Pharmacokinetics of Cetuximab After Administration of Escalating Single Dosing and Weekly Fixed Dosing in Patients with Solid Tumors

Cetuximab (C225, Erbitux, ImClone Systems, Inc., New York, NY and Bristol-Myers Squibb, Princeton, NJ) is a chimeric monoclonal antibody directed against the extracellular domain of the human epidermal growth factor receptor (EGFR; ref. 1). The binding of natural ligands (EGF and transforming growth factor-α) to EGFR results in receptor dimerization, activation of the intracellular tyrosine kinase domain of the receptor and downstream signal transduction cascades, such as the mitogen-activated protein kinase pathway, and stimulation of cellular proliferation. Preclinical studies show that cetuximab blocks the binding of ligands to the EGFR, resulting in competitive inhibition and down-regulation of receptors on the cell surfaces, which consequently leads to inhibition of EGFR activation (2–4).

Several clinical studies have indicated that cetuximab is active in the treatment of epithelial malignancies, including metastatic colon cancer and head and neck squamous cell carcinoma (5–12). Cetuximab in combination with irinotecan is currently approved for the treatment of EGFR-expressing metastatic colorectal cancer that is refractory to irinotecan-based chemotherapy and also as monotherapy in patients unable to tolerate irinotecan-based chemotherapy (6). Cetuximab is also approved in combination with radiation for locally advanced head and neck squamous cell carcinoma and as monotherapy in patients who have progressed on platinum-based therapy (8, 12).

Initial phase I trials of cetuximab focused on identifying doses associated with saturation of clearance, which would presumably reflect complete occupancy of EGFRs (13, 14). These studies show that clearance of cetuximab is eliminated rather slowly from the systemic circulation, with a median half-life (t_1/2) of 7 days for both 200 and 400 mg/m² dose levels (13). Clearance of cetuximab was found to decrease with increasing dose, with an apparent saturation of clearance within the 200 to 400 mg/m² dose range (13, 14). However, Robert et al. (14) noted that, with weekly 250 mg/m² dosing, however, trough levels in some patients decreased below those estimated to yield saturation of clearance. These estimates in cetuximab clearance after a single dose were based on pharmacokinetic sampling protocols that included few data points after 96 hours (from infusion initiation). Because of the prolonged t_1/2 of cetuximab compared with the limited duration of pharmacokinetic sampling, these estimates of cetuximab pharmacokinetic variables may not be accurate.

In a phase IB study of cetuximab and cisplatin in patients with recurrent head and neck squamous cell carcinoma, tumor tissues without concurrent pharmacokinetics were obtained from 12 patients at baseline, 24 hours after the first infusion, and 24 hours before the third infusion of cetuximab (15). Using immunohistochemical analysis with a murine EGFR antibody, as well as an EGFR autophosphorylation assay, significant or even complete occupancy of EGFRs was observed for patients receiving weekly doses of 250 mg/m² when preceded by loading doses of 400 or 500 mg/m². Based on this small data set and the earlier phase I studies, the schedule of a 400 mg/m² loading dose followed by a weekly maintenance dose of 250 mg/m² was adopted for subsequent phase II and III trials with cetuximab.

The primary objective of this study was to characterize the pharmacokinetics of a single dose of cetuximab over a 10-fold range (50, 100, 250, 400, or 500 mg/m²) using a prolonged serum sampling protocol. Additional objectives included assessment of cetuximab trough levels obtained with repeated weekly doses of 250 mg/m², evaluation of toxicity and antitumor responses, and monitoring formation of human antichimeric antibodies (HACA).

Eligibility criteria. Patients were eligible if they had a histologically confirmed diagnosis of an advanced solid tumor of epithelial origin, had measurable or evaluable disease, had received at least one prior chemotherapy regimen for metastatic disease, were ≥18 years of age, and had an Eastern Cooperative Oncology Group performance status of ≤2. Requirements for adequate organ function included an absolute neutrophil count ≥1,500/μL, platelets ≥100,000 μL, hemoglobin ≥9.0 g/dL, creatinine ≤1.5 times upper limit of normal, total bilirubin ≤1.5 times upper limit of normal, aspartate aminotransferase and alanine aminotransferase ≤2.5 times upper limit of normal, and an ejection fraction within normal limits. Patients were excluded if they had prior EGFR-directed therapy, had history of monoclonal antibody therapy, had prior surgery, chemotherapy, or radiotherapy within 4 weeks of enrollment (6 weeks for nitrosourea, mitomycin, and liposomal doxorubicin), or had significant comorbid conditions. The protocol was approved by the Institutional Review Board at each participating institution. All patients gave written informed consent before treatment.

