Commentary|Articles|November 12, 2026

DeLLphi-305 Sets a New Standard, and the Bar Is Rising in SCLC

Author(s)Sabrina Serani, Jacob Sands, MD

Jacob Sands, MD, of Dana-Farber Cancer Institute, discusses survival gains, durability, and safety of tarlatamab plus durvalumab in the phase 3 DeLLphi-305 trial.

Jacob Sands, MD, associate chief of the Lowe Center for Thoracic Oncology at Dana-Farber Cancer Institute and assistant professor of medicine at Harvard Medical School in Boston, Massachusetts, discussed the primary analysis of the phase 3 DeLLphi-305 trial (NCT06211036), presented at the 2026 European Society for Medical Oncology (ESMO) Congress,¹ in an interview with Targeted Onc^TM. The study evaluated tarlatamab-dlle (Imdelltra) plus durvalumab (Imfinzi) vs durvalumab alone as first-line maintenance therapy for patients with extensive-stage small cell lung cancer (ES-SCLC) whose disease had not progressed after first-line platinum, etoposide, and durvalumab. Sands also discussed the prespecified subgroup findings in patients with liver and brain metastases and how he expects the treatment paradigm to evolve.

Targeted Onc: What are the most important key takeaways from DeLLphi-305?

Jacob Sands, MD: In the DeLLphi-305 trial, we've seen unprecedented improvement in median overall survival [OS], with a median of 21.1 vs 12.7 months. This is a hazard ratio [HR] of 0.42. We've also seen unprecedented improvement in median progression-free survival [PFS], 7.4 months vs 1.9 months, an HR of 0.34. I think also to call out, we've seen real durability to this for a substantial number of patients: 38% of patients on the tarlatamab-durvalumab arm [had] ongoing disease control, progression-free survival at 1 year, compared [with] 10% from the durvalumab arm. From a toxicity standpoint, we're seeing quite a manageable profile. And I know we say that about every trial, but 50% cytokine release syndrome [CRS], overwhelmingly grade 1. We see 21 patients, about 7%, with grade 2, and 1 single patient, 0.4%, with grade 3. Tocilizumab [Acemtra] was only used in 3.9%. Low-flow oxygen, 1.4%. So we're not talking about a lot of high-grade CRS or even a lot of management of a CRS with things like tocilizumab. With ICANS [immune effector cell–associated neurotoxicity syndrome], we saw 3 patients, 1% of patients, that had ICANS. So for CRS and ICANS, the numbers really look a little bit better, or in ICANS, substantially better, than what we've seen from prior studies with tarlatamab. So overall, unprecedented improvement in median PFS and OS, along with durability in a substantial percentage of patients and a manageable toxicity profile. This sets carboplatin, etoposide, and durvalumab, followed by tarlatamab and durvalumab, as the new standard of care for first-line extensive-stage small cell lung cancer.

Were any of these findings unexpected?

I’ll point out that first of all, the median overall survival really looked similar. It was within the ballpark of what we saw from DeLLphi-303 [NCT05361395]. That was the phase 1b. Now, that was reported last year at [the World Conference on Lung Cancer] with a median overall survival of 25.3 months. Now, that made a splash. That really brought a lot of attention to tarlatamab in the maintenance setting and made people really look forward to seeing what the randomized data would show, and this landed within that region. Now, I'll point out here at ESMO, the update from DeLLphi-303 with more mature data, that median overall survival has increased from 25.3 to now 37.9 months. The median overall survival is actually a little more than a year more than what we saw from that initial data. Now, that's exciting because in DeLLphi-305, the median overall survival of 21.1 months is with about 16 months of median follow-up. So that tail of the curve will likely evolve, and this is not the final for the median overall survival. So that initial look at the overall survival curves was very exciting, but was kind of in line with what that first look at DeLLphi-303 looked like.

The median progression-free survival I found surprising. Median PFS of 7.4 months, that's better than what we saw from DeLLphi-303, and the 38% ongoing PFS at 1 year—this is a substantial percentage of patients that are really getting incredible durability. I should also highlight that with the immunotherapy agents, we sometimes get oligoprogression. This is a little bit different than what we see from the cytotoxics so much. So a single site or a couple sites that we can then radiate and continue the systemic therapy. That's essentially the physician saying the patient is still benefiting from the drug even though there was technically a progression event. And to that point, 44% of patients continue on the tarlatamab-durvalumab combination at the last data cut, which is to say that although I've been highlighting the 38% ongoing progression-free survival at 1 year, I'd say the number of patients that are benefiting from this combination in a durable way is likely a higher percentage of patients, and I look forward to when we can parse the data a bit more. This is really just a first look, but I think there's a lot more to pull out of the data.

What is the significance of the findings in the liver and brain metastases groups in the prespecified subgroup analyses?

When looking at specific subgroups within DeLLphi-305, some of the stratification factors were presence or absence of liver metastases and/or brain metastases, and we looked at this within PFS and OS subgroups as well. Now, baseline characteristics: about 20% of patients in both arms had baseline brain metastases. Now, this could have been asymptomatic, untreated brain mets, which I'd highlight. Tarlatamab has demonstrated good [central nervous system] efficacy, and so these patients were eligible to enroll into the study. Baseline characteristics also included about a third of patients in both arms that had baseline liver metastases. Now, that was from the initiation of induction therapy. Enrollment to the trial was after induction, but this is essentially presence or absence of brain and/or liver metastases prior to that induction therapy.

What we saw from the subgroups is that whether patients had brain mets or not, or liver mets or not, in both of those they did better with the incorporation of tarlatamab. These looked meaningful. Of course, there's not a statistical plan for these smaller subsets, but on the forest plot, these are not crossing 1 and actually look similar whether they have brain metastases or not, liver metastases or not. Related to this, in DeLLphi-304 [NCT05740566], we saw data for presence or absence of brain mets in the control arm with chemotherapy. Those curves separate out in the way that we've always seen, but in the tarlatamab arm, the curves entirely overlap each other, which adds to this new paradigm, where presence or absence of brain mets does not seem to be prognostic to survival in patients that are getting treated with tarlatamab.

How do you see the treatment paradigm of small cell lung cancer changing in the next 5 to 10 years?

Looking ahead to the treatment paradigm, starting here with DeLLphi-305, this represents a new standard of care in the treatment of patients with extensive-stage small cell lung cancer: carboplatin, etoposide, and durvalumab, followed by tarlatamab and durvalumab. Now, there are a lot of drugs in development. We have the antibody-drug conjugates, which I would say are essentially, or will, compete with platinum/etoposide in the first-line setting. That's kind of a surprising thing. Five years ago, to say, are we going to potentially have something that we would randomize to go against platinum/etoposide, which has been the most reliable therapy up to this point? That was hard to imagine, but here we are with [antibody-drug conjugates] that are showing really impressive response rates in the second-line setting that, frankly, I anticipate to look better in the first-line setting. We also have bispecific checkpoint inhibitors that are looking quite promising, and so I think those will be competing against durvalumab within the current regimen I've described, or atezolizumab [Tecentriq], another PD-L1 inhibitor, in the frontline setting. And then tarlatamab—now, I expect [the combination] to be approved and widely used. There are other T-cell engagers, so I think we're going to see what is the combination that potentially could beat the DeLLphi-305 regimen. But frankly, the bar is getting much higher with this regimen.