In an interview with MedTech Spectrum, Lawrence Florin, CEO of Hesperos, Inc., explains how the company’s Human-on-a-Chip® (HoaC) platform is addressing the limitations of animal models in drug development. Using human cell–based single- and multi-organ systems, Hesperos delivers safety, efficacy, PK/PD, and toxicity data that meet regulatory standards and support IND filings. With the addition of real-time digital twins, first demonstrated through the Malaria-on-a-Chip model, the company is advancing translational science and working to scale these innovations for broader disease coverage and integration into mainstream pharmaceutical R&D and clinical workflows.

How does Hesperos’ Human-on-a-Chip® platform overcome current limitations in disease modeling?

Despite decades of reliance on animal models, the stark reality is that over 90 percent of drugs that appear safe and effective in animal studies ultimately fail in human clinical trials. This high attrition rate highlights the fundamental disconnect, that animal biology (including humanized animal models) often fails to replicate human-physiological responses. These limitations are well known, but until recently, researchers have not had alternatives. However, the advent of new approach methods (NAMs) —such as Hesperos’ Human-on-a-Chip® (HoaC) single- and multi-organ systems —that have been endorsed by FDA, the NIH and other international regulatory health agencies and scientific organizations, ushers in a new era of drug (along with cosmetics and nutraceutical) development along with chemical and food safety testing.

Hesperos’ HoaC platform uses human cells that have been harvested from patients or have been differentiated from progenitor (stem) or fibroblast cells into 2- or 3-dimensional human organ (cellular) constructs. Generally, such organ-on-a-chip systems can be used in carefully designed microphysiological systems to assess the safety of new drugs (or nutraceuticals, food ingredients and/or chemical products). 

What sets Hesperos’ single and multi-organ system models apart is that we not only can evaluate safety for drugs and other products, but we enhance this capability to be able to assess both on and off-target toxicity as well as measure clinically relevant functional readouts (i.e., efficacy parameters) in addition to pharmacokinetic and pharmacodynamic profiling, and more recently, generate a digital twin. Moreover, this set of unique capabilities has permitted our clients to use the efficacy (and other data) data to support their INDs with FDA and move their products into clinical trials.

As mentioned, animal models often are not translatable, this is particularly problematic in rare diseases where these models often are not only inadequate but do not exist. Thus, the ability to incorporate normal and/or diseased patient cells into the HoaC organ systems can help ensure that the experimental results will reflect, to a far higher degree of probability, the efficacy and safety of a new drug (or other product) in patients.

Hesperos, is committed to the internationally recognized 3Rs initiative to replace, refine and reduce the use of animals in research by relying on human-relevant systems to improve decision-making, by providing better information and accelerating development timelines in a cost-effective manner. The advantages of NAMs may be readily evident, but the confirmation of regulatory acceptance of the data generated should help overcome the inertia of pharmaceutical companies adopting these approaches into their standard business practices. 

What distinguishes Hesperos’ digital twin breakthrough from other in silico simulation models?

Digital (medical) twins utilize data that have been generated and synthesize it with newer data to refresh the human physiological model to predict the structure and behavior of products used to improve health and/or treat disease. Thus, digital twins help researchers evaluate the profile of a new drug (or other product) before advancing it into clinical trials with patients. Moreover, once clinical trials begin, digital twins can help reduce their time and cost, the most expensive and time-consuming product development phase of development, by reducing the numbers of patients that need to be enrolled.

What distinguishes a digital twin’s approach from traditional in silico, simulation, modeling is that digital twins require continuous (near real time) updates while conventional simulation models extrapolate from static datasets that are only periodically updated (if ever). Thus, while simulation modeling plays an important role in product development, to test hypotheses and design development programs, digital twins offer a new, real-world, technique that will better inform product development built on a continuous feedback loop.

Hesperos’ work, done in collaboration with leading scientists and institutions, and funded in part by the Gates Foundation, relied on continuous data generated by our newly introduced Malaria-on-a-Chip (MoaC) organ system model. The MoaC used human liver, spleen, and endothelial tissue along with infected blood to replicate the lifecycle of the parasite responsible for the deadliest form of malaria and determine the safety, efficacy and pharmacokinetic and pharmacodynamic profiles of multiple antimalarial agents. 

This groundbreaking work is an important first step in the development and future enhancements of organ-on-a-chip systems to transform preclinical testing by incorporating digital twins as a standard mechanism to improve, accelerate and lower the cost of development of new therapies. 

Given the platform’s demonstrated ability to predict human drug responses—including efficacy, toxicity, and immune signaling—what are the regulatory implications for replacing or supplementing animal testing in preclinical development?

