A pivotal bridging study of lurbinectedin as second-line therapy in Chinese patients with small cell lung cancer

This single-arm, multi-center clinical trial aimed to evaluate the safety, tolerability, DLT, recommended dose (RD), preliminary efficacy, and pharmacokinetics (PK) characteristics of lurbinectedin, a selective inhibitor of oncogenic transcription, in Chinese patients with advanced solid tumors, including relapsed SCLC. Patients with advanced solid tumors were recruited in the dose-escalation stage and received lurbinectedin in a 3 + 3 design (two cohorts: 2.5 mg/m2 and 3.2 mg/m2, IV, q3wk). The RD was expanded in the following dose-expansion stage, including relapsed SCLC patients after first-line platinum-based chemotherapy. The primary endpoints included safety profile, tolerability, DLT, RD, and preliminary efficacy profile, while the secondary endpoints included PK characteristics. In the dose-escalation stage, ten patients were included, while one patient had DLT in the 3.2 mg/m2 cohort, which was also the RD for the dose-expansion stage. At cutoff (May 31, 2022), 22 SCLC patients were treated in the ongoing dose-expansion stage, and the median follow-up was 8.1 months (range 3.0–11.7). The most common grade ≥ 3 treatment-related adverse events (TRAEs) included neutropenia (77.3%), leukopenia (63.6%), thrombocytopenia (40.9%), anemia (18.2%), and ALT increased (18.2%). The most common severe adverse events (SAEs) included neutropenia (27.3%), leukopenia (22.7%), thrombocytopenia (18.2%), and vomiting (9.1%). No treatment-related deaths occurred. The Independent Review Committee (IRC)-assessed ORR was 45.5% (95% CI 26.9–65.3). Lurbinectedin at the RD (3.2 mg/m2) showed manageable safety and acceptable tolerability in Chinese patients with advanced solid tumors, and demonstrates promising efficacy in Chinese patients with SCLC as second-line therapy. Trial registration: This study was registered with ClinicalTrials.gov NCT04638491, 20/11/2020.


Study design, participants, and treatments
This is a single-arm, open-label, multicenter, bridging study, including dose-escalation and dose-expansion stages.The main eligibility criteria are described in the Supplementary Appendix.In the dose-escalation stage, the patients recruited by the study investigator in Jilin Cancer Hospital received lurbinectedin in a classical 3 + 3 design.Two dose levels (DLs): DL1 (2.5 mg/m 2 ) and DL2 (3.2 mg/m 2 ), were evaluated as both were administered as a 1-h intravenous infusion once every 3 weeks; no dose escalation beyond DL2 was allowed.The RD was defined as the highest dose at which < 1/3 patients displayed dose-limiting toxicities (DLTs, which are detailed below) in Cycle 1.Primary granulocyte colony-stimulating factor (G-CSF) prophylaxis was not allowed during Cycle 1.If > 1 patient had neutropenia-associated DLTs in Cycle 1 at DL2, then this dose level would be used as the starting dose for a second dose escalation with primary G-CSF prophylaxis during Cycle 1.After defining RD, the study entered the dose-expansion stage, and patients were recruited by the study investigators in eight hospitals in the Chinese mainland and treated with single-agent lurbinectedin at the RD defined in the doseescalation stage (with or without G-CSF support).
In the dose-escalation stage, DLTs were defined as any adverse events (AEs) or laboratory examination abnormalities related to lurbinectedin that occurred in the first cycle and followed the required criteria (graded per NCI-CTCAE 5.0): grade 4 neutropenia for ≥ 3 days; ≥ grade 3 febrile neutropenia; ≥ grade 3 neutropenia combined with sepsis or other serious infections; grade 4 thrombocytopenia, or grade 3 thrombocytopenia complicated by obvious hemorrhage or requiring platelet transfusion; grade 4 anemia; other grades 3/4 nonehematological AEs suspected to be related with lurbinectedin except nausea/vomiting (unless no obvious remission after treatment for 2 weeks), grade 3 diarrhea for < 24 h or could be obviously relieved within 2 weeks through medical intervention, grade 3 weakness for < 5 days, allergic reactions, hair loss, and simple biochemical abnormalities unrelated with the clinical disease.
The study protocol was approved by an independent local ethics committee of each participating hospital.The study was done in accordance with the Declaration of Helsinki, the Good Clinical Practice international guidelines, and the local regulations for clinical trials.Signed informed consent was obtained from all patients before any study-specific procedure started.

