PID2022)
  • Associazione Italiana per la Ricerca sul Cancro (A26813, C9380/A26813)
  • National Institutes of Health (RO1DK56621)
  • HORIZON EUROPE Framework Programme
  • Japan Society for the Promotion of Science (21K07950)
  • Agència de Gestió d'Ajuts Universitaris i de Recerca
  • National Cancer Institute (R01CA273932, R01CA285544)
  • National Institute of Diabetes and Digestive and Kidney Diseases (R01DK056621, R01DK128289)
Topics
  • Health Sciences › Medicine › Hepatology › Hepatocellular Carcinoma Treatment and Prognosis
  • Health Sciences › Medicine › Oncology › Cancer Immunotherapy and Biomarkers
  • Health Sciences › Medicine › Epidemiology › Hepatitis B Virus Studies
Keywords
hepatocellular carcinoma, adjuvant immunotherapy, neoadjuvant immunotherapy, immune checkpoint inhibitors, atezolizumab, bevacizumab, early-stage liver cancer, tumor immune microenvironment, cancer vaccines, recurrence, hepatectomy, liver transplantation, local ablation, biomarkers, clinical trial design
MeSH terms
Carcinoma, Hepatocellular, Humans, Immunotherapy, Liver Neoplasms, Neoplasm Recurrence, Local, Neoadjuvant Therapy, Tumor Microenvironment

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-09-28.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-09-28.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review