As incidence rates for many common cancers continue to rise – breast, prostate, colorectal, pancreatic, and others – life sciences teams need high-quality cancer real-world data (RWD) to power their research and help develop lifesaving treatments. Building on our expertise in curating RWD for hematologic cancers, which are blood-based cancers such as multiple myeloma, COTA recently launched real-world datasets for solid tumors to help our life sciences partners accelerate studies and bring novel therapies to market, faster.
We sat down with COTA’s Chief Medical Officer and oncologist, Dr. C.K. Wang, to discuss solid-tumor RWD – and how combining it with powerful artificial intelligence (AI) models will support drug development for people with cancer. Read on to learn more.
Responses have been edited for clarity and length.
Q: How does COTA’s expertise in hematological cancer RWD translate to solid tumors?
CKW: COTA has developed a rigorous RWD abstraction methodology and quality management system for hematological malignancies, both of which translate well to solid tumors. Hematologic cancer data is particularly difficult to work with, as these cancers have their own unique, disease-specific data elements such as risk scores and response criteria. Additionally, their treatment approaches are complex. Comparatively, solid tumors are much more common and have more uniform staging approaches, better-documented outcome measures, and relatively less complicated treatment pathways. We have refined our methods through working with hematologic cancer data, and we have applied the learnings when building our solid tumor dataset.
Q: How will solid tumor RWD support life sciences teams during drug development?
CKW: Most cancer RWD today comes from the community treatment setting, where biomarker testing is less comprehensive than at academic medical centers. As a result, most real-world datasets for both solid and blood cancers lack the necessary complement of biomarker data needed to understand and develop treatments for complex disease states.
This biomarker data is crucial for life sciences companies seeking data to help unlock novel biomarkers in novel disease settings that haven’t yet been addressed. COTA’s network of both community and academic cancer centers bridges that gap, giving life sciences teams the high-quality RWD – including genomic data – they need to understand tumors’ genetic makeup and develop treatments where none exist today.
Q: As we continue to see cuts to federal cancer research funding, how can RWD and AI continue to support life sciences research?
CKW: While we can’t predict the future, it’s likely that we’ll see increased pressure on pharma, biotech, and academic research institutions to run more studies with fewer resources. Traditional clinical trials and postmarketing studies, which can cost billions of dollars and take years if not decades to complete, are likely to be unsustainable. For years, the industry has been seeking faster and better ways to get the right therapies to the right patients as quickly, safely, and cost-effectively as possible.
RWD and AI enable faster and less expensive cancer research, allowing researchers to test drug targets and identify biomarkers before taking a drug to trial. That way, they only invest in the studies most likely to succeed and have an impact on patient care. AI is accelerating research further by enabling faster data and insight generation. The process of preparing datasets for research can be time consuming, but AI can help automate aspects of data curation and abstraction. Of course, we still need human experts to validate these models, but AI can reduce the time human experts need to spend analyzing medical records to flag inaccuracies or missing data points, for example. The faster you can prepare a dataset, the sooner you can analyze it to generate new insights. In addition, AI-based analytic platforms are enabling more researchers to study RWD faster. Researchers can simply type questions into these platforms as they would a web search, where the model rapidly scans datasets to answer their query, pulling out interesting trends or novel findings.
With today’s new external pressures, we have an opportunity to advance the use of RWD and AI across the drug development lifecycle – from early discovery through post-marketing studies – to drive more efficient cancer research.
Q: In which solid-tumor areas are you seeing particularly interesting insights emerge from COTA’s RWD?
CKW: Our database shows high rates of biomarker testing for early-stage non-small cell lung cancer. As life sciences companies increasingly focus on this area, biomarker testing will be critical to tease out targeted therapies and their benefits in the non-metastatic setting.
We’re also seeing great interest in ovarian cancer. There hasn’t been much innovation in this disease in the past decades, but with a new influx of potential therapies and FDA approvals, there has been a renewed attention in the space. I look forward to seeing how our rich biomarker data can support treatment development for these patients.
Q: What unmet needs in solid-tumor cancer care can be addressed with high-quality, real-world biomarker data?
CKW: Most R&D in the solid tumor space has focused on metastatic cancers. This is certainly an area of unmet need, as most people who develop metastatic disease will succumb to their disease. By comparison, people with earlier-stage cancers have a better prognosis and higher survival rates. There’s potential to cure their disease through treatment, but the therapeutic landscape for early-stage cancers, for most diseases, hasn’t evolved much in the last decade.
COTA’s RWD is extremely valuable in the early-stage disease setting. Our rich breast cancer dataset has already validated an AI pathology solution to improve risk stratification. In early-stage lung cancer, because we see a relatively high rate of biomarker testing, our data can help drive treatment innovation to improve patient outcomes while decreasing treatment-related morbidity.
—
Want to learn more about COTA’s oncology real-world data? Reach out here to start a conversation with our team!
