EBV is among the most prevalent and persistent infections in humans. Approximately 95% of the world’s population experiences persistent, asymptomatic EBV infection throughout their lifetime, primarily affecting lymphocytes and oropharyngeal epithelial cells (21). There is evidence that the respiratory tract serves as the primary site of EBV hosting. The detection of EBV in bronchoalveolar fluid further suggests that lung tissue may serve as a primary reservoir for EBV (22). The life cycle of EBV encompasses two distinct phases: the latency period and the lytic phase (alternatively described as cleavage followed by reactivation). Following the initial infection which is typically asymptomatic, EBV establishes a lifelong persistent infection within the host. During latency, the EBV genome is replicated as an episome in the S phase of the cell cycle. Subsequently, these replicated episomes are distributed to daughter cells during cell division. The latent EBV expresses gene products that may possess carcinogenic properties, including Epstein-Barr virus nuclear antigen 1 (EBNA1), latent membrane protein-1 (LMP1) and microRNAs (miRNAs) (23–25). Notably, these gene products exhibit high expression levels in PLEC (15), indicating that the latent infection of EBV plays a significant role in the initiation and progression of PLEC. Osorio et al. addressed the epidemiological and experimental evidence of a potential role of EBV (26). However, the precise carcinogenic mechanisms underlying these associations remain to be elucidated through further investigation.

While EBV positivity has been considered a defining feature of PLEC, emerging studies have reported rare cases of EBV-negative PLEC. These EBV-negative cases challenge the current understanding of PLEC pathogenesis and suggest the existence of alternative oncogenic pathways. Studies from independent cohorts yielded similar results: patients with high EBV DNA before pretreatment or positive EBV DNA after treatment had significantly poorer progress free survival (PFS). Circulating EBV DNA levels provide prognostic value for survival and treatment response in patients with PLEC (27). Hence, elucidating the precise role of EBV in PLEC development is essential not only for understanding its typical viral-driven mechanism but also for distinguishing it from phenotypically similar but etiologically distinct tumors.

The role of EBV in tumorigenesis correlates with the expression levels of the latent genes and the extent of genomic aberrations in the host (28). EBV contributes to PLEC pathogenesis through the expression of multiple gene products, including latent proteins, non-coding RNAs, and virus-encoded miRNAs. In PLEC, EBV is primarily found in a latency type II infection pattern, with consistent expression of key latent genes such as LMP1, latent membrane protein 2A (LMP2A), and EBNA1, whereas EBNA2 is usually absent (29). In addition, EBV-encoded small RNAs (EBERs) and miRNAs are upregulated in tumor tissues. These products cooperatively regulate host signaling, immune evasion, and epigenetic modifications (30). The key viral gene products and their functional classifications are summarized in Table 1 .

LMP1 is a constitutively active oncoprotein that mimics a tumor necrosis factor receptor (24). LMP1 plays a crucial role in the pathogenesis of EBV-associated tumors, such as nasopharyngeal carcinoma (NPC) and Hodgkin lymphoma (39, 40). Multiple studies have consistently reported that LMP1 expression is upregulated in PLEC tissues (15, 29, 41), reinforcing its oncogenic potential in a ‘pulmonary’ context. Mechanistically, LMP1 activates the tumor necrosis factor receptor-associated factor (TRAF)-mediated Nuclear Factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling pathway (42, 43). In line with this, Hong et al. observed frequent deletions of TRAF3 and NFKBIA in PLEC samples, suggesting that disruption of the LMP1–TRAF3 axis may represent a key molecular mechanism driving PLEC tumorigenesis (44). Unlike in NPC, where LMP1 activates both canonical and non-canonical NF-κB pathways primarily via TRAF2, recent findings suggest that in PLEC, TRAF3 downregulation may drive more sustained non-canonical NF-κB signaling, which may contribute to a more immunosuppressive tumor environment and resistance to apoptosis in PLEC.

LMP2A has been identified as the typical latency type II transcript encoded by EBV, and it is expressed in PLEC (29). Preliminary genetic studies suggested that LMP2A was not essential for in vitro growth transformation of B cells (32). However, recent studies have demonstrated that LMP2A can function as a simulated receptor, thereby promoting the malignant transformation of B cells. Nevertheless, the precise oncogenic mechanism of LMP2A remains to be elucidated. Studies have shown that LMP2A can indirectly regulate the expression of LMP1 by modulating the activity of NF-kB in epithelial cells (45). Consequently, the interaction of LMP2A with LMP1 in PLEC may enhance its oncogenic potential.

