Feliza Mirasol is the science editor for 

Feliza Mirasol

Sanjay Konagurthu AAPS Interview Part Two

, held from Nov. 9–12 in San Antonio, Texas, 

, PhD, senior director,Science and Innovation, Thermo Fisher Scientific, about the presentation “

AI/ML Model Informed Early-Stage Drug Development.

 Dr. Konagurthu talks about integrating a systematic 

 framework to accelerate decision making regarding 

He says this framework provides insights into a final dosage form, as drug loading can impact both physical and chemical stability—and how much drug substance can be contained in a 

“Drug loading is an important aspect when it comes to physical and chemical stability,” Dr. Konagurthu says. “With some of these modeling tools we have deployed and implemented, we can predict not only the favorable excipients and polymers to screen prior to any experimentation, we can also predict how much drug loading one can achieve.”

Dr. Konagurthu also addresses how molecular dynamics simulations and quantum mechanical calculations inform complex 

“If the drug is bound too tight, it won't release,” he says. “If it's too loose, it doesn't really stabilize in the amorphous form. So, these are the series of calculations and various descriptors we assign in addition to the physical chemical properties ... to be able to do these calculations and narrow down the formulation design space. And all of this is done 

 of Dr. Konagurthu’s interview with PharmTech Group, in which he discusses how artificial intelligence and machine learning can help solve the dilemma of poor solubility.

 for all of our AAPS PharmSci 360 coverage.

Editor's note: This transcript is a lightly edited rendering of the original audio/video content. It may contain errors, informal language, or omissions as spoken in the original recording.

Hello, I'm Sanjay Konagurthu. I'm Senior Director of Science and Innovation at Thermo Fisher.

When we do things like quantum calculations, we look at the confirmer space. So what that does is give us insight into how a molecule folds upon itself. That gives us an idea about the accessible surface area of the molecule and the excipient polymers to interact. So that's one piece of it.

And we look at the various descriptors, like the classical hydrogen bond donors, acceptors, aromatic, hydrophobic. We look at potential energy surface charge models. These are like the sticky points of a molecule that essentially give us insight into how a molecule and an excipient or a polymer are interacting at a molecular level.

And then we look at molecular dynamic. Simulations, looking at interactions between the drug and the polymer. If the drug is bound too tight, it won't release. If it's too loose, it doesn't really stabilize in the amorphous form. So these are the series of calculations and various descriptors we assign in addition to the physical chemical properties, like I mentioned, to be able to do these calculations and narrow down the formulation design space.

. And so we are saving an incredible amount of trial-and-error and unnecessary experimentation and getting to the goal quicker.

Drug loading is an important aspect when it comes to physical and chemical stability. With some of these modeling tools we have deployed and implemented, we can predict not only the favorable excipients and polymers to screen prior to any experimentation, we can also predict how much drug loading one can achieve.

So what that does is, our priority is, you get an insight into what your final dosage form is going to look like because the drug loading can impact both physical and chemical stability, and how much drug loading you have can impact how much drug you can get in a tablet or a capsule. Those are the interplays that we can get insight into based on these AI/ML modeling tools that we have implemented at Thermo Fisher.

Videos

The conclusion of this interview with Sanjay Konagurthu focuses on simulations and calculations that go beyond simple screening.

Molecular Dynamics and Quantum Calculations: AAPS Thermo Fisher Interview, Part Two

Sanjay Konagurthu AAPS Interview, Part One

, PhD, senior director, Science and Innovation, Thermo Fisher Scientific, about the conference presentation “

In the above interview—the first in a two-part installment—Dr. Konagurthu discusses the potential of 

 using artificial intelligence (AI) and machine learning (ML).

“The big advantage is time and cost savings and resource savings for the industry and our customers,” Konagurthu says. “Time is money, literally speaking, and what we bring to the table is reducing risk in product development, saving time and resources, minimizing API consumption. It's an ultra-material-sparing approach by leveraging some of these 

 modeling tools like AI, leveraging AI/ML for supporting product development and formulation development.”

