Notice: Test mode is enabled. While in test mode no live donations are processed.
QuantHealth has raised $45 million in Series B funding to expand its artificial intelligence platform for simulating clinical trials before patients are enrolled. The investment represents another important milestone in the movement toward simulation-first clinical development, where pharmaceutical and biotechnology companies use predictive models to test trial designs, identify risks and evaluate potential outcomes before committing substantial capital to a real-world study.
The funding round was led by Qumra Capital, with participation from Pitango HealthTech, Sanofi Ventures, Artofin Venture Capital Fund, Bertelsmann Healthcare Investments, GC Ventures, NewHealth Ventures, Shoni Top Ventures and Esplanade Ventures. According to the company’s [official Series B announcement] the capital will support research and development, larger datasets, additional scientific validation, workforce expansion and new products covering both clinical development and commercial planning.
The announcement follows reported eightfold sales growth during 2025 and growing adoption among major pharmaceutical companies. QuantHealth says it now works with 12 of the world’s 20 largest pharmaceutical companies, signalling that AI-driven clinical trial simulation is moving beyond experimental pilot projects and into enterprise-level drug-development strategy.
Clinical trials are traditionally designed using evidence from earlier studies, published research, epidemiological data, clinical expertise and assumptions about how selected patient groups may respond to an investigational treatment. Even with extensive preparation, sponsors may discover serious weaknesses only after recruitment has begun.
QuantHealth aims to move some of that learning to an earlier stage.
Its platform combines clinical, biomedical, epidemiological, pharmacological and real-world patient information to create virtual representations of possible trial participants. Research teams can then simulate different versions of a proposed trial by changing variables such as eligibility criteria, treatment arms, endpoints, patient characteristics, dosing strategies and study duration.
The company’s [clinical simulation platform] is designed to support protocol optimization, indication selection, enrollment prediction, market forecasting, asset valuation and portfolio prioritization. QuantHealth says its data resources cover approximately 350 million lives, more than 100,000 drug and mechanism data points and large collections of clinical-trial evidence.
This does not mean that simulated patients can replace actual clinical-trial participants. Human trials remain necessary to establish whether a medicine is safe and effective. Instead, simulation can help researchers decide which real-world trials are most scientifically justified, how they should be structured and where major risks may exist.
As explained in [Fierce Healthcare’s report on QuantHealth’s $45 million Series B] the platform operates before a trial begins. This is important because decisions made during protocol design can influence recruitment, cost, statistical power, patient safety and the probability that a trial will produce meaningful evidence.
Drug development is expensive, uncertain and slow. QuantHealth’s funding announcement states that more than 90% of medicines entering clinical development eventually fail, with efficacy and safety problems accounting for a substantial share of those failures. While this figure comes from the company’s announcement and should be interpreted in context, the underlying problem is widely recognized: many promising drug candidates do not successfully complete clinical development.
A simulation-first approach could help companies identify avoidable weaknesses earlier.
For example, a virtual trial may suggest that the planned patient population is too broad to demonstrate a meaningful treatment effect. Researchers could then explore whether a subgroup defined by a biomarker, disease stage or previous treatment history is more likely to benefit. Simulations may also identify endpoints that are unlikely to capture clinically meaningful improvement or reveal that the planned sample size is inadequate.
Thousands of potential protocol variations can be evaluated digitally before one is selected for real-world implementation. This creates an opportunity to replace a portion of costly trial-and-error development with earlier, evidence-informed decision-making.
QuantHealth reports that it has simulated more than 600 trials and achieved predictive accuracy of up to 90% in certain applications. The company plans to expand from 30 to more than 40 indications in areas including oncology, cardiometabolic disease and inflammatory conditions. These are company-reported results rather than a universal guarantee of performance, making continued prospective validation and transparent reporting essential.
QuantHealth intends to deploy its Series B capital across three connected areas.
The company plans to develop new AI models, expand its datasets and increase scientific validation. Validation is especially important in healthcare because an apparently accurate retrospective model may perform differently when used prospectively, in a new therapeutic area or among populations that were poorly represented in its training data.
Clinical development teams need to know not only whether a model produces a prediction, but also how reliable that prediction is, what evidence supports it and where uncertainty remains.
