- © 2011 by American Society of Clinical Oncology
Randomized, Placebo-Controlled, Phase III Study of First-Line Oxaliplatin-Based Chemotherapy Plus PTK787/ZK 222584, an Oral Vascular Endothelial Growth Factor Receptor Inhibitor, in Patients With Metastatic Colorectal Adenocarcinoma
- J. Randolph Hecht⇓,
- Tanja Trarbach,
- John D. Hainsworth,
- Pierre Major,
- Elke Jäger,
- Robert A. Wolff,
- Katherine Lloyd-Salvant,
- György Bodoky,
- Kelly Pendergrass,
- William Berg,
- Bee-Lian Chen,
- Tarja Jalava,
- Gerold Meinhardt,
- Dirk Laurent,
- David Lebwohl and
- David Kerr
- From the David Geffen School of Medicine at University of California, Los Angeles, Santa Monica, CA; Sarah Cannon Cancer Center, Nashville, TN; The University of Texas MD Anderson Cancer Center, Houston, TX; Virginia Cancer Institute, Richmond, VA; Kansas City Cancer Center, Kansas City, MO; Novartis Oncology, East Hanover, NJ; West German Cancer Center, University Hospital Essen, Essen; Medizinische Klinik II – Onkologie, Frankfurt; Bayer Schering Pharma AG, Berlin, Germany; Juravinski Cancer Centre, Hamilton, Ontario, Canada; Szent Laszlo Hospital, Budapest, Hungary; Bayer Schering Pharma Oy, Espoo, Finland; and Sidra Medical and Research Centre, Doha, Qatar.
- Corresponding author: J. Randolph Hecht, MD, David Geffen School of Medicine at University of California, Los Angeles, 2825 Santa Monica Blvd, Ste 221, Santa Monica, CA 90404; e-mail: JRHecht{at}mednet.ucla.edu.
Abstract
Purpose
Purpose PTK787/ZK 222584 (PTK/ZK; vatalanib), an orally active, multitargeted angiogenesis inhibitor, has shown tolerability and promising activity in early-phase studies, which led to a phase III trial in combination with oxaliplatin, fluorouracil, and leucovorin (FOLFOX4).
Patients and Methods
Patients and Methods Patients (N = 1,168) with previously untreated metastatic colorectal cancer were randomly assigned 1:1 to receive FOLFOX4 plus PTK/ZK or placebo (ClinicalTrials.gov identifier: NCT00056459). Stratification factors included WHO performance status (0 v 1 or 2) and serum lactate dehydrogenase (LDH; ≤ v > 1.5× the upper limit of normal). The primary end point was progression-free survival (PFS). Secondary end points included overall survival (OS) and overall response rate (ORR).
Results
Results PFS, OS, and ORR were not statistically improved with PTK/ZK (P > .05). Median PFS by central review was 7.7 months with PTK/ZK versus 7.6 months with placebo (hazard ratio [HR], 0.88; 95% CI, 0.74 to 1.03; P = .118); median OS was 21.4 months with PTK/ZK versus 20.5 months with placebo (HR, 1.08; 95% CI, 0.94 to 1.24; P = .260). In an exploratory post hoc analysis of PFS in patients (n = 158 per arm) with high serum LDH, a potential marker of hypoxia, PFS was longer with PTK/ZK versus placebo (7.7 v 5.8 months, respectively; HR, 0.67; 95% CI, 0.49 to 0.91; P = .009).
Conclusion
Conclusion Although the efficacy objectives of this study were not met, a subgroup of patients who may potentially benefit from small-molecule vascular endothelial growth factor receptor inhibitor therapy has been identified and further research is warranted.
