INTRODUCTION

Combined hepatocellular-cholangiocarcinoma (cHCC-CCA) is a primary liver malignancy characterized by the unequivocal coexistence of hepatocellular carcinoma (HCC) and cholangiocarcinoma (CCA) components within a single tumor. This rare entity accounts for 0.4-5.0% of all primary liver cancers, with an estimated incidence of approximately 0.05 per 100,000 per year.^1-3 The 2019 World Health Organization (WHO) classification of digestive system tumours substantially revised the diagnostic framework for cHCC-CCA, eliminating the previously recognized subtypes with stem cell features and introducing more stringent diagnostic criteria requiring unequivocal presence of both differentiation lineages.^1,4

Although the precise etiologic drivers unique to cHCC-CCA remain unclear, this tumor entity shares common risk factors with HCC, most notably chronic hepatitis B virus (HBV) and hepatitis C virus (HCV) infection and liver cirrhosis.^2,3,5 Cirrhosis has been reported in 36-68% of cHCC-CCA patients across published series, and chronic viral hepatitis is present in the majority of cases, particularly in East Asian cohorts where HBV infection predominates.^3,6 In Western populations, HCV infection and metabolic dysfunction-associated steatotic liver disease (MASLD)/metabolic dysfunction-associated steatohepatitis (MASH) represent more prevalent underlying etiologies.^2,5 These regional differences in etiologic profiles have implications for interpreting heterogeneity across studies, as the underlying liver disease background may influence tumor biology, molecular subtype distribution, and treatment tolerance. Additional reported risk factors include alcohol-related liver disease, primary sclerosing cholangitis, and liver fluke infection, though their specific contribution to cHCC-CCA carcinogenesis as distinct from pure HCC or CCA remains to be elucidated.^2,6

The fundamental therapeutic challenge in cHCC-CCA arises from its dual nature. HCC and intrahepatic CCA (iCCA) are treated with entirely different systemic regimens: atezolizumab plus bevacizumab or durvalumab plus tremelimumab for HCC, and gemcitabine plus cisplatin (GemCis) with or without durvalumab or pembrolizumab for biliary tract cancer (BTC).^7-10 Critically, patients with cHCC-CCA have been excluded from virtually all landmark clinical trials for both HCC (SHARP, REFLECT, IMbrave150, HIMALAYA) and CCA (ABC-02, TOPAZ-1), with the sole exception of KEYNOTE-966 which included only 13 cHCC-CCA patients without separate subgroup analysis.^9,11

Recent advances in comprehensive genomic profiling have demonstrated that cHCC-CCA exists along a molecular continuum, with individual tumors classifiable as HCC-like, CCA-like, or ambiguous based on their genomic features.^12-14 This molecular heterogeneity provides a rational basis for personalized treatment selection, but also explains the variable responses observed across different regimens in retrospective studies. In this review, we comprehensively evaluate the current evidence on systemic chemotherapy, immunotherapy, and molecularly targeted therapy for advanced cHCC-CCA, and propose an emerging framework for molecular classification-guided treatment.

EVOLUTION OF CLASSIFICATION AND ITS THERAPEUTIC IMPLICATIONS

The 2019 WHO reclassification

The 5th edition WHO classification (2019) introduced significant changes to the definition of cHCC-CCA.¹ Most notably, the stem cell feature subtypes (typical, intermediate-cell, and cholangiolocellular) recognized in the 4th edition (2010) were eliminated. Under the current framework, cHCC-CCA requires the unequivocal presence of both hepatocytic and cholangiocytic differentiation within the same tumor, confirmed by morphology on hematoxylin and eosin staining and supported by immunohistochemistry using hepatocytic markers (HepPar-1, ARG1, GPC3) and cholangiocytic markers (CK7, CK19).^1,4,15 The cholangiolocellular carcinoma subtype was reassigned under small duct iCCA unless an unequivocal HCC component coexists. The intermediate cell carcinoma, characterized by monotonous cells with features between hepatocytes and cholangiocytes, was newly recognized.^1,2

These reclassification changes carry direct implications for treatment. By sharpening diagnostic criteria, the current WHO definition likely selects a more molecularly homogeneous population than older classifications. However, retrospective studies spanning the classification change must be interpreted cautiously, as patient populations may differ depending on which criteria were applied.^2,5,15

An important consideration when interpreting the existing literature is the diagnostic heterogeneity across studies. Under current WHO criteria, a definitive diagnosis of cHCC-CCA requires histopathological confirmation demonstrating both hepatocytic and cholangiocytic differentiation.^1,15 However, some earlier cohorts included patients diagnosed based on radiological criteria without mandatory histopathological confirmation,¹⁶ potentially allowing inclusion of patients with pure HCC or iCCA misclassified as cHCC-CCA. Furthermore, several studies span different classification eras, enrolling patients diagnosed under pre-2019 WHO criteria that included the now-eliminated stem cell feature subtypes.^1,15 These diagnostic inconsistencies may have influenced patient selection and reported outcomes, and should be carefully considered when comparing results across studies.

