BCG instillation initiates a potent innate immune response marked by the recruitment and activation of macrophages, neutrophils, natural killer (NK) cells, and dendritic cells (DCs). Macrophage polarization is central to the immunologic outcome: M1 macrophages promote antitumor activity through pro-inflammatory cytokine release and direct tumoricidal effects, whereas M2 macrophages facilitate immune evasion and tumor progression via immunosuppressive signaling. The dynamic interchange between M1 and M2 phenotypes is regulated by cytokines and interactions with regulatory T cells (Tregs) within the TME. Neutrophils contribute to tumor cell death through the release of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and tumor necrosis factor-alpha (TNF-$\alpha$), and further amplify immune responses by activating NK cells, T cells, and macrophages via secretion of cytokines. BCG-activated NK cells (CD3⁻/CD56⁺) mediate direct cytotoxicity against tumor cells through the release of perforin and interferon-gamma (IFN-$\gamma$), which promotes the recruitment and differentiation of CD8⁺ T cells. NK cells also enhance CD8⁺ T-cell responses by upregulating major histocompatibility complex class I (MHC-I) expression on tumor cells, increasing their susceptibility to cytotoxic killing. DCs serve as a critical link between innate and adaptive immunity by activating NK cells and presenting tumor antigens to CD8⁺ T cells, and they sustain CD8⁺ T-cell function through interleukin-12 (IL-12) secretion. BCG further promotes DCs survival via nuclear factor kappa-light-chain-enhancer of activated B cells (NF-$\kappa$B), dependent anti-apoptotic pathways, although its role in CD4⁺ T-cell differentiation remains incompletely understood [7].

Fig. 1 summarizes the key components and interactions involved in the innate immune response following BCG instillation.

Fig. 1.

Brief description of innate immune response following BCG instillation. (A) Macrophages exhibit a dual role in the bladder tumor microenvironment, with their antitumor or pro-tumorigenic effects largely dependent on their polarization state. M2 macrophages promote immune evasion and tumor progression, while M1 macrophages support antitumor immunity. (B) Neutrophils contribute to tumor cell death through the release of TRAIL and TNF-$\alpha$, and further enhance immune activation by stimulating T cells, NK cells, and macrophages through the secretion of various cytokines. (C) BCG-activated NK cells (CD3⁻/CD56⁺) exert direct cytotoxicity on tumor cells via perforin release and indirectly promote CD8⁺ T-cell activation through IFN-$\gamma$ secretion. NK cells also upregulate MHC class I expression on tumor cells, rendering them more vulnerable to CD8⁺ T-cell-mediated cytotoxicity. (D) Dendritic cells serve as an important link between innate and adaptive immunity by activating NK cells and presenting antigens to CD8⁺ T cells. BCG enhances DC survival via NF-$\kappa$B-mediated anti-apoptotic pathways. The influence of DCs on CD4⁺ T-cell differentiation remains unclear and may require additional signals beyond cytokine profiles. BCG, Bacillus Calmette-Guérin; NK cells, Natural killer cells; TRAIL, Tumor necrosis factor-related apoptosis-inducing ligand; TLRs, Toll-like receptors; IL, Interleukin; TNF-$\alpha$, Tumor necrosis factor alpha; IFN-$\alpha$, Interferon-alpha; IFN-$\gamma$, Interferon gamma; MHC-I, Major histocompatibility complex class I; DCs, dendritic cells; Th1, type 1 T helper; Th2, type 2 T helper; PRRs, Pattern recognition receptors; CD, Cluster of differentiation.

The adaptive immune response to BCG therapy involves the coordinated activation of multiple immune cell subsets, including CD4⁺ and CD8⁺ T cells, Tregs, gamma delta T cells ($\gamma$$\delta$ T cells), and B cells. Th1-polarized CD4⁺ T cells play a central role in initiating antitumor immunity, while CD8⁺ cytotoxic T cells contribute to direct tumor cell lysis. Tregs, characterized by FOXP3 and CD25 expression, suppress CD8⁺ T-cell activity and are enriched in BCG-refractory tumors, potentially contributing to immune escape. $\gamma$$\delta$ T cells represent a distinct T-cell subset that differs structurally and functionally from conventional T cells. They recognize antigens independently of MHC molecules, enhance DCs function, and secrete IFN-$\gamma$, thereby supporting antitumor responses and improving BCG efficacy. B cells may also exert a protective role within the TME, as high B-cell infiltration has been associated with reduced recurrence rates. However, the contribution of humoral immunity remains uncertain due to conflicting data regarding the role of B cells in tumor response [7].

Fig. 2 summarizes the key cellular components and interactions involved in the adaptive immune response following BCG instillation.

Fig. 2.

Brief description of adaptive immune response following BCG instillation. BCG therapy activates a range of innate immune cells, including natural killer cells, macrophages, neutrophils, and dendritic cells. Dendritic cells function as key antigen-presenting cells, driving CD4⁺ T-cell differentiation toward a Th1 phenotype and initiating IFN-$\gamma$ secretion. IFN-$\gamma$ enhances tumor immunogenicity by promoting apoptosis and increasing tumor antigen presentation, while $\gamma$$\delta$ T cells contribute to immune activation. Although B cells may produce anti-BCG antibodies, their role in adaptive immunity remains incompletely understood. BCG, Bacillus Calmette-Guérin; Treg, regulatory T cell; IFN-$\gamma$, Interferon-gamma; $\gamma$$\delta$ T cells, Gamma delta T cells; Th1, type 1 T helper; Th2, type 2 T helper; IgG, Immunoglobulin G. The dotted black arrow represents repeated stimulation of immune cells.

