first-in-human phase i study of lurbinectedin (pm01183) in ... · cancer therapy: clinical...

Cancer Therapy: Clinical

First-In-Human Phase I Study of Lurbinectedin (PM01183) inPatients with Advanced Solid Tumors

María Elena Elez1, Josep Tabernero1, David Geary3, Teresa Macarulla1, S. Peter Kang3, Carmen Kahatt2,Arturo Soto-Matos Pita2, Carlos Fernandez Teruel2, Mariano Siguero2, Martin Cullell-Young2,Sergio Szyldergemajn2, and Mark J. Ratain3

AbstractPurpose: Lurbinectedin (PM01183) binds covalently to DNA and has broad activity against tumor cell

lines. This first-in-human phase I study evaluated dose-limiting toxicities (DLT) and defined a phase II

recommended dose for PM01183 as a 1-hour intravenous infusion every three weeks (q3wk).

Experimental Design: Thirty-one patients with advanced solid tumors received escalating doses of

PM01183 following an accelerated titration design.

Results: PM01183was safely escalated over 200-fold, from0.02 to5.0mg/m2.Dosedoublingwas utilized,

requiring15patients andninedose levels to identifyDLT.The recommendeddosewas 4.0mg/m2,withoneof

15 patients havingDLT (grade 4 thrombocytopenia). Clearancewas independent of body surface area; thus, a

flat dose of 7.0 mg was used during expansion. Myelosuppression, mostly grade 4 neutropenia, occurred in

40% of patients but was transient and manageable, and none was febrile. All other toxicity was mild and

fatigue, nausea and vomitingwere themost common at the recommendeddose. Pharmacokinetic parameters

showedhigh interindividual variation, though linearitywasobserved.Ator above the recommendeddose, the

myelosuppressive effect was significantly associated with the area under the concentration-time curve from

time zero to infinity (white blood cells, P¼ 0.0007; absolute neutrophil count, P¼ 0.016). A partial response

was observed in one patient with pancreatic adenocarcinoma at the recommended dose.

Conclusion:Aflat dose of 7.0mg is the recommended dose for PM01183 as a 1-hour infusion q3wk. This

dose is tolerated and active. Severe neutropenia occurred at this dose, although it was transient and with no

clinical consequences in this study. Clin Cancer Res; 20(8); 2205–14. �2014 AACR.

IntroductionLurbinectedin (PM01183; PharmaMar) is a novel syn-

thetic tetrahydroisoquinoline structurally related to ectei-nascidins (1, 2). PM01183 showed in vitro broad antitumoractivity against several representative human cell lines, withhalf-maximal inhibitory growth concentrations (IG50)within the picomolar range (3). In vivo studies showedsignificant growth inhibition of a wide variety of humancancer xenografts in athymic mice (3).PM01183 binds covalently to the DNA minor groove,

forming adducts that are capable of inducing mainlydouble-strand breaks, but also interfering with normalprotein synthesis machinery at the mRNA level. Accumu-lation of DNA damage delays progression through the

S–G2 phase of the cell cycle and ultimately triggers apo-ptotic cell death (4–7).

Preclinical toxicology studies revealed that PM01183primarily caused myelosuppression, inducing anemia,bone marrow depletion, and atrophy of the lymphoidsystem. Other toxicities were species dependent, such ashepatotoxicity, which was not observed in beagle dogs.Effects tended to be dose dependent and transient, althoughthe maximum-tolerated dose (MTD), when normalized tobody surface area (BSA), differed significantly from onespecies to another. The female Sprague-Dawley rat wasthe most sensitive of all species tested, with a MTD of0.18 mg/m2. As recommended by the International Con-ference of Harmonization (ICH; both European MedicinesAgency andU.S. FoodandDrugAdministration) guidelines(8–10), approximately one tenth of this dosewas selected asa safe starting dose for this first-in-human study.

Given its broad in vitro activity against human solid tumorcell lines and its in vivo activity in xenografted humantumors, together with its acceptable and predictable pre-clinical toxicology profile, an accelerated titration designwas used in this first-in-human, phase I clinical trial to sparemost patients frombeing treated at potentially subclinicallyinactive doses. The trial was aimed at identifying the dose-limiting toxicities (DLT) and determining the MTD and the

Authors' Affiliations: 1Vall d'HebronUniversityHospital andVall d'HebronInstitute of Oncology (VHIO), Universitat Aut�onoma de Barcelona, Barce-lona; 2PharmaMar, S.A., Colmenar Viejo, Madrid, Spain; and 3Departmentof Medicine, The University of Chicago, Chicago, Illinois

Corresponding Author: Mark J. Ratain, 5841 S. Maryland Avenue,MC2115, University of Chicago Medicine, Chicago, IL 60637. Phone:773-702-4400; Fax: 773-702-3969; E-mail:[email protected]

doi: 10.1158/1078-0432.CCR-13-1880

�2014 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 2205

on January 16, 2021. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 21, 2014; DOI: 10.1158/1078-0432.CCR-13-1880

http://clincancerres.aacrjournals.org/

recommended dose for phase II trials when PM01183 isadministered intravenously over one hour every 3 weeks(q3wk) to patients with advanced solid tumors. The phar-macokinetic profile, safety, and preliminary antitumoractivity were secondary objectives.

