PD‐1 blockade treatment in melanoma: Mechanism of response and tumor‐intrinsic resistance

Abstract

Malignant melanoma is characterized by high immunogenicity, genetic heterogeneity, and diverse pathological manifestations, affecting both skin and mucosa over the body. Pembrolizumab and nivolumab, both anti‐PD‐1 monoclonal antibodies, were approved by the US FDA for unresectable or metastatic melanoma in 2011 and 2014, respectively, with enduring and transformative outcomes. Despite marked clinical achievements, only a subset of patients manifested a complete response. Approximately 55% of melanoma patients exhibited primary resistance to PD‐1 antibodies, with nearly 25% developing secondary resistance within 2 years of treatment. Thus, there is a critical need to comprehensively elucidate the mechanisms underlying the efficacy and resistance to PD‐1 blockade. This review discusses the fundamental mechanisms of PD‐1 blockade, encompassing insights from T cells and B cells, and presents resistance to anti‐PD‐1 with a particular focus on tumoral‐intrinsic mechanisms in melanoma.

PD‐1 antibodies relieve the inhibitory effect on T cells and effectively prolong the survival time, which is a landmark treatment for advanced melanoma. However, due to various inherent factors or adaptive changes of tumor cells, only a subset of patients can benefit from the treatment. This review discusses the fundamental mechanism of PD‐1 antibodies and tumor‐intrinsic resistance mechanisms.

1. INTRODUCTION

Melanoma, originating from melanocytes, is the most lethal form of skin cancer. In Asian populations, primary melanomas originate predominantly from acral sites, accounting for more than 50% of cases.

The programmed cell death 1 receptor (PD‐1), also known as CD279, is well known among numerous immune checkpoint molecules. PD‐1 is expressed widely on immune cells including activated T cells, B cells, exhausted T cells, regulatory T cells (Treg cells), as well as natural killer cells (NK cells) and natural killer T cells (NKT cells). ¹ It has two ligands, programmed death‐ligand 1 (PD‐L1/CD274/B7‐H1) and programmed death‐ligand 2 (PD‐L2/CD273/B7‐DC). While PD‐L1 exhibits widespread expression on both hematopoietic and non‐hematopoietic cells, as well as within the tumor microenvironment (TME), PD‐L2 demonstrates more restricted expression patterns. ² The physiological role of the PD‐1 pathway is to maintain self‐tolerance to autoantigens and regulate immune responses to avoid excessive tissue damage. ³ , ⁴ However, within the tumor microenvironment, melanoma cells exploit PD‐L1 expression to evade immune surveillance that prevents their proliferation, invasion, and migration. ⁵ The PD‐1 pathway significantly impacts T‐cell function through various mechanisms (Figure 1):

1. Intracellular signaling: Within the intracellular region of PD‐1, the immunoreceptor tyrosine‐based switch motif (ITSM) becomes phosphorylated, facilitating the recruitment and activation of Src homology 2 domain‐containing protein tyrosine phosphatase 1/2 (SHP‐1/2). The recruitment of SHP‐2 cooperates with CD3 and/or CD28‐ signaling to impede the phosphorylation of the ZAP70/CD3ζ signalosome induced by the CD4⁺ T‐cell receptor (TCR). ⁶ , ⁷

2. Modulation of the PI3K/AKT pathway: the PI3K/AKT pathway is typically induced by the costimulatory molecule CD28. The PD‐1 pathway interferes with it through the ITSM motif, thereby affecting critical CD4⁺ T‐cell survival, proliferation, and metabolism. ⁸

3. Arrest cell cycle: PD‐1 inhibits the Ras/MEK/ERK and PI3K/AKT pathway to reduce the transcription of S‐Phase Kinase‐Associated Protein 2 (SKP2), leading to CD4⁺ T‐cell cycle arrest at the G1 phase. ⁹

