randomized phase ii study of dulanermin in combination with paclitaxel, carboplatin, and bevacizumab in advanced non–small-cell lung cancer Randomized Phase II Study of Dulanermin in Combination With Paclitaxel, Carboplatin, and Bevacizumab in Advanced Non–Small-Cell Lung Cancer

Randomized Phase II Study of Dulanermin in Combination With Paclitaxel, Carboplatin, and Bevacizumab in Advanced Non–Small-Cell Lung Cancer

  1. Fiona Blackhall
  1. Jean-Charles Soria, South Paris University, Institut Gustave Roussy, Villejuif; Jean-Louis Pujol, Centre Hospitalier Universitaire de Montpellier, Montpellier, France; Zsuzsanna Márk, Tüdőgyógyintézet Törökbálint, Törökbálint; Barna Szima, Vas Megyei Markusovszky Lajos Általános, Rehabilitációs és Gyógyfürdő Kórház, Egyetemi Oktatókórház, Szombathely; István Albert, Mátrai Gyógyintézet, Mátraháza; Erzsébet Juhász, Országos Korányi Tbc és Pulmonológiai Intézet, Budapest, Hungary; Petr Zatloukal, Charles University, Faculty Hospital Bulovka and Postgraduate Medical Institute, Prague, Czech Republic; Jerzy Kozielski, Katedra i Klinika Chorób Płuc i GruŸlicy Slaskiego Uniwersytetu Medycznego, Zabrze, Poland; Nigel Baker and Dominic Smethurst, Amgen, Uxbridge; Fiona Blackhall, Manchester Research Centre, The Christie National Health Service Foundation Trust, Manchester, United Kingdom; Yong-jiang Hei and Yang Pan, Amgen, Thousand Oaks; and Avi Ashkenazi, Howard Stern, and Lukas Amler, Genentech, South San Francisco, CA.
  1. Corresponding author: Jean-Charles Soria, MD, SITEP, Département de Médecine, Institut Gustave Roussy, 39 rue Camille Desmoulins, 94805 Villejuif, France; e-mail: soria{at}igr.fr.

  1. Presented in part at the 46th Annual Meeting of the American Society of Clinical Oncology, June 4-8, 2010, Chicago, IL.

 
Next Section

Abstract

Purpose To evaluate the efficacy and safety of dulanermin combined with paclitaxel and carboplatin (PC) and bevacizumab (PCB) as first-line treatment for advanced or recurrent non–small-cell lung cancer (NSCLC).

Patients and Methods Patients with squamous NSCLC and/or CNS metastases received PC every 3 weeks alone (arm 1) or with dulanermin 8 mg/kg for 5 days (arm 2). Patients with nonsquamous NSCLC received PCB alone (arm 3) or with dulanermin 8 mg/kg for 5 days (arm 4) or 20 mg/kg for 2 days (arm 5). The primary end point was the objective response rate (ORR).

Results Overall, 213 patients were randomly assigned (arm 1, n = 41; arm 2, n = 39; arm 3, n = 42; arm 4, n = 40; arm 5, n = 41). The ORR in arms 1 to 5 was 39% (95% CI, 24% to 56%), 38% (95% CI, 24% to 54%), 50% (95% CI, 35% to 65%), 40% (95% CI, 25% to 56%), and 40% (95% CI, 25% to 56%), respectively. The odds ratio for ORR was 1.04 (P = 1.000) for arm 1 versus arm 2, 1.53 (P = .391) for arm 3 and versus arm 4, and 1.53 (P = .391) for arm 3 versus arm 5. The most common grade ≥ 3 adverse events were neutropenia, asthenia, anemia, thrombocytopenia, and hemoptysis. Of 161 available serum samples, a trend toward increased caspase-cleaved cytokeratin-18 was observed after dulanermin treatment in cycles 1 and 2. Among 84 patients evaluated for GalNT14 expression, there was a trend toward favorable progression-free survival and overall survival with dulanermin treatment in those with high GalNT14 expression.

Conclusion The addition of dulanermin to PC and PCB did not improve outcomes in unselected patients with previously untreated advanced or recurrent NSCLC.

Previous SectionNext Section

INTRODUCTION

The addition of bevacizumab to first-line carboplatin and paclitaxel (PCB) improves survival and tumor response in patients with nonsquamous non–small-cell lug cancer (NSCLC).1 However, the 2-year survival rate with PCB is 23%, and bevacizumab is not licensed for patients with squamous cell histology.2 Development of novel therapies in NSCLC is warranted.

