What Is CTLA-4 and How Does It Fight Cancer?

CTLA-4 (cytotoxic T-lymphocyte-associated protein 4) is a receptor on the surface of T cells that acts as a brake on the immune system. In cancer therapy, drugs that block CTLA-4 release that brake, letting T cells attack tumors more aggressively. The first such drug, ipilimumab, became the first treatment ever shown to extend survival in advanced melanoma, and its success launched the entire modern field of checkpoint immunotherapy. But the story of CTLA-4 is more nuanced than “block the brake, kill the cancer,” involving mechanisms researchers are still untangling and side effects that remain a serious clinical challenge.

The Immune Brake That Tumors Exploit

Your immune system has a built-in safety system to prevent T cells from attacking your own tissues. CTLA-4 is one of its key components. When a T cell encounters something foreign, it needs two signals to become fully activated: one from recognizing the foreign material, and a second “costimulatory” signal delivered when a molecule called CD28 on the T cell binds to partner molecules (CD80 and CD86) on nearby immune cells. CTLA-4 competes with CD28 for those same partners, but instead of activating the T cell, it shuts the response down. CTLA-4 binds CD80 with roughly ten times the grip strength that CD28 does, so when CTLA-4 is present, it wins the competition and the T cell stays quiet.¹

This competition happens early, primarily in the lymph nodes where immune responses get started, rather than out in the body’s tissues where T cells do their actual fighting.² That timing matters. CTLA-4 essentially shapes which T cells get activated in the first place, acting as a gatekeeper at the start of the immune response rather than a dimmer switch at the end.³

CTLA-4 also does something more dramatic than simply outcompeting CD28. Through a process called transendocytosis, CTLA-4 physically strips CD80 and CD86 molecules off the surface of antigen-presenting cells and pulls them inside the T cell, where they are destroyed. This leaves the antigen-presenting cell with fewer costimulatory molecules available for other T cells, dampening the immune response even further.⁴ Recent work has shown that CD86 may actually be the more important target of this stripping process in practical terms, because CD86 is the dominant costimulatory molecule on many immune cells and CTLA-4 removes it very efficiently.⁵

How Blocking CTLA-4 Fights Cancer

Tumors survive partly by exploiting CTLA-4’s braking function. If the immune system’s gatekeepers are keeping T cells from fully activating against tumor proteins, then blocking CTLA-4 with an antibody should, in theory, unleash those T cells. That theory drove the development of ipilimumab, the first anti-CTLA-4 antibody approved for cancer treatment. But the anti-tumor effect turns out to involve more than just releasing the brake on activation.

Within tumors, a large proportion of the T cells present are regulatory T cells, which express high levels of CTLA-4. These regulatory T cells suppress anti-tumor immune responses from the inside. When an anti-CTLA-4 antibody binds to CTLA-4 on these cells, it doesn’t just block the receptor’s function; it also tags those regulatory T cells for destruction by macrophages in the tumor. This killing depends on the antibody’s ability to engage immune cells through its Fc region, meaning the antibody’s structure matters, not just its ability to bind CTLA-4.⁶ Critically, this depletion happens selectively inside the tumor, not throughout the body, because the macrophages needed for the killing are concentrated in the tumor microenvironment.

Research has shown that both mechanisms contribute. Blocking CTLA-4’s suppressive signaling and depleting regulatory T cells inside the tumor are each necessary, but neither alone is sufficient for the best outcomes.⁷ This dual mechanism helps explain why simply blocking the receptor with a molecule that can’t engage macrophages produces weaker anti-tumor effects.

What the Clinical Data Show

The clinical story of anti-CTLA-4 therapy begins with ipilimumab in advanced melanoma, a disease that was almost uniformly fatal before immunotherapy. In a pooled analysis of over 1,800 patients from multiple trials, median overall survival with ipilimumab was about 11 months, but the real headline was the tail of the survival curve: roughly one in five patients was still alive at three years, and survival rates held steady from year three onward, with follow-up extending up to ten years.⁸ That plateau was unprecedented. Before ipilimumab, virtually no treatment for metastatic melanoma produced durable responses lasting years. The stable survival curve after three years suggested that the roughly 20% of patients who made it that far may have been effectively cured.⁹

Longer follow-up data from phase II studies confirmed that five-year survival rates held relatively steady from year five out to year six, reinforcing the idea that responders maintained durable benefit.¹⁰

Why CTLA-4 Blockade Is Often Combined with PD-1 Blockade

CTLA-4 and PD-1 are both immune checkpoints, but they operate at different stages and in different locations. CTLA-4 acts early, in the lymph nodes, shaping which T cells get activated. PD-1, the other major checkpoint target, acts later, in the tissues where T cells encounter their targets. Because the two brakes are independent, blocking both at the same time can produce stronger anti-tumor effects than blocking either alone.

