After years of failed attempts to develop anticancer drugs directed at human epidermal growth factor receptor 3 (HER3, also called erbB-3), trial data on three promising assets were presented at the American Society of Clinical Oncology (ASCO) meeting earlier this year. These include an antibody–drug conjugate (ADC) from Daiichi Sankyo, a monoclonal antibody (mAb) from Elevation Oncology and a HER2–HER3 bispecific mAb from Merus—all of which are showing unprecedented degrees of efficacy, even when used as single agents.

Mutant HER2 and HER3 depend on each other to drive oncogenesis. Adapted with permission from J. Diwanji et al. Nature 600, 339–343 (2021), Springer Nature.

“It’s exciting to see drugs that target HER3 showing clinical activity,” says Pasi Jänne, a lung cancer specialist at the Dana-Farber Cancer Institute who has been involved in testing anti-HER3 therapies for over a decade.

Underpinning each therapeutic candidate is an antibody-based drug that fell flat in earlier trials. But companies alighted upon two winning strategies for revitalizing these assets. They either added toxic warheads, as Daiichi did, or focused on tackling a genetically defined type of HER3-addicted cancer. Others, meanwhile, are outfitting mAbs with radioactive particles, fashioning bifunctional fusion proteins and developing vaccines to elicit natural immune responses to HER3 (Table 1).

Table 1 Select HER3-targeted therapies in active development

HER3, like other members of the ‘HER’ protein family, plays critical roles in tumor progression and drug resistance. But HER3 stands apart from its more famous kin—the well-known anticancer targets epidermal growth factor receptor (EGFR, also called HER1) and HER2—in lacking any tyrosine kinase activity of its own. Instead, for its oncogenic functions, HER3 relies on its ability to form dimers with other receptors to induce downstream signaling events.

That road to tumorigenesis often starts with HER3 binding its ligand, a growth factor called neuregulin-1 (NRG1). But not in all cases. If HER3 dimerizes with the likes of HER2 and EGFR, it can still form cancer-promoting protein complexes even in the absence of NRG1—which might explain why the first generation of anti-HER3 therapies all flunked out in clinical development.

Former front-runners in the HER3-targeting race, drugs such as elgemtumab from Novartis and duligotuzumab from Roche, all targeted the extracellular domain where the receptor binds NRG1. This interaction shut down ligand-dependent signaling cascades. But as cancer biologist Mark Moasser and his colleagues at the University of California, San Francisco, reported earlier this year, that domain is not needed for HER3–partner dimerization. Oncogenic signalling requires only that internal kinase domains interact and pair up, Moasser showed, and “that’s why most of these drugs failed,” he says.

Daiichi, of Tokyo, had advanced its fully human IgG1 mAb patritumab into phase 3 trials, but suspended development in 2016 after finding that the treatment offered no additional benefit to patients with lung cancer who were also receiving an EGFR inhibitor. The company then reformulated the antibody as an ADC, using a stable tetrapeptide-based cleavable linker to tack on a topoisomerase I inhibitor. (The same warhead, a type of exatecan derivative, is found in the company’s HER2-directed ADC Enhertu (fam-trastuzumab–deruxtecan-nxki) approved for metastatic breast and gastric cancers.)

Now known as patritumab deruxtecan, the HER3-targeted ADC is undergoing phase 3 testing as a treatment for advanced lung cancers that harbor activating mutations of EGFR. It is also in earlier-stage trials for patients with HER3-expressing metastatic breast cancer.

Seribantumab, too, has had a roundabout clinical trajectory. Developed initially by Merrimack Pharmaceuticals of Cambridge, Massachusetts, the IgG2 mAb reached phase 2 testing in breast, lung and ovarian cancers, but ultimately proved to have little therapeutic value in those settings. Merrimack suspended further development and, in 2019, sold the drug’s rights to Elevation, a New York–based company founded by CEO Shawn Leland.

From the outset, Elevation’s express purpose was to develop a HER3-targeted mAb for those rare patients whose tumors are reliant on NRG1 signaling. That occurs when cancers harbor fusions containing the NRG1 gene, a rearrangement found in approximately 0.2% of solid tumors, most frequently in invasive mucinous adenocarcinomas of the lung and in pancreatic ductal adenocarcinomas.

“Leland considered licensing five different abandoned mAbs before deciding on the Merrimack agent.”

Leland considered licensing five different abandoned mAbs before deciding on the Merrimack agent. “Seribantumab checked all the boxes in terms of having a best-in-class profile and being a phase-2-ready asset,” he says. Elevation launched a 75-person, registration-enabling trial within 14 months of its licensing deal. Among the first 12 evaluable study participants, all of whom harbor NRG1 gene fusions, 2 showed complete tumor shrinkage, 2 had partial responses and 7 saw their tumors neither shrink nor grow.

