PD-1 (or CD279) is a 55-kDa transmembrane protein composed of 288 amino acids. Its structure includes an extracellular N-terminal IgV-like domain, a transmembrane domain and a cytoplasmic tail (26). The cytoplasmic tail contains two tyrosine-based motifs located at the N- and C-terminal regions (27). Its ligand, PD-L1 (also known as CD274 or B7-H1), is a member of the B7 family, which functions as a 33-kDa type I transmembrane glycoprotein. The extracellular region of PD-L1 consists of 290 amino acids, featuring both IgV and IgC structural domains (28). Under normal physiological conditions, PD-L1 is expressed by macrophages, various activated T- and B-cells, dendritic cells and certain epithelial cells, especially in response to inflammatory conditions (29).

The PD-1 pathway functions as a key negative feedback mechanism in normal immunity. T-cell activation, triggered by the engagement of T-cell receptors (TCRs) with antigen-major histocompatibility complex (MHC) and costimulatory signals, leads to the upregulation of PD-1 on the T-cell surface. Binding of PD-1 by PD-L1 from antigen-presenting cells subsequently transmits inhibitory signals that suppress T-cell activation, curb cell proliferation and diminish cytokine release, thereby contributing to the maintenance of immune tolerance (30). In the context of tumor immune escape, cancer cells exploit this regulatory mechanism through upregulating PD-L1 expression by means of various mechanisms, such as in response to IFN-γ within the tumor microenvironment (31). Tumor-surface PD-L1 engages PD-1 on tumor-infiltrating T cells, delivering inhibitory signals that lead to both T-cell dysfunction and an impairment of the immune-mediated clearance of tumor cells (32). Consequently, the suppression of T-cell activity both enables sustained tumor proliferation and promotes tumor disease progression.

The activity of the PD-1/PD-L1 pathway is regulated through multiple layers of control, including gene transcription, post-transcriptional modifications, PTMs and exosomal transport (33). Among these, PTMs, such as glycosylation, phosphorylation, ubiquitination, palmitoylation, SUMOylation and acetylation, fulfill particularly notable roles in modulating the stability and protein-protein interactions of PD-1 and PD-L1(34). Notably, substantial evidence over the past decade has established that the expression levels of PD-1 and PD-L1 are markedly influenced by ubiquitin-mediated proteasomal degradation. As a key regulatory mechanism, ubiquitination is essential for controlling diverse cellular processes, including protein stability, receptor internalization and immune responses (35).

The subsequent part of the present review delineates the sophisticated roles of E3s and DUBs in dynamically shaping the tumor immune microenvironment (TIME). It details a molecular ‘tug-of-war’, wherein specific E3s [for example, speckle-type POZ protein (SPOP), F-box protein (FBXO)22 and F-box and WD repeat domain containing 7 (FBW7)] target PD-L1 and PD-1 for degradation to potentiate anti-tumor immunity. By contrast, counteracting DUBs [for example, ubiquitin-specific peptidase 22 (USP22), ubiquitin-specific peptidase 51 (USP51) and the DUB OTU and ubiquitin aldehyde binding 2 (OTUB2)] stabilize these checkpoints to foster immune evasion; however, these functions are context-dependent. The present review explores how this ubiquitin-mediated regulation extends beyond the checkpoints themselves to orchestrate key oncogenic signaling pathways [for example, the phosphoinositide 3-kinase (PI3K/AKT) and mitogen-activated protein kinase (MAPK) signaling pathways] and modulate broader immune responses, including phagocytic clearance mediated via cluster of differentiation (CD)47. Through highlighting context-dependent regulators such as ubiquitin-specific peptidase 7 (USP7) and ubiquitin-specific peptidase 12 (USP12), the discussion underscores the therapeutic potential of strategically manipulating this equilibrium, which can be achieved either through augmenting E3 activity or by inhibiting specific DUBs, thereby recalibrating the TIME and overcoming resistance to immunotherapy in lung cancer (Fig. 2). Notably, although TIM-3 and TIGIT represent clinically relevant immune checkpoints in lung cancer, their direct regulation by ubiquitin-modifying enzymes remains largely unexplored.

