Two of today's trials report the same shape of result: a clear, statistically robust improvement in progression-free survival, and an overall survival curve that has not separated. It is worth being precise about what that combination does and does not mean, because it is the commonest pattern in modern oncology and the commonest source of over-claiming in a clinic conversation.
Progression-free survival is time until a scan shows growth or the patient dies. When it improves and survival does not, three explanations compete. The trial may be immature, and survival may separate later — often true, and the honest position while waiting. Effective subsequent lines may dilute the difference, since patients who progress on the control arm go on to receive something else. Or the delay in radiological progression may simply not translate into longer life, which happens more often than the field likes to acknowledge.
The practical consequence is in how you frame the offer. A patient hearing 'this drug works' will assume it means more time alive. What has been shown is more time before the scan changes, and — this is the part worth saying aloud — more time on treatment, with its toxicity, its infusions and its cost. Where serious adverse events double, as in both of today's trials, the patient is trading known present harm for a benefit measured on imaging. Some will take that trade, particularly if delaying progression means delaying symptoms. Others will not. Both are reasonable, and neither can be chosen properly if the endpoint is described as though it were survival.
- State the endpoint explicitly when discussing a new agent: progression-free survival is not survival.
- Ask whether the survival data are immature or genuinely flat before concluding either way.
- Weigh the toxicity, which is certain and immediate, against the imaging benefit, which may not translate.
- Where delayed progression means delayed symptoms, say so — that is a real patient-centred benefit.
- Consider what subsequent lines the control arm received; effective crossover dilutes survival differences.
The statistics, in plain English
Progression-free survival is a composite of radiological progression and death, and radiological progression usually dominates it — so the endpoint is largely determined by how often scans are done and how they are read. That is why blinded independent radiology review matters, and why an unblinded trial's progression-free survival advantage is weaker evidence than the same result from a blinded one. Overall survival needs many more events and much longer follow-up to reach significance, so an interval that crosses 1.0 early in a trial's life genuinely can mean 'not yet known' rather than 'no effect'. The way to tell them apart is to look at how many deaths have occurred, not at the p value.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for oncology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free