Also at the ASCO meeting, Merus of Utrecht, the Netherlands, reported clinical success with its HER2×HER3 bispecific zenocutuzumab as a treatment for NRG1-fusion-positive cancers. The Merus drug yielded partial responses in 27 of 79 trial participants, with another 28 participants experiencing stable disease control.

A handful of other HER3-targeted mAbs and bispecifics remain in early-stage trials, mostly in China. Like other antibody-based candidates before them, these agents generally target the extracellular region of HER3. This is also the case with HMBD-001 from Hummingbird Bioscience of Singapore.

But while other HER-3-directed agents generally target the NRG1-binding region of HER3, Hummingbird designed its humanized IgG1 mAb to block the interface where HER3 normally dimerizes with signaling partners. This allows HMBD-001 to shut down both ligand-dependent and ligand-independent HER3-fueled tumor growth, resulting in superior tumor control when compared to clinical-stage competitors in a variety of mouse models.

“That’s what gives us confidence that HMBD-001 will succeed in HER3-driven cancers where others may have seen historically disappointing results,” says Hummingbird co-founder and CSO Jerome Boyd-Kirkup. Initial data from the company’s first-in-human trial involving patients with advanced HER3-expressing solid malignancies, including NRG1-rearranged cancers, are expected later this year.

Patients with metastatic lesions in the brain are unlikely to benefit, however, because mAbs and bispecifics do not readily cross the blood–brain barrier. “You can’t rely on antibodies alone to go after these tumors,” notes Lali Medina-Kauwe, a cancer biologist at Cedars-Sinai Medical Center in Los Angeles. That’s why she favors a type of HER3-targeted nanoparticle delivery paired with chemotherapy drugs such as doxorubicin. In a preprint, she and her colleagues showed how the approach could reduce the growth of breast metastatic breast tumors. Medina-Kauwe had been advancing the strategy through a startup she founded called Eos Biosciences, but the company went belly-up during the pandemic.

Others are pursuing small-molecule drugs that target the defective kinase domain of HER3 and lock the protein in a conformation that prevents it from forming heterodimers. In 2020, for example, AstraZeneca scientists, in collaboration with researchers at the Francis Crick Institute in London, reported the fruits of a 100,000-compound drug screen that looked for HER3-binding molecules capable of interfering with HER2–HER3 complex formation. They found some promising hits. But according to study author Audrey Colomba, now at the University College London Drug Discovery Institute, no further optimization steps were ever performed. Other academic groups have also described proteolysis-targeting chimeric molecules that induce HER3 degradation.

Another outside-the-box HER3-targeting approach comes from tumor immunologist H. Kim Lyerly of Duke University in North Carolina, who for years has been working on a cancer vaccine to activate HER3-specific immunity. In mouse models, Lyerly and his colleagues showed that his vaccine strategy could elicit T cell and polyclonal antibody responses. Those immune responses then led either to complete tumor elimination or to loss of HER3 expression, which rendered the cancers more susceptible to other targeted therapies.

A DNA-plasmid-based vaccine is now being evaluated in an academic-sponsored trial at Duke, funded by the US Department of Defense. Replicate Bioscience, a startup in San Diego, California, that Lyerly cofounded two years ago, also plans to incorporate HER3 antigens (alongside others) into its lead product candidate, a self-replicating RNA vaccine designed to prevent or combat mutations associated with resistance to endocrine therapy in patients with breast cancer.

“HER3 is increasingly appreciated as a mechanism causing cancers to be more aggressive and resistant to treatment.”

“HER3 is increasingly appreciated as a mechanism causing cancers to be more aggressive and resistant to treatment,” says Replicate cofounder and CEO Nathaniel Wang. And by educating the immune system to recognize and attack HER3-expressing tumors, Wang hopes his company’s technology will thwart that resistance mechanism from taking root in vaccinated patients. A first-in-human trial is planned for 2023.

HER3 is additionally the focus of at least one drug candidate outside oncology. JK07 is an experimental agent from Salubris Biotherapeutics, of Gaithersburg, Maryland, that combines an anti-HER3 IgG1 mAb with an active peptide fragment of NRG-1. The recombinant fusion protein is designed to take advantage of NRG-1’s potential to combat heart failure—the growth factor spurs the formation of HER2–HER4 dimers in heart muscle cells to promote cardiac regeneration, while minimizing its signaling via HER3, which can lead to diarrhea, nausea and other side effects

“We’re coming at this from the cardiovascular perspective,” says SalubrisBio CEO Sam Murphy. And although the point of including a HER3-targeted mAb in JK07 is to create a more selective HER4 agonist, with half-life extension as an added bonus, “the net effect,” notes Murphy, “should be minimal to no activation of HER3.” And that is ultimately the goal of HER3-targeted therapies in oncology, too.