4. Upregulation of negative molecules: PD‐1 upregulates the expression of various negative regulators that contribute to T‐cell dysfunction. For instance, it inhibits T‐cell function in chronic viral infections by activating Basic leucine zipper ATF‐like transcription factor (BATF). Silencing BATF would improve cytokine secretion of both HIV‐specific CD4⁺ and CD8⁺ T cells, and proliferation of CD8⁺ T cells. ¹⁰

5. Impact on T‐cell motility: PD‐1 signaling increases the motility of T cells, which disrupts the stability and duration of interactions between T cells and dendritic cells or tumor cells. This reduced interaction impedes effective T‐cell activation. ¹¹ , ¹² , ¹³ This phenomenon has been observed in inflammatory responses and pancreatic autoimmune reactions to affect CD4⁺ T cells, while in viral infections is particularly impacted on CD8⁺ T cells.

6. Impairment of cytotoxic function: the PD‐1 pathway diminishes CD8⁺ T‐cell sensitivity to TCR signals by restraining the Ca²⁺ flux caused by TCR/peptide–major histocompatibility complex (MHC) complexes. ¹⁴

7. Metabolic reprogramming: PD‐1 signaling inhibits activated CD4⁺ T‐cell glycolysis while enhancing fatty acid oxidation (FAO), resulting in metabolic reprogramming that prevents effector T‐cell development. ¹⁵

8. Regulation of T‐cell subsets: PD‐1 pathway activation induces the transformation of T helper type 1 (Th1) cells into immunosuppressive regulatory T (Treg) cells. ¹⁶

9. Induction of apoptosis: PD‐1 pathway mediates apoptosis in tumor‐antigen‐specific cytotoxic T lymphocytes (CTLs). ¹⁷

10. Exhaustion of tumor‐infiltrating T lymphocytes (TILs): the prolonged PD‐1 pathway results in TIL exhaustion and deprivation of immune surveillance capacity. ¹⁸

FIGURE 1.

Schematic of the molecular interactions and downstream signaling that attenuate of T‐cell function caused by PD‐1 (programmed cell death protein 1), pMHC (peptide–major histocompatibility complex), PD‐L1 (programmed death‐ligand 1), TCR (T‐cell receptor), ZAP70 (zeta chain of T‐cell receptor associated protein kinase 70), MEK (mitogen‐activated extracellular signal‐regulated kinase), ERK (extracellular regulated protein kinases), SKP2 (S‐phase kinase‐associated protein2), CDKs (cyclin‐dependent kinases), PD‐1 (programmed cell death 1 receptor), ITIM (immunoreceptor tyrosine‐based inhibition motif), ITSM (immunoreceptor tyrosine‐based switch motif), P (phosphorylation), SHP‐2 (Src homology‐2 domain‐containing protein tyrosine phosphatase‐2), CPT1A (carnitine palmitoyl transferase 1A), FAO (fatty acid oxidation), BATF (basic leucine zipper ATF‐like transcription factor), PI3K (phosphoinositide 3‐kinase), AKT (protein kinase B).

There is no consensus at present on how the PD‐1 pathway regulates T cells due to the intricate intracellular interactions involved. Further research is warranted to elucidate the complex regulatory mechanism comprehensively.

2. FUNDAMENTAL MECHANISMS

Antitumor CD8⁺ T cells that fail to eradicate tumor cells often undergo dysfunction or exhaustion, characterized by a diminished effector function. PD‐1 antibodies modulate the metabolic landscape within tumors by inhibiting glycolysis, thereby increasing the availability of glucose to reinvigorate the glycolytic capacity of exhausted T cells in the blood and TME. ¹⁹ , ²⁰ , ²¹ Additionally, anti‐PD‐1 therapy can repress the exhausted phenotype of γδT cells, a subset of innate‐like lymphocytes (ILLs), which exhibit more rapid migration to tumor sites compared with CD8⁺ T cells due to their lack of restriction to classical MHC molecules. ²²