Preclinical work in mouse tumor models demonstrated that concurrent blockade of both checkpoints produced synergistic anti-tumor activity, meaning the combined effect was greater than what you’d expect from simply adding the two individual effects together. Importantly, this synergy only appeared when both drugs were given at the same time; giving them sequentially did not produce the same benefit.¹¹ The combination increased the ratio of tumor-killing T cells to regulatory T cells within tumors and boosted immune signaling molecules like interferon-gamma far beyond what either drug achieved alone.

In the landmark CheckMate 067 trial, which followed patients with advanced melanoma for ten years, the combination of nivolumab (a PD-1 inhibitor) plus ipilimumab produced a median overall survival of nearly 72 months, compared with about 37 months for nivolumab alone and roughly 20 months for ipilimumab alone. By the end of the trial, over a third of patients receiving the combination were still alive.¹² These results cemented the combination as a standard of care in melanoma and spurred ongoing investigation in many other cancer types.¹³

The improvement comes at a cost. Combination therapy produces higher rates of immune-related side effects than either drug alone.¹⁴ That trade-off is central to ongoing research aimed at identifying which patients truly need both drugs and which would do just as well with a single agent.

The Side Effect Problem

Because CTLA-4 normally keeps the immune system from attacking the body’s own tissues, blocking it can trigger autoimmune-like reactions called immune-related adverse events. A systematic review and meta-analysis found that roughly 72% of patients treated with ipilimumab experienced some form of immune-related side effect, and about 24% experienced severe ones. The most common were skin reactions like rash and itching, followed by inflammation of the colon. Less frequently, the liver, pituitary gland, and thyroid were affected. Rare but serious events included nerve damage and blood cell abnormalities. Death from immune-related side effects occurred in under 1% of patients.¹⁵

Side effects were dose-dependent: patients on the higher dose of ipilimumab (10 mg/kg) had rates of any immune-related event near 79%, versus about 61% for the lower dose (3 mg/kg). These events typically began around ten weeks after starting treatment, though different organs were affected on different timelines. Skin reactions tended to appear first, followed by gastrointestinal problems, with hormonal disruptions arriving somewhat later.

The severity of these side effects is a major reason researchers are working to develop smarter versions of anti-CTLA-4 drugs. The current approach sends the antibody everywhere in the body, but the therapeutic action is needed primarily in the tumor and the local lymph nodes. All that off-target activity in healthy tissues is what drives most side effects.

Why Some Patients Don’t Respond

Despite the striking long-term survival seen in responders, the majority of patients treated with anti-CTLA-4 therapy don’t experience lasting benefit. Understanding resistance is one of the most active areas of research. Several mechanisms have been identified, and most fall into one of three categories.

First, the tumor itself can lose the ability to respond to immune signals. A study of melanoma patients found that three-quarters of those who didn’t respond to ipilimumab had genetic losses affecting the interferon-gamma signaling pathway in their tumors, meaning the tumor cells had lost the molecular machinery needed to respond when T cells attacked. None of the patients who did respond had these defects.¹⁶

Second, T cells themselves can become exhausted, accumulating multiple additional inhibitory receptors beyond CTLA-4, or undergoing metabolic changes that reduce their fighting capacity. Third, the tumor microenvironment can be stacked with suppressive cells and metabolites that dampen immune responses regardless of checkpoint blockade.¹⁷ Often, multiple resistance mechanisms operate simultaneously in the same patient, which is part of why this problem is so difficult to solve.

Early work on biomarkers suggests that the immune cell profile in a patient’s blood before treatment may help predict who will respond. Researchers have found that the distribution of memory T cell subsets differs between responders and non-responders, and these patterns are distinct from the biomarkers that predict response to PD-1 blockade.¹⁸ This is still an emerging field, and no blood test reliably guides treatment selection yet, but the data suggest one may eventually be feasible.

Next-Generation Anti-CTLA-4 Drugs

Much of the current development effort is aimed at keeping the anti-tumor activity of CTLA-4 blockade while reducing the systemic side effects. Two main approaches are in preclinical testing. The first uses “probody” technology, where the antibody is designed as a prodrug that stays inactive until it encounters enzymes that are abundant in the tumor microenvironment but scarce in healthy tissues. In mouse models, a probody version of anti-CTLA-4 depleted regulatory T cells within tumors, boosted tumor-specific killer T cell responses, and showed reduced toxicity compared to conventional anti-CTLA-4 antibodies.¹⁹

The second approach uses antibodies engineered to be active only in the acidic environment found inside tumors. Because most healthy tissues maintain a neutral pH while tumors tend to be acidic, these “conditionally active” antibodies bind CTLA-4 strongly in the tumor but barely at all in normal tissue. In animal testing, these showed similar anti-tumor effects to conventional antibodies but with substantially reduced toxicity in primate safety studies when combined with a PD-1 inhibitor.²⁰

