When Radiology Fails to Fix Clinical Trial Imaging Workflow at Sites

02.03.26 07:36 PM

What Most Think, Why It’s Wrong, & What Cancer Centers Really Need

Originally posted on The Cancer Letter February 27, 2026

Jeffrey Sorenson, CEO of Yunu

Executive Summary:

Radiology departments have tried for years to “own” or “fix” clinical trials imaging assessment workflows by extending clinical tools (PACS, worklists, dictation systems) and habits (speed, seamlessness, few clicks) into the research world. The intent is great: protect radiologists’ time, reduce friction, and avoid burnout. But the result is predictable: fragmented processes, delayed reads, protocol errors, endless follow-up emails, and costly rework for study teams.

Cancer centers need to stop pretending that a homegrown system with an acronym, or a bolted-on point solution, integrated to the radiology PACS, equals a “system.” What’s required is a purpose-built, end-to-end clinical trials imaging platform that lives on the web, connects investigators, coordinators, and readers, and enforces trial-specific workflows, measurements, and data structures—without forcing everything through radiology’s clinical pipeline. [1,2,6,7]

How We Got Here:
Applying Clinical Paradigms to Research Realities

Radiology is central to cancer care. Radiologists are also chronically overworked. For decades, they have focused (rightly) on speed and efficiency in their clinical workflow: A scan is received, appears on a worklist, the radiologist reads it, generates a report, and then proceeds to the following scan. Workflow optimization is about shaving seconds because volumes are relentless, staffing is short, and margins for rest are thin.

When clinical trial research grew as an additional “ask,” radiology attempted to extend the same logic: “Let’s make research feel just like clinical. Integrate it. Minimize clicks. Don’t make me leave my seat. Give me a one-and-done report.” [8,9]

Again, great intentions. Wrong domain. Clinical trials imaging is not just another flavor of clinical imaging. It’s a parallel world with:

Protocol-specific response criteria that can be modified and vary for each trial (RECIST, RANO, Lugano, irRECIST, etc.) [3,4,5]
Defined timing windows, target lesions, longitudinal comparisons, and change thresholds
Auditable source data requirements and investigator workflows [2,10]
Data handoffs to sponsors, CROs, and central review organizations
Heavy documentation and query resolution cycles, often months after the read

Trying to bend research clinical trials into a clinical PACS mold has created more friction and frustration for everyone, especially outside radiology.

What Radiology Usually Thinks Is Needed—And Why It Doesn’t Solve the Core Problems

"Never Leave the Seat”

What they think: Staying within the PACS workstation is sacred. Productivity dies when you switch systems.

Why it’s wrong: Clinical worklists are optimized for fast, unstructured, qualitative reads. Research reads are longer, quantitative, and stringently protocol-driven. The PACS can’t encode trial-specific instructions, track lesions over cycles of therapy, or generate audit-ready data. Purpose-built, web-based research platforms can—and they can be accessed anywhere. Confining research reads to the PACS is mandating the wrong tool for the job.

“Seamless Integrations Are Everything”

What they think: If we integrate our PACS and EHR to the research system, we’ll be fine.

Why it’s wrong: Integration shaves seconds, not hours. Most NCI-designated centers often do <10 research reads/day, and yet the unmanageable workflow makes this a highly disruptive process. Saving 10 seconds per read by “fewer clicks” is meaningless when each read takes 30 minutes due to manual PACS reading processes that trigger 30 more minutes of follow-ups per read to resolve issues. These occur because PACS can’t:

Track patients’ response assessments longitudinally across cycles of therapy
Pre-fetch the correct comparison source imaging data to be ready to read
Provide worklists to direct reads to trained readers on a trial-by-trial basis
Display and guide compliance with protocol criteria or align with past measurements
Perform response calculations
Drive traceable communication and collaboration directly with coordinators and investigators

Using PACS may shave seconds, but it ignores the hours of wasted time for radiologists and study teams. Integrations don’t fix the workflow gaps when it’s an entirely different way of working.

“Fewer Mouse Clicks = Better”

What they think: If we just cut clicks with AI lesion measurement or custom buttons, we’ve improved the process.

Why it’s wrong: A RECIST read has 1–5 target lesions. Click counts for these types of reads are already low; saving two clicks per lesion with AI is trivial. The real delays are that the reader:

Isn’t guided to the right slice/lesion from the last time point
Has to transpose measurements into the cancer center's source files manually
Responds to endless clarification emails because nothing is connected

Some radiology departments purchase separate tumor measurement point solution tools, but are still left with no cancer center workflows, no site-specific investigator workflows, no web access, and no protocol flexibility. Ultimately, those systems create another silo, driving more manual re-entry and more confusion.