Beyond simple protein degradation, the ubiquitin system has emerged as a sophisticated signaling network that dynamically shapes the lung cancer immune landscape. E3s function as molecular switches, the activity of which directly determines tumor immune recognition and therapeutic susceptibility (36); however, their effects are not uniformly pro-immunity. SPOP exemplifies how existing drugs can be repurposed to manipulate this system. The ability of canagliflozin to disrupt the sodium-glucose cotransporter-2 (SGLT2)-PD-L1 axis redirects the activity of SPOP toward PD-L1 degradation, thereby creating a window for enhanced T-cell attack (37). Similarly, FBXO22 operates at the intersection of DNA damage response and immunotherapy. The inhibition of cyclin-dependent kinase 5 was unexpectedly shown to strengthen antitumor immunity through FBXO22-mediated PD-L1 control (38). This system demonstrates marked versatility in terms of its regulatory strategies. Although atrophin-1 interacting protein 4 (AIP4) has been shown to collaborate with metastasis suppressor 1 (MTSS1) to channel PD-L1 toward lysosomal degradation (39), ring finger protein 182 (RNF182) and casitas B-lineage lymphoma (Cbl) proteins employ indirect mechanisms. RNF182 dismantles the transcriptional activator p65, whereas Cbl proteins serve to constrain the upstream signal transducer and activator of transcription 3 (STAT3)/AKT/extracellular signal-regulated kinase (ERK) signaling pathway. These diverse approaches converge on the same goal, namely modulating the presence of PD-L1 at the immune synapse.

Not all E3s, however, are involved in pushing toward immune activation. DCUN1D1 (defective in cullin neddylation 1 domain containing 1), also known as SCCRO, functions as an essential co-E3 component of the neddylation E3 complex that promotes cullin neddylation and subsequent CRL activation (40). This E3 enhances PD-L1 expression through activation of the focal adhesion kinase (FAK) signaling pathway, thereby highlighting the delicate balance within this network (41). The collective activity of these competing forces ultimately determines whether the TIME permits or prevents effective immune surveillance, thereby positioning the ubiquitin system as a central arbitrator of immunotherapeutic outcomes in lung cancer.

Beyond transcriptional control, PD-L1 expression is dynamically shaped by DUBs that stabilize PD-L1 through distinct PTMs, thereby creating a notable barrier to effective immunotherapy. Multiple DUBs converge on PD-L1 protection through specialized mechanisms. USP51 directly stabilizes PD-L1 by removing ubiquitin chains at K280/K281, which has the effect of driving dual resistance to both chemotherapy and immunotherapy (42). By contrast, USP22 employs a dual-layer stabilization mechanism: Not only does it deubiquitinate PD-L1 at multiple lysine residues, but it also stabilizes constitutive photomorphogenesis 9 signalosome subunit 5 (CSN5). CSN5 itself functions as a JAMM-family metalloprotease deubiquitinase that directly binds the C-terminal region of PD-L1 and cleaves its ubiquitin moieties via MPN-domain enzymatic activity, thereby preventing proteasomal degradation. Consequently, USP22 fortifies PD-L1 stability through two convergent routes, direct deubiquitination of PD-L1 and indirect stabilization of its upstream guardian CSN5, creating a reinforced USP22-CSN5-PD-L1 signaling circuit that amplifies immune evasion (43,44). OTUB2 operates as an early checkpoint protein that intercepts PD-L1 in the endoplasmic reticulum (ER) to subvert ER-associated degradation, which thereby ensures that it is trafficked to the cell surface (45).

The regulatory landscape has been shown to extend beyond direct protein stabilization. Ataxin-3, a protein encoded by the human ATXN3 gene that primarily functions as DUB, determines PD-L1 upregulation through transcriptional control. It promotes PD-L1 upregulation through two distinct microenvironmental cues. Upon IFN-γ stimulation, it deubiquitinates and stabilizes both interferon regulatory factor 1 (IRF1) and signal transducer and activator of transcription 3 (STAT3), thereby preserving their ability to bind the PD-L1 promoter and enhance its transcription. Under hypoxic conditions, ATXN3 functions as a deubiquitinase for hypoxia-inducible factor-2α (HIF-2α), preventing its proteasomal degradation; the accumulated HIF-2α subsequently translocates to the nucleus and drives PD-L1 transcription through hypoxia-response elements in the PD-L1 promoter. These convergent transcriptional programs ultimately increase PD-L1 abundance and foster immune evasion (46). Similarly, ubiquitin C-terminal hydrolase L1 was shown to enhance PD-L1 expression through AKT-P65 pathway activation, which creates an immunosuppressive microenvironment that dampens T-cell efficacy (47). Notably, USP5 exemplifies how DUB function is integrated with broader cellular systems. Classically known as isopeptidase T (IsoT), USP5 specifically disassembles unanchored polyubiquitin chains, generated during proteasomal processing or aberrant E2/E3 activity, into monoubiquitin units for recycling. When these free chains accumulate, they competitively bind proteasomal ubiquitin receptors (such as S5a/Rpn10), obstructing the recognition and degradation of authentic ubiquitinated substrates and thereby disrupting global proteostasis. By clearing this inhibitory pool, USP5 preserves proteasome capacity and maintains the cellular monoubiquitin reservoir, which indirectly sustains PD-L1 stability and links the general ubiquitin-proteasome system to specific immune checkpoint control. Beyond proteostasis, USP5 also participates in DNA double-strand break repair, Wnt/β-catenin signaling, and NF-κB-mediated inflammatory responses, further coupling immune checkpoint regulation to genome integrity and oncogenic signaling networks (48,49).

