Building the Future of Evidence Generation

Drug development has more data than ever, but more data alone does not lead to better evidence.

Clinical records, genomic sequences, imaging repositories and real-world treatment histories span health systems across dozens of countries. AI can interrogate these sources at a speed and scale no research team could match. The real test is knowing which data is fit for the question, the decision and the outcome a programme is trying to achieve.

That challenge sits at the centre of how Mukhtar Ahmed guides BC Platforms.”The fundamental difference is getting to the right data to answer the question,” he says. “It’s not about going out to hunt for data; it’s about understanding the core problem, defining the research thesis and then finding the exact data needed to answer it.”

That discipline of starting with the research question anchors BC Platforms’ approach. It helps life sciences organisations define the evidence required for an objective, then connect to distributed datasets capable of providing it. These sources rarely share a common format or governance model. BC Platforms sources and harmonises data, preserves local controls and enables highly sophisticated analysis without requiring institutions to relinquish ownership of sensitive patient information.

For more than 25 years, the company has supported secure data access and an evidence architecture designed around the desired outcome, with the data, analytics and governance model shaped accordingly.

From Research Question to Research-Ready Evidence

What evidence is needed to move a candidate from phase two to phase three? What will support market access after approval? What does effective safety surveillance require?

Each question calls for a different evidence pathway, patient population, source mix and analytical approach. Reaching the right answer means finding relevant data, preserving its context and shaping it into evidence fit for the decision.

For pharmaceutical companies, this requires a more deliberate portfolio-level view: starting with who the research is intended to serve, where the patients and data are located and what outcome the programme needs to achieve. The goal is not simply cost reduction, but speed to outcome.

Much of that information sits with hospitals, biobanks, diagnostic centres and other healthcare organisations. These custodians manage patient data under strict institutional, privacy and residency requirements.

BC Platforms uses federation to make that information available for research without centralising sensitive patient data or moving it outside custodian controls.

“To conduct meaningful research, you have to get to the source of the data,” explains Ahmed. “That invariably means working closely with global data custodians. What federation allows you to do is access that data across countries and organisational units, without the data ever leaving the custodian’s boundaries.”

That model allows analysis to extend across organisations and jurisdictions while each custodian retains control of its information. BC Platforms supports this through technology deployed across participating sites, aligning research access with privacy, governance and residency requirements.

Access, however, is only the beginning. Evidence may need to draw on clinical data from hospitals, biobanks, diagnostic centres and healthcare systems, then combine it with public sources, sponsor-provided datasets and consumer, social or behavioural data. BC Platforms brings these sources together through capabilities that harmonise and integrate the data, apply the right ontologies and metadata models, support federated analysis and create secure environments for patient-level research.

Together, these capabilities support the shift from collecting more data to using the right data to inform the decisions that matter across discovery, clinical development, regulatory, market access and post-approval needs.

This connected architecture is designed for evidence generation that is continuous rather than confined to a single study. As new patients enter health systems, diagnoses change, treatments progress and follow-up records accumulate, evidence can be refreshed, reused and extended to support future research and decision-making.

Curation, quality control, harmonisation and planned data refreshes therefore form part of the same operating model. They allow researchers to update cohorts and analyses as new information becomes available, preserving value beyond a single study, decision point or research question.

“We have built strength in areas such as oncology and rare disease, where longitudinal, multimodal evidence is critical to understanding complex patient populations,” says Ahmed. “While drug research remains our primary focus, the same evidence infrastructure can also support adjacent areas where secure access to high-quality health data is essential.”

Putting the Model to Work

A global biopharmaceutical company developing a biomarker-targeted therapy for non-small cell lung cancer partnered with BC Platforms to generate real-world evidence across the UK, Germany and Spain. The task was to define biomarker-specific patient populations across markets where clinical documentation and testing practices differed and where the evidence had to meet HTA and EU Joint Clinical Assessment requirements.

BC Platforms designed a longitudinal cohort study spanning 1,200 patients across 12 data partners, harmonising EHR, laboratory and pathology data within a secure trusted research environment. The cohort is designed for planned data refreshes, creating a reusable evidence foundation for regulatory engagement, HTA readiness and launch planning.

In Japan, BC Platforms partnered with NTT Life Science Corporation to launch the Japan Precision Medicine Platform, a national infrastructure designed to support secure access, sharing and analysis of clinical data across the country. It enables research across organisations in areas ranging from oncology to rare disease and provides a foundation for precision medicine collaboration.

The Next Architecture

While AI expands the horizons of evidence generation, it also drastically raises the stakes for data quality.

“If you feed the wrong data into a model, your output will naturally be flawed. It really comes back to defining the use cases you are trying to solve and then ensuring your approach, using AI, the right foundation models, the right toolkits and building the right ontologies, is aligned with that,” says Ahmed.

Ahmed sees the next stage of evidence generation taking shape around an analytical nucleus: models trained on relevant data and metadata within therapeutic areas, capable of answering recurring research questions without having to rebuild the work from scratch.

“In five years, it will all be about the data, not just the source data, but the metadata,” he says. “Whoever can assemble that analytical backbone, that core nucleus, will be in a very strong position.”

He extends that logic beyond addressing research problems to anticipating them earlier in the innovation cycle.

“You have to understand what you are trying to achieve and then get the right tools to address it,” says Ahmed. “The goal is to help prevent issues before they occur, which means being actively involved in the innovation cycle.”

That is the position BC Platforms has built. It is a position recognised by Pharma Tech Outlook Europe as Advanced Drug Development Technology of the Year in Europe 2026.

Instead of treating data as something to buy, store and hope will answer every question, BC Platforms provides the connected evidence architecture for an AI-enabled future: one that starts with the problem, then brings together distributed health data, analytics, models and context to support faster innovation and better patient outcomes.