preheader BJMO

Clinical and molecular characteristics of long-term survivors in the CASPIAN trial

In the CASPIAN trial, durvalumab plus platinum-etoposide (EP) showed a sustained overall survival benefit over EP alone in patients with extensive-stage small cell lung cancer (ES-SCLC). In a new post-hoc analysis, three times as many patients were estimated to be alive at the three years when treated with durvalumab plus EP versus EP alone, with the majority still receiving durvalumab at data cut-off. As such, these findings further establish durvalumab plus EP as first-line standard of care for ES-SCLC.

In the phase III CASPIAN trial, first-line durvalumab plus platinum-etoposide (D+EP) significantly improved the overall survival (OS) vs. EP alone in patients with extensive-stage small cell lung cancer (ES-SCLC). Furthermore, a numerical improvement in OS was observed with durvalumab plus tremelimumab plus EP (D+T+EP) vs. EP. Unfortunately however, to date the biomarkers that predict efficacy of immune checkpoint blockade have not been well characterised in SCLC. Therefore, the clinical and molecular characteristics of patients who experienced long-term survival in CASPIAN were assessed.

Study design

The CASPIAN study is a phase III, global, randomised, open-label, active-controlled, multicentre study. Patients with treatment-naïve ES-SCLC (N= 805) were randomised 1:1:1 to D+EP, D+T+EP or EP alone. The primary endpoint of the trial was overall survival. Analysis of tissue tumour mutational burden (tTMB) and PD-L1 were prespecified exploratory endpoints. Clinical characteristics were assessed in long-term survivors (LTS, N= 95), defined as patients still alive at the 22 March 2021 data cut-off (median follow-up 39.4 months). PD-L1 expression, tTMB and presence or absence of the HLA-DQB1*03:01 allele were assessed in the biomarker evaluable populations according to OS landmark (18 and 36 months).


At the time of data cut-off, there were more than three times more LTS in the D+EP arm, than in the EP arm (16%, 14% and 5% in the D+EP, D+T+EP and EP arms, respectively). LTS had a slightly higher incidence of some favourable prognostic factors at baseline. However, there were also patients with baseline brain or liver metastases who did achieve long-term survival, although their frequency was lower among LTS compared to the ITT populations. More LTS received ≥4 cycles EP than ITT, with a higher median overall treatment exposure in the immunotherapy plus EP arms. In total, 46 LTS (27 D+EP; 19 D+T+EP) were still receiving durvalumab treatment at the data cut-off. In both immunotherapy + EP arms, LTS were more likely to have experienced an objective response than the ITT population. Also the 12- and 24-months PFS rates were much higher among LTS, across all three treatment arms. Despite longer exposure, LTS did not experience an increase in serious adverse events (AEs) vs. the ITT population. In addition, the distribution of serious AEs across system organ classes was similar in LTS and the ITT population.

In the D+T+EP arm only, incidence of PD-L1 expression on ≥1% of tumour cells (TC) or immune cells (IC) was enriched in patients with OS ≥18 months vs. OS <18 months; enrichment was maintained with OS ≥36 months vs. < 36 months, although based on small patient numbers. In contrast, the incidence of high or low tTMB (>10 mut/Mb or ≤10 mut/Mb) was not consistently enriched in patients with OS ≥18 vs. < 18 months or OS ≥36 vs. < 36 months. Finally, in the D+T+EP arm only, presence of the HLA-DQB1*03:01 allele was enriched in patients with OS ≥18 vs. < 18 months or OS ≥36 vs. < 36 months.


In CASPIAN, there were more than three times as many LTS in the D+EP arm vs. the EP arm. In general, LTS were more likely to have favourable prognostic characteristics than patients in the ITT population, although the LTS cohort also included patients with brain or liver metastases at baseline. Notably, LTS were more likely to have completed EP induction and to have achieved an objective response and a longer PFS than the ITT population in both the D+EP and D+T+EP arms. There was no evidence of cumulative toxicity for long-term survivors. In the D+EP arm, there was no association of PD-L1 expression, tTMB or HLA-DQB1*03:01 presence with OS ≥ 18 months or ≥ 36 months. However, in the D+T+EP arm, PD-L1 expression (TC or IC ≥1%) and HLA-DQB1*03:01 presence (but not high tTMB) were enriched in patients with OS≥ 18 months. Further investigation in warranted to understand the potential role of these and other biomarkers in SCLC.


Reinmuth N, et al. Durvalumab (D) ± tremelimumab (T) + platinum-etoposide (EP) in 1L extensive-stage (ES) SCLC: characteristics of long-term survivors in the CASPIAN study. Presented at ELCC 2022; Abstract 141O.

Speaker Niels Reinmuth

Niels Reinmuth

Niels Reinmuth, MD, PhD, Asklepios Lung Clinic, Munich-Gauting, Germany


See: Keyslides

Back to Top