| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Report From the FDA |
Authors' Affiliations: Offices of 1 Oncology Drug Products, 2 Biotechnology Products, 3 Biostatistics, and 4 Clinical Pharmacology, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, Maryland
Requests for reprints: Ruthann M. Giusti, Division of Biologic Oncology Products, Office of Oncology Drug Products, Center for Drug Evaluation and Research, Food and Drug Administration, 10903 New Hampshire Avenue, Building 22, Room 2316, Silver Spring, MD 20993-0004. Phone: 301-796-1359; E-mail: ruthann.giusti{at}fda.hhs.gov.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Experimental Design: Food and Drug Administration reviewed a single, open-label, multicenter trial in which 463 patients with epidermal growth factor receptor–expressing metastatic colorectal cancer who had progressed on or following treatment with a regimen containing a fluoropyrimidine, oxaliplatin, and irinotecan were randomized (1:1) to receive best supportive care (BSC) with or without panitumumab (6 mg/kg every other week) administered until disease progression or intolerable toxicity. Progression and response were confirmed by an independent review committee masked to treatment assignment. At progression, patients in the BSC-alone arm were eligible to receive panitumumab.
Results: Although median progression-free survival (PFS) was similar in both treatment arms (
8 weeks), the mean PFS was 50% longer among patients receiving panitumumab than among those receiving BSC alone (96 versus 60 days, respectively) and the objective response rate in patients receiving panitumumab was 8%. However, no difference in overall survival was shown between the two study arms.
Conclusions: Panitumumab received accelerated approval based on improvement in PFS and an independently confirmed response rate of 8%, similar to that observed with other active agents at this advanced stage of disease. Confirmation of clinical benefit will be required for full approval.
In this report, we describe the accelerated approval of panitumumab, which, like cetuximab, binds specifically to the human EGFR. In contrast to cetuximab, panitumumab is a fully human IgG2
monoclonal antibody and was therefore expected to be less immunogenic and to have reduced Fc domain-based effector functions, such as antibody-dependent cellular cytotoxicity and complement-dependent cellular cytotoxicity.
EGFR, a transmembrane glycoprotein, is a member of a subfamily of type I receptor tyrosine kinases, which includes EGFR (HER1, c-ErbB-1), HER2/neu, HER3, and HER4. EGFR is constitutively expressed in many normal epithelial tissues, including the skin and hair follicle. Overexpression of EGFR is also detected in many human cancers, including those of the colon and rectum. Interaction of EGFR with its normal ligands (e.g., EGF and transforming growth factor-
) leads to phosphorylation and activation of a series of intracellular tyrosine kinases, which in turn regulate transcription of molecules involved with cellular growth, survival, motility, proliferation, and transformation. Binding of panitumumab to EGFR competitively inhibits the binding of ligands for EGFR, resulting in inhibition of cell growth, induction of apoptosis, decreased proinflammatory cytokine and vascular growth factor production, and internalization of the EGFR. In vitro assays and in vivo animal studies show that panitumumab inhibits the growth and survival of selected human tumor cell lines expressing EGFR (1).
Panitumumab was evaluated for pharmacologic activity in human tumor cell lines in vitro and in human tumor xenografts in immunodeficient mice and for toxicity and pharmacokinetics in immunodeficient mice and cynomolgus monkeys. Tissue binding studies showed that panitumumab bound with moderate to strong intensity to surface EGFR in tissue samples of epithelial origin from both human and cynomolgus monkey. Treatment of tumor-bearing immunodeficient mice with panitumumab alone or in combination with several different biological or chemotherapy regimens resulted in delayed tumor growth of human colon, epidermoid, breast, or pancreatic cancers (1). No antitumor effects of panitumumab were observed in xenograft models using human tumors with levels of EGFR expression below 10,000 receptors per cell, suggesting that a threshold level of EGFR expression is required for tumor response to panitumumab (2).
