DeLLphi-304 was the first study to explore tarlatamab, a bispecific T-cell engager, as a second-line treatment as compared to standard of care chemotherapy in patients with recurrent or pre-treated small cell lung cancer. The study was first presented and published last summer and for the key endpoint, overall survival, and progression free survival it showed clear superiority of tarlatamab as compared to the three choices of standard chemotherapy. There was a reduction in the risk of death by 40%. It was a randomised open-label global multicentre phase III study.
What results were being presented?
At ESMO we are presenting more detailed data on the toxicity, on the adverse events, because they are very different between T-cell engager, immunotherapy and chemotherapy. So what we found overall is that if you’re looking at AEs, non-specifically tarlatamab has much less grade 3 or higher adverse events and much less treatment interruptions or dose reduction as compared to chemotherapy. It has clearly a different toxicity profile. So with tarlatamab, as expected, there are cytokine release syndrome events but they occur predominantly in the first cycle and don’t reoccur later. Most of them are mild to moderate and they don’t lead to severe outcomes for patients. So because of that, in the conduct of the study, the monitoring period for the CRS event was reduced from 48-hour at inpatients to 6-8 hours outpatients. So a part of the study population had this outpatient monitoring and we could compare not formally but just side by side the incident severity and outcome of CRS events and there was no apparent difference which gives a signal that you could deliver this drug also on an outpatient basis.
With chemotherapy, as expected, there was much more haematotoxicity and that is not just in cycle 1, that is maintained throughout the study treatment. The other adverse event that is of interest with cell-mediated immunotherapies is the so-called ICANS – neurological symptoms. We know that with cellular therapies in lymphoma, for example, or leukaemia this is a quite frequent and worrisome event, but with tarlatamab it was really rare, a very few percent of patients had it. Actually, there were 15 patients of 200-and-something and it was mild to moderate and it was completely reversible in 93% of those. It coincided in almost all of them also with a CRS event. So it’s fortunately not a big issue in tarlatamab treatment as it is with other agents in leukaemia and lymphoma treatment.
There is one neurological AE which is frequent with tarlatamab which is dysgeusia, so alteration of taste. That is occurring in 23% of patients. It’s mild in three quarters and moderate in one quarter but it does not lead to treatment discontinuation so it doesn’t prevent patients from taking the drug.
What impact might these results have?
The impact is that tarlatamab is a new standard of care in second-line treatment for small cell lung cancer. It is already approved in the United States and we are hoping for European approval maybe next summer so that our patients can get it, hopefully sooner or later depending on the health system access to this drug, because this is such a strong survival benefit which we haven’t seen in the past in small cell lung cancer.
Moreover, the compound is now explored in studies as first-line treatment in combination with standard first-line therapies. One of the studies is already fully enrolled. So it is quite likely that in the next few years it will move also towards earlier lines of treatment so that more patients get access to it than those that make it into second-line therapy.
What is the take home message?
Tarlatamab clearly is the standard of care for platinum pre-treated patients with small cell lung cancer with a really profound superiority in overall survival. Our detailed analysis shows that the side effects are as expected and fortunately largely confined to the first two doses, 2-3 doses, and are very manageable so that it can also be given outside of experienced study centres once it’s approved and reimbursed.
