- Design
- phase 3, multicentre, open-label randomised trial, prespecified interim analysis
- Population
- 451 patients with small-cell lung cancer progressing after first-line platinum-based therapy
- Primary outcome
- overall survival
- Effect
- median 13.3 vs 9.4 months, stratified HR 0.46 (95% CI 0.35-0.62); PFS 7.4 vs 2.8 months, HR 0.29
Four hundred and fifty-one patients with small-cell lung cancer progressing after first-line platinum were randomised 1:1 to tambotatug pelitecan, an antibody-drug conjugate directed at B7-H3, or to topotecan. The trial was open-label and this is the prespecified interim analysis.
Median overall survival was 13.3 months (95% CI 12.1 to not estimable) against 9.4 months (7.7-10.5), a stratified hazard ratio for death of 0.46 (0.35-0.62). Progression-free survival was 7.4 against 2.8 months (hazard ratio 0.29, 0.23-0.37). Confirmed objective response was 59.1% against 9.7%. Grade 3 or higher adverse events were less common on the conjugate, 55.4% against 77.9%.
Relapsed small-cell lung cancer is a setting where almost nothing has worked, and topotecan is the comparator because there has been no better one. A 59.1% response rate against 9.7% is not a marginal improvement in that context. Two things temper it: open-label design, which affects investigator-assessed response more than it affects survival, and an interim analysis, where the effect size is typically larger than at final analysis. The trial was conducted largely in China with national and industry funding, and the drug is not approved anywhere. Neither caveat undoes a halving of the hazard of death.
- Confirmed response 59.1% against 9.7% with topotecan, in a setting where response is rare.
- Grade 3 or higher events were less frequent on the conjugate, not more.
- Open-label, so investigator-assessed response is the softest of the three endpoints.
- An interim analysis tends to overstate the effect relative to the final result.
- Not approved in India or elsewhere; nothing to offer a patient today.
Why it matters
Second-line small-cell lung cancer has resisted every improvement for two decades, and this is the first comparison to move survival and toxicity in the same direction.
Don't overread it
Open-label interim analysis of an unapproved drug - the survival signal is strong, the response rate is the endpoint most affected by unblinding.
The statistics, in plain English
A hazard ratio of 0.46 with an upper confidence limit of 0.62 is a large and well-separated effect, and the upper bound of the median survival could not be estimated because too few patients in that arm had died - which is itself a favourable sign rather than a gap. Interim analyses stop early precisely when the effect looks large, and that selection means the true effect is usually somewhat smaller than the one that triggered the analysis.
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