Patient-centric approaches to vaccine development

What is the overarching goal of the patient-centric approach to vaccine CMC development?

CP: The overarching goal of the patient-centric approach [1] [2] is to prioritize patient safety and product efficacy when designing a product and related CMC characteristics and associated quality expectations. This can be achieved by leveraging product characterization and prior knowledge, and generating nonclinical and clinical evidence. A thorough understanding of the product is crucial for developing a control strategy ensuring that the required quality is consistently achieved at production and maintained throughout the entire shelf-life of the product. According to ICH Q8 (R2), in a ‘minimal’ approach, product specifications are a primary means of control and are based on batch data available at the time of registration; an enhanced patient-centric approach integrates specifications into the overall quality control strategy, basing them on desired product performance supported by relevant data.

What are critical quality attributes (CQAs) and why are they important?

CC: CQAs are physical, chemical, biological, or microbiological properties of the vaccine that must be maintained within specific limits to ensure the desired quality of the product. They are crucial because they directly impact the safety and efficacy of the vaccine, guiding the development process to ensure consistent control and monitoring.

Take ‘integrity’ of an mRNA vaccine as an example. Ensuring the integrity of the mRNA is vital for vaccine efficacy, as degraded mRNA may lead to the production of non-functional antigens, potentially failing to provide protection. Advanced analytical methods, such as high-performance liquid chromatography, are used to monitor mRNA integrity during manufacturing and storage. Throughout the vaccine’s lifecycle, mRNA integrity levels may vary without adversely impacting efficacy, if values are within appropriate specification limits. Therefore, degradation kinetics of mRNA integrity is among the CQAs determining vaccine shelf-life.

Can you explain the main strategies within the patient-centric approach and their significance?

CP: The patient-centric approach comprises both proactive and reactive strategies. The proactive strategy focuses on generating data to establish relationships between patient effects and product characteristics through controlled experimentation via both non-clinical and clinical studies and integrating this information via model-based approaches. The reactive strategy monitors clinical development signals and real-world product usage, assessing observations at the population level. Both strategies ensure continuous evaluation and adjustment of CQAs to align vaccine characteristics with patient outcomes.

What role does model-informed vaccine development (MIVD) play in enhancing the patient-centric approach to vaccine development?

RD: MIVD [3] utilizes quantitative in silico approaches integrated with data from preclinical and clinical studies. This methodology supports decision-making and regulatory considerations, optimizes development processes, minimizes uncertainty, and improves the probability of clinical trial success by refining doses, dosing regimens, formulations and precisely targeting patient sub-populations. The MIVD toolbox includes classical statistical and translational models, mechanistic quantitative systems pharmacology models that combine systems biology and vaccine pharmacology models, scientific machine learning, and model-based meta-analysis, among other methodologies.

How can the challenge of limited data from animal and human studies in vaccine development be addressed?

RD: At the heart of patient-centricity is an iterative evidence-based process. This process accumulates data and insights from in vitro, in vivo, and in silico studies to validate or refine hypotheses. Comprehensive data collection—including analytical information, research observations, clinical trials, and real-world evidence—and using in silico models to evaluate, connect, validate, and interpret data ensures alignment of vaccine specifications with patient needs and safety standards. A key highlight to emphasize is that in silico models support the iterative process of data accumulation, enable predictive simulations, bridge gaps in data, reduce uncertainty, and enhance decision-making throughout the vaccine development lifecycle.

What is the purpose of triaging quality attributes in the patient-centric development framework?

FM: Triaging involves listing all quality attributes of interest based on the quality target product profile (QTPP) for the purpose of identifying potential CQAs. This process confirms the criticality of these attributes, ensuring that those with a significant impact on safety and efficacy are considered in the decision-making process, thereby streamlining the development and regulatory approval of the vaccine.

How can evidence-based data for confirmed CQAs be generated?

FM: For confirmed CQAs, a comprehensive plan incorporating both direct measures (impact on disease) and indirect measures (immune markers) of safety and efficacy should be developed. This involves conducting in vitro and in vivo studies focusing on essential properties such as purity, identity, potency, safety signals, and immune biomarkers. The iterative process refines CQA ranges through data from bespoke non-clinical and clinical studies, supported by statistical and modeling analyses.

RD: An important related aspect of MIVD in this context is the learn–confirm–predict cycle, which is an iterative process that enhances decision-making and reduces uncertainty throughout vaccine development. This process involves learning from existing data, confirming findings through additional studies, and predicting outcomes using in silico models. By leveraging prior knowledge, validating hypotheses, and forecasting outcomes, this process helps to optimize clinical trial designs and refine doses. The iterative nature allows for continuous improvement and adaptation to new information. Overall, it lowers uncertainty and improves the success rates of clinical trials, thus accelerating the development of safe and effective vaccines.

