I've spent more than 25 years in oncology clinical research, including my time as Department Administrator for Investigational Cancer Therapeutics at The University of Texas MD Anderson Cancer Center, the world's largest Phase 1 program. Here's what I learned there: the people who run a protocol will find its problems. The only question is when. Either they find them before the protocol is final, or they find them after sites have started screening patients. The second one costs a lot more.
The cost of finding problems late
The data are clear. In a Tufts Center for the Study of Drug Development (Tufts CSDD) study, 57% of protocols had at least one substantial amendment, and sponsors themselves judged 45% of those amendments avoidable. The median direct cost of one amendment was $141,000 in Phase II and $535,000 in Phase III. More recent Tufts data show about three in four protocols get amended, each amendment takes roughly 260 days to implement, and oncology has higher amendment rates than other therapeutic areas. Cancer centers report that amendments received before activation commonly add four to six weeks to start-up.
Eligibility criteria change most often, and in early-phase oncology they determine how fast you move. An analysis of 231 Phase I dose-escalation trials in solid tumors found that trials with restrictive eligibility criteria took a median of 26 months to complete accrual, compared with 17 months for less restrictive trials. They also explored fewer dose levels (a median of 6 versus 9). Rates of dose-limiting toxicities, serious adverse events, and responses did not differ significantly between the two groups.
So tighter criteria meant nine more months of accrual and no measurable safety benefit. That's a cost you can avoid.
Some amendments can't be prevented. New safety data and regulatory requests come when they come. Design problems are different. Criteria that exclude the patients who actually walk into the clinic, visit schedules sites can't staff, and sampling schedules patients won't tolerate can all be caught before the protocol is final. You just have to ask the right people early enough.
What KOL and PI input adds
This isn't only my opinion. Regulators say the same thing. ICH E8(R1), the 2021 guideline from the International Council for Harmonisation (the body through which FDA, the European Medicines Agency and other regulators set shared standards for clinical trials), states that "clinical study design is best informed by input from a broad range of stakeholders," and that clinical investigators and potential participants "have valuable insights into the feasibility" of a study. It also states that study designs "should be operationally feasible and avoid unnecessary complexity".
KOLs and PIs are more than experienced researchers. They're oncologists treating the target population every day, and they can answer questions a protocol team usually can't answer on its own:
Do the eligibility criteria match the patients we actually see in clinic?
Is the PK sampling schedule workable for outpatients?
Can sites staff the visit schedule?
Are the required biomarker tests and imaging available locally, and how long do results take?
In early-phase oncology these questions matter more, not less. As the Phase I data show, criteria that are tighter than they need to be slow accrual with no significant difference in safety.
Timing is everything. The input is most useful while changes are still cheap: during synopsis development and before the protocol is submitted. After submission, every change becomes an amendment.
How WINNIEDEL ONCOLOGY does it
I built WINNIEDEL ONCOLOGY to fix the bottlenecks that keep promising cancer research from getting off the ground, and bringing KOLs in late is one of the biggest. Our KOL/PI engagement service puts practicing oncologists in front of sponsors before the protocol is finalized. It works in four steps
You show us your plan for the science.
We find the right KOLs and PIs with relevant experience, fast.
You bring them into protocol design, advisory boards, on-site roles, or whatever else you need.
Your trial launches with that real-world input built into the design.
What this means for you as a sponsor:
Matched expertise. You get insight from the specialists your protocol targets.
One network for protocol review and site selection. Our KOL network mirrors our Global Oncology Site Universe, so the investigators who review your protocol also inform which sites get selected.
Flexible engagement. Need a PI for an ADC advisory board next week? Done. Need a radiopharma KOL to check dosing windows? Also done. KOLs and PIs can join for a full protocol review, an advisory board, or one specific question.
Direct access. I know the people who run these cancer centers. They pick up the phone because it's us calling, not a generic vendor blast.
Continuity. If the indication shifts or you add expansion cohorts, we bring in the right expert without losing momentum or restarting the search.
No endless vetting. No unnecessary fluff. Just the right clinicians in the right rooms at the right time, because cancer patients don't have time to wait for us to fix a protocol we could have gotten right the first time.
Contact WINNIEDEL ONCOLOGY to talk about KOL and PI input for your protocol.