With conventional drug discovery efforts failing to deliver on some critical medical needs, such as non-opioid treatments for chronic pain, there may be value in exploring plants associated with traditional medicines to discover new drug molecules and mixtures. Gb Sciences is connecting plant-based medicine with modern in silico drug discovery tools, leveraging a unique AI-enabled platform to mine plant species to uncover minimum essential mixtures of molecules with therapeutic benefits and reduced side effect profiles. The company’s President, Andrea Small-Howard, Ph.D., sat down to discuss Gb’s evolution from a cannabis company to a plant-inspired biopharmaceutical company and what makes their approach truly unique, with Pharma’s Almanac’s David Alvaro, Ph.D.

David Alvaro (DA): Can you start off with an overview of the history of Gb Sciences and how the mission has evolved?

Andrea Small-Howard (ASH): We originally started as a company that was producing cannabis-based products for the medicinal cannabis market. Within the medical cannabis world, we sought to apply scientific principles to learn as much as we could about how to unlock the therapeutic potential of the cannabis plant and how to standardize these products. Our goal was to use in silico processes and cell and animal models to characterize the therapeutic potential of different cannabis strains (or varieties).

We used a range of different bioassays, as well as patient-reported outcome data to describe the therapeutic potential of these strains. As a result, we were able to identify cannabis varieties that seemed to have therapeutic potential for specific diseases. We worked with a very well-respected analytical lab, not only to analyze which compounds were in each variety, but also to help determine whether the best way to use cannabis as medicine was using the whole plant, a whole-plant extract, or even a single ingredient.

Each time we studied a strain with therapeutic potential for a specific disease process and applied a reductionist approach, we found that mixtures outperformed individual ingredients. In every strain that we explored in depth, there were compounds that were therapeutically beneficial, others that were therapeutically neutral, and others that actually reduced the net effectiveness of the mixture.

We realized that whole plant extract was probably not the best approach for a product held up to pharma standards, but, on the other hand, a single-ingredient drug would by necessity lose these synergies in the mixtures that we demonstrated at the cell and animal levels. A mixture represented a happy medium. Also, we realized that we could best standardize a therapeutic product by further reducing these mixtures to what we call a minimum essential mixture (MEM).

Over the years, we have filed quite a few patents in which we have chosen a clinical indication and then we demonstrated molecular synergies in a particular cell assay for a specific mixture of individual plant-inspired ingredients. These synergies go beyond the legal burden of obviousness — a scenario in which combining A and B gives you the benefits of both A and B — into effects that are akin to biological magic. We have had six patents issued to date and three more that are likely to be issued soon.

In pursuing that MEM strategy, we ended up leaving the cannabis plant behind, and since then we have sold all of our cannabis-touching assets and doubled down on our biopharmaceutical pipeline. From a cannabis-producing company focused on patients, we transitioned into a company developing prescription pharmaceutical drugs.

Along the way, we were also training an AI-enabled drug discovery tool. In the early days, as we were feeding and training the tool, it was sort of lagging, but it has since bloomed into a full-blown force, and today our discovery work is no longer really about cannabis per se; it’s about traditional medicines in general.

To the best of our knowledge, the database for our AI-enabled drug discovery engine — Phytomedical Analytics for Research Optimization at Scale platform, or PhAROS™ — represents the first time anyone has integrated multiple major traditional medical systems into a single database and figured out ways to search across the different verticals. The ‘big idea’ behind PhAROS as a predictive tool is our in silico convergence analysis capabilities, meaning that, in cross-referencing traditional systems that evolved over thousands of years in geographically distinct areas, any commonality of ingredients or mixtures of ingredients for a particular indication likely isn’t the result of chance.

DA: How did these two approaches converge on candidate therapeutics for chronic pain?

ASH: In our cannabis work, we started with an individual strain for which we had patient-reported outcome data suggesting that it was helping people with chronic pain. We did a very traditional biochemical search, starting with cell-based assay screens of whole plant extract, after which we divided the extract into a terpene-containing fraction and a cannabinoid-containing fraction.

We were very surprised that the terpene-containing fraction had more bioactivity than the cannabinoid-containing fraction. We then took a deeper dive into the components of each fraction and their ratios. As a result, we developed chronic pain formulations that are largely terpene-based. We’ve demonstrated at the cell and animal levels that these pure terpene mixtures exhibit synergies. Additionally, we have some MEMs where we added a single cannabinoid to a subset of the terpenes, and we got some positive preliminary preclinical results.

