Solutions
About Us
Insights
Careers
Contact us
Contact Us
Customer Support
Customer Support

From Regulators to Reimbursement: What the EMA-FDA AI Principles Mean for HEOR

In January 2026, the European Medicines Agency (EMA), together with the U.S. Food and Drug Administration (FDA), have taken an important step by publishing the “Guiding Principles of Good AI Practice in Drug Development.” This document is more than a technical checklist — it is a clear signal that regulators are getting serious about how artificial intelligence (AI) should be developed, validated, governed, and, ultimately, trusted across the medicines lifecycle.

While the principles are formally framed around drug development, their implications go well beyond non-clinical and clinical domains. For Health Economics and Outcomes Research (HEOR), this guidance offers something the field has long needed: a credible regulatory blueprint for responsible AI use that could help agencies move from cautious experimentation to structured adoption.

 

Why this matters now

AI is already being used across HEOR — whether for real-world evidence generation, economic modeling, patient segmentation, or long-term outcome prediction. Yet, despite methodological innovation, acceptance by HTA bodies and payers remains uneven. One of the key barriers is not capability, but confidence: confidence in transparency, robustness, reproducibility, and governance.

By articulating shared principles for AI use, the EMA and its partners are laying the groundwork for that confidence. Importantly, they are doing so in a way that aligns closely with the questions HTA agencies ask every day: What is this model for? What risks does it introduce? Can we trust the outputs? And how do we manage it over time?

 

A bridge to HEOR: Learning from regulatory leadership

We have already seen how regulatory clarity can accelerate adoption. The UK, for example, has actively explored how AI can be used to support evidence generation and decision-making in health systems. EMA-FDA’s principles create an opportunity to extend this momentum across Europe and beyond — including into HEOR and HTA decision frameworks.

Although all ten principles are relevant, four stand out as particularly transformative for HEOR.

 

Four principles with outsized impact on HEOR

1. Human-centric by design

This principle explicitly anchors AI development in ethical and human-centric values. For HEOR, this is critical. Economic models and real-world analyses directly influence access, reimbursement, and, ultimately, patient outcomes.

A human-centric approach reinforces that AI in HEOR should support, not replace, expert judgement. It legitimizes hybrid workflows where analysts, clinicians, patients, and decision-makers remain central, while AI enhances scale, speed, and insight. This framing directly addresses common HTA concerns about “black box” decision-making.

 

2. Risk-based approach

Not all AI use cases carry the same consequences, and this principle explicitly recognizes this. For HEOR, this principle is particularly powerful.

Using AI to automate literature screening does not pose the same risk as using it to inform long-term survival extrapolations or pricing decisions. A risk-based approach allows proportionate validation, governance, and oversight — making AI adoption more realistic and scalable for both developers and agencies.

This is precisely the kind of nuance HTA bodies need to move beyond binary “acceptable/not acceptable” positions on AI.

 

3. Risk-based performance assessment

Closely linked, the EMA and FDA emphasize that performance assessment should consider the complete system, including human-AI interaction, and be tailored to the intended context of use.

For HEOR, this reframes validation away from abstract accuracy metrics and toward decision relevance. The key question becomes: Is this AI fit-for-purpose for the policy or reimbursement decision it supports? This aligns naturally with HTA thinking and opens the door to more pragmatic, decision-focused validation frameworks.

 

4. Life cycle management

Perhaps the most underappreciated principle in HEOR today is life cycle management. The EMA highlights the need for ongoing monitoring, re-evaluation, and management of issues such as data drift.

HEOR models are often treated as static artefacts, yet AI-enabled models evolve as data, clinical practice, and populations change. Recognizing AI as a living system — not a one-off submission — could fundamentally change how HTA agencies think about post-submission evidence generation, managed entry agreements, and reassessment over time.

 

From drug development to HTA: An opportunity not to miss

This guidance is explicitly focused on drug development, but its principles are intentionally broad and collaborative. They invite extension, adaptation, and harmonization across jurisdictions and evidence domains.

For HEOR, this is an opportunity. By aligning AI methods with regulatory expectations early — rather than waiting for explicit HTA-specific rules — the field can help shape how agencies evaluate AI-enabled evidence. In doing so, HEOR can move from being a passive recipient of regulation to an active contributor to responsible AI adoption.

 

Looking ahead

AI will not replace HEOR expertise — but it will increasingly shape how evidence is generated, synthesized, and interpreted. These guiding principles offer a shared language to discuss trust, risk, and value. If agencies apply similar thinking to HEOR, we may finally see a path toward consistent, transparent, and confident use of AI in reimbursement and access decisions.

In that sense, this guidance is not just about AI in drug development. It is about preparing the entire evidence ecosystem — including HEOR — for a future where intelligent systems are used responsibly, transparently, and in service of better patient outcomes.

 

Interested in learning more?

Watch our recent webinar, “AI in HEOR: Case Studies on Navigating Regulatory and HTA Guidance,” on demand, featuring experts Dalia Dawoud, Manuel Cossio, Sheena Singh, and Cale Harrison:

The What, When, and Why of the Changes to NICE Methods: Is the Devil in the Details?

Following weeks of anticipation, NICE officially announced in December that the recently rumored increase of its standard cost effectiveness threshold will take effect beginning April 2026.

 

What’s changing and when?

The standard cost effectiveness threshold range that NICE committees use to judge whether a medicine is cost effective will increase by 25% from 20–30K GBP per QALY gained to 25–35K GBP per QALY gained.

NICE stated in its webinar on December 3, 2025, that the Department of Health and Social Care (DHSC) will consult on powers to direct NICE to enact this change starting April 2026, in a targeted change to regulation. This consultation opened on December 9, 2025, and will close on January 13, 2026.

NICE stressed that this targeted change will not mean any broader intervention from government ministers in its methods or decisions. It also confirmed that it is proposing to the government that the new threshold applies across all NICE guidance (Digital, HealthTech, Guidelines) and was awaiting further details. NICE also confirmed in the webinar that it was not aware of any proposals to change the thresholds used to evaluate Highly Specialized Technologies (HSTs) for ultra-rare diseases.

