Phase I study of lurbinectedin in combination with weekly paclitaxel with or without bevacizumab in patients with advanced solid tumors

Lurbinectedin and paclitaxel showed synergism in preclinical studies and have non-completely overlapping toxicity profiles. This phase I trial evaluated a combination of paclitaxel and lurbinectedin with/without bevacizumab in advanced tumors. This trial was divided into Group A, which evaluated weekly paclitaxel (60 or 80 mg) plus lurbinectedin (3.0–5.0 mg flat dose [FD] or 2.2 mg/m2) every 3 weeks in advanced solid tumors; and Group B, which evaluated bevacizumab (BEV, 15 mg/kg) added to the recommended dose (RD) defined in Group A in advanced epithelial ovarian or non-small cell lung cancer (NSCLC). 67 patients (A, n = 55; B, n = 12) were treated. The RD was paclitaxel 80 mg/m2 on Day (D)1,D8 plus lurbinectedin 2.2 mg/m2 on D1. At this RD, myelotoxicity was reversible and manageable, and most non-hematological toxicities were mild/moderate. Adding BEV did not notably change tolerability. Twenty-five confirmed responses were observed: 20/51 evaluable patients in Group A (overall response rate [ORR] = 39% at all dose levels and at the RD), and 5/10 evaluable patients in Group B (ORR = 50%). Most responders had breast (n = 7/12 patients), small cell lung (SCLC) (n = 5/7), epithelial ovarian (n = 3/9) and endometrial cancer (n = 3/11) in Group A, and epithelial ovarian (n = 3/4) and NSCLC (n = 2/6) in Group B. Clinical benefit rate was 61% in Group A (58% at the RD), and 90% in Group B. No major pharmacokinetic drug-drug interactions were observed. Paclitaxel/lurbinectedin and paclitaxel/lurbinectedin/BEV are feasible combinations. Further development is warranted of paclitaxel/lurbinectedin in SCLC, breast, and endometrial cancer, and of paclitaxel/lurbinectedin/BEV in epithelial ovarian cancer. Supplementary Information The online version contains supplementary material available at 10.1007/s10637-022-01281-z.


Introduction
Lurbinectedin (Zepzelca ® ) is a synthetic tetrahydroisoquinoline alkaloid structurally related to trabectedin. It inhibits oncogenic transcription primarily through binding to the exocyclic amino group of guanine-rich DNA sequences around promoters of protein-coding genes, thereby altering the 3D DNA structure and evicting oncogenic transcription factors from their binding sites, thus halting their aberrant transcription programs [1][2][3]. Lurbinectedin adducts can stop transcribing (phosphorylated) RNA polymerase II, decreasing mRNA synthesis and inducing the ubiquitination and degradation of RNA polymerase II inhibition [4]. Lurbinectedin adducts may also trick the nucleotide excision repair system, favoring the production of DNA double-strand breaks and triggering apoptotic cell death [5]. Lurbinectedin monotherapy has been approved in the U.S. and other countries for the treatment of adult patients with metastatic small cell lung cancer (SCLC) and disease progression on or after platinum-based chemotherapy.
The first-in-human phase I study defined a recommended dose (RD) of 7 mg flat dose (FD) for single-agent lurbinectedin as a 1-h intravenous (i.v.) every 3 weeks (q3wk) [6]. Severe but transient neutropenia, and mild fatigue, nausea and vomiting, were common at this RD. The pharmacokinetic (PK) profile of lurbinectedin showed dose linearity, high interpatient variability, and a long median half-life (70.6 h at the RD).
Preclinical studies showed improved antitumor activity for lurbinectedin with taxanes. In vivo, synergism was observed with lurbinectedin and paclitaxel in mice bearing gastric, ovarian, non-small cell lung cancer (NSCLC), breast or prostate xenografted tumors [7]. Lurbinectedin and paclitaxel have toxicity profiles that are not completely overlapping. Both are cytochrome CYP3A4 substrates; hence, PK interactions by competitive inhibition of this enzyme system cannot be discarded.
Bevacizumab (BEV) is a humanized monoclonal antibody against the circulating vascular endothelial growth factor that inhibits tumor angiogenesis. BEV primarily acts in the tumor microenviroment, with very little hematological toxicity. The combination of chemotherapy with BEV has been associated with improved clinical activity compared with chemotherapy alone [8][9][10][11][12][13].
The aim of this phase I study was to determine the maximum tolerated dose (MTD) and the RD, safety profile, activity and PK of lurbinectedin combined with weekly paclitaxel, with or without BEV, in advanced solid tumor patients.