We’ve also explored chronic pain using our PhAROS program. We loaded our database with many diverse traditional medical systems, representing five of the seven continents so far, and it was fortuitous that pain is easy to search for, because every culture has a word for pain. Our in silico convergence analysis for pain found 121 compounds that were conserved among the traditional medical systems, and we categorized them by the type of molecule and the kind of plants.

Interestingly, terpenes emerged as major drivers for pain, confirming our cannabis work but on a global scale. We found four other categories of compounds that are also therapeutically important in pain treatments, so we’ve been using analytics, as well as wet lab experiments, to uncover the best mixtures that demonstrate synergistic effects. In addition to the chronic pain formulation based on cannabis that’s further developed in our pipeline, we are developing another formulation derived from a more broad-based traditional medicine approach with no cannabinoids.

DA: Have terpenes and cannabinoids proven challenging to formulate into drug products?

ASH: Speaking of terpenes, one of the tricky things that came out of the cannabis work for pain was that terpenes are very volatile, and hence really incredibly hard to formulate, because they float away on you. We’ve been working with a team of scientists from the University of Seville to make oral nanoparticles that stabilize the terpene molecules. We’ve explored using PLGA (poly(lactic-co-glycolic acid)) nanoparticles that are PEGylated such that, if you ingest it orally, it can provide time-released dosing. In our preliminary proof-of-concept work on these nanocarriers, we found that a single oral dose provided 11 days of consistent relief from pain in a mouse model, compared with peak activity between three and nine hours and no relief after 24 hours without the carrier. This relatively new category of therapeutic compounds — terpenes — requires a novel approach for therapeutic delivery simply to cage it, but it’s a great benefit that the nanocarriers also provide a sustained-release mechanism.

Controlled release is really a big deal for pain therapies. If you speak with people who suffer from chronic pain, one of the things that they always will tell you is that tolerance is a huge problem for them with the current standard of care, which is opioids. As a result, they end up needing to take more and more of the same drugs to achieve the same effect. Our candidate is not only not an opioid; it has kinetics that are well-suited for people with chronic pain.

DA: For both the terpene-based formulation and the one containing the cannabinoid, is the mechanism of action on pain pathways well understood?

ASH: Absolutely. We’ve published a bunch of papers demonstrating that terpenes interact with the TRP (transient receptor potential) channels located in peripheral neural bundles. Myrcene and other molecules in our formulations are strong ligands activating TRPV1, which is a pretty well-characterized drug target. In addition, the compounds in our MEM target other functional TRP receptors in peripheral sensory neurons. Nociception involves both the perception of pain, which is central, and the sensing of pain, which is peripheral. We designed these products to act at the pain-sensing level, but there’s also evidence of these compounds acting on CB1 and CB2 in the brain, so we can’t rule out the possibility that our MEM is targeting both aspects of pain.

We have demonstrated that the terpenes and the cannabinoids are able to desensitize the TRP channels — via ligand-based desensitization. We’ve also done molecular docking studies demonstrating the putative binding pockets for both the cannabinoids and for terpenes, although not all cannabinoids and terpenes dock. There are five terpenes with somewhat similar shapes and sizes that seem to be able to hit that pocket in TRPV1. The terpene-binding site overlaps with the binding site for CBD in TRPV1, which provides for potential allosteric interactions in the presence of both ligands. Notably, the cannabis-derived ligands are effective enough to trigger the TRPV1 receptor but without flipping the TRPV1 receptor into a secondary pore-dilation state that causes cell death, which has been an issue for capsaicin-based therapies that target TRPV1.

DA: I think it’s no secret that there has been a serious need for non-opioid treatments for chronic pain for some time. What can you tell me about some of the other approaches in development today and how you see your drug differentiated from them?

ASH: As I mentioned earlier, there is a capsaicin-based pain therapy targeting TRPV1 in the works, which has this issue with triggering neuronal cell death as a part of the mechanism of action. There are a handful of other companies exploring cannabinoids for pain, but mostly single molecules, typically synthetic versions of THC alone or CBD alone, or THC and CBD combined. Another company is exploring combinations of synthetic THC with already approved analgesics, mostly opioids.

