Standigm uses AI to discover potential Crohn’s disease treatment

Crohn’s disease is an insidious, formidable condition. In the US alone, it impacts more than half a million patients, with approximately 33,000 new cases diagnosed every year (according to the Crohn’s and Colitis Foundation​), and the effects can be debilitating.

Standigm—a company that employs artificial intelligence (AI) in drug discovery—is setting its sights on coming up with a drug candidate to treat Crohn’s disease. To learn more about how Standigm is putting AI to work to explore treatment possibilities for several conditions, including Crohn’s disease, Outsourcing-Pharma connected with Hanjo Kim, senior vice president of global strategy and head of medicinal chemistry with the company.

OSP: Could you please share a little bit about Standigm—who you are, what you do, key capabilities, and what sets you apart from the competition?​

HK: Standigm is a Seoul-based workflow AI drug discovery company co-founded in 2015 by three drug discovery science and technology experts. The three joined forces to bridge their areas of expertise across AI, chemical engineering, and systems biology to eliminate some of the false starts in drug discovery by employing unique AI algorithms.

Standigm employs AI to detect patterns and trends relevant to human biology across vast biomedical databases. It uses three leading proprietary technologies, including hit generation-lead optimization, drug repurposing, and target discovery, and then applies novel design solutions to go from new disease targets to innovative molecules, in record time.

Although most AI-based drug discovery companies are focused on the creation of new compounds and the re-creation of new drugs, only a few companies have the technology to discover new targets with AI. Standigm has developed an ASK platform that can discover new targets to complete the search for new drug candidates with AI technology and to strengthen technological differentiation from other AI-based drug development companies. And, Standigm BEST, an AI platform for novel compounds can generate the 150-dimensional latent space constructed by various chemical features, including Al-based conditional modification and hit identification models.

Standigm has expanded its operations across the world, recently opening offices in the US. and UK. This expansion has enabled Standigm to carry out multiple projects simultaneously, using its workflow AI technology to meet the needs of various novel scenarios of early drug discovery. The ultimate goal of its workflow AI is to complete the closed-loop AI system in which any data produced in projects are fed back for continuous improvement while reducing the resources to secure patentable lead compounds inhibiting a novel target.

OSP: What has the history of Crohn’s disease treatment looked like in recent years—have there been any notable developments, or has the landscape been relatively quiet?​

HK: Crohn’s disease (CD) is an idiopathic, chronic, transmural inflammatory process of the bowel that can affect any part of the gastrointestinal (GI) tract from the mouth to the anus. The clinical symptoms include abdominal pain, weight loss, malaise, diarrhea, and/or rectal bleeding. The treatment of CD focuses on ameliorating symptoms while minimizing morbidity and maintaining intestinal continuity through the joint efforts of gastroenterologists and surgeons. The number of prevalent cases of CD in the USA, France, Germany, UK, Italy, and Spain will increase by 7%, from 1.4m to 1.5m, in 10 years. The prevalence of CD is highest in the United States at 243 per 100,000 persons.

The pathophysiology of CD is complex and involves many biological pathways including inflammation, the regulation of mucosal immunology, and autophagy. Immunomodulators and biologics increased significantly over recent decades. Despite their efficacy, these drugs are associated with toxicity issues, infractions, and malignancies.

The major drug classes used to treat CD are the 5-ASAs, corticosteroids, immunosuppressants, TNF-α inhibitors, CAM inhibitors, IL-12/23 inhibitors, and antibiotics. 5-ASAs are predominantly used to treat mild CD owing largely to their safety and low price, although experts are unconvinced of their efficacy. A short course of corticosteroids (less than six months) is generally effective to induce remission in moderate to severe CD, despite the safety concerns associated with their chronic use.

TNF-alpha inhibitors are the first-line biologics for CD. Physicians have a lot of experience with them so there is comfort in prescribing them and there is confidence in the agents’ efficacy and safety. Immunosuppressants, including 6-MP, azathioprine, and methotrexate, are mostly used for the maintenance of remission, largely due to their slow onset of action.

Physicians often prescribe a combination of a TNF-alpha inhibitor and an immunosuppressant (e.g., azathioprine) to achieve better disease control. The CAM inhibitor Entyvio (vedolizumab) has a better safety profile than the dose Tysabri (Natalizumab), the first-in-class anti-CAM therapy (approved for CD only in the US). The CAM inhibitors offer an alternative MOA for moderate to severe CD, especially in patients who have failed TNF-alpha inhibitors.

