Discovery and Immunobiological Foundations of the PD-1 Pathway
The programmed cell death protein 1 (PD-1) pathway emerged as a therapeutic target through foundational discoveries in tumor immunology that culminated in the 2018 Nobel Prize in Physiology or Medicine awarded to James Allison and Tasuku Honjo for their work on immune checkpoint blockade. The PD-1 receptor, expressed on activated T cells, binds to its ligands PD-L1 and PD-L2 to deliver inhibitory signals that terminate immune responses and maintain peripheral tolerance. Tumors exploit this pathway to evade immune destruction by upregulating PD-L1 expression through intrinsic oncogenic signaling (PI3K/PTEN, WNT/β-catenin, Ras, hedgehog pathways) and adaptive immune resistance mechanisms driven by interferon-gamma exposure in the tumor microenvironment1.
Early diagnostic development focused on PD-L1 immunohistochemistry as a predictive biomarker, though the field has struggled with assay standardization challenges. Multiple commercially available assays (22C3, 28-8, SP142, SP263) exhibit generally equivalent performance except for SP142, which demonstrates lower sensitivity. Critical limitations include lack of standardized controls, varied scoring systems distinguishing tumor cell expression (tumor proportion score, TPS) from immune cell expression (combined positive score, CPS) with different thresholds applied across tumor types, and reliance on absolute percentage cutoffs that fail to capture continuous biomarker-efficacy relationships1. Approximately 45% of PD-L1-positive patients respond to PD-1 monotherapy versus 15% of PD-L1-negative patients, yet this enrichment is insufficient for patient selection in most malignancies2.
Representative PD-1 Therapeutics and Clinical Benchmarks Across Major Tumor Types
The current therapeutic landscape encompasses at least 15 approved PD-1 monoclonal antibodies globally, with 12 approved in China versus four in the United States, reflecting aggressive domestic innovation in Asian markets21. Leading agents include pembrolizumab (Merck, Keytruda), nivolumab (Bristol-Myers Squibb, Opdivo), cemiplimab (Regeneron/Sanofi, Libtayo), dostarlimab (GSK/AnaptysBio, Jemperli), toripalimab (TopAlliance/Coherus, Loqtorzi) and tislelizumab (BeiGene, Tevimbra)21.
In advanced melanoma, the CheckMate 067 ten-year follow-up established nivolumab plus ipilimumab as the gold standard with median overall survival of 71.9 months and 52% ten-year survival in BRAF-mutant patients, while nivolumab monotherapy achieved median overall survival of 36.9 months with median duration of response exceeding eight years. Remarkably, 64% of nivolumab plus ipilimumab patients remained treatment-free at ten years, demonstrating potential for functional cure in a subset22.
In first-line metastatic non-small cell lung cancer (NSCLC) with PD-L1 tumor proportion score of 50% or higher, monotherapy PD-1 antibodies achieve median overall survival ranging from 15 to 26 months, objective response rates of 27% to 39%, and five-year overall survival of 11% to 24%. The EMPOWER-Lung 1 trial of cemiplimab demonstrated median overall survival of 26.1 months versus 13.3 months for platinum doublet chemotherapy (hazard ratio 0.57), with five-year overall survival of 24% versus 11% 22.
In intermediate or poor-risk renal cell carcinoma, the CheckMate 214 nine-year follow-up of nivolumab plus ipilimumab demonstrated five-year overall survival of 43% and nine-year overall survival probability of 31% with sustained progression-free survival benefit (96-month progression-free survival probability 23% versus 9% for sunitinib). Counterintuitively, the trial demonstrated greatest benefit in the PD-L1 less than 1% population (five-year overall survival 22% versus 8%, hazard ratio 0.63), challenging the paradigm that PD-L1 expression universally predicts PD-1 monotherapy benefit22.
Recent Regulatory and Label Expansion Milestones in PD-1/Checkpoint Therapy
Recent regulatory developments between March 2025 and March 2026 include FDA approval of tislelizumab plus platinum-based chemotherapy for PD-L1-positive esophageal squamous cell carcinoma, achieving median overall survival of 16.8 months versus 9.6 months for chemotherapy alone (34% reduction in risk of death)23. Summit Therapeutics' ivonescimab, a PD-1/VEGF bispecific antibody, received FDA Biologics License Application acceptance for EGFR-mutated NSCLC after tyrosine kinase inhibitor progression, with regulatory decision expected by November 202625. The European Medicines Agency authorized Opdualag (nivolumab plus relatlimab fixed-dose combination targeting PD-1 and LAG-3) for first-line advanced melanoma in patients with tumor cell PD-L1 expression less than 1%, demonstrating 6.7 months progression-free survival versus 3 months for nivolumab alone24.
