Thinking on linkers for antibody-drug conjugates

In the chemical world, the tiny hyphen in antibody-drug conjugates—or ADCs, as they’re commonly known—does a lot more work than the perfunctory punctuation suggests. That hyphen stands for the chemical linker that ties an ADC’s tumor-targeting antibody to its cytotoxic drug. While linkers may seem like little more than chemical tethers, ADC makers are discovering that linker chemistry has an important role to play in these cancer therapies.

The ADC concept has been around since the early 1900s, but it took nearly a century for the first ADC to make it to market. To date, 11 ADCs approved by the US Food and Drug Administration are on the market, and 266 ADC drug candidates are being actively tested in clinical trials around the world, according to Hanson Wade’s Beacon platform, which provides information about clinical trials.

In theory, ADCs selectively kill tumor cells, sparing healthy cells: The antibody portion of the molecule gloms onto a specific target on a tumor cell’s surface. The cell then engulfs the ADC. Once it’s inside the cell, chemical, biological, or enzymatic processes release the drug portion, usually referred to as the payload. But it’s the type of linker used to hold the antibody and the drug together that determines how the payload gets cut loose from the ADC. So much of the action—from a chemistry perspective—lies with the linker.

Keep it together

ADCs’ generic names have two parts—a first name for the antibody and a second name for the linker-payload combination. Mylotarg, the first ADC approved by the FDA, is known generically as gemtuzumab ozogamicin. Early ADC linkers were very simple, such as organic chains with acid-labile hydrazone groups. They were just there to connect the antibody to the drug, says Hyunyong Cho, chief scientific officer of Pinotbio, a company working in the ADC linker-payload space.

These early linkers were susceptible to cleavage before reaching their target, so they released their toxic payload in a way that damaged healthy cells and sickened patients. Another common problem was that because ADCs have greasy or sticky payloads, they tended to aggregate when drugmakers wanted to add multiple payloads to an ADC.

Now, as drugmakers regard the linker as an important part of the ADC, companies like Pinotbio are trying to add even more value to the linker. “We have invested a significant amount of time and resources to improve linker technologies for ADCs,” says Pinotbio’s CEO, Dooyoung Jung.

Pam James, vice president of product with Vector Laboratories says although ADCs seem simple conceptually, they are quite nuanced. “Even with the exact same antibody and payload, if you change the linker, you get a completely different outcome,” she says. Linkers “add more value than just sticking payload to an antibody.”

Last year, Vector Labs acquired Quanta BioDesign, a company that makes polyethylene glycol (PEG) linkers. PEG, which is hydrophilic, can enhance the properties of ADCs. “PEG has a little bit of a bad reputation as this big gemisch of different sizes of chains,” James says, but it’s possible to create specific PEG architectures to optimize what the linker is doing for the ADC. For example, PEG groups can shield the sticky or greasy hydrophobic ADC payload, preventing aggregation. They can also protect the trigger points where a linker gets cleaved, preventing early payload release.

Companies entering the ADC space often think the molecules are plug and play, James says. But it’s a mistake to think that it’s simply a matter of using any linker to join a payload to an antibody that can reach the right target. “It’s just not really that way. For the success of the ADC field, there has to be more willingness to try these different linker platforms,” she says.

Scientists at Seagen, an industry leader in ADCs that was acquired by Pfizer last year, have seen this firsthand in the laboratory. Seagen researchers used the linker-payload combination called vedotin in drugs like Adcetris (brentuximab vedotin) and Padcev (enfortumab vedotin).

Vedotin contains both the cytotoxic payload monomethyl auristatin E (MMAE) and a valine-citrulline dipeptide that gets cleaved by cathepsin enzymes. But Sharsti Sandall, a cell biologist formerly with Seagen and now with Pfizer, says tinkering with that linker can bring about big changes to an ADC.

Seagen scientists developed a different linker that replaces the dipeptide with a glucuronide group, which is cleaved by β-glucuronidase enzymes instead of cathepsins, and added a PEG chain to boost solubility.

“What we found is that just by changing the linker, you changed the way and the manner in which the MMAE was released in the cell,” Sandall says. With the glucuronide linker, MMAE was better retained in the cell than with the dipeptide linker. “To me, that was just incredible,” she says. “It’s the same drug. Why would it be so different?”

While the team has a hypothesis—that the linkers are cleaved at different times in the lysosome, leading to different rates of escape to the endosome—they’re not exactly sure. When it comes to choosing a linker, Sandall says, “I’d love to tell you we’ve figured out the rules for this, but we haven’t. It tends to be empirical, and it tends to be something you just have to start testing.”

The stability question

One area that ADC makers are still grappling with is how stable to make the linker. Sandall says that with the MMAE payload, making the linker more stable makes the ADC more toxic.

“So far, the more unstable drugs are the ones that are working clinically, but there’s this push right now preclinically to go more stable,” she says. “I’m curious to see how that will play out for other payloads.”

Stuart Barnscher, senior director of preclinical programs in the ADC therapeutic development department at Zymeworks, has been working on ADC linker technology for more than a decade. He says that in the 2010s, ADC makers were very concerned about linker stability.