The US (and international) regulatory landscape(s) is/(are) rapidly evolving. Recently, both NIH and FDA have reaffirmed their support of using, non-animal, new approach methods (NAMs), such as organ-on-a-chip systems, in vitro assays, and in silico computational modeling to improve the drug development process (and by extension testing of foods and chemicals and other products) by providing data that better inform how these products will affect patients and do so faster and less expensively than conventional, often poorly translatable, animal models.

Hesperos’ Human-on-a-Chip® (HoaC) single- and multi-organ system platform uses human-based organ cell constructs to recapitulate human physiology. Unlike other systems that primarily focus on blunt safety measures, our advanced models generate intricate safety data (including on- and off-target toxicity), clinically relevant functional (i.e., efficacy), PK/PD and now, digital twins. 

The comprehensiveness of our services has enabled our HoaC platform to support multiple IND filings, primarily for rare diseases, that have advanced into clinical trials, and in four cases wherein there already was a foundation of data about the drug, the program moved directly into Phase 2 (and one drug already has moved into Phase 3, with a second expected to do so in the coming months). Thus, we have demonstrated that the FDA will accept data generated from our HoaC platform, and over time from other NAM solutions.

Notwithstanding this success, in the near-term, animal models will continue to be used, particularly for whole animal safety evaluations. However, as our HoaC system continues to evolve and other NAMs become qualified, collectively they will continue to meet the goals of the 3Rs initiatives to replace, refine and reduce the use of animals in research and substitute more informative human-based in vitro models and assays and accompanying in silico computational modeling.

What challenges remain in scaling or personalizing these digital twins for broader disease categories beyond malaria, and how close are we to applying this technology in routine precision medicine?

The processes needed to scale digital medical twins for a wide number of diseases, including disease subtypes, requires, but is not limited to: developing appropriate organ-on-a-chip systems, procuring and being able to properly sustain the appropriate organ cells (including cells obtained from patients), understanding of the underlying disease mechanism and assembling a body of relevant data to permit the creation of the needed digital twin.

Adoption of digital twins into preclinical and clinical research requires confirmation of the utility of the models and the associated in silico computational datasets by compiling relevant (sub)population modeling to qualify the systems and assure acceptance by health authorities. Concurrently, data outputs digital twins can be harnessed by clinicians at the bedside to make better informed decisions regarding how to treat their patients thus bringing precision medicine into mainstream clinical practice.

How does the integration of PK/PD modeling and in vitro to in vivo extrapolation (IVIVE) enhance translational relevance, and what role do collaborations (e.g., with MMV and academic centers) play in advancing this work?

The integration of PK/PD modeling and IVIVE within our HoaC platform increases its translational relevance by providing more actionable insights to researchers (and potentially clinicians) by providing information about a drug’s profile including its absorption, distribution and metabolism (including active metabolites) that impact its pharmacokinetic and pharmacodynamic safety and efficacy profiles. Often researchers face difficulties in finding the appropriate dosage range of a new entity due to any of these or other factors such as the metabolomics of patients, or due to drug-drug interactions.

Properly designed organ system models combined with ever improving PK/PD profiling datasets will help to accelerate the adoption of these tools to bridge the gap between in vitro data and clinical outcomes, enabling us to predict therapeutic efficacy, identify potential toxicities, and optimize dosing strategies before clinical trials begin.

Collaborations with academic centers and patient focused (disease) groups such as Medicines for Malaria Venture (MMV) play a crucial role in the development, refinement and eventual usage of these in vitro / in silico solutions. Hesperos strongly believes that the efforts of academic research centers are integral to future success. In fact, two of our co-founders, J. Hickman, PhD and Michael Shuler, PhD, retain their academic appointments at University of Central Florida and Cornell University, respectively. Further, patient organizations not only offer insights from patients and caregivers, but also funding for selected projects along with a cadre of highly respected experts that can provide intellectual support to these endeavors.

From a commercialization and adoption standpoint, what are the next steps Hesperos envisions for bringing this platform into mainstream pharmaceutical R&D pipelines and clinical decision-making workflows?

From a commercialization perspective, our near-term focus is to continue to develop and refine the Malaria-on-a-Chip multi-organ system model and use it as the foundational blueprint to build other models that are fit-for-purpose for specific disease indications. We welcome the opportunity to perform the development work needed with biopharmaceutical sponsor companies so that the resultant models can help accelerate their development pipelines by proving more relevant and predictive information, more efficiently (i.e., increased speed and lower cost).

Hesperos also is actively involved in another noteworthy, FDA-driven, initiative. We are developing an in vitro model that will be qualified to serve as drug development tool (DDT). DDTs are systems that the FDA has review and ‘qualified’, whereas a qualified DDT is one that may be used to support product development programs and to address related manufacturing requirements, among other uses, with the approval of the agency.