Materials and MethodsThe study was conducted at the Vall d’Hebron University

Hospital (Barcelona, Spain) and the University of Chicago(Chicago, IL) in accordance with the ICH Good ClinicalPractice guidelines, and was approved by the respectiveResearch Ethics Committees.Written informed consent wasobtained before any procedure. The study is registered atClinicalTrials.gov as NCT00877474.

Eligibility criteriaEligible patients were ages �18 years with histologically

proven advanced solid tumor refractory to standard therapyor with no such therapy available; had recovered fromprevious toxicities to grade �1 (excluding alopecia); andhad Eastern Cooperative Oncology Group performancestatus (ECOG PS) �2 and adequate bone marrow, hepatic,and renal function.

Exclusion criteriaPatients were excluded if they had received any other

antitumor therapy within 3 weeks before inclusion, couldachieve tumor-free survival or significant palliation bysurgery, had progressive central nervous systemmetastases,were immunocompromised, were pregnant or lactatingwomen, or had increased cardiac risk, significant neurologicor psychiatric disorders, active infection, significant non-

neoplastic liver disease, or known hypersensitivity to anydrug component.

Study treatmentPM01183 was supplied as a powder concentrate in 0.2

and 1.0 mg vials and reconstituted in 1 and 2 mL of sterilewater for injection, respectively. The reconstituted solutionwas further diluted with 5% dextrose solution for infusion(0.2mg vials) or either 5%glucose or 0.9% sodium chloridesolution for infusion (1.0mg vials). Antiemetic prophylaxiswas not allowed before nausea and/or vomiting were deter-mined to be treatment-related toxicities. Intrapatient doseescalation was not allowed under any circumstances. Doselevels were rounded to integer numbers if appropriate.Treatment was administered until disease progression,unacceptable toxicity, treatment delay >2 weeks for anyreason (except in cases of obvious patient benefit), inter-current illness that precluded continuation of the study,patient refusal, and/or noncompliance with studyrequirements.

Dose escalationThe starting dose chosen for this first-in-human studywas

0.02 mg/m2. According to the accelerated dose-escalationdesign defined for this study, cohorts of at least one evalu-able patient were assigned to each dose level and doseescalation proceeded at 100% dose increments, as previ-ously used (11). In the event of clinically relevant grade�2toxicities in any patient in any cohort during the acceleratedtitration phase, cohorts of at least three patients were to betreated and up to 50% dose escalation was allowed if noDLTs had occurred. Once a DLT was observed at any doselevel, the design switched to a classical 3þ3 design withfurther dose increases limited at a maximum of 25% incre-ments, provided the MTD had not been reached yet.

Dose-limiting toxicitiesDLTs were defined as: grade 4 neutropenia for >7 days or

with fever (�38.5�C), sepsis or severe infection, grade 4thrombocytopenia, any grade �3 nonhematologic toxicity(excluding nausea/vomiting without antiemetic treatment,grade 3 diarrhea lasting <24 hours without treatment, grade3 asthenia for <5 days, hypersensitivity reactions, andnonclinically relevant isolated biochemical abnormalities),>2-week delay in a subsequent PM01183 administrationdue to toxicity. Following the EuropeanMedicines Agency’sGuidance on the Evaluation of Anticancer Medicinal Pro-ducts in Man (CPMP/EWP/205/95), the MTDwas prospec-tively defined as the lowest dose level explored at which onethird or more of patients experienced DLTs; and the recom-mended dose was the highest dose level explored below theMTDwith less than one third of patients experiencing DLTsin cycle 1.

Study assessmentsAt baseline, the patient’s relevant medical history was

determined. A complete physical examination was con-ducted, including the ECOG performance status score

Translational RelevanceLurbinectedin (PM01183) is a novel synthetic tetra-

hydroisoquinolinewithbroad antitumor activity in vitro.This first-in-human study defined 7.0 mg flat dose as thephase II recommended dose with PM01183 adminis-tered as 1-hour intravenous infusion every 3 weeks. Theprimary toxicity at the recommended dose was myelo-suppression, mainly consisting of nonfebrile grade 4neutropenia in 40% of patients. Despite this severeneutropenia, episodes were transient and shortlasting,and had limited consequences on treatment schedule.Other common PM01183-related adverse events at therecommended dose were mild/moderate fatigue, nau-sea, and vomiting. The pharmacokinetic profile showeddose proportionality within the dose range explored andno drug accumulation between cycles. One confirmedpartial response as per Response Evaluation Criteria inSolid Tumors v.1.0 with tumor marker normalizationwas found in a previously refractory pancreatic cancerpatient. Prolonged disease stabilizationwith evidence oftumor shrinkage occurred in three patients with soft-tissue synovial sarcomas and malignant melanoma.

Elez et al.

