- © 2007 by American Society of Clinical Oncology
Phase II Trial of Sunitinib in Patients With Metastatic Colorectal Cancer After Failure of Standard Therapy
- Leonard B. Saltz,
- Lee S. Rosen,
- John L. Marshall,
- Robert J. Belt,
- Herbert I. Hurwitz,
- S. Gail Eckhardt,
- Emily K. Bergsland,
- Daniel G. Haller,
- A. Craig Lockhart,
- Caio M. Rocha Lima,
- Xin Huang,
- Samuel E. DePrimo,
- Edna Chow-Maneval,
- Richard C. Chao and
- Heinz J. Lenz
- From the Memorial Sloan-Kettering Cancer Center, New York, NY; Premiere Oncology, Santa Monica; University of California, San Francisco; Pfizer Global Research and Development, La Jolla; University of Southern California, Los Angeles, CA; Lombardi Cancer Center, Washington, DC; Oncology Hematology of Kansas City, Kansas City, MO; Duke University Medical Center, Durham, NC; University of Colorado Cancer Center, Aurora, CO; University of Pennsylvania, Philadelphia, PA; Vanderbilt University Medical Center, Nashville, TN; and University of Miami, Miami, FL
- Address reprint requests to Leonard B. Saltz, MD, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; e-mail: saltzl{at}mskcc.org
Abstract
Purpose Sunitinib is an oral, multitargeted receptor tyrosine kinase inhibitor of the vascular endothelial growth factor receptor and multiple other growth factor receptors. We assessed the safety and efficacy of sunitinib in patients with metastatic colorectal cancer after failure of standard therapy.
Patients and Methods Eighty-four patients were enrolled onto this two-stage phase II trial and were stratified by whether they had received prior bevacizumab (n = 43) or not (n = 41). Treatment comprised sunitinib 50 mg orally daily for 4 weeks, followed by 2 weeks off treatment, in repeated 6-week cycles.
Results By Response Evaluation Criteria in Solid Tumors criteria, one patient, who was in the prior bevacizumab cohort, achieved a partial response, and 13 patients (two in the prior bevacizumab cohort and 11 in the no prior bevacizumab cohort) achieved stable disease lasting ≥ 22 weeks. Median time to progression in the prior bevacizumab and bevacizumab-naïve cohorts was 2.2 months (95% CI, 1.9 to 2.3 months) and 2.5 months (95% CI, 2.3 to 3.1 months), respectively, whereas median overall survival time was 7.1 months (95% CI, 4.9 to 10.6 months) and 10.2 months (95% CI, 8.2 to 15.3 months), respectively. The most common adverse events were fatigue, diarrhea, nausea, vomiting, and anorexia. Twenty-six patients (32%) required dose reduction to 37.5 mg/d, and one patient required dose reduction to 25 mg/d.
Conclusion Sunitinib did not demonstrate a meaningful single-agent objective response rate in colorectal cancer refractory to standard chemotherapy. However, the mechanisms of action and acceptable safety profile of sunitinib warrant further study in combination with standard regimens for metastatic colorectal cancer.
INTRODUCTION
Sunitinib (SU11248; Sutent; Pfizer Inc, La Jolla, CA) is an oral, multitargeted tyrosine kinase inhibitor that selectively inhibits the tyrosine kinases of the vascular endothelial growth factor (VEGF) receptor and platelet-derived growth factor receptor family members, as well as stem-cell factor receptor (KIT), glial cell line–derived neurotrophic factor receptor (rearranged during transfection; RET), colony-stimulating factor receptor (CSF-1R), and FMS-like tyrosine kinase-3 (FLT3; Pfizer, Inc; data on file).1-4 This compound has shown activity in imatinib-refractory gastrointestinal stromal tumor and advanced renal cell carcinoma.5,6 The clinical activity seen thus far with sunitinib may involve both a direct antitumor effect and an anti-VEGF effect.1,2,7-10
Fluorouracil has been the cornerstone of treatment of colorectal cancer for many decades, and additional therapies have become available only relatively recently. These include irinotecan, oxaliplatin, capecitabine, cetuximab, and bevacizumab. In the clinic, inhibition of VEGF with the monoclonal antibody bevacizumab significantly improves survival and other outcomes in patients receiving chemotherapy for metastatic colorectal cancer.11-13 In addition, preclinical studies have demonstrated the role of platelet-derived growth factors in modulating angiogenesis and inducing production of VEGF14-17 and have shown that the coordinated inhibition of both growth factors results in reduced tumor vascularization and greater tumor regression.18,19
Despite the introduction of these new treatments, the 5-year survival rate for metastatic colorectal cancer remains just 10%,20 and additional active agents are needed to improve the prognosis of this population. Here, we report the results of an open-label, phase II study assessing the safety and efficacy of sunitinib in patients with advanced, unresectable colorectal cancer after failure of standard therapy.
