If you've heard an oncologist mention "immunotherapy" in the last few years, there's a good chance they meant a checkpoint inhibitor. Drugs in this class — Keytruda, Opdivo, Tecentriq and others — have become some of the most important cancer treatments of the era, and they earned their inventors a Nobel Prize. But the way they work is genuinely counterintuitive, and the marketing names give nothing away. Here's the idea, without the acronym soup.
The brakes, not the gas
Most older cancer drugs are like pressing the gas pedal: chemotherapy poisons fast-dividing cells, radiation burns them, targeted drugs jam a specific switch. Checkpoint inhibitors do something different. They don't touch the cancer at all. Instead, they remove a brake that the cancer has been using to stop your immune system from attacking it.
Your immune system has built-in safety checks — "checkpoints" — that stop it from attacking your own healthy tissue. Without them, you'd get autoimmune disease. These checkpoints are essentially off-switches on your T-cells. The problem is that many tumors have learned to press those off-switches on purpose, telling the immune system, "nothing to see here, stand down." A checkpoint inhibitor blocks that trick, so the immune cells wake back up and recognize the cancer for what it is.
PD-1 and PD-L1, finally explained
These two terms come up constantly, and they're simpler than they sound. PD-1 is a receptor — an off-switch — sitting on the surface of a T-cell. PD-L1 is the matching key. When a cell displays PD-L1 and it touches the PD-1 switch on a T-cell, the T-cell powers down. This is a normal, healthy mechanism: it's how your body keeps immune cells from running wild.
Cancer cells exploit it by coating themselves in PD-L1. Every T-cell that comes to investigate gets switched off before it can attack. A PD-1 or PD-L1 inhibitor is an antibody that physically gets in the way of that handshake — it covers either the switch or the key, so the off-signal never goes through. The T-cell stays awake and does its job.
This is also why you'll hear about "PD-L1 testing." Tumors with a lot of PD-L1 are relying heavily on this particular brake, so blocking it tends to help more. Tumors with little PD-L1 may be hiding in other ways, which is part of why these drugs don't work for everyone.
CTLA-4, the other brake
PD-1 isn't the only checkpoint. CTLA-4 is a second off-switch that acts earlier, in the lymph nodes, while T-cells are first being trained for battle. The drug ipilimumab blocks CTLA-4. Because the two brakes work at different stages, doctors sometimes combine a CTLA-4 blocker with a PD-1 blocker — releasing two brakes at once often produces stronger responses, at the cost of more side effects.
Why the side effects look like autoimmune disease
Here's the logical consequence of the whole mechanism: if you take the brakes off the immune system, sometimes it attacks healthy tissue too. So the side effects of checkpoint inhibitors are unlike chemotherapy's. Instead of hair loss and nausea, you see inflammation that can appear almost anywhere — the skin, the gut (colitis), the thyroid and other hormone glands, the lungs, occasionally the liver.
These reactions are usually manageable when caught early, often with steroids that calm the over-active immune response. But they can be serious, and they can show up weeks or even months after treatment starts. That's why patients on these drugs are told to report new symptoms promptly rather than waiting — a persistent cough or a change in bowel habits means something different on immunotherapy than it would otherwise.
Why it works brilliantly for some and not at all for others
This is the central puzzle of the field. For some people with melanoma or lung cancer, checkpoint inhibitors produce durable remissions that last years — a near-miracle for cancers that were once rapidly fatal. For others with the same diagnosis, they do nothing.
The difference comes down to whether the immune system can recognize the tumor in the first place. Cancers with many mutations tend to look more "foreign" and respond better. Tumors that are cold — invisible to the immune system, or walled off from it — often don't respond, because releasing the brakes does no good if no immune cells ever show up. A great deal of current research is about turning cold tumors hot: combining checkpoint inhibitors with vaccines, other immunotherapies, or radiation to get the immune system to notice the cancer at all.
The short version
Checkpoint inhibitors don't kill cancer directly — they block the off-switches (PD-1, PD-L1, CTLA-4) that tumors use to shut down your immune defenses, letting your own T-cells do the attacking. That's why their side effects look like autoimmune flare-ups, and why they work spectacularly for some patients and not others. To see which combinations and cancers are being tested now, our cancer immunotherapy trials overview and immunotherapy combinations guide follow the active research.