Clin Cancer Res; 20(8) April 15, 2014 Clinical Cancer Research2206




assessment both at baseline and at the beginning of eachtreatment cycle. Hematology tests [differential white bloodcells (WBC), hemoglobin, and platelets] and some bio-chemistry tests (liver function tests, creatinine, serum elec-trolytes, glucose, blood urea nitrogen, uric acid, and crea-tine phosphokinase; CPK) were conducted at baseline andweekly throughout the treatment period. Other biochem-istry tests (total protein, albumin,a1-acid glycoprotein, andC-reactive protein), coagulation tests, and urinalyses wereperformed at baseline and at the beginning of each treat-ment cycle. If any grade �3 laboratory abnormalitiesoccurred, the hematologic counts or appropriate chemistrytest(s) were repeated at least every 2 to 3 days until recovery.Electrocardiograms and left ventricular ejection fractionassessments were conducted at baseline, and repeatedwhileon treatment only if clinically indicated. All adverse events

and laboratory variables were assessed at baseline andduring treatment. Antitumor activity was evaluated every2 cycles (6 weeks).

Pharmacokinetic analysesBlood samples were collected on day 1 of cycle 1 at

predose, 5 minutes before infusion end, and between0.25 and 168 hours postinfusion. In addition, blood sam-pleswere also collected following the same sampling onday1 of cycle 2 during cohort expansion at the recommendeddose. Samples were centrifuged at 2500 � g for 15 minutesat 4�C and plasma was separated and stored at�20�C untilanalysis. The urine produced during the first day of treat-ment was collected and a 20-mL aliquot was stored at�20�C until analysis. Concentrations of PM01183 weredetermined using a liquid chromatography system coupled

Table 1. Baseline characteristics of patients (n ¼ 31)

RD (n ¼ 15) Total (n ¼ 31)

GenderMale 8 (53%) 21 (68%)Female 7 (47%) 10 (32%)

Median age (range; y) 59 (21–75) 61 (21–77)ECOG performance status0 7 (47%) 14 (45%)1 8 (53%) 17 (55%)

Median albumin (range; g/dL) 4.0 (3.4–4.7) 4.0 (3.3–4.7)Median hemoglobin (range; g/dL) 12.6 (9.7–14.5) 12.8 (9.1–15.7)No. of patients with LDH > ULN 6 (43%) 15 (50%)Median LDH (range; x ULN)a 1.2 (1.0–3.8) 1.3 (1.0–3.8)

Primary tumorColorectal cancer 7 (47%) 17 (55%)Pancreatic cancer 3 (20%) 4 (13%)Biliary tract cancer .— 2 (7%)Soft tissue sarcoma 1 (7%) 2 (7%)Otherb 4 (27%) 6 (19%)

Median number of metastatic sites (range) 3 (1–5) 3 (1–5)Sites of diseaseVisceral 1 (7%) 8 (26%)Nonvisceral 2 (13%) 3 (10%)Both 12 (80%) 20 (65%)

Prior treatmentChemotherapy 15 (100%) 31 (100%)Biological therapy 10 (67%) 22 (71%)Radiotherapy 8 (53%) 16 (52%)Investigational therapy 4 (27%) 9 (29%)

Number of prior chemotherapy-containing regimensMedian 3 3�2 5 (33%) 12 (39%)4 or more 5 (33%) 10 (32%)

Abbreviations: LDH, lactate dehydrogenase; RD, recommended dose.aLDH values from patients with >ULN at baseline.bAdenocarcinoma of unknownprimary site, head and neck squamous cell carcinoma,melanoma,mesothelioma, prostate cancer, andsalivary gland adenocarcinoma (n ¼ 1 each).

First-In-Human Phase I Study of Lurbinectedin (PM01183)

www.aacrjournals.org Clin Cancer Res; 20(8) April 15, 2014 2207




with electrospray ionization liquid chromatography/tan-dem mass spectrometry method. The limit of quantitationwas 0.05 ng/mL.

Statistical analysisPatient accrual was presented descriptively. Demo-

graphics and baseline characteristics were summarizedfor all patients and analyzed by dose. Continuous vari-ables were presented with summary statistics (mean,median, and range) and categorical variables in frequencytables. Treatment exposure, delays, and reductions wereevaluated using counts, percentages, and detailed listings.The total number of included patients, the number ofpatients evaluable for DLTs, and the number of patientswith any DLT were summarized by dose. Descriptivestatistics were used for the evaluation of safety. Theincidence and grade of adverse events and laboratoryabnormalities were classified according to the NationalCancer Institute Common Terminology Criteria forAdverse Events (NCI-CTCAE) v.3.0 (12) and the eventswere coded using the Medical Dictionary for RegulatoryActivities (MedDRA) v.11. Antitumor activity was evalu-ated as per the Response Evaluation Criteria in SolidTumors (RECIST) v.1.0 (13). Response rates were char-acterized using descriptive statistics. Time-related para-meters were analyzed according to the Kaplan–Meiermethod. SAS v9.2 (SAS Institute Inc) was used to generatethese outputs. Noncompartmental pharmacokinetic

parameters were calculated using WinNonLin v.5.2 (Phar-sight). All individual pharmacokinetic parameters weretabulated and summarized. Categorical and continuouscovariates were explored against pharmacokinetic para-meters using the PROCMIXED procedure under SAS v.8.2(SAS Institute Inc). Ninety-five percent confidence inter-vals for the slope and p test with a level of significance of0.05 were calculated to detect differences and correla-tions. Pharmacokinetic graphics were performed using Rv.2.15.0 (R Development Core Team) and the plottingpackage ggplot2 v.0.9.3 (14). The interindividual coeffi-cient of variation (CV) was calculated as the ratio of theSD to the mean for the entire population. Similarly, theintraindividual coefficient of variation was determinedfor each patient.