Severe dermatologic and gastrointestinal toxicities were noted at all dose levels in cynomolgus monkeys treated weekly for 4 to 26 weeks with doses of panitumumab corresponding to 1.25- to 10-fold greater than the human dose of 6 mg/kg panitumumab when adjusted for body weight and occurred with increased frequency and severity as both the dose and duration of panitumumab increased. Toxicities included moderate to severe diarrhea, dehydration, hypomagnesemia, hypocalcemia, and hypophosphatemia. Extensive skin toxicities, including skin rash, erythema, suppurative dermatitis, erosions, sloughing, and ulcerations, were observed. In several studies, skin toxicity resulted in early mortalities and necessitated palliative care and/or euthanasia. The dermatologic toxicities observed in these studies are consistent with the pharmacodynamic effects of panitumumab in inhibiting critical intracellular pathways involved with the activation and function of EGFR expressed on skin cells and subsequent inhibition of epidermal cell growth and maturation.6
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
10% of tumor cells; this was amended after enrollment of 99 patients (21% of the study population) to permit enrollment of patients with at least 1+ staining in 1% of tumor cells. The study was sized to achieve 90% power for a two-sided 1% significance level test given a hazard ratio (panitumumab plus BSC:BSC) of 0.67, a 50% relative median increase, 2.5 versus 3.75 months, assuming exponential PFS. Patients were randomized 1:1 to receive panitumumab at a dose of 6 mg/kg given once every 2 weeks plus BSC (n = 231) or BSC alone (n = 232) until investigator-determined disease progression. Randomization was stratified based on Eastern Cooperative Oncology Group (ECOG) performance status (0-1 versus 2) and geographic region (Western Europe, Eastern/Central Europe, or other). The primary end point of the trial was PFS, with secondary end points of estimation of objective response rate, response duration, OS, and toxicity profile. The primary analyses of PFS, OR, and response duration were based on events confirmed by the IRC composed of a panel of radiologists and a medical oncologist who were masked to treatment assignment. On investigator-determined disease progression, patients in the BSC-alone arm were eligible to receive panitumumab. In these patients, radiologic imaging and clinical information were collected until disease progression was confirmed by the IRC. In the primary analysis of PFS, the IRC-determined time to progression event was used for patients in the BSC arm even if that event occurred while the patient was receiving panitumumab following an unconfirmed investigator-determined progression event.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
65 years, 99% were Caucasian, 86% had a baseline ECOG performance status of 0 or 1, and 67% had colon cancer. All patients were diagnosed with metastatic disease at the time of the primary diagnosis. The median number of prior therapies for metastatic disease was 2. Patients randomized to receive panitumumab were similar to those randomized to receive BSC alone in baseline demographic and disease characteristics (Tables 1
and 2
).
|
|
|
|
OS and best objective response were cosecondary end points. There were 19 partial responses among the 231 patients receiving panitumumab (8%) and no responses among patients receiving BSC alone (P < 0.0001). The median duration of response was 17 weeks with a 95% confidence interval (95% CI) of 16.4 to 25.3. A total of 250 deaths occurred by the cutoff date (panitumumab plus BSC = 119; BSC = 131). There was no significant difference between arms in OS. Median survival times for the panitumumab plus BSC group and BSC-alone group were 193 and 184 days, respectively (Fig. 2 ). No correlation was identified between the number of receptor molecules and panitumumab (or cetuximab) response or PFS.
|
In the randomized clinical trial, the exposure to panitumumab was short term with a median duration of therapy of 56 days (range, 1-421 days). Fifty percent of patients on the randomized trial received four or fewer panitumumab infusions. The median weight-adjusted cumulative dose was 26.1 mg/kg (range, 0.19-151.3 mg/kg) and the median average weight-adjusted dose delivered was 6.0 mg/kg (range, 0.19-6.56 mg/kg).
Although analysis of the clinical trial data was confounded due to crossover of patients randomized to the BSC arm, there did not seem to be a difference in the numbers of deaths on the panitumumab arm (Table 4 ). Most deaths were, as expected, attributed to progressive disease. Commonly occurring adverse events observed in the randomized clinical trial are shown in Table 5 .
|
|
The median time to the development of skin/eye-related toxicity was 14 days; the time to most severe skin/eye-related toxicity was 15 days after the first dose of panitumumab; and the median time to resolution after the last dose of panitumumab was 84 days. Severe skin toxicity necessitated dose interruption in 11% of panitumumab-treated patients. Sepsis, septic death, and abscesses requiring incisions and drainage were reported after the development of severe dermatologic toxicities.