PATIENTS AND METHODS
Patients
The study population included patients with incurable histologically or cytologically proven colorectal adenocarcinoma. Eligible patients received prior treatment with irinotecan, oxaliplatin, and a fluoropyrimidine, administered in any combination or sequence in the adjuvant or advanced disease settings. Patients may also have received treatment with bevacizumab or cetuximab for advanced disease. Evidence of resistance to prior treatment was required and was defined either by relapse or disease progression (as determined by the treating clinician) during or within 6 months of completing the most recent regimen with irinotecan, oxaliplatin, or fluoropyrimidine or by disease progression during or within 6 months of finishing bevacizumab therapy, if prior bevacizumab had been received. Other inclusion criteria included the following: age of 18 years or older; performance status of 0 or 1 (Eastern Cooperative Oncology Group); unidimensionally measurable disease; no prior anticancer therapy for at least 3 weeks before study entry; absolute neutrophil count ≥ 1.5 × 109/L; platelets ≥ 100 × 109/L; hemoglobin ≥ 9 g/dL; and adequate hepatic, renal, and cardiac function.
Patients were excluded if they had received prior treatment with more than three cytotoxic chemotherapy regimens for advanced disease; had been previously treated with any tyrosine kinase inhibitors or VEGF inhibitors other than bevacizumab; had known brain or leptomeningeal disease; had clinically significant cardiovascular disease or prolonged QTc interval; or were pregnant or breastfeeding.
Study Design and Treatment
This study was an open-label, multicenter, two-stage, phase II clinical trial. Patients were assigned to one of two cohorts, depending on whether or not they had received prior bevacizumab treatment. The study was reviewed and approved by the institutional review boards of each participating center, and all patients gave written informed consent before enrollment onto this trial.
The primary end point of the study was the objective tumor response rate, with responses defined by Response Evaluation Criteria in Solid Tumors criteria.21 Secondary end points included time to tumor response, duration of response, time to tumor progression (TTP), and overall survival.
Sunitinib was administered at a starting dose of 50 mg once daily for 4 consecutive weeks, followed by 2 weeks off treatment, in repeated 6-week cycles. Depending on individual patient tolerance, the once-daily dose of sunitinib could be reduced to 37.5 mg and subsequently to 25 mg as needed.
Study Assessments
Tumor response was assessed at baseline (within 21 days of starting treatment), at the end of cycles 2, 4, 6, and 8, and at the end of treatment. All responses were confirmed 4 weeks or longer after first documentation. Patients were observed every 2 months for 1 year from the first dose of study therapy for information on disease progression, survival status, and additional cancer treatments.
Safety was assessed at regular intervals by recording adverse events, hematology and clinical chemistry parameters, physical examination, and echocardiogram or multigated acquisition scans. Adverse events were graded using the National Cancer Institute Common Terminology Criteria of Adverse Events (version 3).
Blood samples were collected for pharmacokinetic studies and soluble protein analyses. Plasma concentrations of sunitinib and its active metabolite SU12662 were determined using a liquid chromatography/mass spectrometry method at BASi (West Lafayette, IN), with a lower limit of detection of 0.1 ng/mL for each compound. Plasma samples were analyzed for soluble proteins (VEGF-A, soluble VEGF receptor-2 [sVEGFR-2], soluble VEGF receptor-3 [sVEGFR-3], and soluble KIT [sKIT]) using sandwich enzyme-linked immunosorbent assay kits (R&D Systems, Minneapolis, MN).