Diagnostic biomarkers

Nestin has emerged as a promising diagnostic biomarker. Calderaro et al.¹⁷ demonstrated that nestin immunohistochemical expression discriminates cHCC-CCA from pure HCC with an area under the curve of 0.85 (sensitivity 75%, specificity 93%). High nestin expression (>30% positive neoplastic cells) was independently associated with worse overall survival (OS) and disease-free survival. Serum alpha-fetoprotein (AFP) elevation is observed in a subset of cHCC-CCA with HCC-like molecular features, while carbohydrate antigen 19-9 (CA19-9) elevation correlates with CCA-dominant tumors. Dual elevation of both markers occurs in approximately 15% of patients and is suggestive but not diagnostic.^2,6,17

MOLECULAR LANDSCAPE AND THERAPEUTIC IMPLICATIONS

Genomic heterogeneity: the HCC-like vs. CCA-like spectrum

Comprehensive genomic profiling studies have established that cHCC-CCA is not a single molecular entity but rather a spectrum bridging HCC and iCCA. The landmark study by Xue et al.¹² profiled 133 cHCC-CCA cases, identifying TP53 (49%) and TERT promoter mutations (23%) as the most frequent alterations, and defined distinct molecular subtypes: combined-type cHCC-CCA with iCCA-like features (EpCAM, KRT19 overexpression, KRAS mutations) and mixed-type tumors with HCC-like features (AFP, GPC3, SALL4 expression).

Joseph et al.¹³ sequenced 20 cHCC-CCA and 10 iCCA cases, demonstrating that cHCC-CCA genomics are remarkably similar to HCC, even in the CCA component, with TERT promoter mutations (80%), TP53 (80%), RTK/Ras/PI3K pathway genes (55%), and cell cycle genes (40%). Critically, TERT promoter mutations were identified in both HCC and CCA components of the same tumor, supporting TERT as an early founding event. No isocitrate dehydrogenase 1/2 (IDH1/2), fibroblast growth factor receptor 2 (FGFR2), or BRCA1 associated protein 1 (BAP1) alterations, typical iCCA mutations, were detected in any cHCC-CCA.

The largest clinical genomic profiling study by Murugesan et al.¹⁴ used FoundationOne CDx (Foundation Medicine, Boston, MA, USA) on 73 cHCC-CCA, 4,975 CCA, and 1,470 HCC cases. A machine learning classifier categorized 58% of cHCC-CCA as HCC-like, 16% as CCA-like, and 26% as ambiguous. HCC-like tumors were enriched for TERT, CTNNB1, and MYC alterations, while CCA-like tumors harbored FGFR2 (25%), IDH1 (25%), and ARID1A (25%) alterations. Notably, 24.6% of all cHCC-CCA harbored potentially actionable genomic alterations, including BRCA2, ERBB2, MET, FGFR2, and IDH1.

Calderaro et al.¹⁸ applied deep learning to histopathology images from 405 cHCC-CCA cases, successfully reclassifying tumors as HCC or iCCA phenotypes. The artificial intelligence-based predictions correlated with clinical outcomes and genetic alterations, offering a practical tool for tumor reclassification that could ultimately guide treatment allocation. While this molecular classification framework provides a biologically compelling basis for differential treatment selection, it is important to note that the clinical outcome data available to date have not been stratified by molecular subtype. Consequently, the direct association between molecular phenotype (HCC-like vs. CCA-like) and differential response to specific systemic regimens remains hypothesis-generating and has yet to be prospectively validated.^12,14,18

Immune microenvironment

The most comprehensive immune profiling of cHCC-CCA was performed by Nguyen et al.¹⁹ on 96 patients, identifying two distinct immune subtypes: immune-high (IH; 57%) and immunelow (IL; 43%). IH tumors displayed overexpression of genes related to immune cell recruitment, adaptive and innate immunity, antigen presentation, and cytotoxicity. The IH subtype showed activation of gene signatures predictive of immunotherapy response in HCC and was an independent predictor of improved OS. These findings provide the strongest biological rationale for immune checkpoint inhibitor (ICI)-based therapy in cHCC-CCA and suggest that approximately 60% of patients may derive benefit from immunomodulating approaches.

Programmed death-ligand 1 (PD-L1) expression data specific to cHCC-CCA remain limited. Available case reports suggest that PD-L1 does not reliably predict immunotherapy response, as clinical responses have been observed in PD-L1-negative tumors.^20,21 Tumor mutational burden is generally low to moderate, and microsatellite instability-high (MSI-H) status is rare.¹⁴

Actionable genomic alterations

The identification of potentially targetable alterations in approximately 25% of cHCC-CCA mandates comprehensive genomic profiling for all patients with advanced disease.¹⁴ Key actionable alterations include FGFR2 fusions/rearrangements (present in 25% of CCA-like cHCC-CCA, overall to 5%), IDH1 mutations (25% of CCA-like cases), human epidermal growth factor receptor 2 (HER2)/ERBB2 amplification, and rare NTRK fusions, BRAF V600E mutations, and MSI-H status. FGFR2 expression was demonstrated in 21.3% of cHCC-CCA specimens by Sasaki et al.^22, comparable to small duct iCCA (25.7%) and significantly higher than HCC (0%).