Myeloid-derived suppressor cells (MDSCs), which are immature precursors of neutrophils and basophils, undergo abnormal expansion in response to tumor-derived signals that inhibit their differentiation. Unlike transient expansion during acute inflammation, cancer-associated MDSCs remain persistently activated and inhibit T-cell responses [15]. Beyond their immunosuppressive functions, MDSCs contribute to tumor progression by enhancing angiogenesis, remodeling the extracellular matrix, and supporting tumor growth [16]. Accordingly, bladder cancer patients demonstrate increased circulating MDSCs and robust infiltration of these cells within tumor tissue [17]. In a clinical study of NMIBC patients undergoing BCG therapy, urine samples collected before and four hours after BCG instillation revealed that a low T-cell to MDSC ratio (1) correlated with higher recurrence rates and a Th2-skewed cytokine profile [18].

BCG has been shown to induce the production of chemokines such as C‑X‑C motif chemokine ligand 8 (CXCL8) and C‑C motif chemokine ligand 22 (CCL22) by urothelial cells, macrophages, and dendritic cells, thereby promoting the recruitment of MDSCs to the tumor microenvironment [19]. Furthermore, combining BCG with anti‑PD‑L1 therapy has been shown to reduce MDSC infiltration within the tumor, highlighting the complex dynamics of immune modulation in the TME [20]. Together, these observations suggest that BCG’s capacity to reshape the TME through modulation of MDSC populations may play an important role in determining its therapeutic efficacy.

Tumor‑associated macrophages (TAMs) play a central role in tumor progression, particularly when polarized toward the M2 phenotype. M2‑TAMs suppress CD8⁺ T‑cell activity through PD‑L1 expression and the secretion of immunosuppressive cytokines such as IL‑10 and TGF‑$\beta$1, promoting regulatory T‑cell recruitment. They also remodel the extracellular matrix, secrete VEGF to drive angiogenesis, and interact with cancer and endothelial cells to form tumor microenvironment of metastasis (TMEM) structures that facilitate metastatic dissemination [21]. TGF-$\beta$ secreted by TAMs inhibits the activity of NK cells, DCs, and effector T cells, while promoting the differentiation of CD4⁺ T cells into a Th2 phenotype, further enhancing immunosuppression [22].

There are four distinct subtypes of M2-TAMs, each with unique phenotypic characteristics. Some subtypes secrete epidermal growth factor (EGF), which activates epidermal growth factor receptor (EGFR) signaling in tumor cells, enhancing pseudopod formation and promoting metastasis. Importantly, TAMs can facilitate DNA repair in tumor cells following chemotherapy-induced damage, thereby contributing to the development of drug resistance [23]. TAMs originate from hematopoietic stem cells in the bone marrow or from erythromyeloid progenitors in the yolk sac and fetal liver, and their functional roles may vary depending on their origin [24]. Clinically, a high density of TAMs infiltrating the TME has been associated with poor outcomes in patients with NMIBC undergoing BCG therapy. In a study involving 71 patients with NMIBC, higher TAM counts within tumors were associated with decreased recurrence-free survival [25]. Another study involving 99 NMIBC patients showed that high TAM infiltration in the tumor stroma was linked to a twofold increased risk of BCG failure [26]. Additional data suggest that the presence of M2-polarized TAMs within the TME may compromise the therapeutic effectiveness of BCG in NMIBC, with their abundance being linked to tumor recurrence as demonstrated by immunofluorescence-based profiling of M1 and M2 macrophage subsets [27].

Regulatory T cells (Tregs) are immunosuppressive CD4⁺CD25⁺ lymphocytes that arise either as thymus-derived natural Tregs or from naïve CD4⁺ T cells in the periphery. Their differentiation is driven by T-cell receptor recognition of self-antigens presented by MHC molecules, and they are characterized by high expression of the transcription factor FOXP3 and dependence on IL-2 for survival [28].

Tregs suppress antitumor immunity through multiple complementary mechanisms, including secretion of immunosuppressive cytokines such as TGF-$\beta$ and IL-10, which inhibit NK cells, DCs, effector T cells, and $\gamma$$\delta$ T cells [29]. They also induce apoptosis of antigen-presenting cells (APC) via perforin and granzyme B release and limit effector T-cell proliferation through competitive consumption of IL-2 [30]. In addition, Tregs express immune checkpoint molecules, including such as cytotoxic T‑lymphocyte-associated protein 4 (CTLA‑4) and lymphocyte activation gene‑3 (LAG‑3), which suppress APC activation by engaging CD80/CD86. Treg-derived ATP is converted to adenosine by CD39 and CD73, further suppressing immune function through A2A receptor signaling on T cells and Antigen‑presenting cells (APCs) [29, 30]. Although PD-1 is expressed on Tregs, its functional role within the TME remains incompletely defined. Emerging evidence suggests that PD-1 blockade may enhance Treg suppressive activity, implicating PD-1 signaling as a potential modulator of Treg-mediated immune suppression in cancer [31].