ResultsPatient characteristics and treatment

Thirty-one patients were included between March 2009and September 2010; patient characteristics are shownin Table 1.

A total of 97 PM01183 cycles were administered at 10dose levels, of which 59 cycles (61%) were administered atthe recommended dose, with a median of two cycles perpatient; the median dose intensity was 2.3 mg flat dose/week and median relative dose intensity was 96.3% of theplanned dose.

Table 2. Drug-related adverse events (�10% of patients) and laboratory abnormalities (hematologic andbiochemical) at the recommended dose for phase II studies

PM011834.0 mg/m2 (or 7.0 mg FD)

Per patient (n ¼ 15) Per cycle (cyc ¼ 51)

NCI-CTCAE grade 1–2 3 4 Total 1–2 3 4 Total

Anorexia 2 (13%) — — 2 (13%) 3 (6%) — — 3 (6%)Diarrhea 3 (20%) — — 3 (20%) 4 (8%) — — 4 (8%)Fatigue 7 (47%) — — 7 (47%) 12 (24%) — — 12 (24%)Nausea 7 (47%) — — 7 (47%) 18 (35%) — — 18 (35%)Vomiting 7 (47%) — — 7 (47%) 10 (20%) — — 10 (20%)

Hematologic laboratory abnormalitiesAnemia 10 (67%) 4 (27%) — 14 (93%) 42 (82%) 4 (8%) — 46 (90%)Neutropenia 3 (20%) 2 (13%) 6 (40%) 11 (73%) 7 (14%) 14 (27%) 7 (14%) 28 (55%)Thrombocytopenia 9 (60%) — 1 (7%) 10 (67%) 25 (49%) 1 (2%) 1 (2%) 27 (53%)

Biochemical laboratory abnormalitiesALT increased 9 (60%) 1 (7%) — 10 (67%) 18 (35%) 1 (2%) — 19 (37%)AP increased 10 (67%) — — 10 (67%) 25 (49%) — — 25 (49%)AST increased 6 (40%) 1 (7%) — 7 (47%) 10 (20%) 1 (2%) — 11 (22%)CPK increased 2 (13%) — — 2 (13%) 6 (12%) — — 6 (12%)Creatinine increased 2 (13%) 1 (7%) — 3 (20%) 3 (6%) 1 (2%) — 4 (8%)Total bilirubin increased 3 (20%) — — 3 (20%) 8 (16%) — — 8 (16%)

NOTE: The number of patients/cycle with each adverse event is specified.Hematologica and biochemical abnormalities are shown regardless of their relationship to treatment.Abbreviations: AP, alkaline phosphatase; FD, flat dose.

Elez et al.





Dose-limiting toxicities and recommended doseNo grade 2 or higher toxicities were observed during

escalation from the starting dose of 0.02 mg/m2 untildose level VII (1.3 mg/m2). Both patients treated at 2.6mg/m2 (dose level VIII) had grade 1/2 nausea and/orvomiting, thus, standard antiemetic prophylaxis (ondan-

setron and dexamethasone, both intravenous, 20–30minutes before PM01183 infusion) was made compul-sory in all subsequent dose levels. DLTs were firstobserved in two of three patients treated at dose level IX(5.0 mg/m2) and consisted of grade 4 transaminase[alanine aminotransferase (ALT) and/or aspartate

4,300

4,000

3,000

2,000

1,000

0

CA

19.9

(IU

/mL

)

D-15

BL Cycle 1 Cycle 2 Cycle 3 Cycle 4 Cycle 5 Cycle 6 Cycle 7 Cycle 8 Cycle 9

Decrease 50% Baseline

C

A B

Upper normal limit

Cycle 10

D15 D15D1 D1D8 D8 D15D1 D8 D15 D15D22D1 D1D8 D8 D15D1 D8 D15D1 D8 D15D1 D8 D15 D22D1 D8 D15 D217D1 D8

Figure 1. Tumor shrinkage observed in one patientwith pancreas adenocarcinomawho achievedPRwith PM01183. Three lesionswere observed in the liver atbaseline (A). After three cycles of treatment with PM01183, two of these lesions disappeared and the third decreased from 29.8 mm to 16.5 mm (i.e., a tumorshrinkage of 45%; B). Changes in serum CA19-9 levels from baseline (BL) to cycle 10 (C).






aminotransferase (AST)] increases (n ¼ 2) and a combi-nation of grade 4 neutropenia, grade 3 fatigue, and grade 2diarrhea that resulted in patient withdrawal from the study(n ¼ 1). According to the protocol definition, dose level IX(5.0 mg/m2) was defined as the MTD and dose escalationwas terminated, consequently PM01183 dose was reducedby 20% to 4.0 mg/m2 and cohort scheme was thenswitched to a classical 3þ3 design. None of the first sixpatients treated in this cohort had DLTs, though nonfebrilegrade 4 neutropenia occurred in three patients. No rela-tionship of plasma clearance and BSA was observed; there-fore, nine additional patients were given a flat dose of 7.0mg, approximately equivalent to 4.0 mg/m2 for mostpatients. One of 15 treated patients had a DLT (grade 4thrombocytopenia), therefore confirming 4.0 mg/m2 or7.0 mg flat dose as the recommended dose for PM01183given as a 1-hour infusion q3wk.

