Combination lurbinectedin and doxorubicin versus physician's choice of chemotherapy in patients with relapsed small-cell lung cancer (ATLANTIS): a multicentre, randomised, open-label, phase 3 trial

Background: Lurbinectedin is a synthetic marine-derived anticancer agent that acts as a selective inhibitor of oncogenic transcription. Lurbinectedin monotherapy (3·2 mg/m2 every 3 weeks) received accelerated approval from the US Food and Drug Administration on the basis of efficacy in patients with small-cell lung cancer (SCLC) who relapsed after first-line platinum-based chemotherapy. The ATLANTIS trial assessed the efficacy and safety of combination lurbinectedin and the anthracycline doxorubicin as second-line treatment for SCLC. Methods: In this phase 3, open-label, randomised study, adult patients aged 18 years or older with SCLC who relapsed after platinum-based chemotherapy were recruited from 135 hospitals across North America, South America, Europe, and the Middle East. Patients were randomly assigned (1:1) centrally by dynamic allocation to intravenous lurbinectedin 2·0 mg/m2 plus doxorubicin 40·0 mg/m2 administered on day 1 of 21-day cycles or physician's choice of control therapy (intravenous topotecan 1·5 mg/m2 on days 1-5 of 21-day cycles; or intravenous cyclophosphamide 1000 mg/m2, doxorubicin 45·0 mg/m2, and vincristine 2·0 mg on day 1 of 21-day cycles [CAV]) administered until disease progression or unacceptable toxicity. Primary granulocyte-colony stimulating factor prophylaxis was mandatory in both treatment groups. Neither patients nor clinicians were masked to treatment allocation, but the independent review committee, which assessed outcomes, was masked to patients' treatment allocation. The primary endpoint was overall survival in the intention-to-treat population. This trial is registered with ClinicalTrials.gov, NCT02566993, and with EudraCT, 2015-001641-89, and is complete. Findings: Between Aug 30, 2016, and Aug 20, 2018, 613 patients were randomly assigned to lurbinectedin plus doxorubicin (n=307) or control (topotecan, n=127; CAV, n=179) and comprised the intention-to-treat population; safety endpoints were assessed in patients who had received any partial or complete study treatment infusions (lurbinectedin plus doxorubicin, n=303; control, n=289). After a median follow-up of 24·1 months (95% CI 21·7-26·3), 303 patients in the lurbinectedin plus doxorubicin group and 289 patients in the control group had discontinued study treatment; progressive disease was the most common reason for discontinuation (213 [70%] patients in the lurbinectedin plus doxorubicin group vs 152 [53%] in the control group). Median overall survival was 8·6 months (95% CI 7·1-9·4) in the lurbinectedin plus doxorubicin group versus 7·6 months (6·6-8·2) in the control group (stratified log-rank p=0·90; hazard ratio 0·97 [95% CI 0·82-1·15], p=0·70). 12 patients died because of treatment-related adverse events: two (<1%) of 303 in the lurbinectedin plus doxorubicin group and ten (3%) of 289 in the control group. 296 (98%) of 303 patients in the lurbinectedin plus doxorubicin group had treatment-emergent adverse events compared with 284 (98%) of 289 patients in the control group; treatment-related adverse events occurred in 268 (88%) patients in the lurbinectedin plus doxorubicin group and 266 (92%) patients in the control group. Grade 3 or worse haematological adverse events were less frequent in the lurbinectedin plus doxorubicin group than the control group (anaemia, 57 [19%] of 302 patients in the lurbinectedin plus doxorubicin group vs 110 [38%] of 288 in the control group; neutropenia, 112 [37%] vs 200 [69%]; thrombocytopenia, 42 [14%] vs 90 [31%]). The frequency of treatment-related adverse events leading to treatment discontinuation was lower in the lurbinectedin plus doxorubicin group than in the control group (26 [9%] of 303 patients in the lurbinectedin plus doxorubicin group vs 47 [16%] of 289 in the control group). Interpretation: Combination therapy with lurbinectedin plus doxorubicin did not improve overall survival versus control in patients with relapsed SCLC. However, lurbinectedin plus doxorubicin showed a favourable haematological safety profile compared with control. Funding: PharmaMar.