Treatment plan and toxicity evaluation. This was a phase I, multi-institutional, open-label study of cetuximab administered as a single dose followed by a weekly fixed dose. The study had two phases of treatment. For the initial phase of the study, patients were randomized to 50, 100, 250, 400, or 500 mg/m² of cetuximab as a 2-hour i.v. infusion for the purpose of single-dose pharmacokinetic evaluation. Eight patients were enrolled at each dose level. The second phase consisted of continuous weekly administrations of cetuximab as a 1-hour infusion at a fixed dose of 250 mg/m² starting 3 weeks later (on day 22). All patients were premedicated with 50 mg i.v. diphenhydramine. Cetuximab was supplied by ImClone Systems as an injectable solution in 50 mL vials containing 2 mg/mL cetuximab.

Toxicities were evaluated and graded using National Cancer Institute Common Toxicity Criteria version 2.0. Dose reductions due to unacceptable toxicity were allowed starting with the second weekly dose of cetuximab and defined as a weekly dose of 200 mg/m² for the first dose reduction and 150 mg/m² for the second dose reduction. Treatment modifications occurred as follows: for grade 3 rash, cetuximab could be omitted for up to two consecutive infusions with no dose reduction and resumed when the rash improved to grade ≤2. For grade 3 nonhematologic toxicities, except rash and allergic reaction/hypersensitivity, treatment was withheld until the toxicity resolved to grade ≤1 and then reinstituted at the next lower dose level. For grade 4 nonhematologic toxicity, including rash, cetuximab was discontinued.

Clinical evaluation. At baseline, a history, physical examination, electrocardiogram, multiple-gated acquisition scan, and laboratory tests (including a complete blood count with differential, chemistry, and urinalysis) were obtained. The latter were repeated every 3 weeks. Imaging of involved cancer sites was done within 4 weeks of enrollment and every 6 weeks. Treatment continued until disease progression or unacceptable toxicity. Modified WHO criteria were used to assess response (16).

Pharmacokinetic sampling and analysis. In the single-dose part of the study, serum samples for pharmacokinetics were collected before dose and at 1 hour, 1 hour and 58 minutes (2 minutes before the end of infusion), 2 hours and 30 minutes, 3 hours, 4 hours, 6 hours, 8 hours, 24 hours, 48 hours, 96 hours, 168 hours, 264 hours, 336 hours, 432 hours, and 504 hours after administration. In the multiple-dose part of the study (in which all patients received 250 mg/m² cetuximab weekly), serum samples for trough concentrations of cetuximab were obtained before each fixed weekly dose (up to a maximum of 10 doses).

Serum concentrations of cetuximab were measured using a validated ELISA. Briefly, EGFR-coated polystyrene microtiter plates were loaded with serum samples, washed, and incubated with a horseradish peroxidase–conjugated rabbit anti-human IgG. After elution of unbound conjugate, chromogenic substrate was added, and the reaction product was measured using a microtiter plate reader at 450 nm. Calibration curves were linear in the concentration range of 0.475 to 14.25 μg/mL. For the determination of the interday and intraday precision and accuracy of the assay, human serum quality control samples were used at a low (1.425 μg/mL), an intermediate (5.7 μg/mL), and a high concentration (11.4 μg/mL) of cetuximab. Intraday precision of the means of the quality control samples ranged from 2.8% to 5.6%, whereas the interday precision was in the range of 2.1% to 5.6%. Intraday accuracy of the means of the quality control samples ranged from 91.3% to 93.8%, whereas the interday accuracy was in the range of 93.2% to 100.4%.

Pharmacokinetic variables were calculated by a noncompartmental approach using Kinetica version 2.2 (InnaPhase Corp., Philadelphia, PA). Michaelis-Menten kinetic variables (V_m and K_m) for the three higher dose levels were analyzed using a time-concentration model according to the method of Gibaldi and Perrier (17). The nlme nonlinear random effects package for the R statistical system was used to obtain the variable estimates (18).

