In parallel with the development of personalized precision oncology treatments targeting key relevant genomic alterations, strategies acting on the immune system have demonstrated significant clinical activity (1). The approved therapies in this field target immunosuppressive proteins, therefore inducing the activation of the immune system (2). Among the different components of the immune system, human leukocyte antigen (HLA) proteins, including both class I and class II molecules, are key participants in immune regulation and immune escape in the cancer process (2, 3).

There are two major classes of HLA molecules: class I and class II. Class II HLA molecules, including HLA-DQ, HLA-DR, and HLA-DP, are primarily expressed on antigen-presenting cells (APCs) such as dendritic cells, macrophages, and B cells. These molecules play a critical role in presenting extracellular antigens to CD4+ T helper cells, thereby initiating and modulating adaptive immune responses. Dysregulation of HLA class II expression has been implicated in immune evasion by tumors, as reduced expression can impair the activation of CD4+ T cells and weaken antitumor immunity (3).

Class I HLA molecules are further divided into classical and non-classical subtypes (2). Classical type I HLA molecules (HLA-A, HLA-B, and HLA-C), are essential for immunosurveillance and cancer immunotherapy as T cells are presented with cellular antigens by this subtype (2, 4, 5). The non-classical HLA molecules, HLA-E, HLA-F, and HLA-G have immunosuppressive properties in contrast to traditional HLA-I molecules (2, 4, 5). HLA-E, found on chromosome 6, is formed by the association of a heavy chain and β2-microglobulin. This molecule presents peptides derived from the leader sequences of other MHC class I molecules, as well as potentially from pathogen-derived peptides (6). Its primary role is to enable Natural Killer (NK) cells to monitor the expression of the other class I HLA molecules, limiting NK action while also inhibiting T-cell cytotoxicity (6). Thus, HLA-E serves as a critical link between memory-driven adaptive immunity and the rapid response of innate immunity (5, 7–9).

Although it is frequently prevalent in trophoblast and tumor cells, HLA-E can be expressed at low levels in all nucleated cells (7, 10). Growing evidence indicates that tumors can use HLA-E expression to evade NK cell detection even in the absence of conventional class I HLA (8). A poor prognosis has been linked to HLA-E expression in various malignancies including colorectal cancer, gastric cancer, gynecological cancers, non-small cell lung cancer (NSCLC) and breast cancer (11–16).

Additionally, the tumor microenvironment (TME) is a dynamic ecosystem comprising immune cells, stromal components, and signaling molecules that collectively influence tumor progression and therapeutic responses (17–20). Within this setting, HLA-E can further facilitate immune evasion by engaging inhibitory receptors on NK and T cells, underscoring its relevance in shaping antitumor immunity and response to immunotherapies. In certain tumors, HLA-E expression can diminish the protective role of tumor-infiltrating lymphocytes, thus contributing to immune escape (20). Re-education strategies targeting the pro-tumor TME may help restore antitumor immune functions (17). This evolving understanding highlights HLA-E as a potential therapeutic target, particularly in combination with immunotherapies aimed at counteracting T cell exhaustion and enhancing treatment outcomes (18, 19).

HLA-E has shown significant therapeutic potential due to its expression in tumor cells compared to healthy tissues (21). By inhibiting cytotoxic NK cells and a subset of CD8 T lymphocytes through interaction with the NKG2A/CD94 heterodimer, it presents a promising target for novel therapeutic antibodies currently under evaluation in clinical studies (21). Targeting this pathway is vital because NK cells perform cytotoxic functions, facilitate antibody-mediated cellular cytotoxicity (ADCC), and recruit dendritic cells into tumors enhancing communication between the innate and adaptive immune systems (21).

Considering the importance of this family of proteins and particularly the relevant role of HLA-E for therapeutic purposes we conducted a systematic review and meta-analysis to investigate the predictive relationship between HLA-E and survival in solid tumors.

This systematic review and meta-analysis were performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) (22) and was registered in PROSPERO (CRD42024527598).

