ADC Linkers: Role, Types, and Impact on Cancer Therapy

The linker in an antibody-drug conjugate (ADC) is the chemical bridge connecting a tumor-targeting antibody to a cell-killing drug, and it has an outsized influence on whether the therapy works or causes harm. An ideal linker stays intact while the ADC circulates through the bloodstream, then releases its toxic payload precisely when it reaches the tumor.¹ That sounds simple, but getting the chemistry right has turned out to be one of the hardest engineering problems in modern oncology, and much of the progress in ADC design over the past decade comes down to smarter linker choices.

Why the Linker Matters More Than You Might Think

ADCs are sometimes described through a “magic bullet” analogy, first proposed over a century ago: an antibody finds and binds to a marker on cancer cells, delivering a potent poison directly to the tumor while sparing healthy tissue.² In practice, each part of the ADC (the antibody, the linker, and the cytotoxic drug) has to be chosen and tuned carefully. But the linker is often the component that makes or breaks a therapy, because it controls two competing demands: plasma stability and payload release. If the linker breaks apart too early, the drug leaks into the bloodstream and causes systemic toxicity, much like conventional chemotherapy. If the linker is too stable, the drug never fully detaches from the antibody once inside the cancer cell, and the treatment loses potency.

Existing linkers frequently release their payloads in the wrong place, leading to off-target toxicity that limits how much drug a patient can tolerate.¹ This is why so much research effort has gone into designing linkers that respond to specific conditions found only inside tumors or tumor cells, rather than conditions present throughout the body.

Cleavable Versus Non-Cleavable Linkers

The broadest distinction in ADC linker design is between cleavable and non-cleavable types. Both are used in approved drugs, and each has real advantages and trade-offs.

Cleavable linkers are designed to break apart in response to conditions that differ between healthy tissue and the tumor microenvironment. The most common triggers are:

Acid-sensitive (hydrazone): These linkers degrade in the low-pH environment found inside lysosomes, the cellular compartments where the ADC ends up after being swallowed by the cancer cell. The problem is that mildly acidic conditions also exist elsewhere in the body, so these linkers can be somewhat leaky.
Enzyme-cleavable (peptide): These contain short amino acid sequences recognized and cut by enzymes like cathepsin B, which are abundant inside lysosomes. The valine-citrulline dipeptide is one of the most widely used designs. Enzyme-cleavable linkers tend to offer a good balance of stability in plasma and rapid release inside the cell.
Disulfide: These break apart in the high-glutathione (reducing) environment inside cells. They offer reasonable selectivity but can be sensitive to the reducing conditions found in blood plasma as well.

Non-cleavable linkers take a fundamentally different approach. They do not break apart on their own. Instead, the entire antibody must be degraded inside the cell’s lysosome before the drug is freed, still attached to an amino acid residue from the antibody. The advantage is excellent plasma stability, since there is no chemical trigger that could accidentally fire in the bloodstream. The trade-off is that the released metabolite (drug plus amino acid fragment) is typically charged and cannot easily cross cell membranes, which limits how far the drug can spread beyond the cell that originally internalized the ADC.

The Bystander Effect and Why It Matters for Real Tumors

Tumors are not uniform. Within a single mass, some cells express the target antigen that the ADC’s antibody recognizes, and some do not. If the linker releases a drug that can only kill the cell it was delivered into, the antigen-negative neighbors survive. Over time, those surviving cells can repopulate the tumor.

This is where cleavable linkers and membrane-permeable payloads create what is called the “bystander effect.” After the drug is released inside one cancer cell, it can diffuse out and kill neighboring cells that the antibody never directly bound. In laboratory spheroid models, ADCs with cleavable linkers and permeable payloads showed drug-induced cell death spreading well beyond the directly targeted cells on the spheroid’s periphery, reaching bystander cells deeper in the tissue. By contrast, ADCs with non-cleavable linkers, such as T-DM1 (the drug Kadcyla), showed killing only in the directly targeted surface cells, with no spread into bystander populations.³

The clinical relevance of this distinction has become especially visible in breast cancer treatment. Kadcyla uses a non-cleavable linker and a payload (DM1) that cannot easily cross cell membranes, so its activity is largely confined to cells that directly internalize the ADC. Enhertu, a newer ADC targeting the same protein (HER2), uses a cleavable linker and a membrane-permeable payload, which enables a bystander effect and broader tumor killing, even in cancers with lower or more heterogeneous HER2 expression.⁴ This difference in linker and payload design is widely considered one of the key reasons Enhertu has shown activity in patient populations where Kadcyla had limited effect.