In patients with NMIBC treated with intravesical BCG, increased intratumoral Treg infiltration has been associated with shorter recurrence‑free survival, suggesting a potential predictive role for Treg density in tumor recurrence [25]. Consistently, elevated levels of PD‑L1⁺ Tregs were detected in urine during BCG therapy, indicating that BCG may promote Treg recruitment and generate an alternative source of PD‑L1-expressing cells that could facilitate recurrence [32]. Histologic analyses of BCG‑unresponsive tumors further demonstrate dense Treg infiltration in non‑responding lesions [33].

The PD‑1/PD‑L1 axis represents a central immune‑regulatory pathway within the bladder cancer TME. Tumor‑mediated engagement of PD‑1 on effector T cells by PD‑L1 suppresses T‑cell activation and facilitates immune evasion [34]. BCG therapy induces PD‑L1 expression on tumor cells and APCs, including macrophages and DCs, through IL‑6- and IL‑10-mediated Signal Transducer and Activator of Transcription 3 (STAT3), activation, as well as toll-like receptor 4 (TLR4)‑dependent extracellular signal-regulated kinases (ERK) signaling [35]. In parallel, BCG stimulates immune cells to release IFN-$\gamma$, a cytokine known to influence PD-L1 regulation through its interaction with IFN-$\gamma$ receptors expressed on tumor cells. Collectively, these mechanisms promote PD‑1/PD‑L1 interactions that drive CD8⁺ T‑cell dysfunction and impair antitumor immunity [36].

In bladder cancer models, BCG increases PD‑L1 expression on tumor cells, and its combination with anti‑PD‑L1 therapy enhances both the number and function of CD8⁺ T cells [20, 37]. Clinically, PD‑L1 expression correlates with high‑grade tumors, disease progression, and BCG unresponsiveness [38, 39]. Importantly, in a cohort of NMIBC patients identified as BCG-unresponsive, tumor specimens collected before and after BCG instillations revealed PD-L1 overexpression. Interestingly, these tumors also showed dense infiltration by CD8⁺ T cells, suggesting that BCG may impair T cell function without affecting their recruitment to the TME [40]. A recent study analyzing tissue microarrays from 432 patients with BCG-naïve HR-NMIBC found that only 7% of patients had tumors that expressed PD-L1. PD-L1 positivity was defined as membranous staining in $\ge$5% of tumor-infiltrating immune cells within the tumor area. Importantly, PD-L1 expression was not associated with treatment failure following adequate BCG therapy. However, it was linked to more aggressive clinical features, including higher-risk disease and advanced tumor stage. A notable limitation of the study was the use of a single antibody across all tumor samples, which may not adequately capture microenvironmental heterogeneity or allow for nuanced analysis of PD-L1 expression in relation to effector immune cell populations [41].

In high-grade bladder cancer tissues, the malignant phenotype-associated glycan sialyl-Tn (STn), a tumor-associated antigen, is markedly overexpressed. This overexpression is associated with increased infiltration of DCs within the TME, which exhibit an immature phenotype characterized by low expression of MHC class II, CD80, and CD86. These immature DCs demonstrate impaired antigen-presenting capacity and fail to effectively activate immune effector cells, thereby contributing to a diminished antitumor immune response [42].

In a cohort of 94 bladder cancer patients treated with intravesical BCG, primary tumors from patients who experienced disease recurrence exhibited reduced STn expression. Although STn may impair DC-mediated antitumor immunity, it could potentially enhance BCG internalization into tumor cells, supporting the hypothesis that BCG may exert a direct cytotoxic effect on cancer cells. In certain cases, this direct cellular effect may compensate for, or even surpass, the antitumor immune response elicited by BCG therapy. However, these findings should be interpreted with caution due to the limited sample size and the possibility that the observed outcomes may reflect a direct tumoricidal effect of BCG, rather than an immune-mediated mechanism, in a subset of cases [43].

One key mechanism in cancer progression is the upregulation of cyclooxygenase-2 (COX-2) within tumor tissues, which enhances the inflammatory response through increased production of prostaglandin E2 (PGE2), a pro-tumorigenic molecule. PGE2 impairs the function of APCs, promotes a Th2-skewed immune response, and contributes to immunosuppression. Additionally, PGE2 exerts paracrine effects within the TME by inducing angiogenesis, primarily through the upregulation of VEGF secretion [44].

In a bladder cancer model, PGE2 has been shown to be overproduced within the TME, leading to increased infiltration of MDSCs and inhibition of APC function, collectively contributing to a weakened antitumor immune response [45]. In vitro studies have shown that COX-2 inhibitors suppress bladder cancer cell proliferation in a COX-2-dependent fashion. Additional data suggest that COX-2 inhibitors may also exert antitumor effects through mechanisms beyond PGE2 inhibition, including the direct induction of cancer cell apoptosis [46]. Celecoxib was evaluated as an adjuvant therapy for bladder cancer in a randomized clinical trial involving patients with intermediate- or high-risk NMIBC, known as the BOXIT trial. Participants (n = 472) received standard-of-care treatment, BCG induction and maintenance for high-risk disease, and mitomycin C induction for intermediate-risk disease, and were randomized 1:1 to receive either celecoxib 200 mg twice daily or placebo for two years. While recurrence-free rates were similar between the two groups, time to recurrence in patients with T1 tumors was longer in the celecoxib arm. Despite comparable induction and maintenance protocols, not all patients completed the full treatment course as indicated. Furthermore, COX-2 expression levels in tumor tissues were not assessed, which may have influenced the observed outcomes [47]. As PGE2 primarily exerts its effects PGE2 receptor (EP), selective inhibition of these receptors may enhance antitumor strategies and offer a safer alternative to celecoxib, particularly given the cardiovascular safety concerns associated with its use in elderly cancer patients [46].