This intricate network reveals how DUBs function as sophisticated modulators of tumor immunity. The diverse range of mechanisms that they employ, ranging from direct deubiquitination to transcriptional reprogramming and pathway activation, demonstrate their potential as therapeutic targets. Nevertheless, the impact of DUB inhibition needs to be evaluated in a context-specific manner, as the net effects elicited depend on the particular substrate and cellular environment.

Multiple E3s form a complex regulatory network that both modulates the TIME and responses to anti-PD-1/PD-1 immunotherapy in lung cancer. However, these enzymes exert divergent effects on immune surveillance. Tripartite motif containing (TRIM)28 promotes the recruitment of myeloid-derived suppressor cells (MDSCs) and confers anti-PD-1 resistance. It catalyzes K63-linked ubiquitination of receptor-interacting serine/threonine kinase 1, thereby activating nuclear factor-κB (NF-κB) signaling (50). This process is dependent on its E3 activity. Additionally, the DNA-binding protein TRIM35 enhances immune surveillance in NSCLC. It mediates the K63-linked ubiquitination of lysine specific demethylase 1, which suppresses its demethylase activity. This inhibition subsequently leads to an upregulation of the transcription of the protein ER-Golgi intermediate compartment 1, stabilizes IFN-γ receptor 1 and activates IFN-γ signaling, ultimately leading to an increased expression of MHC-I (51). Similarly, FBW7 ubiquitinates PD-1 via K48-linked polyubiquitination, thereby targeting it for degradation and enhancing anti-PD-1 efficacy (52). This interaction is facilitated by cyclin-dependent kinase 1-mediated phosphorylation of PD-1 at the residue S261. Furthermore, tryptophan metabolism also engages in this cross-talk: Tryptophan or indoleamine 2,3-dioxygenase inhibitors have been shown to promote the tryptophanylation and activation of the E3, thyroid hormone receptor interactor 12 (TRIP12) at K1136. TRIP12 subsequently degrades nuclear factor of activated T cells 1 (NFATc1), which functions as a key transcription activator of PD-1(53). This pathway enhances CD8⁺ T-cell-mediated cytotoxicity and inhibits tumor growth.

The roles of DUBs in lung cancer immunity are context-dependent, as exemplified by the contrasting, yet complementary, functions of USP7 and USP12. Although both are cysteine proteases within the USP family, their inhibition leads to divergent outcomes in the tumor microenvironment, thereby highlighting the nuanced regulation of immunotherapy efficacy. The targeted inhibition of USP7 presents a dual therapeutic advantage; on one hand, it directly suppresses tumor growth through the activation of the p38 MAPK pathway, whereas, on the other hand, it concurrently effects an upregulation of PD-L1 expression (54). This seemingly paradoxical increase in PD-L1 serves to sensitize the tumor to anti-PD-1 antibodies, which thereby transforms the microenvironment into a more responsive state for combination immunotherapy.

Alternatively, USP12 acts as a guardian of antitumor immunity. Its knockdown exacerbates tumor progression through fostering a powerfully immunosuppressive niche. This occurs through enhanced macrophage recruitment, hypervascularization and T-cell inactivation. Mechanistically, a depletion of USP12 disrupts the stability of the protein phosphatase Mg²⁺/Mn²⁺ dependent 1B, which serves as a negative regulator of NF-κB signaling (55). The subsequent hyperactivation of NF-κB then drives the overexpression of pro-tumor chemokines, including CXCL1, CXCL8, CCL2 and CCL5, which promote the recruitment of tumor-associated macrophages and myeloid-derived suppressor cells (55). These chemokines, in turn, orchestrate the immunosuppressive landscape and drive resistance to immune checkpoint blockade (55). Therefore, a loss of USP12 function has the effect of crippling the responsiveness of tumors to anti-PD-1 treatment.