Furthermore, CD8⁺ T cells have the ability to differentiate into Tc17 cells. Tc17 cells play a crucial role in autoimmune diseases, but whether they participate in antitumor immune responses directly remains unclear. ²³ In an in vivo melanoma study, it was found that IL‐17‐producing cells promoted the terminal exhaustion of CD8⁺ T cells and tumor progression. ²⁴ However, other melanoma studies have suggested that, due to the microenvironment‐dependent lineage plasticity and longer lifespan of Tc17 cells, they remain of interest in the antitumor process. PD‐1 signaling strongly inhibits the differentiation of Tc17 cells and suppresses their stemness and plasticity toward Tc1‐like cells. Therefore, PD‐1 antibodies may act as a dual attack on antitumor immunity. ²⁵

While T cells are the primary targets of anti‐PD‐1 therapy, the expression of PD‐1 extends beyond T cells to include innate lymphoid cells (ILCs), which play a role in early immune responses in mucosal tissues. In the context of melanoma metastasis, upregulated PD‐1 expression on ILCs has been associated with impaired cytokine secretion. PD‐1 blockade has been shown to restore cytokine secretion, including granulocyte–macrophage colony‐stimulating factor (GM‐CSF) that recruits neutrophils and macrophages, as well as TNF‐α that targets tumor vasculature and augments tumor oxidative stress. The combination of IL‐33 and anti‐PD‐1 therapy has emerged as a strategy to potentiate the antitumor activity of ILC2s by promoting their proliferation and subsequently inhibiting PD‐1 expression. ²⁶ , ²⁷ , ²⁸

The formation of tertiary lymphoid structures (TLS) within non‐lymphoid tissues has been associated with a favorable prognosis in melanoma. PD‐1 antibodies increase the production of CXCL13, the ligand for CXCR5, recruiting CXCR5⁺ B cells and CXCR5⁺ follicular helper T cells (Tfh) to form TLSs. B cells serve as professional antigen‐presenting cells and have the ability to differentiate into plasma cells that secrete antibodies targeting an array of tumor antigens. Meanwhile B cells may be more helpful to antitumor immunity via antigen presentation to T cells. PD‐1 antibodies are able to induce the production of ICI‐Responsive B cells, a cluster of memory B cells, which have the potential to sustain long‐term antitumor responses. ²⁹ , ³⁰ Actually, the role of B cells in melanoma immunotherapy remains contentious due to their heterogeneous composition within the TME. Utilizing B cell‐specific PTEN‐deficient mice to study regulatory B cells (Bregs) found that B1a Bregs produce IL‐10 to inhibit the secretion of T helper 1 cytokines by tumor‐infiltrating CD8⁺ T cells. ³¹ , ³²

3. TUMOR‐INTRINSIC RESISTANCE MECHANISMS

Patients who exhibit primary resistance have never been able to derive therapeutic benefits from PD‐1 blockade. While some patients initially respond positively, tumor cells undergo adaptive changes under immune pressure, leading to acquired resistance. Effective antitumor immune responses hinge upon a series of events: initially, tumor cells must express tumor‐specific antigenic peptides, which are subsequently presented on the cell surface by MHC. Finally, the peptide–MHC complex must be recognized by antigen‐specific CD8⁺ T cells. PD‐1 antibodies primarily target T cells, so a myriad of tumor‐intrinsic factors may impede T‐cell infiltration or induce dysfunction, posing challenges to the efficacy of PD‐1 blockade therapy. The following will mainly focus on tumor‐intrinsic resistance mechanisms discovered in melanoma research.

3.1. Loss of Melanoma‐Associated Antigens

Genetic changes within cells instigate the emergence of neoantigens, rendering tumor cells discernible as foreigners by the immune system. In the context of melanoma patients, a greater abundance of neoantigens frequently correlates with heightened reactivity and improved prognostic outcomes. ³³ Factors such as elevated mutational burden, microsatellite instability (MSI‐H), and deficiency in mismatch repair (dMMR) contribute to the generation of neoantigens. Moreover, for neoantigens to elicit robust antitumor responses, they must evoke high TCR polyclonality and be recognized recurrently in the blood and TME. This phenomenon may explain why melanoma cases characterized by a lower abundance of neoantigens still exhibit favorable responses to immune checkpoint therapy. ³⁴ , ³⁵