What is the significance of the QTPP in the patient-centric development framework?

CC: The QTPP serves as the foundational framework for drug and vaccine design, providing a forward-looking summary of quality characteristics critical for ensuring safety and efficacy. It guides the shaping of product attributes and the development of processes and analytical methods, evolving through the development stages from initial drafts to a complete version by launch, ensuring alignment with patient needs.

How can the challenge of setting specifications with limited manufacturing experience be addressed?

CP: A shift towards a patient-centric development framework that prioritizes patient safety and efficacy over traditional process benchmarks can address this challenge. By leveraging existing knowledge and insights and generating new evidence, vaccine specifications can be reoriented to focus on patient outcomes rather than solely on manufacturing experience and stability data from limited lots, thereby ensuring robust product quality.

How can changes in product attributes throughout the vaccine’s lifecycle be handled?

FM: An iterative approach should be adopted to consistently review and reassess CQAs, enhancing knowledge through additional in vitro, in vivo, in silico, and clinical studies. This process ensures that any changes in product attributes, such as vaccine thermostability and degradation under various storage conditions, are evaluated for their impact on safety and efficacy, maintaining alignment with patient needs and regulatory requirements.

What is the importance of immune biomarkers in the context of a patient-centric vaccine development?

RP: Immune biomarkers are crucial for predicting vaccine efficacy, assessing safety, and guiding the development of safe and effective vaccines. They serve as surrogate endpoints that provide insights into the immune response elicited by the vaccine, helping to establish correlations between immune markers and clinical outcomes, such as protection against disease, thereby supporting evidence-based decision-making.

Which kind of models or approaches can inform patient-centric vaccine development?

RP: There are many such approaches. As an example, controlled human infection models (CHIMs) can provide valuable input on key immune parameters and serve as proof-of-concept to assess potential vaccine candidates. CHIM studies can be used to validate biomarkers associated with protection, support dose selection, and provide early evidence of vaccine efficacy, thereby de-risking later-stage clinical trials. However, not all diseases have CHIMs available, therefore, analysis of samples from patients at early infection stages and exposed uninfected individuals can support biomarker discovery and validation.

MF: Preclinical-to-clinical translation involves using data from animal studies to predict human responses [3]. This process helps bridge the gap between preclinical findings and clinical outcomes, ensuring that the vaccines are accurately assessed before human trials, thereby reducing the risk of adverse events and improving the likelihood of clinical success. Based on early clinical data readout and using the learn-confirm-predict paradigm, MIVD approaches can contribute to dose selection by using quantitative models to predict the optimal dose that balances immunogenicity and reactogenicity. This approach allows for the identification of dosing regimens that maximize efficacy while minimizing adverse effects, ensuring that the vaccine is both safe and effective for the target population.

Along similar lines, what is the role of dose–response modeling in patient-centric vaccine development?

RD: The right choice of vaccine dose can be the difference between clinical success or failure for a vaccine. Dose–response modeling plays a critical role in understanding the relationship between vaccine dose and the resulting immune response. By analyzing dose-response curves, researchers can determine the optimum dose and dosing regimens to achieve the desired immunogenicity with minimal side effects, thereby enhancing patient safety and efficacy. Vaccine developers are embracing this approach for enhancing probability of clinical success. A published example is using model-based approaches to guide the pediatric dose selection of a SARS-CoV-2 vaccine [4].

MF: Two key points to highlight here. First, vaccine development teams typically consider the appropriate dose for a new vaccine from the beginning by leveraging prior knowledge on similar products or platforms. This stage, often referred to as ‘Phase 0’, provides an opportunity to apply model-based approaches, gain quantitative insights, guide experimental design, and increase the chances of clinical success. Second, vaccine dose-responses often differ between animal models and humans, necessitating translational frameworks to bridge the two. Therefore, for both these reasons, it is crucial to embed MIVD approaches into vaccine development from the outset.

How do correlates of protection (CoP) to predict vaccine efficacy in new clinical trials fit with MIVD approaches?