However, the first proposal in the DHSC consultation document refers explicitly to all NICE guidance:

“Do you agree or disagree that the power to direct NICE about the standard cost-effectiveness threshold should apply to all NICE guidance that makes recommendations on health spending? This includes technology appraisal and highly specialised technology evaluation recommendations.

As part of the timeline announced by NICE (see figure), which is subject to consultation, NICE confirmed that in early 2026 it will consult on how this change will be implemented.

 

Anticipated timeline to implement the announced changes (Source: NICE webinar on December 3, 2025)

 

In addition to an increase of its cost effectiveness threshold, NICE also announced it will start using a new EQ-5D-5L UK value set that has been developed by asking 1,200 members of the public to judge different health states and is anticipated to be published in a peer-reviewed publication by March 2026. This change, however, will follow the standard approach to making modular updates to its methods including a public consultation on the proposed change before its full implementation.

NICE’s announcement came in parallel with an announcement from the UK government about the successful closure of a trade deal with the US that includes this change, alongside an agreement regarding the tariff that UK pharmaceutical manufacturers will pay when exporting medicines to the US.

 

Why these changes?

NICE’s methods changes are anticipated to reshape the market access environment in the UK and beyond. The US-UK trade deal, of which this threshold change is part, may convince pharma companies to continue their presence in the UK and to maintain the UK’s positioning in the launch sequence after previously threatening to pull out of the UK market under pressure from the newly announced US tariffs and policies such as the MFN external reference pricing policy.

According to the UK government’s press release announcing NICE threshold changes:

“This is supported by confirmation that — thanks to strong UK support for innovation — the UK has secured mitigations under the US’ ‘Most Favoured Nation’ drug pricing initiative so that we will continue to ensure access to the latest treatments. This will encourage pharmaceutical companies from around the world to prioritise the UK for early launches of their new medicines, meaning British patients could be among the first globally to access breakthrough treatments.”

 

The anticipated impact

These NICE methods changes will have far reaching impact on the assessment of cost effectiveness of medicines in the UK, with likely spillover effects on other countries’ practices as well.

The higher WTP threshold expands headroom for treatments near previous ICER cut-offs, improving the feasibility of charging higher prices for innovative therapies. However, the unchanged discount rate limits the full advantage of this increase. This means more flexibility on price, but continued pressure on future value. It remains to be seen whether this increased threshold will also apply to other NICE guidelines apart from its technology appraisal (TA) program. What has been confirmed is that the threshold change will not lead to any reviews of completed appraisals.

NICE’s adoption of the EQ-5D-5L UK value set will also reshape patient-reported outcomes strategy. Utilities derived from EQ-5D directly influence QALY calculations and ICERs. By reflecting more nuanced health states, EQ-5D-5L supports a more accurate calculation of QALYs. Trials that currently collect EQ-5D-3L data may need a new mapping function to align with the new set. Future trials should prioritize EQ-5D-5L and ensure high completion rates for PRO instruments, as missing data will become even more critical.

From a patient perspective, this means their lived experience is better represented in HTA decisions. For pharma companies, it means interventions that improve pain, anxiety, and functional independence can show their full value in cost-effectiveness models.

 

Regional impact

It is not clear how Europe will respond to these changes on both sides of the Atlantic, but what is clear is that actions will need to be taken to minimize the impact of these changes on both the favorability of European markets as launch markets and the prices to be charged by pharma companies in these markets, both of which are likely to impact patient access to innovative medicines.

Further, we could speculate that this change could bring prices in the UK closer to France and Germany. The UK has been able to achieve low prices because of the powerful negotiating position of the UK’s single centralized payer for the majority of UK healthcare (the NHS), its deeply embedded health technology appraisal processes through NICE, which acts as the gatekeeper for the reimbursement of drugs, and through long-standing price-control mechanisms that effectively cap the NHS’s spend on innovative medicines — the most recent iteration of which is the Voluntary Scheme for Branded Medicines Pricing, Access and Growth (VPAG), and a fallback Statutory Scheme. The current VPAG scheme requires UK manufacturers to pay an effective clawback rate of 23.5% to the UK Government on “newer medicines” (22.9% clawback plus a 0.6% investment program funding, excluding new active substances) — far higher than comparators such as France (5.7%), Germany (7%), and Spain (7.5%).

 

Have you considered these and other impacts and is your team ready for these changes?

2025 in Perspective: Reflections From Our Newest Colleagues

Every year brings new faces, fresh ideas, and inspiring stories to Cytel. In 2025, these colleagues joined us from across the globe, each bringing unique experiences and ambitions. As the year closes, we asked them to share what stood out, what they’ve learned, and how they see their work shaping something bigger. Their reflections tell a story of connection, growth, and purpose.

 

Joining Cytel: Memorable moments and settling in

For Kasum de Souza Mateus (Senior Biostatistician, FSP) the most memorable part of joining Cytel was simple yet meaningful: “Being able to meet colleagues and mentors in person.” That feeling of connection resonated with many new Cytelians, from Adish Jindal (Senior Recruiter), who described the joy of reconnecting with familiar faces, to Luke Hilliard (Event Manager), who fondly recalls a team meeting: “I really enjoyed the trip to Bruges. It was such a pleasure meeting everyone in person. We came away with some fantastic ideas that we’ve since put into action for our events.”

Others found their defining moments and success in challenges that brought people together. Kanchan Kulkarni (Manager, Accounting) stepped into her role during a major system transition: “One of my most memorable experiences has been leading the Global GL Accounting function across EMEA, APAC, and NA regions during our Oracle ERP transition. It wasn’t just about systems and numbers — it was about connecting people, aligning processes, and building something stronger together.” And for Scott Rogers (CFO), the most powerful moment came during a Town Hall: “I was very moved by the presentation where we heard directly from a patient and understood how our work helped him realize the benefits he was seeing.”

For Macarena Pazos Maidana (Senior Market & Business Development Manager) success came early: “During my third week, I successfully secured a key renewal with a major pharmaceutical client for the East Horizon™ platform. This achievement not only boosted my confidence but also reinforced my belief in the value our solutions bring to the industry.” And Hannes Engberg Raeder (Principal Biostatistician, FSP) found pride in collaboration: “I’m proud of having been able to support one of our partnerships through process improvements that helped strengthen collaboration and overall efficiency.”