HLA-E expression levels, typically elevated in tumor cells compared to healthy tissues, lead to the inhibition of cytotoxic NK cells and CD8 T lymphocytes through interaction with the NKG2A/CD94 heterodimer (2, 21, 31, 32). Consequently, tumor cells may evade NK cell action through HLA-E overexpression (21, 33). This immunosuppressive mechanism emphasizes the potential therapeutic importance of targeting the NKG2A/HLA-E axis (13, 15, 21, 34). This systematic review and meta-analysis investigated the relationship between HLA-E expression and survival in solid tumors. Our analysis revealed no statistically significant association between HLA-E expression and OS. However, HLA-E non-expression was significantly associated with improved DFS. The included studies encompassed a diverse range of solid tumors, with considerable variability in staging and follow-up periods.

HLA-E binding peptides are primarily responsible for controlling HLA-E expression (31, 32). Tumors may upregulate the expression of HLA-E by ensuring a sufficient supply of peptides that stabilize HLA-E, thereby engaging the inhibitory receptor NKG2A on NK cells and certain T cells to protect themselves from immune attack (35, 36). HLA-E can modulate CD8+ T cell responses, and under certain conditions, can induce a Th2-like cytokine profile and activate B cells ( Figure 4 ) (37, 38).

HLA-E exists in different conformational forms, notably the peptide-bound form and the open conformer. The peptide-bound form, associated with β2-microglobulin and a peptide ligand, is the predominant form expressed on the cell surface and interacts with inhibitory receptors such as NKG2A/CD94 on NK cells and CD8+ T cells (36). In contrast, the open conformer lacks a bound peptide and may have altered stability and receptor interactions (39). The conformational state of HLA-E can influence its expression levels, immune recognition, and function in immune regulation (36, 39).

Depending on the receptor and the responding cell, HLA-E engagement can result in immune activation or suppression (31, 40). As a result, several studies have investigated the connection between HLA-E expression and tumors in recent years. Patients with melanoma (41), glioblastoma (42), gastric cancer (12), breast cancer (16), and ovarian cancer (32) were reported to have high expression of HLA-E and the expression level of HLA-E was linked to clinical outcomes. Another study reported a similar correlation between HLA-E expression and disease-free survival in bladder cancer patients (43).

In the current study, HLA-E made no difference in OS. The reasons for the disparity of the results could be due to different histologies as well as differences in methodology. Our results for DFS, however, showed that overexpression of HLA-E was associated with a poor DFS. Across the six studies included in our DFS analysis, a common theme emerged: HLA-E expression was linked to worse DFS outcomes, despite the variability in cancer types and methodologies. Levy et al. evaluated colorectal cancer patients using IHC, finding that high HLA-E expression was significantly associated with shorter DFS, especially in Dukes’ C patients (11). Levy et al. also highlighted the role of HLA-E in immune evasion through NK cell inhibition (11). Similarly, Morinaga et al. (12), Hiraoka et al. (23), Benevolo et al. (28), and Versluis et al. (29), found that higher HLA-E expression correlated with poorer DFS across esophageal squamous cell carcinoma, pancreatic ductal adenocarcinoma, cervical cancer, and endometrial cancer. Lastly, Wang et al. in their mRNA expression analysis in hepatocellular carcinoma patients showed that higher HLA-E mRNA levels were associated with worse DFS (30). In this context, in primary tumors, predominantly from early-stage disease, HLA-E could play a key immunosuppressive role (31) and therefore could be associated with a detrimental prognosis. In line with their potential role as a therapeutic target, treatments targeting their receptor, NKGD2, are currently under research (44).

Therapeutic strategies targeting the NKG2A/HLA-E axis include both monotherapies and combination therapies (21, 45). While direct inhibition of HLA-E could potentially prevent its interaction with the activating NKG2C receptor, the preferred approach has been to block NKG2A (46). Monalizumab, a first-in-class humanized IgG4 antibody targeting the NKG2A/CD94 receptor, has shown promise in preclinical studies by enhancing NK and CD8 T cell activity (44, 45). Monalizumab blocks the interaction of NKG2A/CD94 with HLA-E, thus promoting NK and CD8 T cell activation. This mechanism not only enhances cytotoxic activities but also augments antibody-dependent cellular cytotoxicity (ADCC), a crucial effector function for many anticancer antibodies (46, 47). Early clinical trials with Monalizumab alone or in combination with anti-PD-L1 antibodies like Durvalumab have demonstrated acceptable safety profiles and preliminary signs of immune activation, the COAST study reported improved overall response rates (ORR) and progression-free survival (PFS) with Monalizumab and Durvalumab in unresectable stage III NSCLC patients (48).