Procedures
In the dose-escalation stage, patients were given lurbinectedin from 2.5 to 3.2 mg/m 2 , based on the a 3 + 3 design rule.Finally, a dose of 3.2 mg/m 2 without G-CSF support was defined as the RD for the dose-expansion stage.
In the dose-expansion stage, all patients were given a starting dose of 3.2 mg/m 2 lurbinectedin and antiemetic prophylaxis.Treatment delays and dose reductions were permitted for managing toxic effects at the investigator's discretion.However, patients requiring > 2 dose reductions (from 3.2 to 2.6 mg/m 2 and then to 2.0 mg/m 2 ) were to be withdrawn form the trial.
For both stages, all the patients received lurbinectedin until disease progression [defined by the Response Evaluation Criteria in Solid Tumors (RECIST) criteria] or unacceptable toxicity (as per the investigator's decision), except for excluded patients.The antitumor activities were evaluated using a radiological assessment [contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI) scan] in accordance with RECIST1.1, once every 6 weeks (± 7 days) after the first dose.The tumors were also evaluated by the independent review committee (IRC) through a masked review of the radiological results through de-identified images; the statistical analysis of ORR was mainly based on the IRC results (Fig. 1).

Pharmacokinetic profile evaluation
The PK sampling was conducted on 3 and 21 subjects in the 2.5 mg/m 2 and 3.2 mg/m 2 groups during the trial, respectively.For each subject, 16 and 9 blood samples were collected in the 1st and the 2nd cycles for PK analysis, respectively.

Endpoints
The primary observation endpoints included safety profile, tolerability, DLT, and RD of lurbinectedin in Chinese advanced solid tumor patients, as well as the preliminary efficacy of lurbinectedin at RD as the second-line therapy in Chinese SCLC patients, while the secondary endpoints included pharmacokinetic parameters.

Statistical analysis
The overall objective response, overall survival, and PFS were assessed in the intention-to-treat population, comprising all enrolled participants.The response duration was assessed in all the participants who had a confirmed complete or partial response (PR).Moreover, the safety profile was assessed in the as-treated population, defined as all participants who took at least one study treatment dose.
We used SAS (version 9.4) for all statistical analyses.We used the binomial method (Wilson binomial) to assess the 95% confidence interval of the overall objective response rate (ORR).We also estimated the overall survival, PFS, and response duration using the Kaplan-Meier method.
Based on a sample size of 22 subjects, if the observed ORR was 30.5%, our study would have an 85% power to ensure that the point estimate of ORR is greater than the lower bound of 95% CI for ORR from the PM1183-B-005-14 SCLC cohort study, which is 21.9%.The uncertainty of the ORR estimate was calculated by a 95% exact binomial confidence interval.

Ethics approval and consent to participate
The study was approved by the ethics committee of Jilin Cancer Hospital (202008-044-01).All patients provided written informed consent.