How can AI/ML improve early-stage drug development?

In particular, Konagurthu says, AI/ML can help sift through a field of drug candidates in which the percentage that are 

 has increased in the past several decades from 30% to as high as 90%.

“It's a challenging environment and the probability of success of getting a poorly soluble NCE [new chemical entity] to a commercial product is pretty low,” he says. “It's way less than 10%, probably more like 5%. So, by leveraging some of these AI/ML tools that we've developed, we aim and strive to reduce the risk in product development.”

Check back for Part Two of PharmTech Group’s interview with Konagurthu, in which he drills down further on molecular dynamics simulations and quantum mechanical calculations as tools to be deployed beyond simple screening.

Hello, I'm Sanjay Konagurthu. I'm Senior Director of Science and Innovation at Thermo Fisher. I've been in the industry going on now 27 years.

Background, I'm a chemical engineer by education, PhD. My experience, broadly speaking, is in small-molecule oral drug delivery areas like solubility, bioenhancements, spray drying, hot-melt extrusion, nanotechnology, as well as modified control release, complex dosage forms, etc., ranging from early development through commercialization.

I also lead a predictive modeling and simulation team at Thermo Fisher. We are a global team. We deal with the predictive modeling and simulations for drug product development.

That's a great question. I like to tell people when I first started working in this industry over 20-plus years ago, the number of candidates that were poorly soluble were estimated to be 30% or so. Now, depending on the source that people quote, 70% to 90% of molecules are poorly soluble, so it's more than doubled, if not tripled.

It's a challenging environment and the probability of success of getting a poorly soluble NCE to a commercial product is pretty low. It's way less than 10%, probably more like 5%. So by leveraging some of these AI/ML tools that we've developed, we aim and strive to reduce the risk in product development.

We can get to accelerating the early phase especially by leveraging these 

 AI/ML tools for selection of technologies. What's the best technology that might work for any given molecule? And it's all dependent on the type of molecule. Every molecule is unique.

And also in formulation selection, for example, we deal with complex modalities like protacs, which are on the larger end of the spectrum, and the traditional small molecules. And all these molecules violate, typically, Lipinski's rule of five. So they're literally the brick dust, greaseball-type molecules. By leveraging these AI/ML tools we again, as I said, reduce the risk in product development, get to the milestones faster, and shorten timelines, etc.

The big advantage is time and cost savings and resource savings, for the industry and our customers. Time is money, literally speaking, and what we bring to the table is reducing risk in product development, saving time and resources, minimizing API consumption. It's an ultra-material sparing approach by leveraging some of these 

 modeling tools like AI, leveraging AI/ML for supporting product development and formulation development.

Every API, as I mentioned, is unique. So your structural features are very important, obviously the chemical structure, also fundamental properties like the melting point which have bearing on the crystal lattice, [the] energy of the molecule [is] important. Other properties like pKa, LogP, LogD, all of those are important.

Also, things like permeability are important. So typically we try to get data measured in 

 assays like KCO2 or MDCK, and we can leverage that to predict what's happening in the human body. These are some of the features that are important in some of these calculations that we do.

Sanjay Konagurthu of Thermo Fisher Scientific discusses how AI and ML can help solve the dilemma of poor solubility.

Leveraging AI/ML to Reduce Risk in Drug Development: Part One

Mechanistic models and hypothesis-driven strategies generate optimized efficient solutions for drug development, says Catalent’s Nathan Bennette at AAPS PharmSci 360.