INTRODUCTION
Metastatic colorectal cancer is the second most common cause of cancer death in the Western world.1 The approval of newer chemotherapy agents such as oxaliplatin and irinotecan has significantly improved survival, yet the majority of patients still die in less than 2 years.2,3 Numerous new biologic targets have been examined in the last decade to improve these statistics. Angiogenesis is essential for tumor growth and metastasis, and several studies have revealed a positive correlation between tumor angiogenesis and colorectal cancer stage.4 The vascular endothelial growth factor (VEGF) pathway seems to be a primary regulator of normal and pathologic angiogenesis.5 VEGF-A seems to be the most important member of the VEGF family, followed by three VEGF receptors (VEGFRs).6 Treatment of patients with metastatic colorectal cancer with bevacizumab, a monoclonal antibody to VEGF-A, significantly improved survival when given in combination with fluorouracil (FU) and irinotecan compared with chemotherapy alone.7
VEGFR inhibitors have been shown to have efficacy in the treatment of several malignancies, including renal cell carcinoma and hepatocellular carcinoma.8,9 PTK787/ZK 222584 (PTK/ZK; vatalanib) is a potent, orally active angiogenesis inhibitor that blocks tyrosine kinase signaling of all known VEGF receptors by acting as a competitive inhibitor at the adenosine triphosphate–binding site of the receptor kinase.10 Preclinical studies showed reduction of tumor growth and angiogenesis with PTK/ZK treatment.11,12 Phase I trials indicated that PTK/ZK monotherapy at achievable doses was well tolerated and resulted in changes to tumor blood supply, as measured by dynamic contrast-enhanced magnetic resonance imaging.13,14 Single-agent responses also have been seen in various malignancies.15,16 In addition, a phase Ib trial in which PTK/ZK 1,250 mg once daily was combined with FU, leucovorin (LV), and oxaliplatin (FOLFOX4) showed that this combination was feasible and safe and had promising activity as a first-line treatment in patients with metastatic colorectal cancer.17 Phase II data regarding the efficacy or safety of PTK/ZK plus FOLFOX4 combination therapy were not available when this trial was initiated.
In this report, we prospectively examined the combination of PTK/ZK with FOLFOX4 in patients from the Colorectal Oral Novel Therapy for the Inhibition of Angiogenesis and Retarding of Metastases (CONFIRM 1) trial who were receiving first-line chemotherapy for metastatic adenocarcinoma of the colon or rectum.
PATIENTS AND METHODS
Patients
Eligibility criteria included the following: patients older than age 18 years presenting with histologically or cytologically confirmed metastatic adenocarcinoma of the colon or rectum for first-line chemotherapy and a WHO performance status (PS) of 0 to 2; a life expectancy of at least 12 weeks; measurable lesion(s) as per the modified Response Evaluation Criteria in Solid Tumors (RECIST)18; and provision of written informed consent.
Study Design
This multinational, randomized, double-blind, phase III study (ClinicalTrials.gov identifier: NCT00056459) was sponsored by Novartis and Bayer Schering Pharma as part of a codevelopment program. Patients were randomly assigned to receive PTK/ZK plus FOLFOX4 or placebo plus FOLFOX4. Clinical and study site personnel, patients, and caregivers were blinded to the assigned treatment. The study drug (PTK/ZK) and its placebo were identical in appearance. Random assignment was performed by the sponsor using an automated system.
A preplanned stratified random assignment scheme with a 1:1 ratio was used to stratify for WHO PS (0 v 1 or 2) and serum lactate dehydrogenase (LDH; ≤ v > 1.5× the upper limit of normal), a potential marker of disease burden. FOLFOX4 (day 1: oxaliplatin 85 mg/m2 intravenous [IV] infusion and LV 200 mg/m2 racemate [or 100 mg/m2 l-LV] IV followed by FU 400 mg/m2 IV bolus and FU 600 mg/m2 IV as a 22-hour continuous infusion; day 2: LV 200 mg/m2 racemate [or 100 mg/m2 l-LV] IV over 120 minutes followed by FU 400 mg/m2 IV bolus and FU 600 mg/m2 IV as a 22-hour continuous infusion)3 was administered every 2 weeks in combination with oral PTK/ZK (1,250 mg) or placebo on a once-daily continuous schedule. Predefined dose reductions of chemotherapy, PTK/ZK, or placebo were undertaken in the presence of toxicity. Patients continued on study treatment until the occurrence of disease progression or unacceptable toxicity. Overall survival (OS) and initiation of second-line therapy were observed every 2 months until data cutoff.