Toxicity profile at the recommended doseOf 31 patients, 15 (48%) were treated at the recom-

mended dose and were evaluable for safety. AllPM01183-related adverse events, at the recommendeddose, were grade 1 or 2, the most common were nausea,fatigue, and vomiting despite mandatory antiemetic pro-phylaxis (Table 2). Hematologic abnormalities comprisedanemia (93% of patients, grade 3: 27%, no grade 4 events),neutropenia (73%, grade 3/4: 53%), and thrombocytope-nia (67%, no grade 3 events, grade 4: 7%; Table 2). Grade 4neutropenia occurred in six patients (40%) at the recom-mended dose during cycle 1; the nadir (median count 0.2�

109/L; range, 0.0–0.5� 109/L) occurred on dayþ11 (range,10–15days), and themediandurationwas 3days (range, 1–4 days). No febrile episodes complicated any neutropeniaduring this study. Grade 4 thrombocytopenia occurred inone patient during cycle 1; the nadir (9.2 � 109/l) wasfound on day þ10 and returned to grade �2 within 3days. There were only one cycle delay and two dosereductions at the recommended dose, all due to hema-tologic toxicity. Common biochemical abnormalities,irrespectively of their relationship, were mild alkalinephosphatase (67%) and ALT/AST increases (67% and47%, respectively; Table 2). At the recommended dose,no biochemical abnormalities reached grade 4 severity,and grade 3 only occurred for ALT/AST increases (7%each) and creatinine increase in one patient (7%) withdisease-related obstructive renal failure.

EfficacyTwenty-eight patients were evaluable for efficacy as per

RECIST v.1.0; one confirmed partial response (PR) withtumor marker (CA19-9) normalization occurred in apatient with refractory pancreatic cancer treated at therecommended dose (Figs. 1 and 2). In addition, nine(29%) patients had disease stabilization as best response,which lasted for at least 4 months in five of them. Three ofthese five patients were treated at the recommended doseand all of them achieved longer progression-free survival(PFS) values with PM01183 when compared with theirlast prior therapy (Fig. 2). Overall, six patients showedevidence of some tumor shrinkage while on PM01183

Patient #4

Patient #3

Patient #2

Patient #1

Last prior PFS (mo) PM01183 PFS (mo)

10 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 10

3.1 4.5+

Sorafenib

2.97.3

Fotemustine

2.3 5.3

Trabectedin

1.7 7.8+

Inv. drug

STS

Pancreatic cancer

STS

Malignant melanoma

Partial response Stable disease Progressive diseaseBest RECIST response:

Figure 2. PFS values achievedbefore and after treatment withPM01183. Values are shown onlyfor the four patients who weretreated at the recommended doseand achieved PR (n¼ 1) or diseasestabilization for�4 months (n¼ 3).

Elez et al.





treatment; five of them were treated at the recommendeddose.

PharmacokineticsPharmacokinetic data were available from 31 patients in

cycle 1 and from seven patients treated at the recommendeddose in cycle 2 (Table 3). PM01183 plasma concentrationsincreased linearly within the dose range explored, andclearance was independent of dose (P ¼ 0.451). Pharma-cokinetic variability was high; at the recommended dose,theCVof clearancewas 57% in the first cycle, and somewhatlower (31%) in the second cycle, with a median intraindi-vidual coefficient of variation of 30% (range, 5%–50%).None of the demographic and biochemical variablesexplored in this study had a significant impact on pharma-cokinetic parameters; particularly, no significant relation-ship was found between BSA and clearance (P ¼ 0.889), orbetween gender and clearance (P ¼ 0.074). Urinary excre-tion of unaltered PM01183was almost negligible regardlessof dose; themean percentage of unaltered drug recovered inurine at the recommended dose was 0.23%. To explore theeffect of PM01183 on hematologic parameters, the survivalfraction (ratio of nadir to baseline) was calculated (15) forthe first cycle and used as the dependent variable for furtheranalysis. There was a significant correlation of area under

the concentration-time curve from time zero to infinity(AUC) with both log WBC survival fraction (r ¼�0.73, P¼ 0.0007) and log neutrophil survival fraction (r ¼�0.57, P ¼ 0.016), which was stronger than the rela-tionship with dose (Fig. 3).

DiscussionThis first-in-human study met its goal of defining the

recommended dose of PM01183 in an efficient manner.The accelerated titration design proved to be adequateand safe to allow a swift dose escalation more than 200times, from the conservative starting dose of 0.02 mg/m2,derived from the MTD in the Sprague-Dawley rat. Similarto what has been described with other related compoundssuch as trabectedin, this starting dose, although safe,turned out to be poorly predictive of the human MTD.Should this trial have followed a more classical 3þ3Fibonacci design, it would have required significantlylonger time to reach potentially therapeutic dosages. Incontrast, almost half of participating patients in our studywere treated at the recommended dose, thus maximizingthe exposure to clinically relevant concentrations of theinvestigational treatment, while no unexpected toxicitiesoccurred during escalation in this first-in-human study.