Apoptosis is a potential therapeutic target in cancer.3–5 Apoptosis is regulated through the intrinsic pathway by cell stress and the extrinsic pathway by binding of ligands (eg, apoptosis ligand 2 [Apo2L]/tumor necrosis factor–related apoptosis-inducing ligand [TRAIL]) to proapoptotic death receptors (DRs).6,7 A high proportion of NSCLC tumors express DR4 and DR5.8 Tumor cell apoptosis is increased during bevacizumab treatment,9,10 whereas expression of DR4 and DR5 is unaffected.11 A combination therapy consisting of bevacizumab and a proapoptotic receptor antagonist (PARA) may be appropriate.

Dulanermin (recombinant human Apo2L/TRAIL) is a PARA of DR4 and DR5 with antitumor activity in preclinical models.12–16 Expression of the peptidyl O-glycosyltransferase GalNT14 has been suggested as a potential biomarker of dulanermin activity in NSCLC.17,18 In the first-in-human study, dulanermin monotherapy had acceptable toxicity and was associated with partial response and stable disease in 3% and 53% of patients, respectively, with advanced solid tumors.19 In a phase Ib study, dulanermin combined with PCB was well tolerated and was associated with an overall response rate of 58% and progression-free survival (PFS) time of 7.2 months in patients with advanced nonsquamous NSCLC.20 The objectives of this study were to evaluate the efficacy and safety of dulanermin combined with standard-of-care therapy in previously untreated advanced squamous or nonsquamous NSCLC. Patients with squamous histology and/or CNS metastases (bevacizumab-ineligible2) received PC with or without dulanermin, whereas patients with nonsquamous histology received PCB with or without dulanermin. Tumor GalNT14 expression was evaluated by immunohistochemistry as a predictive biomarker of dulanermin activity. Potential pharmacodynamic biomarkers were also assessed.

Previous SectionNext Section

PATIENTS AND METHODS

Study Design

This phase II portion of a phase Ib/II, multicenter, open-label, randomized study was conducted at 36 European centers. The phase Ib results were reported previously.20 The primary end point was objective response rate (ORR; complete or partial response per modified Response Evaluation Criteria in Solid Tumors version 1.0 [RECIST]21 by independent central assessment). Secondary end points included overall survival (OS), PFS, time to response, duration of response, time to progression (TTP), and pharmacokinetics (reported for the phase Ib study20). Safety end points included adverse events (AEs), laboratory abnormalities, and antidulanermin antibody formation. Exploratory end points included biomarkers of tumor response.

Patients were to receive paclitaxel (200 mg/m2) and carboplatin (area under the curve, 6 mg/mL · min) intravenously on day 1 of each cycle. Up to six 3-week cycles were administered. Patients with squamous NSCLC and/or treated and stable CNS metastases (ineligible for bevacizumab) received PC (arm 1) or PC plus intravenous dulanermin (8 mg/kg once daily for 5 days beginning on day 1 of each cycle; arm 2). Patients with nonsquamous NSCLC and no CNS metastases (eligible for bevacizumab treatment) received either PC plus 15 mg/kg intravenous bevacizumab once every 3 weeks beginning on day 1 (arm 3), PCB plus dulanermin (8 mg/kg once daily for 5 days beginning on day 1; arm 4), or PCB plus dulanermin (20 mg/kg once daily for 2 days beginning on day 1; arm 5). Patients in arm 1 who completed six cycles of PC received no further treatment. Patients in arms 2 through 5 who completed six cycles continued to receive bevacizumab and/or dulanermin until disease progression, intolerable AE, withdrawal of consent, or 24 months from the date the last patient was randomly assitgned. Dose reductions of dulanermin and bevacizumab were not permitted, but treatment could be withheld for toxicity; treatment was discontinued if withheld more than 3 weeks. Chemotherapy cycles could be delayed for toxicity ≤ 3 weeks. Granulocyte colony–stimulating factor was administered for febrile neutropenia or related infections.

Tumor Assessments

Computed tomography or magnetic resonance imaging was performed at screening, every 6 ± 1 week during treatment, at the end-of-study visit, and at long-term follow-up visits. Tumor response was assessed per modified RECIST21 by independent central assessment (Perceptive Informatics, Waltham, MA). Responses were to be confirmed ≥ 4 weeks after criteria were first met. Assessments of stable disease occurred ≥ 49 days after enrollment.

Safety

AEs occurring from enrollment until the end of study or 30 days after the last dose of protocol-specified therapy were graded using National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. Blood samples for the assessment of antidulanermin antibodies were collected preinfusion on day 1 of every cycle.

Biomarker Measurement

Serum samples were collected on days 1, 2, and 8 of cycle 1 and on days 1 and 2 of cycle 2. Caspase-cleaved cytokeratin-18 and total soluble cytokeratin-18 were assessed using M30-Apoptosense and M65 enzyme-linked immunosorbent assays (Peviva AB, Stockholm, Sweden), respectively (Appendix, online only). GalNT14 expression in archival tumor samples was assessed by immunohistochemistry (Appendix, online only).