NANOG, a transcription factor essential for maintaining stem cell pluripotency, may contribute to tumor progression when aberrantly expressed in the cytoplasm of cancer cells. This atypical localization has been implicated in the upregulation of HDAC1 (Histone Deacetylase 1), an epigenetic modifier known to repress immune-related gene transcription. One consequence of this regulatory axis is the suppression of CXCL10, a chemokine that plays a key role in recruiting CD8⁺ cytotoxic T lymphocytes to the TME. In patients with NMIBC receiving intravesical BCG therapy, elevated expression of NANOG and/or HDAC1 is associated with unfavorable clinical outcomes, including reduced recurrence-free and progression-free survival. From an immunological perspective, tumors co-expressing these markers exhibit diminished infiltration of CD8⁺ T cells and a lower abundance of granzyme B⁺ cytotoxic cells, indicative of a more immunosuppressive TME and compromised anti-tumor immunity [48].

Natural Killer Group 2A (NKG2A) is an inhibitory receptor expressed on NK cells, CD8⁺ T cells, and other subsets of T cells. It forms a heterodimeric complex with CD94, another membrane protein also found on immune effector cells. This NKG2A/CD94 complex specifically interacts with HLA-E, a non-classical MHC class I molecule that is frequently overexpressed on cancer cells. The nature of the peptide presented by HLA-E critically influences the outcome of this interaction, potentially delivering either inhibitory or stimulatory signals to immune cells. Notably, cancer cells may present peptides via HLA-E that engage the NKG2A/CD94 receptor complex on cytotoxic immune cells, such as NK cells and CD8⁺ T cells, leading to their functional inhibition and contributing to immune evasion [49, 50]. Upon engagement, the intracellular immunoreceptor tyrosine-based inhibitory motifs (ITIMs) of NKG2A become phosphorylated, leading to suppression of activating signals from receptors such as NKG2D and the T‑cell receptor (TCR), thereby inhibiting the cytotoxic activity of effector immune cells [51]. However, the interaction between NKG2A and HLA-E within the TME is complex and modulated by various factors. These include other HLA molecules and cytokines such as IFN-$\gamma$, which upregulates both NKG2A and HLA-E, and IL-15, which specifically enhances NKG2A expression. Additionally, CD4⁺ T cells may interact with NKG2A via Qa-1, the murine homolog of human HLA-E, leading to inhibition of NK cell activity [49].

Recent studies have investigated the role of the NKG2A/HLA-E axis in patients with NMIBC treated with intravesical BCG. In BCG-unresponsive cases, tumors were heavily infiltrated by NKG2A⁺ NK and T cells, and the addition of autologous tumor organoids restored antitumor activity, suggesting a potential therapeutic role for NKG2A blockade in overcoming BCG failure [52].

Monalizumab is a humanized monoclonal antibody that blocks NKG2A, thereby enhancing NK cell cytotoxic function. It can be combined with cetuximab, an anti- EGFR antibody, to promote antibody-dependent cellular cytotoxicity (ADCC), or with durvalumab, an anti-PD-L1 antibody, to further boost NK and CD8⁺ T cell-mediated cytotoxicity. This therapeutic strategy has shown promising results in metastatic colorectal cancer and non-small cell lung cancer [49, 53], with potential for extension to NMIBC and advanced bladder carcinoma.

CD6 is a surface glycoprotein expressed on T lymphocytes, some NK cells, and subsets of B cells. It binds to CD166, also known as activated leukocyte cell adhesion molecule (ALCAM), which is expressed on tumor cells, APCs, and endothelial cells. CD6 plays a key role in T cell activation, proliferation, and adhesion to APCs and endothelial cells. Additionally, CD6 can modulate TCR-mediated signaling and, under certain conditions, exert a negative regulatory effect on T cell responses [54].

A study using single cell RNA sequencing of NMIBC tumor in patients, both BCG-naïve and BCG-unresponsive disease, has revealed elevated expression of the CD6-ALCAM immune-regulatory pathway prior to BCG therapy. These findings suggest that pre-existing inflammatory signaling within the TME may contribute to resistance to BCG [55]. These findings suggest that inhibition of the CD6-ALCAM axis may restore antitumor immunity and represents a promising therapeutic strategy to overcome BCG unresponsiveness.

In a separate study on T-cell lymphoma, intratumoral injection of a CD6-targeted antibody-drug conjugate (CD6-ADC) led to regression of subcutaneous nodules, demonstrating potent antitumor activity through CD6-ALCAM pathway inhibition [56]. These insights may inform future in vitro and clinical investigations targeting CD6-ALCAM signaling in BCG-unresponsive NMIBC.