Shifting from Trial-and-Error to Rational Design Drives Efficiency in Pharmaceutical Formulation

American Association of Pharmaceutical Scientists

, director, Scientific and Technical Advisory, Catalent Pharma Services, discusses a paradigm shift toward rational design—a method grounded in conceptual and mechanistic modeling. According to Bennette, the rational design approach brings clarity and efficiency to the development of new drug 

Instead of the traditional trial-and-error method, Bennette advocates a structured approach. “The rational design concept is to use models—conceptual models and mechanistic models—to develop more focused hypotheses and then targeted experimentation to more efficiently get at the solution,” he explains. This shift begins with the creation of conceptual models to understand the problem in depth, followed by mechanistic (quantitative or mathematical) models that reveal the fundamental physical and chemical processes at play, he explains.

Semi-empirical models, often realized via computer simulations, Bennette notes, further refine hypothesis testing and solution optimization. Bennette shares a recent example in which his team faced a drug with poor oral absorption. Through 

 experimentation, the team isolated the rate-limiting step—diffusion across the unstirred water layer—and designed formulations specifically to enhance that process. Bennette notes that the 

How are rational design models transforming pharmaceutical problem solving?

Bennette believes this efficiency-first mindset yields tangible industry advantages. “If you can come up with these really focused hypotheses and work on those problem statements, you get there faster. You get there with less expenditure of time and energy and resource, and, ultimately, I think you come up with a better product,” he says.

Bennette adds that, as pharmaceutical companies face mounting pressure to deliver innovations with speed and efficacy, rational design frameworks promise to shape industry standards for problem solving and formulation optimization.

Nathan Bennette, Director, Scientific and Technical Advisory, Catalent Pharma Services

Bennette is a subject matter expert on establishing developability and non-parenteral formulation strategies for poorly bioavailable molecules, including protein degraders and others described as “beyond rule of 5”. He has worked in drug development for over 15 years in a range of technical and leadership roles and has broad experience with formulation of small-molecule therapeutics. Bennette applies his passion for mechanistic understanding of drug delivery to design systems tailored to each molecule and target product profile, utilizing a range of oral technologies such as amorphous solid dispersions, controlled-release platforms, nanocrystalline formulations, multiparticulates, and lipid-based systems. He has experience working across the entire drug development lifecycle from preclinical and early phase development to late phase clinical studies and commercialization.

Editor's note: This transcript is a direct, unedited rendering of the original audio/video content. It may contain errors, informal language, or omissions as spoken in the original recording.

Hi, Nathan Bennette, I work in our scientific advisory function at catalent pharma solutions, and my role there is to be sort of an in house consultant working with customers to define the problem statement and develop technical strategy for formulations.

So in sort of traditional problem solving and pharma development and formulation and process development in particular, a lot of times, we have historically done a lot of trial and error, where you form a hypothesis, but then you run a large number of experiments trying everything under the sun to develop a solution to the problem. The rational design concept is to use models, conceptual models and mechanistic models to develop more focused hypotheses and then targeted experimentation to more efficiently get at the solution.

The question, it's principally around efficiency. Those trial and error experimentation methods require huge amounts of effort and to arrive at a robust solution eventually. But these mechanistic, driven, hypothesis, driven experiment designs allow us to use the best practices in science to build out our understanding of the system and more efficiently come to a more optimal solution.

I think of it in sort of three layers. First, you have conceptual models, you have to form a picture in your head of what is actually happening in the system that you're studying. Once you've done that, you can start to form a real hypothesis about what might be going wrong with your with your molecule, for example, as you're trying to develop your formulation, all formulation development is a problem solving exercise, and it's important to know what problem you're trying to solve. So this conceptual model allows you to have a framework for forming a hypothesis. Then there's another level that I call mechanistic models. Some people call them quantitative models, mathematical models. These are the basic physics that underlie the physical situation of your drug molecule and the question you're trying to answer maybe it's from an absorption perspective or from a stability perspective. And so once you have your quantitative mechanistic model, you can start to form a more specific hypothesis. How fast does this process happen to? What extent can we expect this process to proceed. And then finally, we have sort of empirical or semi empirical models where that mechanism and concept have been reduced to an actual computer program, and you can run simulations. So these tools are all available for a range of different questions that are relevant to pharma development, like pharmacokinetic absorption or chemical stability or process modeling for different pharmaceutical manufacturing processes. So we use the starting from conceptual to mechanistic to semi empirical models. We use this whole framework to describe the problems that we're trying to solve and develop actual formulations to resolve that.