SYSTEMIC CHEMOTHERAPY

Gemcitabine plus platinum-based regimens

Gemcitabine plus platinum-based chemotherapy represents the most extensively studied systemic regimen for cHCC-CCA. The evidence derives entirely from retrospective studies.

Salimon et al.¹⁶ reported the French AGEO multicenter experience with 30 patients receiving gemcitabine/platinum (GEMOX 60%, GEMOX plus bevacizumab 30%, GemCis 10%), demonstrating a median progression-free survival (PFS) of 9.0 months, median OS of 16.2 months, and an objective response rate (ORR) of 28.6%, the most favorable outcomes reported for chemotherapy in cHCC-CCA (Table 1). However, 27% of cases were diagnosed radiologically rather than histologically, which may have introduced selection bias.

Kobayashi et al.²³ conducted a 15-center Japanese study of 36 pathologically confirmed cHCC-CCA patients. GemCis achieved a median OS of 11.9 months vs. only 3.5 months for sorafenib (all 5-sorafenib-treated patients had progressive disease). Multivariate analysis demonstrated sorafenib was significantly inferior to platinum-containing regimens (hazard ratio [HR], 15.83; 95% confidence interval, 2.25-111.43; P=0.006).

Trikalinos et al.²⁴ reported the Washington University experience with 68 treated patients. Gemcitabine/platinum achieved a disease control rate (DCR) of 78.4%, median PFS of 8.0 months, and median OS of 11.5 months, which was significantly superior to gemcitabine/5-fluorouracil (DCR, 38.5%; P=0.014) and sorafenib (DCR, 20%; median OS, 9.6 months).

HCC-directed tyrosine kinase inhibitors

Sorafenib has shown consistently limited efficacy, with ORR of 0-10% and median OS of 3.5-10.7 months across most studies.^23-25 The notable exception is the largest single study by Kim et al.²⁶ from Asan Medical Center, which evaluated 99 patients (62 sorafenib, 37 chemotherapy) and found no statistically significant differences between sorafenib and cytotoxic chemotherapy (median OS, 10.7 vs. 10.6 months, P=0.34; ORR, 9.7% vs. 21.6%, P=0.14). This discrepancy may reflect differences in patient selection, tumor dominant phenotype, and viral hepatitis prevalence across study populations.

Lenvatinib shows more promising preliminary data. Tanabe et al.²⁷ reported a Japanese multicenter retrospective study of 14 patients receiving first-line lenvatinib, demonstrating an ORR of 42.9% and DCR of 92.9%, exceeding both sorafenib and chemotherapy in prior reports. However, these results require cautious interpretation given the small sample size and potential selection of HCC-dominant tumors.

The European multicenter cohort by Gigante et al.²⁵ compared 25 tyrosine kinase inhibitor (TKI)-treated and 54 chemotherapy-treated cHCC-CCA patients, with platinum-based chemotherapy achieving median OS of 11.9 months vs. 8.3 months for TKIs, although differences were not statistically significant after multivariate adjustment. Pomej et al.²⁸ reported the largest European multicenter cohort (101 patients, four institutions), with cytotoxic chemotherapy achieving median OS of 15.5 months vs. 5.3 months for sorafenib, a strong trend (P=0.052) that did not reach statistical significance.

Summary of chemotherapy evidence

Across all available studies, gemcitabine plus platinum-based regimens demonstrate the most consistent efficacy, with median OS ranging from 10.2 to 16.2 months and ORR of 15-29%. Sorafenib shows highly variable results (median OS, 3.5-10.7 months), with the weight of evidence suggesting inferiority to platinum-based chemotherapy in most populations. The variability across studies likely reflects the molecular heterogeneity of cHCC-CCA itself, tumors with HCC-like genomics may respond differently to HCC-directed versus CCA-directed regimens.^29-31

Notably, there is heterogeneity across studies regarding the relative efficacy of platinum-based chemotherapy versus TKIs. While some studies demonstrated significantly superior outcomes with gemcitabine/platinum regimens compared to sorafenib,^23,25 others reported no statistically significant difference between chemotherapy and TKI-based approaches.²⁶ This inconsistency may reflect differences in patient selection criteria, histopathological confirmation status, the relative proportion of HCC-dominant vs. CCA-dominant tumors within each cohort, regional variations in etiologic background, and the inherently small sample sizes of studies in this rare disease. These factors should be considered when interpreting comparative efficacy data.

IMMUNOTHERAPY

Atezolizumab plus bevacizumab

The HCC-directed combination of atezolizumab plus bevacizumab has generated the most immunotherapy data in cHCC-CCA. Gigante et al.³² published the first multicentric retrospective study evaluating this combination in 16 patients across seven French centers, reporting a first-line ORR of 33% and median OS of 13.0 months. Satake et al.²¹ described a Japanese case series of six patients with first-line atezolizumab/bevacizumab, achieving three partial responses (50%) and one stable disease. Tanabe et al.²⁷ reported seven patients receiving atezolizumab/bevacizumab with an ORR of 14.3% but a notable DCR of 100%.

