Europe's confidence gap
The question repeated throughout the ESMO–CSCO session was whether findings from China-only trials could be applied elsewhere. That is increasingly the wrong question. China is already producing large, randomised studies capable of changing practice. The more useful question, as one presenter put it, is: “Will the evidence apply to patients and practice worldwide?” Differences in pharmacokinetics, tumour biology or safety may require additional investigation. But some of the biggest barriers have little to do with ethnicity. The comparator may no longer reflect the relevant standard. Previous and subsequent treatments may differ. Supportive care, diagnostics and toxicity management may alter how outcomes should be interpreted. Liver cancer – primarily hepatocellular carcinoma - illustrates why these differences matter. In China, more than three-quarters of cases are linked to hepatitis B6, while in Europe the causes are more varied, with metabolic and alcohol-related liver disease playing a much greater role7. A Chinese trial may therefore recruit patients with quite different characteristics from those European oncologists see in practice. The evidence may be robust, but the questions it leaves unanswered can be different. And this works both ways: Chinese physicians are also increasingly interested in understanding how global trial results apply to their own patients. None of these challenges is exclusive to China-only trials. Standards of care can move faster than pivotal trials in any region. The China–Europe issue is sharper because evidence may need to cross two gaps at once: time, as the treatment landscape moves, and context, as the evidence moves between healthcare systems.

A trial can therefore be scientifically credible yet difficult to translate into a European decision. Not because the study is China-only, but because the question it answered may no longer match the question Europe needs answered. Commissioning a new global Phase III trial is one response. It is also the most expensive, measured in money, development time and delayed patient access. The response should instead be proportionate to the uncertainty: sometimes a pharmacokinetic or safety cohort may be enough; sometimes a regional cohort, comparative study or external control could close the gap; and sometimes a new randomised study will still be necessary.
