In recent years, the tumour microenvironment has become a major focus of study in CCA. CCA is highly desmoplastic and typically presents with a stroma rich in cancerassociated fibroblasts (CAFs), along with a variety of immune cells [2]. Similar to other tumours with a prominent stromal component, the combination of the immune cell profile and the specific CAF subtypes present in the tumour region has a significant impact on disease progression and aggressiveness [3,4]. Furthermore, the establishment of immunotherapy as a standard treatment, in combination with chemotherapy, has been one of the most important milestones in advancing disease management [5]. Therefore, understanding the contribution of the stroma to therapy efficacy will be key to improving and developing new treatment strategies.

In this issue of Clinical and Molecular Hepatology, Ye et al. [6] propose a novel stromal signature for stratifying intrahepatic CCA tumours, aimed at predicting post-surgery prognosis. Using surgical resections from a training and a validation cohort (total=599), the authors created a training series of tissue microarrays, which were analysed using immunohistochemical (IHC) staining for alpha-smooth muscle actin (aSMA) and collagen deposition (aniline) to assess the presence and abundance of CAFs. In addition, they employed markers for leukocytes (CD3, CD4, and CD8 – T-cells), as well as for myeloid populations (CD68 – macrophages) and CD66b (neutrophils).

The data first revealed that CAF status alone had a significant impact on both recurrence-free survival (RFS) and overall survival (OS). By classifying patients into four distinct groups, the analysis showed that those with low aSMA staining combined with high collagen deposition had the longest survival rates, while patients with low collagen deposition but high aSMA-positive CAF infiltration exhibited the poorest prognosis. These findings align with previous single-cell sequencing analyses of intrahepatic CCA, which demonstrated that aSMA and extracellular matrix (ECM) production do not always correlate when clustering CAFs into different subtypes [7,8]. It is plausible that a dense ECM, devoid of supportive cells such as CAFs, may act as a physical barrier, limiting tumour expansion, while the presence of reactive CAFs within the stroma may be a key requirement not only to promote tumour proliferation but also to recruit other pro-tumourigenic populations into the tumour microenvironment.

The migration and infiltration of various immune populations into tumour nodules is one of the most critical factors influencing tumour behaviour. In highly desmoplastic tumours like CCA, CAFs play a pivotal role in secreting factors that determine which immune cells are recruited or excluded from the tumour region. In this study, the authors expanded their previous analysis by integrating CAF status with the profiling of lymphoid and myeloid populations present within the tumours. They assessed the abundance of anti-tumour T-cell populations (CD4+, CD8+), pro-tumour regulatory T-cells (FOXP3+), as well as macrophages (CD68+) and neutrophils (CD66b+). By integrating immune cell abundance alongside CAF status and the annotated clinical data, the most significant variables were identified and further examined using a recursive partitioning algorithm.

These data align with the transcriptomic characterization of iCCA, which supported the contribution of stroma and immune infiltrate as the key markers for iCCA classifiers, which, not only impact prognosis, but can also inform therapeutic algorithm [9]. However, while the clinical application of gene signatures remains more challenging, the definition of a score based on multiplex-immunohistochemistry increases clinical applicability. Therefore, the study by Ye et al. has the advantage of providing a cost-effective and easy-to-use classifier.

This analysis revealed six distinct tumour groups, which predicted RFS and OS in ascending order of risk. Group I, characterised by high CD8+ T-cell infiltration, low aSMA reactivity, and high collagen deposition, presented the best prognosis. In contrast, Group VI, defined by low CD8+ infiltration and high neutrophil presence, displayed the worst survival prediction. Thus, the presence of CD8+ T-cells emerged as the most critical determinant of survival, while additional risk factors for both RFS and OS included macrophage and neutrophil abundance, as well as the presence of aSMA-positive CAFs.

While CD8+ T-cells have long been recognised for their pivotal anti-tumour role, it is important to emphasise the authors’ findings regarding the role of neutrophils. It is increasingly clear that neutrophils are among the most abundant immune populations in the tumour microenvironment, and their pro-tumourigenic role is being unravelled across various tumour types. In lung and breast cancers, for instance, neutrophils have been shown to promote tumour invasion by degrading and remodelling the ECM [10-12], thereby removing the physical barrier surrounding the tumour. In hepatocellular carcinoma (HCC), neutrophils modulate cancer growth and educate the tumour immuno-microenvironment to change response to immunotherapy [13]. However, while a few studies have previously noted the presence of neutrophils in iCCA, their mechanisms of action remain largely unclear [14]. Furthermore, this study further supports the idea that the poor prognosis associated with a phenotype characterised by low CD8+ T-cell levels and significant neutrophil infiltration - especially those with ECM-degrading capabilities - is largely independent of the CAF status in these patients.

Interestingly, this study found that FOXP3+ T-cells did not play a significant role in determining patient outcomes when compared to other markers and were not a decisive factor in generating the stromal signature groups. This contrasts with previous reports, where Treg infiltration was found to impact CD8+ T-cell activity in the tumour microenvironment and contribute to metastasis in the lymph nodes. Given that the authors found neutrophils—but not Tregs— to influence survival in patients with lymph node metastases, further investigation is needed to clarify the role of Tregs and explore whether other cell types, such as gamma-delta T-cells, might be contributing to the immunosuppressive environment.

Despite the survival impact of phenotypes such as Group VI, it is important to note that most patients are distributed across a range of phenotypes where their CAF and ECM status plays a crucial role. aSMA is not highly expressed in a significant proportion of CAFs, and it has not been identified as a relevant CAF subtypes in iCCA [15]. Therefore, strengthening this analysis by further profiling different CAF subtypes using a more specific panel of markers, such as IL-6, CD74, periostin, or endothelial markers, is deemed necessary. Alternatively, employing more general CAF markers, such as PDGFRα or FAP, would improve the detection of larger quantities of CAFs or bring to light CAF populations not identified previously in inert-defined tumours.

Despite the assessment of prognostic scores maintain their validity in the management of iCCA, the development of predictive markers of response to adjuvant systemic treatment are necessary to optimize the therapeutic strategies currently available. It will be important in the future to understand how the score proposed in this paper will perform in presence of adjuvant treatment. Recent data on advanced iCCA undergoing chemotherapy has revealed the impact of stromal and immune cells on the benefit from treatment. Specifically, CD68+ cells, which are included in the score presented by Ye et al., were shown to display transcriptomic differences between long survivors and rapid progressors after chemotherapy treatment, suggesting that not only the number, but the heterogeneous functionality of these cells may affect tumour biology [16].

Finally, by highlighting the interplay between CAFs and immune cells as a major prognostic factor in iCCA, this study reinforces the need to direct future research in two key directions. The first involves deeper profiling and understanding of the various stromal components and their roles in tumour progression. The second seeks to target the stroma, especially CAFs, to not only inhibit tumour support but also modulate the immune populations recruited to the tumour microenvironment, thus enhancing the efficacy of both chemotherapy and emerging immune-based cancer therapies in cholangiocarcinoma.