Chronic stimulation of CD8⁺ T cells leads to functional exhaustion, characterized by reduced cytotoxicity and diminished secretion of TNF‑$\alpha$, granzyme B, and IFN‑$\gamma$ [57]. This exhausted state is marked by upregulation of inhibitory receptors, including PD‑1 and CTLA‑4, which are highly expressed in tumors from BCG‑unresponsive patients and in infiltrating CD8⁺PD‑1⁺ T cells [58]. In bladder cancer, field cancerization may further exacerbate exhaustion by increasing tumor mutational burden and neoantigen exposure, thereby sustaining immune activation and promoting T‑cell dysfunction [59].

Strandgaard et al. [60] performed a multiomics analysis of urinary and tumor samples collected before and after BCG therapy in a cohort of 156 patients with NMIBC. They identified an immune activation signature following BCG therapy, characterized predominantly by elevated levels of IFN$\gamma$-inducible chemokines, including CXCL9, CXCL10, and CXCL11. Their findings also revealed that patients with high-grade recurrences exhibited elevated urinary levels of immunoinhibitory proteins (CD70, CD5, PD-1) and significantly higher CD8⁺ T cell exhaustion scores (p 0.05), based on genetic profiling correlated with T cell exhaustion [60].

Collectively, these findings suggest that BCG‑unresponsive disease is characterized by a paradoxical immune landscape in which intense pro‑inflammatory signaling coexists with profound T‑cell exhaustion following BCG stimulation. In contrast, durable responders appear to harbor a more favorable baseline TME, defined by lower CD4:CD8 and Th2:Th1 ratios [61].

Cancer‑associated fibroblasts (CAFs) in the bladder cancer TME arise from resident fibroblasts activated by tumor‑derived factors, particularly TGF‑$\beta$1 and insulin-like growth factor 1 (IGF-1). Once activated, CAFs promote tumor growth by secreting interleukin-1$\beta$ (IL-1$\beta$) and TGF‑$\beta$1, remodeling the extracellular matrix (ECM) to increase tissue stiffness, and facilitating invasion. CAFs further drive disease progression by releasing monocyte chemoattractant protein-1 (MCP-1/CCL2) and hepatocyte growth factor (HGF), which enhance metastasis and modulate immune cell recruitment [62].

In addition, CAFs release extracellular vesicles (EVs) that suppress CD8⁺ T cell proliferation and reduce the secretion of key pro-inflammatory cytokines, including IFN-$\gamma$, IL-2, and TNF-$\alpha$, thereby modulating the immune response. These EVs also carry PD-L1, which interacts with PD-1 receptors on CD8⁺T cells, impairing their cytotoxic function [63]. CAFs themselves can express PD-L1, further contributing to immunosuppression through PD-1/PD-L1 signaling [64].

Following BCG therapy, fibroblasts contribute to granuloma formation and inflammatory signaling within the TME, yet their direct modulation by BCG remains unstudied, representing an important knowledge gap given their role in immune evasion and therapeutic resistance [65].

Figs. 3,4 summarize the key mechanisms of immune evasion that contribute to BCG failure in patients with NMIBC.

Fig. 3.

