Tarlatamab gave an mOS of 13.6 months against 8.4 with chemotherapy in SCLC at extended DeLLphi-304 follow-up

Survival at two years roughly doubled against chemotherapy

The first analysis of DeLLphi-304, at a median follow-up of less than 12 months, had shown tarlatamab improving OS, progression-free survival (PFS) and patient-reported outcomes against chemotherapy. The extended analysis, with a data cutoff of 18 February 2026, gives Kaplan-Meier estimated OS rates of 55.9% at 12 months and 30.3% at 24 months with tarlatamab, against 39.0% and 14.9% with chemotherapy.

Chemotherapy in the control arm was topotecan, lurbinectedin or amrubicin, given until progression, unacceptable toxicity or death. The trial was open-label, and PFS was assessed by investigators.

A one-month median PFS gap, with 14.2% against 2.3% progression-free at two years

Median PFS differed by one month between the arms, at 4.2 months with tarlatamab and 3.2 months with chemotherapy, with median follow-up of 22.2 and 21.9 months. The estimated PFS rate at 12 months was 20.6% with tarlatamab and 5.9% with chemotherapy.

Sustained disease control for 52 weeks or longer was seen in 24.4% of patients given tarlatamab and 4.3% of those given chemotherapy. These figures describe a subset of patients with prolonged disease control in each arm.

Ongoing responses in 31.1% of tarlatamab responders at cutoff, against 3.8%

Of the 90 responders to tarlatamab, 28 still had an ongoing response at data cutoff, against 2 of the 52 responders to chemotherapy. The abstract does not give median duration of response for the extended analysis.

“With extended follow-up, tarlatamab continues to demonstrate sustained and durable clinical benefit and a favorable long-term tolerability profile versus chemotherapy, in patients with SCLC after first-line platinum-based chemotherapy, supporting its use as the standard of care in this setting.”Rudin et al., Annals of Oncology (2026)

Cytokine release syndrome and ICANS clustered early and were mostly low grade

Treatment-related adverse events led to discontinuation in 3.2% of the tarlatamab arm and 6.6% of the chemotherapy arm. Grade 3 or higher events became less frequent over time, and no serious treatment-related adverse event had its onset at 12 months or later.

Cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome (ICANS) occurred early in treatment and were predominantly low grade. The abstract does not give their rates.