Immunogenicity studies. A validated double-antigen radiometric assay was used to detect the formation of HACA. Cetuximab was immobilized onto polystyrene beads, which were then incubated with serum samples. ¹²⁵I-labeled cetuximab was added to the samples, and the radioactivity derived from the bound ¹²⁵I-labeled cetuximab was used to calculate the results. The lower limit of quantitation was 5.5 mg/mL (19, 20). Serum samples were obtained before the first, second, and fourth dose of cetuximab and then every 6 weeks (before infusion).

Statistical analysis. The correlation between the clearance of cetuximab and dose, age, sex, weight, and body surface area (BSA) was analyzed using linear regression. The relationship between cetuximab dose, systemic exposure, C_max, and clearance and rash of any grade or diarrhea of any grade, which occurred in the first 22 days, was evaluated by logistic regression. The occurrence of grade 1, 2, and 3 rash at any time in relation to clinical outcome was analyzed by Fisher's exact test. The relationship of trough values and clinical outcome was assessed using a random effects model, and the correlation of the trough value on day 85 with rash was analyzed using a two-sample t test.

General. Between October 2002 and March 2003, 40 patients were randomized to receive a single dose of cetuximab at one of five levels. Patient characteristics are shown in Table 1. Five patients did not enter the subsequent fixed dose phase of the study due to disease progression (3), clinical deterioration (1), and a grade 3 hypersensitivity reaction (1). One patient required a dose reduction to 200 mg/m² after the first fixed dose. Study therapy was interrupted in eight patients due to toxicity during the fixed dose phase. A median of four infusions and a median cumulative dose of 1,375 mg/m² were given during the fixed dose phase. The majority of patients discontinued the study due to disease progression. Other reasons for discontinuation included death (3), toxicity (2), and clinical deterioration without documented progression (1).

Toxicities. Toxicities are summarized in Table 2. Most toxicities were grade 1 or 2. Three patients experienced grade ≥3 cetuximab-related toxicity. A patient who received a single dose of cetuximab at 400 mg/m² experienced a grade 3 hypersensitivity reaction on day 1, requiring discontinuation of therapy. A patient who was treated with a 250 mg/m² initial dose also had a grade 3 hypersensitivity reaction on day 1. Another patient developed grade 2 painful bilateral periungual lesions on the fingers that started on day 85 after receiving six doses of cetuximab and withdrew from further treatment. Concomitant grade 3 rash and grade 3 hyperkinesia occurred in a patient on day 147.

Pharmacokinetics. Plasma samples for single-dose pharmacokinetic analyses were obtained from 39 patients. In one patient, samples were not available as a result of study discontinuation after a grade 3 hypersensitivity reaction. Following infusion of a single dose, cetuximab serum concentrations reached a maximum at ∼3 hours and then declined slowly. The last measurable concentrations in the 50 and 100 mg/m² dose groups were at approximately 96 and 168 hours, respectively. Individual serum concentrations of cetuximab were quantifiable up to 504 hours in the other dose groups. After a single dose, mean maximum cetuximab concentrations and mean area under the plasma concentration-time curve (AUC_0→∞) increased in a greater than dose-proportional manner up to 500 mg/m² (Figs. 1 and 2; Table 3). Mean estimates of the terminal elimination t_1/2 also increased with dose, ranging from 27.6 hours in the 50 mg/m² dose group to 132 hours in the 500 mg/m² dose group (Table 3).

Similarly, mean estimates of total body clearance decreased 4-fold over the 10-fold dose range (Table 3). Mean clearance was similar in the 400 and 500 mg/m² dose groups (21.5 and 20 mL/h/m², respectively). Regression analysis of the relationship between clearance and dose predicted that clearance would be reduced by 90% at a dose of 260 mg/m² (Fig. 3). With regard to volume of distribution, individual V_ss values in all dose groups ranged from 2.4 to 3.1 L/m², indicating minimal distribution of cetuximab into the extracellular space (Table 3).

Individual drug concentration-time profiles for each dose cohort were modeled using a random effects approach and were fit to first-order elimination and Michaelis-Menten models. Data from the 50 and 100 mg/m² dose groups fit a simple linear elimination model, whereas for the 250, 400, and 500 mg/m² dose groups there was an initial rapid decrease in serum concentrations over the initial 48 hours followed by a slower elimination phase that fit Michaelis-Menten kinetics. K_m (concentration of cetuximab at which its elimination rate is half of the maximum rate) values for the 250, 400, and 500 mg/m² doses were calculated as 13,000, 16,000, and 23,000 ng/mL, respectively.