Table 3. Noncompartmental pharmacokinetic parameters of PM01183 1-hour three weekly schedule

Dose (mg/m2) n Cmax (ng/mL) AUC (h�ng/mL) t1=2 (h) CL (l/h) Vss (l)

Cycle 10.02 1 1.5 (–) 2.7 (–) 3.9 (–) 15.2 (–) 68.9 (–)0.04 2 4.4 (1.3) 7.5 (0.4) 5.1 (4.3) 10.6 (1.0) 64.3 (62.7)0.08 1 4.1 (–) 5.6 (–) 1.8 (–) 26.6 (–) 53.8 (–)0.16 1 7.3 (–) 24.5 (–) 14.6 (–) 11.9 (–) 161.9 (–)0.32 2 23.1 (5.8) 125.9 (76.5) 53.9 (0.6) 6.9 (5.4) 340.3 (246.3)0.64 2 33.4 (9.1) 125.6 (63.2) 98.2 (116.7) 11.8 (6.8) 754.9 (763.3)1.3 2 70.4 (33.6) 286.0 (30.8) 48.2 (10.9) 7.9 (0.1) 363.4 (98.0)2.6 2 147.0 (69.3) 502.1 (318.7) 39.1 (0.0) 12.2 (8.0) 336.2 (166.9)4.0 or 7.0 mg FD 15 182.2 (85.9) 951.8 (850.0) 62.7 (45.5) 12.5 (7.1) 487.1 (201.9)4.0 mg FD 6 233.2 (103.4) 777.4 (386.6) 50.9 (27.5) 11.9 (5.7) 402.5 (157.0)7.0 mg FD 9 148.2 (54.6) 1068 (1064) 70.6 (54.6) 12.8 (8.2) 543.5 (216.7)5.0 2 276.0 (141.4) 1127 (674.9) 37.7 (11.6) 9.3 (4.9) 314.9 (51.7)7.0a 1 215.0 (–) 559.7 (–) 36.4 (–) 19.3 (–) 688.5 (–)

Cycle 25.6 mg FDb 1 103.0 (–) 1057 (–) 123.4 (–) 5.3 (–) 799.7 (–)7.0 mg FD 6 153.8 (60.4) 532.6 (194.4) 51.4 (9.1) 14.5 (4.4) 639.8 (193.5)

Cycle 1 and 2 at 4.0 or 7 mg FDFemale 10 175.3 (50.7) 566.2 (320.3) 42.5 (21.3) 15.4 (6.3) 814.2 (245.6)Male 11 172.9 (100.9) 1074.0 (937.3) 74.9 (45.1) 10.9 (5.9) 955.5 (413.2)

NOTE: Values are expressed asmean (SD). Sampleswere collected in cycle 1 for all treated patients, and in cycle 2 for patients treatedduring cohort expansion at the recommended dose.Abbreviations: AUC, area under the concentration-time curve from time zero to infinity; FD, flat dose; CL, total body clearance; Cmax,maximum plasma concentration; t1=2, terminal half-life; Vss, volume of distribution at steady state.aAccidental overdose; see footnote of Table 2.bDose was reduced after cycle 1 due to hematological toxicity.






This was in line with prior preclinical toxicology studies,which had shown myelosuppression and hepatotoxicityas primary PM01183 toxicities in mammals.

A flat dose of 7.0 mg (approximately equivalent to 4.0mg/m2) q3wk was the highest dose level explored with lessthan one third of patients experiencing aDLT in cycle 1, andconsequently was defined as the recommended dose forfurther phase II studies. Of note, one of the 15 patientstreated at the recommended dose experienced a DLT (grade4 thrombocytopenia). Nonetheless, grade 4 transient non-febrile neutropenia occurring in 40% of these patientsprecluded any attempt to further escalate the recommendeddose.

The regimen was generally well tolerated. Hematologictoxicity was predictable, manageable, and transient, where-as the schedulewas suitable formost patients to receive theirtreatment on an ambulatory basis. All toxicities other thanmyelosuppression at the recommended dose were usuallylow grade and transient, the most common being nausea/vomiting and fatigue. Primary antiemetic prophylaxis with5-HT3 antagonists plus a steroid was mandatory, althoughsome patients might also need extended treatment in accor-dance with the American Society of Clinical Oncologyguidelines (16).

Myelosuppression was the primary toxicity observedat the recommended dose or above it. Neutropenia

Figure 3. AUC (left) and dose (right) versus neutrophil ratio (top) and WBCs ratio (bottom) in cycle 1 at/over the recommended dose. The solid linerepresents the semi-log regression and the shadow area is the confidence region. The wide confidence regions for dose imply a weak relationship. RD,recommended dose.

