Cancer immunotherapy did not really exist as a serious clinical discipline before checkpoint inhibitors. By 2026, drugs like Pembrolizumab (Keytruda) and Nivolumab (Opdivo) are first-line therapy for melanoma, lung cancer, kidney cancer, head and neck cancer, Hodgkin lymphoma, and a long and growing list of others. They are among the best-selling drugs in the world and have produced the kind of durable, multi-year responses that chemotherapy almost never delivers.
The science that made this possible is one of the great recent stories of modern medicine. This guide unpacks the immunology, the antibodies, and the patient-side reality of PD-1 and PD-L1 blockade. You can explore the antibodies themselves in the Antibody Design Lab.
Why T cells need brakes
T cells are the killers of the adaptive immune system. They learn to recognize the molecular signature of foreign or abnormal cells and then patrol the body looking for matches. This is great for fighting infections, but it is also dangerous: a T cell that misidentifies a healthy cell could trigger autoimmune disease.
To prevent that, evolution built in multiple layers of inhibitory checkpoints. When a T cell starts to attack a target, the target cell can display a “don't kill me” signal, and receptors on the T cell pick that signal up and shut the T cell down. CTLA-4 is the early-phase checkpoint that controls how T cells get activated in lymph nodes. PD-1 is the later-phase checkpoint that controls how activated T cells behave in tissues.
The PD-1 / PD-L1 axis
PD-1 (programmed cell death protein 1) is a receptor on T cells. Its job is to limit T cell activity in peripheral tissues so that ongoing immune responses don't damage healthy bystander tissue. PD-L1 (programmed death-ligand 1) is the ligand that engages PD-1. It is normally displayed on a small subset of cells as part of immune homeostasis.
Cancer cells exploit this system. Tumors that learn to express PD-L1 are essentially flashing a peace sign at every T cell that approaches: “Don't kill me, I'm friendly.” T cells that engage PD-L1 through PD-1 become functionally exhausted. They don't kill the tumor, they don't proliferate, and they slowly drift out of the tumor.
T cell exhaustion is a real molecular state. Exhausted T cells upregulate inhibitory receptors (PD-1, LAG-3, TIM-3, TIGIT), downregulate effector functions, and have characteristic epigenetic changes. They are still alive, but they have stopped fighting.
Releasing the brake
The conceptual breakthrough was that you could rescue exhausted T cells just by blocking the inhibitory signal. If you cover up PD-1 with an antibody, the tumor's PD-L1 has nothing to dock with. The brake is gone. The T cell wakes up, recognizes the tumor as the threat it is, and attacks.
This is exactly what Pembrolizumab and Nivolumab do. They are humanized (Pembrolizumab) or fully human (Nivolumab) IgG4 monoclonal antibodies that bind PD-1 with high affinity and sterically prevent PD-L1 (and PD-L2) from engaging it. The IgG4 framework is chosen specifically to minimize ADCC — you do not want to kill the T cells you are trying to rescue.
Allison, Honjo, and the Nobel Prize
James Allison at MD Anderson and Tasuku Honjo at Kyoto University worked on parallel pieces of this story for two decades. Allison showed that blocking CTLA-4 with an antibody could shrink tumors in mice, leading to the development of Ipilimumab (Yervoy), the first checkpoint inhibitor approved by the FDA in 2011. Honjo discovered PD-1 in 1992 and worked out its inhibitory function on T cells, which led directly to the development of the PD-1 antibodies.
Their shared 2018 Nobel Prize in Physiology or Medicine recognized that they had unlocked an entirely new modality of cancer therapy. In retrospect, the surprising thing is how slow the field was to accept the idea — for years, cancer immunotherapy was widely dismissed as a dead end.
The major drugs
PD-1 antibodies (block the receptor on T cells)
- Pembrolizumab (Keytruda) — humanized IgG4, approved 2014. Now used in dozens of cancer types and is one of the best-selling drugs in the world.
- Nivolumab (Opdivo) — fully human IgG4, approved 2014. The first checkpoint inhibitor approved for non-small cell lung cancer.
- Cemiplimab (Libtayo) — fully human IgG4, approved 2018 for advanced cutaneous squamous cell carcinoma.
PD-L1 antibodies (block the ligand on tumors)
- Atezolizumab (Tecentriq) — humanized IgG1 with a modified Fc to remove ADCC. First PD-L1 inhibitor approved (2016).
- Durvalumab (Imfinzi) — fully human IgG1 with Fc engineering. Used in lung cancer maintenance after chemoradiation.
- Avelumab (Bavencio) — fully human IgG1 with intact Fc, retains some ADCC activity. Used in Merkel cell carcinoma and other indications.
CTLA-4 antibodies (the original checkpoint blocker)
- Ipilimumab (Yervoy) — fully human IgG1, approved 2011 for metastatic melanoma. Often used in combination with Nivolumab to deepen responses, at the cost of more autoimmune side effects.
Who responds, and why some don't
Checkpoint inhibitors are remarkable but they are not magic. Roughly 20 to 40 percent of patients in eligible tumor types experience durable responses. The rest either fail to respond or progress after an initial response. Several factors predict response:
- Tumor mutational burden (TMB). Tumors with many mutations create many neoantigens — abnormal protein fragments T cells can recognize. High TMB tumors (melanoma, smoking-related lung cancer, mismatch-repair deficient tumors) tend to respond best.
- PD-L1 expression. Tumors that already display high PD-L1 are essentially admitting that they have learned to use the PD-1 pathway. Blocking it is more likely to help.
- Pre-existing T cell infiltration. “Hot” tumors with T cells already inside the tumor microenvironment respond better than “cold” tumors that have no immune activity to begin with. You can't release a brake on a car that has no engine running.
- Microbiome and gut bacteria. Multiple independent studies have linked specific gut microbe profiles to checkpoint inhibitor response. Antibiotic use shortly before treatment correlates with worse outcomes.
Combination therapy and the future
The next chapter of checkpoint immunotherapy is combinations. Combining a PD-1 antibody with a CTLA-4 antibody (Nivolumab + Ipilimumab) deepens responses in melanoma and renal cell carcinoma at the cost of higher toxicity. Combining checkpoint inhibitors with chemotherapy, with VEGF inhibitors, with radiation, and with targeted therapies is being tested across essentially every solid tumor.
Newer checkpoint targets — LAG-3, TIM-3, TIGIT — are reaching clinical trials, though early results suggest they are less potent than PD-1 alone. Bispecific antibodies that hit PD-1 plus a second target (PD-1 × VEGF, PD-1 × LAG-3) are an active area of development.
Explore the antibodies
The Antibody Design Lab has dedicated workspaces for the major checkpoint inhibitors:
- Pembrolizumab — humanized anti-PD-1 IgG4
- Nivolumab — fully human anti-PD-1 IgG4
- Atezolizumab — humanized anti-PD-L1 IgG1 with Fc-silent engineering
Each workspace shows the heavy and light chain sequences, the six CDR loops, ImmuneBuilder structure prediction, and AntiFold-based CDR variant design.
Bottom line
Checkpoint inhibitors changed cancer treatment by attacking the problem from a totally new angle. Instead of trying to poison cancer cells directly, they remove the inhibitory signals that let tumors hide from the immune system. Pembrolizumab and Nivolumab are the most clinically successful drugs ever produced by this approach, and the underlying biology — PD-1, PD-L1, T cell exhaustion — has become foundational vocabulary in modern oncology.