EBNA1 is a pivotal factor in the establishment of latent infection of EBV in proliferating B cells. EBNA1 has been demonstrated to stimulate replication of DNA, regulate transcription of both virus and host genes, and tether the virus to cell chromosomes (31). It has been demonstrated that EBNA1 exhibits a high degree of affinity for the recognition site in the human genome. The binding of EBNA1 to its target sequence has been shown to cause local demethylation, thereby promoting the activation of silent cell promoters (46). In a study by Wu et al., the expression of EBNA1 was found to be up-regulated based on genome sequencing of EBV isolated from 78 PLEC patients and 37 healthy controls (29). Despite the established role of EBNA1 as an oncoprotein expressed in all EBV-associated tumors (47), further investigation is required to elucidate its potential carcinogenic effect in PLEC.

Non-coding RNAs are transcripts that are not translated into proteins. EBER1 and EBER2 are EBERs that possess high abundance (typically reaching up to 10⁷ copies per cell) in latent infection (33). Their expression is considered the gold standard for identifying EBV-positive tumors via in situ hybridization (34). Chen et al. examined tumor tissues from 42 patients with PLEC by EBER detection, and 78.6% (33/42) showed positive results (48). Functionally, EBERs contribute to immune modulation and may promote cell survival. Recent findings indicate that EBER2 enhances B-cell growth by upregulating the deubiquitinase UCHL1, a process that may also be relevant in EBV-positive epithelial tumors (36).

EBV encodes two major clusters of miRNAs: miR-BHRF1 and miR-BART. The BamHI A rightward transcript (BART) cluster is preferentially expressed in epithelial tumors such as PLEC and NPC (37). Chen et al. performed RNA sequencing on fresh frozen tissues from eight patients with PLEC, and found that BART5-3P and BART20–3 were upregulated in PLEC, but not BART. The region between BART5-3P and BART20-3P is the dominant integration site of EBV. Functionally, miR-BART5-3p suppresses TP53-mediated apoptosis by directly targeting the 3′UTR of p53 mRNA, while miR-BART20-3p downregulates MICB, an NKG2D ligand, thereby reducing NK cell-mediated tumor surveillance. Of these eight patients, five had low expression of p53 and programmed death-ligand 1(PD-L1), and the prognosis was poor (30). It has been reported that the high expression of miR-BART in NPC promotes cancer development by targeting various cell and viral genes. Although primarily studied in NPC, the upregulation of miR-BART in PLEC suggests shared mechanisms of EBV-driven tumorigenesis (38).

Chen et al. identified 288 integration breakpoints of EBV on chromosomes after whole exome sequencing in 128 PLEC patients. They found that the intergenic regions were the preferred integration sites of EBV in PLEC. EBV was easily integrated into the intergenic and intronic regions of two up-regulated miR-BARTs, namely BART5-3P and BART20-3P. These fragile regions were susceptible to DNA damage, which increases the possibility of EBV DNA insertion into the host genome and contributes to the occurrence of PLEC (30). Wu et al. identified 179 EBV host integration sites by bioinformatics methods, only 7 sites were validated by targeted PCR amplification and Sanger sequencing (29). These validated integration events occurred in three patients in advanced stages of PLEC (stages III and IV), suggesting to us that EBV integration may be somehow associated with the stage of tumor progression. The investigators identified an integration hotspot on chromosome 4q28.3 subband, a finding that is particularly striking because it may point to a key oncogenic mechanism. The hotspot was integrated twice in the same patient, further highlighting its potential importance in tumourigenesis. In addition, another integration breakpoint was located in the adjacent subband 4q31.21, which further narrowed the scope of the study and allowed scientists to more precisely investigate the effects of these integration events on gene expression and function (29). In summary, the above studies not only revealed the integration sites of EBV in host cells, but also preliminarily explored the effects of these integration events on gene expression and function. These findings provide new clues for a comprehensive insight of the mechanism of EBV-associated tumourigenesis, as well as potential targets for future therapeutic strategies.