 

Leaning on advice

Of course, starting something new means leaning on advice from colleagues or mentors, and some words of wisdom stuck. Nicole Sheridan (Manager, Talent Management) shared the famous mantra that shaped her approach: “’Do or do not, there is no try.’ It’s simple, but it completely changed how I think about my work and even life outside of work. I realized it’s not about being perfect but it’s about showing up, committing, and seeing things through. That mindset has really helped me take initiative, stay resilient, and turn ideas into results.

Damian Kowalski (Principal Statistical Programmer, FSP) emphasized collaboration: “Don’t be afraid to ask questions. Collaboration is our strength.” And Sydney Jenkins (Senior Employee Relations & Engagement Partner) shared a perspective that guides her work: “Trust your logic. That perspective reminds me to approach challenges with a clear, rational mindset, even under pressure!”

 

Growth and ambition

This year was not only about settling into their role for our new Cytelians, however. It also marked a year of growth and achievements. Adish honed his global recruitment expertise: “One skill I’m particularly proud of developing in 2025 is my ability to manage global recruitment processes more effectively.” Monica Chaudhari (Associate Director, Biostatistics, FSP) shared a technical milestone: “My first study that I got assigned to was already closed. To help myself support the team through database lock, review of final outputs and drafting of the CSR, I created a swimmers plot summarizing all important endpoints on each subject’s trajectory that helped identify major deviations.”

Valeria Duque Mora (Project Coordinator, Resource Management) reflected on teamwork: “My current team has made a real difference in my daily work. They are the foundation of our success, always supporting each other and sharing new information with kindness and collaboration throughout every process.” For Dominika Wisniewska (Senior Statistical Programmer, FSP), the impact was deeply personal: “I am grateful that Cytel gave me the opportunity to work directly for our client where I work on research within rare diseases and neurology diseases. I am particularly interested in neuro because of personal reasons, and I am happy to participate in maybe discovering new treatments.” And Sankhyajit Sengupta (Senior Statistical Programmer, FSP) embraced learning: “In this very short period of time (three months), I’ve had the opportunity to gain exposure to R programming in live studies and also completed required trainings on R, an important step as the industry is moving in this direction.”

Looking ahead, our new colleagues are already thinking about how to make an even bigger impact in 2026. Kanchan hopes to drive automation and efficiency, Luke dreams of organizing a standalone event, and Ye Miao (Associate Director, Biostatistics, FSP) plans to deepen expertise in R programming to contribute more effectively to data analysis and reporting tasks in his FSP role. Sydney aims to strengthen policy awareness and consistency across the organization, while Macarena is focused on enhancing client retention and satisfaction. Each goal reflects a commitment to making an even bigger impact in year two.

 

Connecting to the bigger picture

Every role at Cytel connects to our mission of improving patient lives. Adish summed it up well: “As a Global Senior Recruiter, I help bring in the talent that powers our mission. Every great hire strengthens our culture, drives innovation, and helps the company achieve its goals globally.” Wyatt Gotbetter (Senior Vice President, Global Head Evidence, Value and Access) described the EVA team’s role: “I like to describe the work of EVA as the essential ‘last mile’ in our client’s drug development journey — after decades of scientific discovery, animal and human trials, and regulatory approvals, we play a vital role in helping ensure patients get access to needed therapies.” And Damian reminded us of the impact behind the data: “Every dataset we program and validate helps ensure reliable insights for clinical trials. It’s amazing to know that our work plays a role in bringing life-saving therapies to patients worldwide.”

 

The voices of our newest colleagues remind us that Cytel is more than a workplace. It’s a community driven by purpose, collaboration, and innovation. Here’s to their continued success and to another year of making a difference together.

Women’s Health Is Society’s Wealth: Unlocking Economic Value When Bridging the Gender Health Gap

When the facts are loud and clear: investing in women’s health could unlock $1 trillion in global gross domestic product annually by 2040, prevent 24 million female life years lost to disability, and yield exponential returns to economy for every investment across obstetrics and gynecology, female and maternal health, immunology, neurology, cardiology, and oncology.

 

Improving global health equity has been increasingly recognized as a strategic priority for different stakeholders in healthcare,1 including policymakers, industry, governments, investors, and global health organizations. Beyond an ethical and human rights imperative, reducing health inequities and ensuring that everyone has a fair opportunity to achieve their full health potential — independent of socioeconomic status, sex, gender, geography, or race/ethnicity — leads to economic and societal benefits and resilient healthcare systems.2 Although progress continues to be made toward improving general health outcomes, Cytel researchers have found that this has not translated equally for men and women.3

 

The gender health gap: A global health crisis

Attention to women’s health inequities and the potential economic impact from closing this health gap remain largely unnoticed. The gender health gap — the long-standing, unfair differences in health outcomes between women and men — has been only recently recognized as a medical and healthcare issue. The underinvestment in female health research, the absence of systematic data collection to understand and document the unique biological needs of females and assess disparities, as well as biases in male-dominant clinical trial programs have all contributed to the neglect of women’s health issues. The survival paradox is documented, with women outliving men but experiencing poorer general health, including mental health; recent data showing that women live approximately five years longer than men does not adequately categorize the fact that women spend more than one-quarter of their lives in poor health.4, 5 This health gap is a global health crisis that affects women of all ages to varying extents depending on geography and income levels.4

The gender health gap is generally defined by the conditions that affect women uniquely, differently, or disproportionately, and are not limited to those related to sexual and reproductive health.4 For example, women from the general population are at significantly greater risk of mental health disorders (e.g., depression, suicide) than men, and women with type 2 diabetes mellitus have a disproportionally higher risk of adverse cardiac events, including mortality.6 Men, on the other hand, are significantly more likely to have adverse events after specific types of surgeries and higher mortality after COVID-19.6 Despite the fact that cardiovascular disease is the top cause of death for women in the US, males outnumber females two to one in related clinical trials.7

 

Quantifying the economic benefit of closing the women’s health gap

Quantifying the economic benefit of closing the women’s health gap for societies and economies is important for several reasons and makes visible the “invisible” topic of women and their health. By attaching a measurable economic gain — such as productivity gains, increased workforce participation, reduced healthcare spending — policymakers and investors can grasp the tangible impact on global economies and growth. As financial pressures are restraining healthcare spending, prioritization of resource allocation where interventions yield the greatest returns to economies, such as women’s health, may be placed higher in the list of investment priorities.7 Therefore, systematic efforts to quantify the economic value when bridging the gender health gap will push reframing health equity as a driver for inclusive and sustainable growth, making it a strategic imperative for governments and businesses and overturn the negligible investment in women’s health (only 5% in 2020).8

 

Our findings: The value of investment to improve women’s health

We conducted a comprehensive literature review that aimed to systematically investigate and summarize quantitative evidence on the economic impact of investments to close the women’s health gap globally. We identified robust evidence to demonstrate that when investment is made to improve women’s health, there is return to this investment by bringing back higher value to economies.