Broader immune checkpoint inhibitor experience

The largest dedicated ICI cohort comes from Jang et al.,³³ who retrospectively evaluated 25 cHCC-CCA patients receiving checkpoint inhibitors (nivolumab 68%, pembrolizumab 20%, atezolizumab/bevacizumab 8%, ipilimumab/nivolumab 4%) at Asan Medical Center. The ORR was 20%, with median PFS of 3.5 months, median OS of 8.3 months, and a median duration of response of 11.6 months, suggesting that responders derive durable benefit.

The European multicenter cohort reported by Pomej et al.²⁸ included seven ICI-treated patients with an ORR of 29% and a median OS of 17.8 months, the highest median OS reported for any systemic regimen in cHCC-CCA, though limited by very small sample size. Diab et al.³⁴ reported an ORR of 83.3% in six hepatitis C virus-positive cHCC-CCA patients receiving programmed cell death protein 1 (PD-1)±cytotoxic T-lymphocyte-associated protein 4 inhibitors, although this likely reflects significant selection bias.

ICI plus chemotherapy combinations

The KEYNOTE-966 trial (pembrolizumab plus GemCis for BTC) included a small number of cHCC-CCA patients (eight in the pembrolizumab arm, five in placebo), but no separate subgroup analysis has been published.⁹ The TOPAZ-1 trial (durvalumab plus GemCis) did not explicitly include cHCC-CCA patients.⁸ Despite this, current National Comprehensive Cancer Network guidelines note that GemCis combined with an ICI is an appropriate choice for first-line therapy of cHCC-CCA, extrapolating from BTC trial data.

ICI plus TKI combinations

Emerging case-level evidence supports combined ICI/TKI approaches. Zhou et al.²⁰ reported a partial response after just two cycles of sintilimab plus lenvatinib plus nab-paclitaxel in a patient with GemCis-resistant metastatic cHCC-CCA, notably in a tumor mutational burden-low, microsatellite stable, PD-L1-negative tumor. These anecdotal reports suggest that ICI/TKI combinations may have activity beyond what individual biomarkers predict, warranting prospective evaluation.

TARGETED THERAPY

FGFR inhibitors

FGFR2 fusions and rearrangements, hallmark alterations of iCCA, are also present in a meaningful proportion of cHCC-CCA. Sasaki et al.²² demonstrated FGFR2 expression in 16/75 (21.3%) of cHCC-CCA, comparable to small duct iCCA (25.7%) and significantly higher than HCC (0%). Murugesan et al.¹⁴ found FGFR2 alterations in 25% of CCA-like cHCC-CCA cases. A case report described persistent response to pemigatinib combined with chemotherapy in a child with cHCC-CCA harboring an FGFR2-PRDM16 fusion.³⁵

Three FGFR inhibitors have been approved for FGFR2-altered iCCA: pemigatinib (ORR 35.5%, FIGHT-202 trial), futibatinib (ORR 42%, FOENIX-CCA2 trial),³⁶ and lirafugratinib (ORR 88% in FGFR inhibitor-naïve CCA in the ReFocus trial). These represent viable options for CCA-like cHCC-CCA harboring FGFR2 alterations, though dedicated data in cHCC-CCA remain extremely limited.

IDH inhibitors

IDH1 mutations are enriched in the CCA-like cHCC-CCA subset (approximately 25% of CCA-like cases) but are rare in HCC-like tumors, yielding an overall prevalence in unselected cHCC-CCA of approximately 3-5%.¹⁴ The ClarIDHy trial demonstrated PFS benefit for ivosidenib in IDH1-mutant cholangiocarcinoma (PFS, 2.7 vs. 1.4 months; HR, 0.37; adjusted OS, 10.3 vs. 5.1 months).³⁷ CCA-like cHCC-CCA cases harboring IDH1 mutations would theoretically be candidates for ivosidenib, though no published reports specifically address its use in cHCC-CCA.

HER2-directed therapy and other targets

HER2/ERBB2 amplification occurs in a subset of cHCC-CCA and represents an actionable target. Trastuzumab deruxtecan received tumor-agnostic approval by the US Food and Drug Administration for HER2 immunohistochemistry 3+ solid tumors based on DESTINY-PanTumor02 results (ORR 56.3% in BTC). MET amplification was found in 30.8% of cHCC-CCA in multiregional sequencing by Na et al.,³⁸ though MET-directed therapies remain investigational in liver cancers. CTNNB1 (Wnt/β-catenin) mutations in HCC-like cHCC-CCA may predict immune exclusion and poor immunotherapy response, an important consideration when selecting ICI-based regimens.^14,19