Mechanisms of Resistance and Clinical Limitations of Current PD-1 Therapeutics
The Society for Immunotherapy of Cancer (SITC) Biomarkers Committee established a four-category framework organizing resistance mechanisms across the tumor-immunity cycle. Category one encompasses T-cell activation failures including lack of antigenicity from low neoantigen load, defects in antigen processing or MHC expression, and JAK1/2 truncating mutations that abolish interferon-gamma signaling. Category two addresses tumor microenvironment access barriers including cold or immune-desert tumors, immune-excluded phenotypes driven by tumor-associated macrophages and cancer-associated fibroblasts, aberrant vessel permeability, hypoxia-induced checkpoint upregulation, and WNT/β-catenin pathway alterations. Category three involves immune suppression through regulatory T cells, myeloid-derived suppressor cells, and suppressive soluble factors including indoleamine-pyrrole 2,3-dioxygenase, adenosine, TGF-β, VEGF, and IL-8-mediated neutrophil recruitment. Category four captures tumor intrinsic resistance mechanisms including TRAF2 alterations, anti-apoptotic proteins, extracellular matrix stiffness, autophagy, and exosomal PD-L1 decoys1.
Specific genetic contexts predict resistance in lung adenocarcinoma. A study performing in-depth immune profiling with next-generation sequencing identified that distinct combinations of STK11, EGFR, and TP53 mutations were major determinants of tumor immune profile and PD-L1 expression. TP53-mutant/STK11-EGFR-wildtype tumors demonstrated highest CD8 T-cell density, highest PD-L1 expression, and upregulated pathways related to T-cell chemotaxis, immune cell cytotoxicity, and antigen processing, with patients achieving prolonged progression-free survival on anti-PD-1 therapy (hazard ratio 0.32). Clinical benefit was further enhanced with strong PD-L1 co-expression, whereas STK11 mutations encoding LKB1 and EGFR mutations were associated with reduced CD8 infiltration and immune activation13.
In metastatic colorectal cancer, immune profiling of MSI-H and POLE-mutated tumors treated with PD-1 blockade revealed that responders (complete or partial response) had significantly higher densities of CD8-positive PD-1-positive tumor-infiltrating lymphocytes than nonresponders (p equals 0.0007), and the percentage of CD4-positive Tbet-positive T-cells with Th1 phenotype was significantly higher in responders (p equals 0.0007). Notably, PD-L1 expression, CD4-positive T-cell density, and CD4-positive FOXP3-positive T-cell density did not significantly differ between responders and nonresponders, indicating that immune cell phenotype and activation state are more predictive than checkpoint molecule expression alone15.
Lesion-to-lesion heterogeneity within individual patients complicates resistance classification. In metastatic melanoma trials, only 23.5% of individual target lesions progressed despite RECIST version 1.1 categorization of progressive disease, suggesting that biomarker sampling from progressing lesions may be systematically biased. Standard formalin fixation and paraffin embedding introduces variability through degradation and epitope masking, while core needle biopsies preserve tumor microenvironment architecture but provide very small tissue fractions leading to high discordance with surgical samples due to tumor heterogeneity1.
Immune-related adverse events represent a distinct safety challenge, with grade 3 or higher treatment-related adverse events occurring in 21.8% of patients receiving PD-1 monotherapy, 47% to 48% receiving PD-1 plus chemotherapy, and 62.6% receiving PD-1 plus CTLA-4 combination in melanoma, predominantly driven by immune-mediated colitis, hepatitis, and endocrinopathies22. A systematic review and meta-analysis of 34 randomized clinical trials involving 18,709 patients demonstrated that immune checkpoint inhibitors as monotherapy show favorable association with patient-reported quality of life (pooled between-groups difference in mean change from baseline to 12 weeks of positive 4.6 points) and can be combined with other anticancer drugs without worsening quality of life outcomes20.
Novel Combination Strategies and Biomarker-Selected Differentiation Opportunities
The clinical trial evidence from 2023 to 2026 demonstrates that monotherapy PD-1 antibody development offers minimal differentiation opportunity due to market saturation with 15 approved products and at least seven biosimilars in development, predominantly concentrated in China where Henlius's nivolumab biosimilar HLX18 received NMPA Investigational New Drug approval in March 20262129. Genuine clinical value accrual is increasingly concentrated in biomarker-selected combination strategies that address specific resistance mechanisms.