Study Assessments
Radiologic tumor assessment was performed every 8 weeks until disease progression using the modified RECIST criteria by local investigators and central review. Adverse events (AEs) were graded according to the National Cancer Institute Common Toxicity Criteria (NCI-CTC) version 2.0.
Study Objectives
The primary objective was to compare the blinded and centrally assessed progression-free survival (PFS) distribution of patients randomly assigned to each treatment (PTK/ZK plus FOLFOX4 v placebo plus FOLFOX4). The key secondary objective was OS. Other secondary objectives included assessment of PFS based on local investigator assessment, time to progression, best overall response rate (ORR; complete response plus partial response), tolerability and safety profile, time to first deterioration in WHO PS, and time to first weight loss greater than 5% relative to baseline. An exploratory post hoc analysis also was performed to examine the effects of PTK/ZK in the randomized strata defined by serum LDH.
Statistical Analyses
The statistical basis for a claim of efficacy was the statistical significance (at the overall 5% significance level) of the treatment comparison for PFS and/or OS in favor of the FOLFOX4 and PTK/ZK treatment group based on the two-sided stratified log-rank test. PFS was defined as time from the date of random assignment to the date of the first observation of documented disease progression during or after discontinuation from study treatment or death as a result of any cause. The study was powered at 90% to detect 25% and 20% reductions in hazard for PFS and OS, respectively. To achieve this result, the analysis of PFS was performed after 562 events, and the analysis of OS was performed after 844 events. Kaplan-Meier plots were used to depict the PFS and OS in each treatment group.
Analysis of PFS and OS was phrased in terms of the null hypothesis (ie, PFS and OS distributions of the two treatment groups were equivalent). In accordance with the stratified random assignment scheme used in this study design, a stratified log-rank test, adjusting for the four strata (WHO PS [0 v 1 or 2] and serum LDH at baseline [≤ v > 1.5× the upper limit of normal) used in the random assignment, was used to test this null hypothesis. A stratified Cochran-Mantel-Haenszel test was performed to determine whether a difference existed in the ORR. The other secondary efficacy variables used the same analysis method that was used for OS.
RESULTS
Patients
Recruitment began on February 20, 2003, and ended on May 27, 2004, with 1,300 patients screened and 1,168 patients randomly assigned. The cutoff date for PFS analysis was October 31, 2004, and the cutoff date for OS analysis was January 13, 2007.
On January 13, 2007, four patients were still receiving PTK/ZK plus FOLFOX4 treatment, and eight patients were continuing on placebo plus FOLFOX4 treatment. Of the 1,156 patients who had discontinued from all study treatments, 58.9% had discontinued because of disease progression, although fewer patients receiving PTK/ZK treatment discontinued for this reason (51.3% v 66.6% in placebo group). Discontinuations of study treatment because of AEs (22.4% in PTK/ZK group v 12.2% in placebo group) and withdrawal of consent (14.7% in PTK/ZK group v 10.1% in placebo group) were more frequent in the PTK/ZK treatment group than the placebo treatment group (Fig 1). No clinically notable differences were seen between the treatment groups in terms of demographic characteristics at baseline (Table 1).
Patient flow diagram and analysis population. (*) Cutoff date for overall survival analysis was January 13, 2007. IVRS, interactive voice response system; PTK/ZK, PTK787/ZK 222584.