MD: Correlates of protection are immune markers that are associated with protection against disease [5]. Identifying these markers is important because they can serve as surrogate endpoints in clinical trials, allowing for the prediction of vaccine efficacy without the need for large-scale efficacy studies. CoP can be used to predict vaccine efficacy by quantitatively correlating the immune responses elicited by the vaccine to protection from disease. By analyzing immune responses across vaccinated individuals, researchers can estimate the vaccine’s efficacy in new clinical trials, thereby accelerating the development process and ensuring that the vaccine meets patient needs. CoP within the context of MIVD also helps to elevate dose-immunogenicity response modelling to dose-efficacy modelling to enable precision dosing strategies across patient sub-populations such as elderly, pediatric, and immunocompromised, thus enabling patient-centric vaccine deployment.

References

1. Algorri M, Cauchon NS, Christian T, O’Connell C, Vaidya P. Patient-centric product development: a summary of select regulatory CMC and device considerations. J. Pharm. Sci. 2023; 112(4), 922–936.  Crossref

2. Krause P, Campa C, Chang A et al. A vision for patient-centric specifications for biologicals. Biologicals 2024; 88, 101796.  Crossref

3. Desikan R, Germani M, van der Graaf PH, Magee M. A quantitative clinical pharmacology-based framework for model-informed vaccine development. J. Pharm. Sci. 2024; 113(1), 22–32.  Crossref

4. Ivaturi V, Attarwala H, Deng W, et al. Immunostimulatory/immunodynamic model of mRNA-1273 to guide pediatric vaccine dose selection. CPT Pharmacometrics Syst. Pharmacol. 2025; 14(1), 42–51.  Crossref

5. Dull P, Plotkin SA, Gilbert P, Cassels F. Correlates of protection for SARS-CoV-2 vaccines. Vaccine Insights 2022; 1(1), 85–93.  Crossref

Biographies

Rajat Desikan is a Scientific Director within Clinical Pharmacology Modelling & Simulation in R&D at GSK, based in the United Kingdom. He leads the integration of advanced modelling approaches, including quantitative systems pharmacology, into clinical-stage drug and vaccine development, with a focus on infectious diseases such as HIV, Hepatitis B and respiratory vaccine portfolios. His work supports internal and regulatory decision-making, including Go/No-Go assessments, dose selection, and clinical trial design. Prior to this role, Rajat has experience in molecular modelling within the discovery space and disease-area modelling across preclinical and clinical stages, working on therapeutic areas such as immune-oncology, autoimmune diseases, and vaccines. His expertise spans diverse drug and vaccine modalities, including small molecules, monoclonal and bispecific antibodies, siRNA, antisense oligonucleotides, mRNA vaccines, and liposome-encapsulated drugs. In addition to his work at GSK, Rajat serves as an associate editor for CPT: Pharmacometrics & Systems Pharmacology, a journal of the American Society for Clinical Pharmacology & Therapeutics (ASCPT).

Rajat Desikan, Clinical Pharmacology Modelling & Simulation, GSK, UK

Cristiana Campa is the External CMC Intelligence Lead in Global Technical R&D at GSK, based in Italy. She has been instrumental in advancing innovative CMC development strategies and integrating QbD principles across GSK Vaccines. Prior to joining GSK, she served as Head of Analytical Development at Novartis Vaccines and has over two decades of experience as a researcher and team leader in academia and industry. She actively contributes to global regulatory and industry groups, including the Expert Working Group for ICH Q6 Guideline revision (specifications), pandemic preparedness initiatives in Vaccines Europe/IFPMA, and CEPI. In addition, she is deeply involved in scientific committees and technical conferences, and she serves on the Board of Directors at the Parenteral Drug Association (PDA). Her leadership and advocacy continue to drive innovation and collaboration across the biopharmaceutical sector.

Cristiana Campa, Vaccines Technical R&D, GSK, Italy

Marc Fourneau is a Senior Director of Translational Statistics and Modelling in Research at GSK Vaccines, based in Belgium. With a BSc in Biology, he joined GSK in 1988 and has since held various roles of increasing responsibility and leadership within the statistical organization, contributing to pre-clinical, clinical, and epidemiological research for the development of numerous vaccines, including Hepatitis A, B, AB, and E; DTPw and DTPa combination vaccines; Hib, HSV, strep, typhoid, flu, dengue, MMRV, and shingles. In his current role, which he has held since 2021, Marc leads efforts to integrate advanced statistical methodologies and modelling approaches to support translational research and vaccine innovation. His extensive experience and long-standing commitment to vaccine development have been instrumental in advancing GSK’s portfolio and addressing global health challenges.