Consequently, a low tumor mutational burden (TMB), sustained microsatellite stability, and effective DNA repair mechanisms may serve as predictive indicators of primary resistance to immunotherapy. Furthermore, persistent tumor‐intrinsic interferon‐gamma (IFN‐γ) signaling induces a state of irreversible dedifferentiation in melanoma cells, thereby downregulating the expression of tumor‐associated antigens. In non–small‐cell lung cancer patients with acquired resistance, the evolutionary loss of mutations encoding neoantigens has been documented. ³⁶

3.2. Loss of MHC Molecules

IFN‐γ, a type II interferon, binds to its receptor through Janus kinases (JAK1/2) and signal transducers and activators of transcription (STATs), initiating the expression of MHC molecules. Consequently, any disruption in IFN‐γ signaling compromises antigen presentation. IFN‐γ signaling also orchestrates the induction of components crucial for antigen‐presenting machinery (APM), such as Transporter associated with Antigen Processing (TAP1/TAP2) and Beta‐2‐microglobulin (β2M). ³⁷ The low level of MHC‐I APM transcription correlated with primary resistance to immune checkpoint blockade (ICB) in melanoma. ³⁸ Mutations in the β2M gene impede proper MHC‐I folding and transportation. ³⁹ Furthermore, elevated intratumoral levels of secretogranin 2 (SCG2) hinder APM transcription, diminishing the sensitivity of melanoma cells to T‐cell cytotoxicity, even upon IFN‐γ treatment. ⁴⁰

Melanocyte inducing transcription factor (MITF) serves as a pivotal regulator of melanoma differentiation, while anexelekto (AXL) induces mesenchymal transition. Transforming growth factor‐beta (TGF‐β) signaling can promote variable dedifferentiation markers like MITF^low and AXL^high. MHC‐I downregulation associated with the dedifferentiation state is a frequent occurrence in both innate and acquired resistance to anti‐PD‐1 therapy. ⁴¹

F‐box protein 11 (FBXO11), an E3 ligase, plays a crucial role in the ubiquitination and proteasome‐dependent degradation of the MHC class II transactivator (CIITA), a transcriptional regulator of MHC‐II expression. In melanoma cell lines, deletion of FBXO11 contributes to the accumulation of CIITA and increased expression of MHC‐II. ⁴²

However, the presence of MHC molecules does not have to confer beneficial effects for patients. Qa‐1b/HLA‐E, a non‐classical MHC‐I molecule, interacts with its receptor KG2A/CD94, which is expressed on the surface of the effector and exhausted CD8⁺ T cells following ICB therapy. HLA‐E/NKG2A axis attenuates CD8⁺ T‐cell cytotoxicity directly. ⁴³ In addition, the expression of MHC‐I shields mouse melanoma cells from NK cell‐mediated killing and MHC‐II facilitates melanoma cell evasion from apoptosis through the combination with Lymphocyte Activation Gene‐3 (LAG‐3). ⁴⁴ , ⁴⁵

3.3. Mutations in Signaling Pathways

Mutations within signaling pathways significantly influence the pathogenesis and progression of melanoma, with Braf^V600E mutations represented the most prevalent driver mutation. V‐raf murine sarcoma viral oncogene homolog B1 (Braf) mutations result in the sustained activation of the mitogen‐activated protein kinase (MAPK) pathway, orchestrating downstream effects for facilitating the secretion of vascular endothelial growth factor (VEGF) and interleukin‐8 (IL‐8), which dampen T‐cell recruitment and their functions. ⁴⁶ Furthermore, mutated BRAF prompts tumor cells to secrete interleukin‐1α and β, inducing the expression of PD‐1 ligands and cyclo‐oxygenase‐2 (COX‐2). PD‐L1, PD‐L2, and COX‐2 are expressed on the tumor‐associated fibroblasts (TAFs) that erect physical barriers between TILs and tumor cells. ⁴⁷ Additionally, BRAF mutations repress the expression of MHC molecules, and BRAF inhibitors can counteract this effect. ⁴⁸ Aberrations in the MAPK pathway also drive the expression of CD73, an enzyme facilitating the generation of immunosuppressive adenosine. Studies have revealed that melanoma cells adopt a mesenchymal‐like phenotype following ICB treatment, concomitant with an upregulation of CD73 expression, which represents a mechanism for adaptive resistance. ⁴⁹

