Adrenocortical carcinoma (ACC) is a rare and aggressive malignancy with an incidence of 0.6-2 inhabitants per million per year (1). Around 60% of patients show steroid hormone excess, which has been associated with poor outcome (2, 3).

The treatment options of advanced ACC are unsatisfactory. After resection with microscopically unaffected margins (R0), 50-80% of patients develop recurrent or metastatic disease (4–7). Mitotane is the only approved treatment for patients to control tumour growth and hypercortisolism. With a response rate of - at best - ~20% in carefully selected cases, mitotane monotherapy is of limited efficacy (8). Even in combination with cytotoxic chemotherapy, the response rate of only ~25% underlines the unmet medical need for novel treatment alternatives (9–11)

Immune checkpoint inhibitors (ICI) have revolutionised the treatment of many cancer entities but, unfortunately, not of ACC. The results of five clinical trials using different ICIs are heterogeneous and overall disappointing with objective response in only 17 (14%) of 121 treated patients (8, 12–16). A similar response rate was confirmed in a large retrospective analysis and a recent meta-analysis (17, 18). In none of these trials, established response markers to ICI such as tumour mutational burden or microsatellite instability were associated with response.

The tumour microenvironment of ACC has been reported to be depleted of T lymphocytes giving rise to the notion of ACC as an immunologically “cold” tumour entity (13), which is considered to contribute to the impaired responsiveness of ACC to immunotherapy.

Recently, our group has used clinical data of steroid hormone secretion together with IHC for T cell markers in 146 ACC tumours and found an immunosuppressive effect of clinical GC excess on T cell infiltration (13).

The Wnt/β-catenin pathway has been suggested as an additional cause for an immunologically silent tumour microenvironment in several types of cancer (19, 20). Activation of the Wnt/β-catenin pathway is a common driver event in ACC and caused by mutations in CTNNB1 itself or by mutations in the upstream pathway gene ZNRF3 ( 21, 22).

Established roles of canonical Wnt/β-catenin signalling include proliferation, apoptosis, migration and tissue homeostasis (14). The key steps and mechanisms of Wnt/β-catenin activation in ACC have recently been reviewed elsewhere (23) and are depicted in Figure 1 . In the final step of this pathway, CTNNB1 binds to the transcription factors TCF/LEF-1 (25) and triggers expression of target genes like c-Myc, cyclin D1 and CDKN1A (26).

Beyond its classical pro-tumourigenic functions, Wnt/β-catenin activation has been found to be associated with T cell-depletion across tumour entities including ACC (27, 28).

Mechanistically, Wnt signalling regulates peripheral T cell activation and differentiation of CD4⁺ T helper and CD8⁺ cytotoxic T cells (CTLs) (29). In melanoma cell lines as well as dendritic cells (DC), experimental overexpression of CTNNB1 has been shown to increase interleukin (IL) IL-10 secretion and to impede the ability of DCs to cross-prime tumour-targeting CD8⁺ T cells (30). Cancer cells secrete Wnt ligands that interact with frizzled (FZD) receptors on DCs, thereby inhibiting the production of IL-12 and promoting the differentiation of regulatory T cells (T_regs) (31, 32).

Conversely, the production of IL-1β by tumour-associated macrophages has been shown to result in the phosphorylation and the subsequent stabilisation of β-catenin in colon cancer (33).

Despite the prominent role of this pathway in ACC, a systematic study of Wnt/β-catenin signalling and its potential interconnections with tumour-infiltrating immune cells has not been conducted so far.

Therefore here, we aimed to elucidate (i) the impact of Wnt/β-catenin pathway activation on the number of tumour-infiltrating T cells, (ii) the relationship between Wnt/β-catenin pathway activation and GC signalling and (iii) the immune-related mechanisms affected by Wnt/β-catenin signalling in the tumour microenvironment. To this end, we used IHC to assess LEF-1, CTNNB1 and GR expression and applied mRNA gene expression analysis of immune-related transcripts in clinically annotated ACC tissue samples.