“Everybody wanted to make hyperstable antibody-drug conjugates with the linkage from the drug-linker to the antibody being rock solid,” as the ADC circulates through the body, Barnscher says. “We did that for the better part of 10 years. Lots of molecules entered the clinic, and not a single drug approved has a completely stable linker.”

An ADC with a rock-stable linker could clear the clinic in the future, Barnscher says, but “thus far, none of the ADCs that have been approved do, and it’s not for lack of trying.”

Yet some researchers do point to linker instability as a problem for ADCs. In 2000, Mylotarg became the first ADC approved by the FDA. But the drug was withdrawn from the market in 2010 at the FDA’s request because of toxicity problems. The FDA reapproved the drug in 2017 with a lower recommended dose, an altered dosing schedule, and a new patient population.

ADC linker expert Fernando Albericio, at the University of KwaZulu-Natal, says that the ease with which Mylotarg’s acyl hydrazone linker is cleaved has likely been the source of this drug’s woes. “In theory, it is very nice to have a cleavable linker. But in practice it is one more extra difficulty in a very, very complex molecule,” Albericio says.

Conservative chemistry

Even as drugmakers recognize that linker chemistry can influence an ADC’s success, they’ve been conservative in introducing new linkers. In the 11 ADCs currently approved by the FDA, there are seven distinct linkers. Barnscher points out that ADC makers Seagen and ImmunoGen, which was acquired by AbbVie in 2023, drove the industry for many years. Other firms looked to mimic their success with similar linkers.

Dowdy Jackson, an expert on ADCs and CEO of Jackson Consulting Group, agrees. “What a lot of people are trying to do right now is maintain the type of linker chemistries that we know work.”

The current darling of the ADC field is Daiichi Sankyo and AstraZeneca’s Enhertu (trastuzumab deruxtecan). The deruxtecan linker-payload features a tetrapeptide linker that gets cleaved by proteases to release a topoisomerase I inhibitor payload called DXd.

Enhertu is used to treat people with several types of HER2-expressing cancer, even those where there are low levels of the HER2 protein. The drug generated $2.6 billion in sales last year. “Everybody wants to have that same type of success,” Jackson says.

To that end, Barnscher says he’s seen a trend where many companies are making ADCs with the molecule that DXd is based on, exatecan, as the payload. Researchers at Daiichi Sankyo originally developed exatecan as a small-molecule topoisomerase I inhibitor, but it was too toxic. Then they tried adapt it to an ADC, but they ran into solubility and aggregation problems. Daiichi Sankyo chemists got around the problem by transforming exatecan’s amine group into hydroxyacetamide, making the derivative DXd. Exatecan recently came off patent, and now many companies are trying to make nonaggregating ADCs with an exatecan payload by linking that molecule to antibodies with solubilizing linkers, which are usually PEG based.

But Barnscher thinks that ADC makers should be more innovative with their linkers. They’ve been relying on linkers that can be cleaved by cathepsins and proteases in the lysosome, he says. Creating linkers that respond to enzymes overexpressed in cancer cells would move the industry further along.

Marc Robillard, CEO of ADC linker company Tagworks Pharmaceuticals, says that in terms of linkers, “the field is really fine-tuning.” Companies that have a promising target and good antibody and are looking to get into the ADC game are likely to go with what’s already been validated clinically, he says. But others who are looking to innovate might want to adjust stability or change how easy it is to add the linker-payload to the antibody.

At Tagworks, scientists are creating ADCs that use click chemistry to release their payloads. The company’s ADCs are designed to bind targets that don’t get internalized but stay on the surface of cells or are in between cells. “These are fantastic tumor markers,” Robillard says.

Tagworks’ linker features a trans-cyclooctene that’s bound to a payload molecule. The idea is that after a patient has been dosed and the ADC has reached its target, the patient then gets a dose of tetrazine, which triggers the trans-cyclooctene to release the payload.

Looking ahead

Ultimately, Barnscher says, making a good ADC is about balance and thinking of the ADC as a holistic molecule. “It’s not just the antibody. It’s not just the linker. It’s not just the payload. It’s how those things all work together,” he says.

Thomas Bruckdorfer, chief scientific officer and vice president of business development for Iris Biotech says the chemistry of ADCs is one aspect of linker technology, but that there’s also the patent landscape to consider. Many companies have patented linker-payload combinations. Iris has a 152-page booklet called Linkerology , which highlights more than 100 linkers that the company sells. Bruckdorfer says Iris made Linkerology in the hope of inspiring ADC makers with options they might try to get around those patents.

Although there’s currently a lot of enthusiasm about ADCs, Jackson, the ADC consultant, says it’s important to be cautious. “I have been through these cycles before with ADCs,” he says. “When I first entered into the ADC space, there was a lot of interest because it was a way to achieve this magic bullet concept.”

But Jackson has seen many ADCs fail once they reach clinical trials. “Right now, we’re at the height of the of the ADC hype,” he says. The roller coaster will descend again, he says, unless there are some new linker-payload chemistries, new conjugation approaches, new targets, or new antibody formats. “I hope there will be some novel innovations in the field that can move it forward.”