A recent report jointly published by the World Economic Forum and the McKinsey Health Institute, for example, highlights that investments in addressing the women’s health gap could not only extend life years and healthy life years, but also have the potential to boost the global economy by $1 trillion annually by 2040.4 These findings were supported by an additional impact analysis conducted by Women’s Health Access Matters across three indications: rheumatoid arthritis, coronary artery disease, and Alzheimer’s disease. Key findings indicated that an investment of $300 million in women’s health research across these three diseases would conservatively result in a $13 billion return to the US economy.7

Over the past 70 years, the influx of women into the workforce has been closely linked to economic growth.4 Since nearly half of the health burden affects women in their working years, this can have serious consequences for the income-earning potential of women, causing a ripple effect on society.4 Economic benefits in the same direction were also documented by simulation studies in other countries such as the United Kingdom whereas limited data were identified for low- and middle-income countries.

We are committed to standing at the forefront of assessing public policy trends and critical policy matters that highlight emerging challenges and seizing opportunities for improving public health. Some examples include our environmental scan of publicly available data repositories to address disparities in healthcare decision-making,9 an umbrella review of the impacts of climate change on maternal health and birth outcomes,10 and blueprints for collective action to close the women health gap.

 

 

Interested in learning more?

Grammati Sarri, Lilia Leisle, and Jeffrey M. Muir will be at the upcoming ISPOR Europe conference in Glasgow, Scotland, where they will present “The Economic Case for Gender Equity: How Closing the Women’s Health Gap Benefits Healthcare Systems and Economies” on Wednesday, November 12, 2025, from 9 to 11:30 a.m. Register below to book a meeting or visit us at Booth #1024 to connect with our experts:

Generative AI in Evidence Synthesis: Harnessing Potential with Responsibility

The integration of AI into the healthcare research landscape is accelerating, with one obvious area of application being evidence synthesis. From early scoping reviews to comprehensive systematic literature reviews (SLRs), AI promises to reduce manual burden and enhance efficiency by saving time. However, it is crucial to understand both the strengths and limitations of using AI in this broad context to ensure compliance, reliability, and scientific rigor.

 

Knowing where it works: A targeted approach

Artificial intelligence, including generative AI models, shines when used for targeted literature reviews (TLRs) or when generating summaries of scientific articles to support evidence-based decision-making at an early development stage. AI can synthesize large volumes of information quickly, offering valuable insights during exploratory or early-phase research.

However, it’s critical to distinguish these from regulatory-facing systematic literature reviews, especially those intended for payer or health technology assessment (HTA) submissions. In this context, SLR extractions have traditionally been completed by two independent human reviewers. This human oversight ensures objectivity and reproducibility, key elements of regulatory compliance.

 

Expertly trained models vs. generalist giants

The current landscape is filled with large generalist language models trained on diverse internet-scale data. While impressive, these models often exhibit hallucinations — the generation of plausible but incorrect or fabricated content — particularly in domain-specific applications like evidence synthesis.

This is why domain-trained expert models are preferred. These models are fine-tuned on biomedical and scientific corpora, ensuring higher reliability and reducing the risk of misinterpretation or erroneous conclusions. They understand field-specific terminology, data structures, and compliance requirements far better than their generalist counterparts.

 

The imperative of data traceability

In evidence synthesis, transparency is non-negotiable. Any AI-generated output must allow users to:

  • Highlight the exact source (i.e., sentence or section) of the original scientific article from which a conclusion or data point was extracted.
  • Compare the model’s interpretation with the source text to identify discrepancies or nuances that could affect meaning or validity.

Using structured tags to annotate key terms, qualifiers, and relationships can make these comparisons clearer and more systematic but also inform advanced search and retrieval activities. By surfacing subtle differences, tagging supports expert review, preserves contextual integrity, and strengthens the reliability and defensibility of the synthesized evidence.

 

Measuring what matters: Precision and beyond

Traditional evaluation metrics like precision, recall, and F1 score (the harmonic mean of precision and recall) remain foundational when assessing AI model performance in literature screening and data extraction.

But in generative contexts — where the task may be summarization, paraphrasing, or abstract reasoning — additional measures become valuable:

  • Answer correctness: Does the output convey a factual, verifiable point?
  • Semantic similarity: How closely does the AI output align in meaning with the ground truth?
  • BLEU, ROUGE, and BERTScore: These Natural Language Processing metrics offer quantitative insights into the quality of generated text, especially for summarization and content generation tasks.

Selecting the right mix of these metrics provides a comprehensive view of model performance and reliability.

 

Where AI makes a difference: Screening and beyond

One of the most promising applications of generative AI in evidence synthesis is in literature screening, or the ability to assess whether a publication (abstract or full text) meets the criteria for inclusion. Studies and pilot implementations suggest that AI can reduce screening time by up to 40%, making it a powerful ally for research teams.

AI tools have been leveraged to assign a probability of inclusion to a title or abstract or full text to guide the screening process but also to allow researchers to quickly understand the impact of modifying search strategies on yield. By automating this repetitive and time-consuming phase, organizations can reallocate expert human resources to higher-value tasks, such as:

  • Resolving ambiguous or context-dependent data extractions
  • Validating nuanced findings and offering insights into implications of these findings
  • Ensuring alignment with HTA submission standards

In this way, AI doesn’t replace human reviewers but augments them, driving efficiency without compromising accuracy.

 

AI with guardrails

Generative AI is reshaping the landscape of evidence synthesis, but its integration must be strategic, measured, and compliant. By combining domain-trained models, robust traceability, appropriate evaluation metrics, and human oversight, organizations can unlock the true value of AI — accelerating workflows without sacrificing quality or compliance.