We think it’s significant that our formulations don’t include THC. THC comes with a lot of baggage — not only issues of legality, but the simple fact that a pain patient may not want to experience the other psychoactive effects of THC. We found that it ultimately was not necessary for the therapeutic effect. We’re leaning heavily into the synergistic terpenes, which is not common, especially our leading candidate formulation that doesn’t include any cannabinoids at all.

Ultimately, most of the drug development work for pain still revolves around opioids — many companies are still doing third- and fourth-generation opioid-based work. But while there is a place for that work, this opioid crisis will never be resolved until there’s something else that doctors and patients can responsibly use. In the wake of the well-publicized opioid crisis and enhanced restrictions on opioid prescriptions, the medical community found that lots of people were still in pain. The solution to the opioid crisis being “don’t prescribe opioids as much” hasn’t really worked for patients.

We believe that the key will be developing novel nonopioid mechanisms to break that cycle. We are hopeful — based on the data we have accumulated to date — that the therapeutic compounds that we’re using should be really beneficial for patients, combining novel ingredients with a novel delivery system to provide sustained release and avoid tolerance issues.

DA: On the macro level, are there any particular reasons why it’s taken the industry so long to move beyond opioids?

ASH: Beyond the mechanistic issues, there’s always the business aspect of a society’s adoption of new things. When different U.S. states started to open up medical cannabis programs, sales of opioid drugs dropped dramatically. Many pharmaceutical companies are making a lot of money selling opioid drugs, so they weren’t exactly throwing their weight behind alternatives like cannabis.

I’d like to believe that this has begun to change. Big pharma is now definitely keeping tabs on work like what we’re doing, but we’re still far from the inflection point where they will really want to invest. Ultimately, I think that everybody wants new and better pain medicine, and the side effect profile of cannabis — where you aren’t likely to overdose — is very attractive. We’re steadily trying to get these novel formulations into the clinic, and I think that when we are able to present human data demonstrating that a single oral dose provides a week of pain relief, it’s going to be hard to shut that program down.

DA: Where are you currently in your preclinical evaluation? Do you yet have a sense of a projected timeline for clinical development?

ASH: We’re in animal trials right now with the National Research Council in Nova Scotia, Canada, testing the individual compounds and the mixtures on their own and in nanoparticles. Early results are very positive. We’ve also spent a lot of time on the chemistry side, working to optimize the shelf-life of these nanoparticles and refine our manufacturing controls. My prediction is that, by the end of this year or early next year, we will have solid animal data on these time-released oral nanoparticles, which will give us a better sense of the larger timeline. At the same time, we’ve been exploring doing preliminary efficacy studies in animals and continuing to perfect the formulations.

We have focused our discussion on pain today, but we are also on our way to the clinic with our lead formulations for the treatment of Parkinson’s disease, which are mixtures of three cannabinoids formulated as oral dissolving tablets using an oral dissolving tablet matrix owned by Catalent Pharma. We are working hard to get these formulations into the clinic for Parkinson’s patients by the end of 2023, and to follow with the pain therapeutics about a year behind that.

On top of that, we have been exploring a kava-inspired product for anxiety. Kava is a plant that grows in the South Pacific region, and people there have been grinding the root and extracting it with water, like a cold-brew tea, for thousands of years. Any time there was a dispute among tribes, they would drink kava together. After they drank the kava, if they still wanted to war, they would war, but most often when you drink kava, you don’t want a war. It provides a blissful experience that seems to be a potent anti-anxiety treatment without psychoactive side effects.

Other groups tried to commercialize kava for anxiety in the past. It is clear that the efficacy of kava extracts cannot be reduced to a single ingredient, but they also found that one of the compounds in the plant extract may cause liver toxicity. We used PhAROS to substitute in an analogous molecule from a different plant in the same family, and we found a couple of different natural “biosimilars,” which we are testing. I recently received new data in an animal model of anxiety demonstrating that some of our MEM reduced anxiety with statistical significance. And this data set also demonstrated significant synergy among the ingredients; if you look at the activity of the individual ingredients, it’s tiny compared with the anti-anxiety activity of mixtures in these preclinical studies.

DA: For these compounds, would you pursue biomanufacturing, or growing plants and performing extractions?