IFN‐γ regulates melanoma gene expression through the Janus kinase–Signal transducer and activator of transcription (JAK–STAT) pathway. Tumor cells harboring mutations in JAK1 demonstrated insensitivity to all types of interferons, whereas those with JAK2 mutations exhibited specific resistance to IFN‐γ. Consequently, either JAK1 or JAK2 mutations confer acquired resistance by disrupting IFN‐γ signaling that mediates cytostasis and cytotoxicity. ³⁹ , ⁵⁰ Moreover, mutations in JAKs reduce the expression of chemokines such as CXCL9, CXCL10, and CXCL11, which are responsible for recruiting T cells. ³⁷ SOX2, a transcription factor, can mediate the durable activation of the JAK–STAT pathway by inhibiting the transcription of suppressor of cytokine signaling 3 (SOCS3) and protein tyrosine phosphatase non‐receptor type 1 (PTPN1), thereby contributing to the overexpression of interferon‐stimulated gene resistance signature (ISG.RS) in melanoma. ⁵¹

In preclinical melanoma models, activation of the Wnt/β‐catenin pathway has been observed to downregulate the expression of chemokine CCL4 that attracts CD103⁺ dendritic cells. The presence of CD103⁺ dendritic cells (DCs) is necessary for the pre‐existence of sufficient T cells in the TME, thus the activation of Wnt/β‐catenin pathway leads to primary resistance. ⁵² Additionally, Wnt/β‐catenin signaling mediates the upregulation of carnitine palmitoyl transferase 1A (CPT1A), enhancing indoleamine‐2,3‐dioxygenase (IDO) activity, promoting regulatory T‐cell (Treg) proliferation, and inhibiting effector T‐cell functions. ⁵³

Another critical pathway implicated in drug resistance is the phosphatidylinositol‐3‐kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathway, caused by phosphatase and tensin homolog (PTEN) mutations. Preclinical and clinical melanoma studies have indicated that deletion or loss‐of‐function mutations of PTEN was associated with impaired CD8⁺ T‐cell proliferation and infiltration, increased VEGF secretion, and elevated levels of anti‐inflammatory cytokines. ⁵⁴

3.4. Expression of PD‐L1

The expression of PD‐L1 on tumor cells is either constitutive or inducible. Constitutive expression is primarily triggered by intrinsic oncogenic signals, while inducible expression occurs in response to IFN‐γ secreted by immune cells within the TME. The absence of constitutive PD‐L1 expression confers primary resistance to anti‐PD‐1 therapies. ³⁷ Developmentally regulated GTP‐binding protein 2 (DRG2) is responsible for the endosomal trafficking of PD‐L1 on the tumor surface. Loss of DRG2 results in a poor response to anti‐PD‐1 therapy for the absence of PD‐L1 expression. ⁵⁵

Melanoma cells evade immune detection and resist immune‐based therapies through upregulating PD‐L1. MAPK signaling cooperates with JUN and STAT3 to regulate constitutive PD‐L1 expression and the PI3K‐AKT pathway regulates PD‐L1 expression in an IFN‐γ‐independent manner, but the effect of BRAF, MAPK/ERK kinase (MEK), and PI3K inhibitors varies substantially on PD‐L1 expression across melanoma cells. The above signaling works at the transcriptional level. CKLF Like MARVEL Transmembrane Domain‐Containing 6 (CMTM6) and CMTM4 positively regulate both constitutive and IFN‐γ‐induced PD‐L1 expression at the posttranslational level. Furthermore, some studies have found that inhibiting NF‐κB and MYC decreased PD‐L1 expression in melanoma cells. Also, the HIF‐1α increase in PD‐L1 expression mediated the resistance to CTL‐mediated killing in a mouse melanoma. ⁵⁶