When used thoughtfully, generative AI becomes more than just a tool — it becomes a partner in advancing scientific research.

 

Meet with us at ISPOR 2025!

Manuel Cossio and Nathalie Horowicz-Mehler will be in Glasgow for ISPOR Europe 2025! Click the link below to book a meeting, or stop by Booth #1024 to connect with our experts:

Breaking Barriers in Rare Disease Research with Generative AI and Synthetic Data

In healthcare innovation, one of the most pressing challenges lies in rare disease research. There are approximately 7,000 rare diseases affecting over 300 million people worldwide. With only a handful of patients dispersed globally, gathering sufficient data to power robust clinical studies or predictive models is a monumental hurdle. However, a solution is emerging at the intersection of generative AI and real-world data (RWD) — a novel approach with the potential to reshape possibilities and unlock insights to address unmet medical needs in rare diseases.

 

The rare disease data dilemma

In the U.S., rare diseases are defined as conditions affecting fewer than 200,000 people. Despite their low individual prevalence, rare diseases collectively impose a significant burden on both patients and healthcare systems.

Research and development in rare diseases often face a vicious cycle: low prevalence leads to data scarcity. Traditional clinical trials are often infeasible and/or statistically underpowered due to the limited pool of participants.

Meanwhile, RWD sources such as electronic health records (EHRs), insurance claims, registries, and patient-reported outcomes offer valuable, albeit messy and fragmented, glimpses into the patient journey. Yet even RWD struggles to paint a complete picture in rare diseases. This is where generative AI steps in.

 

Enter generative AI: Making data where there is none

Generative AI — especially models like Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and, more recently, large foundation models — has a transformative ability: it can learn patterns from limited datasets and generate synthetic yet realistic datasets.

How it works

  1. Learning from RWD: Even small datasets from rare disease patients can be used to train and fine-tune generative models. These models identify patterns, distributions, and time-dependent relationships present in the data.
  2. Synthesizing patients: Once trained, the model can create new, synthetic patient records that preserve the statistical properties and characteristics of the original data. These “digital patients” simulate disease progression, treatment responses, and comorbidities.
  3. Validating realism: Synthetic data must be validated to ensure it reflects the real-world data it was trained on. Techniques like distributional comparison, propensity scoring, and expert validation are used to ensure accuracy and utility.

 

Why synthetic data matters for rare diseases

Synthetic data can enhance rare disease clinical research in many ways, including:

 

1. Augmenting small cohorts

Synthetic data can boost sample sizes for rare disease studies, enabling:

  • Simulation of clinical trials
  • Development of more robust predictive models
  • Generation of synthetic control arms where traditional controls are ethically or logistically impractical

 

2. Enhancing privacy

In rare diseases, patient re-identification is an increased risk due to unique phenotypes or genetic markers. Synthetic data protects patient privacy, while at the same time preserves the utility of the data.

 

3. Facilitating global collaboration

As synthetic data is deidentified, it facilitates data sharing among researchers, institutions and borders, minimizing regulatory hurdles and fostering cross-collaborative discovery.

 

4. Accelerating drug development

Pharma and biotech companies can use synthetic data to:

  • Test drug targeting strategies
  • Model long-term outcomes
  • Conduct in silico trials in the earliest stages of development

 

Challenges and considerations

While promising, this approach is not without its challenges:

  • Bias amplification: Synthetic data reflects the biases of its training data. If the RWD is incomplete or skewed, so will the synthetic outputs be. Strategies to handle bias are essential.
  • Regulatory acceptance: Regulatory bodies are still evaluating how to incorporate synthetic data into approval pathways.
  • Validation standards: There is a need for consistent benchmarks and best practices for validating synthetic data — both in terms of privacy and utility, as well as broader generative AI applications in healthcare.

 

Looking ahead

The marriage of generative AI and RWD opens new doors for rare disease research. With the ability to synthesize patient data that preserves real-world complexity, we can begin to break free from the constraints of scarcity — generating insights, hypotheses, and interventions that were once out of reach.

As we move forward, interdisciplinary collaboration among clinicians, data scientists, regulatory bodies, and patient advocacy groups will be key to harnessing this potential ethically and effectively.

 

Interested in learning more?

Download our complimentary ebook, Rare Disease Clinical Trials: Design Strategies and Regulatory Considerations:

External Control Arms: A Powerful Tool for Oncology and Rare Disease Research

In clinical research, the randomized controlled trial (RCT) has been considered the gold standard. Yet in many areas — especially in oncology and rare diseases — running an RCT with a balanced control arm is not always possible. Patients, physicians, and regulators often face a difficult reality: how do we evaluate promising new therapies when traditional designs aren’t feasible?

This is where external control arms (ECAs) come into play. By carefully drawing on existing data sources and applying rigorous methodology, ECAs can help provide the context and comparative evidence needed to make better decisions.

Here, we will explore why ECAs are particularly valuable in oncology and rare diseases, how they support decision-making and study design, what data sources they can rely on, and which statistical methods are essential to reduce bias. We will also introduce the concept of quantitative bias analysis and conclude with why experienced statisticians are key to the success of this methodology.

 

Why external control arms matter in oncology and rare diseases

Oncology and rare disease research share several challenges that make traditional RCTs difficult:

  • Small patient populations: In rare diseases, the number of eligible patients is often extremely limited. Asking half of them to enroll in a control arm may make recruitment impossible.
  • High unmet need: In oncology, patients and families are eager for new options. Many consider it unacceptable to randomize patients to placebo or outdated standards of care.
  • Ethical constraints: For life-threatening conditions, denying patients access to an experimental therapy can be ethically challenging.
  • Rapidly changing standards of care: In oncology, new treatments are approved frequently. A control arm that was relevant when a trial began may become outdated by the time results are available.

In such contexts, single-arm studies (where all patients receive the experimental therapy) are common. But single-arm results alone are not sufficient. Without a comparator, how do we know if the observed survival or response rate truly reflects an advance? ECAs provide the missing context.

Even when a trial includes a control arm, unbalanced designs — such as smaller control groups or cross-over to experimental treatment — can limit the ability to make clean comparisons. External controls can augment these designs, helping to stabilize estimates and provide reassurance that results are robust.