Twenty-four patients were evaluable for assessment of trough cetuximab concentrations obtained before weekly doses of 250 mg/m². These results indicate that there was significant interpatient variability in cetuximab trough values (mean, 54,800; range, 4,300-157,000 ng/mL; coefficient of variation, 72%). Analysis of individual weekly trough values indicated that these increased slightly over time in most patients, consistent with accumulation of cetuximab plasma concentrations with a weekly dose of 250 mg/m² (Fig. 4). For eight patients who had trough values obtained on both day 29 and day 50, the average accumulation ratio on day 50 (defined as average of trough on day 50 divided by average trough concentration on day 29) was 1.88.

Univariate correlative analyses of single-dose pharmacokinetic variables indicated that clearance was associated with BSA (P = 0.002), weight (P = 0.002), and dose (P < 0.0001) but not with age or sex. The association between clearance and BSA remained statistically significant after adjusting for dose, age, and sex (P = 0.006) but not after adjusting for weight. The systemic exposure of cetuximab (AUC) after a single dose was not associated with BSA in either a univariate analysis (P = 0.76) or after adjusting for dose, age, and sex (P = 0.96).

The occurrence of rash or diarrhea of any grade in the first 22 days did not correlate with the first cetuximab dose, AUC, C_max, or clearance. Similarly, there was no association between rash and trough values of cetuximab on day 85 (n = 8). In addition, cetuximab pharmacokinetic variables did not correlate with the occurrence of stable disease. However, there was a strong association between the occurrence of rash (grade 1, 2, and 3) and disease stabilization (P = 0.002).

Immunogenicity. Samples to determine the presence of HACA were obtained in 39 patients. One patient developed a low level of HACA before the second dose. Anti-cetuximab antibodies were also detected in a second patient, but these were present before the first dose and the level of response was considered low. These results indicate that development of anti-cetuximab antibodies is unlikely to occur with weekly 250 mg/m² dosing.

Response data. Among the 39 assessable patients, there were no complete or partial responses. However, 15 patients experienced stable disease (7 colon, 3 non–small cell lung cancer, 2 breast cancer, 1 anal cancer, 1 cholangiocarcinoma, and 1 unknown). The median duration of stable disease was 12.7 weeks (range, 10.1-35.7). Fourteen of the 15 patients who experienced disease stabilization reported a rash (grade 1, 2, or 3) sometime during their treatment course. Among the 24 patients who did not experience stable disease, 9 patients had a rash.

Our study was done to fully characterize the pharmacokinetic profiles of single doses of cetuximab over a 10-fold range. The duration of pharmacokinetic sampling in our study was considerably longer than in earlier dose-finding studies of cetuximab as a single agent or in combination with cisplatin or radiation (13, 14). Interestingly, short sampling periods, in which the ratio of the sampling duration to the calculated t_1/2 was small, have been done in phase I studies with other antibodies, such as matuzumab (EMD72000), a humanized IgG1 anti-EGFR monoclonal antibody, and trastuzumab, the humanized IgG1 monoclonal antibody that targets human EGFR2, leading to underestimation of the t_1/2 of the respective antibody (21, 22).

Our analysis of drug clearance as a function of dose indicated that 90% saturation of clearance is predicted to occur at 260 mg/m². This finding supports the currently recommended weekly dose of 250 mg/m² and is comparable with a previous estimate of saturation of clearance at a dose range of 200 to 400 mg/m² (13). The mean trough value of 54,800 ng/mL (∼360 nmol/L) observed in our study during 250 mg/m² repetitive weekly dosing is similar to values of 200 and 202 nmol/L observed during weekly dosing of 200 mg/m² previously reported by Baselga et al. (13) and Robert et al. (14), respectively. The mean trough value observed in our study is well above the K_m concentration of 13,000 ng/mL predicted for the 250 mg/m² dose by Michaelis-Menten modeling of single-dose cetuximab serum concentrations. This result suggests that, for many patients, repeated weekly dosing with 250 mg/m² cetuximab will produce serum concentrations that are above those required for at least 50% occupancy of EGFRs. Nevertheless, it should be noted that, for two of our patients, trough levels were consistently near or below this predicted K_m concentration, indicating that a 250 mg/m² dose may be insufficient for receptor saturation in some patients (Fig. 4).