Drug-to-Antibody Ratio and the Hydrophobicity Problem

Loading more drug molecules onto each antibody sounds like it should make the ADC more powerful. In a test tube, that is true: higher drug-to-antibody ratios (DAR) increase cell-killing potency. But in a living body, the math changes. ADCs with very high DAR values get cleared from the bloodstream faster, which reduces the amount of drug that actually reaches the tumor.⁵

The reason is hydrophobicity. Most cytotoxic payloads are greasy, water-repelling molecules, and the more of them you attach, the more the ADC behaves like a fat-soluble particle that the liver eagerly scoops up. Preclinical work with maytansinoid conjugates found that ADCs with a DAR below about six had similar clearance rates, but those loaded with roughly nine to ten drug molecules per antibody were cleared rapidly and accumulated heavily in the liver, reaching two to three times the liver concentration of lower-DAR versions. Those overloaded ADCs actually had worse anti-tumor activity than their lighter counterparts, despite carrying more drug.⁶

Linker design is one of the main tools for managing this problem. By incorporating hydrophilic elements into the linker itself, such as short polyethylene glycol (PEG) chains or charged groups, researchers can partially offset the hydrophobicity that each drug molecule adds. The goal is to keep the ADC’s surface properties close to those of a naked antibody, so the body’s clearance machinery does not treat it as foreign debris.

Premature Payload Loss and Linker Stability

One persistent headache in ADC development is premature deconjugation, where drug molecules fall off the antibody while the ADC is still circulating. Many early ADCs used maleimide-based chemistry to attach the linker to the antibody’s cysteine residues, and these connections turned out to be reversible. The drug-linker can detach through a retro-Michael reaction, transferring the payload to blood proteins like albumin instead of delivering it to the tumor.⁷ This not only reduces the amount of drug reaching the cancer but also introduces systemic toxicity, since the freed payload is now drifting through the bloodstream attached to albumin.

A clever solution emerged in the form of self-hydrolyzing maleimides. By placing a basic amino group next to the maleimide ring, researchers created linkers that undergo rapid ring-opening (hydrolysis) at body temperature and neutral pH. Once the ring opens, the chemical connection becomes permanent, and the retro-Michael escape route is blocked.⁸ This kind of molecular engineering, making a small structural tweak that locks the linker in place, illustrates how linker chemistry has matured from an afterthought into a central design concern.

Overcoming Drug Resistance Through Linker Engineering

Cancer cells develop resistance to ADCs through several routes, and one of the most frustrating is multidrug resistance (MDR). Cells that overexpress the MDR1 drug-efflux pump can actively push cytotoxic payloads back out before they do any damage. The drug gets into the cell, the pump kicks it back out, and the ADC fails.

Linker design offers a way around this. When researchers replaced the standard MCC linker in a maytansinoid ADC with a more hydrophilic PEG4Mal linker, the resulting drug metabolite inside the cell was too hydrophilic for the MDR1 pump to efficiently export. In MDR1-expressing cancer cell lines, the PEG4Mal-linked conjugate was significantly more potent than the original design, while an MDR1 inhibitor barely improved the new conjugate’s performance, confirming that it was already evading the pump on its own.⁹ Conjugates bearing hydrophilic linkers with sulfonate or PEG groups generally showed a wider selectivity window, being equally or more potent against antigen-positive cells while being less toxic to antigen-negative cells, especially against MDR cell lines.¹⁰

The broader point here is that the linker is not just a passive tether. By controlling the chemical character of the metabolite that the cell eventually has to deal with, the linker shapes the intracellular pharmacology of the whole ADC. Linker modification has been shown repeatedly to overcome MDR1-mediated resistance in both cell culture and animal tumor models.¹¹

Conjugation Chemistry and the Push for Homogeneity

Traditional methods for attaching linkers to antibodies rely on the natural amino acids already present on the antibody’s surface, typically lysines or cysteines. The problem is that antibodies have many of these residues, and the chemistry is not selective about which ones get modified. The result is a heterogeneous mixture: some antibody molecules carry two drug molecules, some carry four, some carry eight, and they are attached at different positions. This variability affects how each individual ADC molecule behaves in terms of stability, clearance, and potency.¹²

Site-specific conjugation technologies aim to fix this by attaching drug-linkers at defined positions on the antibody, producing ADCs where every molecule has the same DAR and the same attachment sites. Several platforms have been developed, including engineered cysteine residues (THIOMAB), aldehyde tags (SMARTag), enzymatic methods using transglutaminase, glycan-remodeling approaches, and chemical modification platforms like AJICAP. In preclinical testing, site-specific ADCs generally show cleaner pharmacological profiles than their heterogeneous counterparts. One clinical comparison found that a site-specific conjugate (DAR 2) could be dosed at more than double the dose of a non-site-specific version using the same linker-payload combination, with comparable response rates around 56 to 59 percent. However, the site-specific version also brought higher rates of certain side effects, including ocular toxicities that sometimes forced treatment discontinuation.¹³

Homogeneity matters for manufacturing as well. The inherent heterogeneity introduced by traditional conjugation methods creates challenges for quality control, since every batch is a statistical distribution of species rather than a single defined product. Site-specific approaches simplify analytical characterization and help ensure that what comes out of the factory is consistent from lot to lot.