Patients and methods
Supplementary Information includes details regarding study design and eligibility criteria. In brief, patient accrual began in Group A (paclitaxel and lurbinectedin) at a starting dose of paclitaxel 60 mg/m 2 plus lurbinectedin 3.0 mg flat dose (FD). Treatment initially consisted of escalating doses of paclitaxel as 1-h i.v. infusions on Day (D) 1, D8 and D15, followed by lurbinectedin as a 1-h i.v. infusion on D1, both every three weeks (q3wk). After DL3, the paclitaxel schedule was changed to 1-h i.v. infusions on D1 and D8 q3wk owing to a high incidence of D15 dose omissions in DL3. Furthermore, during dose escalation lurbinectedin was converted to a body surface area (BSA)-based dose. Once the RD for paclitaxel and lurbinectedin had been determined in Group A, patients were enrolled in Group B and received this RD supplemented with BEV 15 mg/kg as a 30-90 min i.v. infusion on D1 q3wk.
In both groups, paclitaxel was discontinued after Cycle 6; patients still on treatment continued receiving lurbinectedin alone (Group A) or lurbinectedin plus BEV (Group B) at the same dose. Treatment was administered until disease progression, unacceptable toxicity, intercurrent illness precluding study continuation, patient refusal and/or non-compliance with study requirements, treatment delay > 15 days (except if with clear clinical benefit), and > 2 dose reductions.

Study assessments
Supplementary Information includes definitions for doselimiting toxicities (DLTs) and details regarding safety, efficacy and PK assessments.
Adverse events (AEs) and laboratory abnormalities were graded with the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) v.4 [14], and coded using the Medical Dictionary for Regulatory Activities (MedDRA) v. 16.0. Antitumor activity was evaluated every three cycles according to the RECIST v.1.1 [15]. The PK analysis was conducted on 12 blood samples collected from each patient during Cycle 1 to quantify lurbinectedin and paclitaxel plasma concentrations.

Statistical analysis
Continuous variables were presented with summary statistics and categorical variables in frequency tables. Timeto-event variables were calculated using Kaplan-Meier approach. Binomial exact distribution was used to calculate 95% confidence intervals (95%CIs) for categorical variables. Blood and plasma concentration-time profiles were analyzed by standard non-compartmental methods. Individual PK parameters were tabulated and summarized.

Patient characteristics
Sixty-nine patients were enrolled in this study: 55 in Group A (paclitaxel/lurbinectedin) and 14 in Group B (paclitaxel/ lurbinectedin/BEV). Patient characteristics at baseline are summarized in Table 1.
Most of the 55 patients (73%) enrolled at all dose levels in Group A were female, with median age 57 years (range, 31-74 years). The most common primary tumors comprised endometrial (n = 13 patients, 24%), breast (n = 12, 22%), epithelial ovarian, NSCLC (n = 9, 16% each) and SCLC (n = 7, 13%). The median number of metastatic sites was 2 (range, 1-6 sites). The median number of lines of all prior therapies for advanced disease was 2 (range, 0-7 lines), and the most common prior therapies were platinum compounds (82%) and taxanes (60%). Thirty-seven patients in Group A were enrolled at the RD. Twenty-six of these patients (70%) were female, with a median age of 61 years (range, 38-74 years). Their most common primary tumors were endometrial (35% of patients), breast (22%), NSCLC (16%), SCLC (14%) and epithelial ovarian (8%). The median number of metastatic   sites was 2 (range, 1-6 sites) and the median number of lines of prior therapy for advanced disease was 1 (range, 0-5 lines), with the most common prior therapies being platinum compounds (87%) and taxanes (65%). Most of the 14 patients (57%) enrolled in Group B were also female. The median age was 57 years (range, 47-72 years). All patients had epithelial ovarian cancer or NSCLC (n = 7 patients each). The median number of metastatic sites was 3 (range, 1-6 sites), and the median number of lines of prior therapy for advanced disease was 2 (range, 0-4 lines). All 14 patients had been pre-treated with platinum compounds; other prior therapies were taxanes (67%) and folic acid analogues (58%).