I have a really good one from recent work that does both. It helped us to get much more quickly to an optimized solution that actually we tested it out in vivo, and it worked really well. So we were developing this compound that had very, very poor oral absorption. And we asked the question first, why? And we started with the conceptual model that every molecule has to get into the GI tract. It has to dissolve it then has to diffuse across the unstirred water layer, up against the epithelium and then partition across the epithelium to get into systemic absorption. So that's our conceptual model. And we asked, Well, why is this molecule having a hard time doing that? What? What is the fundamental rate limiting step? And we learned through a series of in vitro experiments that the rate limiting step was diffusion across that unstirred water layer, the mucus layer. And so with that in mind, we didn't, we didn't go out and try every formulation concept under the sun that would be a trial and error approach. Let's try this. Let's try this. Let's try this. We said, Okay, we know the key problem is diffusion rate. So what formulations can we develop that enhance the diffusion rate? Because that's the key step. So we went in the lab and we developed a few formulations that we proposed would help with that, and we tested them in vitro and demonstrated that they did, in fact, enhance the diffusion rate. And then we took the best of that set into an in vivo study. And what was really cool was that the extent to which. We enhanced diffusion with those formulations, in vitro correlated very well with the in vivo enhancement factor as well. So we saw, we figured out what the right problem was. We developed a conceptual framework for how to resolve that, and we went and tested that specific hypothesis, and much more quickly, much more efficiently, came to a workable solution.

I think I've been hitting a lot on the efficiency, the efficiency of the process of coming with a workable formulation. I think that translates to faster timelines if you're if you're spending enormous amounts of time trying everything in order to find the one thing that works. Then you spend a lot of resource. You spend a lot of time working on stuff that isn't isn't going to be the answer. But if you can come up with these sort of really focused hypotheses and work on those problem statements, you get there faster. You get there with less expenditure of time and energy and resource and ultimately, I think you come up with a better product. It's a it's a more optimized outcome.

Motor neuron connecting to muscle fiber, 3D illustration | Image Credit: © Julian - stock.adobe.com

A positive opinion from the European Medicines Agency’s Committee for Medicinal Products for Human Use (CHMP) has moved a higher-dose regimen of nusinersen closer to approval in Europe, according to a Nov. 17, 2025 announcement by Biogen. The committee’s positive opinion signals a potential shift in treatment approaches for spinal muscular atrophy (SMA), the company noted in a press release (1).

On the back of the committee’s recommendation, a 

 is expected from the European Commission in January 2026. The recommendation reflects a growing interest in dosing strategies that may offer expanded therapeutic benefit for individuals diagnosed with the most common form of the disease, 5q SMA (1).

SMA is a rare, genetic neuromuscular condition that affects people across age groups. With treatment landscapes evolving rapidly, decisions around dosing, long-term outcomes, and real-world effectiveness have become increasingly central for clinicians, researchers, and 

“While we’ve seen great progress over the past decade [Table], there is urgency to do more to address the clear unmet needs of the SMA community,” said Priya Singhal, MD, MPH, executive vice president and head of development, Biogen, in the press release (1). “The CHMP’s positive opinion for the high dose regimen of nusinersen represents a promising advancement in our commitment to support the evolving needs of individuals living with SMA and deliver therapies that can enhance patient outcomes.”

Biogen licensed the global rights to develop, manufacture, and commercialize nusinersen (brand name Spinraza) from Ionis Pharmaceuticals (1).

What study findings are shaping expectations for clinical impact?