Wnt/β-catenin pathway activation is a known driver event in solid tumours including ACC (43). Here, we aimed to determine whether the activation of this pathway is associated with reduced tumoural immune cell infiltration and to identify other potentially associated determinants of tumoural immunity, such as GC signalling.

We first established LEF-1 IHC as a marker of Wnt/β-catenin pathway activation in ACC and demonstrated LEF-1 expression to lack association with T cell infiltration. Surprisingly, LEF1 was higher expressed in GC secreting ACC and associated with shorter patient survival.

Wnt/β-catenin pathway activation signatures have been established from RNA expression data but are difficult to assess in clinical routine, especially when using FFPE samples.

We found LEF-1 expression to be positively correlated to CTNNB1 staining and an easy to assess marker of the Wnt/β-catenin pathway activation. In contrast, the actual expression of activated CTNNB1 is difficult to distinguish visually due to both cytoplasmic and nuclear staining which is particularly true in lipid-rich tumours like ACC. LEF-1 expression is exclusively nuclear and therefore can be readily assessed by IHC. LEF-1 expression may be used as a downstream readout of Wnt/β-catenin pathway activation. High levels of LEF-1 expression correlate with an active Wnt/β-catenin signalling. LEF-1 expression has been previously utilised as a marker in different tumour entities like acute lymphoblastic leukaemia, oral squamous cell carcinoma, and colorectal cancer to identify Wnt/β-catenin activated tumours (44, 45). To our knowledge, LEF-1 has not been studied in ACC.

Activating mutations in genes of the Wnt/β-catenin pathway occur in many solid tumour entities like colorectal cancer (46), melanoma (47) or lung cancer (48). Driver mutations comprise the CTNNB1, ZNRF3, APC or Axin1 genes in most cases. To this date no mutations in LEF-1 have been discovered for any entity.

In ACC, Wnt/β-catenin pathway activation is frequent and occurs via activating mutations in the CTNNB1 gene or inactivating mutations in ZNRF3 ( 21, 49). It must be noted that for our study, mutational data of CTNNB1 and ZNRF3 were unavailable due to the retrospective nature of this study. Still, yet unidentified mechanisms may activate the Wnt/β-catenin pathway in ACC similar to other malignancies. As an example, hematopoietic malignancies drive transcriptional pathway activation through the downstream effector LEF-1 in a β-catenin independent manner (45, 50). Based on gene expression data, Assié et al. have demonstrated ACCs with an activating CTNNB1 mutation to exhibit higher expression of LEF-1 compared to non-tumour adrenal samples and ACCs without CTNNB1 mutation (21).

It has been shown in ovarian and colorectal cancer that Wnt/β-catenin pathway activation leads to significantly lower levels of mature CD8⁺ T cells. In colorectal cancer, higher β-catenin levels led to an inhibition of CCL4, a proinflammatory chemokine, and subsequently lower recruitment of CD103⁺ DCs thus impairing CD8⁺ T cell activation (27, 51, 52). In a pan-cancer study by Luke et al., ACC represented the tumour entity with strongest pathway activity which was in turn related to reduced T cell-driven inflammation (28). Using a gene signature approach to characterise tumoural immune cells, Thorsson et al. demonstrated CTNNB1 mutations to be associated with the immune cell-depleted phenotype of ACC (53). Also in melanoma it was shown that T cell depleted tumours often have an activating CTNNB1 mutation (54). By using high LEF-1 levels as an indicator of Wnt/β-catenin pathway activation we did not observe this negative correlation in our ACC cohort. Previous work implying Wnt/β-catenin pathway activation in immune cell exclusion has relied on gene expression signatures and focused on the relationship between CTNNB1 expression and CD8⁺ T cell infiltration in adrenocortical carcinoma. Liu et al. used TCGA data to show an inverse correlation between CTNNB1 expression and CD8⁺ T cell infiltration in ACC (55). These findings mirror the generally low proportion of CD8⁺ cells in an independent set of ACC specimens observed by Landwehr et al. (13). The mechanism underlying this relationship is unclear. While our study may be too small to capture an association of CTNNB1 expression specifically with CD8⁺ T, we find the number of CD3⁺ T cells to be inversely associated CTNNB1 expression.