 

Supporting internal and regulatory decision-making

ECAs serve multiple purposes:

  1. Internal decision-making:
    • Companies developing new therapies must decide whether to advance to the next trial phase, expand into new indications, or pursue partnerships.
    • ECAs help answer questions like: Is the observed benefit large enough compared to historical data? Do safety signals look acceptable in context?
  2. Regulatory decision-making:
    • Regulatory agencies such as FDA and EMA increasingly accept ECAs as part of submissions, especially in rare diseases and oncology.
    • While not a replacement for RCTs, ECAs can strengthen the evidence package and demonstrate comparative effectiveness in situations where randomization is not feasible.
  3. Helping the medical community:
    • Physicians, payers, and patients need to interpret trial results. An overall survival rate of 18 months in a single-arm study may sound promising, but how does it compare to similar patients receiving standard of care?
    • ECAs help put numbers into perspective, allowing the community to better understand the true value of a new therapy.

 

Designing better studies with ECAs

External controls are not only a tool for analyzing results — they can also improve study design.

  • Feasibility assessments: By examining real-world data or prior trial results, sponsors can estimate expected event rates, patient characteristics, and recruitment timelines. This reduces the risk of under- or over-powered studies.
  • Endpoint selection: Understanding how endpoints behave in historical or real-world settings helps refine choices for the trial, ensuring relevance to both regulators and clinicians.
  • Eligibility criteria: RWD and earlier trial data can reveal which inclusion/exclusion criteria are overly restrictive. Adjusting them can broaden access while maintaining scientific rigor.
  • Sample size planning: By leveraging ECAs, trialists may reduce the number of patients required for an internal control arm, easing recruitment in small populations.

In other words, ECAs can shape trials from the start, rather than being seen only as a “rescue” option after the fact.

 

Sources of external control data

An ECA is only as good as the data it relies on. Broadly, there are three main sources:

  1. Other clinical trials:
    • Prior trials of standard of care treatments can serve as external comparators.
    • Individual patient-level data (IPD) is preferred, but often only summary data is available.
    • These data are typically high quality but may not perfectly match the new study population.
  2. Published studies:
    • Systematic reviews and meta-analyses of the literature can provide comparator data.
    • Useful when IPD is unavailable but limited by reporting standards and heterogeneity across studies.
  3. Real-world data (RWD):
    • Sources include electronic health records, registries, and insurance claims databases.
    • These capture routine clinical practice, reflecting the diversity of real patients.
    • However, RWD often suffers from missing data, variable quality, and lack of standardized endpoints.

Each source has strengths and weaknesses. Often, the best approach is to triangulate across multiple sources, ensuring that conclusions do not rest on a single dataset.

 

The value of earlier clinical trials

Earlier-phase trials (Phase I and II) can be particularly valuable in constructing ECAs. These studies often include control arms, detailed eligibility criteria, and well-captured endpoints.

For rare diseases and oncology, earlier trials may be the only available benchmark. By carefully aligning populations and endpoints, statisticians can extract maximum value from these datasets.

The challenge, of course, is ensuring comparability. Patient populations may differ in prognostic factors, supportive care practices may evolve, and definitions of endpoints may shift over time.

This is where advanced statistical methods become essential.

 

Reducing bias with propensity scoring

One of the key criticisms of ECAs is the risk of bias. Without randomization, patients receiving the experimental therapy may differ systematically from those in the external control.

Propensity score methods are a powerful way to reduce this bias. The idea is simple:

  • For each patient, estimate the probability (the “propensity”) of receiving the experimental treatment based on baseline characteristics.
  • Match or weight patients in the external control group so that their distribution of covariates mirrors that of the trial patients.

This approach creates a “pseudo-randomized” comparison, balancing measured variables. While it cannot eliminate unmeasured confounding, it greatly improves fairness in comparisons.

 

Quantitative bias analysis: Addressing the unmeasured

Even with careful propensity scoring, unmeasured confounding remains a concern. Clinical researchers often ask: What if there are factors we didn’t account for?

This is where quantitative bias analysis (QBA) enters. QBA does not eliminate bias but helps us understand its potential impact.

For example:

  • Analysts can model how strong an unmeasured confounder would need to be to explain away the observed treatment effect.
  • Sensitivity analyses can simulate scenarios with different assumptions about unmeasured variables.

By explicitly quantifying uncertainty, QBA provides transparency. Regulators and clinicians gain confidence that conclusions are robust — or at least, that limitations are clearly understood.

 

The need for experienced statisticians

Constructing an ECA is not a “plug-and-play” exercise. It requires expertise across multiple domains:

  • Data curation: Selecting fit-for-purpose datasets, cleaning and harmonizing variables, and aligning endpoints.
  • Study design: Defining eligibility, follow-up time, and analysis plans that minimize bias.
  • Statistical methodology: Applying techniques like propensity scoring, inverse probability weighting, Bayesian borrowing, and QBA.
  • Regulatory communication: Explaining assumptions, limitations, and sensitivity analyses in language that regulators and clinicians can understand.

In short, ECAs demand both technical skill and strategic judgment. Partnering with experienced statisticians ensures that external controls provide credible, decision-grade evidence rather than misleading comparisons.

 

Final takeaways

External control arms are rapidly becoming an indispensable tool in modern clinical research — especially in oncology and rare diseases, where traditional RCTs often fall short.

They offer:

  • Context for single-arm studies and unbalanced designs.
  • Support for both internal and regulatory decisions.
  • Guidance in study design and feasibility planning.

By leveraging diverse data sources — from earlier trials to real-world evidence — and applying rigorous methods such as propensity scoring and quantitative bias analysis, ECAs can bring clarity and credibility to difficult development programs.

But the value of ECAs depends on how well they are planned and implemented. Done poorly, they risk misleading decisions. Done well, they empower researchers, regulators, and clinicians to make better choices for patients.

As the field evolves, one thing is clear: the expertise of skilled statisticians is the cornerstone of successful ECAs.

 

Interested in learning more?