The CHMP’s opinion is based on data from a multi-part Phase II/III study (DEVOTE), which explored safety, efficacy, and pharmacokinetics of a higher-dose regimen across treatment-naïve individuals and those who had been receiving the existing 12 mg dose. In treatment-naïve infants in the pivotal Part B cohort, investigators reported statistically significant improvements in motor function relative to a matched sham group drawn from an earlier study. Participants treated with the higher dose also experienced a reduction in the risk of death or permanent ventilation. These findings underscore the potential for dose intensity to influence therapeutic outcomes, an area of active discussion within neuromuscular research (1).

In Part C of the study, which followed individuals who transitioned to the higher-dose schedule after years on the standard regimen, investigators observed motor function improvements during the first 302 days of treatment. Because long-term durability and safety remain important considerations, the extension phase of the study continues to assess outcomes over time (1).

“The positive CHMP opinion for the high dose regimen of nusinersen is an important milestone for the SMA community,” said Eugenio Mercuri, MD, PhD, professor of Pediatric Neurology at the Catholic University, Rome, Italy, in the release (1). “Based on the results from the DEVOTE study and my experience with patients receiving this novel regimen, I am confident that high dose nusinersen has the potential to bring meaningful benefits to people living with SMA.”

How could a new dose option affect global regulatory decisions and care models?

Nusinersen is approved in more than 71 countries at its current dose, and the higher-dose regimen has recently received approval in Japan, according to Biogen. The company also stated in its release that regulatory reviews are underway in multiple regions, including the United States, for which a decision is expected in April 2026 (1). If additional regions move forward, health systems may see shifts in treatment planning, administration logistics, and long-term monitoring practices. Intrathecal administration, which requires specialized experience, will remain a key implementation factor for hospitals and clinics (3,4).

Safety observations from the study were generally aligned with both the known profile of nusinersen and common clinical events in SMA populations. Ongoing surveillance will continue to shape clinician guidance, particularly around procedural considerations and laboratory monitoring (1).

 the evidence, decisions around dose flexibility may influence how clinicians personalize therapy and how drug developers design future trials aimed at optimizing motor outcomes, durability, and safety for diverse SMA populations (5).

High Dose Regimen of Nusinersen Receives Positive CHMP Opinion for the Treatment of Spinal Muscular Atrophy

. Press Release. Nov. 17, 2025.
2. Servais, L.; Day, J. W.; De Vivo, D. C.; et al. Real-World Outcomes in Patients with Spinal Muscular Atrophy Treated with Onasemnogene Abeparvovec Monotherapy: Findings from the RESTORE Registry. 


3. Finkel, R. S.; Ryan, M. M.; Pascual Pascual, S. I.; et al. Scientific Rationale for a Higher Dose of Nusinersen. 


4. Cordts, I.; Lingor, P.; Friedrich, B.; et al. Intrathecal Nusinersen Administration in Adult Spinal Muscular Atrophy Patients with Complex Spinal Anatomy. 


5. Finkel, R. S.; Day, J. W.; Pascual Pascual, S. I.; et al. DEVOTE Study Exploring Higher Dose of Nusinersen in Spinal Muscular Atrophy: Study Design and Part A Results. 

U.S. FDA Approves Biogen’s SPINRAZA™ (nusinersen), The First Treatment for Spinal Muscular Atrophy

. Press Release. Dec 23, 2016.
7. Ionis. 

SPINRAZA® (nusinersen) Approved in the European Union

. Press Release. June 1, 2017.
8. Novartis. 

AveXis receives FDA approval for Zolgensma®, the first and only gene therapy for pediatric patients with spinal muscular atrophy (SMA). Press Release

FDA Approves Oral Treatment for Spinal Muscular Atrophy. Press Release

FDA Approves Genentech’s Evrysdi (risdiplam) For Use in Babies Under Two Months with Spinal Muscular Atrophy (SMA)

Articles

The European Medicines Agency's Committee for Medicinal Products for Human Use has given a positive opinion on a higher-dose regimen of nusinersen for spinal muscular atrophy (SMA), potentially altering treatment strategies. This decision, based on the DEVOTE study, showed significant motor function improvements and reduced mortality risk in treatment-naïve infants. The higher-dose regimen, already approved in Japan, is under review in other regions, including the U.S. If approved, it could impact global treatment planning and administration practices, emphasizing the need for specialized intrathecal administration expertise.