Statistical Analysis

The planned sample size of 200 patients (40 per arm) allowed estimation of the ORR within treatment arms with an SE of ≤ 8% (half width of the 95% CIs ≤ 16%) and between treatment arms with an SE of ≤ 11%. ORRs (with 95% CIs) were calculated for each treatment arm, and a Fisher's exact test was used to test the treatment differences. Unadjusted odds ratios of responder/nonresponder for dulanermin arms versus their respective comparator arms were calculated. Differences in Kaplan-Meier estimates of PFS time (from randomization to first disease progression or death), OS, time to response, TTP, and duration of response were assessed using a log-rank test. A Cox proportional hazards model estimated the hazard ratio (HR) between each dulanermin arm and its comparator. Summary statistics are reported for safety data. Patients who received one or more doses of the study drug were included in the safety analysis set. The intent-to-treat analysis set included all randomly assigned patients.

Treatment means postbaseline biomarker sampling time points were compared with cycle baselines to test a time effect by treatment arm using Wilcoxon signed-rank tests. Log-fold ratios of serum biomarker concentrations from baseline to each postbaseline sampling time were compared between treatment arms. Spearman rank correlation coefficient was used to assess correlations between log-fold ratios and tumor measurements. Cox proportional hazards models were used to assess associations between tumor GalNT14 expression and OS and PFS (Appendix, online only).

Patient randomization was performed using an interactive voice response system (ICON Clinical Research, North Wales, PA) at Amgen (Thousand Oaks, CA). The analyses were performed by the biostatistics departments at Amgen (San Francisco, CA) and Amgen (Cambridge, United Kingdom).

Previous SectionNext Section

RESULTS

Patients

Between November 20, 2007, and October 22, 2008, 213 patients were randomly assigned (arm 1 [PC], n = 41; arm 2 [PC plus dulanermin 8 mg/kg for 5 days], n = 42; arm 3 [PCB], n = 44; arm 4 [PCB plus dulanermin 8 mg/kg], n = 43; arm 5 [PCB plus dulanermin 20 mg/kg for 2 days], n = 43; Table 1). Overall, 120 patients received one or more doses of dulanermin, 71 (33%) of whom received the protocol-specified six cycles of treatment. Ninety-three patients received maintenance treatment, nine of whom continued to receive treatment as of December 17, 2009 (Fig 1). The reasons for study discontinuation were disease progression (n = 95 [45%]), AE (n = 34 [16%]), protocol-specified criteria (n = 19 [9%]), death (n = 17 [8%]), administrative decision (n = 11 [5%]), consent withdrawal (n = 10 [5%]), other (n = 7 [3%]), ineligibility (n = 6 [3%]), and protocol deviation (n = 2 [1%]). The median weight-adjusted cumulative doses of dulanermin in arms 2, 4, and 5 were 242 mg/kg (range, 40 to 916 mg/kg), 307 mg/kg (range, 40 to 932 mg/kg), and 301 mg/kg (range, 40 to 1,055 mg/kg), respectively. The median weight-adjusted cumulative doses of bevacizumab in arms 3, 4, and 5 were 126 mg/kg (range, 15 to 427 mg/kg), 113 mg/kg (range, 15 to 355 mg/kg), and 114 mg/kg (15 to 389 mg/kg), respectively.

View this table:
Table 1.

Baseline Demographics and Disease Characteristics

Fig 1.
View larger version:
Fig 1.

CONSORT diagram. D, dulanermin; ITT, intent to treat; NSCLC, non–small-cell lung cancer; PC, carboplatin plus paclitaxel; PCB, carboplatin plus paclitaxel and bevacizumab.

ORR

The ORR was 39% in arm 1, 38% in arm 2, 50% in arm 3, 40% in arm 4, and 40% in arm 5 (Table 2). The median time to response was 4.0 months for arm 1, 4.3 months for arm 2, 2.8 months for arm 3, 4.1 months for arm 4, and 4.5 months for arm 5. The median duration of response was 4.6 months for arm 1, 4.2 months for arm 2, 7.0 months for arm 3, 10.5 months for arm 4, and 8.9 months for arm 5. The odds ratios for these end points were not significantly affected by the addition of dulanermin to PC and PCB in unadjusted (Table 2) and adjusted (age, sex, and smoking status) Cox proportional hazards models (data not shown).

View this table:
Table 2.

Tumor Response

PFS and OS

At the time of this analysis, 71% of patients had disease progression or had died (Table 2). Median PFS was 6.1 months in arm 1, 5.5 months in arm 2, 7.3 months in arm 3, 8.6 months in arm 4, and 9.5 months in arm 5 (Fig 2). Median OS was 10.1 months for arm 1, 9.8 months for arm 2, 15.1 months for arm 3, 13.9 months for arm 4, and 14.3 months for arm 5. Median TTP was 6.6 months for arm 1, 5.9 months for arm 2, 7.9 months for arm 3, 11.9 months for arm 4, and 9.6 months for arm 5. These end points were not significantly affected by the addition of dulanermin to either PC or PCB in unadjusted (Table 2) and adjusted (age, sex, and smoking status) Cox proportional hazards models (data not shown).