“One-and-Done Reports”

What they think:The radiology report is the final output. Do it once, move on.
Why it’s wrong: A clinical read and a clinical trial research read are not the same report. They can (and often do) differ legitimately. Trial protocols define progression differently from clinical practice. Doing both simultaneously doesn’t improve productivity, and mixing them increases error rates.

A Better Model:

A small, trained set of radiologists (internal or external) performs research reads part-time or in reading cores.
Clinical radiologists stay focused on clinical throughput.
Research data stays protocol-compliant, traceable, and separate, without compromising patient care.

The Myth of “We Already Have a System”

Most large cancer centers claim to have a "system" (often accompanied by a catchy acronym). In reality, what's called a system is usually a combination of a PACS for image viewing, CD-ROMs or cloud drives (such as Dropbox or Box) for file transfer, Excel or paper forms for measurements and source data, Salesforce or SharePoint for task tracking, and email or phone calls for all other purposes.

This is not a system. It’s a Frankenstein patchwork of inefficient, error-prone manual processes and point solutions—accessible mostly to radiology (if anyone else at all). It’s fragile, non-cloud-native, and incapable of scaling beyond the main campus. [1]

Observed reality:

Top cancer centers either use Yunu as their end-to-end software platform or have painstakingly built internal patchwork tools that still don’t extend even to their own organizations’ satellites and affiliates. Community hospitals rely on radiology groups outside the health system, adding more complexity and a need for connected workflows. Unfortunately, many have no research system whatsoever beyond PACS and paper.

Bottom Line:

An acronym and a SharePoint site are not a workflow. Honestly, they are not worthy of their confidence-building acronyms. Success is not “We built something”; it’s “Our investigators and sponsors get accurate, timely clinical trial research reads every time, with minimal rework, minimal queries, and full auditability.”

What’s Really Needed—A Purpose-Built, End-to-End Clinical Trials Imaging Platform

Purpose-built for clinical trial imaging is defined by a few key characteristics:

1. Protocol-Centric Workflows that
Encode flexible, modifiable criteria (RECIST, RANO, etc.) and visit schedules up front.
Enforce correct, protocol-compliant lesion selection, measurements, and calculations—no guesswork.

2.Reader Guidance & Traceability to
Auto-navigate readers to the correct images and prior targets.
Capture every measurement and decision point with timestamps and users.

3. Protocol-Centric Workflows that
Coordinators, investigators, radiologists, and QA staff to work in one place.
No “radiology-only” silos. Sponsors/CROs can receive endpoints immediately.

4. De-Identification & Secure Sharing to
Automate compliant de-identification and streamline internal or external reads.
Eliminate CD-ROMs, ad hoc file-sharing, and emailing PHI.

5. Concurrency & Audit Support to
Log every action. Queries can be resolved with context, not guesswork.
Simplify re-reviews, adjudications, and reader switching when required.

6.Data Delivery & Analytics so
Results flow downstream automatically; no manual transposition.
Dashboards display status, turnaround times, and bottlenecks in real time.

Stop forcing clinical trial imaging research through clinical pipes. Seek out the best workflow for the research pipeline, and every stakeholder wins.

Cost & Impact Summary Direct Costs of the Current Approach

*[6,7]

*These are directional; substitute your center’s numbers to get exact totals.

The Organizational Impact of Poor Workflow

Radiology-centric PACS and other patchwork systems for trial imaging have significant reverberations throughout the organization. Radiology burnout increases because they battle a process not designed for them to work well in a cross-functional manner. Investigators lose trust when data is delayed or inconsistent, which can damage trial momentum. It is not the radiologists' fault – they are given an impossible task to manage multiple trial reads across multiple different protocols without purpose-built tools.

Sponsors leave, revenue leaks, and errors abound. If your center can’t deliver clean, on-time reads, sponsors will go elsewhere and work with sites that can. Every research read that takes 2–3 times longer than necessary costs real money, either in staff hours or lost trials. Ultimately, protocol deviations become the norm. Errors lead to deviations, and excessive deviations can result in data exclusion or costly remediation.

The Upside of Doing It Right

When data capture and delivery are streamlined and harmonized, trial staff experience time savings of 80% or more, which opens up opportunities to enroll more patients and conduct more trials. More trials result in increased revenue for the cancer center and more patients becoming eligible for potentially life-saving or life-extending treatments. Radiologists can improve efficiency with guided lesion tracking and structured protocols, boosting productivity by up to 50%. This results in more bandwidth to read a greater number of clinical exams, ultimately leading to increased RVUs and radiology revenue. All stakeholders see fewer queries & re-reads because the data is complete, consistent, and auditable at the point of capture. Most importantly, patients benefit when their oncologist has precise and timely information in advance of their office visit. [6,7]

Sites become more attractive to sponsors, and scalability is unlocked. The ability to implement one platform, one time, and deploy it everywhere across all current and future trials makes scaling not only possible but an immediate reality. Coupled with the ability to deliver research endpoints accurately and rapidly, sites become highly differentiated and desirable to sponsors.