Results
Between November 24, 2020, and May 31, 2022, a total of 32 patients were enrolled; all were treated with lurbinectedin and included for the evaluation of safety analysis.Ten patients with various advanced solid tumors (4 breast cancer, 2 SCLC, 2 sarcomas, 1 rectal cancer, and 1 pancreatic cancer) were included in the dose-escalation stage, while 22 relapsed SCLC patients after first-line platinum-based chemotherapy were enrolled in the doseexpansion stage, respectively.One baseline CNS metastasis patient was included in the dose-expansion stage because of the missed diagnosis of a minor metastatic lesion in the MRI scan of the brain during the screening.
The investigators judged this patient as a potential beneficiary of lurbinectedin and thus continued the treatment, albeit this case was considered a protocol deviation.Among the 22 patients, 17 (77.3%)were males, and 20 (90.9%) of them were diagnosed at the extensive stage; the median age was 58; the chemotherapy-free interval (CTFI) was < 90 days in 8 (36.4%) patients and > 90 days in 14 (63.6%)patients.There were 14/22 (63.6%) patients who previously received immunotherapy before their dose-expansion stage enrollment (Tables 1, 2).At cutoff, the treatment cycles comprising total and median cycles were 43, 3 (range 1-14) and 132, 6 (range 1-12) in the dose-escalation and dose-expansion stages, respectively.The median total exposure dosage was 13.3 mg (range 5-89.6 mg) and 28.4 mg (range 5.4-74.5 mg) in the dose-escalation and dose-expansion stages, respectively.In the dose-escalation stage, no patients received ≥ 6 cycles in the 2.5 mg/m 2 cohort, but 3/7 (42.9%) patients received ≥ 6 cycles in the 3.2 mg/m 2 cohort.There were 12/22 (54.5%) patients who received ≥ 6 cycles in the dose-expansion stage.No dose reduction occurred in the dose-escalation stage, while the doses were reduced in 7 (31.8%)patients and 8 (6.1%) cycles in the dose-expansion stage because of treatment-related AEs.Similarly, treatment-related dose delay occurred in 2 (20.0%) patients and 2 (4.7%) cycles, as well as 8 (36.4%) patients, and 12 (9.1%)cycles in the dose-escalation and dose-expansion stages, respectively.The most common causes for both the dose reduction and delay were hematological toxicities (neutropenia, leukopenia, and thrombocytopenia).Moreover, two patients in the 3.2 mg/m 2 cohort in the dose-escalation stage discontinued treatment because of treatment-emergent adverse events (TEAEs) (gastrointestinal disorders and hepatotoxicity), while only 1 (4.5%) patient in the dose-expansion stage discontinued the treatment due to TEAEs (grade 4 ALT increased and grade 3 AST increased).