CHMP supports an enhanced nusinersen regimen, according to Biogen, indicating possible SMA treatment shifts and international regulatory decisions.

Implications on SMA Care of High-Dose Nusinersen

At AAPS PharmSci 360 2025, Elly Zhou says digital twins helped forecast the effects of drugs on human drugs via a digital control arm.

Achieving a Personalized Digital Control Arm for Preclinical Drug Evaluation 

In this interview regarding the poster presentation “A Novel Approach to Preclinical Drug Development Using Digital Twins of Ex Vivo Human Lungs” at 

, Elly (Xuanzi) Zhou, MS, PhD-Candidate, University Health Network, Toronto, Canada, discusses a project that seeks to address significant limitations inherent in traditional randomized control trials, namely, she says, the limited availability of human organs for testing and the high inter-subject variability that often complicates result interpretation. Furthermore, Zhou notes that conventional evaluation models like animal models or organoids often fail to fully capture the complexity of the human system.

The central innovation of her research is using digital twins to create a personalized digital control for every organ being treated. The digital twin leverages large-scale multimodal datasets to train machine learning models that accurately forecast future lung and organ function, Zhou notes. Specifically, after a human lung receives a drug treatment, the digital twin generates the hypothetical untreated effects—the outcome 

 the organ was not treated. This, she adds, serves as a robust digital control arm, enabling direct comparison between the observed treatment effects and the digital twin-generated untreated effects within the same lung. By reducing the reliance on traditional cohort-based analysis and the number of human lungs needed for control groups, this approach is designed to reduce study size and accelerate clinical trials, according to Zhou.

The foundation of the data collection is an

 lung perfusion (EVLP) system, an artificial circuit developed by thoracic surgeons at Toronto General Hospital in 2008. EVLP keeps human lungs alive outside the body by perfusing the organ through vasculature and keeping the lungs ventilated at 37 degrees Celsius. This method allowed Zhou and colleagues to test human lungs directly with a condition of interest, such as perfusing a lung with disease on the circuit. The resulting data are considered highly valuable because they are“clean and capture the lungs in isolation,” free from the systemic effects of the rest of the human body, Zhou explains.

The models incorporate multiple layers of data, including lung physiology (air pressure, compliance, gas exchange), biochemistry (pH level, electrolytes), imaging data (X-rays taken during EVLP), proteomic analysis, metabolic biomarkers, and transcriptomic pathways, says Zhou. This rich dataset is used to generate personalized forecasts. As she summarized,"If you are treating a lung in real life,… you give the treatment and then you observe the treatment effects in terms of organ function and behavior, but... digital twins… generate… untreated effects to serve as this personalized digital control for this organ being treated." Furthermore, the EVLP system demonstrated utility in Zhou and team’s study beyond testing, having been used as a therapeutic platform to perform gene editing using CRISPR technology on human lungs.

*Editor’s Note: This transcript is a direct, unedited rendering of the original audio/video content. It may contain errors, informal language, or omissions as spoken in the original recording.



 Hi everyone, my name is Elly. I'm a third year PhD candidate at the University of Toronto in Canada, and my supervisor is Dr. Andrew Sage. And my project is focused on developing digital twins of ex vivo human lungs to potentially accelerate preclinical drug development.

Yeah. So if you think about a randomized control trial where you have the two arm design, you have the control and the treatment. So essentially what happens is that we face two major challenges. The first is the limited availability of human organs for testing, and a second is, we're looking at a significant variability within the inter subject conditions that can cause the result interpretation to be very difficult.