CLINICAL TRIALS AND FUTURE DIRECTIONS

No prospective clinical trial has been designed specifically for cHCC-CCA. The international multicentre survey by Claasen et al.¹¹ confirmed that well-defined treatment policies are lacking, with marked interdisciplinary discrepancies in management. The rarity of cHCC-CCA (incidence approximately 0.05 per 100,000/year) creates fundamental challenges for conventional randomized trial design. International collaborative registries and adaptive platform trials represent the most feasible paths toward prospective evidence generation.^2,11,31

Potentially relevant active trials include tumor-agnostic basket studies (DESTINY-PanTumor02 for HER2-positive tumors, KEYNOTE-158 for MSI-H/tumor mutational burden-high tumors), BTC-specific trials that may include CCA-like cHCC-CCA if diagnosed as iCCA, and the ReFocus trial (lirafugratinib for FGFR2 alterations). Future research priorities include: prospective validation of the HCC-like vs. CCA-like molecular classification as a treatment selection tool, dedicated cohorts within BTC or HCC trials, identification of predictive biomarkers beyond molecular subtyping, and evaluation of combination strategies (ICI plus chemotherapy, ICI plus TKI) in molecularly defined subsets.^18,19,29

Additionally, an emerging area of interest is the multimodal approach combining locoregional therapies (LRT) with systemic agents for unresectable cHCC-CCA. Although data specific to this setting remain very limited, a recent retrospective cohort study reported potential survival benefits from combining interventional treatments with immunotargeted therapy in unresectable cHCC-CCA, suggesting that LRT-systemic combinations may warrant further investigation as a therapeutic strategy. Prospective studies are needed to define the optimal combination strategies, patient selection criteria, and sequencing of locoregional and systemic therapies in this context.³⁹

PROPOSED MOLECULAR CLASSIFICATION-BASED TREATMENT ALGORITHM

While no validated treatment algorithm exists, synthesizing available molecular and clinical evidence supports the following rational framework for advanced cHCC-CCA.

First, all patients with advanced cHCC-CCA should undergo comprehensive genomic profiling to identify actionable alterations and classify tumors as HCC-like versus CCA-like.^14,18 Second, if targetable mutations are identified (FGFR2 fusion, IDH1 mutation, HER2 amplification, NTRK fusion, BRAF V600E, MSI-H), matched targeted therapy should be considered.^36,37 Third, for CCA-like tumors (enriched for FGFR2, IDH1, ARID1A, TERT/CTNNB1 wild-type), CCA-directed regimens (GemCis±durvalumab or pembrolizumab) are rationally preferred; for HCC-like tumors (enriched for TERT, CTNNB1, MYC), HCC-directed regimens (atezolizumab/bevacizumab, durvalumab/tremelimumab, or lenvatinib-based combinations) may be more appropriate.^14,19,29 Fourth, the approximately 57% of tumors classified as immune-high may derive particular benefit from ICI-based strategies, while Wnt/β-catenin-activated tumors may be immune-excluded.¹⁹

Pragmatic alternative when comprehensive genomic profiling is not feasible

When comprehensive genomic profiling is not feasible, clinicians may employ a pragmatic surrogate-based approach to estimate the dominant tumor component and guide initial treatment selection. Imaging characteristics provide valuable surrogates: hypervascular enhancement with washout patterns consistent with Liver Imaging Reporting and Data System (LI-RADS) category 5 suggest an HCC-dominant phenotype, whereas peripheral rim enhancement or mass-forming patterns consistent with LI-RADS category M favor a CCA-dominant phenotype.^2,6 Serum tumor markers complement imaging assessment, with AFP elevation favoring an HCC-dominant phenotype and CA19-9 elevation favoring a CCA-dominant phenotype.^6,17 Integrating these clinical and radiological surrogates, clinicians may reasonably select HCC-directed regimens (atezolizumab/bevacizumab or durvalumab/tremelimumab) for tumors with HCC-dominant features and CCA-directed regimens (gemcitabine/cisplatin plus durvalumab or pembrolizumab) for CCA-dominant presentations.

However, it must be emphasized that this surrogate-based approach has not been prospectively validated, and the inherent spatial and histological heterogeneity of cHCC-CCA means that imaging and serum markers may not accurately reflect the full molecular landscape of the tumor.^2,38 Comprehensive genomic profiling should therefore still be pursued whenever feasible, and the surrogate-based approach should be regarded as a pragmatic interim strategy rather than a definitive substitute for molecular characterization.

This framework, while biologically rational, requires prospective validation. The deep learning-based phenotyping approach¹⁸ offers a practical tool for tumor reclassification but remains investigational. Critically, it should be noted that the proposed molecular classification-guided treatment recommendations are based on biological plausibility and indirect evidence extrapolated from parent tumor types (HCC and iCCA), as most existing clinical outcome data for cHCC-CCA were not stratified by molecular subtype. Prospective validation with molecular subtype-stratified outcome data is essential before this framework can be adopted as standard clinical practice.