Pharmacodynamic Biomarker Analysis

Serum samples were available from 194 patients in the phase II study. Numeric increases in serum caspase-cleaved cytokeratin-18 were observed 1 day after treatment in cycles 1 and 2 for patients who received dulanermin versus control (arm 1 v 2; arm 3 v arms 4 and 5; Fig 3). Similar increases were observed for total soluble cytokeratin-18 (data not shown). There were no correlations between log-fold changes from baseline in serum cytokeratin-18 and changes in tumor measurements (data not shown).

Fig 3.
View larger version:
Fig 3.

Box plots of median (horizontal lines) and mean (plus symbols) log-fold change from baseline in circulating caspase-cleaved cytokeratin-18 (using M30 ELISA) on day 2 of cycles 1 (A) and 2 (B). Arm 1, PC; Arm 2, PC + D 8 mg/kg for 5 days; Arm 3, PCB; Arm 4, PCB + D 8 mg/kg for 5 days; Arm 5, PCB + D 20 mg/kg for 2 days. D, dulanermin; PC, paclitaxel and carboplatin; PCB, paclitaxel, carboplatin, and bevacizumab.

GalNT14 Expression Biomarker Analysis

Archived tumor samples for determination of GalNT14 expression by immunohistochemistry were available from 84 patients, which included seven from the phase Ib study to increase the sample size for this hypothesis-generating analysis. Twenty-eight (33%) of these had elevated GalNT14 expression. There was a statistically insignificant trend toward favorable OS (HR, 0.77; 95% CI, 0.25 to 2.35) and PFS (HR, 0.58; 95% CI, 0.21 to 1.60) in GalNT14-positive patients who received dulanermin versus those in control arms (adjusted for age and sex; Table 4). In the control arms, PFS was shorter in GalNT14-positive patients compared with GalNT14-negative patients.

View this table:
Table 4.

Association Between Tumor GalNT14 Expression Status and Overall Survival and Progression-Free Survival

Previous SectionNext Section

DISCUSSION

In this phase II study, the addition of dulanermin to PC or PCB did not improve ORR (the primary end point), PFS, OS, duration of response, time to response, or TTP. These results are disappointing given the results of the phase 1b study in which dulanermin combined with PCB had acceptable toxicity and encouraging efficacy (ORR, 58%; PFS, 7.2 months) compared with historical controls.20 This study joins a number of phase II studies failing to demonstrate efficacy with targeted agents in NSCLC after encouraging results in phase I studies.22–27 The addition of the DR5 agonist monoclonal antibodies drozitumab and mapatumumab to PCB or PC in patients with previously untreated advanced NSCLC did not improve ORR, OS, and PFS.23,24 The differences in interpretation between the phase Ib and II results may have been influenced by the lack of control in the phase Ib study, small sample size, or blunting effects of the respective central versus investigator assessments. This estimation study was not designed to unequivocally assess differences in efficacy between the treatment arms, but rather to guide hypothesis generation for phase III evaluation.

There is increasing support that NSCLC histology may affect outcomes and safety among patients treated with cytotoxic or targeted therapies.29 The ORR among patients with nonsquamous histology who received dulanermin (arms 4 and 5, 40%) was relatively consistent with that of the patients with nonsquamous histology in the PCB arm of the E4599 study1 (35%) and in the cisplatin/gemcitabine plus bevacizumab arm of the AVAiL study30 (34% and 30%, respectively). The ORR in the PCB arm (arm 3, 50%) was greater than in E4599 (35%). The ORR in the PC arm was also higher (39% v 17%) than in another randomized study of advanced NSCLC (although the histology criteria were different).31 The lower ORR in arms 4 and 5 (40%) versus arm 3 (50%) may have resulted from differences in bevacizumab exposure. PFS and OS in the PCB arm of E4599 and OS in the combination arms of AVAiL were shorter than in arms 4 and 5. Although OS in the PCB arm was longer than in E4599, it is consistent with the OS reported in a recent phase II study in which patients received bevacizumab 15 mg/kg every 3 weeks (14.0 months) or the small molecule angiogenesis inhibitor motesanib 125 mg once daily (14.0 months).32