While it is established that Wnt/β-catenin activation is associated with poor prognosis in ACC, we observed lacking association between CTNNB1 nucleocytoplasmic staining and clinical parameters in our cohort. We therefore suggest that LEF-1 may represent a better marker of Wnt/β-catenin pathway activation in ACC. The activation of this pathway is closely related to cortisol excess associated with lesser T cell infiltration. It must be acknowledged that specifically the infiltration with CD8⁺ T cells may be induced by crosstalk of the Wnt/β-catenin pathway not transduced via LEF1. This phenomenon of signal crosstalk has been reviewed by Zhou et al., highlighting the complex interactions between the Wnt/β-catenin pathway activation and immune exclusion (20).

On the other hand, GC excess has been demonstrated to be strongly associated with depletion of T cell subtypes in ACC, both at the mRNA (56) and protein expression levels (13). Here, we found that LEF-1 protein expression is associated with the presence of clinical GC excess. Hence, any association of Wnt/β-catenin pathway activation will be confounded by direct GC excess.

However, while we observed a direct association between CTNNB1 and CD3⁺ immune cells, LEF-1 was not linked to any of the immune cell subtypes but interestingly associated with impaired expression of antigen presentation in tumours suggesting modulation of immune recognition.

We observed a less favourable survival for patients with high levels of LEF-1 but not high CTNNB1 expression.

Based on TCGA data, we hypothesized CTNNB1 expression levels to have a significant impact on overall survival. While the clinical characteristics of our cohort are similar with those of the TCGA cohort, TCGA has studied mRNA expression whereas we used protein expression by IHC on FFPE tissue samples. It is likely that mRNA expression does not fully reflect pathway activation given the complexity of pathway regulation and signalling crosstalk. In colorectal cancer, breast cancer and hepatocellular cancer, the activation of the Wnt/β-catenin pathway leads to a worse overall survival as high levels of CTNNB1 promotes metastasis and epithelial-mesenchymal transition (EMT) (57).

In contrast to a study by Wu et al., we did not find any correlation between the expression levels of GR and the immune cell infiltration levels or overall survival of the ACC patients (37).

In our study, we employed IHC to analyse LEF-1 expression, similar to the approach used in their West China Hospital (WCH) cohort. However, it is important to acknowledge the differences between the two studies, including variations in cohort size, clinical characteristics, and group classification methods. While the WCH cohort defined “high” expression as the top quartile, our study utilized the median expression level as the threshold for classification.

We found a positive correlation between cortisol secretion and LEF-1 expression levels, supporting our hypothesis that the Wnt/β-catenin pathway contributes significantly to immune exclusion in ACC, potentially through the mechanism of cortisol excess. These results not only corroborate previous findings but also strengthen the rationale for targeting these pathways in future treatment strategies.

The main limitations of our study is the relatively small number of samples. This is related to the rarity of the disease where large cohorts of clinically annotated patients are difficult to accrue. Further, in the absence of methods to directly assess intra-tumoural steroid concentration, we classified the samples according their clinical features.

Collectively, our work shows that LEF-1 can be used as a routine biomarker of Wnt/β-catenin pathway activation. Activation is not directly related to T cell infiltration but GC secretion in ACC and might play a role in antigen presentation, which needs further investigation.

The potential of LEF-1 expression as a biomarker for Wnt/β-catenin pathway activation could aid in patient selection for treatment with Wnt/β-catenin inhibitors, such as DCR BCAT. The promising results observed by Ganesh et al., where combination therapy with CTLA-4 and PD -1 antibodies led to complete remission in the majority of mice with Wnt-driven tumours, further underscore the therapeutic potential of targeting this pathway in ACC (58).