Post-progression treatment considerations

Post-progression treatment for cHCC-CCA remains poorly defined due to the absence of prospective data. Several pragmatic strategies may be considered based on expert opinion and extrapolation from HCC and BTC literature.^11,31 First, phenotype-based treatment switching represents a rational approach: patients who progress on an HCC-directed first-line regimen (e.g., atezolizumab/bevacizumab) may benefit from switching to CCA-directed therapy (e.g., gemcitabine/cisplatin-based combinations), and vice versa. This strategy is predicated on the rationale that the non-dominant component may drive acquired resistance and subsequent disease progression.²⁹ Second, re-biopsy at the time of progression should be considered when clinically feasible, as clonal evolution during treatment may shift the dominant molecular phenotype or reveal newly emergent actionable alterations.^38,40 Liquid biopsy (circulating tumor DNA analysis) offers a less invasive alternative to tissue re-biopsy and may provide complementary information about emerging resistance mechanisms and new targetable alterations, although its clinical utility in cHCC-CCA specifically remains to be validated. Third, for patients whose tumors harbor actionable genomic alterations not targeted in the first-line setting, matched targeted therapy represents a viable post-progression option.^14,36,37 Finally, enrollment in clinical trials, particularly basket studies with molecular selection criteria, should be actively pursued.¹¹

CONCLUSION

Systemic therapy for advanced cHCC-CCA stands at an inflection point. The field has evolved from empiric extrapolation to early molecular classification-guided treatment, but prospective evidence remains absent. Three key insights emerge from the current evidence. First, the molecular dichotomy of cHCC-CCA, with tumors classifiable as HCC-like or CCA-like in approximately 75% of cases, provides a rational basis for differential treatment selection that should be tested prospectively. Second, immunotherapy signals are sufficiently strong (ORR 20-33%, with durable responses) to support ICI-based approaches as a preferred strategy, particularly given that 57% of tumors display immune-high profiles predictive of checkpoint inhibitor benefit. Third, the finding that nearly one-quarter of cHCC-CCA tumors harbor actionable genomic alterations mandates comprehensive molecular profiling in all patients with advanced disease.^14,19,29

Importantly, it should be acknowledged that the molecular classification (HCC-like vs. CCA-like) framework, while biologically compelling, remains largely unvalidated in terms of treatment efficacy. Most clinical outcome data in the existing literature were not stratified by molecular subtype, and the association between molecular phenotype and differential response to specific systemic regimens is therefore hypothesis-generating rather than evidence-based. Prospective clinical studies incorporating molecular stratification in their design are essential to bridge this critical knowledge gap.

The most urgent need is for international collaborative prospective studies, whether through dedicated cHCC-CCA trials, cHCC-CCA-specific cohorts within basket trials, or adaptive platform designs, to transform this evidence base from retrospective observations to practice-changing data. Until such data become available, a rational molecular classification-based approach remains the most feasible strategy for managing this challenging malignancy.^2,11,31

Article information

Conflicts of Interest

The authors declare no competing interests.

Ethics Statement

This review article is fully based on articles which have already been published and did not involve additional patient participants. Therefore, IRB approval is not necessary.

Funding Statement

Not applicable.

Data Availability

Not applicable.

