As of May 2018, the Pharmaceutical Research and Manufacturers of America estimated there were more than 1100 oncology treatments in clinical trials or awaiting FDA approval in the United States. And there have been 63 new cancer treatments approved in the past five years, according to an IQVIA Institute report on global oncology trends. There are many players in cancer drug discovery, and each company tries to solve extremely difficult puzzles before the competition does.

The Cambridge Healthtech Conference on Drug Discovery Chemistry, which will be held in San Diego this April, will bring together speakers from a number of these companies to discuss their myriad paths to discover new drugs, from fragment-based approaches to machine learning–driven candidate screening.

Roderick E. Hubbard, PhD, who splits his time between an academic post at the University of York’s department of chemistry and pharmaceutical company Vernalis, will be presenting on his work at both institutions involving the Bcl-2 family of proteins, which some cancer cells generate to prevent apoptosis.

Vernalis currently has candidates targeting Bcl-2 proteins in Phase I clinical trials for various cancers, according to Hubbard, but he says his talk will also focus on the efforts to make the discovery platform that made those drug candidates possible.

“Modern methods in biology are identifying an increasing number of new, unprecedented potential targets for drug discovery,” he says. “The challenge for is how to establish [early] a robust drug discovery platform for the targets.”

Vernalis’s approach to this problem has been fragment-based discovery.

“You can find small compounds, or fragments, that bind to most sites on most targets,” Hubbard says. “The challenge for difficult targets like these protein-protein interactions is having robust assays in place and strategies to know what to do with the fragments.”

Key immune response drivers

FLX Bio will discuss research on inhibiting general control nonderepressible 2, or GCN2, a protein kinase essential for pathway tumors to maintain amino acid nutrient levels and suppress the immune response while proliferating.

DNA hypodermic needle
Drug developers hope to develop small molecule drugs that can interfere with oncogenic mechanisms. For example, scientists have developed inhibitors of transcription factors implicated in cancer growth. [wildpixel/Getty Images]
“We’re focused on identifying key drivers of the immune response, particularly within the tumor microenvironment,” FLX Bio CEO Brian Wong, MD, PhD, says. The FLX Bio pipeline consists primarily of small molecules, and spearheaded by FLX475, the company’s lead asset. It’s a “first-in-class and best-in-class CCR4 antagonist,” Wong notes, “and it is now in a Phase I/II study.”

CCR4 is a regulatory T-cell receptor that tumors target via chemokines—such as CCL17 and CCL22—recruiting those T cells from the circulation to suppress the normal antitumor immune responses of effector T cells. FLX Bio has reported that FLX475 given in conjunction with traditional checkpoint inhibitor immunotherapies shrank tumors more than immunotherapy alone in preclinical models.

“The industry has been interested in this target for a number of years,” Wong says, “but hasn’t had the insight into the biochemistry to be able to develop a compound that can potently and selectively address the target.”

Machine learning

To crack the problem, FLX Bio has invested heavily into machine learning as a platform for breaking down the complex interplay of immune cells in the tumor microenvironment in a novel and systematic way, according to Wong. “We have been able to dissect various pathways out and assign priorities to them,” he says. “From that first path of work, these two targets came out, namely CCR4 and GCN2.”

FLX Bio is also using its machine learning platform to match patients who are likely to respond to the treatment targets uncovered in the drug discovery process. Wong says this is a key strategy for finding new drugs in a future where low-hanging fruit has been plucked. “I don’t think we’re going to have a situation where a mechanism is good across a majority of cancers,” he points out. “Much more precision will need to be applied to cancer immunology than in the past.”

Aurigene Discovery Technologies is also interested in the small molecule approach to immunotherapy, according to chief scientific officer Murali Ramachandra, PhD. His conference presentation will discuss some of the 16 different small molecules the company has in development, primarily candidates targeting the immunosuppressive TIGIT T-cell receptor and adenosine signaling pathways. The company’s most advanced asset, CA-170, currently in Phase II trials in conjunction with Boston’s Curis, targets the programmed death ligand 1, or PD-L1, that cancers utilize to inhibit T-cell activity.