Unc‐51 like kinase 1 (ULK1) is correlated with poor survival and upregulation of post anti‐PD‐1 therapy in melanoma patients. It phosphorylates interferon regulatory factor 1 (IRF1), enhancing its binding affinity to the promoter regions of PD‐L1. ⁵⁷

TRAF6, a non‐conventional E3 ubiquitin ligase, promotes PD‐L1 transcription through the formation of a YAP1/TFCP2 transcriptional complex. Suppression of TRAF6 by bortezomib can decrease PD‐L1 expression. ⁵⁸

Upon blocking the PD‐1/PD‐L1 interaction, T cells are stimulated to produce IFN‐γ. IFN‐γ signaling‐mutant tumor cells that lack of PD‐L1 expression tend to dominate under immune selective pressure. However, wild‐type tumor cells with exogenous PD‐L1 expression can still compensate for therapeutic resistance. ⁵⁹

3.5. Additional Disadvantageous Factors in Anti‐PD‐1 Treatment

The effectiveness of anti‐PD‐1 treatment in melanoma patients is influenced by various factors. One such factor is the tumor‐derived CD155, which serves as a ligand for CD226. This interaction induces the ubiquitination, internalization, and proteasomal degradation of CD226, a molecule associated with improved progression‐free survival in melanoma patients undergoing ICB therapy. The loss of CD226 is a complementary mechanism contributing to T‐cell dysfunction, in addition to the overexpression of immune checkpoints. ⁶⁰

The presence of CD58, inherently expressed by melanoma cells, is crucial for T‐cell infiltration into the TME and facilitates intratumoral T‐cell differentiation at the early stages of tumorigenesis. The absence of an intact CD58–CD2 axis leads to ICB resistance. Additionally, CMTM6 coordinates the balance of signaling stability between CD58 and PD‐L1. If CD58 is lost, PD‐L1 will escape from the lysosomal degradation via the stable interaction with CMTM6, leading to a higher level of PD‐L1. ⁶¹

The melanocortin‐1 receptor (MC1R), a G protein‐coupled receptor (GPCR) expressed on melanoma, inhibits the transcription of chemokines such as CXCL9, CXCL10, and CXCL11, resulting in diminished T‐cell infiltration in the TME via the activation of the GPCR–GNAS–PKA axis. ⁶²

The type I IFN/IFNAR (interferon‐alpha/beta receptor)/NOS2 (nitric oxide synthase 2) signaling pathway may run through the whole course of anti‐PD‐1 treatment. Melanoma patients in an advanced stage of resistance to anti‐PD‐1 show a notable increase in NOS2. ⁶³ NOS2 converts l‐arginine to l‐citrulline to product nitric oxide (NO) that plays a role in melanoma development. ⁶⁴ It can be seen that NOS2 contributes to the acquired resistance to anti‐PD‐1 blockade and anti‐PD‐1 therapy increases interferons (IFNs) to regulate NOS2 production.

The cyclo‐oxygenase‐2 (COX2)/microsomal prostaglandin E synthase‐1 (mPGES1)/prostaglandin E2 (PGE2) pathway is frequently overactivated in melanoma. mPGES1 inhibits the type I IFN pathway to hinder the infiltration of CD8a⁺ DCs that can secret effector CD8⁺ T‐cell‐attracting chemokines. ⁶⁵ PGE2 attracts various immunosuppressive cells and enhances the capacity of invasion and metastasis. ⁶⁶ Compared with receiving PD‐1 blockade alone, the combination with COX inhibitors (COXIBs; and nonsteroidal anti‐inflammatory drugs [NSAIDs]) or selective mPGES1 inhibitors can improve the therapeutic effect significantly. Furthermore, mPGES1 inhibitors were safer and more effective than COXi in clinical trials. ⁶⁷

