Utilizing Multimodal Real-World Data to Advance Patient Care in Oncology through AI Innovations

Multimodal real-world data means collecting different types of patient information from many places. This includes clinical records from electronic health systems (EHR), genomic sequences, transcriptomic data, pathology images, lab results, and even social information about health. Unlike data from clinical trials, real-world data shows what happens to patients in normal healthcare settings. It gives a fuller picture of how diseases develop and how patients respond to treatments.
Combining these different types of data allows for more detailed study than before, when only one or two data sources were used. For example, mixing genomic sequences with pathology images and clinical results helps doctors classify cancer types better and choose the right treatments.
In the U.S., companies like Caris Life Sciences and Tempus have built huge databases with lots of multimodal cancer data. Caris offers molecular profiling like Whole Exome Sequencing (WES) and Whole Transcriptome Sequencing (WTS) for each patient they test. Their database has more than 50 petabytes of cancer data, including DNA, RNA, protein, and imaging data.
Tempus also manages a large real-world dataset, with over 8 million anonymous research records. This includes 2 million pathology images and many matched clinical and genomic data. Their platform, Tempus Lens, uses conversational AI to help researchers and doctors access this information.
These data collections are much bigger than old public resources like The Cancer Genome Atlas, by more than 60 times. This gives new chances to improve cancer diagnosis and treatment.

How AI Enables Oncology Research and Patient Care

Artificial intelligence (AI) helps manage and analyze the large amounts of multimodal data. Machine learning can find patterns that humans might miss. This helps find new biomarkers, predict how patients will respond to treatment, and match patients to clinical trials.
For example, AI models help drug companies and doctors by confirming biomarkers, defining which patients can join clinical trials, and simulating trial results. This lowers the risk of trial failure and speeds up new drug development.
Tempus works with over 200 drug companies, including most top cancer drug makers. This teamwork supports drug development and personalized medicine by mixing molecular data with real-world clinical evidence.
Caris Life Sciences and Flatiron Health have joined their data. Caris provides molecular data while Flatiron gives clinical results. This creates one of the largest multimodal datasets for cancer research. Big patient groups help answer difficult research questions and improve clinical trials.
AI also helps find biomarkers that guide treatment choices, especially in immuno-oncology. Tempus’s Immune Profile Score (IPS) is an AI tool that predicts patient response to immune therapies within about 18 months.
Doctors can use these AI tools to pick the right patients for treatments. This helps improve precision medicine and might increase patient survival and well-being.

Benefits for Medical Practices and Oncology Care Providers

• Better Patient Stratification: Different data types let doctors categorize patients into precise groups based on molecular and clinical profiles. This leads to personalized treatment plans.

• Improved Diagnostic Accuracy: AI digital pathology tools help identify tumor features faster. For example, AI can find microsatellite instability (MSI)-high status in prostate cancer, allowing for timely treatment.

• Enhanced Treatment Monitoring: Combining long-term clinical data with molecular changes allows doctors to watch disease progress and treatment success closely. They can adjust care as needed.

• Efficient Clinical Trial Enrollment: AI helps quickly find eligible patients for clinical trials. This raises enrollment rates and gives patients access to new treatments.

• Reduced Healthcare Disparities: Using social health data and real-world evidence, AI can show underserved patient groups and suggest ways to improve fairness.

These benefits are important for U.S. oncology practices because there are many cancer cases. It is necessary to optimize healthcare and control costs.

AI and Workflow Coordination in Oncology Practices

Another growing area is using AI and automation to make oncology clinic workflows better. Companies like Simbo AI create AI tools for front-office tasks such as managing calls, scheduling appointments, and engaging patients.
In busy oncology clinics, automating phone calls can lower wait times and reduce abandoned calls. This frees administrative staff to do more complex work. Since many cancer patients need frequent visits, smooth communication improves patient satisfaction and follow-up.
AI platforms like Tempus Next help care teams find care gaps, schedule screenings on time, and avoid delays in diagnosis and treatment. AI can review electronic health records and lab results to spot missed tests or referrals.
AI also aids pathology labs by automating tissue use and slide analysis. This cuts down waiting times for reports and improves test selection.
Clinics using AI workflow tools often see gains in efficiency and better patient outcomes. Care teams can focus more on medical decisions instead of paperwork.

Real-World Use Cases and Industry Perspectives

• David Spetzler, President of Caris Life Sciences, says combining molecular and clinical data helps understand patient journeys better and supports personalized cancer therapies.

• Stephanie Reisinger, Senior Vice President at Flatiron Health, calls the Caris-Flatiron partnership a platform for building large clinical-omics datasets needed for drug development.

• Susan Galbraith, Executive Vice President at AstraZeneca, notes that working with Tempus increased knowledge of tumor biology and helped clinical trials succeed more.

• Jorge DiMartino, Chief Medical Officer at Kronos Bio, praises Tempus for speeding up drug candidate development and testing drug combinations.

• Jan van de Winkel, CEO of Genmab, is hopeful about expanding partnerships with Tempus to find new cancer biomarkers and drug targets.

These views show a trend in U.S. cancer care: big data partnerships and AI are creating new ways to develop drugs and improve patient care.

Regulatory Compliance and Data Privacy Considerations

Medical administrators and IT managers in oncology must consider data governance and privacy when using multimodal real-world data and AI. Many datasets from Tempus and Caris are de-identified to keep patient privacy while helping research.
AI tools and data platforms must follow HIPAA rules and meet FDA standards for diagnostic devices and decision tools.
It is important to work with vendors who meet these standards. Many platforms include audit trails and security features to ensure responsible data use.

The Role of AI in Multi-Modal Data Fusion and Analytics

The challenge is to combine different data types to get useful results. AI uses methods like semi-supervised and weakly supervised learning to understand interactions between genomic, clinical, and imaging data.
For example, liquid biopsy results, which track tumor DNA in the blood, can be combined with patient history and imaging. This helps doctors monitor minimal residual disease (MRD) without invasive methods. It supports early relapse detection and treatment changes.
New techniques, including generative AI models like GPT-4, are also starting to be used in cancer research. These models help combine data and form hypotheses quickly, but real-world testing is still needed to confirm their usefulness.

Implications for Oncology Practice Administrators and IT Professionals in the U.S.

• Infrastructure Readiness: Data platforms need secure and scalable cloud systems that can store large genomic and imaging files.

• Interoperability: AI tools must work smoothly with existing EHR and lab systems for good workflow.

• Staff Training: Clinical and admin staff need to learn how to use AI results properly and understand their limits while keeping patients informed.

• Patient Privacy: Strong data protection and following privacy laws are key, especially with sensitive genomic data.

• Vendor Collaboration: Work with trusted AI and data partners who follow rules, are clear about their methods, and provide support.

Summary

Using multimodal real-world data together with AI is changing cancer care in the United States. Large partnerships and advanced data platforms allow better patient profiles, smoother clinical trials, and improved therapies.
Oncology leaders need to use AI workflow tools with multimodal data analysis to improve operations and patient care. There are challenges in combining data and following regulations, but companies like Tempus, Caris Life Sciences, Flatiron Health, and Proscia offer a path forward.
By managing technology use carefully and focusing on data security, medical administrators and IT staff can help their practices take advantage of these improvements. Using real-world clinical data, molecular information, and AI tools will be important for better cancer care and research across the country.