Statistical Analysis
This study used Simon's Minimax two-stage design22 to test the null hypothesis that the true overall response rate was ≤ 5% (which would not be clinically meaningful), as opposed to the alternative hypothesis that the true overall response rate was ≥ 15%. Up to 63 patients were planned for each cohort (prior bevacizumab and no prior bevacizumab cohorts) to assess the overall response rate with 85% power and α = .05. Stage 1 comprised 38 patients recruited to each cohort. If two or more objective tumor responses were observed in a given cohort, an additional 25 patients would be enrolled onto that cohort in stage 2.
The study population for all analyses included every patient who received at least one dose of study medication. TTP was defined as the time from first sunitinib dose to first documentation of objective tumor progression. Overall survival was defined as the time from first sunitinib dose to death as a result of any cause. Time-to-event and overall survival data were summarized using the Kaplan-Meier method.
RESULTS
Patients were enrolled between December 2003 and November 2005. Many of the patients accrued to the bevacizumab-naïve cohort began their treatment for colorectal cancer before the availability or widespread use of bevacizumab in the treatment of this disease. In total, 82 patients were treated, 42 of whom had received prior bevacizumab therapy. Baseline patient characteristics are listed in Table 1 and show a median age of 57 years, a slight male predominance, good performance status, and the colon as the primary tumor site in the majority of patients (n = 66; 80.5%). All patients had received prior treatment with irinotecan, oxaliplatin, and a fluoropyrimidine, and 63 patients (76.8%) had also received treatment with another therapeutic agent (excluding bevacizumab) before study entry, including cetuximab (17 patients [40.5%] in the prior bevacizumab cohort and three patients [7.5%] in the bevacizumab-naïve cohort).
Baseline Patient Characteristics
A total of 68 patients (83%) completed at least one cycle of treatment, and 55 patients (67%) received additional treatment cycles. The median number of treatment cycles was one (range, zero to four cycles) for patients previously treated with bevacizumab and two (range, zero to nine cycles) for patients who had not received bevacizumab therapy. Overall, the median daily dose was 50 mg in both patients who had received prior bevacizumab (range, 34.4 to 50 mg) and patients who were bevacizumab naïve (range, 39.5 to 50 mg).
At the time of data analysis, no patients remained on study treatment. The median duration from the time of first sunitinib dose to termination from study (including 28 days after the last dose) was 2.3 months (range, 0.3 to 5.6 months) and 2.8 months (range, 0.5 to 12.0 months) in the prior bevacizumab and bevacizumab-naïve cohorts, respectively. In the prior bevacizumab cohort, reasons for stopping treatment included disease progression in 36 patients (86%), adverse events in three patients (7%), and consent withdrawn in three patients (7%). In the bevacizumab-naïve cohort, discontinuations were a result of disease progression in 32 patients (80%), adverse events in four patients (10%), and consent withdrawn in two patients (5%); in addition, one patient was lost to follow-up, and another patient completed treatment as per protocol.
Efficacy
One patient in the prior bevacizumab cohort achieved a confirmed partial response (Table 2; Fig 1). This 38-year-old female patient initially received irinotecan, fluorouracil, and leucovorin over 6 months, followed 4 months later by treatment with oxaliplatin, fluorouracil, leucovorin, and bevacizumab over 4 months; she then enrolled onto the current study. Tumor assessments showed a 6-cm peripancreatic lymph node that decreased from 6 cm at baseline to 1.5 cm in cycle 2, to 1.6 cm in cycle 3, and to 0 cm in cycle 4, at which time she had developed a new nodal mass indicating progressive disease.
Maximum percent change in target lesion size by patient during sunitinib treatment (n = 72; 10 patients withdrew from the study before their first restaging scan). *One patient with SD has a maximal change of zero, and is not visible in this graph. PR, partial response; SD, stable disease; RECIST, Response Evaluation Criteria in Solid Tumors Group.
Best Clinical Response to Sunitinib Treatment
There were no objective responses in the bevacizumab-naïve cohort; however, 11 patients (28%) experienced stable disease for at least 22 weeks. The maximum percent changes from baseline in target lesions from 72 patients (10 patients discontinued the study before the first tumor assessments at cycle 2 and had no postbaseline scans) are shown in Figure 1.