So what digital twins can do is that it can generalize a personalized control for every organ that's been treated. So essentially after you provide a drug to treat the human lung, and then what you observe is essentially the observed treatment effects. And then for the digital twin part is that it's going to generate the untreated effects as if the organ was not treated.

So essentially this will allow us to reduce the reliance on a cohort based analysis where we can do direct comparisons between the untreated and treatment effects within the same lung. So this will reduce the study size because this will reduce the number of human lungs we use as controls. So in that way we'll be able to have a faster clinical trial.

So if we think about the traditional drug evaluation models, for example, animal models or organoids, they typically don't fully or often fully capture this. Complexity within the human body or system. So what digital twins can do is that we actually leverage a large scale multimodal data set to train the machine learning models to accurately forecast lung function, organ function of future time points.

So in this way we'll be able to essentially say what is the untreated effects of this drug in this given lung that's been treated? And in addition on this EVLP system, ex vivo lung perfusion, we’ll actually be able to test human lung with the condition of interest directly.

Like for example, we can perfuse a human lung with disease directly on the circuit and observe its behavior, the organ function.

So for this model, we actually incorporate multiple layers of data. So we have lung physiology data, for example, lung, air pressure, compliance, gas, like oxygenation exchange and also biochemistry data such as pH level, acid base and different electrolytes. We also have imaging data, x-ray images because on this EVLP profusion system, we can actually take x-ray images of the lungs.

And in addition to that, we also have proteomic analysis metabolic biomarkers and also transcriptomic pathways. And they have all been incorporated into the model.

Yeah. And so what digital twin does is its going to serve as this robust and also effective digital control arm in the study. So if you can imagine if you are treating a lung in real life while you give the treatment and then you observe the treatment effects in terms of organ function and behavior but what digital twins do, is that it's going to generate this, I guess, untreated effects to serve as this personalized digital control for this organ being treated. So by doing this comparison between the treatment effects and digital twin-generated, untreated effects, we can actually do a direct comparison to say, what is this drug's impact on this organ been treated.

Yeah, so what EVLP system is Ex vivo lung perfusion, which means we're trying to keep a set of human lungs alive outside the human body. So what we do is that we have this artificial circuit that was originally developed in 2008 by a group of thoracic surgeons at Toronto General Hospital. So what we do is we perfuse the organ through vasculature, keep the lungs ventilated through the trachea, keep everything at 37 degrees Celsius in a physiologically protective environment.

So during the process we'll be able to collect a set of clean multimodal lung function data. That can be used to train the digital twin model. And also this data is very clean in the sense that we're actually just looking at the lungs in isolation without systemic effects of the human body.

And in addition to that, at our hospital, we actually have demonstrated that the EVLP system can be used as a therapeutic platform in the work or in the study led by Dr. Shaf Keshavjee at Toronto General Hospital, we have shown that you can actually do gene editing on a set of human lung using CRISPR technology during EVLP

At AAPS PharmSci 360 2025, Siddhant Sojitra, Alexion, defines an agitation model to reliably test stability for early-stage biologics.

A Scientifically Justified Agitation Stress Model for Early-Stage Biologics 

Siddhant Sojitra, Associate Scientist III, Injectable Drug Product Development, Alexion Pharmaceuticals, discusses the primary role played by scale-down agitation in the development of high-concentration biologics by simulating mechanical stress on drug products in this interview regarding the poster presentation “

Defining a Scale-Down Agitation Model for Early-Stage High Concentration Biologics Formulation Development

. This stress, often encountered as shaking and mixing, occurs throughout the drug product lifecycle, including during manufacturing, transportation, storage, and administration, Sojitra and colleagues 

. Utilizing this model offers crucial insights into how varying levels of agitation impact drug products. Ultimately, the goal is to develop more stable formulations capable of sustaining these mechanical forces without compromising the drug's quality or effectiveness, he adds.