Mechanisms of immune evasion contributing to BCG failure in NMIBC. (A) Immunosuppressive functions of MDSCs. BCG stimulates the release of chemokines such as CXCL8 and CCL22 from urothelial cells, macrophages, and dendritic cells. These chemokines recruit MDSCs to the tumor microenvironment, where they remain persistently activated and suppress T-cell responses, contributing to immune evasion. (B) Immunosuppressive functions of TAMs. M2-polarized TAMs contribute to tumor progression by suppressing immune responses and remodeling the extracellular matrix within the tumor microenvironment. These cells express PD-L1 and secrete immunosuppressive cytokines such as IL-10 and TGF-$\beta$, which inhibit effector immune cells and promote the recruitment of regulatory T cells. Additionally, they release pro-angiogenic factors like VEGF, facilitating neovascularization and metastatic spread. TAMs also engage in direct interactions with cancer and endothelial cells to form tumor microenvironment of metastasis (TMEM) structures, which support tumor cell dissemination. (C) Immunosuppressive functions of regulatory T cells. Tregs express CTLA-4, which binds to CD80/CD86 on APCs. This interaction inhibits APC activation and reduces their ability to stimulate effector T cells. Tregs express high levels of CD25, the alpha chain of the IL-2 receptor. By consuming IL-2, they deprive other T cells of this crucial growth factor, limiting their proliferation and survival. Tregs release ATP, which is enzymatically converted to adenosine by CD39 and CD73. Adenosine binds to A2A receptors on T cells and APCs, suppressing their activation and cytokine production. Tregs can induce apoptosis in APCs and effector T cells by releasing granzyme B and perforin. (D) Immunosuppressive functions of PD-1/ PD-L1 pathway. BCG binds to TLR4 on tumor cells, activating the ERK signaling cascade and inducing PD-L1 overexpression. Concurrently, BCG activates APCs, including macrophages and DCs, leading to the secretion of IL-6 and IL-10. These cytokines induce STAT3 phosphorylation, which subsequently enhances PD-L1 expression on APCs. BCG also stimulates immune cells to secrete IFN-$\gamma$, which acts on tumor cells via IFN-$\gamma$ receptors, contributing to additional PD-L1 expression. The cumulative effect of these pathways facilitates PD-1/PD-L1 interactions, leading to CD8⁺ T cell dysfunction and impaired anti-tumor immune responses. (E) Immunosuppressive functions of tolerogenic DCs. In high-grade bladder cancer tissue, the malignant phenotype-associated glycan STn is overexpressed. This overexpression correlates with an increased presence of immature DCs in the tumor microenvironment, characterized by low expression of MHC-II, CD80, and CD86. These immature DCs are functionally impaired in their ability to activate other immune cells, unlike mature DCs, thereby contributing to a reduced antitumor immune response. (F) Immunosuppressive activity mediated by the secretion of immunomodulatory molecules into the tumor microenvironment. Increased PGE2 production in tumor microenvironment impairs APCs function, skews immunity toward a Th2 phenotype, suppresses antitumor responses, and induces angiogenesis via VEGF secretion. In the bladder cancer TME, cytoplasmic NANOG expression in cancer cells promotes transcription of HDAC1, leading to downregulation of CXCL10, a chemokine critical for recruiting CD8⁺ T cells. BCG, Bacillus Calmette-Guérin; MDSCs, Myeloid-derived suppressor cells; CXCL8, C-X-C motif chemokine ligand 8; CCL22, C-C motif chemokine ligand 22; DCs, Dendritic cells; TAM, Tumor-associated macrophages; VEGF, Vascular Endothelial Growth Factor; ECM, extracellular matrix; PD-L1, Programmed death-ligand 1; IL, Interleukin; TGF-$\beta$, Transforming growth factor-beta; Tregs, regulatory T cells; CD, Cluster of differentiation; CTLA-4, Cytotoxic T-Lymphocyte-Associated Protein 4; APCs, Antigen-presenting cells; MHC, Major Histocompatibility Complex; PD-1, Programmed Death-1; TLR4, Toll-like receptor 4; STAT3, Signal transducer and activator of transcription 3; ERK, Extracellular signal-regulated kinases; IFN-R, IFN receptor; IFN-$\gamma$, Interferon-gamma; STn, Sialyl-Tn antigen (cancer-associated glycan); PGE2, prostaglandin E2; Th2, type 2 T helper; HDAC1, Histone Deacetylase 1; NANOG, Nanog Homeobox (a transcription factor involved in stem cell pluripotency); CXCL10, C-X-C Motif Chemokine Ligand 10; NK cells, Natural killer cells; CD8⁺ T cells, Cluster of differentiation 8 positive T cells. Note that the red horizontal line indicates functional blockade, and the ‘X’ symbols denote inhibition or cessation of the corresponding function.

Fig. 4.

Mechanisms of immune evasion contributing to BCG failure in NMIBC (continued). (A) Immunosuppressive signaling via the NKG2A/HLA-E checkpoint axis. This involves the inhibitory receptor NKG2A, which is expressed on NK cells and CD8⁺ T cells. NKG2A forms a heterodimer with CD94 that binds to HLA-E, a non-classical MHC class I molecule frequently overexpressed on cancer cells. Peptides presented by HLA-E modulate this interaction, triggering phosphorylation of the ITIMs within NKG2A, thereby suppressing cytotoxic immune responses in the tumor microenvironment. (B) Immune suppression mediated through the CD6-ALCAM signaling pathway. Immune suppression in bladder cancer may be mediated through the CD6-ALCAM signaling pathway, wherein CD6⁺ T cells and NK cells interact with ALCAM-expressing tumor cells. This interaction modulates TCR signaling and may negatively regulate T cell activation and proliferation within the tumor microenvironment. (C) Immune suppression mediated by effector T cell exhaustion. Prolonged stimulation of CD8⁺ T cells by tumor antigens leads to functional exhaustion, characterized by reduced cytokine secretion and increased expression of inhibitory receptors on T cells, such as PD-1 and CTLA-4. (D) Immune suppression mediated by cancer associated fibroblasts. CAFs within the tumor microenvironment promote tumor progression by remodeling the ECM and secreting cytokines and growth factors that enhance cancer cell proliferation and metastasis. They also suppress CD8⁺ T cell function through the release of EVs carrying PD-L1 and by directly expressing PD-L1, leading to impaired cytotoxic activity and diminished pro-inflammatory cytokine production. TCR, T Cell Receptor; NKG2A, Natural killer group 2A; NK cells, Natural killer cells; CD8⁺ T cells, Cluster of differentiation 8 positive T cells; CD, Cluster of differentiation; HLA-E, Human leukocyte antigen-E; MHC, Major histocompatibility complex; ITIMs, Immunoreceptor tyrosine-based inhibitory motifs; ALCAM, Activated leukocyte cell adhesion molecule; PD-1, Programmed cell death protein 1; CTLA-4, Cytotoxic T-lymphocyte-associated protein 4; CAF, Cancer-associated fibroblast; TGF-$\beta$1, Transforming growth factor beta 1; IL-1$\beta$, Interleukin-1 beta; ECM, Extracellular matrix; MCP-1/CCL2, Monocyte chemoattractant protein-1/Chemokine (C-C motif) ligand 2; HGF, Hepatocyte growth factor; EV, Extracellular vesicle; IFN-$\gamma$, Interferon gamma; PD-L1, Programmed death-ligand 1; TNF-$\alpha$, Tumor necrosis factor alpha. Note that the red horizontal line indicates functional blockade, and the ‘X’ symbols denote inhibition or cessation of the corresponding function.