In summary, USP7 inhibition creates a favorable condition that may be combined with existing immunotherapies, whereas USP12 activity is essential for maintaining the capacity to respond to such treatments. This delineation underscores that a strategic targeting of the ubiquitin system requires a precise understanding of each component's unique role in balancing tumor suppression and immune activation.

Ubiquitin-dependent fine-tuning of the PI3K/AKT and MAPK cascades orchestrates the amplitude and duration of signaling that ultimately dictate the PD-1/PD-L1 checkpoint threshold in lung cancer. Within the PI3K-AKT axis, E3s such as WW domain-containing E3 ubiquitin protein ligase 2 promote Akt activation via mediating phosphatase and tensin homolog degradation. This consequently leads to an upregulation of PD-1 expression (56). On the other hand, PD-1 engagement itself attenuates Akt phosphorylation (57), thereby inducing metabolic quiescence in T cells (58). Similarly, ubiquitination has been shown to regulate multiple levels of the MAPK cascade. For example, Cbl-mediated ubiquitination of the epidermal growth factor receptor (EGFR) restricts upstream signaling, whereas the ubiquitin-dependent turnover of MAPK kinase/MAPK shapes downstream output (59). Collectively, these events lead to reduced ERK activity and lower PD-L1 expression (60). Therefore, dynamic ubiquitination and deubiquitination cycles not only modulate oncogenic signaling, but they also serve to calibrate the balance between T-cell exhaustion and reinvigoration, providing a mechanistic rationale for targeting ligases and DUBs to enhance immune checkpoint blockade.

DUBs markedly shape the immunosuppressive lung cancer microenvironment through controlling the stability of pivotal immune checkpoint molecules. Ubiquitin-specific peptidase 2 (USP2) directly deubiquitinates and stabilizes CD47, thereby stabilizing the protein and preserving its anti-phagocytic signal. Either genetic or pharmacological inhibition of USP2 has been shown to promote CD47 degradation, which thereby shifts macrophages toward a pro-inflammatory M1 phenotype, also enhancing CD8⁺ T-cell infiltration (61). In parallel, ubiquitin-specific peptidase 8 (USP8) circumvents the TNF receptor-associated factor 6-mediated K63-linked ubiquitination of PD-L1. Inhibiting USP8 directs PD-L1 toward K48-linked polyubiquitination and degradation, while simultaneously enhancing the NF-κB-dependent expression of MHC-I. Elevated MHC-I surface density increases the presentation of tumor-derived antigenic peptides to CD8⁺ T cells, thereby augmenting antigen presentation and priming cytotoxic T-cell responses (62). Consequently, the dual targeting of USP2 and USP8 disrupts two complementary ubiquitin-dependent immune barriers, namely phagocytic checkpoint and T-cell suppression, thereby effectively converting an immune-excluded tumor niche into an inflamed microenvironment that is susceptible to PD-1 checkpoint blockade.

Notably, the fate of the immune checkpoint PD-L1 is not sealed at its production; rather, its fate is perpetually decided by a PTM ‘tug-of-war’ that is orchestrated within lung cancer cells. One the one hand, the E3s strive to pull PD-L1 toward its degradation, whereas, on the other hand, the DUBs counteract the activity of the E3s, and their function is to ensure the survival and continued activity of PD-L1. This dynamic equilibrium acts as a ‘master regulator’ of tumor immune evasion.

E3s modulate PD-L1 in various ways to cause its degradation, thereby exposing the tumor to immune attack. Ligases such as SPOP (activated by canagliflozin), FBXO22 and AIP4 in combination with MTSS1 directly bind to PD-L1. They decorate PD-L1 with ubiquitin chains, thereby marking it for proteasomal or lysosomal disposal. Other players, such as Cbl-b/c-Cbl and RNF182, however, take a more indirect approach. They do not target PD-L1 itself, but interrupt its production; Cbl-b/c-Cbl dampens the signaling pathways (such as the STAT3/AKT/ERK pathways) that drive PD-L1 transcription (63), whereas RNF182 downregulates a key transcription factor, p65, that is required for the activation of the PD-L1 gene (64).