Your Next Steps (Practical Checklist)

Final Thought

Cancer centers are judged by their ability to deliver accurate and timely clinical trial imaging data that advances trials and benefits both patients and sponsors. Radiology’s clinical efficiency mindset does not translate to research success. It’s time to stop pretending it does. Select a system designed for clinical trials imaging, connect the stakeholders who actually use the data, and measure success by the speed and quality of your endpoints, not by how many clicks a radiologist made.

The ambition to “transform” research imaging through clinical tools failed. The good news? The fix is clear, implementable, and already in use at over a quarter of all NCI-designated cancer centers that have refused to settle for less. In fact, you can check out Yunu’s full end-to-end platform in less time than it took to read this article. Ready to find a better way?

Request A Personalized Demo of Yunu

We look forward to showing you our complete imaging workflow platform that helps synchronize and accelerate the work between cancer centers, sponsors, and imaging teams.

What can I expect?

A brief discussion of your current workflow, needs, and goals.
A live demo to review features and functionality with our clinical experts.
Q&A on how the Yunu system can meet the needs of your organization.

Click here to start a conversation

References

1. Sorenson, J. (2025, July 18). Imaging’s ride to the bottom in clinical trials—and why it matters now. The Cancer Letter. https://cancerletter.com/sponsored-article/20250718_4/

2. Fotenos, A. (2018, September 20). Update on FDA approach to safety issue of gadolinium retention after administration of gadolinium-based contrast agents (Presentation by Anthony Fotenos, MD, PhD, Lead Medical Officer, Division of Medical Imaging Products). U.S. Food and Drug Administration. https://www.fda.gov/media/116492/download

3. Eisenhauer, E. A., Therasse, P., Bogaerts, J., Schwartz, L. H., Sargent, D., Ford, R., Dancey, J., Arbuck, S., Gwyther, S., Mooney, M., Rubinstein, L., Shankar, L., Dodd, L., Kaplan, R., Lacombe, D., & Verweij, J. (2009). New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1). European Journal of Cancer, 45(2), 228–247. https://doi.org/10.1016/j.ejca.2008.10.026

4. Wen, P. Y., Macdonald, D. R., Reardon, D. A., Cloughesy, T. F., Sorensen, A. G., Galanis, E., DeGroot, J., Wick, W., Gilbert, M. R., Lassman, A. B., Tsien, C., Mikkelsen, T., Wong, E. T., Chamberlain, M. C., Stupp, R., Lamborn, K. R., Vogelbaum, M. A., van den Bent, M. J., & Chang, S. M. (2010). Updated response assessment criteria for high-grade gliomas: Response Assessment in Neuro-Oncology Working Group. Journal of Clinical Oncology, 28(11), 1963–1972. https://doi.org/10.1200/JCO.2009.26.3541

5. Cheson, B. D., Fisher, R. I., Barrington, S. F., Cavalli, F., Schwartz, L. H., Zucca, E., & Lister, T. A. (2014). Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: The Lugano classification. Journal of Clinical Oncology, 32(27), 3059–3067. https://doi.org/10.1200/JCO.2013.54.8800

6. Herzog, T. J., Wahab, S. A., Mirza, M. R., Pothuri, B., Vergote, I., Graybill, W. S., Malinowska, I. A., York, W., Hurteau, J. A., Gupta, D., González-Martin, A., & Monk, B. J. (2024). Concordance between investigator and blinded independent central review in gynecologic cancer clinical trials. International Journal of Gynecological Cancer. Advance online publication. https://www.international-journal-of-gynecological-cancer.com/article/S1048-891X(24)01777-8/fulltext

7. Yunu. (2024). 50% Imaging Error Rate in Clinical Trials Discussed by Expert Panelists from 5 NCI-Designated Cancer Centers. Retrieved from https://www.yunu.io/blogs/post/50-imaging-error-rate-in-clinical-trials-discussed-by-expert-panelists-from-5-nci-designated-cancers

8. Hicks, L. (2022, February 18). Disrespect from colleagues is a major cause of burnout, radiologists say. Medscape. https://www.medscape.com/viewarticle/968776?form=fpf

9. Rula, E. Y. (2024, July 3). Radiology workforce shortage and growing demand: Something has to give. American College of Radiology. https://www.acr.org/Clinical-Resources/Publications-and-Research/ACR-Bulletin/Radiology-Workforce-Shortage-and-Growing-Demand-Something-Has-to-Give

10. International Council for Harmonisation Expert Working Group. (2025). ICH harmonised guideline: Guideline for good clinical practice E6(R3). https://database.ich.org/sites/default/files/ICH_E6%28R3%29_Step4_FinalGuideline_2025_0106.pdf