The level of kelch‐like ECH‐associated protein 1 (KEAP1) in circulating tumor cells (CTCs) is elevated in complete response (CR) patients but nearly undetectable in patients with progressive disease (PD). By contrast, genes regulated by nuclear factor erythroid 2‐related factor 2 (NRF2) are overexpressed in PD patients' CTCs, which is associated with KEAP1 loss. This indicates that KEAP1‐NRF2 pathway is involved in melanoma resistance mechanisms to ICB. ⁶⁸

Midkine (MDK), secreted by melanomas, stimulates the generation of chemoattractants and cytokines in an autocrine manner, which polarizes macrophages to a tolerant phenotype and leads ultimately to CD8⁺ T‐cell dysfunction. MDK not only activates protumoral NF‐κB signaling but suppresses antitumoral genes associated with IFN‐γ. So MDK mediates ICB resistance from both the secretome and transcriptome. ⁶⁹

In anti‐PD‐1 resistant murine melanoma cell lines and databases of metastatic melanoma, high expression of cytidine deaminase (CDA) has been found to be associated with anti‐PD‐1 resistance. In pancreatic cancer, CDA has been shown to promote the synthesis and release of uracil nucleotides. uridine diphosphate (UDP) activates P2Y6 (a high‐affinity receptor for UDP) on the surface of tumor‐associated macrophages (TAMs), leading to an immunosuppressive state. A previous study has reported that the UDP–P2Y6 axis promotes melanoma lung metastasis by recruiting neutrophils. However, the role of UDP in anti‐PD‐1 resistance in melanoma requires further investigation. ⁷⁰

There are additional problems caused by DRG2 depletion in addition to PD‐L1 expression. After anti‐PD‐1 treatment, DRG2‐depleted murine tumor cells will replace the previously dominant anti‐PD‐1 responsive population, leading to a TME characterized by low levels of effector‐like T cells and IFN‐γ‐expressing CD8⁺ T cells, and high levels of regulatory‐like T cells. ⁵⁵

The metabolism of the melanoma leads to oxygen and nutrient depletion in the TME, which reprograms the metabolism of T cells and elevates the glycolysis of tumor cells. Deprivation of glucose is unfavorable for effector T cells and M1 macrophages to play a role in tumor killing. ⁷¹ This metabolic stress in melanoma can induce the increasing expression of fat mass and obesity‐associated protein (FTO), which attenuates tumor‐cell sensitivity to IFN‐γ and promotes the intrinsic expression of immune checkpoint molecules. ⁷²

4. CONCLUSION AND PERSPECTIVES

PD‐1 antibodies target the compromised immune system of melanoma patients. PD‐1 antibodies restore the inherent antitumor capability of T cells, repair the impaired cytokine secretion capacity of ILCs, promote B cell infiltration into tumors, and facilitate the formation of TLS, all of which are strongly correlated with the responsiveness to immunotherapy and patient prognosis. The mechanism of PD‐1 antibodies has been extensively documented. In contrast, the mechanism of resistance is much more complex.

Adaptive changes inside tumor cells causing immune escape and drug resistance are summarized in Table 1. There are other factors within the TME, including the lack of neoantigen‐specific T cells, numerous immune‐suppressive cells, cytokines and the expression of other immune checkpoints, which all contribute to the inhibition of antitumor immune responses. The gut microbiome has emerged as a popular issue in recent studies, which have shown that the absence of specific gut bacteria is associated with unfavorable responses in melanoma patients with anti‐PD‐1 treatment. Clinical trials combining fecal microbial transplantation (FMT) from healthy donors with PD‐1 antibodies are currently underway. ⁷³ , ⁷⁴

TABLE 1.

Mechanisms of tumor‐cell intrinsic resistance.