The median TTPs in the prior bevacizumab and bevacizumab-naïve cohorts were 2.2 months (95% CI, 1.9 to 2.3 months) and 2.5 months (95% CI, 2.3 to 3.1 months), respectively (Fig 2). Approximately 50% of patients in both treatment groups had experienced progression at the time of the first restaging computed tomography scan (ie, day 28 of cycle 2, or 10 weeks after starting treatment). The median overall survival time was 7.1 months (95% CI, 4.9 to 10.6 months) in the prior bevacizumab cohort and 10.2 months (95% CI, 8.2 to 15.3 months) in the bevacizumab-naïve cohort (Fig 2).
Time to tumor progression (TTP) and overall survival in patients with advanced colorectal carcinoma after treatment with sunitinib (Kaplan-Meier analysis).
Pharmacokinetics
There were no differences observed in the pharmacokinetics of patients previously treated with bevacizumab compared with patients who did not receive prior bevacizumab therapy. Patients achieved and maintained steady-state trough (Ctrough) plasma concentrations of sunitinib and SU12662 throughout the dosing periods for multiple cycles. Median Ctrough concentrations of sunitinib and SU12662 combined ranged from 59 to 164 ng/mL. Neither sunitinib nor SU12662 accumulated across dosing cycles. Ctrough plasma levels of sunitinib and SU12662 combined were similar to those reported previously.23 Both patient populations demonstrated median concentrations within the range of 50 to 100 ng/mL, which has been shown to inhibit target receptor tyrosine kinases in preclinical models.4
Plasma Soluble Protein Biomarkers
Plasma levels of VEGF, sVEGFR2, sVEGFR3, and sKIT were serially measured at baseline and at regular intervals over several cycles. Significant changes (P < .00001) in the mean plasma levels for all four soluble proteins were noted in the first treatment cycle among the entire study population. Compared with baseline, mean levels of VEGF increased by approximately 3.9-fold after one cycle of sunitinib treatment, whereas at the same time, mean levels of sVEGFR-2 (45.2%), sVEGFR-3 (40.5%), and sKIT (27.3%) decreased. Of note, VEGF baseline levels differed between the bevacizumab-naïve and prior bevacizumab cohorts; baseline VEGF levels were more than seven-fold higher in the prior bevacizumab cohort than in the bevacizumab-naïve cohort (433 v 57.4 pg/mL, respectively; P < .05) and seemed to correlate inversely with the time since cessation of bevacizumab treatment (higher VEGF levels correlated with shorter time intervals since the last dose). Linear regression analysis of trough drug levels and changes in these proteins indicated a moderate direct correlation with VEGF (R2 = 0.36) and weaker inverse correlations with sVEGFR-2 (R2 = 0.28) and sVEGFR-3 (R2 = 0.13); no correlation between trough levels and sKIT changes was observed (R2 < 0.01). At the end of the first cycle, patients with tumor size reduction had significantly higher induction in VEGF (P = .027) and greater reduction in sVEGFR-3 (P = .0001) than patients with tumor size increases; this trend was stronger in the bevacizumab-naïve cohort compared with the prior bevacizumab cohort.
Safety and Tolerability
Overall, 38 patients (46%; 25 [60%] with and 13 [33%] without prior bevacizumab) experienced a delay or change in dosing as a result of adverse events during the study. The dose of sunitinib was reduced from 50 to 37.5 mg in 26 patients (32%) and subsequently to 25 mg/d in one of these 26 patients. Dosing was delayed in 33 patients (40%). Adverse events that most commonly led to dose reductions or delays included GI disorders (16 patients, 20%), including vomiting in six patients (7%) and nausea in three patients (4%); general disorders and administration site conditions (14 patients, 17%), including fatigue in 11 patients (13%); blood and lymphatic system disorders (six patients, 7%), including thrombocytopenia in four patients (5%) and neutropenia in three patients (4%); and skin and subcutaneous tissue disorders (five patients; 6%), including hand-foot syndrome in four (5%) patients.
Seven patients (9%; three in the prior bevacizumab group and four in the bevacizumab-naïve group) discontinued treatment after an adverse event. In the prior bevacizumab group, two patients (5%) discontinued after treatment-related adverse events (grade 3 thrombocytopenia and grade 3 diarrhea, which were attributed to both the underlying disease and to sunitinib). In the bevacizumab-naïve group, one patient (3%) experienced a grade 3 perianal abscess (also attributed both to the disease and sunitinib).