Nadofaragene firadenovec (Adstiladrin) is an intravesical gene therapy that employs a non-replicating adenoviral vector to deliver the human interferon alpha-2b (IFN$\alpha$2b) gene directly into urothelial cells. The formulation includes Syn3, a surfactant that enhances viral uptake and transgene delivery, resulting in localized production of IFN$\alpha$2b to stimulate anti-tumor immune responses [66]. IFN$\alpha$2b promotes apoptosis in cancer cells by upregulating TRAIL (TNF-related apoptosis-inducing ligand), which activates caspase-8 and caspase-10 through death receptor-mediated signaling [67, 68]. Within the TME, IFN$\alpha$2b also inhibits angiogenesis by suppressing the expression of basic fibroblast growth factor (bFGF) in cancer cells [69]. Beyond its direct cytotoxic effects, IFN$\alpha$ enhances the function of DCs and NK cells, increases MHC class I expression on tumor cells, and facilitates CD8⁺ T cell-mediated killing through improved antigen recognition. Since bladder cancer cells frequently downregulate MHC I, the activation of NK cells by IFN$\alpha$2b provides an additional and potent mechanism of action for nadofaragene firadenovec [70].

The Phase III multicenter clinical trial (NCT02773849) evaluating the efficacy of intravesical nadofaragene firadenovec led to its FDA approval for patients with high-risk BCG unresponsive disease with CIS [71]. Patients received 75 mL of nadofaragene firadenovec at a concentration of 3 $\times$ 10¹¹ viral particles/mL, administered once every three months for up to 12 months, with continued treatment permitted in the absence of high-grade recurrence. A total of 103 patients with BCG-unresponsive CIS were included in the efficacy analysis. The complete response (CR) rate at any time was 55%, with CR rates of 24% and 19% at 12 and 24 months, respectively. Grade III treatment-related adverse events (TRAEs) occurred in 4% of patients, including approximately 3% with urinary symptoms (urgency, incontinence, bladder spasms) and 1% with syncope or hypertension. No TRAEs exceeding Grade III were reported [71, 72].

Nogapendekin alfa inbakicept-pmln (Antkiva) (N-803), also known as IL-15 superagonist or ALT-803, is a cytokine complex composed of IL-15, the sushi domain of the interleukin-15 receptor alpha (IL-15R$\alpha$), and the Fc portion of IgG1. This engineered fusion enhances the biological activity and half-life of IL-15, thereby amplifying its anti-tumor effects through activation of CD8⁺ T cells and NK cells [73]. IL-15 exerts its immunostimulatory effects by interacting with a receptor complex consisting of IL-15R$\alpha$ (CD25), IL-2R$\beta$ (CD122), and IL-2R$\gamma$ (CD132). Notably, N-803 mimics the physiological trans-presentation of IL-15, allowing it to selectively stimulate immune cells that express IL-2R$\beta$$\gamma$, such as NK cells and CD8⁺ T cells, while bypassing IL-2R$\alpha$, which is highly expressed on Tregs. This selective receptor engagement minimizes Treg expansion and preserves anti-tumor immunity [74]. Recent evidence suggests that intravesical BCG therapy induces trained immunity, a form of innate immune memory that enhances responsiveness within the TME. This trained response is more robust than that of naïve immune cells [75]. The synergistic potential of combining BCG with N-803 lies in the ability of N-803 to further amplify this trained immunity, potentially improving the efficacy of intravesical immunotherapy for NMIBC.

In the phase II, open-label, multicenter QUILT 3.032 study, patients with BCG-unresponsive NMIBC with CIS received intravesical therapy consisting of 400 µg of N-803 combined with 50 mg of BCG. Treatment was administered once weekly for six weeks, followed by additional courses once weekly for three weeks at months 4, 7, 10, 13, and 19. Among the CIS cohort (n = 84), the CR rate at any time was 71%. CR rates at 12 and 24 months were 45% and 37%, respectively. In a safety analysis of patients who received intravesical N-803 monotherapy (n = 10), one patient experienced a Grade III TRAE (cerebrovascular accident). No adverse events exceeding Grade III were reported [76].

Cretostimogene grenadenorepvec also known as CG0070 is a tumor-selective oncolytic adenovirus designed to target bladder cancer cells with dysregulated retinoblastoma (Rb) protein pathways. The virus selectively replicates in these malignant cells, leading to direct tumor cell lysis. This process releases tumor-associated antigens, including neoantigens, into the local TME. Concurrently, the virus delivers a transgene encoding granulocyte-macrophage colony-stimulating factor (GM-CSF), which is synthesized by infected tumor cells. GM-CSF promotes the recruitment and activation of APC, such as DCs and macrophages, facilitating the uptake and presentation of neoantigens to T lymphocytes [77, 78]. This dual mechanism, tumor destruction and immune stimulation, enhances both local and systemic anti-tumor immune responses.

In the phase III BOND-003 trial, patients with BCG-unresponsive NMIBC with CIS (n = 112) received intravesical cretostimogene grenadenorepvec (CG0070) at a dose of 1 $\times$ 10¹² viral particles. Treatment consisted of weekly instillations for six weeks, followed by maintenance courses of three weekly doses every three months during year one, and every six months during year two. The CR rate at any time was 75%, with estimated CR rates of 46% and 42% at 12 and 24 months, respectively. No grade III TRAEs were reported in the study [79].