Clinical trials have also explored various combinations of Monalizumab with other therapeutic agents in the recurrent or metastatic setting (46, 49, 50). Other monoclonal antibodies like S095029 and HY-0102 are in clinical development (51, 52). However, tumors may compensate for NKG2A/HLA-E blockade by upregulating other immunosuppressive molecules like PD-L1 or IDO, thereby preserving an immune-evasive niche (18, 20).

To the best of our knowledge, our study is the first meta-analysis to discuss HLA-E and its role in the outcomes of solid cancer patients, including OS and DFS. It offers valuable insights into the prognostic significance of HLA-E expression in various solid tumors. However, to contextualize the interpretation of our findings, it is important to highlight several limitations. A significant limitation is the pronounced heterogeneity among the included studies, encompassing diverse tumor types (e.g., colorectal, pancreatic, esophageal, brain, renal cell, gastric, endometrial, cervical, and hepatocellular carcinomas) and varying disease stages. Each of these cancer types possesses unique biological behaviors, prognostic factors, treatment modalities and responses to treatment. The variability in cancer stages across studies further compounds this issue. This heterogeneity in tumor types and stages may confound the relationship between HLA-E expression and survival outcomes, thereby limiting the ability to generalize our findings. There was substantial variation in the methods used to quantify HLA-E expression. While the majority of studies employed IHC to detect HLA-E levels in tissue samples, others utilized mRNA expression analyses, and the positivity thresholds can varied widely. Importantly, the specificity of the MEM-E/02 antibody for HLA-E detection warrants consideration, as potential cross-reactivity with other HLA class I molecules or recognition of different conformational forms of HLA-E could impact the accuracy of HLA-E expression assessment (53). Such discrepancies in detection techniques, antibody specificity, and positivity criteria can lead to inconsistencies in categorizing patients as HLA-E positive or negative, potentially impacting the results. Moreover, our study did not correlate HLA-E expression with T cell and NK cell infiltration or the expression of HLA-E-specific receptors on immune cells, which limits our understanding of the immunological context and mechanisms of immune evasion (31). Furthermore, HLA-E expression can be influenced by underlying polymorphisms, such as the commonly studied HLA-E01:01 and HLA-E01:03 alleles, which differ in expression levels and surface stability (8, 35, 43). These variations may modulate immune recognition and the extent of immune evasion within the TME, further highlighting the complexity of HLA-E’s prognostic and therapeutic role. Much of the research that comprises our analysis is retrospective, which brings with it inherent limitations including selection bias, dependence on pre-existing data sources and confounding variables. Additionally, relying solely on published literature could introduce publication bias. These limitations could affect the validity of our findings. Although the results of our meta-analysis shed light on the predictive importance of HLA-E expression in solid tumors, they do not clarify the underlying mechanisms responsible for these correlations. To shed light on the causative pathways and biological significance of our findings, further research is necessary to fully understand the intricate interactions between HLA-E expression, tumor biology, immune microenvironment, and treatment modalities.

Our systematic review and meta-analysis provide comprehensive insights into the prognostic significance of HLA-E expression in solid tumors. Despite the lack of statistical significance in the relationship between HLA-E expression and OS, our results point to a possible predictive role for HLA-E non-expression for improved DFS suggesting that targeting the NKG2A/HLA-E axis could be a valuable strategy in enhancing antitumor immune responses, which could be particularly beneficial for early-stage diseases. However, several challenges remain. The variability in clinical outcomes highlights the need for more research to optimize treatment combinations and identify the patient populations that would benefit most from NKG2A/HLA-E axis inhibition. In particular, prospective studies and clinical trials evaluating HLA-E as a therapeutic target will be instrumental. Moreover, understanding the underlying mechanisms of immune evasion and the role of HLA-E in different tumor microenvironments is crucial for developing effective treatments. Further investigation is essential to develop these approaches, optimize combination strategies, and elucidate the mechanisms driving the variable clinical outcomes across different cancer types and stages.