Elez et al.





correlated with dose, but particularly with AUC. Thenadir occurred during the second week of the cycle in allcases, usually on day þ11 (range, 10–15 days), andgrade 4 lasted a median of 3 days (range, 1–5 days),thus allowing enough time to recovery without causingtreatment delays in the majority of cases. Of note, nofebrile neutropenia occurred, probably reflecting thisrelatively short duration and the relatively good condi-tion of the patients participating in this study. Even ifthe eligibility criteria were standard for phase I, nopatient treated at the recommended dose in this studywas older than 75 years, had albumin level below 3.4g/dL, or had an ECOG performance status score >1. Atthis time, it is speculative wherever PM01183-relatedmyelosuppression is equally well tolerated in less select-ed patients; this remains to be tested prospectively. Inaddition, almost one third of patients treated at therecommended dose in this study had no neutropeniaat all, which probably reflected the high interindividualpharmacokinetic variability. Myelosuppression does notseem to be cumulative, as most patients who experi-enced it did so during the first cycle, and no dosereductions or delays were required because of toxicityamong patients receiving more than three cycles at therecommended dose or higher doses. The pharmacoki-netic profile showed dose proportionality within thedose range tested and no drug accumulation betweencycles was found, thus suggesting that administrationevery 3 weeks is safe and appropriate.There was high interindividual variability in the clear-

ance of PM01183, the source of which is yet unclear.PM01183 undergoes extensive microsomal metaboliza-tion in vitro; although the specific metabolic pathwaysand cytochromes involved remain to be fully elucidated.Theoretically, genetic differences in cytochromes or inmembrane transporters at the protein expression levelmight contribute, at least partially, to explain the phar-macokinetic findings; ongoing pharmacogenomic studieswill help to better characterize this observation. Renalexcretion is a very minor route of elimination. Thus,changes in renal function are very unlikely to have animpact on the pharmacokinetic behavior of PM01183. Ofnote, the severity of the myelosuppression correlated withAUC measured during cycle 1; thus, understanding thecauses of this variability might help to optimize patientexposure and thus reduce toxicity in the future. Finally,none of the explored demographic characteristics in thisstudy, particularly BSA, had any significant impact onpharmacokinetics. Therefore, PM01183 flat dosing wassuccessfully implemented at the recommended dosecohort expansion. Nonetheless, due to the limited num-ber of patients treated in this study, none had a BSAoutside the 1.47 to 2.48 m2 range. At the present, cautionshould be exercised when using the 7.0 mg flat doserecommended dose in populations different than thosestudied herein. In addition, further analysis to check anypotential difference in clearance between males andfemales is warranted (see Table 3).

Antitumor efficacy was confirmed, despite extensive priortreatment. Most patients included in this study had colo-rectal tumors and had received three or more previous linesof chemotherapy for advanced disease and therefore werevery unlikely to respond to any therapy at that point.Nevertheless, evidence of antitumor activity in this studywas found in a refractory pancreatic cancer patient who hada confirmedPRwith tumormarker normalization, aswell asin three patients (twowith soft-tissue synovial sarcomas andone with a malignant melanoma) who had prolongeddisease stabilizations with some radiological evidence oftumor shrinkage. Comparison of the PFS achieved onPM01183 with that on the immediately prior treatmentline suggests that the stable disease might be due toPM01183 treatment (as opposed to an indolent naturalhistory, see Fig. 2).

In conclusion, PM01183 administered as a 1-hour infu-sion q3wk to adult patients with refractory solid tumorswaswell tolerated. It is active at its recommended dose of 7.0mgflat dose, and caused primarily severe but reversible, limit-ed, and predictable neutropenia in this study. This scheduleis now being evaluated in several phase II clinical trials; theresults, when available, will provide valuable additionalinformation on safety, feasibility, and pharmacokineticbehavior in a much wider patient population, as well asconfirm its preliminary antitumor activity in specific diseasesettings.

Disclosure of Potential Conflicts of InterestS.P. Kang is employed byMerck&Co Inc. A. S.-M. Pita is employed by and

has ownership interest (including patents) in PharmaMar. M. Siguero hasownership interest (including patents) in PharmaMar. S. Szyldergemajn isemployed by PharmaMar. M.J. Ratain has received a commercial researchgrant from PharmaMar. No potential conflicts of interest were disclosed bythe other authors.

Authors' ContributionsConception and design: M.E. Elez, J. Tabernero, C. Kahatt, C. Fernandez-Teruel, M. Siguero, S. Szyldergemajn, M.J. RatainDevelopment of methodology: M.E. Elez, J. TaberneroAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): M.E. Elez, J. Tabernero, D. Geary, T. Macarulla,S.P. Kang, C. Fernandez-Teruel, M.J. RatainAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): M.E. Elez, J. Tabernero, A.S.-M. Pita,C. Fernandez-Teruel, M. Siguero, M. Cullell-Young, S. Szyldergemajn,M.J. RatainWriting, review, and/or revision of the manuscript: M.E. Elez, J. Taber-nero, S.P. Kang, C. Kahatt, A.S.-M. Pita, C. Fernandez-Teruel, M. Siguero,M. Cullell-Young, S. Szyldergemajn, M.J. RatainAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): D. Geary, C. Fernandez-Teruel,M. Siguero, M.J. RatainStudy supervision: J. Tabernero, D. Geary, S. Szyldergemajn, M.J. Ratain

Grant SupportThis work was supported by Pharma Mar, S.A., Colmenar Viejo, Madrid,

Spain.The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received July 9, 2013; revised January 13, 2014; accepted January 24, 2014;published OnlineFirst February 21, 2014.