Treatment administration
In Group A, 55 enrolled patients were treated at five dose levels: DL1 (n = 3 patients), DL2 (n = 3) and DL3 (n = 6) with paclitaxel on D1, D8 and D15 q3wk, and DL4 (n = 6) and DL5 (n = 37) with paclitaxel on D1 and D8 q3wk (see Supplementary Information). A total of 392 treatment cycles were administered at all dose levels (median: 5.0 cycles per patient); 20 patients (36%) were still on treatment after Cycle 6 and switched to single-agent lurbinectedin. At DL5 (the RD), 256 treatment cycles were given (median: 5.0 cycles per patient) and 15 patients (41%) switched to single-agent lurbinectedin after Cycle 6. Most treatment discontinuations (44 of 55 patients [80%] at all dose levels; 30 of 37 patients [81%] at DL5 [the RD]) were due to disease progression. No treatment-related discontinuations occurred at the RD. The median time on treatment was 4.8 months, both at all dose levels and at the RD. The median dose intensities (DIs) at the RD were 49.1 mg/m 2 /week for paclitaxel and 0.7 mg/ m 2 /week for lurbinectedin, and the median relative DIs compared to the initially planned dose were 92% and 98%, respectively. At the RD, 6 patients (16%) had cycle delays and 7 patients (19%) had study drug dose reductions due to treatment-related reasons (mostly hematological toxicity).
In Group B, 12 of the 14 enrolled patients received a total of 111 treatment cycles at a dose of paclitaxel 80 mg/m 2 on D1 and D8 and lurbinectedin 2.2 mg/m 2 on D1 q3wk supplemented with BEV 15 mg/kg on D1 q3wk (median: 9.5 cycles per patient). Two patients were withdrawn from the study before receiving the first dose due to disease-related bowel obstruction (n = 1), and patient refusal (n = 1). Eight patients (67%) were still on treatment after Cycle 6 and switched to lurbinectedin and BEV. Most treatment discontinuations (8 of 12 patients [67%]) were due to disease progression. One discontinuation was due to treatment-related thrombocytopenia. The median time on treatment was 7.9 months. The median DIs were 50.3 mg/ m 2 /week for paclitaxel, 0.7 mg/m 2 /week for lurbinectedin and 5.0 mg/m 2 /week for BEV; the median relative DIs were 94%, 100% and 99%, respectively. Three patients (25%) had cycle delays and 2 patients (17%) had study drug dose reductions owing to treatment-related reasons (mostly hematological toxicity).

Dose escalation and recommended dose
Fifty-two treated patients in Group A were evaluable for DLTs (see Supplemental information). Three patients were non-evaluable because they did not receive a complete Cycle 1 due to disease-related events (n = 2), or because of lack of laboratory assessment for DLT evaluation during Cycle 1 (n = 1). Paclitaxel was initially administered at a schedule of D1, D8 and D15 q3wk. No DLTs occurred at the first two dose levels. Two of 6 patients at DL3 (paclitaxel 60 mg/ m 2 , lurbinectedin 5.0 mg FD) had delayed DLTs (grade 3 or 4 neutropenia after Cycle 1). After DL3, the paclitaxel schedule was changed to D1 and D8 q3wk owing to the finding of a high incidence (4 of 6 patients) of paclitaxel dose omissions due to treatment-related neutropenia during

Safety
All treated patients were evaluable for safety. Treatmentrelated AEs and laboratory abnormalities at the RD in Group A and in Group B are shown in Table 2.
The best response to paclitaxel and lurbinectedin with or without BEV was compared with response to the last prior therapy in evaluable patients treated at the RD in Group A and in Group B. Overall, 9 of 33 patients (27%) at the RD in Group A and 6 of 10 patients (60%) in Group B showed greater antitumor activity with paclitaxel and lurbinectedin with or without BEV compared to the last prior therapy (Fig. 2). The tumor types of these 15 patients were NSCLC (n = 6, all in Group B), breast, SCLC (n = 3 each), epithelial ovarian (n = 2), and endometrial cancer (n = 1).
Clinical benefit for > 12 months was observed in 5 patients: 4 in Group A (3 at the RD) and one in Group B. These patients had SCLC (n = 2), epithelial ovarian, endometrial cancer, and NSCLC (n = 1 each), had received 1-3 prior chemotherapy lines, and were given 18-35 cycles of study treatment each (all with PR as best response). Four of these patients showed no signs of disease progression prior to study termination.