The most commonly experienced adverse events, regardless of causality, were consistent with those associated with metastatic colorectal cancer and the known adverse effects of sunitinib treatment. Grade 3 or 4 events reported by five or more patients included fatigue (n = 14; 17%), diarrhea (n = 11; 13%), thrombocytopenia (n = 7; 9%), hypertension (n = 6; 7%), and vomiting (n = 5; 6%).
Treatment-related adverse events occurring in at least 10% of patients are listed in Table 3; the adverse events most often reported included fatigue, diarrhea, nausea, vomiting, and anorexia. Grade 3 treatment-related adverse events experienced by five or more patients were fatigue (n = 12; 15%), diarrhea (n = 9; 11%), and thrombocytopenia (n = 7; 9%). No treatment-related grade 4 events were reported in more than one patient.
Treatment-Related Adverse Events Experienced by ≥ 10% of Patients by Worst Toxicity Grade According to NCI CTCAE (N = 82)
No patients died during the study while receiving study treatment. Five patients (6%) died within 28 days of the last dose of study drug (defined as the on-study period); all of these deaths were a result of disease progression.
DISCUSSION
Inhibition of the VEGF signaling pathway has been validated as an approach to treating patients with colorectal cancer by results achieved with the anti-VEGF monoclonal antibody bevacizumab.11,12,24,25 The development of bevacizumab in the colorectal cancer population was unusual in oncology drug development because the antibody was combined with chemotherapy for first-line treatment of these patients immediately after phase I studies without preliminary evidence of efficacy in higher lines of treatment.26 However, it was in this setting that adding bevacizumab to chemotherapy first demonstrated a significant and substantial survival benefit of 4.7 months in patients with advanced colorectal cancer, despite a relatively modest increase of 10% in the objective response rate.11
Patients in this sunitinib study were heavily pretreated. In this population, one patient who had been previously treated with bevacizumab had a confirmed partial response, and 13 patients achieved stable disease for 22 weeks or longer, suggestive of a clinical benefit even for those without significant tumor shrinkage, including one patient with stable disease at 10.6 months. TTP was similar in the prior bevacizumab and bevacizumab-naïve groups, although a greater number of patients achieved durable stable disease in the bevacizumab-naïve patient group. The patient with a confirmed partial response exhibited a maximum decrease of 75% in tumor size (sum of longest dimensions in the target lesions), and the 13 patients with durable stable disease showed maximum decreases in tumor sizes ranging from 3% to 51%.
The safety profile in this study was similar to that seen with sunitinib in other patient populations.5,27,28 Most treatment-related adverse events were mild or moderate (ie, grades 1 or 2).
Analysis of circulating proteins that might serve as biomarkers of antiangiogenic activity indicates a pharmacodynamic effect of sunitinib. The effects on VEGF and VEGFRs were generally consistent with results from similar biomarker analyses reported in phase II studies in metastatic breast cancer and metastatic renal cell carcinoma, in which the same dose and schedule for sunitinib was used.29,30 In this study, differences in baseline VEGF levels were observed between the prior bevacizumab and bevacizumab-naïve cohorts; this may indicate feedback induction after withdrawal of bevacizumab (rather than residual effect of reduced VEGF clearance, because the enzyme-linked immunosorbent assay used in this study does not readily detect VEGF bound to bevacizumab [Pfizer Inc, data on file]).31 However, the increased levels of free VEGF may have also resulted from dissociation of bound VEGF from bevacizumab. Additionally, a correlation between drug trough levels and plasma biomarkers was observed, supporting a potential association between biomarkers and adverse events that has been similarly postulated in previous studies.32 These results suggest that trough drug levels, soluble biomarkers, or both may be useful in a model of sunitinib dosing. Finally, preliminary indication of potential association between tumor size reductions and changes in VEGF and sVEGFR-3 was observed.