EG-70 (detalimogene voraplasmid) is a non-viral plasmid-based immunotherapy administered directly into the bladder. Following instillation, the plasmid is primarily internalized by bladder epithelial cells. Once inside, the plasmid is transcribed, leading to the expression of two key immunostimulatory components: IL-12, a cytokine that promotes adaptive immune responses, and double-stranded RNA (dsRNA) mimics, which activate retinoic acid-inducible gene I (RIG-I), initiating innate immune signaling pathways [80]. EG-70 comprises a plasmid, a dually derivatized chitosan polymer that complexes with the plasmid, and a polyethylene glycol-polyglutamic acid coating that surrounds the resulting complex [81].

In the phase II LEGEND trial involving patients with BCG-unresponsive CIS NMIBC (n = 26), EG-70 was administered intravesically at a dose of 40 mg per treatment, delivered in a 50 mL volume at a concentration of 0.8 mg/mL. The treatment schedule included four instillations given during weeks 1, 2, 5, and 6 of a 12-week induction cycle. Patients who remained free of disease progression after this initial cycle were eligible to receive up to three additional 12-week cycles of therapy. Preliminary results demonstrated an overall CR rate of 71%, with CR rates of 67% at 3 months and 47% at 6 months. Notably, no TRAEs of grade III or higher were reported [82].

Fig. 6 illustrates emerging immunomodulatory therapies for BCG-unresponsive NMIBC, along with their proposed mechanisms of action.

Fig. 6.

Mechanisms of action of emerging immunomodulatory therapies for BCG unresponsive NMIBC. (A) Mechanism of action of intravesical nadofaragene firadenovec: Following bladder instillation, the adenoviral vector utilizes Syn3, a surfactant, to facilitate entry into urothelial cells. Once internalized, the vector delivers the IFN$\alpha$2b gene to the nucleus, resulting in local overexpression of interferon alpha-2b. IFN$\alpha$2b binds to interferon receptors (IFNAR) on tumor cells, inducing the expression of TRAIL, which activates caspase-8 and caspase-10, thereby triggering apoptosis. It also suppresses angiogenesis by downregulating bFGF expression in tumor cells. Additionally, IFN$\alpha$2b enhances the function of DCs and NK cells, strengthening both innate and adaptive immune responses to cancer. Furthermore, IFN$\alpha$2b upregulates MHC class I molecules on tumor cells, increasing their immunogenicity and promoting CD8⁺ T cell-mediated cytotoxicity. (B) Mechanism of action of intravesical IL-15 superagonist, N-803 (nogapendekin alfa inbakicept-pmln; Anktiva). N-803 is a fusion cytokine complex that enhances IL-15 activity and stability by combining IL-15, the sushi domain of IL-15R$\alpha$, and the Fc portion of IgG1. It activates CD8⁺ T cells and NK cells by mimicking the trans-presentation mechanism of IL-15, specifically targeting these immune effector cells through their expression of the IL-2 receptor $\beta$$\gamma$ complex (IL-2R$\beta$$\gamma$). (C) Mechanism of action of intravesical Cretostimogene grenadenorepvec (CG0070). Cretostimogene grenadenorepvec selectively replicates in bladder cancer cells with Rb pathway defects, causing tumor cell lysis and release of neoantigens into the tumor microenvironment. Infected cells also produce GM-CSF, which recruits antigen-presenting cells to initiate a systemic anti-tumor immune response. This dual mechanism enhances tumor cytotoxicity by promoting CD8⁺ T cell-mediated immune responses. (D) Mechanism of action of intravesical EG-70 (detalimogene voraplasmid). EG-70 is a non-viral plasmid-based immunotherapy delivered intravesically, where it is taken up by bladder epithelial cells, leading to the expression of interleukin-12 and double-stranded RNA mimics. These molecules activate innate immune pathways via RIG-I and stimulate adaptive immunity by promoting CD8⁺ T cell-mediated tumor clearance. NMIBC, non-muscle invasive bladder cancer; IFN$\alpha$2b, Interferon alpha-2b; Syn3, Surfactant used to enhance adenoviral transduction; IFNAR, Interferon alpha and beta receptors; TRAIL, TNF-related apoptosis-inducing ligand; bFGF, Basic fibroblast growth factor; DCs, Dendritic cells; NK cells, Natural killer cells; MHC class I, Major histocompatibility complex class I; CD8⁺ T cells, Cluster of differentiation 8 positive T cells; AR, Adenovirus receptor; IL-15, Interleukin-15; IL-15R$\alpha$, IL-15 receptor alpha; IL-2R$\beta$, IL-2 receptor beta; IL-2R$\gamma$, IL-2 receptor gamma; Fc, Fragment crystallizable; CG0070, cretostimogene grenadenorepvec; Rb, Retinoblastoma; GM-CSF, Granulocyte-macrophage colony-stimulating factor; DCP, Derivatized chitosan polymers; PEG-PA, polyethylene glycol-polyglutamic acid; RIG-I, retinoic acid-inducible gene I; EG-70, detalimogene voraplasmid. Note that the red solid arrows represent the direct cytotoxic effect on cancer cells exerted by activated CD8⁺ T cells, while the red dashed arrows indicate the delivery of genetic material into the nucleus. The solid black arrows denote the impact on the target cell.