PD-1 plus CTLA-4 combinations demonstrate greatest benefit in PD-L1 low or negative populations that would otherwise be resistant to monotherapy. The CheckMate 9LA trial of nivolumab plus ipilimumab plus two cycles of chemotherapy in stage IV NSCLC showed particular benefit in the PD-L1 less than 1% population with five-year overall survival of 16% versus 5% for chemotherapy (hazard ratio 0.65)22. Similarly, in renal cell carcinoma, nivolumab plus ipilimumab achieved greatest benefit in PD-L1 less than 1% population with five-year overall survival of 22% versus 8% (hazard ratio 0.63)22. This suggests that limited chemotherapy primes immune response in non-inflamed tumors through immunogenic cell death while CTLA-4 blockade enhances T-cell priming.
PD-1 plus VEGF inhibition represents a mechanistically distinct approach. The HARMONi-303 trial demonstrated that ivonescimab, a PD-1/VEGF bispecific antibody, achieved statistically significant superiority over pembrolizumab monotherapy in PD-L1 one percent or higher NSCLC with median progression-free survival of 11.14 months versus 5.82 months (hazard ratio 0.51, p less than 0.00001), representing the first head-to-head Phase III victory over pembrolizumab monotherapy in this setting. Notably, ivonescimab showed consistent benefit regardless of PD-L1 expression level (TPS 1 to 49% hazard ratio 0.54; TPS 50% or higher hazard ratio 0.46), suggesting VEGF inhibition overcomes PD-L1-low resistance by enhancing vascular normalization and T-cell infiltration22. In EGFR-mutant NSCLC after tyrosine kinase inhibitor progression, the HARMONi-A trial demonstrated median progression-free survival of 7.1 months versus 4.8 months for chemotherapy alone (hazard ratio 0.46), with particular benefit in brain metastases (hazard ratio 0.40)22.
Novel checkpoint combinations beyond CTLA-4 have yielded mixed results requiring refined biomarker selection. The nivolumab plus relatlimab (LAG-3 inhibitor) adjuvant melanoma trial CheckMate 67W failed to demonstrate recurrence-free survival benefit versus nivolumab alone (12-month recurrence-free survival 75% versus 72%) despite metastatic melanoma approval, with higher grade 3 or higher adverse events (19% versus 8%)22. Multiple TIGIT plus PD-1/PD-L1 programs failed including tiragolumab plus atezolizumab in SKYSCRAPER-01 (median overall survival 22.9 months versus 16.7 months, hazard ratio 0.81, not significant) and belrestotug plus dostarlimab in GALAXIES Lung-201, which was terminated after failing to meet progression-free survival improvement criteria22. In contrast, domvanalimab (TIGIT inhibitor) plus zimberelimab (PD-1 inhibitor) showed encouraging 12-month overall survival improvement in PD-L1 50% or higher NSCLC (68% versus 57% versus 50% for chemotherapy, hazard ratio 0.64 versus monotherapy), though five-year data are pending22.
Bispecific antibody formats offer potential for improved therapeutic index. The rilvegostomig PD-1/TIGIT bispecific program demonstrated objective response rate of 61.8% in PD-L1 50% or higher NSCLC with only 10.5% grade 3 or higher treatment-related adverse events, suggesting that bispecific format may reduce toxicity compared to dual antibody combinations22. The dataset encompasses more than 40% of Phase I/II pipeline programs in bispecific or multispecific formats (PD-1/VEGF, PD-1/CTLA-4, PD-1/TIGIT, PD-1/LAG-3, PD-1/TIM-3), with China-based companies leading in trispecific and quadruple-target constructs21.
Perioperative settings provide high-value differentiation opportunities through pathologic response endpoints and circulating tumor DNA-adaptive strategies. The CheckMate 816 five-year overall survival data published in 2025 represent the first Phase III demonstration of overall survival benefit in neoadjuvant NSCLC, with neoadjuvant nivolumab plus platinum doublet chemotherapy achieving pathologic complete response of 24% versus 2.2% (odds ratio 13.94), major pathologic response of 36.9% versus 8.9%, median event-free survival of 59.6 months versus 21.1 months (hazard ratio 0.68), and five-year overall survival of 65% versus 55% (hazard ratio 0.72). Presurgical circulating tumor DNA clearance was associated with improved overall survival (56% versus 35% clearance rate), suggesting circulating tumor DNA-adaptive strategies could refine perioperative treatment22. The RATIONALE-315 trial of tislelizumab plus platinum doublet chemotherapy in resectable stage II/IIIA NSCLC achieved higher major pathologic response than CheckMate 816 (56.2% versus 15.0%) and pathologic complete response of 40.7% versus 5.7% in a Chinese population22.