Tumor‐cell intrinsic resistance mechanisms	Changes in the cell
Loss of melanoma‐associated antigens	Low TMB
	Stable microsatellite
	MMR
	Intrinsic IFN‐γ signaling
	Loss of antigen encoding mutations
Loss of MHC molecules	Loss of components of AMP
	High levels of SCG2
	TGF‐β signaling activation
	High levels of FBXO11
	Mutations in IFN‐γ signaling
T‐cell exclusion	Activation of oncogenic signaling pathways
	PTEN loss
	PD‐L1 expression
	MC1R expression
	Loss of CD58
	Loss of CD226
Insensibility to T cells	Mutations in IFN‐γ signaling
	High levels of FTO
Contribute to immunosuppressive environment	Increasing NOS2/NO
	Overactivating mPGES1/PGE2 pathway
	Secreting MDK
	DRG2 depletion

Efforts have been intensified to innovate combination strategies targeting cell intrinsic and extrinsic factors to overcome resistance and enhance therapeutic efficacy. The combination of PD‐1 antibodies with other immune checkpoint inhibitors has helped some patients overcome resistance. Additionally, combining targeted therapies such as BRAF inhibitors and MEK inhibitors also has shown promising clinical efficacy (Table 2). Although such combination approaches have yielded response rates of up to 58% in clinical, patients often endure severe immune‐related adverse events (irAEs) and are susceptible to relapse. ⁷⁵ Various factors are a context‐dependent double‐edged sword in immunotherapy due to the diverse phenotypes and durations of therapeutic effects. Combination strategies certainly deserve further development, but they only address symptoms rather than the root cause. Therefore, it is imperative to further research the underlying mechanisms of primary and adaptive resistance for more effective interventions.

TABLE 2.

Phase III clinical trials related to combination strategies of PD‐1 antibodies in melanoma.

Combination strategies	Trail number	Patient condition	Treatment arms	Primary outcome
Anti‐PD1 plus anti‐CTLA4	NCT01844505	Unresectable stage III or IV melanoma, no prior systemic anticancer therapy	Nivolumab + Ipilimumab vs. Nivolumab vs. Ipilimumab	Median PFS (m): 11.50 vs. 6.87 vs. 2.89 24‐month OSR: 64% vs. 59% vs. 45% 24‐month PFSR: 0.43 vs. 0.37 vs. 0.12
Anti‐PD1 plus anti‐LAG 3	NCT03470922	Unresectable stage III or IV melanoma, no prior systemic anticancer therapy	Nivolumab + Relatlimab vs. Nivolumab	Median PFS (m): 10.12 vs. 4.63
Anti‐PD1 plus BRAF + MEK inhibitors	NCT02967692	Unresectable stage III or IV cutaneous melanoma with BRAF V600 mutations, no prior systemic anticancer therapy	Spartalizumab + Dabrafenib + Trametinib vs. Placebo + Dabrafenib + Trametinib	Median PFS (m): 16.2 vs. 12.0

In summary, ICB therapy offers more advantages compared with conventional treatments. Deeper investigations of the interactions between the immune system and tumor cells are needed to maximize the benefits of PD‐1 blockade, facilitating the development of effective and tailored therapeutic strategies for melanoma patients.

AUTHOR CONTRIBUTIONS

Tong Wang: Writing – original draft; writing – review and editing. Wenjie Ma: Supervision. Zijian Zou: Methodology; supervision. Jingqin Zhong: Supervision. Xinyi Lin: Supervision. Wanlin Liu: Supervision. Wei Sun: Funding acquisition; resources; supervision. Tu Hu: Funding acquisition; resources; supervision. Yu Xu: Resources; supervision. Yong Chen: Funding acquisition; project administration; resources; supervision; writing – review and editing.

FUNDING INFORMATION

This study was supported by the National Key R&D Program of China (Grant No. 2023YFC2506400), National Natural Science Foundation of China (Grant No. 82373385), the National Natural Science Foundation of China (82373385), and the National Natural Science Foundation of China (82303608).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

ETHICS STATEMENT

Approval of the research protocol by an Institutional Review Board: N/A.

Informed Consent: N/A.

Registry and the Registration No. of the study/trial: N/A.

Animal Studies: N/A.

Wang T, Ma W, Zou Z, et al. PD‐1 blockade treatment in melanoma: Mechanism of response and tumor‐intrinsic resistance. Cancer Sci. 2025;116:329‐337. doi: 10.1111/cas.16398