Mechanical stress testing is essential because constant agitation leads to significant negative impacts on product quality, primarily protein aggregation and particulates formation, Sojitra notes. The shaking can expose the protein’s hydrophobic patches, causing nucleation and aggregation, and may also unfold the protein structure. Aggregation severely compromises the protein’s potency, while the formation of particulates raises the risk of immunogenicity and can violate regulatory guidelines, the study team notes in their poster.

To assess the stability of biologic formulations against these stresses, quality attributes (QAs) are demonstrated through various testings, including analysis of appearance, turbidity, size/charge variance, aggregates formation, and particulates formation, says Sojitra. The researchers specifically utilized analytical methods, such as size-exclusion ultra-performance liquid chromatography, to measure high molecular-weight species (HMWS), micro flow imaging to detect subvisible particles (SVP), and optical density (OD) to assess turbidity.

The study systematically evaluated three physical agitation models: an orbital shaker, a multichannel vortexer, and a bench-top shipping simulator (VR5500). Comparisons showed that both the orbital shaker (at 200 RPM) and the vortexer (at 1200 RPM) subjected samples to more severe mechanical stress than the VR5500, making them suitable for robust formulation development studies. A key finding was that the different models impacted distinct QAs. The orbital shaker was found to predominantly promote the formation of HMWS, whereas the vortexer was more likely to induce the generation of SVP.

Based on the study's findings, the optimal, scientifically justified agitation model for early-stage formulation development was defined. Recognizing the need for smaller sizes due to material constraints in early development, the standard configuration selected was the 2R vial. The data support using the 2R vials with a minimum fill volume of 1 mL, positioned in a horizontal placement, and subjected to orbital shaker agitation at ambient temperatures for up to 24 hours. This setup provides a reliable and consistent method to assess the impact of mechanical stress, thereby helping to identify stable formulations early in the development pipeline, Sojitra and team stated.

Check out Sojitra’s take on the roles and barriers to AI and digital technology in biologics, our 



I am Siddhant Sojitra. I work as an associate scientist three in drug injectable drug product development team at Alexion, AstraZeneca. I have three plus years of working experience in the pharmaceutical industry now. My specialization includes biologics drug product formulation development, especially high concentration monoclonal antibody-based formulations.

I have an engineering background. I did my masters in chemical engineering from Northeastern University of Boston.

So the primary role of scale-down agitation model is to simulate agitation stress on drug products. This mechanical stress is in the form of agitation, which means drug products are getting exposed to shaking, mixing, and this is often encountered during manufacturing, transportation, storage, administration.

So that's where the scale-down agitation model comes in. Having this model gives us crucial insights about how different label of agitation impacts drug products, and can also help us to develop formulations much more stable version of drug products, which can sustain all this shaking without compromising its quality and effectiveness.

The major impact on drug product quality are protein aggregation and particulates formation. So because of this constant shaking and agitation, proteins often get exposed to its hydrophobic patches, which can nucleate segregation. It can also unfold protein, and because of this aggregation, it impacts protein's potency, and particulates formation increases immunogenicity risk, and it can also violate guidelines.

Drug products’ quality attributes are demonstrated by various testings, such as appearance and turbidity, measuring size and charge variance, aggregates formation, particulates formation. This is how the drug products quality attributes are demonstrated.

So for early-stage stress testing, we are keeping material constraints in mind. We focused on smaller size vials, which are widely used and our data shows that at least 1 mL in 2R vials when shaken horizontally, using an orbital shaker up to 24 hours, provides a reliable approach to study early-stage stress testing.

Following mechanical stress, there are various testings done, which includes appearance, turbidity, particulate formation, using microflow imaging techniques or light obscuration techniques, aggregates formation using size-exclusion chromatography. And this gives us an idea how drug product is behaving and how the impact of agitation is on drug product.

Siddhant Sojitra, Alexion, defines in an AAPS PharmSci 360 2025 presentation an agitation model to reliably test stability for early-stage biologics.