Conversely, DUBs antagonize E3-mediated ubiquitination by removing ubiquitin chains to rescue PD-L1 from degradation. DUBs such as USP51, USP5 and OTUB2 act as notable defenders. They directly intercept PD-L1, and strip off its ubiquitin chains. A key tactical maneuver is employed by OTUB2, which interacts with PD-L1 during its processing in the endoplasmic reticulum (ER). This prevents its degradation before it even reaches the cell surface. Some DUBs, such as USP22, have been shown to provide a layered defense. USP22 not only directly deubiquitinates and stabilizes PD-L1, but also enhances the stability of CSN5, which itself functions as a PD-L1 stabilizer. This establishes a cooperative positive feedback loop that amplifies PD-L1 abundance (43).

The ultimate level of PD-L1 to be found on the tumor cell surface (and therefore, its ability to suppress T-cells) is the net result of this continuous ‘tug-of-war’. When E3s gain the upper hand, immunity is enhanced (51,52), but when DUBs dominate, the tumor persists. This refined understanding of the crosstalk moves us beyond a static list and reveals a dynamic signaling network. The most promising therapeutic strategies that are coming into prominence now aim to disrupt this balance. In terms of influencing this balance, clinicians can either inhibit the stabilizing DUBs (for example, with USP51 or OTUB2 inhibitors) or activate specific E3s. Through strategically intervening in this tug-of-war, it is possible to tip the scales in favor of the immune system and thereby overcome resistance to immunotherapy.

The success of anti-PD-1 therapy in lung cancer is therefore governed by a dynamic equilibrium in the ubiquitin system, where E3s and DUBs engage in a molecular tug-of-war to control immune recognition. E3s predominantly propel the system toward immune activation (65) through dismantling inhibitory checkpoints. FBW7 and TRIP12 have been shown to degrade both PD-1 and its transcriptional regulator, NFATc1. By contrast, TRIM35 potentiates T-cell surveillance by enhancing IFN-γ signaling. On the other hand, DUBs provide nuanced, context-dependent counterbalances to the action of the E3s. The contrasting roles of USP7 and USP12 exemplify this complexity. Inhibiting USP7 creates a synthetically vulnerable state wherein tumor cells become sensitized to immune checkpoint blockade: USP7 inhibition concurrently suppresses tumor growth through p38 MAPK activation and upregulates PD-L1 expression, thereby priming the tumor microenvironment for enhanced anti-PD-1 efficacy (54). By contrast, preserving USP12 is essential to prevent the collapse of the immune microenvironment into a state of profound immunosuppression (55).

Therefore, the ubiquitin system acts not merely as an on/off switch, but as a sophisticated signaling hub. Therapeutic success hinges on being able to strategically manipulate this hub. This strategic manipulation may be achieved either through reinforcing the ‘pro-immunity’ force of key E3s or by precisely inhibiting the ‘pro-resistance’ function of specific DUBs. Such interventions tip the scales in favor of sustained antitumor immunity (65).

ICIs perform a pivotal role in the management of different types of lung cancer, especially NSCLC. PD-1 inhibitors, such as nivolumab and pembrolizumab, are currently employed in clinical practice based on phase III randomized controlled trials (92,93), while PD-L1 inhibitors, such atezolizumab, have also demonstrated efficacy (94). Furthermore, for the first-line treatment of NSCLC, tremelimumab (Imjudo), a CTLA-4 inhibitor, has been used in combination with duvalumab (95). Drug resistance has gradually emerged with the widespread application of immunotherapy, although only a small proportion of patients with lung cancer respond to immunotherapy, and this in itself poses a new challenge in the treatment of lung cancer (14,96).

Berberine (BBR) is a naturally occurring isoquinoline quaternary alkaloid that is isolated from Coptis chinensis, a species of goldthread flowering plant native to China. It exhibits broad pharmacological properties, including anti-cancer, anti-microbial, anti-diabetic and anti-inflammatory activities (97). In terms of the underlying mechanism, BBR has been shown to enhance T-cell-mediated tumor cell death through downregulating the expression of PD-L1 in cancer cells. Specifically, BBR interacts with the glutamate residue E76 within CSN5, thereby inhibiting its DUB activity. This interaction promotes PD-L1 ubiquitination, and subsequent proteasomal degradation, which effectively disrupts the PD-1/PD-L1 immune checkpoint axis (97). Furthermore, BBR has been shown to impair the MPN domain of CSN5, which is critical for CSN5-mediated stabilization and deubiquitination of PD-L1(98). In addition to its direct effects on PD-L1, BBR also enhances anti-tumor immunity through promoting the infiltration and activation of tumor-specific T cells. It has also been shown to suppress the expansion and function of immunosuppressive cell populations, such as MDSCs and regulatory T cells (97).

