AstraZeneca PLC (NASDAQ: AZN) and Amgen Inc. (NASDAQ: AMGN) have reported Phase 3 findings showing that IMFINZI (durvalumab) combined with IMDELLTRA (tarlatamab-dlle) reduced the hazard of death by 58% compared with IMFINZI alone as first-line maintenance treatment for extensive-stage small-cell lung cancer. Presented at the European Society for Medical Oncology Congress 2026 on October 24 and simultaneously published in the New England Journal of Medicine, the DeLLphi-305 results demonstrated median overall survival of 21.1 months with the combination versus 12.7 months with durvalumab alone.
The findings provide randomised evidence supporting the use of a DLL3-directed T-cell engager earlier in the treatment pathway, before patients experience disease progression following initial chemoimmunotherapy. Tarlatamab is already approved in the United States for extensive-stage small-cell lung cancer that has progressed on or after platinum-based chemotherapy, but its use in the first-line maintenance regimen evaluated in DeLLphi-305 remains investigational.
The results are particularly consequential because extensive-stage small-cell lung cancer remains an aggressive malignancy with substantial relapse risk, even after initial response to platinum chemotherapy and immune checkpoint inhibition. The survival improvement indicates that introducing tarlatamab during maintenance treatment can materially alter outcomes in the studied population. However, the combination also introduces additional toxicity monitoring and administration requirements that will be important in future regulatory and clinical assessments.
What do the DeLLphi-305 Phase 3 survival results reveal about first-line maintenance treatment?
DeLLphi-305 is a global, randomised, Phase 3 study involving 563 patients with extensive-stage small-cell lung cancer who had not experienced progression following induction treatment with durvalumab, platinum chemotherapy and etoposide. Participants were randomly assigned to receive continued IMFINZI in combination with tarlatamab or IMFINZI alone. Treatment continued until disease progression or unacceptable toxicity.
The primary endpoint was overall survival, providing a direct assessment of whether the addition of tarlatamab extended life compared with established immunotherapy maintenance. At a planned interim analysis with approximately 16 months of median follow-up, the trial reported an overall survival hazard ratio of 0.42. The 95% confidence interval ranged from 0.32 to 0.54, with a p-value below 0.001.
Median overall survival was 21.1 months with the combination compared with 12.7 months for durvalumab alone, representing an 8.4-month difference in the median outcomes reported at the analysis. Estimated survival at 12 months was 79% in the combination group and 53% in the control group. These results provide both relative and absolute measures of the survival advantage observed during the study.
The hazard ratio indicates that the risk of death over the observation period was substantially lower in the combination arm. It should not be interpreted as meaning that 58% fewer patients died at every individual time point. The one-year survival estimates provide a complementary assessment, demonstrating a 26-percentage-point difference between the study groups.
AstraZeneca characterised the median overall survival as the longest reported in a Phase 3 extensive-stage small-cell lung cancer study. That assessment depends on the populations, treatment settings and endpoints included in the comparison, and it should not be taken as a head-to-head demonstration of superiority over unrelated regimens. The stronger conclusion is that DeLLphi-305 established a statistically significant survival improvement against its own randomised control arm.
Why does the progression-free survival benefit strengthen the DeLLphi-305 findings?
The trial also demonstrated a statistically significant improvement in progression-free survival, a key secondary endpoint. Median progression-free survival reached 7.4 months with tarlatamab plus durvalumab compared with 1.9 months for durvalumab alone. The hazard ratio was 0.34, corresponding to a 66% lower hazard of disease progression or death.
At 12 months, an estimated 38% of patients receiving the combination remained alive without disease progression, compared with 10% receiving durvalumab alone. This finding indicates that the treatment improved disease control beyond the initial maintenance period for a greater proportion of participants. The consistency between the progression-free and overall survival results provides a coherent efficacy profile.
The study also reported a confirmed objective response rate of 26.4% with the combination versus 4.1% with durvalumab alone among patients who had measurable disease at maintenance randomisation. Median duration of response was 12.7 months and 5.6 months, respectively. These results suggest that tarlatamab can produce additional tumor responses even after patients have already completed initial chemoimmunotherapy.