Infusion reactions. Infusional toxicity was defined as any event described as "allergic reaction" or "anaphylactoid reaction" at any time during the study or any event described as "allergic reaction," "anaphylactoid reaction," "fever," "chills," or "dyspnea" occurring on the first day of dosing. Vital signs and temperature were measured within 30 min before initiation and on completion of the panitumumab infusion.
In the randomized, controlled clinical trial, 4% of patients experienced infusion reactions, and in 1% reactions were graded as severe [National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events grade 3-4]. Fatal infusion reactions have not been reported to date with panitumumab. However, there is insufficient data to conclude that the rate or severity of infusion reactions differs substantially from other agents in the class or that patients who experience an infusion reaction with cetuximab can be safely treated with panitumumab.
Pulmonary fibrosis. Pulmonary fibrosis was reported as a rare adverse event following panitumumab treatment and has been observed cetuximab (see cetuximab prescribing information). Pulmonary fibrosis occurred in 1% (2 of 1,467) of patients enrolled in all panitumumab clinical studies. One of these two cases occurred in a 43-year-old male patient with non–small cell lung cancer and a history of idiopathic pulmonary fibrosis treated with panitumumab in combination with carboplatin and paclitaxel chemotherapy. The patient developed worsening shortness of breath concurrent with computed tomography evidence of worsening pulmonary fibrosis and partial disease response and died with respiratory failure and worsening pulmonary fibrosis after four doses of panitumumab. In the second case, the patient developed cough and wheezing 8 days following the initial panitumumab dose, exertional dyspnea on the day of the seventh panitumumab dose, and persistent symptoms and computed tomography evidence of pulmonary fibrosis following the eleventh dose. In this case, panitumumab was administered as monotherapy. After the initial fatality was reported, patients with a history of interstitial pneumonitis, pulmonary fibrosis, evidence of interstitial pneumonitis, or pulmonary fibrosis were excluded from clinical studies. Therefore, the risk of panitumumab use in such patients is uncertain.
Diarrhea. Panitumumab is approved for use as monotherapy. In the randomized clinical trial, diarrhea (any grade) was observed in 21% of patients receiving panitumumab and in 11% of patients receiving BSC alone. Grade 3 to 4 diarrhea was observed in 2% of patients receiving panitumumab and in no patients receiving BSC alone. When panitumumab was used in combination with irinotecan, the incidence and severity of chemotherapy-induced diarrhea increased. In a study of 19 patients receiving panitumumab in combination with irinotecan, bolus 5-fluorouracil, and leucovorin, the incidence of NCI Common Toxicity Criteria grade 3 to 4 diarrhea was 58% and was fatal in one patient. In a study of 24 patients receiving panitumumab plus irinotecan, 5-fluorouracil, and leucovorin, the incidence of NCI Common Toxicity Criteria grade 3 diarrhea was 25%. The safety and effectiveness of panitumumab when used in combination with other anticancer agents has not been established.
Electrolyte depletion. In the randomized, controlled clinical trial of panitumumab, median magnesium levels decreased by 0.1 mmol/L in the panitumumab arm; hypomagnesemia (NCI Common Toxicity Criteria grade 3 or 4) requiring oral or i.v. electrolyte repletion occurred in 2% of patients. Hypomagnesemia occurred 6 weeks or longer after the initiation of panitumumab. In some patients, hypomagnesemia was associated with hypocalcemia. Patients' electrolytes should be periodically monitored during and for 8 weeks after the completion of panitumumab therapy.
Pharmacokinetics
The panitumumab concentration-time profile was best described by a two-compartmental pharmacokinetic model with linear and nonlinear clearance pathways, likely to be mediated by the reticuloendothelial system and EGFR, respectively.