Join Alexander Schacht, Steven Ting, and Vahe Asvatourian for their upcoming webinar, “Beyond the Standard Clinical Trial in Early Development: When and Why to Consider External Controls” on Thursday, October 16 at 10 a.m. ET:

Addressing Evidentiary Gaps with Advanced Quantitative Methods

By Hoora Moradian, Victor Laliman-Khara, Peter Wigfield, Michael Dolph, and Michael Groff

 

Global health technology assessment (HTA) bodies are setting higher standards for rigorous evidence to support access decisions. In this evolving landscape, generating meaningful health economic (HE) models and indirect treatment comparison (ITC) analysis is critical — particularly in rare and chronic disease settings. However, traditional modeling techniques often fall short, prompting the need for more advanced and adaptable approaches.

Here, we discuss finding the right method for various market access scenarios, given your indication, patient population characteristics, and data gaps.

 

Addressing heterogeneity across multiple studies

Victor Laliman-Khara

Problem: A sponsor has an asset that is entering a crowded market (3rd or 4th entrant), with a significant number of comparators and a related shift in standard of care. While network meta-analyses (NMA) are broadly accepted, one of the challenges is the limited possibility to adjust for between-study heterogeneity. While Matching-Adjusted Indirect Comparison (MAIC) and Simulated Treatment Comparison (STC) exist, they only allow for pairwise comparison, limiting comparative effectiveness to two treatments.

Solution: Population-adjusted indirect comparisons (PAICs) at the aggregate level — such as Multilevel Network Meta-Regression (ML-NMR) and Network Meta-Interpolation (NMI) — are increasingly used to address heterogeneity across multiple studies. These methods are particularly well-suited for situations where population adjustments are necessary across a network of trials, and where standard NMA falls short in accounting for treatment effect modifiers and between-study differences. Unlike traditional approaches such as MAIC or STC, which are limited to pairwise comparisons, ML-NMR and NMI extend the capability to more complex networks, enabling more robust and generalizable estimates.

 

Accurately reflecting treatment sequencing

Michael Groff

Problem: Many clients struggle to accurately reflect treatment sequencing in their decision problem modeling — especially in indications such as chronic inflammation with multiple therapeutic options. Capturing sequences can make it difficult to fully exploit the value proposition of a product, and poor structuring often leads to models that oversimplify the clinical pathway, become overly complex, or drive runaway costs that make the intervention appear less favorable.

Solution: Apply a structured, evidence-based methodology to transparently map and model clinical sequences, ensuring all clinically relevant transitions, such as disease progression, treatment pathways, and health state changes, are incorporated without compromising credibility, usability, or cost realism.

 

Handling pairwise comparisons in rare disease indications

Hoora Moradian

Problem: Assets in an indication where the referent trial and comparator trial have sparse data and/or heterogeneous populations present a challenge, especially in pairwise comparisons. This is common in rare disease indications, where standard PAIC approaches often struggle to produce reliable insights.

Solution: Apply innovative strategies for pairwise comparisons when traditional methods fall short:

  • G-Computation, a flexible and well-established method, can be applied when there is poor covariate overlap between study populations.
  • MAIC with random forest-based weighting can be used when the overlap is poor and the sample size is very small, delivering improved robustness over traditional methods.

 

Accounting for health equity

Mike Dolph

Problem: Accounting for health inequity is growing in importance in certain areas with a public health focus such as vaccination programs.

Solution: As health systems increasingly prioritize equity-informed decision-making, distributional cost-effectiveness analysis (DCEA) has emerged as an important extension of the traditional cost-effectiveness analysis, reflecting a growing recognition that efficiency alone is not enough to guide healthcare decisions. This approach allows decision-makers to quantify and weigh the trade-offs between improving overall health and reducing health inequalities.

 

Managing immature survival data

Peter Wigfield

Problem: Survival data, particularly in early-stage diseases, are often immature, and the emergence of more effective treatments is likely to prolong this trend.

Solution: State transition models (especially multistate frameworks) offer benefits that are difficult to achieve with conventional partitioned survival methods alone. With real-world evidence evolving rapidly, the relative ease of incorporating external evidence sources is a significant advantage, enabling the real-world value of a product to be realized. However, multiple limitations (particularly related to indirect treatment comparisons and model fits) also need to be considered.

 

Final takeaways

At Cytel, we specialize in solving complex evidence challenges, whether it’s navigating complex markets, rare diseases, or evolving treatment pathways. Our advanced methodologies, including ML-NMR, NMI, structured sequencing frameworks, equity-informed modeling (DCEA), and multistate survival models, empower sponsors to generate credible, realistic, and actionable insights. These tools go beyond traditional approaches to support better decisions and stronger value demonstration.

 

Interested in learning more?

Watch the authors’ recent webinar, “Leveraging the Right Advanced Quantitative Methods to Address Evidentiary Gaps,” on demand:

Breathing Easier: How Wearables Are Revolutionizing Patient-Reported Outcomes in Respiratory Disease

The rise of wearable technology is transforming how clinicians track chronic respiratory diseases like asthma and COPD (chronic obstructive pulmonary disease). Traditionally, managing these conditions has relied heavily on intermittent clinic visits and subjective symptom reports. But what if we could continuously monitor how patients breathe, move, and feel — right from their homes?

Enter wearables: smart devices that collect real-time physiological and behavioral data. These devices typically work in tandem with smartphone apps that prompt patients to complete patient-reported outcome (PRO) measures — allowing for integrated, real-time tracking of a full range of patient-relevant outcomes. When combined, these tools offer a powerful new lens for respiratory health.

 

Why PROs matter in respiratory disease

PROs are essential for understanding the true impact of respiratory disease on daily life. PRO measures like the Asthma Control Test (ACT), COPD Assessment Test (CAT), and modified Medical Research Council (mMRC) Dyspnea Scale help patients communicate their symptoms and limitations. Yet, these snapshots — typically completed during in-clinic visits — often miss the nuances of fluctuating symptoms and the effects of lifestyle or environment.

This is where wearables shine: they offer objective, continuous, real-world data that can complement traditional PROs — typically administered in-clinic on paper or electronically — by adding daily context and physiological insight to self-reported symptoms. By enabling patients to complete PRO measures remotely, often via smartphone apps, paired with real-time wearable data, we gain a fuller, more continuous picture of their health and functioning.