The response analysis involved fewer patients than the overall survival assessment because eligibility required measurable disease at randomisation. It should therefore not be interpreted as the response rate among every individual enrolled in the trial. The endpoint remains useful for demonstrating antitumor activity during maintenance, but overall survival provides the more consequential evidence of clinical benefit.
The progression-free survival results also help distinguish the combination from a treatment strategy that merely prolongs survival without affecting measurable disease control. Both endpoints favoured the investigational regimen, supporting the biological rationale for introducing DLL3-directed therapy before progression.
How does tarlatamab’s DLL3 mechanism complement durvalumab’s PD-L1 blockade?
Tarlatamab is a bispecific T-cell engager designed to bind delta-like ligand 3, or DLL3, on small-cell lung cancer cells and CD3 on T cells. By connecting these two cell populations, the medicine is intended to activate T-cell-mediated destruction of DLL3-expressing cancer cells. DLL3 is frequently expressed in small-cell lung cancer, making it a therapeutic target in a malignancy with relatively limited molecularly targeted treatment options.
Durvalumab operates through a different immune mechanism. It blocks programmed death-ligand 1, or PD-L1, helping prevent tumor-associated immune suppression through the PD-1 pathway. Its established role in extensive-stage small-cell lung cancer involves combination treatment with platinum chemotherapy and etoposide followed by maintenance immunotherapy.
Combining these mechanisms creates a therapeutic strategy that pairs immune checkpoint blockade with direct recruitment of T cells to tumor cells. The potential advantage is that tarlatamab may generate tumor-directed immune activity while durvalumab helps preserve aspects of the antitumor immune response. The clinical trial establishes the efficacy of the regimen but does not independently quantify the contribution of every proposed biological interaction.
The timing of administration is important. In DeLLphi-305, patients entered the maintenance comparison only after completing induction treatment without disease progression. This population differs from patients starting treatment with newly diagnosed extensive-stage disease and from those receiving tarlatamab after platinum-based treatment failure.
The results therefore support a specific treatment sequence rather than unrestricted use of the combination. Whether the regimen can benefit patients who progress during induction therapy, have different prior treatments or present with particular complications would require separate evidence.
What do the findings suggest for patients with brain or liver metastases?
AstraZeneca reported that the survival and progression-free survival benefits were observed across prespecified subgroups, including patients with brain or liver metastases. These populations are clinically important because extensive-stage small-cell lung cancer frequently spreads beyond the thorax, and metastatic involvement can complicate treatment and prognosis.
The trial permitted patients with treated or untreated asymptomatic brain metastases at baseline, providing evidence relevant to a broader population than studies excluding such patients entirely. However, the findings cannot automatically be extended to individuals with symptomatic or unstable central nervous system disease who would not have met the trial’s eligibility requirements.
Subgroup consistency can strengthen confidence that the overall trial result was not driven exclusively by one clinical category. Nevertheless, the magnitude and precision of treatment effects within individual metastatic subgroups need to be evaluated through the complete data. A favourable overall result does not establish identical efficacy across every patient population.
Brain metastases also have operational implications for a treatment involving T-cell engagement. Neurological symptoms may arise from the underlying cancer, prior treatment or immune-related toxicity, making careful assessment important. This consideration is particularly relevant because tarlatamab is associated with immune effector cell-associated neurotoxicity syndrome.
The contribution of the subgroup findings will depend on how the therapy performs in routine oncology practice, where patients may have more complex neurological or organ-function characteristics than those enrolled in controlled trials.
What safety risks could influence adoption of tarlatamab plus IMFINZI?
The combination’s efficacy advantage must be considered alongside its additional toxicity burden. In DeLLphi-305, Grade 3 or higher treatment-related adverse events occurred in 20.9% of patients receiving tarlatamab plus durvalumab, compared with 10.8% receiving durvalumab alone. The companies reported no treatment-related fatalities in the combination group.
Cytokine release syndrome was reported in 49.6% of patients receiving the combination, although most events were Grade 1 or Grade 2. Grade 3 cytokine release syndrome occurred in 0.4% of combination-treated patients, and no Grade 4 or fatal cases were reported. Immune effector cell-associated neurotoxicity syndrome occurred in 1.1% of patients, with no Grade 3 or higher cases disclosed.