Demographics and Clinical Characteristics of Study Population
Efficacy
The primary objective of this study was to compare the PFS distribution (as assessed by the central review) of patients treated with PTK/ZK plus FOLFOX4 versus those treated with placebo plus FOLFOX4. For the overall comparison of the intent-to-treat population, no statistically significant difference was noted between the two treatment groups (Fig 2A). The stair-step pattern for the Kaplan-Meier curve was a result of the timing of the required tumor evaluations. Median PFS was 7.7 months for the PTK/ZK group and 7.6 months for the placebo group (hazard ratio [HR], 0.88; 95% CI, 0.74 to 1.03; P = .118). A preplanned comparison of PFS based on the investigator assessment, a secondary end point, showed a statistically significant difference between the treatment groups in favor of patients treated with PTK/ZK (HR, 0.83; 95% CI, 0.70 to 0.98; P = .026).
Kaplan-Meier curves of (A) progression-free survival based on central radiologic assessment and (B) overall survival. HR, hazard ratio; PTK/ZK, PTK787/ZK 222584; FOLFOX4, oxaliplatin, fluorouracil, and leucovorin.
Analysis of OS, the key secondary end point, was performed after 844 deaths had been observed. For the overall comparison of OS, no statistically significant difference was seen between the two treatment groups (HR, 1.08; 95% CI, 0.94 to 1.24; P = .260). Median OS was 21.4 months for the PTK/ZK group and 20.5 months for the placebo group (Fig 2B).
Other secondary end points included ORR, which showed no statistically significant differences between the two treatment groups. The time to deterioration of WHO PS was longer in the placebo group (HR, 1.55; 95% CI, 1.16 to 2.08; P = .003), as was the time for weight loss of more than 5% relative to baseline (HR, 1.91; 95% CI, 1.58 to 2.30; P < .001).
Patients were stratified by serum LDH levels before random assignment. An exploratory post hoc analysis of PFS in patients with high serum LDH revealed that, for the overall comparison, median PFS was longer in patients treated with PTK/ZK than with placebo (7.7 v 5.8 months, respectively; HR, 0.67; 95% CI, 0.49 to 0.91; P = .009; Fig 3A). Analysis of the OS distribution of patients with high serum LDH did not reveal a meaningful difference between the two treatment groups. Median OS was 14.8 months in the PTK/ZK group and 14.6 months in the placebo group (HR, 1.04; 95% CI, 0.82 to 1.33; P = .751; Fig 3B). There were no differences in PFS or OS between the PTK/ZK and placebo arms in the low serum LDH group (Supplemental Table).
Kaplan-Meier curves of (A) progression-free survival based on central radiologic assessment and (B) overall survival in patients with high levels of serum lactate dehydrogenase. HR, hazard ratio; PTK/ZK, PTK787/ZK 222584; FOLFOX4, oxaliplatin, fluorouracil, and leucovorin.
Exposure and Relative Dose-Intensity
The mean and median overall exposure to study drug were shorter for patients in the PTK/ZK group than the placebo group, and the overall relative dose-intensity was lower for all components of the study treatment in the PTK/ZK group than in the placebo group (Table 2).
Overall Exposure to Study Drug and Overall Relative Dose-Intensity of Study Treatment
Safety
A higher percentage of patients in the placebo group than the PTK/ZK group died during the study or within 28 days of study treatment termination (6.4% v 4.8%, respectively). However, the percentage of patients who died because of disease progression was the same in both treatment groups (3.3%).
Serious AEs and other clinically significant AEs were more frequent in the PTK/ZK treatment group than in the placebo treatment group, as were discontinuations from the study because of these AEs (Table 3).
Deaths, Serious AEs, Clinically Significant AEs,and Discontinuations
NCI-CTC grade 3 or 4 AEs were reported for 85.3% of patients in the PTK/ZK group compared with 77.5% of patients in the placebo group. Of these AEs, the most frequently reported were neutropenia, hypertension, and diarrhea (Table 4). Overall, more patients in the PTK/ZK group than the placebo group experienced NCI-CTC grade 3 or 4 AEs; the most notable differences were for hypertension (23.0% v 6.8%, respectively), diarrhea (15.4% v 11.1%, respectively), dizziness (7.4% v 2.3%, respectively), and pulmonary embolism (5.7% v 1.7%, respectively).