Separately, researchers are also focusing on antibody engineering to optimize the Fc region, the part of the antibody that interacts with macrophages. Since depleting regulatory T cells inside the tumor depends on this interaction, designing antibodies with enhanced Fc function could improve efficacy. However, timing becomes critical: if the anti-CTLA-4 antibody is given while killer T cells are actively expressing CTLA-4 during activation, the enhanced Fc function can inadvertently destroy the very cells you want fighting the tumor. In mouse studies, delaying tumor antigen exposure by several days after antibody treatment avoided this problem and significantly improved survival.²¹

The Gut Microbiome Connection

One of the more unexpected findings in anti-CTLA-4 research is the role of gut bacteria. Certain bacterial species appear to influence both how well the therapy works and how severe its side effects are. Higher levels of Bacteroides fragilis and bacteria from the Faecalibacterium genus in the gut have been linked to stronger anti-tumor responses and lower rates of colitis during CTLA-4 blockade. Conversely, certain species like Bacteroides intestinalis have been associated with greater toxicity.²²²³

Bacteria involved in vitamin B synthesis and polyamine transport have also been linked to colitis resistance without sacrificing therapeutic benefit. This has opened up a whole line of investigation into whether modifying the microbiome, through probiotics, fecal transplants, or dietary changes, could improve the safety profile of checkpoint therapy. The work is still early, but it adds an entirely new dimension to thinking about how to optimize these treatments.

When You Want More CTLA-4, Not Less

In cancer, the goal is to remove CTLA-4’s brake on the immune system. But in autoimmune diseases and organ transplantation, the opposite problem exists: the immune system is too active and needs to be reined in. This has led to a parallel line of medicine where CTLA-4 is used as a drug rather than a target. Abatacept is a fusion protein that mimics CTLA-4, binding to CD80 and CD86 on antigen-presenting cells and blocking them from activating T cells through CD28. It is approved for the treatment of rheumatoid arthritis.²⁴ A more potent version called belatacept, engineered to bind CD86 more tightly, is approved for preventing rejection in kidney transplant recipients.²⁵

There’s an ironic complication in these drugs. Because abatacept and belatacept work by blocking CD80 and CD86, they block the interaction of these molecules with both CD28 (the activating receptor, which you want to block) and CTLA-4 itself (the inhibitory receptor, which you don’t want to interfere with). By inadvertently preventing CTLA-4 from doing its natural suppressive job, these drugs may partially undermine their own effectiveness.²⁶

The existence of people born with only one working copy of the CTLA-4 gene underscores how important this molecule is for immune balance. These individuals develop a syndrome featuring overactive T cells, infiltration of organs by immune cells, and progressive loss of certain immune cells from the blood. Unlike mice missing one copy (which appear normal), humans are far more sensitive to reduced CTLA-4 levels.²⁷ This finding is a useful reminder that CTLA-4 blockade in cancer therapy isn’t hitting a dispensable target; it is deliberately removing a molecule the immune system genuinely depends on, which is why the side effects are real and expected rather than incidental.

Cost and Access

Even when anti-CTLA-4 therapy works, the question of who can access it looms large. A cost-effectiveness analysis of nivolumab plus ipilimumab for advanced non-small cell lung cancer found that the combination added roughly $202,000 in cost per patient compared to chemotherapy, for an additional half year of quality-adjusted life. The resulting cost per quality-adjusted life-year was over $400,000, well above the commonly used threshold of $100,000 that health economists consider acceptable. In modeling, the combination only became cost-effective if monthly treatment costs were slashed by about 80%.²⁸ These numbers will shift as patents expire and biosimilars enter the market, but for now, the economics of combination checkpoint therapy remain a barrier in many health systems.

Checkpoint Therapy for Dogs

The CTLA-4 pathway isn’t unique to humans. Dogs develop spontaneous cancers that share many features with human tumors, including similar genetic patterns, immune signatures, and metastatic behavior. This has made them a valuable population for testing immunotherapy approaches in a setting that more closely mirrors real human cancer than standard mouse models do. Researchers have developed fully canine anti-CTLA-4 antibodies, designed to be non-immunogenic in dogs, that block CTLA-4 from binding its targets and enhance T cell activity.²⁹ The key binding site on CTLA-4 is conserved across humans, mice, and dogs, which makes cross-species translation feasible.³⁰ Beyond helping pets with cancer, these trials also serve as a bridge for testing drug combinations and identifying biomarkers of response in a way that is impossible in traditional clinical trials, where tumor heterogeneity and natural disease progression are hard to replicate in controlled lab settings.