Life Science

Leveraging Digital Chemistry to Create Best-in-Class Drug Candidates

For people with chronic medical conditions, a reliable therapeutic that treats their disease while preserving a good quality of life can be transformative. Patients are thankful when a medication helps relieve symptoms, even if that treatment comes with dosing challenges such as injections or infusions required for most biologic drugs. In our current era of precision medicine and targeted therapies, we can minimize side effects while at the same time, improving or maintaining the outcomes we expect from treatment. The following is a real-world example of how drug hunters at Morphic Therapeutic worked with Schrödinger’s team to combine digital chemistry with structural biology for the discovery and optimization of new molecules—based on known treatment mechanisms for inflammatory bowel disease (IBD)—producing a best-in-class drug candidate.

“A truly best-in-class treatment would allow patients to remain at home, live a more normal life, and not have to schedule around the availability of doctors and infusions.”

1. Identify the unmet need  

We looked at a currently approved approach to treating IBD: an antibody (Entyvio® or vedolizumab) that’s been helping patients for years. It works by keeping T cells in the blood and preventing them from entering the gut, where they can do harm by mistaking healthy gut cells as invaders that T cells are meant to destroy. In the body, T cells move through the highway of your blood, and the treatment ensures they stay on that path and don’t erroneously take an exit ramp into the gut. With this proven mechanism of action, Entyvio has become a validated treatment for IBDs like ulcerative colitis and Crohn’s disease. 

On a cellular level, a problem arises when two types of cells interact: the autoimmune T cells in the blood and the endothelial cells that surround the bloodstream. Gut homing T cells use an integrin pathway called alpha 4 beta 7 to make the connection with abundant receptors on the endothelial surface called MAdCAM-1. If allowed to interact with MAdCAM-1, the alpha 4 beta 7 pathway enables the release of T cells into the gut. Entyvio binds to alpha 4 beta 7 and prevents this interaction, thus its mechanism of action is “alpha 4 beta 7 inhibition.” 

With a comprehensive understanding of the disease pathway, Morphic set out to develop new approaches. More than 150,000 patients have used Entyvio successfully, but there’s room for improvement. Entyvio is an antibody, so it is delivered by infusion. This process requires a visit to the doctor and an IV treatment that takes time to infuse into the body. Patients and clinicians might both look at the dosing regimen for this treatment as an inconvenience and a major life-long time commitment. 

At Morphic, we saw an opportunity to create an oral medication that inhibits the same alpha 4 beta 7-MAdCAM-1 interaction. A truly best-in-class treatment would allow patients to remain at home, live a more normal life, and not have to schedule around the availability of doctors and infusions. 

“We needed to create a novel molecule at the center of our Venn diagram, a molecule that has never existed.”

2. Get the Venn diagram right 

Our team at Morphic knew that any oral, small-molecule alpha 4 beta 7 inhibitor would need to balance potency (efficacy), selectivity (how strong the molecule prefers to interact with target of interest), and pharmacokinetics (how a treatment is absorbed, distributed and eliminated from the body, often called PK). We also knew about the risk of a severe, rare nervous system side effect known as progressive multifocal leukoencephalopathy (PML), which can occur if our inhibitors are not selective over a closely related alpha 4 beta 1 target. Long-term blocking of the interaction between alpha 4 beta 1 and the VCAM-1 receptor is believed to trigger the rare side effect.

The solution is to find the molecule that does all that perfectly. No big deal, right?

Of course, it’s not that easy. Experts in the alpha 4 beta 7 integrin space have delivered a variety of potent compounds, but very few that are also selective and even fewer that also have good PK properties. We needed to create a novel molecule at the center of our Venn diagram, a molecule that has never existed. But designing a molecule to achieve all these (and more) specific criteria presents a major challenge. Fortunately, we already had a significant advantage—the world’s largest collection of proprietary crystal structures of these historically difficult to target integrins. This essentially revealed the fingerprints of the binding pockets, which made it possible for us to leverage digital chemistry approaches in partnership with Schrödinger. 

Thus, we began an important collaboration. Our own deep expertise in biology, structural biology and medicinal chemistry at Morphic and Schrödinger’s state-of-art computational platform were all critical to developing a potentially meaningful new medicine for patients. 