Consistent with the phase Ib study,20 the addition of dulanermin to PCB was tolerable, and there was no marked difference in toxicity between the dulanermin arms and the control arms. The incidence of treatment-related deaths was consistent with that reported in the PCB arm of the E4599 study.1 The increased incidence of grade ≥ 3 and serious AEs in arm 2 versus arm 1 suggests that the AE profile with dulanermin varies with tumor histology. Certain frequent grade ≥ 3 AEs (neutropenia, asthenia, anemia, and thrombocytopenia) were reported in the phase 1b study20 and in studies of PARAs alone or combined with PC or PCB.6,23,27 Grade ≥ 3 hemoptysis, which was not reported in the phase Ib study,20 was associated with four deaths (two were dulanermin-related), all of which occurred in bevacizumab-ineligible patients (arm 2). The rate of grade ≥ 3 hemoptysis among patients who received bevacizumab was consistent with that reported in the E4599 study.1 Given the small sample size, the relationship of hemoptysis to dulanermin is unclear. hemoptysis, hypertension, and intestinal perforation have been reported in patients with advanced NSCLC treated with bevacizumab as monotherapy or combined with chemotherapy.1,2,30

In combination with PC, dulanermin enhances NSCLC tumor cell apoptosis in vitro33,34 and inhibits orthotopic NSCLC xenograft growth.14 Transient increases in activated caspase 3/7 and caspase-cleaved cytokeratin-18, an indicator of apoptosis,35,36 were observed after treatment in the dulanermin first-in-human study (Pan et al, manuscript submitted for publication). Circulating total cytokeratin-18 and caspase-cleaved cytokeratin-18 have shown potential utility as prognostic or predictive biomarkers for a variety of solid tumors.37–40 Although change in circulating caspase-cleaved cytokeratin-18 was not associated with efficacy in this study, the numeric increases in circulating levels of this apoptosis-related biomarker in dulanermin-treated patients suggests an increase in epithelial cell apoptosis and a biologic effect by dulanermin combined with PC or PCB. However, these results are limited by the small sample size.

The presence of DR4 and/or DR5 is necessary, but not necessarily sufficient, for induction of apoptosis by dulanermin.17 Although we detected a range of cellular DR5 expression and distribution by immunohistochemistry (data not shown), DR4 expression was not assessed because an adequately sensitive and selective monoclonal antibody was unavailable.

High GalNT14 mRNA and protein expression in tumor cell lines are associated with Apo2L/TRAIL sensitivity.17,18 Although there was no significant interaction between tumor GalNT14 expression and patient outcome among dulanermin-treated patients, a trend toward increased PFS and OS was observed in GalNT14-positive patients in dulanermin arms. The decreased PFS among GalNT14-positive versus GalNT14-negative patients in the control arms suggests that GalNT14 expression may be associated with poor prognosis in advanced NSCLC. Consistent with recent studies in NSCLC,41,42 these results are limited by a low (33%) tumor sample ascertainment rate and the small sample size. Additional studies to evaluate GalNT14 expression as a biomarker predictive of response to PARA/extrinsic pathway–targeted therapy in NSCLC are needed.

In conclusion, these results indicate that the addition of dulanermin to PC and PCB does not improve efficacy in unselected patients with previously untreated advanced squamous or nonsquamous NSCLC.

Previous SectionNext Section

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: Nigel Baker, Amgen (C); Dominic Smethurst, Amgen (C); Yong-jiang Hei, Amgen (C); Avi Ashkenazi, Genentech (C); Howard Stern, Genentech (C); Lukas Amler, Genentech (C); Yang Pan, Amgen (C) Consultant or Advisory Role: Fiona Blackhall, Amgen (C) Stock Ownership: Nigel Baker, Amgen; Dominic Smethurst, Amgen; Yong-jiang Hei, Amgen; Avi Ashkenazi, Roche; Howard Stern, Roche; Lukas Amler, Genentech; Yang Pan, Amgen Honoraria: None Research Funding: None Expert Testimony: None Other Remuneration: None

Previous SectionNext Section

AUTHOR CONTRIBUTIONS

Conception and design: Avi Ashkenazi, Lukas Amler, Yang Pan

Provision of study materials or patients: Jean-Charles Soria, Fiona Blackhall

Collection and assembly of data: Jean-Charles Soria, Zsuzsanna Márk, Barna Szima, István Albert, Erzsébet Juhász, Jean-Louis Pujol, Jerzy Kozielski, Nigel Baker, Dominic Smethurst, Howard Stern, Lukas Amler, Yang Pan

Data analysis and interpretation: John-Charles Soria, Petr Zatloukal, Jean-Louis Pujol, Nigel Baker, Dominic Smethurst, Yong-jiang Hei, Avi Ashkenazi, Howard Stern, Lukas Amler, Yang Pan, Fiona Blackhall

Manuscript writing: All authors

Final approval of manuscript: All authors

Previous SectionNext Section

Acknowledgment

We thank Jeff Wiezorek and William Novotny (trial design); Ren Xu, Matt Peach, and Liming Sui (biomarker analysis); Lovely Goyal and Xinqun Yang (statistical analysis); and Benjamin Scott, whose work was funded by Amgen, for assistance in the preparation of this article.