Although the immune system can largely control EBV infection, the virus cannot be eliminated. To produce new viral progeny, EBV is reactivated from the latently infected cells. After analyzing the copy number variations of 46 cases of PLEC, Hong et al. found that the narrow region of 9p21.3 (chr9: 22028316-22041442) showed a focal and significant deletion, and the nearby region within 9p21.3 also showed a high-frequency deletion involving the type I interferon (IFN) gene. Type I IFN is a frontline defense against viral infection and a key component of host-virus confrontation. The level of CD8⁺ tumor-infiltrating lymphocytes (TILs) in tumors with the 9p21.3 deletion was lower than that in tumors without the 9p21.3 deletion. This suggests that frequent loss of type I IFN gene may lead to a lack of host immune response to the virus and persistent EBV infection in PLEC (44). These findings suggest that deletions in immune-regulatory loci, such as type I IFN genes on 9p21.3, may contribute to persistent EBV infection and immune evasion in PLEC (49).

EBV-encoded oncoproteins regulate cellular epigenetic mechanisms to reprogram viral and host epigenetic genomes, particularly in the early stages of infection (50). Abnormal epigenetic modifications mainly include CpG methylation and histone modification. In epithelial cells, LMP1 can upregulate DNA methyltransferase (46). Although LMP1 is highly expressed in PLEC (15, 41, 44), its carcinogenic effect through the regulation of epigenetic mechanisms needs further investigation.

EBV contributes to PLEC tumorigenesis through a sequential cascade involving epigenetic reprogramming, oncogenic signaling activation, and dysregulation of host gene expression. Thus, we propose a model for EBV-driven PLEC as shown in Figure 1 . Initially, EBV infection alters epigenetic marks including DNA methylation and histone modifications in bronchial or alveolar epithelial cells, thereby disrupting normal gene expression to promote uncontrolled proliferation, and impair apoptosis. These epigenetic changes facilitate the activation of multiple oncogenic pathways, such as NF-κB, PI3K/AKT, and JAK/STAT, mediated by viral proteins like LMP1 and LMP2A. Furthermore, EBV-encoded gene products—including EBNA1, BARF1, and BART miRNAs—enhance tumor progression by fostering immune evasion, chronic inflammation, and resistance to apoptosis. Together, these mechanisms form an integrated network driving the initiation and advancement of PLEC.

Numerous copy number variations (CNVs) have been identified in PLEC, and their biological and clinical impacts are summarized in Table 2 . The CNVs, characterized by frequent gains in oncogenic regions and losses in tumor suppressor loci, play a crucial role in shaping the molecular landscape and immunogenicity of PLEC.

Based on the mutation profile, key signaling pathways associated with PLEC have been suggested in PLEC, such as cell cycle, JAK/STAT and NF-κB pathways. Mutations or deletions of TP53, amplification of MDM2 and CCND1, deletions of CDKN2A/B and RB1 are implicated in the dysregulation of cell cycle in PLEC (42, 44, 53, 54). The frequent dysregulation of the JAK/STAT pathway in PLEC is primarily due to the deletion of CISH, followed by mutations or deletions of PTPRD, and mutations or amplification of JAK2 (44). CISH encodes a cytokine-induced SH2-containing protein from the SOCS family, which serves as a key negative regulator of the JAK/STAT pathway (55). PTPRD encodes a tumor suppressor that negatively regulates JAK/STAT pathway by dephosphorylation and inactivation of STAT3 (56). In addition to deletions of key negative regulators of NF-κB signaling pathway (including TRAF3, CYLD, NFKBIA, and NLRC5), somatic mutations or amplification of several components of the canonical NF-κB pathway, including FADD, TRAF2, TRAF6 and CARD11, have been identified in PLEC. FADD is an apoptosis adaptor protein that activates the NF-κB signaling pathway through recruitment of caspase-8. TRAF2 and TRAF6 are members of the TRAF protein family that mediate NF-κB signaling pathway activation and participate in the regulation of inflammation, antiviral response and apoptosis (57).

Interestingly, studies have identified mutual exclusivity between LMP1 overexpression and the three key pathway (cell cycle, JAK/STAT and NF-κB) aberrations (44). Furthermore, several regulatory genes involved in immune escape, such as CD274 (PD-L1) and PDCD1LG2 (PD-L2), were found to be amplified in PLEC (44, 58). These findings suggest that both somatic alterations and viral factors may collaborate in the tumorigenesis and progression of PLEC.