ASH: We are taking the biomanufacturing route. We no longer need plants because we’re working with Purisys (Purisys, Inc. Athens, Georgia), and they are providing us with all our cannabis-inspired ingredients. While cannabis is a pretty hearty crop, the earlier attempts to commercialize kava ran into supply chain issues, because it only grows well in a particular niche, and quality control issues from suppliers compromised their commercialization efforts. That experience demonstrated that getting away from the actual plant may help with the success of the launch of these kinds of products. Purisys, and other similar companies, can synthesize many of the plant-inspired ingredients we need at scale. Any time you’re dealing with plant-derived products, you’ve got to deal with a lot of regulatory paperwork about sourcing for the plants and such, whereas, for the synthetically produced cannabinoid homologs, although cGMP manufacturing has its own paperwork, it’s the kind of paperwork that regulatory agencies are more used to.

DA: This is all very exciting work on its own merits, but it strikes me as potentially the proof of concept of a whole new drug discovery paradigm. Are there other indications at the top of your list or other promising plant species that you’re thinking about exploring in the future?

ASH: Looking at our whole pipeline, Parkinson’s will be our first to the clinic. Pain is in a sweet spot, there’s a real need for new therapies, and I think that our approach is very different than what other people are doing. I’m very excited about the kava-inspired anxiety drugs.

We also have an anti-inflammatory program that was born out of COVID. We originally started working with Dr. Norbert Kaminski at Michigan State University, who was following a cohort of HIV and AIDS patients that were developing HIV-associated neurocognitive disorder (HAND).

We used PhAROS to determine if there were compounds within the cannabis plant that would help to downregulate key inflammatory cell processes and cytokines in the pathway that cause hyperinflammatory reactions to viruses, and we found that THC was the strongest single ingredient for decreasing inflammatory responses. But the minor cannabinoids and terpenes were more interesting, because their “selectivity” to specific aspects of the inflammatory cascade could be used to design more targeted anti-inflammatory therapies, almost like immune checkpoint inhibitors. Then, when the pandemic hit, we began trying to develop mixtures that would specifically downregulate the cytokines that cause the most damage in hospitalized COVID patients. Because no one should be medically immune compromised during a pandemic, our COVID-related anti-inflammatory mixtures were designed to downregulate these cytokines doing the most damage, but without impacting plasmacytoid dendritic cells and other immune functions that are important for mounting an antiviral response. Based on in silico and cell data, we designed 24 different combinations that were able to suppress the immune system. Eight of them performed in the selective way that we hoped in preclinical trials. We’re really excited about this program as well, even though we still have a lot of research to do — but the idea of using these molecules like checkpoint inhibitors is novel and powerful. Because our therapy doesn’t target the virus, it targets the hyperinflammatory side effects secondary to a viral infection, this therapy is potentially useful for future viral outbreaks as well.

A little bit further behind that, we have a heart program based on some elegant work from our collaborators at the University of Hawaii and Makai Biotechnology, who demonstrated a reduction of cardiac hypertrophy in a rodent model of heart disease using our favorite channel, TRPV1.

And yes, it just keeps going. We’re interested in depression. We’ve also been increasingly moving beyond cannabis to embrace other plant-inspired therapies, and we are undergoing some appropriate rebranding.

DA: Obviously, bringing your pain therapeutic to the market will have a huge impact on so many chronic pain patients. But what do you anticipate as the bigger picture impacts of truly proving the validity of this approach to drug discovery?

ASH: I want to shift the paradigm. Traditional pharmaceutical development has been driven for a long time by the concept of a magic bullet — the idea that any disease state can be reduced to a single receptor-based reaction. I think that that is a flawed concept. On the other hand, simple models were the only option back when this paradigm was constructed, because we didn’t have the kind of computing power that we do today, so you had to be able to reduce the complexity in your experimental design to actually be able to interpret the results.

Well, the world has changed. The information age is here. Like other industries that have been revolutionized by the application of big tech, this is an exciting time for a lot of scientists. The availability of this computing power should enable work in more complex, biological models of human diseases. We should also be honest with ourselves that even single-ingredient drugs don’t have a single target in the body, or there wouldn’t be side effects.

What Gb Sciences is trying to do is create and study complex models of human diseases, which enables us to predict the effects of multiple ingredients that target multiple aspects of each of these diseases. What if you could pre-factor in the side effect profile to make them beneficial — make side effect profiles work for you — and really think about how these drugs interact in your initial conception of the product, as opposed to later, when it’s on the market. Our mission and our challenge for the world is to create a new kind of medicine, which can really address the complexity of disease.