Previous SectionNext Section

Appendix

Previous SectionNext Section

Materials and Methods

Measurement of serum cytokeratin-18.

Whole blood (8.5 mL) was collected in Vacutainer (Becton Dickinson, Franklin Lake, NJ) tubes on days 1 (pretreatment), 2 (pretreatment), and 8 of cycle 1; pretreatment on days 1 and 2 of cycle 2; and 30 ± 3 days after the last dose was administered. Serum was separated from blood cells by centrifugation at 1,500 × g at 4°C for 15 minutes and stored at −70°C until the time of assay. Serum caspase-cleaved cytokeratin-18 and serum soluble cytokeratin-18 were measured using the M30-Apoptosense and M65 enzyme-linked immunosorbent assays, respectively (Peviva AB, Stockholm, Sweden), as described previously (Olofsson MH, et al: Clin Cancer Res 13:3198-3206, 2007; Hägg M, et al: Invest New Drugs 20:253-259, 2002).

To test a time effect for each treatment arm, mean postbaseline serum cytokeratin-18 concentrations at each time point were compared with the baseline concentration in the respective cycle. Log-fold ratios of baseline serum cytokeratin-18 concentrations were compared at each postbaseline sampling time. Spearman's rank correlation coefficient was used to investigate the correlation between log-fold ratios of cytokeratin-18 concentrations from baseline and tumor response measurements (absolute and percentage change from baseline at week 6 and at the time of best response per central assessment). The sum of the log ratios of cytokeratin-18 concentrations at cycle 1 (day 2) and at cycle 2 (day 2) was also used for analysis. The Spearman rank correlation coefficient was also used to assess the correlation of tumor measurements and cytokeratin-18 concentrations between baseline and postbaseline sampling time points. No multiplicity adjustments were used to conclude statistical significance. A criterion of P < .01 was used to screen results for biomarker hypothesis generation.

Measurement of archival tumor GalNT14 expression.

Formalin-fixed paraffin embedded sections (5-μm thick) from archival tumor samples were collected for analysis of GalNT14 expression. The development of the GalNT14 immunohistochemistry assay has been described previously (Stern HM, et al: Clin Cancer Res 16:1587-1596, 2010).

Stained slides were scored by a pathologist who estimated the percentage of tumor cells with a Golgi staining pattern in each of four categories of staining intensity: 0, no stain; 1+, weak stain; 2+, moderate stain; 3+, maximal stain). If the sum of the percentages of tumor cells with ≥ 1+ staining was ≥ 10%, then the patient was categorized as GalNT14 positive; otherwise, patients were categorized as GalNT14 negative. For both GalNT14-positive and GalNT14-negative patients, Cox proportional hazards models were used to estimate the hazard ratios (adjusted for age and sex) for overall survival and progression-free survival (per central assessment) between patients who received dulanermin (arms 2, 4, and 5) versus those who received control treatments (arms 1 and 3) in the GalNT14-positive and GalNT14-negative patients.

Previous SectionNext Section

Footnotes

  • Supported by Amgen and Genentech.

  • Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

  • Clinical trial information can be found for the following: NCT00508625.