Extracellular Cleavage and ADCs That Do Not Need Internalization

For most ADCs, the assumed mechanism is that the antibody binds its target, the whole complex is pulled inside the cell through internalization, and the linker is cleaved inside lysosomes. But not all tumor antigens internalize well, and some tumors have stromal or vascular targets that sit outside cells entirely. This has spurred interest in linkers designed to be cleaved in the extracellular space by enzymes present in the tumor microenvironment.

One surprising finding was that the valine-citrulline peptide linker, long assumed to require intracellular cathepsin B for cleavage, can actually be cut by proteases in the extracellular matrix. Researchers showed that ADCs targeting non-internalizing antigens could still release their payload and kill tumor cells through extracellular proteolytic cleavage of this linker.¹⁴ This challenged a foundational assumption in ADC design and opened up the possibility of targeting antigens that had previously been considered unsuitable.

Building on this concept, newer linker platforms incorporate peptide sequences specifically designed to be cut by matrix metalloproteinases (MMPs), enzymes that are overexpressed in many solid tumors. One recent design created bivalent antibody constructs held together by an MMP-cleavable linker. In the tumor microenvironment, MMPs cut the linker, splitting the construct into two smaller cytotoxic species that can penetrate tissue more easily.¹⁵ This kind of design blurs the line between an ADC and a prodrug activated by tumor biology.

Safety Implications and the Linker’s Role in Side Effects

ADC side effects are not caused solely by the antibody targeting the wrong cells or the payload being inherently toxic. The linker contributes independently. Interstitial lung disease (ILD), a potentially serious inflammatory condition of the lungs, has emerged as a notable risk with several ADCs, particularly those using topoisomerase inhibitor payloads. An analysis in non-small-cell lung cancer found that the incidence of drug-induced ILD was not directly correlated with how much of the target antigen the tumor expressed, suggesting that factors beyond target binding, including the linker and payload characteristics, play a role.¹⁶

This is an active area of investigation, and there is no consensus yet on exactly how much linker choice contributes to ILD risk versus the payload or the antibody. But the observation that ILD does not track with antigen expression levels is an important clue. It suggests that linker stability in the lung, the rate and location of payload release, and the physical properties of the free drug all feed into the safety equation in ways that researchers are still working to untangle.

Aqueous Solubility and the Phosphate Linker Approach

A recurring practical challenge in ADC development is solubility. Many potent cytotoxic payloads are poorly soluble in water, and when these molecules are attached via hydrophobic linkers to an antibody, the resulting ADC can aggregate, losing both stability and manufacturability. One approach to this problem is incorporating phosphate groups into the linker itself. Phosphate-modified cathepsin B-cleavable linkers were shown to improve the aqueous solubility of drug-linker intermediates while maintaining strong plasma stability and rapid payload release inside lysosomes.¹⁷ Beyond improving manufacturing convenience, better solubility can also expand the range of payloads that can be used in ADC designs, since drugs that were previously too hydrophobic to conjugate cleanly become viable candidates.

Dual-Payload ADCs and the Role of Multi-Functional Linkers

Tumors can develop resistance to a single cytotoxic mechanism the same way bacteria develop antibiotic resistance: by adapting to survive the specific insult. One emerging strategy to pre-empt this is building ADCs that carry two different payloads, each with a distinct mechanism of cell killing, on the same antibody. Achieving this requires linkers that can accommodate two attachment points, each bearing a different drug molecule. Approaches include multi-functional linkers that present two chemically distinct reactive handles, as well as methods using non-canonical amino acids or enzyme-mediated conjugation to place each payload at a precisely defined site on the antibody. The goal is to deliver both drugs simultaneously into the same cancer cell, making it much harder for the cell to evolve resistance to both at once.¹⁸

Dual-payload ADCs are still largely in the preclinical stage, but they represent a logical extension of the principles that have driven single-payload linker design: specificity, stability, and controlled release. The linker in these constructs is doing even more work than usual, managing two independent drug molecules while keeping the whole assembly stable in circulation.

Computational Approaches to Linker Design

Designing a linker involves balancing so many competing demands, including stability, hydrophobicity, cleavage kinetics, metabolite properties, and manufacturing feasibility, that trial and error can be painfully slow. Machine learning and physics-based computational methods are increasingly being used to accelerate this process. These tools can predict how a given linker-payload combination will affect the ADC’s overall properties, suggest optimal conjugation sites, and estimate ideal drug-to-antibody ratios, reducing the number of compounds that need to be synthesized and tested in the lab.¹⁹ Given that each ADC is really a three-way optimization problem between antibody, linker, and payload, computational screening of the linker design space could meaningfully shorten development timelines for the next generation of conjugates.