Discussion
Our study was the first clinical study conducted on lurbinectedin in Chinese patients, as well as the first doseexpansion study with an FDA-approved dosage (3.2 mg/m 2 , 1 h IV q3wk, without G-CSF primary prophylaxis) in Asian SCLC patients as second-line therapy.
The safety profile analysis in the dose-escalation stage suggested that hematological toxicities (e.g., neutropenia, leucopenia, and thrombocytopenia) were the main adverse reactions; however, they were manageable and tolerable as well as coincident with the findings from other studies.Notably, a case of grade 3 atrial fibrillation was reported as an SAE in the 2.5 mg/m 2 group.However, based on a previous study (NCT02451007), significant effects of lurbinectedin administration on the QT interval of patients with solid malignancies at a dose of 3.2 mg/m 2 q3wk have been ruled out 15 .The lurbinectedin's safety profiles observed in our dose-expansion stage were similar to the basket trial results of the SCLC cohort, although with higher incidences of hematological and liver disorders.The different incidences may be attributed to the ethnic differences between the two studies and could be explained by the characteristic antitumor and metabolic activities of lurbinectedin in the body.Besides, according to the Investigator's Brochure of Lurbinectedin (Aug, 2023), elevated creatine phosphokinase (CPK) and muscular adverse events have been reported in lurbinectedin historical studies.Specifically, CPK increases were reported in 9.2% of patients treated with single-agent lurbinectedin at 3.2 mg/m 2 q3wk in phase II and III studies.Of these, only two patients (0.4%) had grade 3 CPK increase.Episodes of rhabdomyolysis (grade 3 and grade 4) were reported in two patients treated with single-agent lurbinectedin, but not at the RD of 3.2 mg/m 2 www.nature.com/scientificreports/q3wk.Consistently, in our study, only two cases of CPK increased (both grade1), one case of muscular weakness (grade 1), and one case of limb pain (grade 1) were reported.No case of rhabdomyolysis was reported in our study.Since no new safety limitations were found, lurbinectedin's toxicities in our study were manageable and acceptable, as evidenced by the majority of the adverse reactions being relieved by appropriate therapeutic management.The incidences of treatment-related dose delay and reduction were also acceptable in our study.Moreover, the median cycles administered in our dose-expansion stage were similar to that in the basket trial (6 vs. 4).Notably, we did not include G-CSF primary prophylaxis, but it could be considered in future clinical practices, as G-CSF support could promote better tolerability 16 .The efficacy of lurbinectedin was evaluated in both dose-escalation and dose-expansion stages, but mainly in the latter category, i.e., in the Chinese SCLC population, as second-line therapy.Regarding the efficacy evaluation in the dose-escalation stage, it was noted that two sarcoma patients benefited from lurbinectedin administration at 3.2 mg/m 2 , i.e., one soft tissue sarcoma patient had a confirmed PR as the best response, and another one with renal leiomyosarcoma displayed long-term stable disease.These two encouraging results were different from a previous phase II study results (NCT02448537), in which 42 metastatic soft tissue sarcoma patients (mostly leiomyosarcoma) were enrolled; however, 12 of them after treatment with lurbinectedin 3.2 mg/m 2 did not respond.This might become a reminder of the indications of lurbinectedin development in the future.In the dose-expansion stage, all 22 enrolled patients showed a relapse after first-line platinum-based chemotherapy and were treated with the same dosage (i.e., 3.2 mg/m 2 , 1 h IV, q3wk) as the phase II basket trial.At the cutoff date, the result showed encouraging responses with an ORR of 45.5% (10/22, 95% CI 26.9-65.3)based on both IRC and investigator assessments, which was similar to the phase II basket trial result (35.2% by the investigator; 30.5% by IRC), and higher than the ORR (16.9%) 17 reported with topotecan which was the only evidence-based standard of care in SCLC second-line therapy before the FDA's approval of lurbinectedin.Notably, based on multiple small sample studies, the estimated ORR of topotecan in Chinese SCLC patients as second-line therapy ranged from 5%-30%.Other efficacy endpoint results in our study, e.g., mDOR, DCR, mPFS, and mOS, also presented similar or superior trends compared to the basket trial results (IRC-assessed) and the historical data of Table 3. TEAEs worst grade ≥ 3 per patient in dose-escalation stage and dose-expansion stage.TEAE treatment-emergent adverse events, MedDRA medical dictionary for regulatory activities (v.25.0),SOC system organ class, PT preferred term.For SCLC patients having a first-line therapy relapse, CTFI is considered the strongest predictor of outcome, as patients with the sensitive disease (CTFI ≥ 90 days) have a tumor response rate of 25% to additional chemotherapy, whereas patients with the resistant disease (CTFI < 90 days) exhibit tumor response rates < 10% 18 .The CTFI subgroup analysis in our study demonstrated that the lurbinectedin response in SCLC patients with the sensitive disease (CTFI ≥ 90 days), was better than that in the resistant population (CTFI < 90 days): 50% vs. 37.5%, by IRC and investigator assessments.This finding was consistent with (or even superior to) the basket trial results (sensitive vs. resistant: 43.3% vs. 13.3%, by IRC and 45.0% vs. 22.2%, by the investigator).Furthermore, our survival results (mOS, sensitive vs. resistant: 11.0 months vs. 9.2 months) were similar to the basket trial results (mOS, sensitive vs. resistant: 11.9 months vs. 5.0 months).
Immunotherapy in SCLC has considerably improved, as evidenced by the FDA's approval for atezolizumab (2019) 19 or durvalumab (2020) 20 in combination with chemotherapy for the first-line therapy of extensive-stage SCLC 12 .The SCLC patients' percentage who had prior immunotherapy (e.g., PD-1/PD-L1 inhibitors) in our study (dose-expansion stage) was much higher than the basket trial result (SCLC cohort): 63.6% (14/22) vs. 7.6% (8/105), which could be attributed to the different timings of both the studies.In patients who had prior immunotherapy, 35.7% (5/14) patients in our study and 62.5% (5/8) patients in the basket trial had responses (PRs, IRC assessed).Additionally, these results might indicate greater lurbinectedin's efficacy in patients who had prior immunotherapy but should be interpreted with caution due to the limited sample size in both studies 4 .
Regarding the PK profile, lurbinectedin's exposure in plasma increased with increased dosage, and the plasma concentration reached C max at the end of infusion and decreased rapidly after the infusion.With an administration interval of 21 days, the plasma concentration before the second administration was below the Lower Limit of Quantification, and no drug accumulation was observed after two administration cycles.Compared with the basket trial results, the PK profile in Chinese patients was similar to that of Caucasian patients.Due to the small sample size, no exploration was conducted in our study to evaluate the relationships between lurbinectedin exposure and clinical safety and efficacy in Chinese SCLC patients.However, the exposure-response (E-R) analyses have been done by Pharma Mar to determine the correlation between lurbinectedin exposure and safety endpoints from phase I to III studies (n = 692) and also with efficacy endpoints from study B-005 (n = 99), and the results supported a favorable benefit-risk profile for lurbinectedin 3.2 mg/m 2 q3wk 21 .
Since this was a bridging study with a small sample size, the potential bias cannot be excluded.However, as per the ICH guidance "E5 Ethnic Factors in the Acceptability of Foreign Clinical Data", this study was designed by minimizing the duplication of clinical studies and supplying medicines expeditiously to patients for their benefit.In this regard, this study might have concluded at this stage.However, further evaluations of lurbinectedin in a broader population are warranted for greater validation.