Our results also indicated that weekly dosing at 250 mg/m² may lead to drug accumulation. In some patients, we observed a 4-fold increase in trough concentrations during weekly infusions (Fig. 4). These results suggest that biweekly administration of cetuximab, as well as elimination of the loading dose, could be considered. Indeed, biweekly dosing without the standard loading dose is being evaluated (23). Panitumumab, a fully human IgG2 anti-EGFR monoclonal antibody, is currently under evaluation in clinical trials on an every 2-week administration schedule, as serum concentrations and exposures were similar between weekly and biweekly schedules (24).

Correlative pharmacokinetic studies indicated that BSA and weight were significantly associated with clearance, although one was not more highly correlated than the other, supporting the use of either BSA or weight in calculating individual cetuximab doses. This is a finding that has not been reported in previous phase I studies with cetuximab. BSA dosing is not used with panitumumab (weight based) or matuzumab (fixed dosing). In contrast, the lack of correlation between pharmacokinetic variables of cetuximab and either rash or response does not support the use of serum drug monitoring for individualization of treatment.

Pathways of metabolism for anticancer antibodies have been described (25). EGFR antibodies are subject to a saturable antigen-specific elimination process based on internalization of the ligand-receptor complex and removal from circulation. At a certain serum concentration, EGFRs will be saturated. Phase I studies of cetuximab show nonlinear pharmacokinetics and clearance decreases with increasing doses, implying saturation of the elimination pathways (13, 14). Dosing of the antibodies based on the hypothesis of progressive saturation of EGFRs or antigen (EGFR) sink has been applied to other antibodies, including panitumumab. Similar to cetuximab, panitumumab exhibits nonlinear pharmacokinetics (26). Clearance of panitumumab is saturated at 2.5 mg/kg administered once weekly, and 100% occurrence of skin rash is achieved on this schedule, reflecting saturation of EGFRs.

Cetuximab has the potential to induce an immune response. Several studies conducted have obtained sampling for HACA. Pooled data indicate that very few patients (4%) have a positive antibody response to cetuximab (27). Our data were consistent with previous reports of low detection of HACA.

The most common toxicity that occurs with cetuximab treatment is a follicular rash that occurs in an acneiform distribution (face, scalp, chest, and upper back). Several studies suggest that rash correlates with clinical response in patients treated with cetuximab and other EGFR-targeted agents, including the small molecule tyrosine kinase inhibitors, such as erlotinib and gefitinib (6, 28–30). Consistent with these previous reports, we observed a strong association between rash and stabilization of tumor growth. The occurrence of rash has also been examined in relation to survival in four phase II studies in which cetuximab was used alone or with chemotherapy in patients with metastatic colorectal cancer, head and neck squamous cell carcinoma, and pancreatic cancer. The results of this analysis showed that patients with rash had significantly longer survival than patients without rash (31). The correlation between rash and clinical response needs validation in larger prospective trials, and “dose-to-rash” studies are currently being conducted. In a phase I/II trial (EVEREST study), patients with metastatic colorectal cancer pretreated with irinotecan who have grade ≤1 rash are randomized to the current Food and Drug Administration–approved dosing of cetuximab plus irinotecan versus escalating doses of cetuximab (up to 500 mg/m²/wk) and irinotecan until grade ≥2 rash in an effort to achieve higher response rates (32).

The clinical relevance of the cetuximab dose that results in full saturation of elimination pathways (which is hypothesized to best approximate the dose that leads to tumor receptor saturation and presumably lead to optimal clinical response) is not clearly established. The use of rash for titrating the dose of cetuximab is currently being explored, and the correlation between skin toxicity and clinical outcome needs further investigation. With the use of additional tests, such as pharmacogenomics and proteomics, we hope to identify predictors of response to treatment with cetuximab and other EGFR-targeted agents.

We thank the research nurses and pharmacy at the Cancer Institute of New Jersey (New Brunswick, NJ), Elizabeth Arlt for data management, and Novlette Simmons for pharmacokinetic collections.