European Pharmaceutical Manufacturer sat down with John McDermott, VP scientific consulting at Quotient Sciences, to explore how Translational Pharmaceutics is reshaping the journey from preclinical development to first-in-human studies.
Quotient Sciences
How does Quotient Sciences’ integrated approach to drug development, Translational Pharmaceutics, expedite the transition from preclinical development to clinical testing?
One of the most important things today is there are an ever-increasing number of molecules entering drug development and making that transition from preclinical to clinical development.
We're chasing more challenging targets, and ones that are increasingly difficult to drug. Building knowledge about the molecule, including a biopharmaceutical risk assessment, is incredibly important to help new molecules achieve their potential. This is important right from the outset to ensure the very first experiments are successful.
Having specialisms in biopharmaceutics and formulation are key to this process. Individuals who can understand how to formulate a molecule for preclinical toxicology studies, and then translate that information into dosage form design for first in human study make that process happen.
That first in human study is focussed on demonstrating safety, tolerability, pharmacokinetics. Increasingly we are also seeing customers seek pharmacodynamic data to get early signals of efficacy in early development – essentially looking to understand whether a drug is a “quick win” or “fast fail”. Why put precious dollars into a development program that's not likely to achieve a therapeutic effect? These are the types of conversations we have with our customers, and looking to best understand their goals is all part of how we design and deliver their programs from the outset.
Quotient Sciences’ approach is integrated. In our case, integration is bringing the CDMO service world close to the CRO service world. It is horizontal integration, providing breadth of service, being both a CDMO and CRO – as well as vertical integration bringing depth of service, and deep acumen in each of the areas we support within formulation, manufacturing, and clinical testing.
Through Translational Pharmaceutics, our integrated drug development platform, we bring formulation development, drug product manufacturing, and clinical testing into the same program of work, managed by a single project manager. That means we can manufacture batches of drug products in our facility, release them, and dose them in our clinics to trial volunteers in less than seven days.
Translational Pharmaceutics is proven to accelerate early development studies by taking scale up of drug product and long-term stability studies off the critical path for dosing. These two aspects alone can save four to six months of time getting to the first subject, first dose.
Another aspect is being able to save high-value or hard to produce API, giving customers another way to control their investment as they identify the formulation that's right for onward scale into phase two studies.
Ultimately, Translational Pharmaceutics also lets us, and our customers, be even more data-driven. Data from a dosing period can influence what we do next in a way that traditional development manufacturing processes just can't. That could also be done with many of the complex formulation technologies, such as modified release, that are necessary with the molecules that we're facing today.
You've recently announced some collaborations - one in the UK and one in the US. What do these collaborations bring to the company?
Collaborations are one avenue that we have used this year to extend our science and services. Earlier this year, we announced a joint venture with CPI, the UK Centre of Process Innovation, to advance the development of mRNA based therapeutics.
It is known that mRNA therapies have a complex supply chain, often requiring multiple providers for synthesis, LNP formation and clean up, and sterile fill-finish let alone the cold chain requirements for these drugs.
The mRNA-based vaccines created to address the COVID-19 pandemic are a great example of how these development programs can be accelerated. However, those were unprecedented times, and normal operating modes have now for the most part returned.
The lessons-learned and pace of COVID-19 vaccine development we hope can be applied to regular therapeutics, and we know in our case Translational Pharmaceutics has shown that kind of speed. For almost two decades, we’ve consistently shown how we can streamline the supply chain, taking a molecule from API to human-ready dosage form and through Phase I trials in accelerated time by eliminating unnecessary steps that create distance between the drug product manufacturer and the clinic. We're very excited to be working in this space to bring our Translational Pharmaceutics platform into mRNA manufacturing.
The second collaboration is a partnership with Biorasi, a US-based CRO that is focused on global patient trials. We have made this move as a response to increasing industry trends that we've observed in terms of getting efficacy data in the first in human study. Yes, we're a contract manufacturer, but we're also a contract research organisation, and we’ve built a body of work over the last 20 years running first in human studies and accelerating those programs demonstrating safety, tolerability, and pharmacokinetics in healthy volunteers.
The key question is, what about efficacy? Is the drug actually working? The trend that's starting to emerge, again, is back to having a quick win or fast fail.
By bringing patients into a first in-human study, we get to evaluate that drug during its first year of clinical development to assess whether it actually has an impact, and if not, provide our clients with the data so they can confidently choose to cease investment or pursue other paths.
You said you can do a trial with a few people. Does that still give just as accurate a result?
It depends very much upon the therapeutic area, but we are focussed on incorporating exploratory assessments in small numbers of patients
To give a past example, we've conducted first in-human studies in respiratory disease where we have incorporated some trial participants who have asthma. Forced expiratory volume is one variable we can measure in these types of studies. If there is a measured improvement relative to placebo, we can use that as an earlier indicator of efficacy. There's a signal there that allows you to follow it through for further clinical testing.
Another example: a significant proportion of our work in recent years has been in GLP-1s. We've manufactured the drug product, including for injectable systems, and designed our trials to bring in volunteers with increased BMI. Through a clinical program and treatment over the course of 8 to 12 weeks, we have been able to demonstrate reductions in body mass index.
Again, both of these are aspects of involving patient cohorts in first-in-human studies, showing positive clinical outcomes, and ultimately justifying future investment.
With the GLP-1s, has the demand affected the company?
There's a lot of interest and competition in the space. The end goals for every customer have been safety, tolerability, and efficacy. Ultimately, does the drug work at the tested dosage? Can side effects be reduced or eliminated?