P5091 is a selective USP7 inhibitor, which demonstrates potent immunomodulatory activity via reprogramming tumor-associated macrophages through activation of the p38 MAPK signaling pathway, which has the effect of promoting their polarization toward the M1 phenotype. This molecular mechanism enhances cytotoxic T lymphocyte-mediated tumor immune surveillance and ultimately suppresses tumor progression (54). Notably, USP7 inhibition causes an upregulation of PD-L1 expression within the tumor microenvironment, and this creates a synergistic therapeutic window when combined with PD-1 blockade immunotherapy (54).

Cullin 4A (Cul4A) is an 87-kDa protein, that belongs to a family of evolutionarily conserved cullin proteins and has been shown to be associated with the ubiquitin proteasome pathway (99). The virus protease inhibitor PIK-93 functions as a potent PD-L1 destabilizing agent through facilitating the formation of the CUL4A-E3 complex with PD-L1. This molecular interaction promotes the K48-linked polyubiquitination of PD-L1, thereby leading to its subsequent proteasomal degradation through the UPS (100). Administering PIK-93 has been shown to cause a notable downregulation of PD-L1 expression in M1-polarized macrophages, thereby augmenting their tumoricidal activity. In both syngeneic and human peripheral blood mononuclear cell-derived xenograft models, the combined administration of PIK-93 and PD-L1 blockade therapy has been demonstrated to potentiate T-cell-mediated antitumor immunity. This was evidenced both by an enhanced infiltration of TILs and a suppression of tumor progression (100). This combinatorial approach served to promote the establishment of an immunologically active tumor microenvironment, which led to a synergistic improvement of the therapeutic efficacy of immune checkpoint inhibition (100).

Metformin is a widely used anti-diabetic agent. Emerging research has demonstrated that it exhibits notable anti-tumor activity across multiple types of cancer, including lung, prostate and colorectal malignancies (101). Moreover, mechanistic studies have revealed that metformin-mediated stabilization of the adaptor protein, stimulator of interferon genes (STING), and the enhancement of T-cell cytotoxicity are both axis inhibition protein 1 (AXIN-1)-dependent. The genetic ablation of this scaffold protein led to an abrogation of these effects. In serine/threonine kinase 11-mutant lung cancer models, metformin was shown to synergize with the PD-1 blockade via suppressing RNF5-dependent K48-linked polyubiquitination, and the subsequent degradation of STING. This process was shown to require AXIN-1(102). Furthermore, metformin was found to potentiate tumor-specific T-cell responses in vitro, and to augment the therapeutic efficacy of PD-1 inhibitors in both preclinical models and experimental settings.

The antidiabetic drug canagliflozin has been shown to suppress tumor growth by reducing the expression of PD-L1 in an ‘on-target’ manner. In terms of the underlying mechanism, SGLT2 maintains PD-L1 stability during endocytic recycling through its direct interaction with it; however, canagliflozin was shown to have a two-fold effect, namely to disrupt the interaction between SGLT2 and PD-L1 and to promote the interaction between PD-L1 and the E3, SPOP (37).

PD-L1 also forms a complex with glycogen synthase kinase-3β (GSK-3β) and the E3 β-transducin repeat-containing protein (TrCP), leading to phosphorylation-dependent degradation of the UPS (103). The annual herb Centipeda minima (CM) and its active component 6-O-angeloylplenolin (6-OAP) were found to upregulate the expression of PD-L1 via inhibiting GSK-3β/β-TRCP-mediated ubiquitination and degradation, which led to further increases in CD8 T-cell infiltration (104). Therefore, mechanistically speaking, the induced expression of PD-L1 and the enhanced cytotoxicity of CD8 T cells underlie the beneficial effects of 6-OAP-rich CM in NSCLC. Preclinical studies have demonstrated that CM and 6-OAP significantly enhance ICI-induced tumor burden reduction and prolong overall survival in NSCLC-bearing mice when administered in combination with anti-PD-L1 antibody (104). These findings position CM as a promising candidate for combination with ICIs, and a phase I clinical trial (NCT05735028) is currently underway to evaluate its safety and efficacy in patients with lung cancer. As shown in Fig. 3, these diverse strategies converge on a unified therapeutic principle: Strategically tipping the ubiquitin equilibrium toward checkpoint degradation or microenvironment remodeling can overcome immunotherapy resistance in lung cancer (Table II and Fig. 3).