Conclusions
Our study met its primary efficacy endpoint, i.e., IRC-assessed ORR in the dose-expansion stage, which was higher than the phase II basket trial result of the SCLC cohort and much superior to the topotecan's available data.The survival benefits of first relapsed Chinese SCLC patients from lurbinectedin treatment (3.2 mg/m 2 , 1 h IV, q3wk) were also observed, albeit some data were immature at cutoff.Together with a manageable and acceptable safety profile, our encouraging results can pave the way for lurbinectedin's application as second-line therapy in Chinese adult patients with SCLC.

Figure 1 .
Figure 1.Study design and subject flow.PK pharmacokinetics.

Table 1 .
Baseline characteristics in the dose-escalation stage.BMI body mass index, ECOG Eastern Cooperative Oncology Group, IRC Independent Review Committee, CNS central nervous system, SCLC Small Cell Lung Cancer. 2.

Table 2 .
Baseline characteristics in the dose-expansion stage.BMI body mass index, ECOG Eastern Cooperative Oncology Group, IRC Independent Review Committee, CNS central nervous system, CTFI chemotherapy-free interval.

Table 4 .
Overall efficacy of lurbinectedin treatment assessed by IRC and the investigator in the doseescalation stage.RECIST response evaluation criteria in solid tumors, IRC Independent Review Committee.*Patients censored due to disease not progressed or still being alive.

Table 5 .
Efficacy assessed by IRC and the investigator in the dose-expansion stage.RECIST response evaluation criteria in solid tumors, IRC Independent Review Committee, CTFI chemotherapy-free interval; one patient was not evaluable because of withdrawal.*Patients censored due to disease not progressed or still being alive.topotecan 17 : 4.2 months vs. 5.1 months vs. 4.2 months, 90.9% vs. 61.9% vs. 61.5%,5.6 months vs. 3.5 months vs. 3.5 months, and 11.0 months vs. 9.3 months vs. 7.8 months, although some of our results were immature at cutoff.