These findings indicate that severe cytokine release syndrome and neurotoxicity were relatively uncommon within the trial, but the overall frequency of cytokine release syndrome remains substantial. The distinction between common low-grade events and less frequent severe complications is important because both can influence monitoring requirements and clinical workload.
Tarlatamab has an established boxed warning for cytokine release syndrome and neurological toxicities, including immune effector cell-associated neurotoxicity syndrome. These risks are relevant to its already approved indication and remain important when considering use earlier in the disease course. Durvalumab also carries established warnings relating to immune-mediated adverse reactions affecting multiple organ systems.
In DeLLphi-305, patients underwent monitoring in a healthcare setting following tarlatamab infusions on the first and eighth days of the first treatment cycle. Monitoring lasted one to two hours in some regions and six to eight hours in others, including Europe. The regimen’s eventual clinical implementation will depend on the authorised dosing and monitoring instructions, which may differ from the procedures used in the trial.
The median treatment duration was approximately 46.9 weeks in the combination arm compared with 8.1 weeks for durvalumab alone. Longer treatment exposure is consistent with improved disease control but may also increase cumulative administration and monitoring demands. A commercial programme would need to account for both the costs of treatment and the resources required to manage immune-related toxicity.
How could the results change the treatment sequence for extensive-stage small-cell lung cancer?
Tarlatamab received accelerated United States approval in May 2024 for adults with extensive-stage small-cell lung cancer progressing on or after platinum-based chemotherapy. The Food and Drug Administration subsequently granted traditional approval in November 2025 following confirmatory evidence from the DeLLphi-304 trial. These approvals established its role after disease progression, rather than as part of first-line maintenance treatment.
DeLLphi-305 investigates whether introducing tarlatamab earlier provides greater benefit than waiting until the disease progresses. The strong overall survival result supports that development strategy, although it does not establish the optimal sequence for every patient. Future treatment decisions may need to account for whether earlier use changes the effectiveness of subsequent therapy.
An important commercial consideration is the difference between adding tarlatamab to established maintenance treatment and replacing an existing standard. Durvalumab is already used after induction chemoimmunotherapy, meaning that the proposed regimen would add another medicine and associated monitoring. Its uptake would therefore involve evaluating the incremental survival benefit against the additional treatment burden.
The emerging evidence could influence the position of T-cell engagers in solid tumor oncology more broadly. These treatments have already demonstrated clinical activity in selected settings, but moving them into earlier treatment lines requires randomised evidence that the benefit outweighs additional toxicity. DeLLphi-305 provides such evidence for a specific small-cell lung cancer population.
The companies will need regulatory authorisation before promoting the combination for first-line maintenance use. The trial’s success does not change the existing approved indication for tarlatamab or automatically extend durvalumab’s label to include the new combination.
What do the results mean for AstraZeneca and Amgen’s oncology strategies?
For AstraZeneca, the findings support continued development of IMFINZI across different small-cell lung cancer settings. The medicine already has an established role in extensive-stage disease, and the company has advanced it in limited-stage small-cell lung cancer following chemoradiotherapy. The new combination provides another opportunity to extend its use within the treatment pathway.
For Amgen, the result supports efforts to broaden the clinical application of IMDELLTRA beyond relapsed disease. Earlier treatment may involve a larger eligible population and potentially longer treatment duration, although commercial opportunity will depend on approval, prescribing criteria and treatment access. The existing United States approval provides a commercial foundation, but the first-line maintenance indication remains a separate regulatory objective.
The development also demonstrates the strategic value of combining assets owned by different pharmaceutical companies. Each partner contributes a medicine with a distinct mechanism, and the clinical programme tests whether their combined activity offers benefits beyond established treatment. The eventual economics will depend on the relevant collaboration arrangements and commercial responsibilities.
AstraZeneca and Amgen have not established a new approved first-line maintenance combination through the October announcement. The immediate next step is to engage regulatory authorities and determine the appropriate submission pathway using the detailed Phase 3 evidence.
The 8.4-month median overall survival difference is a substantial clinical finding in the studied population, particularly given the aggressive nature of extensive-stage small-cell lung cancer. Its implementation will depend on preserving the demonstrated benefit while establishing practical administration, monitoring and access arrangements for a regimen that introduces T-cell engagement earlier in treatment.