Author Contributions

Conceptualization: JYH

Data curation: JYH

Investigation: JYH, DHS, SYH

Methodology: JYH, DHS, SYH

Writing - original draft: JYH

Writing - review & editing: JYH, DHS, SYH

Approval of final manuscript: JYH, DHS, SYH

References

• 1. Nagtegaal ID, Odze RD, Klimstra D, Paradis V, Rugge M, Schirmacher P, et al. The 2019 WHO classification of tumours of the digestive system. Histopathology 2020;76:182−188.ArticlePubMedPMCPDF
• 2. Beaufrère A, Calderaro J, Paradis V. Combined hepatocellular-cholangiocarcinoma: an update. J Hepatol 2021;74:1212−1224.ArticlePubMed
• 3. Schizas D, Mastoraki A, Routsi E, Papapanou M, Tsapralis D, Vassiliu P, et al. Combined hepatocellular-cholangiocarcinoma: an update on epidemiology, classification, diagnosis and management. Hepatobiliary Pancreat Dis Int 2020;19:515−523.ArticlePubMed
• 4. Brunt E, Aishima S, Clavien PA, Fowler K, Goodman Z, Gores G, et al. cHCC-CCA: consensus terminology for primary liver carcinomas with both hepatocytic and cholangiocytic differentation. Hepatology 2018;68:113−126.PubMedPMC
• 5. Choi JH, Ro JY. Combined hepatocellular-cholangiocarcinoma: an update on pathology and diagnostic approach. Biomedicines 2022;10:1826. ArticlePubMedPMC
• 6. Vij M, Veerankutty FH, Rammohan A, Rela M. Combined hepatocellular cholangiocarcinoma: a clinicopathological update. World J Hepatol 2024;16:766−775.ArticlePubMedPMC
• 7. Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim TY, et al. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med 2020;382:1894−1905.ArticlePubMedPMC
• 8. Oh DY, He AR, Qin S, Chen LT, Okusaka T, Vogel A, et al. Durvalumab plus gemcitabine and cisplatin in advanced biliary tract cancer. NEJM Evid 2022;1:EVIDoa2200015. PubMed
• 9. Kelley RK, Ueno M, Yoo C, Finn RS, Furuse J, Ren Z, et al. Pembrolizumab in combination with gemcitabine and cisplatin compared with gemcitabine and cisplatin alone for patients with advanced biliary tract cancer (KEYNOTE-966): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet 2023;401:1853−1865.PubMed
• 10. Abou-Alfa GK, Lau G, Kudo M, Chan SL, Kelley RK, Furuse J, et al. Tremelimumab plus durvalumab in unresectable hepatocellular carcinoma. NEJM Evid 2022;1:EVIDoa2100070. PubMed
• 11. Claasen MPAW, Ivanics T, Beumer BR, de Wilde RF, Polak WG, Sapisochin G, et al. An international multicentre evaluation of treatment strategies for combined hepatocellular-cholangiocarcinoma. JHEP Rep 2023;5:100745. ArticlePubMedPMC
• 12. Xue R, Chen L, Zhang C, Fujita M, Li R, Yan SM, et al. Genomic and transcriptomic profiling of combined hepatocellular and intrahepatic cholangiocarcinoma reveals distinct molecular subtypes. Cancer Cell 2019;35:932−947.e8.ArticlePubMedPMC
• 13. Joseph NM, Tsokos CG, Umetsu SE, Shain AH, Kelley RK, Onodera C, et al. Genomic profiling of combined hepatocellular-cholangiocarcinoma reveals similar genetics to hepatocellular carcinoma. J Pathol 2019;248:164−178.ArticlePubMedPDF
• 14. Murugesan K, Sharaf R, Montesion M, Moore JA, Pao J, Pavlick DC, et al. Genomic profiling of combined hepatocellular cholangiocarcinoma reveals genomics similar to either hepatocellular carcinoma or cholangiocarcinoma. JCO Precis Oncol 2021;5:1285−1296.ArticlePubMedPMC
• 15. Sciarra A, Park YN, Sempoux C. Updates in the diagnosis of combined hepatocellular-cholangiocarcinoma. Hum Pathol 2020;96:48−55.ArticlePubMed
• 16. Salimon M, Prieux-Klotz C, Tougeron D, Hautefeuille V, Caulet M, Gournay J, et al. Gemcitabine plus platinum-based chemotherapy for first-line treatment of hepatocholangiocarcinoma: an AGEO French multicentre retrospective study. Br J Cancer 2018;118:325−330.ArticlePubMedPMCPDF
• 17. Calderaro J, Di Tommaso L, Maillé P, Beaufrère A, Nguyen CT, Heij L, et al. Nestin as a diagnostic and prognostic marker for combined hepatocellular-cholangiocarcinoma. J Hepatol 2022;77:1586−1597.PubMed
• 18. Calderaro J, Laleh NG, Zeng Q, Maille P, Favre L, Pujals A, et al. Deep learning-based phenotyping reclassifies combined hepatocellular-cholangiocarcinoma. Nat Commun 2023;14:8290. PubMedPMC
• 19. Nguyen CT, Caruso S, Maille P, Beaufrère A, Augustin J, Favre L, et al. Immune profiling of combined hepatocellular-cholangiocarcinoma reveals distinct subtypes and activation of gene signatures predictive of response to immunotherapy. Clin Cancer Res 2022;28:540−551.ArticlePubMedPDF
• 20. Zhou N, Lei CF, Tan SR, Huang QY, Zhang SY, Liang ZX, et al. Case report: remarkable response to sintilimab, lenvatinib, and nab-paclitaxel in postoperative metastatic chemotherapy-resistant combined hepatocellular-cholangiocarcinoma. Front Pharmacol 2023;14:1190967. ArticlePubMedPMC
• 21. Satake T, Shibuki T, Watanabe K, Sasaki M, Imaoka H, Mitsunaga S, et al. Case report: atezolizumab plus bevacizumab for combined hepatocellular-cholangiocarcinoma. Front Oncol 2023;13:1234113. ArticlePubMedPMC