I first started working on GLP-1 molecules in the early 2010s, with several oral medications in this space. Because most of our work has been in early development, we have seen some of these popular drugs at early stages, including some that have not yet made it to market. Recent improvements in drug delivery technology and peptide engineering have made oral peptide delivery even more feasible. OSD (oral solid dose) forms are also a better route for patients that may fear needles, or just find a pill or tablet easier to take every day.
A proven method for oral peptides is coformulation with a permeation enhancer. We’ve done formulation work in this space, as well as provided PBPK (physiologically based pharmacokinetic) modelling and simulation support to our Translational Pharmaceutics programs.
Oral delivery of peptides poses absorption challenges, typically addressed by adding functional excipients like salcaprozate sodium (SNAC). Determining optimal permeation enhancer levels requires clinical studies, though. Preclinical models are unreliable for human bioavailability predictions, and varying concentrations in the GI tract add complexity.
In one project, done with a leading pharma client, we redeveloped a peptide from SC (subcutaneous) injection to a tablet to improve patient access and compliance. A custom PBPK model describing both the peptide and permeation enhancer was created. This enabled us to predict new formulations, then refine using clinical testing and resulting clinical data. This integrated approach improved understanding of the role of the permeation enhancer and has received positive client feedback encouraging further exploration.
You have clinics in both the US and the UK? What would you say is the importance of having numerous clinics, and do you have plans to expand that further?
Quotient Sciences is a global company, and yes, our footprint is in both the UK and US. Today, we've got clinics in Nottingham, UK and in Miami, Florida.
Where our clinics are located are strategic for our business. The origins of our business is in the UK, in Nottingham, but we expanded into the US in 2017 to create a global presence. Before we acquired facilities in Miami and Philadelphia, we had a significant number of customers based in the US. We often hear: ‘When is Quotient Sciences coming to the US? How can we do Translational Pharmaceutics in the US?’
The other advantage is that there are different regulatory environments in each country. That does sometimes allow us to make recommendations or give advice to our customers as to how to optimise the delivery of their study too.
We will continue to invest in and run clinical studies in both the UK and US, and are also expanding our capabilities through partnerships as discussed, such as the one recently announced with Biorasi, to reach more patients.
What are the differences between the regulations in the UK and the US?
The structure and the underpinning early development regulatory framework that exists in the US is different to Europe and the UK, each with strategic benefits.
It’s commonly reported that the US has faster timelines, which is true in part but the UK is catching up, and a UK Clinical Trial Application can be achieved without an IND opening meeting for example.
There's been quite a lot of press about the performance of MHRA and concerns about the UK’s competitiveness for clinical research. As a company that offers both CDMO and CRO services, we have worked with the MHRA for the last several years and have seen their improvements first-hand. We are also actively collaborating with other organisations, including the UK Department of Health & Social Care and the Health Research Authority, who are responsive and committed to improving clinical trial timelines and scientific advice.
Just this month, the MHRA reported the total review time for initial clinical trial submissions is only 41 days. 99% of applications are being reviewed within statutory timelines. In our case, our clinical trial approval times have improved significantly over the past six months, as well.
Moving on to AI, what is the importance of it in the industry?
Artificial intelligence is something that can be quite polarising: you can be scared to death of it, be sceptical of it, or embrace it, but in any case it's here to stay. It also is generational: my mother thinks AI is the equivalent of having the robots taking over, while my daughter thinks it's the best thing she's ever seen. I'm a scientist, so I go in with a critical brain when I start to look at platforms like ChatGPT, Copilot, and similar tools. Essentially, I really want to try and understand how it works.
There are day-to-day business benefits we all know, like writing emails and general improvements for productivity, but in the space of pharmaceutical development, applications may range—molecule discovery, speeding up data analysis, improving capabilities of modelling and simulation software (many software platforms are now adding AI-based features, M&S software is no exception), improving formulations, improving manufacturing processes to reduce waste or gain efficiency (e.g. reduce batch failures, automate processes), and so-forth.
We work routinely with companies that are using AI in drug discovery to try and identify new molecules and the use case is very clear here. Also, I recently had the benefit of recruiting someone from an AI-driven company for a role on our PBPK modelling team. They were using AI to help identify new targets for the treatment of certain rare and orphan diseases. Fabulous work.
In the formulation development space, Quotient Sciences have recently engaged in a program to develop formulations using machine learning. This is not seeking to take away the role of the formulator, but helping the formulator reduce their bias in the compositions that they're selecting, as well as add meaningfully to the process—new ideas, creativity, speed...
To give an example, there's a feasibility study in a program we are working on at the moment where we used an AI-based, machine learning tool to develop a range of formulations for a clinical investigation. That direction got us from a standing start in the laboratory, through to formulations that were correct and ready for evaluation in the clinic, about 60% of the time in the tests that we ran. This meaningfully reduced the number of batches and the time that it would typically take to achieve the same goal.
Showing the time and cost savings in this way can be impactful for our customers. It might not be the right application for every single program, but early wins and failures like this allow us to iterate and make it better over time.
AI is here to stay. We have to embrace it, and learn how to use it. In the formulation and manufacturing space, prompting and testing is going to be key. There are likely gains to be made in the development of IP about how you design prompts and test models, as well.
Are there any things that scientists need to be careful of with AI?
That's what I mean by the prompting. If you don't prompt it correctly, it has the potential to misdirect development activities. At the same time, it’s important to not let human bias take over. Just because you wouldn't choose something as a human doesn't mean it's the wrong thing to try. You will always need the human in loop to make those decisions and ultimately assess those data, but it is important when you are looking at these things and trying to embrace these things, you need to be aware of the potential for human bias in directing these experiments and ultimately give it a fair chance.