  • Received May 24, 2011.
  • Accepted August 24, 2011.
Previous Section
 

REFERENCES

  1. ↵
    1. Sandler A,
    2. Gray R,
    3. Perry MC,
    4. et al.
    (2006) Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med 355:2542–2550.
    CrossRefMedline
  2. ↵
    1. Johnson DH,
    2. Fehrenbacher L,
    3. Novotny WF,
    4. et al.
    (2004) Randomized phase II trial comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 22:2184–2191.
    Abstract/FREE Full Text
  3. ↵
  5. ↵
    1. Rowinsky EK
    (2005) Targeted induction of apoptosis in cancer management: The emerging role of tumor necrosis factor-related apoptosis-inducing ligand receptor activating agents. J Clin Oncol 23:9394–9407.
    Abstract/FREE Full Text
  6. ↵
    1. Ashkenazi A,
    2. Herbst RS
    (2008) To kill a tumor cell: The potential of proapoptotic receptor agonists. J Clin Invest 118:1979–1990.
    CrossRefMedline
  7. ↵
    1. Ashkenazi A,
    2. Holland P,
    3. Eckhardt SG
    (2008) Ligand-based targeting of apoptosis in cancer: The potential of recombinant human apoptosis ligand 2/tumor necrosis factor-related apoptosis-inducing ligand (rhApo2L/TRAIL) J Clin Oncol 26:3621–3630.
    Abstract/FREE Full Text
  8. ↵
    1. Spierings DC,
    2. de Vries EG,
    3. Timens W,
    4. et al.
    (2003) Expression of TRAIL and TRAIL death receptors in stage III non-small cell lung cancer tumors. Clin Cancer Res 9:3397–3405.
    Abstract/FREE Full Text
  9. ↵
    1. Mizobe T,
    2. Ogata Y,
    3. Murakami H,
    4. et al.
    (2008) Efficacy of the combined use of bevacizumab and irinotecan as a postoperative adjuvant chemotherapy in colon carcinoma. Oncol Rep 20:517–523.
    Medline
  10. ↵
    1. Wedam SB,
    2. Low JA,
    3. Yang SX,
    4. et al.
    (2006) Antiangiogenic and antitumor effects of bevacizumab in patients with inflammatory and locally advanced breast cancer. J Clin Oncol 24:769–777.
    Abstract/FREE Full Text
  11. ↵
    1. Yildiz R,
    2. Benekli M,
    3. Buyukberber S,
    4. et al.
    (2010) The effect of bevacizumab on serum soluble FAS/FASL and TRAIL and its receptors (DR4 and DR5) in metastatic colorectal cancer. J Cancer Res Clin Oncol 136:1471–1476.
    CrossRefMedline
  12. ↵
    1. Hylander BL,
    2. Pitoniak R,
    3. Penetrante RB,
    4. et al.
    (2005) The anti-tumor effect of Apo2L/TRAIL on patient pancreatic adenocarcinomas grown as xenografts in SCID mice. J Transl Med 3:22.
    CrossRefMedline
    1. Pollack IF,
    2. Erff M,
    3. Ashkenazi A
    (2001) Direct stimulation of apoptotic signaling by soluble Apo2l/tumor necrosis factor-related apoptosis-inducing ligand leads to selective killing of glioma cells. Clin Cancer Res 7:1362–1369.
    Abstract/FREE Full Text
  14. ↵
  16. ↵
    1. Ashkenazi A,
    2. Pai RC,
    3. Fong S,
    4. et al.
    (1999) Safety and antitumor activity of recombinant soluble Apo2 ligand. J Clin Invest 104:155–162.
    Medline
  17. ↵
  18. ↵
    1. Stern HM,
    2. Padilla M,
    3. Wagner K,
    4. et al.
    (2010) Development of immunohistochemistry assays to assess GALNT14 and FUT3/6 in clinical trials of dulanermin and drozitumab. Clin Cancer Res 16:1587–1596.
    Abstract/FREE Full Text
  19. ↵
    1. Herbst RS,
    2. Eckhardt SG,
    3. Kurzrock R,
    4. et al.
    (2010) Phase I dose-escalation study of recombinant human Apo2L/TRAIL, a dual proapoptotic receptor agonist, in patients with advanced cancer. J Clin Oncol 28:2839–2846.
    Abstract/FREE Full Text
  20. ↵
    1. Soria JC,
    2. Smit E,
    3. Khayat D,
    4. et al.
    (2010) Phase 1b study of dulanermin (recombinant human Apo2L/TRAIL) in combination with paclitaxel, carboplatin, and bevacizumab in patients with advanced non-squamous non-small-cell lung cancer. J Clin Oncol 28:1527–1533.
    Abstract/FREE Full Text
  21. ↵
    1. Therasse P,
    2. Arbuck SG,
    3. Eisenhauer EA,
    4. et al.
    (2000) New guidelines to evaluate the response to treatment in solid tumors: European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92:205–216.
    Abstract/FREE Full Text
  22. ↵
    1. Camidge DR,
    2. Herbst RS,
    3. Gordon MS,
    4. et al.
    (2010) A phase I safety and pharmacokinetic study of the death receptor 5 agonistic antibody PRO95780 in patients with advanced malignancies. Clin Cancer Res 16:1256–1263.
    Abstract/FREE Full Text
  23. ↵
    1. Karapetis CS,
    2. Clingan PR,
    3. Leighl NB,
    4. et al.
    (2010) Phase II study of PRO95780 plus paclitaxel, carboplatin, and bevacizumab (PCB) in non-small cell lung cancer (NSCLC) J Clin Oncol 28:546s, suppl; abstr 7535.
    Search Google Scholar
  24. ↵