References1. Rinehart KL, Holt TG, Fregeau NL, Keifer PA, Wilson GR, Perun TJ Jr,

et al. Bioactive compounds from aquatic and terrestrial sources. J NatProd 1990;53:771–92.

2. Manzanares I, Cuevas C, Garcia-Nieto R, Marco E, Gago F. Advancesin the chemistry and pharmacology of ecteinascidins, a promising newclass of anti-cancer agents. Curr Med Chem Anticancer Agents 2001;1:257–76.

3. Leal JF, Martinez-Diez M, Garcia-Hernandez V, Moneo V, Domingo A,Bueren-Calabuig JA, et al. PM01183, a new DNA minor groovecovalent binder with potent in vitro and in vivo anti-tumour activity.Br J Pharmacol 2010;161:1099–110.

4. Guillen MJ, Catalu~na O, Palomares M, Lopez R, Cuevas C, Aviles P. Invivo combination studies of PM01183 with alkylating, antimetabolites,DNA-topoisomerase inhibitors and tubulin binding agents [abstract].In: Proceedings of the AACR 102nd Annual Meeting; 2011 April 2–6;Orlando FL. Philadelphia, PA: AACR; 2011. Abstract nr 3538.

5. Soares DG, Machado MS, Rocca CJ, Poindessous V, Ouaret D,Sarasin A, et al. Trabectedin and its C subunit modified analoguePM01183 attenuate nucleotide excision repair and show activitytoward platinum-resistant cells. Mol Cancer Ther 2011;10:1481–9.

6. Aviles P, Galmarini C, Cuevas C, Guillen MJ, Frapolli R, Uboldi S, et al.Mechanism of action and antitumor activity of PM01183. 100th AnnualMeeting of the American Association for Cancer Research (AACR);2009; Denver, Colorado; abstract 2679.

7. Soares DG, Larsen AK, Escargueil AE. The DNA damage response tomonofunctional anticancer DNA binders. Drug Discov Today 2012;9:59–67.

8. U.S. Food and Drug Administration Guidance for industry. Estimatingthe maximum safe starting dose in initial clinical trials for therapeutics

in adult healthy volunteers. 2005. http://www.fda.gov/Drugs/Guidan-cecomplianceRegulatoryInformation/Guidances/default.htm.

9. The EuropeanAgency for the Evaluation ofMedicinal ProductsHumanMedicines Evaluation Unit. Note for guidance on the pre-clinicalevaluation of anticancer medicinal products. 1998. http://www.iss.it/binary/scf1/cont/CPMP_SWP_997_96.pdf.

10. Guidance on non-clinical safety studies for the conduct of humanclinical trials and marketing authorization for pharmaceuticals. 2008.http://www.fda.gov/RegulatoryInformation/Guidances/ucm129520.htm.

11. Undevia SD, Vogelzang NJ, Mauer AM, Janisch L, Mani S, Ratain MJ.Phase I clinical trial of CEP-2563 dihydrochloride, a receptor tyrosinekinase inhibitor, in patients with refractory solid tumors. Invest NewDrugs 2004;22:449–58.

12. National Cancer Institute (2006) Common Terminology Criteria forAdverse Events v3.0 (CTCAE). http://www.eortc.be/services/doc/ctc/ctcaev3.pdf.

13. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS,Rubinstein L, et al. New guidelines to evaluate the response totreatment in solid tumors. European Organization for Research andTreatment of Cancer, National Cancer Institute of the United States,National Cancer Institute of Canada. J Natl Cancer Inst 2000;92:205–16.

14. Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York:Springer; 2009.

15. Ratain MJ, Schilsky RL, Conley BA, Egorin MJ. Pharmacodynamics incancer therapy. J Clin Oncol 1990;8:1739–53.

16. Kris MG, Hesketh PJ, Somerfield MR, Feyer P, Clark-Snow R, KoellerJM, et al. American Society of Clinical Oncology guideline for antie-metics in oncology: update 2006. J Clin Oncol 2006;24:2932–47.


Elez et al.




2014;20:2205-2214. Published OnlineFirst February 21, 2014.Clin Cancer Res María Elena Elez, Josep Tabernero, David Geary, et al. Patients with Advanced Solid TumorsFirst-In-Human Phase I Study of Lurbinectedin (PM01183) in

Updated version

10.1158/1078-0432.CCR-13-1880doi:

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/20/8/2205.full#ref-list-1

This article cites 9 articles, 3 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/20/8/2205.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/20/8/2205To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-13-1880

http://clincancerres.aacrjournals.org/content/20/8/2205.full#ref-list-1

http://clincancerres.aacrjournals.org/content/20/8/2205.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/20/8/2205


first-in-human phase i study of lurbinectedin (pm01183) in ... · cancer therapy: clinical...

Documents