Following the recommended dose regimen (6 mg/kg given once every 2 weeks as a 1-h infusion), panitumumab concentrations reached steady-state levels by the third infusion with mean (±SD) peak and trough concentrations of 213 ± 59 and 39 ± 14 µg/mL, respectively. The mean (±SD) area under the concentration-time curve (AUC0-tau) and clearance were 1,306 ± 374 µg·mL/d and 4.9 ± 1.4 mL/kg/d, respectively. The elimination half-life was 7.5 days (range, 3.6-10.9 days). Age (21-88 years), gender, race (15% nonwhite), mild-to-moderate renal dysfunction, and mild-to-moderate hepatic dysfunction had no apparent effect on the pharmacokinetics of panitumumab.
An exposure-response relationship could not be established because of the low overall response rate and the limited pharmacokinetic data. The relationship between panitumumab dose and the incidence of integument/eye toxicities within 28 days of panitumumab treatment was assessed using logistic regression. Panitumumab exposure was correlated with the incidence and duration of integument/eye toxicity but not with the duration of severe integument/eye toxicity. There was no correlation between the duration of integument/eye toxicity and trough concentrations of panitumumab identified.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The recommendation for accelerated approval for panitumumab as a single agent for patients in this subgroup was based on a clinically modest but highly statistically significant prolongation in median PFS shown in a randomized trial and confirmed by an IRC masked to treatment assignment. Consistent with this finding, an improvement in overall response rate was shown among those treated with panitumumab compared with those treated with BSC alone, suggesting the benefit to be real. Response rates of 8% to 13% were observed in four additional phase 2 trials in a similar patient population submitted for review by the sponsor, including two studies with an independent confirmation of response. Response rates of the same magnitude have been observed with cetuximab in this patient population (4, 5).
Approximately 30% of the patients evaluated for entry into the randomized clinical trial were deemed ineligible due to failure to meet criteria for EGFR expression. Because this represents a substantial proportion of the general population of patients with metastatic colorectal cancer in whom the safety and effectiveness has not been studied, the indications and usage and other relevant aspects of the labeling clearly state that panitumumab therapy is intended for use only in patients whose tumors express EGFR.
Drug approval in the United States requires adequate and well-controlled studies showing that a drug is both safe and effective for the indication for which approval is sought (Federal Food, Drug, and Cosmetic Act, Kefauver-Harris Amendment of 1962) and is based on the demonstration of a clinical benefit (e.g., prolongation of survival or improvement in disease-related symptoms) or an effect on an established surrogate for clinical benefit (6, 7). Randomized studies with 5-fluorouracil–based chemotherapy show that improvements in PFS are generally accompanied by improvements in OS (8–10). This observation is the basis for the Food and Drug Administration's Oncology Drug Advisory Committee's recommendation that PFS, although not an established surrogate end point, is a surrogate end point that is likely to predict effects on OS, although it failed to do so for this study.
The reason(s) for the failure to show that an effect on PFS resulted in an effect on OS in this study remains unclear. Panitumumab may, in fact, have no real effect on survival in advanced, treatment-refractory metastatic colorectal carcinoma when compared with BSC alone. A higher percentage of subjects randomized to the BSC-alone arm (59%) than in the panitumumab plus BSC arm (36%) had an early unscheduled tumor assessment done before the first scheduled assessment at 8 weeks. A post hoc sensitivity analysis was done in which the date of progression was moved to the date of the closest postrandomization scheduled assessment. This analysis was compromised by the discordance between locally and centrally assessed determinations of disease progression and had no effect on OS. A third possibility is that the large proportion of patients in the control arm, half of whom initiated panitumumab therapy before the first study assessment, obscured detection of an effect on survival.
The sponsor has committed to conduct a randomized trial of chemotherapy alone versus chemotherapy and Vectibix in second-line treatment of metastatic colorectal cancer.
|
Footnotes |
|---|
Note: The views expressed herein are independent and do not necessarily represent the views of the U.S. Food and Drug Administration.
5 http://www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm?fuseaction=Search.Label_ApprovalHistory#apphist 
6 Panitumumab: pharmacology and toxicology review. 2006. Available from: http://www.fda.gov/cder/foi/nda/2006/125147s0000_PharmR.pdf.
Received 5/31/07; revised 10/ 9/07; accepted 11/ 6/07.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