AEs of Maximum NCI-CTC Grade 3 or 4 Occurring in at Least5% of Patients From Either Treatment Group
Three patients in the PTK/ZK group were diagnosed with reversible posterior leukoencephalopathy syndrome (RPLS). No additional patients with RPLS were identified in this study after the recommendation to tightly manage hypertension was implemented. RPLS is associated with other angiogenesis inhibitors, such as bevacizumab and sunitinib.19,20 Similar to the results of the other reported trials, all the patients with RPLS recovered.
DISCUSSION
The addition of PTK/ZK to FOLFOX4 did not improve PFS or OS. However, PTK/ZK did increase toxicity, and more patients withdrew from treatment because of events other than disease progression in the PTK/ZK arm. In light of the activity of bevacizumab in colorectal cancer and of other small-molecule VEGFR inhibitors in other malignancies, together with evidence of biologic effects and single-agent activity of PTK/ZK in colorectal cancer, these negative clinical results were not anticipated.
There are a number of potential explanations. Patients receiving PTK/ZK had more toxicity, leading to reduced exposure to chemotherapy, which may have counteracted any salutary effect of PTK/ZK. PFS is also particularly sensitive to patient discontinuation of study treatment as a result of causes other than disease progression. Some of these toxicities, such as hypertension, occur during treatment with all types of anti-VEGF drugs, and others, such as nausea and fatigue, are observed more frequently with small-molecule inhibitors and may be a result of inhibition of other targets.
Since the initiation of this trial, more has been learned about the pharmacokinetics of PTK/ZK. The half-life of PTK/ZK is relatively short (4 to 6 hours),21 and daily dosing may not adequately inhibit the target VEGFRs, which would have an effect on PFS. This information is particularly important in small-molecule VEGFR inhibitors because their use is associated with an increase in VEGF levels.22 Twice-daily dosing of PTK/ZK is reported to be well tolerated and to significantly increase trough levels.23
Another possible etiology for our disappointing results is the choice of chemotherapy partner. Although oxaliplatin-containing regimens are the most commonly used combination chemotherapies in metastatic colorectal cancer, the addition of bevacizumab in the NO16966 trial, which included 1,401 patients, resulted in a smaller improvement in PFS in the first-line setting than was observed with older regimens with no improvement in OS. However, these results do not explain the differences observed between the negative PTK/ZK and positive bevacizumab trials with second-line FOLFOX therapy.24,25
One interesting result from this study is an apparent benefit of PTK/ZK treatment on PFS in patients with high serum LDH levels. Although the analysis is post hoc, it is strengthened by the use of LDH as a stratification factor for random assignment and by the similar results seen in the Colorectal Oral Novel Therapy for the Inhibition of Angiogenesis and Retarding of Metastases in Second Line (CONFIRM 2) trial.24 LDH has been shown to be a potential marker of poor prognosis for many cancers, including colorectal cancer,26,27 which is why it was used in our random assignment strategy. LDH is critical for anaerobic metabolism, and its expression, like VEGF, is regulated by hypoxia and hypoxia-inducible factor 1α.26 The LDH-5 isozyme has been shown to correlate with total serum LDH and a more aggressive phenotype in patients with colorectal cancer.27 It is possible that high serum LDH levels may identify tumors that are more dependent on abnormal angiogenesis and may be more susceptible to VEGF inhibition. Although high baseline serum LDH was predictive of improved PFS in a recent randomized trial of gemcitabine and erlotinib with or without bevacizumab,28 Hurwitz et al7 found no correlation between the effect of bevacizumab and LDH in metastatic colorectal cancer. Further study of the predictive value of high serum LDH in patients treated with antiangiogenic agents is warranted.
