The immune system has a remarkable ability to distinguish self from non-self, destroying foreign invaders while sparing healthy tissue. Cancer cells, however, evolve mechanisms to evade immune detection — essentially putting on a molecular disguise. Checkpoint inhibitors work by removing this disguise, allowing the immune system to recognize and attack cancer cells.
The Immune Checkpoint System
Under normal conditions, immune checkpoints act as natural brakes on the immune system, preventing excessive immune responses that could damage healthy tissue. T cells express checkpoint receptors (such as PD-1 and CTLA-4) that, when engaged by their ligands, send inhibitory signals that dampen immune activity. This system is essential for preventing autoimmunity.
Cancer cells exploit this system by expressing checkpoint ligands on their surface. When a tumor cell displays PD-L1, for example, it engages PD-1 on approaching T cells, effectively telling them to stand down. The T cell receives an inhibitory signal and fails to attack, even though it has recognized the cancer cell as abnormal. This mechanism of immune evasion is now understood to be one of the hallmarks of cancer.
How Checkpoint Inhibitors Restore Anti-Tumor Immunity
Checkpoint inhibitors are monoclonal antibodies that block the interaction between checkpoint receptors and their ligands. By occupying these molecular docking sites, they prevent the inhibitory signal from being transmitted, allowing T cells to remain activated and attack cancer cells. There are three main classes currently in clinical use.
Anti-PD-1 antibodies (pembrolizumab/Keytruda, nivolumab/Opdivo) bind to PD-1 on T cells, blocking PD-L1 from engaging it. This is the most widely used class, with FDA approvals across more than 20 tumor types. Anti-PD-L1 antibodies (atezolizumab/Tecentriq, durvalumab/Imfinzi, avelumab/Bavencio) work from the other side, binding PD-L1 on tumor cells to prevent it from engaging PD-1. Anti-CTLA-4 antibodies (ipilimumab/Yervoy, tremelimumab) block CTLA-4, a checkpoint receptor involved in earlier stages of T-cell activation, providing a complementary mechanism of immune activation.
Combining anti-PD-1 with anti-CTLA-4 (the ipilimumab-nivolumab combination) provides dual checkpoint blockade that can achieve deeper and more durable responses than either agent alone, though at the cost of increased immune-related side effects.
Which Cancers Respond Best?
Not all cancers respond equally to checkpoint inhibitors. Response generally correlates with factors that increase immune visibility: high tumor mutational burden (generating more neoantigens for T cells to recognize), PD-L1 expression (indicating the tumor is actively suppressing immunity), microsatellite instability (creating a high neoantigen load), and viral-driven cancers (where viral antigens provide additional immune targets).
Cancers with the highest checkpoint inhibitor response rates include melanoma (high mutational burden from UV damage), Hodgkin lymphoma (near-universal PD-L1 amplification), Merkel cell carcinoma (viral-driven), non-small cell lung cancer with high PD-L1 expression, and MSI-high colorectal cancer. In contrast, cancers with low mutational burden and immunosuppressive microenvironments — such as pancreatic cancer and glioblastoma — have shown limited response to checkpoint inhibitors as single agents.
Biomarkers Guiding Immunotherapy Selection
Several biomarkers help clinicians predict which patients are most likely to benefit from checkpoint inhibitors. PD-L1 expression, measured by immunohistochemistry (IHC), is the most widely used but imperfect predictor — some PD-L1-negative patients respond, while some PD-L1-positive patients do not. The Combined Positive Score (CPS) and Tumor Proportion Score (TPS) are the two main PD-L1 scoring methods, each used for different tumor types.
Microsatellite instability (MSI-high/dMMR) is the strongest predictor of checkpoint inhibitor response and has led to the first tumor-agnostic FDA approval: pembrolizumab for any MSI-H solid tumor regardless of where it originated. Tumor mutational burden (TMB-high, defined as 10 or more mutations per megabase) is another tumor-agnostic biomarker. Research continues to develop better predictive biomarkers, including immune gene signatures and the composition of the gut microbiome.
Looking Forward
The checkpoint inhibitor field continues to evolve rapidly. Newer checkpoint targets beyond PD-1 and CTLA-4 include LAG-3 (targeted by relatlimab, now approved in combination with nivolumab for melanoma) and TIGIT (under investigation in multiple tumor types). Combinations of checkpoint inhibitors with other treatment modalities — targeted therapy, antibody-drug conjugates, radiation, and cellular therapies — represent the next frontier.
PipelineEvidence tracks checkpoint inhibitor approvals across all 50 tumor types covered on this platform. Explore our complete therapy database to see the current landscape of approved immunotherapy agents.