 

What wearables can measure

Modern wearables can track a range of data relevant to respiratory care, including:

  • Physical activity (steps, walking time, exertion)
  • Heart rate and heart rate variability
  • Respiratory rate and breathing patterns
  • Sleep quality and disruptions
  • Environmental exposures (via linked apps or sensors)

While wearables provide continuous physiological data, PROs are typically captured via separate smartphone apps or digital platforms, where patients log symptoms, functioning, or side effects on a scheduled or event-triggered basis.

When patients report increased fatigue or shortness of breath, wearables can confirm whether activity levels dropped, sleep was disrupted, or physiologic stress markers changed — giving clinicians a fuller picture of disease impact and progression.

 

Applications in COPD and asthma

One of the most promising areas for wearables in respiratory care is pulmonary rehabilitation (PR). PR is a cornerstone therapy for COPD and increasingly recommended for severe asthma. However, adherence and engagement outside clinical settings can be challenging.

Wearables like Fitbit or Garmin devices are being used in PR programs to:

  • Monitor daily activity levels
  • Set and track exercise goals
  • Deliver motivational feedback
  • Correlate physical activity trends with PROs such as dyspnea and fatigue

Recent studies suggest that integrating wearables into PR not only boosts patient motivation but also correlates with improved self-reported symptoms and quality of life.

Another area of growth is early detection of exacerbations. New wearable patches and multi-sensor systems can detect subtle changes in respiratory rate, coughing, or oxygen saturation — sometimes days before a patient would seek help. When combined with self-reported symptoms like increased breathlessness or wheezing, these alerts could trigger early intervention and reduce hospitalizations.

 

Case in point: A digital lifeline for COPD patients

In one pilot program, COPD patients were equipped with a wearable sensor that tracked activity, respiratory patterns, and heart rate. They also submitted weekly symptom reports via an app. When wearable data indicated decreased activity and rising respiratory rate, and the patient-reported worsening breathlessness, clinicians were alerted and could intervene early — often adjusting treatment or scheduling a check-in before an exacerbation worsened.

This “digital safety net” approach is gaining traction as a way to personalize care and improve outcomes, especially in vulnerable or remote populations.

 

Challenges to widespread use

Despite their promise, wearables in respiratory care face several hurdles:

  • Data integration: Many devices still don’t seamlessly connect with electronic health records (EHRs).
  • Clinical validation: While feasibility is proven, more large-scale trials are needed to show that wearable-enhanced PRO monitoring improves long-term outcomes.
  • Implementation: Providers may require training in how to teach their patients to utilize wearables and the associated smartphone apps that collect PRO data, meaning that time spent on these activities should be considered billable.
  • Equity and access: Not all patients have smartphones, internet access, or feel comfortable using digital devices — particularly older adults, those in underserved or rural communities, and individuals facing technological or connectivity barriers.
  • Privacy and regulation: Health data from consumer-grade devices must be handled securely, and many wearables are not yet classified as medical devices.

 

The road ahead

With increasing support from healthcare systems, regulators, and tech companies, the future looks bright for wearable-assisted respiratory care. Remote patient monitoring is now reimbursable in countries like the U.S., and smart integration with PRO tools is making these technologies more usable and impactful.

As clinicians and researchers continue to validate these tools, we can expect wearables — and the PRO data they pair with — to become a routine part of respiratory disease management. Smartphone apps are now central to this ecosystem, not just for data capture but for delivering care.

Trustworthy AI in Action: Predicting Stroke Risk Transparently with Claims-Based Machine Learning

In recent years, deep learning and large neural networks have garnered most of the attention in the machine learning (ML) community. Their ability to model complex, high-dimensional data is indeed impressive. But in healthcare — where decisions can have serious consequences and interpretability is paramount — simpler, transparent models like logistic regression still have an important role to play.

Not every problem requires a black box. When it comes to predicting disease risk using structured data, such as insurance claims, traditional models can offer accuracy and insight.

 

Claims databases: An untapped resource for disease risk prediction

Claims databases are an increasingly valuable source of real-world data (RWD). Unlike clinical trial data, which is highly controlled but limited in scale and scope, administrative claims datasets cover millions of lives over multiple years, reflecting real patient behavior and care patterns.

These databases include information on diagnoses, procedures, prescriptions, and demographics — elements that, while lacking granular clinical detail, can still reveal important patterns in disease progression and risk. The scale of these datasets allows for robust statistical modeling, even for rare outcomes.

 

The case for explainable machine learning in claims-based risk prediction

When working with claims data, models like logistic regression, Lasso, or Ridge regression are not just sufficient — they are often ideal. These models:

  • Produce coefficients that quantify the relationship between features and outcomes.
  • Allow for transparent understanding of why a prediction was made.
  • Are easier to validate and communicate to clinicians, payers, and regulators.

In contrast, deep learning models often deliver slightly higher accuracy at the cost of interpretability — a trade-off that may not be acceptable in regulated healthcare environments.

 

A real-world example: Predicting stroke risk with claims data

In a recent study, Cytel used data from over 2.5 million insured individuals to predict the risk of stroke hospitalization. Using only claims-based features such as age, medication use, comorbidities (e.g., diabetes, hypertension), and health service utilization, we compared the performance of several models, including:

  • Logistic Regression
  • Regularized linear models (Lasso and Ridge)
  • XGBoost (a state-of-the-art ML algorithm)

The results? All models achieved similar predictive performance, with area under the ROC curve (AUC) values around 0.81. Logistic regression — simple, explainable, and well-established — performed on par with XGBoost, demonstrating that advanced complexity wasn’t necessary to achieve meaningful predictive power.

 

Transparency enables trust and action

What sets models like logistic regression apart is their explainability. Stakeholders can see precisely how risk factors like atrial fibrillation, hypercholesterolemia, or age contribute to predicted stroke risk. This level of clarity is essential not only for clinicians making decisions, but also for data governance, compliance, and patient communication.

In a time when “black box” AI models are under increasing scrutiny, explainable models offer a pragmatic path forward — especially when paired with large-scale real-world datasets like claims data.

 

Keep it simple, keep it transparent

Healthcare doesn’t just need powerful algorithms — it needs trustworthy ones. As our study shows, standard machine learning models remain highly relevant, especially when applied to well-structured real-world data. Claims databases, in particular, offer a rich foundation for developing these models and making preventive healthcare smarter, earlier, and more accessible.