Market Dynamics, Biosimilar Competition, and China's National Reimbursement Drug List Strategy
China's National Healthcare Security Administration released the 2020 National Reimbursement Drug List excluding all multinational PD-1 drugs including pembrolizumab, nivolumab, durvalumab, and atezolizumab despite negotiations, instead prioritizing three Chinese domestic manufacturers with aggressive price concessions. BeiGene's tislelizumab secured inclusion at 74,100 yuan annual treatment cost (80% discount), Junshi's toripalimab at 54,900 yuan (70% plus discount), and Hengrui's camrelizumab at 49,800 yuan (85% discount)26. This pattern of excluding imported PD-1/PD-L1 inhibitors has persisted through successive NRDL updates for the past three years, reflecting China's strategic preference for domestic innovation and manufacturing capacity building27. The global pembrolizumab market was valued at 36.8 billion USD in 2025 and is projected to reach 98.4 billion USD by 2034 at 11.6% compound annual growth rate, though China's NRDL exclusion limits regional revenue28.
Biosimilar development is advancing rapidly to capture value from first-generation PD-1 antibodies as patent exclusivity windows close. At least seven biosimilar programs targeting pembrolizumab and nivolumab are in development predominantly in China, with Amgen's ABP-206 representing the most advanced Western biosimilar effort in Phase III for nivolumab21. Shanghai Henlius announced NMPA approval for Investigational New Drug application to initiate Phase 1 clinical trial of HLX18, a nivolumab biosimilar, in March 2026, following prior approval for Phase 1 initiation in the United States as of December 2025. Global nivolumab sales reached approximately 11.785 billion USD in 2025, representing substantial market opportunity for biosimilar entry29. Sandoz announced a global collaboration agreement with Henlius in April 2025 to commercialize an ipilimumab biosimilar with milestone-based consideration totaling up to 301 million USD, positioning the ipilimumab biosimilar as complementary to Sandoz's proposed nivolumab biosimilar currently in integrated Phase I/III development30.
The broader PD-1 and PD-L1 inhibitor market was valued at 74.16 billion USD in 2026 and is projected to reach 142.04 billion USD by 2031 at 13.88% compound annual growth rate, reflecting continued label expansion, combination therapy adoption, and emerging market penetration despite biosimilar competition and NRDL pricing pressure32. More than 50 companies maintain active PD-1 programs across USA and China, with approximately 60% of programs in Phase I/II indicating ongoing clinical exploration of novel combinations and formats, though only approximately 10% have reached approval or late-stage regulatory filing, suggesting high attrition and competitive selection pressure21.
Clinical Attractiveness Assessment for New Entrants and Future Value Accrual
The evidence base suggests that undifferentiated PD-1 monotherapy offers limited opportunity for new entrants unless paired with meaningful differentiation, such as novel formulations, biomarker-enriched indications, or combination strategies. The class is highly crowded globally, and pricing pressure is especially intense in China, where NRDL inclusion commonly requires substantial negotiated price reductions. Ivonescimab provides a notable example of next-generation differentiation: in the phase III HARMONi-2 trial, the PD-1×VEGF bispecific outperformed pembrolizumab in PD-L1-positive advanced NSCLC, suggesting that value creation is more likely to come from multi-target biology than from incremental optimization of PD-1 binding alone 2126.
Future clinical value is most likely to accrue through biomarker-guided combinations, next-generation bispecifics, and lifecycle expansion into earlier-stage settings. Combinations such as PD-1 plus CTLA-4 in selected PD-L1-low NSCLC populations, PD-1/VEGF co-targeting in defined resistance settings, and perioperative immunotherapy regimens supported by pathologic response and ctDNA analyses exemplify this trend. However, not all mechanistically attractive combinations succeed: TIGIT and LAG-3 programs have shown mixed or negative phase III results, underscoring the need for rigorous biomarker development and confirmatory trials 22.