Pharmacokinetics
All patients were sampled for PK analysis and were suitable for non-compartmental analysis (NCA). Parameters obtained for paclitaxel and lurbinectedin at each dose level are shown in Supplemental Information. Wide variability was observed in lurbinectedin and paclitaxel total clearance (CL). No dose linearity could be established for lurbinectedin maximum concentration (C max ) and area under the concentration-time curve (AUC). Potential drug-drug interactions (DDIs) between paclitaxel and lurbinectedin could not be fully ruled out, since a slight decrease in the CL of each drug was observed at the high AUC of the other drug. No statistically significant differences were observed in the PK parameters of paclitaxel and lurbinectedin in the presence or absence of BEV. Additional details are provided in Supplemental Information.

Discussion
This clinical trial defined the RD for phase II studies of paclitaxel plus lurbinectedin combination at paclitaxel 80 mg/m 2 on D1 and D8, and lurbinectedin 2.2 mg/m 2 on D1, q3wk.
The safety profile of the paclitaxel plus lurbinectedin combination was predictable. Myelotoxicity at the RD was common but reversible and manageable, and most nonhematological toxicities were mild/moderate. As expected, severe myelotoxicity and some non-hematological toxicities (vomiting, and especially neuropathy and alopecia) were slightly more frequent at this RD compared to the established dose for lurbinectedin monotherapy (3.2 mg/m 2 q3wk) [16,17]. This may be attributed to the addition of paclitaxel, as these toxicities are commonly reported with single-agent paclitaxel [18][19][20]. No patients treated with the combination discontinued treatment due to toxicity, thereby further suggesting an acceptable safety profile.
As expected, the addition of BEV 15 mg/kg q3wk to this RD increased the incidence of some treatment-related Fig. 2 Swimmer plot of best response as per RECIST to study treatment vs. last prior therapy at the RD in Group A (paclitaxel plus lurbinectedin) (n = 33) and in Group B (paclitaxel plus lurbinectedin and BEV) (n = 10). The tumor type, last prior therapy, and total number of prior lines (in parenthesis) of each patient is shown at the left of the figure. BEV, bevacizumab; CAV, cyclophosphamide, doxorubicin and vincristine; CR, complete response; CSF1R, colonystimulating factor-1 receptor; FGFR, fibroblast growth factor receptor; HSP, heat shock protein; NA, not available; NE, not evaluable; NSCLC, non-small cell lung cancer; PD, progressive disease; PDL1, programmed death ligand-1; PFS, progression-free survival; PLD, pegylated liposomal doxorubicin; PR, partial response; RD, recommended dose; RECIST, Response Evaluation Criteria In Solid Tumors; SCLC, small cell lung cancer; SD, stable disease; TTP, time to progression; UK, unknown events (e.g., hypertension, gastrointestinal events, embolism and febrile neutropenia), and severe neutropenia. Similar increases have been reported in clinical trials comparing chemotherapy plus BEV vs. chemotherapy alone in patients with solid tumors [8,11,21,22]. Despite this additional toxicity, which was manageable, the triple combination of paclitaxel, lurbinectedin and BEV was generally well tolerated.
The addition of BEV improved response rate to the combination at the RD in patients with epithelial ovarian cancer (from 33 to 75%) and NSCLC (from 25 to 33%). This is in line with the finding, in previous studies, of a 15-20% improvement in ORR with the addition of BEV to chemotherapy vs. chemotherapy alone in patients with recurrent ovarian cancer [12,22,35] and previously untreated NSCLC [9,36,37].
The PK parameters of paclitaxel and lurbinectedin in the present study were generally similar to those reported elsewhere [6,38]. The absence of linearity found for lurbinectedin C max and AUC was probably due to the dose levels explored being very close. DDIs between paclitaxel and lurbinectedin could not be ruled out, although their clinical relevance would be marginal, based on the slight changes in CL and the large variability of PK parameters observed. BEV had no significantly effects on the PK profile of either lurbinectedin or paclitaxel.
In conclusion, weekly paclitaxel 80 mg/m 2 on D1 and D8 combined with lurbinectedin 2.2 mg/m 2 on D1 q3wk, with or without the addition of BEV 15 mg/kg on D1, showed a manageable overall safety profile and promising antitumor activity in patients with selected advanced solid tumors. These results support further development of this combination without BEV in the treatment of SCLC, breast, and endometrial cancer, and with added BEV in the treatment of epithelial ovarian cancer.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.