• 22. Sasaki M, Sato Y, Nakanuma Y. Expression of fibroblast growth factor receptor 2 (FGFR2) in combined hepatocellular-cholangiocarcinoma and intrahepatic cholangiocarcinoma: clinicopathological study. Virchows Arch 2024;484:915−923.ArticlePubMedPMCPDF
• 23. Kobayashi S, Terashima T, Shiba S, Yoshida Y, Yamada I, Iwadou S, et al. Multicenter retrospective analysis of systemic chemotherapy for unresectable combined hepatocellular and cholangiocarcinoma. Cancer Sci 2018;109:2549−2557.ArticlePubMedPMCPDF
• 24. Trikalinos NA, Zhou A, Doyle MBM, Fowler KJ, Morton A, Vachharajani N, et al. Systemic therapy for combined hepatocellular-cholangiocarcinoma: a single-institution experience. J Natl Compr Canc Netw 2018;16:1193−1199.ArticlePubMed
• 25. Gigante E, Hobeika C, Le Bail B, Paradis V, Tougeron D, Lequoy M, et al. Systemic treatments with tyrosine kinase inhibitor and platinum-based chemotherapy in patients with unresectable or metastatic hepatocholangiocarcinoma. Liver Cancer 2022;11:460−473.ArticlePubMedPMCPDF
• 26. Kim EJ, Yoo C, Kang HJ, Kim KP, Ryu MH, Park SR, et al. Clinical outcomes of systemic therapy in patients with unresectable or metastatic combined hepatocellular-cholangiocarcinoma. Liver Int 2021;41:1398−1408.ArticlePubMedPDF
• 27. Tanabe N, Saeki I, Yamaoka K, Kawaoka T, Tomonari T, Tani J, et al. Efficacy of lenvatinib and atezolizumab bevacizumab combination therapy in patients with combined hepatocellular-cholangiocarcinoma. Anticancer Res 2025;45:1117−1125.ArticlePubMed
• 28. Pomej K, Balcar L, Shmanko K, Welland S, Himmelsbach V, Scheiner B, et al. Clinical characteristics and outcome of patients with combined hepatocellular-cholangiocarcinoma - a European multicenter cohort. ESMO Open 2023;8:100783. ArticlePubMedPMC
• 29. Azizi AA, Hadjinicolaou AV, Goncalves C, Duckworth A, Basu B. Update on the genetics of and systemic therapy options for combined hepatocellular cholangiocarcinoma. Front Oncol 2020;10:570958. ArticlePubMedPMC
• 30. Leoni S, Sansone V, Lorenzo S, Ielasi L, Tovoli F, Renzulli M, et al. Treatment of combined hepatocellular and cholangiocarcinoma. Cancers (Basel) 2020;12:794. ArticlePubMedPMC
• 31. Hubner RA, Reeves HL, Edeline J. Combined hepatocellular-cholangiocarcinoma - more questions than answers. Liver Int 2021;41:1186−1188.ArticlePubMedPDF
• 32. Gigante E, Bouattour M, Bedoya JU, Regnault H, Ziol M, Assenat E, et al. Atezolizumab and bevacizumab for non-resectable or metastatic combined hepatocellular-cholangiocarcinoma: a multicentric retrospective study. United European Gastroenterol J 2024;12:429−439.ArticlePubMedPMC
• 33. Jang YJ, Kim EJ, Kim HD, Kim KP, Ryu MH, Park SR, et al. Clinical outcomes of immune checkpoint inhibitors in unresectable or metastatic combined hepatocellular-cholangiocarcinoma. J Cancer Res Clin Oncol 2023;149:7547−7555.ArticlePubMedPMCPDF
• 34. Diab O, Khan M, Abbasi S, Saeed A, Kasi A, Baranda JC, et al. Efficacy of immunotherapy in hepatocholangiocarcinoma (HCC-CC): proof of concept. J Clin Oncol 2021;39(15 Suppl): e16194. Article
• 35. He GQ, Li Q, Jing XY, Li J, Gao J, Guo X. Persistent response to combination therapy of pemigatinib and chemotherapy in a child of combined hepatocellular-cholangiocarcinoma with FGFR2 fusion. Mol Cancer 2024;23:269. ArticlePubMedPMCPDF
• 36. Goyal L, Meric-Bernstam F, Hollebecque A, Valle JW, Morizane C, Karasic TB, et al. Futibatinib for FGFR2-rearranged intrahepatic cholangiocarcinoma. N Engl J Med 2023;388:228−239.ArticlePubMed
• 37. Abou-Alfa GK, Macarulla T, Javle MM, Kelley RK, Lubner SJ, Adeva J, et al. Ivosidenib in IDH1-mutant, chemotherapy-refractory cholangiocarcinoma (ClarIDHy): a multicentre, randomised, double-blind, placebo-controlled, phase 3 study. Lancet Oncol 2020;21:796−807.ArticlePubMedPMC
• 38. Na HY, Kim JH, Kim H, Cho JY, Han HS, Jang ES, et al. Multiregional analysis of combined hepatocellular-cholangiocarcinoma reveals histologic diversity and molecular clonality. Histopathology 2024;84:402−408.ArticlePubMed
• 39. Lin YS, Wu LY, Lin LH, Yang X, Liu FY, Wu YQ, et al. Interventional treatment combined with immunotargeted therapy in unresectable combined hepatocellular-cholangiocarcinoma: a real-world retrospective cohort study. Front Immunol 2025;16:1591127. ArticlePubMedPMC
• 40. Wang A, Wu L, Lin J, Han L, Bian J, Wu Y, et al. Whole-exome sequencing reveals the origin and evolution of hepato-cholangiocarcinoma. Nat Commun 2018;9:894. ArticlePubMedPMCPDF

Figure & Data

References

Citations

Citations to this article as recorded by  

• Curative Effect Evaluation of Targeted Therapy and Chemotherapy for Non-Resectable Combined Hepatocellular-Cholangiocarcinoma: A Systematic Review and Meta-Analysis
  永豪 林
  Advances in Clinical Medicine.2026; 16(05): 920.     CrossRef