CA-170, being a small molecule rather than a monoclonal antibody, is part of what sets Aurigene’s strategy apart, according to Ramachandra. “I believe small molecule approaches for targeting immune checkpoint pathways could offer unique advantages,” he maintains, including oral bioavailability and potential for simultaneously targeting related checkpoint proteins. “Additionally, small molecule agents offer significant advantages in terms of cost of goods.”

Discovering small molecule drug candidates in immuno-oncology is difficult, according to Ramachandra, in part because of the inherent difficult in matching small molecules with a relatively large receptor–ligand interface. “We took the approach of truncating high-affinity peptides or critical fragments from the interface to arrive at the shortest pharmacophore,” he says. “The shortest pharmacophore was further transformed into druggable molecules either by converting it into a peptidomimetic or an amino-acid-inspired small molecule.”

Protein degradation

Aurigene is also working on candidates for a targeted protein degradation approach, Ramachandra says, “that allows us to go after targets that are not considered ‘druggable’ or not ideally suited for inhibitors.”

Protein degradation strategies, particularly utilizing the ubiquitin proteasome system, are the raison d’être for newcomer Cullgen, which was founded last March, according to Michael B. Plewe, Ph.D., vice president of medicinal chemistry.

“It’s a very hot and interesting field,” he notes. “Some people believe that targeted protein degradation is going to be the future of drug discovery.”

Plewe will be presenting research on protein degradation strategies making use of the ubiquitin proteasome, the human body’s existing system for getting rid of old or malformed proteins. Proteins to be degraded are marked with the regulatory protein ubiquitin via the E3 ubiquitin ligase. Cullgen is developing ligands that bind to other, tumor-specific proteins, and it is linking them to E3 ligase. The permutations of this system, adds Plewe, allow for a potentially vast number of drug candidates to be researched.

“All the companies are using similar ubiquitin proteasome degradation systems, but there are different ways of making degraders, he continues. There are so many ways of assembling the degraders together, based on the E3 ligase ligand space and on the ligand that binds to the protein of interest. I believe every company has their own unique niche for how to assemble those degraders.”

While Cullgen is not ready to reveal its current drug candidates until after filing patents, Plewe noted two of the company’s founders—Yue Xiong, PhD, of the University of North Carolina at Chapel Hill, and Jian Jin, PhD, at Mount Sinai in New York—have published on protein degradation approaches targeting anaplastic lymphoma kinase, or ALK.

“We know that this specific protein is expressed only in the embryonic stage and is usually no longer expressed in adults,”  Plewe says. “But in cancer, it is highly active. So, by removing it, it should be failsafe to completely degrade it.”

This approach is so new, according to Plewe, that the large target-interaction databases necessary to train machine learning approaches do not yet exist, and the company instead is relying on the deep experience of the founders to suggest new targets to pursue. Plewe is confident about the new approach. “It works, I tell you,” he says with a laugh. “We see solid tumor growth inhibition in mice for multiple degraders using this system.”

Gene expression regulators

Genentech will be sending Karen Gascoigne, PhD, discovery oncology scientist, to the April conference to present on research into gene expression regulators, the histone acetyl-transferases CBP/P300.

“[CBP] is particularly important in tumor cells for controlling gene expression and expression of oncogenes, but it also plays a role in some of the immune cells we might be interested in modulating,” she says. “This will be the first time we will be presenting that work. I will probably leave it till then to discuss that.”

Genentech is heavily invested in the development of multiple oncology drug technologies, from immunotherapies to protein degradation techniques, but Gascoigne’s work represents the company’s overarching strategy, one of seeking to understand the biology first, according to Shiva Malek, PhD, director and principal scientist, discovery oncology.

“It’s really driven by the biology and the pathways we have interest in, and then the technology is born out of that,” she explains. “Our thinking is that we need to leverage cancer genomics to understand the underlying biology of tumor cells, and using that information, identify novel ways to directly target tumors.”

Previous articleTaming the RNA Zoo While Saving Rare Transcripts
Next articleCirculating Tumor Cells and the Liquid Biopsy