Sunitinib failed to demonstrate a significant single-agent objective response rate in patients with heavily pretreated, metastatic colorectal cancer. However, the mode of action of sunitinib, along with an acceptable safety profile, should facilitate its incorporation into standard regimens for treating metastatic colorectal cancer in the clinic, particularly in patients less heavily pretreated. Biomarker exploration in this study compared favorably with other disease-specific phase II clinical trials, suggesting that there is nothing unusual about the way colorectal cancer patients respond to the drug. Further research will hopefully elucidate the clinical significance of these biomarkers so as to perhaps select patients more likely to receive benefit from sunitinib. Studies combining sunitinib with infusional fluorouracil, leucovorin, and oxaliplatin, and with infusional fluorouracil, leucovorin, and irinotecan regimens in patients with metastatic colorectal cancer are ongoing.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. 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: Xin Huang, Pfizer Inc (C); Samuel E. DePrimo, Pfizer Inc (C); Edna Chow-Maneval, Pfizer Inc (C); Richard C. Chao, Pfizer Inc (C) Consultant or Advisory Role: Leonard B. Saltz, Pfizer Inc (C), Roche (C), Genentech (C), Bayer (C); John L. Marshall, Pfizer Inc (C); Herbert I. Hurwitz, Pfizer Inc (C), Roche (C), Genentech (C), Amgen Inc (C); S. Gail Eckhardt, Pfizer Inc (C); Daniel G. Haller, Pfizer Inc (C); Heinz J. Lenz, Pfizer Inc (C), ResponseGenetics (C) Stock Ownership: Xin Huang, Pfizer Inc; Samuel E. DePrimo, Pfizer Inc; Edna Chow-Maneval, Pfizer Inc; Richard C. Chao, Pfizer Inc; Heinz J. Lenz, ResponseGenetics Honoraria: John L. Marshall, Pfizer Inc; Herbert I. Hurwitz, Genentech, Roche; Caio M. Rocha Lima, Pfizer Inc; Heinz J. Lenz, Pfizer Inc Research Funding: Leonard B. Saltz, Pfizer Inc, Roche, Bayer, Taiho, ImClone Systems Inc, Bristol-Myers Squibb Co, Genentech; Lee S. Rosen, Pfizer Inc; John L. Marshall, Pfizer Inc; Robert J. Belt, Pfizer Inc; Herbert I. Hurwitz, Pfizer Inc, Roche, Genentech, AstraZeneca, GlaxoSmithKline, Sanofi aventis, Amgen Inc, Bristol-Myers Squibb Co, Adnexus, Cephalon, Sunisis; S. Gail Eckhardt, Pfizer Inc; Daniel G. Haller, Pfizer Inc; Heinz J. Lenz, Pfizer Inc Expert Testimony: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Leonard B. Saltz, Samuel E. DePrimo, Richard C. Chao, Heinz J. Lenz
Administrative support: Edna Chow-Maneval
Provision of study materials or patients: Leonard B. Saltz, Lee S. Rosen, John L. Marshall, Robert J. Belt, Herbert I. Hurwitz, S. Gail Eckhardt, Emily K. Bergsland, Daniel G. Haller, A. Craig Lockhart, Caio M. Rocha Lima, Heinz J. Lenz
Collection and assembly of data: Xin Huang, Samuel E. DePrimo, Edna Chow-Maneval, Richard C. Chao
Data analysis and interpretation: Leonard B. Saltz, Herbert I. Hurwitz, Xin Huang, Samuel E. DePrimo, Richard C. Chao
Manuscript writing: Leonard B. Saltz, Samuel E. DePrimo, Edna Chow-Maneval, Richard C. Chao, Heinz J. Lenz
Final approval of manuscript: Leonard B. Saltz, Lee S. Rosen, John L. Marshall, Robert J. Belt, Herbert I. Hurwitz, S. Gail Eckhardt, Emily K. Bergsland, Daniel G. Haller, A. Craig Lockhart, Caio M. Rocha Lima, Xin Huang, Samuel E. DePrimo, Edna Chow-Maneval, Richard C. Chao, Heinz J. Lenz
Acknowledgments
Editorial assistance was provided by ACUMED (Tytherington, United Kingdom) and was funded by Pfizer Inc.
Footnotes
-
Supported by Pfizer Inc.
-
Presented in part at 42nd Annual Meeting of the American Society of Clinical Oncology, June 2-6, 2006, Atlanta, GA.
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
- Received May 30, 2007.
- Accepted July 30, 2007.