In conclusion, this trial did not show improved efficacy with the addition of PTK/ZK (1,250 mg once daily) to standard FOLFOX chemotherapy in the first-line treatment of patients with metastatic colorectal cancer. Research into the relationship between serum LDH and efficacy also may help to identify patients who would benefit from small-molecule VEGFR inhibitors and other antiangiogenic therapies. Worldwide, many trials with anti-VEGF inhibitors currently are under way, and careful drug development and patient selection may lead to improved outcomes for patients with metastatic colorectal cancer.
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: William Berg, Novartis (C); Bee-Lian Chen, Novartis (C); Tarja Jalava, Bayer Schering Pharma (C); Gerold Meinhardt, Bayer Schering Pharma (C); Dirk Laurent, Bayer Schering Pharma (C); David Lebwohl, Novartis Consultant or Advisory Role: Tanja Trarbach, Schering-Plough (C), Novartis (C); Pierre Major, Amgen (C), Novartis (U); Robert Wolff, Amplimed (C); György Bodoky, Schering-Plough (C), Novartis (C); David Kerr, Bayer Schering Pharma (U) Stock Ownership: William Berg, Novartis; Bee-Lian Chen, Novartis; Gerold Meinhardt, Bayer Pharmaceuticals; Dirk Laurent, Bayer Schering Pharma; David Lebwohl, Novartis Honoraria: Tanja Trarbach, Schering-Plough, Novartis; György Bodoky, Schering-Plough Research Funding: Joel Randolph Hecht, Novartis; Tanja Trarbach, Amgen, Eli Lilly, Saladex; John D. Hainsworth, Novartis; Pierre Major, Amgen, Novartis; Robert Wolff, Eli Lilly; Kelly Pendergrass, Novartis Expert Testimony: Pierre Major, Novartis (U) Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Bee-Lian Chen, Dirk Laurent, David Lebwohl, David Kerr
Provision of study materials or patients: J. Randolph Hecht, Tanja Trarbach, John D. Hainsworth, Pierre Major, Elke Jäger, Robert A. Wolff, György Bodoky, Kelly Pendergrass, David Kerr
Collection and assembly of data: J. Randolph Hecht, Tanja Trarbach, Katherine Lloyd-Salvant, Tarja Jalava, David Lebwohl
Data analysis and interpretation: J. Randolph Hecht, Tanja Trarbach, William Berg, Bee-Lian Chen, Tarja Jalava, Gerold Meinhardt, Dirk Laurent, David Lebwohl, David Kerr
Manuscript writing: J. Randolph Hecht, Tanja Trarbach, John D. Hainsworth, Pierre Major, Elke Jäger, Robert A. Wolff, Katherine Lloyd-Salvant, György Bodoky, Kelly Pendergrass, William Berg, Bee-Lian Chen, Tarja Jalava, Gerold Meinhardt, Dirk Laurent, David Lebwohl, David Kerr
Final approval of manuscript: J. Randolph Hecht, Tanja Trarbach, John D. Hainsworth, Pierre Major, Elke Jäger, Robert A. Wolff, Katherine Lloyd-Salvant, György Bodoky, Kelly Pendergrass, William Berg, Bee-Lian Chen, Tarja Jalava, Gerold Meinhardt, Dirk Laurent, David Lebwohl, David Kerr
Acknowledgment
We thank the Colorectal Oral Novel Therapy for the Inhibition of Angiogenesis and Retarding of Metastases (CONFIRM 1) Study Group investigators; Andrea Wagner, Martina Poethig (Bayer Schering Pharma), Christian Jacques, Andrew Henry, Carol Gano, Michael Shi, Jay Mei, and Honeylet Wortman-Vayn (Novartis) for study design, conduct, and management; and ICON Clinical Research (United Kingdom) Limited and Scientific Connections for medical writing assistance.
Footnotes
-
See accompanying editorial on page 1938 and article on page 2004
-
Supported by Novartis and Bayer Schering Pharma as part of a codevelopment program.
-
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: NCT00056459.
- Received March 19, 2010.
- Accepted December 2, 2010.