Studies have shown that approximately 73.9% of Chinese patients with NSCLC harbor at least one actionable mutation, as defined by the National Comprehensive Cancer Network guidelines, including mutations in the epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), Kirsten rat sarcoma viral oncogene (KRAS), v-raf murine sarcoma viral oncogene homolog B1 (BRAF), and repressor of silencing 1 (ROS1) (60). EGFR mutations, including point mutations and small insertions/deletions, were the first identified pharmaceutically targetable mutations in NSCLC and remain the most widely used predictive biomarker for EGFR tyrosine kinase inhibitors. The most common EGFR mutations associated with sensitivity to tyrosine kinase inhibitors include the deletion of exon 19 (approximately 45% of patients with EGFR mutations) and the L858R mutation in exon 21 (approximately 40%) in NSCLC (61). In PLEC, the identified EGFR mutations include the L858R mutation in exon 21, non-L858R mutation in exon 21, deletions in exon 19, mutations in exon 20, and mutations in exon 18 ( Table 3 ) (54, 62–64). Overall, alterations in EGFR and ALK genes are relatively rare in primary PLEC However, there may still be some special cases with EGFR or ALK gene alterations that interact with other factors or are related to individual differences of patients, tumor heterogeneity, or detective methods (65). The current data tentatively suggests that EGFR-targeted therapy is not suitable for patients with advanced PLEC due to lack of the typical driver mutations as observed in NSCLC. The low prevalence of classic driver mutations in PLEC implies that PLEC may be driven by a distinct tumorigenic pathway, rather than by conventional oncogenic mutations found in NSCLC.

The tumor microenvironment (TME) refers to the complex milieu surrounding a tumor, encompassing cellular components, extracellular matrix (ECM), and vascular networks, which collectively influence tumor initiation, growth, and metastasis ( Figure 2 ). TME usually includes immune cells, stromal cells, ECM and other secreting molecules, blood and lymphovascular network. The immune cell population within the TME includes T cells, B cells, tumor-associated macrophages (TAMs), dendritic cells (DC), natural killer cells (NK), neutrophils and myeloid suppressor cells (MDSCs), and others. The various subsets of immune cells in the TME exhibit distinct functions, which influence tumor progression through multiple mechanisms (69). Research has found that the recurrence risk of stage I NSCLC may be attributed to alterations in the immune and metabolic microenvironment (70). PLEC is a malignant epithelial tumor characterized by pronounced lymphocytic infiltrates. Kasai et al. observed TILs in EBV-positive PLEC. Most CD3-positive T cells were labeled as CD8 and TIA-1 positive but were negative for granzyme-B, indicating that TILs are resting cytotoxic T-lymphocyte (CTLs) (71). Chang et al. analyzed the lymphocyte composition in the stroma surrounding PLEC tumor cells by immunohistochemistry, and they found that CD8⁺ cells and B cells were present in all cases. In total, the number of stained cells of CD8 positive cells exceeded that of B cells (72). The results of the study by Kobayashi et al. showed that in PLEC, infiltrated T cells were more than B cells, and CD8 positive cells were more than CD4 positive cells (73). Yo Kawaguchi reported a case about a 70-year-old male with PLEC, in which tumor regression occurred during treatment with selective serotonin reuptake inhibitors (SSRIs). Subsequently, histological examination revealed infiltration of CD3⁺, CD4⁺ and CD8⁺ lymphocytes around the tumor. It has been hypothesized that SSRIs may activate these lymphocytes, potentially leading to spontaneous tumors regression (74). There are also studies reporting the impact of neoadjuvant therapy on PD-L1 expression and CD8⁺ lymphocyte density in NSCLC (75).Studies have demonstrated that EBV-specific CD8⁺ TILs in EBV-driven PLEC exhibit heterogeneity and partial deficiency in PD-1 expression (76). Furthermore, other researchers have identified polyclonal plasma cells, polymorphonuclear granulocytes, and CD56⁺ NK cells, alongside CD3⁺ T-lymphocytes and CD79a⁺ B-lymphocytes among the infiltrating cells in PLEC (77, 78). To date, the relationship between lymphocyte infiltration and patient prognosis remains debated in PLEC. The distribution of various lymphatic subpopulations in PLEC requires further elucidation through single-cell sequencing and large multicenter studies (79).