    1. Von Pawel J,
    2. Harvey JH,
    3. Spigel DR,
    4. et al.
    (2010) A randomized phase II trial of mapatumumab, a TRAIL-R1 agonist monoclonal antibody, in combination with carboplatin and paclitaxel in patients with advanced NSCLC. J Clin Oncol 28:539s, suppl; abstr LBA7501.
    CrossRef
    1. Goss GD,
    2. Arnold A,
    3. Shepherd FA,
    4. et al.
    (2010) Randomized, double-blind trial of carboplatin and paclitaxel with either daily oral cediranib or placebo in advanced non-small-cell lung cancer: NCIC clinical trials group BR24 study. J Clin Oncol 28:49–55.
    Abstract/FREE Full Text
    1. Goss G,
    2. Shepherd FA,
    3. Laurie S,
    4. et al.
    (2009) A phase I and pharmacokinetic study of daily oral cediranib, an inhibitor of vascular endothelial growth factor tyrosine kinases, in combination with cisplatin and gemcitabine in patients with advanced non-small cell lung cancer: A study of the National Cancer Institute of Canada Clinical Trials Group. Eur J Cancer 45:782–788.
    CrossRefMedline
  27. ↵
    1. Leong S,
    2. Cohen RB,
    3. Gustafson DL,
    4. et al.
    (2009) Mapatumumab, an antibody targeting TRAIL-R1, in combination with paclitaxel and carboplatin in patients with advanced solid malignancies: Results of a phase I and pharmacokinetic study. J Clin Oncol 27:4413–4421.
    Abstract/FREE Full Text
  28. Reference deleted.
  29. ↵
    1. Langer CJ,
    2. Besse B,
    3. Gualberto A,
    4. et al.
    (2010) The evolving role of histology in the management of advanced non-small-cell lung cancer. J Clin Oncol 28:5311–5320.
    Abstract/FREE Full Text
  30. ↵
    1. Reck M,
    2. von Pawel J,
    3. Zatloukal P,
    4. et al.
    (2009) Phase III trial of cisplatin plus gemcitabine with either placebo or bevacizumab as first-line therapy for nonsquamous non-small-cell lung cancer: AVAiL. J Clin Oncol 27:1227–1234.
    Abstract/FREE Full Text
  31. ↵
    1. Schiller JH,
    2. Harrington D,
    3. Belani CP,
    4. et al.
    (2002) Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med 346:92–98.
    CrossRefMedline
  32. ↵
    1. Blumenschein GR Jr,
    2. Kabbinavar F,
    3. Menon H,
    4. et al.
    (2011) A phase II, multicenter, open-label randomized study of motesanib or bevacizumab in combination with paclitaxel and carboplatin for advanced nonsquamous non-small-cell lung cancer. Ann Oncol 22:2057–2067.
    Abstract/FREE Full Text
  33. ↵
  34. ↵
    1. Odoux C,
    2. Albers A,
    3. Amoscato AA,
    4. et al.
    (2002) TRAIL, FasL and a blocking anti-DR5 antibody augment paclitaxel-induced apoptosis in human non-small-cell lung cancer. Int J Cancer 97:458–465.
    CrossRefMedline
  35. ↵
  36. ↵
    1. Leers MP,
    2. Kolgen W,
    3. Bjorklund V,
    4. et al.
    (1999) Immunocytochemical detection and mapping of a cytokeratin 18 neo-epitope exposed during early apoptosis. J Pathol 187:567–572.
    CrossRefMedline
  37. ↵
    1. Dive C,
    2. Smith RA,
    3. Garner E,
    4. et al.
    (2010) Considerations for the use of plasma cytokeratin 18 as a biomarker in pancreatic cancer. Br J Cancer 102:577–582.
    CrossRefMedline
    1. De Petris L,
    2. Branden E,
    3. Herrmann R,
    4. et al.
    (2011) Diagnostic and prognostic role of plasma levels of two forms of cytokeratin 18 in patients with non-small-cell lung cancer. Eur J Cancer 47:131–137.
    CrossRefMedline
  40. ↵
    1. de Haas EC,
    2. di Pietro A,
    3. Simpson KL,
    4. et al.
    (2008) Clinical evaluation of M30 and M65 ELISA cell death assays as circulating biomarkers in a drug-sensitive tumor, testicular cancer. Neoplasia 10:1041–1048.
    Medline
  41. ↵
    1. Khambata-Ford S,
    2. Harbison CT,
    3. Hart LL,
    4. et al.
    (2010) Analysis of potential predictive markers of cetuximab benefit in BMS099, a phase III study of cetuximab and first-line taxane/carboplatin in advanced non-small-cell lung cancer. J Clin Oncol 28:918–927.
    Abstract/FREE Full Text
  42. ↵
    1. Douillard JY,
    2. Shepherd FA,
    3. Hirsh V,
    4. et al.
    (2010) Molecular predictors of outcome with gefitinib and docetaxel in previously treated non-small-cell lung cancer: Data from the randomized phase III INTEREST trial. J Clin Oncol 28:744–752.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. Published online before print October 17, 2011, doi: 10.1200/JCO.2011.37.2623 JCO vol. 29 no. 33 4442-4451
  1. Abstract
  2. » Full Text
  3. PDF
  4. Protocol

Classifications

Navigate This Article

Current Issue

  1. July 20, 2013, 31 (21)
  1. Alert me to new issues of JCO
    • Advertisement
    • Advertisement
    • Advertisement