“The ability to collaborate through LiveDesign was key—it meant that no matter when or where inspiration might strike, we had a platform to capture and share ideas.”

3. Finding answers with collaboration, computational predictions, and predictive modeling 

At Morphic, integration of structural biology and computational chemistry plays a key role in all our drug discovery programs and medicinal chemistry campaigns. We focus on key interactions to enable the design of compounds that are more potent and more selective than what’s currently available or in development elsewhere. By collaborating with the Schrödinger team, we were able to focus specifically on data-driven, hypothesis-based designs.

Incorporating accurate physics-based predictions and machine learning models into our compound design and synthesis workflows, we accelerate the synthesis and testing of the most promising drug candidates. We also refine our future design cycle at a faster pace using computational chemistry. Our partnership with Schrödinger allowed us to optimize several important parameters simultaneously in silico: potency, selectivity and permeability. This approach yielded higher chances of finding molecules that combine high cellular potency—which could be indicative of a drug being more effective at lower doses in humans—with permeability—which is typically correlated to how efficiently a drug is absorbed in the body. 

We also captured all the project data in LiveDesign, including compound designs and hypotheses, computational models and predictions, and experimental data. This enterprise informatics platform allowed Morphic and Schrödinger teams to collaborate seamlessly and effectively across different geographical locations and time zones. The ability to collaborate through LiveDesign was key—it meant that no matter when or where inspiration might strike, we had a platform to capture and share ideas. For example, when picking up soup with my son and explaining the process by which cucumbers become pickles, I realized we could radically change the design of one molecular scaffold using key new insights. I was able to quickly draw the new scaffold designs into LiveDesign for the full team to view and begin exploring that hypothesis.

Those collaborative efforts led us to deliver multiple potential drug candidates and culminated in MORF-057, which is currently in Phase 2 trials for treatment of ulcerative colitis. MORF-057 has proven to be highly potent against the alpha 4 beta 7 protein target and highly selective against the key off-target alpha 4 beta 1 protein. The dosing is optimized for an acceptable benefit-to-risk ratio for an oral treatment, and patient trials are continuing. We’re really excited about MORF-057. 

4. Keep working it

You have to start smart. The key to creating a truly best-in-class treatment is to find a solid starting place. That process begins with knowledge of the disease, an understanding of molecular interactions of chemical matter with the target of interest, and a commitment to improving outcomes for the patient. 

Next, you keep refining and looking for better options. When you start with the best candidates that have the highest potency, selectivity and permeability, you start ahead of the learning curve and increase the odds for success. With collaborative efforts and high-speed computing, you can evaluate many options computationally and rapidly optimize compounds for even better properties. 

You’ve done your best; now it’s time to test. When you’ve used all the computational and predictive tools available, it’s time to create the best potential drug candidates, test them experimentally and pick the one best molecule to progress into the challenging (and costly) process of pre-clinical and clinical development. During these stages, a drug candidate is evaluated for safety and efficacy first in a lab setting and later in Phase 1-3 clinical trials. By sticking to high standards and meeting multiple stringent parameters in our compound optimization process, we have delivered a drug candidate in MORF-057 that we believe will be a meaningful medicine for patients. We take great pride in being pioneers of a potential best-in-class alpha 4 beta 7 inhibitor to help IBD patients.

Author Photo: Matt Bursavich

Matt Bursavich

Matt Bursavich is a seasoned drug hunter and the VP, Head of Chemical Sciences at Morphic Therapeutic. Morphic is leading the development of a new generation of oral integrin-targeting drugs, informed by decades of world-leading expertise and a unique integrin discovery platform. Morphic believes oral drugs targeting the integrin protein family can transform the treatment paradigm for patients suffering from serious chronic diseases, including autoimmune, cardiovascular and metabolic diseases, as well as fibrosis and cancer. Morphic is focused on scientific excellence and close collaboration with partners, including Schrödinger. Such partnerships expand the scope of discovery activity and potential pipeline breadth. 

Sign Up Today


Sign up to receive quarterly updates with the latest from Extrapolations.