TAMs are among the tumor-infiltrating immune cells (TIIC) within tumor microenvironment. TAMs are derived from MDSCs or monocytes and play pivotal roles in promoting tumor growth, metastasis, angiogenesis, and immunosuppressive functions (80). PLEC is closely associated to a variety of mononuclear inflammatory cells, including a substantial population of TAMs, which contribute to its pathophysiology. The study of Wong et al. demonstrates that the expression of monocyte chemoattractant protein-1 (MCP-1) in PLEC tumor cells plays a critical role in the recruitment of TAMs in PLEC (81). Wang et al. show that pre-treatment monocyte-to-lymphocyte ratios (MLR) may serve as an independent prognostic marker in patients with PLEC and could guide the optimization of treatment strategies (82). TAMs in PLEC tumors are thought to contribute to the inhibition of anti-tumor immune response. However, given the limited sample sizes in previous studies, this hypothesis requires further validation in larger cohort studies.

The programmed cell death protein 1 (PD1) - PD-L1 axis presents a critical immune checkpoint pathway, which can be hijacked by cancer cells to evade immune surveillance (83, 84). PD-L1 expression is a biomarker for improving the survival of advanced PLEC and the potential effectiveness of immunotherapy (85). Blockade of PD1-PD-L1 axis in NSCLC has led to durable objective response and significantly improved survival compared to conventional therapies (86). The combination of PD-1/PD-L1 and lymphocyte activation gene 3 (LAG-3) is associated with the clinical activity of immune checkpoint inhibitors (ICIs) in metastatic PLEC (87). Studies have shown that PD-L1 expression in PLEC is elevated compared to the average levels observed in NSCLC, suggesting that immunotherapy may be a promising treatment strategy for PLEC (18, 64). However, the prognostic value of PD-L1 expression in PLEC tumor cells remains controversial (88, 89). The discrepancies in findings across studies may be attributable to several factors, including the types of PD-L1 antibodies used, varying thresholds for defining positive expression, and differences in specimen collection methods ( Table 4 ). These differences highlight the urgent need for standardized PD-L1 assessment protocols in PLEC, Multi-center, large-sample clinical studies aiming to address the consistency of efficacy of different PD-L1 expression cut-offs and PD-L1 antibodies have become important for the immunotherapy of PLEC.

Currently, there are no standardized treatment protocols for PLEC due to its rarity. However, its association with EBV infection, high PD-L1 expression, and activation of oncogenic signaling pathways such as NF-κB suggests several potential therapeutic strategies ( Figure 3 ).

ICIs have shown promise in EBV-associated tumors and may offer clinical benefit in PLEC (90, 91). Case series and retrospective analyses have reported durable responses to PD-1/PD-L1 blockade in patients with advanced or relapsed disease. Nonetheless, variability in PD-L1 detection methods, cut-off thresholds, and tumor heterogeneity complicates patient selection. Refinement of biomarker assessment—including co-expression of cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), Lymphocyte Activation Gene 3 (LAG3), or tumor mutational burden (TMB) status—may help identify responders more accurately (92).

Given that LMP1-mediated activation of the NF-κB pathway is a hallmark of EBV-driven oncogenesis, this axis represents another promising target. Preclinical studies in related EBV-positive malignancies have explored the use of NF-κB pathway inhibitors, such as proteasome inhibitors or natural compounds (93), though their clinical efficacy in PLEC remains untested (94, 95).

In addition, EBV-directed therapies are under development. Approaches such as lytic induction therapy combined with antiviral agents (e.g., ganciclovir), adoptive transfer of EBV-specific CTLs, and EBV peptide vaccines have shown encouraging results in NPC and post-transplant lymphoproliferative disorders (96–98). These modalities could be translated to PLEC with further validation (18, 95).

Combination therapies, such as ICIs plus chemotherapy, radiotherapy, or anti-angiogenic agents, may also enhance antitumor responses by modulating both tumor and immune compartments (99). Given the immunologically active microenvironment of PLEC, multimodal regimens may offer a rational approach, though prospective clinical trials are needed (92).

Briefly, these emerging strategies highlight the importance of tailoring treatment based on both viral and immune profiles in PLEC. Future research should focus on validating these approaches in clinical settings and establishing disease-specific therapeutic guidelines.