Entyvio is more gut-specific, so it has a better safety profile than that of Tysabri and is predominantly prescribed to moderate to severe CD patients who cannot tolerate or are refractory to TNF-alpha inhibitors Stelara (ustekinumab), which has been approved for moderate to severe CD, is the first-in-class IL-12/23 inhibitor to launch for CD. In general, it has a positive review on efficacy and safety in moderate to severe CD and will be anticipated to be continually used primarily as a later-line biologic following the failure of TNS-alpha inhibitors.

The disease course of CD is intermittent. Ongoing disease activity and recurring flares put patients at risk for structural bowel damage and a poor prognosis. Acute therapy tries to produce CD remission and block the inflammatory process, while maintenance therapy aims to keep the illness in remission for a long time and prevent flare-ups.

Mucosal healing is of growing importance as a clinical trial outcome and measure of treatment success. Novel treatment methods, strategies, and goals are being tested in order to enhance the long-term disease outcome.

OSP: Why is Crohn’s disease worthy of more, better treatment options?​

HK: Diagnosed prevalent cases will exceed 1.55 million in 2029 due to changes in disease risk and population demographics. As a result, new medicines, including numerous first-in-class therapies, will drive the CD therapy market forward. The necessity for alternate treatments for patients who have failed many biological medicines will promote their adoption. The emerging therapies together will be anticipated to generate major-market sales of close to $3.5b in 2029.

Conventional therapies for mild to moderate CD are largely genericized, limiting the potential for new therapies targeting this population. Numerous drugs targeting various immune and inflammatory mediators have failed to demonstrate efficacy in late-stage clinical trials.

For instance, a substantial percentage of moderate to severe patients treated with TNF-alpha inhibitors either do not respond or lose response over time. Therefore, additional treatment options for moderate to severe patients who have failed TNF-alpha inhibitors are required. More effective maintenance therapy for moderate-to-severe CD is needed because CD patients usually require long-term medications to maintain remission and prevent disease flare-ups.

Long-term efficacy without safety issues is widely wanted, as are the limitations of present treatments (such as TNF-alpha inhibitors, IL-12/IL-23 inhibitors, CAM, and JAK inhibitors, as well as corticosteroids). Novel CD oral medicines, new treatments for fistulizing illness, and agents with improved mucosal healing rates are all being studied as crucial factors in enhancing CD treatment.

OSP: How did Standigm come to target Crohn’s disease?​

HK: Initially, from collaborative research with SK Chemicals, Standigm identified and validated repurposed drug candidates for rheumatoid arthritis disease. Since these candidates are likely to have anti-inflammatory and antirheumatic properties, they were also expected to be effective for Crohn's disease. Standigm conducted basic validations to seek possibilities in Crohn’s disease and filed a patent to cover usage in this indication. Standigm is under discussion with SK Chemicals regarding further studies in Crohn’s.

OSP: Specifically, could you please share some detail about how AI can be put to use to advance care for Crohn’s?​

HK: Standigm Insight is a proprietary AI platform for drug repurposing and was used to find new treatments for Crohn's disease. Standigm Insight has been developed to predict new indications of existing drugs based on the idea that drugs can be classified from their gene expression perturbation to cellular phenotypes. Large-scale transcriptomic profiles of thousands of existing drugs were secured and utilized to train an indication prediction model. Existing drugs could be labeled as their current indications which were used as training examples.

Standigm Insight has learned hidden patterns between transcriptomic profiles and drug indications and then predicts unknown novel indications of existing drugs based on their perturbed gene expression profiles. Our other AI models such as the drug-target binding prediction model “STAR” and biological knowledge graph-based therapeutic path prediction model Standigm ASK can also be used in a combined and synergistic way.

OSP: Could you share any other noteworthy accomplishments Standigm has made with its AI-centric drug discovery—are there any other hopeful candidates your company and tech have uncovered?​

HK: Standigm’s noteworthy pipelines are AI-driven drug candidates which respectively target mitochondrial diseases, Parkinson’s disease, NASH, and cancer. Standigm is going to submit investigational new drugs (IND) for these drug candidates in 2023. Moreover, there are pipelines under license out (L/O) discussion with Stadigm’s customers, including AI-driven NASH drug candidates.

Standigm believes the role of AI in those efforts includes saving time and lowering the cost of discovery. With Standigm’s AI technology, the drug discovery process can be finished three times faster and ten times cheaper. So far, Standigm has run 52 AI-based drug discovery projects, targeting 9 indications. But above everything else, Standigm’s AI-based drug discovery can secure backup scaffolds that can enhance the project's success rate.