QuantHealth plans to increase the number of diseases and clinical indications its platform can simulate. Broader coverage could make the technology relevant to more pharmaceutical pipelines, including complex areas in which patient responses differ substantially.
However, expanding across diseases also creates a higher validation burden. A model that performs well in one oncology indication cannot automatically be assumed to perform equally well in cardiometabolic disease, infectious disease or maternal health.
QuantHealth’s ambitions extend beyond protocol design. The company plans to support decisions from early clinical development through market positioning and commercialization.
This could allow drug developers to examine not only whether a study is likely to meet its endpoint, but also how a treatment may compare with future competitors, which patient groups could benefit most and whether the eventual product profile is likely to address a meaningful clinical need.
African governments and medicines regulators should begin preparing frameworks for the responsible use of AI-generated evidence in clinical development.
Regulators will need technical capacity to evaluate model documentation, data provenance, validation methods, uncertainty estimates and performance across population groups. They should also clarify when simulation results can inform protocol discussions and when conventional evidence remains mandatory.
Data-governance requirements should address where health information is stored, who can access it, how it is de-identified and whether local institutions receive meaningful scientific benefit from its use.
NGOs and philanthropic organizations should consider financing independent evaluations of clinical simulation systems in settings relevant to the Global South. Such studies should test whether predictions remain reliable across different countries, health systems and population groups.
Funders can also require projects to include African principal investigators, local data scientists, patient representatives and research institutions as genuine decision-making partners. Providing access only to a finished software product will not create sustainable capacity.
Universities and research centres should invest in clinical data science, biostatistics, pharmacology, epidemiology and responsible AI. The goal should be to enable African scientists to build, audit and govern simulation technologies—not merely provide data for systems developed elsewhere.
Commercial organizations deploying AI-driven clinical trial simulations should publish clear evidence about model performance, including unsuccessful predictions and known limitations.
They should demonstrate that their datasets are representative of the populations in which recommendations will be used. Performance should be reported separately across clinically relevant groups rather than presented only as a single global accuracy figure.
Companies should also treat affordability and post-trial access as part of development strategy. Faster trials will have limited public-health value if resulting medicines remain unavailable to the countries that contributed participants, data or research infrastructure.
AI simulations can create persuasive charts, probability estimates and virtual outcomes. That apparent precision can encourage decision-makers to place too much confidence in a model.
Every simulation depends on assumptions and underlying data. Missing information, unmeasured social factors, changing standards of care or underrepresented populations can produce misleading results.
A model may also reproduce historical inequalities. If previous clinical trials rarely included rural African communities, pregnant women, children or people with multiple health conditions, training a system on those historical trials will not automatically correct the imbalance.
For that reason, AI outputs should be reviewed by multidisciplinary teams that include clinicians, statisticians, pharmacologists, ethicists, regulators and patient representatives. Sponsors should document how the model influenced decisions and monitor whether its predictions were confirmed when the real trial was conducted.
The announcement that QuantHealth raises $45M in Series B funding is a significant vote of confidence in simulation-first clinical development. It suggests that investors and pharmaceutical companies increasingly see predictive modeling as a practical way to improve trial design, reduce expensive failures and make earlier decisions about promising therapies.
The technology could help researchers explore more options before exposing patients to risk. It could also support faster development of treatments for cancer, inflammatory disease, cardiometabolic conditions and potentially diseases that receive inadequate commercial investment.
For Africa and the wider Global South, however, the outcome will depend on implementation. AI-driven clinical trial simulations must be validated with diverse data, governed transparently and connected to stronger local research systems. Governments, NGOs, funders, universities and companies should use this moment to build African trial capacity, improve regulatory expertise and ensure that affected communities participate in shaping the evidence.
The objective should not be faster drug development alone. It should be better-designed research, safer trials and more equitable access to effective medicines.
DiMe is developing practical AI governance tools to help healthcare organisations adopt, monitor and manage artificial i
Portugal becomes the first EU state to join HealthAI, advancing safer, more transparent and globally coordinated AI regu
Discover how AI, smart yoga and genomics are making healthcare more personalised, preventive and accessible.
