The Problem Isn’t “China IITs.” It’s Trials Without Quality Systems.

Recent headlines about patient deaths in Chinese investigator-initiated trials raise legitimate questions about patient safety. They should.

But they risk missing the more important lesson: IIT is a pathway, not a definition of quality. What matters is how a trial is designed, reviewed and run, regardless of where it takes place.

Consider the cases behind the headlines.

In one study, ethics approval came before a toxicology report that showed significant liver injury. In another, a patient death was not disclosed appropriately. There were cases where the PI and the hospital EC blamed each other for not conducting the proper review or taking enough ownership. There were non-clinical packages that did not hold together, animal models that did not match the indication, comparability gaps between non-clinical and clinical material.

Early human CGT research will never be risk-free, but none of these were scientific failures. They were avoidable failures in process.

The distinction is important: these were not failures of IIT as a concept. They were failures in how individual IITs were run.

Compare, for example, another recent “China IIT” that also unfortunately had a death among its patients (attributed to progression of the underlying cancer). That programme went on to become part of a ~$1 billion acquisition. Yes, that’s AstraZeneca and EsoBiotec. The difference was how the trial was run. The reviewers and investigators followed proper process, reported the toxicity and death, published the results and maintained the appropriate research boundaries.

That raises the more useful question: why are some trials run well and others not?

Part of the answer is the lack of adequate guidelines. The trials in the headlines all happened prior to China’s Order 818, at a time when there was significant variance in process and review criteria. The problem is not just how things were measured, but also what to measure, when to review, and what process was needed for adequate safety. As we shall see in our next article, Order 818 addresses many of these gaps.

But a bigger part of the problem is ‘how’ the trials were run. And this is why professional third-party oversight is critical for quality assurance.

• It brings experience: knowing what to look for, such as when apparently acceptable animal data or dose calculations deserve another challenge. In CGT in particular, judgement matters; a result that looks acceptable on paper may not adequately predict what happens in a human patient.
• It brings execution know-how: PIs and sites are not interchangeable. They differ in disease expertise, CGT experience, patient access, infrastructure and operational capability. There are significant pitfalls to navigate across site selection, logistics, approval, and execution (examples highlighted in our next article). Knowing how to navigate the local landscape, match the right PI and site to the specific study requirements, and be close to the study to manage and react quickly can materially affect both execution and quality.
• It brings accountability: single ownership with a holistic view across the innovation originator, PI, hospital, CMC, safety, operations and data workstreams, rather than assuming somebody else has checked each component, or operating in silos. Importantly, the incentive differs: success depends on delivering many studies consistently, not simply getting one particular project started as quickly as possible.
• And it brings operational discipline: compliant processes, appropriate QC, active monitoring, traceable data, documented handovers and timely escalation when something does not look right. This problem is particularly common when running GMP-like setups without professional CMC support, or IIT studies without adequate medical and operations support.

This matters because even leading hospitals and PIs face practical challenges, and most biotechs lack the resources to connect those individual pieces into one coherent quality system.

When conducting trials, speed should never be the only goal. What must happen is adequate oversight from PI/site selection and protocol design, through site readiness, trial monitoring, quality management, data and reporting.

Speed is valuable. Speed without a quality system is not.

So rather than asking how good “China IITs” are, the better question is: Who ran the study? What standards did they apply? And who was independently responsible for making sure those standards were followed? That is the standard by which any trial, Chinese, American or otherwise, should be judged.

The next question is, what are the future standards to consider? Sign up to find out.