Marc Fourneau, Research and MDS Statistics, GSK, Belgium

Ricardo Palacios is the Clinical Project Lead (Senior Director) for Early Bacterial Projects at GSK’s Vaccine & Infectious Diseases Research Unit in Siena, Italy. He provides clinical leadership and strategic direction for the development of vaccines and monoclonal antibodies targeting bacterial infections, guiding projects from conception to early clinical stages. Prior to joining GSK, Ricardo led vaccine development for arboviruses and respiratory viruses at the Instituto Butantan in São Paulo, Brazil, where his contributions to the phase III dengue vaccine trial was recognized by the New England Journal of Medicine as one of the Notable Articles of 2024. During the COVID-19 pandemic, he represented the Developing Countries Vaccine Manufacturing Network (DCVMN) on the COVAX Clinical SWAT team, designing a pivotal trial that supported the WHO’s Emergency Use Listing for Sinovac’s COVID-19 vaccine, CoronaVac, while also supporting clinical sites and trials funded by WHO across Asia, Africa and Latin America. A physician with a PhD in Infectious Diseases, Ricardo also holds a BSc in Social Sciences and a specialization in bioethics.

Ricardo Palacios, Clinical Science Bacterial Vaccines, GSK, Italy

Martine Douha is a Lead Statistician for CMC projects at GSK, based in Belgium. She has been instrumental in integrating statistical methodologies into product, process, and analytical development, supporting successful vaccine development and regulatory approvals, particularly for LVV vaccines. Previously, she served as a clinical project statistician at GSK Biologicals from 2003–2018, where she oversaw vaccine development trial designs, analyses and regulatory submissions for projects such as Priorix-Tetra, Shingrix, RSV and nicotine vaccines. With over 25 years of experience in the biopharmaceutical industry and advanced degrees in Mathematics and Applied Statistics, Martine remains a strong advocate for aligning clinical and CMC specifications with patient-centric goals while driving innovation across research and development.

Martine Douha, Vaccines CMC Statistics, GSK, Belgium

Frédéric Mathot is a Scientific and Technical Lead (Associate Director) for drug product development at GSK’s Vaccine Research & Development Centre in Rixensart, Belgium. He provides strategic leadership on drug product development and associated processes for new vaccine candidates, collaborating with key internal and external stakeholders. Over the past decade, he has focused on advancing drug product development for nucleic acid-based platforms, including adenoviral/ MVA vectors and mRNA/LNP modalities, working with partners such as BARDA, WuXi Biologics, and CureVac. A pharmacist with a PhD in Pharmaceutical Sciences from UCLouvain, Frédéric also leads initiatives to onboard novel technologies, such as AI/ML platforms, alternative devices, and innovative drying technologies, while contributing to academia through PhD supervision and lecturing at UCLouvain.

Frédéric Mathot, Vaccines Technical R&D, GSK, Belgium

Carlo Pergola is a Senior Director of Global Product Development at GSK, based in Italy and Belgium. He leads cross-functional teams in advancing vaccine drug product technical development through clinical phases and achieving product approvals, with expertise in CMC development and QbD. Prior to joining GSK, Carlo served as Laboratory Head of Vaccine Chemistry and Formulation at Novartis Vaccines and held senior roles at Merck KGaA, where he directed formulation and analytical strategies for biotech therapeutics. Holding a PhD in Pharmaceutical Sciences, Carlo has also contributed to academic research at the University of Naples, Tübingen, and Jena. His work integrates advanced methodologies to accelerate drug development and ensure alignment with global regulatory standards.

Carlo Pergola, Vaccines Technical R&D, GSK, Italy

Authorship & Conflict of Interest

Contributions: The named author takes responsibility for the integrity of the work as a whole, and has given their approval for this version to be published.

Acknowledgements: None.

Disclosure and potential conflicts of interest: This work was sponsored by GlaxoSmithKline Biologicals SA. All authors are employees of the GSK group of companies and declare owning GSK shares.

Funding declaration: The author received no financial support for the research, authorship and/or publication of this article. 

Article & Copyright Information

Copyright: Published by Vaccine Insights under Creative Commons License Deed CC BY NC ND 4.0 which allows anyone to copy, distribute, and transmit the article provided it is properly attributed in the manner specified below. No commercial use without permission.

Attribution: Copyright © 2025 Rajat Desikan, Cristiana Campa, Marc Fourneau, Ricardo Palacios, Martine Douha, Frederic Mathot, Carlo Pergola. Published by Vaccine Insights under Creative Commons License Deed CC BY NC ND 4.0.

Article source: Invited.

Revised manuscript received: May 20, 2025.

Publication date: Jun 27, 2025.