Meliora Therapeutics is building the full stack solution for MOA identification and polypharmacology
Molecular perturbations and phenotypic readouts for scalable MOA identification
Meliora Therapeutics recently announced its impressive $11M seed round led by HOF Capital and ZhenFund (press release). We are lucky to have participated in the financing and are eager to continue supporting their endeavors. Herein, we will discuss the problem being addressed, the solution being developed, the competitive differentiation offered, and a futuristic application enabled by Meliora Therapeutics.
Check out the Wall Street Journal exclusive on Meliora’s seed financing!
Highlights & Takeaways
Successful drugs rely on robust mechanism-of-actions (MOAs) to ameliorate disease, yet there are no methods to rigorously characterize MOAs at scale. Furthermore, disease pathology is often complex and may require multiple targets to be hit simultaneously for a meaningful clinical response.
Meliora Therapeutics is building a platform technology to infer relationships between a given pharmacologic intervention and perturbed phenotypic readouts, which ultimately allows the deconvolution of MOAs at scale. It also offers the ability to rapidly evaluate a given drug’s capacity for polypharmacology. This computational genomics drug discovery engine enables Meliora to identify the true target(s) of given compounds and use these as scaffolds for efficient engineering of potent drug candidates.
A competitive differentiation that Meliora has comes from understanding the phenotypic effects of many compounds and creating the corresponding molecular maps to enrich further drug discovery efforts. In contrast, other drug discovery startups leveraging AI are generally developing methods for target identification or molecular modeling, not in understanding the relationships between compound chemistry and phenotypic effect.
A futuristic application enabled by their technology is generative drug design, which is essentially the inverse of MOA identification. Rather than taking extant compounds and figuring out the target effect, a scientist could request new compounds for a desired target effect.
Source: Meliora Therapeutics
Problem being addressed by Meliora Therapeutics
As biomedical research advances, we have gained significant knowledge about the molecular drivers of tumor development (ref; ref). This has encouraged the exploration of targeted molecules to treat cancer, yielding medicines like ibrutinib, dasatinib, venetoclax, trastuzumab, nivolumab, and ipilimumab. Understanding the mechanism-of-action (MOA) of drug candidates has been an enduring goal in cancer drug discovery (ref). However, the translation of these drug candidates has been severely limited: approximately 97% of clinical-stage cancer drugs fail to receive regulatory approval (ref). Moreover, the pharmaceutical industry is hampered by the significant costs of discovering and developing drugs (ref; ref; ref; Ergo Bio Insights 2021).
Furthermore, discovering drugs with potency against multiple targets simultaneously is very difficult despite the highly attractive proposition of synergistic effects derived from the success of combination therapies (ref; ref). This is also known as polypharmacology, in which many diseases are likely polygenic in nature and thus require therapeutics that hit multiple targets simultaneously. Multi-target drugs tend to be serendipitously found, and are oftentimes determined late into clinical development or even after commercial launch. A rigorous approach for systematically discovering polypharmacological drugs is desperately needed.
Solution being developed by Meliora Therapeutics
Meliora is creating a platform technology to deconvolute therapeutic MOA at scale. To improve the probability of success and reduce the costs of translation (by selectively advancing better drug candidates), validating therapeutic MOA is crucial. It is quite common that drug candidates are mischaracterized and thus suboptimal (ref; ref). For example, OTS964 is a clinical-stage drug candidate which putatively inhibits PBK, but the team revealed that it instead potently blocks CDK11. Likewise, the team has identified several other drug candidates that were mischaracterized by leveraging high-throughput perturbation techniques (e.g. CRISPR, RNAi) and biologically-relevant phenotypic screening. At a high-level, their approach is to “pattern match” the phenotypic readout of a given compound to those of well-characterized drugs or molecular perturbations. Importantly, the team has a deep understanding of the technical strengths and limitations of these perturbation techniques (ref), as well as how genetic alterations correspond with clinical outcomes (ref).
The team has already identified 22 hits (of 25 experimentally tested) that have potency against the in silico target. These serve as scaffolds for engineering best-in-class drug candidates, or even drug candidates for novel target(s). Multiple compounds were already tested in early clinical trials and have clean safety profiles. These represent de-risked opportunities to create new therapeutic programs and advance drug candidates into the clinic. With further resources, the team can quickly assemble a therapeutic pipeline for various malignancies. Furthermore, some scaffolds have the potential for designing polypharmacology, i.e. molecules which can hit multiple targets for synergistic effects (ref).
Competitive differentiation offered by Meliora Therapeutics
An important distinction and competitive edge that Meliora holds is a platform technology that focuses on mapping phenotypic effect to drug candidates, i.e. characterization of the therapeutic MOA. Competitors tend to focus on mapping phenotypic effect to genetic modulation, which falls short of linking to a given pharmacological intervention. Importantly, the AI-driven direct link of phenotypic effect to pharmacological intervention is what gives Meliora confidence that it can discover and develop drugs with higher probability of success. The defensibility of their approach is rooted in proprietary machine learning algorithms involving multi-modal molecular fingerprints to generate n-dimensional MOA maps.
Futuristic application enabled by Meliora Therapeutics
Meliora Therapeutics is building the full stack solution for MOA identification. They can leverage this for discovering promising chemical scaffolds and creating new drugs with potency against multiple targets simultaneously. Now consider the inverse of MOA identification: instead of characterizing target effects for a given drug, can we de novo generate drugs with pre-specified MOAs? With enough phenotypic data collected through an automated laboratory, this should be within the realm of possibilities! In the future, Meliora could have an incredible engine for generative drug design through amassing perturbed phenotypic readouts, thereby revolutionizing drug discovery.
Author information
Ergo Bio closely follows innovation in the biotechnology space and evaluates interesting drugs and deals. It is run by Vandon T Duong (LinkedIn), feel free to connect! I am a biotech enthusiast and a molecular engineer by training. I am also an avid consumer of news and research around precision medicine.
Apply to join our investment syndicate and help support incredible founders and their startups! Read our related notes about risks associated with drug discovery startups, community building in biotech venture, biotech venture models, and due diligence in biotech investments. We will make a societal difference by syndicating high-conviction investments in impactful startups and supporting their incredible founders!
Ergo Bio pages
Disclaimer
This article serves informational purposes only and should be treated strictly as educational material, not as investment recommendation or legal advice. The information presented may be inaccurate or out-of-date. The contributing authors and editors disclaim liability for any errors or omissions. Any opinions expressed may change without notice. Ergo Bio LLC reserves all rights to the content generated through this resource hub (Ergo Bio Insights).