How Multimodal Real-World Data is Shaping the Future of Oncology Treatment and Research

For medical practice administrators, owners, and IT managers in the United States, understanding how this data shapes cancer care is important.
Multimodal RWD includes clinical records, genomic data, pathology images, treatment outcomes, and other patient information. When all this data is combined and analyzed using advanced artificial intelligence (AI) techniques, it can help healthcare providers make better decisions.
This article looks at the role of multimodal RWD in changing oncology practices and shows how AI and workflow automation help support this change.

What is Multimodal Real-World Data (RWD)?

Multimodal RWD means different types of patient data collected from regular clinical settings. This includes electronic health records (EHRs), genomic sequencing, imaging data, pathology slides, and biochemical markers.
Unlike data from traditional clinical trials, which are controlled and limited, real-world data shows how patients actually experience treatments and their outcomes.
Caris Life Sciences, a company focused on personalized cancer care, has built one of the largest databases. It combines molecular, clinical, and claims data.
Their database has over 580,000 clinico-genomic profiles from more than 57 tumor types and more than 4.45 million digitized pathology slides.
This combined approach helps describe tumors better and gives doctors a fuller view of each patient’s condition.

The Impact of Multimodal RWD on Oncology Research

Multimodal RWD helps research aimed at customizing cancer treatments for individual patients.
Traditional clinical trials often study large groups of patients with similar conditions.
But precision oncology knows cancer is different even in the same tumor type. Patients may have different molecular causes, reactions to treatment, and outcomes.
Real-world data lets researchers study these smaller groups in more detail.
For example, Tempus is a company that uses AI to join clinical, molecular, and behavioral data. They work to find new treatment targets and find where care is missing.
More than 65% of Academic Medical Centers and over 50% of oncologists in the US use Tempus’s platform.
AI models can analyze large patient groups to find new biomarkers that show which patients will respond to certain treatments.
Tempus created an Immune Profile Score (IPS) biomarker that helps find patients who might benefit from immune checkpoint inhibitors.
This shows how multimodal data can meet clinical needs not addressed before.

Enhancing Drug Development and Clinical Trial Design

One key use of multimodal RWD is to improve how clinical trials are designed and how drugs are developed.
With advances in molecular and computational biology, oncology trials are moving away from strict drug-centered designs.
Many trials now allow multiple treatment options and change while running based on patient responses.
This method makes trials more efficient and lets more patients qualify.
People who might not have met traditional trial rules may now join if their biomarkers or genetic profiles suggest they could benefit.
Tempus has identified over 30,000 patients who might be eligible for trials using AI analysis of multimodal data.
Flatiron Health uses AI to create “digital twins”—virtual patient models that show how treatments might work.
This method improves how accurate trials are, speeds up recruitment, and cuts costs.
By mixing real-world data with traditional trial data, Flatiron helps deliver more personalized care and find new treatments.

Multi-Omics Integration and Computational Pathology

Multi-omics data includes genomics, transcriptomics, proteomics, and metabolomics, giving a deeper view of tumors.
The McGill Centre for Translational Research in Cancer collects this kind of data and mixes it with real-world patient data from over 11,000 cancer cases across 16 Canadian centers.
This provides a big data set for research focusing on precise cancer treatment.
AstraZeneca is working on using computational pathology to improve cancer diagnosis.
Their Quantitative Continuous Scoring (QCS) platform uses deep learning to study tissue samples at the single-cell level.
This helps detect biomarkers and sort patients better.
These technologies study how tumor cells interact with the immune system and other nearby cells, which is important for making combined therapies.
Also, circulating tumor DNA (ctDNA) tests help detect mutations without invasive procedures and monitor tumor changes during treatment.
This real-world method helps doctors adjust treatments quickly and improve results.

Addressing Challenges in Multimodal Data Usage

Multimodal RWD has potential but also faces challenges.
Data quality, privacy concerns, and differences in data sources are main problems for healthcare providers.
Work is ongoing to build standardized data storage and rules for data security and sharing without risking patient privacy.
Authors like K. Verkerk and E.E. Voest say that creating special RWD databases can help extend or improve drug uses and confirm biomarkers for custom therapies.
Also, combining real-world data with AI analytics can speed up government decisions and conditional drug approvals, helping patients get access to new treatments faster.

AI-Driven Automation and Workflow Optimization in Oncology Practices

AI use in oncology is not just for research and trials.
It also helps make office work easier and improves how patients are handled.
AI tools can automate tasks like patient scheduling, answering calls, and data entry.
These jobs often take a lot of time and can have mistakes.
For administrators and IT managers, AI tools such as Simbo AI’s phone automation can reduce the stress of many calls.
These tools help patients get timely replies and make scheduling more efficient.
This type of automation fits well with the growing complexity of cancer care where quick communication is very important.
AI can also help find patients who qualify for clinical trials by checking real-time EHRs and other data.
This lowers manual chart reviews and speeds up recruiting patients for trials.
In pathology labs, AI automation speeds up image analysis, quality checks, and report writing.
For example, Aignostics created AI tools that classify tumor cells versus immune cells and remove image errors.
This increases diagnostic accuracy and the number of cases handled.
These improvements shorten wait times and help doctors make decisions faster.
Using this technology also improves compliance and standardizes data, which is important for practices joining value-based care programs and research networks.

The Growing Role of Healthcare IT in Managing Multimodal Data

Handling the large and varied data from multimodal methods needs strong IT systems.
Health IT managers must make sure data is stored safely, easy to find, and can be shared across different systems.
Big oncology data sets can reach petabyte sizes, as with Tempus managing over 300 petabytes.
Handling this amount requires cloud platforms, advanced encryption, and scalable databases that follow HIPAA rules.
Standards like HL7 FHIR help systems exchange clinical and genomic data smoothly.
It is also important that staff get proper training to use these technologies.
Medical administrators may need to work with drug companies and research groups to set up data sharing agreements and support precise cancer studies using practice data.

Real-World Impact: Case Examples and Collaborations

Using multimodal real-world data has already started to affect cancer care in many US places.
Tempus works with over 200 drug companies and 65% of US Academic Medical Centers to improve patient outcomes with data-driven insights.
They recently got FDA clearance for Tempus ECG-AF, an AI tool that detects heart risk, showing AI’s growing acceptance in clinics.
The McGill Centre for Translational Research in Cancer in Canada serves as an example by combining multi-omics data with real-world clinical data to drive biomarker discovery and improve treatment plans.
Caris Life Sciences’ database helps make personalized treatment choices by linking molecular profiles with clinical outcomes.
Their work with over 849,000 cancer cases shows how large and relevant these efforts are.

Practical Considerations for Oncology Practices in the US

For medical administrators and owners, using multimodal real-world data means investing in systems that can handle complex data.
They also need to partner with tech providers who know AI and data management.
These choices affect budgets, IT staff, and training for clinicians.
Protecting healthcare data privacy is very important when working with outside AI vendors and researchers.
Clear rules about patient consent and data privacy help keep compliance and maintain patient trust.
Clinically, involving oncologists and care teams with decision tools based on multimodal data leads to a more informed and personalized way to manage patients.