- © 2006 by American Society of Clinical Oncology
Phase I/II Study of Cetuximab in Combination With Cisplatin or Carboplatin and Fluorouracil in Patients With Recurrent or Metastatic Squamous Cell Carcinoma of the Head and Neck
- Jean Bourhis,
- Fernando Rivera,
- Ricard Mesia,
- Ahmad Awada,
- Lionel Geoffrois,
- Christian Borel,
- Yves Humblet,
- Antonio Lopez-Pousa,
- Ricardo Hitt,
- M. Eugenia Vega Villegas,
- Lionel Duck,
- Dominique Rosine,
- Nadia Amellal,
- Armin Schueler and
- Andreas Harstrick
- From the Institut Gustave Roussy, Villejuif; Centre Alexis Vautrin, Vandoeuvre les Nancy; Centre Paul Strauss, Strasbourg, France; Hospital Universitario Marqués de Valdecilla, Santander; Institut Català d’Oncologia, Duran I Reynals, L’Hospitalet; Hospital de la Sant Creu i de Sant Pau, Barcelona; Hospital Universitario de Octubre, Madrid, Spain; Institut Jules Bordet; University Hospital Saint-Luc, Brussels, Belgium; and Merck KGaA, Darmstadt, Germany
- Address reprint requests to Jean Bourhis, MD, PhD, Institut Gustave Roussy, Radiothérapie, 39 rue Camille Desmoulins, 94800 Villejuif, France; e-mail: bourhis{at}igr.fr
Abstract
Purpose This was an open, randomized, multicenter, phase I/II study to investigate the safety and tolerability of cetuximab in the first-line treatment of recurrent/metastatic squamous cell carcinoma of the head and neck (SCCHN).
Patients and Methods Treatment comprised cetuximab (initial dose 400 mg/m2 with subsequent weekly doses of 250 mg/m2) in combination with 3-week cycles of either cisplatin (100 mg/m2) or carboplatin (area under the curve, 5), each in combination with a 5-day infusion of fluorouracil (FU) at escalating doses of 600, 800, and 1,000 mg/m2/d. The study was divided into two phases: A, the first two cycles (6 weeks) focusing on the safety and tolerability of combination therapy; and B, the remaining time for those benefiting from therapy until disease progression or intolerable toxicity.
Results Fifty-three patients were enrolled onto the study. The incidence of dose-limiting toxicities in phase A was acceptable. The most common grade 3/4 adverse events in both groups were leucopenia (38%), asthenia (25%), vomiting (14%), and thrombocytopenia (15%), which are consistent with the known safety profiles of cetuximab, cisplatin/carboplatin, and FU. The overall response rate among patients was 36%, with no clear trend toward an increased efficacy at the highest dose of FU, and no impact of the concomitant chemotherapy regimens on cetuximab pharmacokinetics.
Conclusion The combination of cetuximab, cisplatin/carboplatin, and FU was reasonably well tolerated and active in recurrent/metastatic SCCHN, and merits additional investigation. An FU dose of 1,000 mg/m2/d in combination with cisplatin or carboplatin can be recommended for additional studies.
INTRODUCTION
Head and neck carcinomas, the majority of which are of the squamous cell histologic subtype (SCCHN), account for 6% of all cancers worldwide (excluding nonmelanoma skin cancers).1 In Europe alone in 2002, head and neck cancers accounted for an estimated 68,000 deaths.1 The main cause of death for patients with locally advanced head and neck tumors are locoregional recurrences, the majority of which are in the radiotherapy field. Metastatic recurrences are less frequent and account for 15% to 20%. Median survival for patients with recurrent disease is 6 months, independent of the use of palliative chemotherapy for their treatment. Consequently, these patients are suitable candidates for new experimental therapeutics.2
The cornerstone of treatment for patients with recurrent/metastatic SCCHN is systemic chemotherapy. Cisplatin is one of the most active agents for the treatment of this disease.3 With cisplatin-based chemotherapy, response rates of approximately 35% can be expected.4,5 Clearly, this leaves considerable room for improvement in the treatment of recurrent/metastatic disease. Other active chemotherapy agents, such as the taxanes, have shown good activity in SCCHN, but seem to be no more effective than cisplatin-based chemotherapy.6 In addition, taxane-based regimens have shown significant myelotoxicity and neurotoxicity.7,8
In recent years, the use of molecular-targeted therapy has attracted increasing amounts of attention, particularly in the treatment of advanced or recurrent disease. The epidermal growth factor receptor (EGFR) is one of a number of molecules that has been identified as an important target for cancer therapy. It is expressed at high levels in a number of solid tumors, including SCCHN, where it is associated with poor survival.9,10 Cetuximab (Erbitux; Merck KGaA, Darmstadt, Germany) is an immunoglobulin G1 monoclonal antibody that specifically targets the EGFR with high affinity and competitively inhibits endogenous ligand binding.11 Nonclinical studies have shown that it potentiates the effects of radiation in vitro and in vivo12 and several chemotherapeutic agents, including cisplatin.13 Importantly, however, the pharmacokinetics (PK) of cetuximab and chemotherapy do not seem to be adversely affected by coadministration.14,15 Cetuximab has already shown promising results in the treatment of both recurrent/metastatic16-19 and locoregionally advanced SCCHN.20,21
In view of the fact that cisplatin/carboplatin and fluorouracil (FU) combination therapy is the treatment of choice for patients with recurrent/metastatic SCCHN, this phase I/II study sought to investigate the safety and tolerability of adding cetuximab to this standard combination. FU dose escalation was carried out as part of the study to determine the maximum-tolerated dose to use for the combination in future investigations.
PATIENTS AND METHODS
Study Design
This was a multicenter, randomized, open-label, phase I/II study, conducted at nine centers in three European countries (Belgium, France, and Spain), to investigate the safety and efficacy of the combination of cetuximab, platinum (cisplatin or carboplatin), and FU in recurrent/metastatic SCCHN. All patients received an intravenous infusion of cetuximab (400 mg/m2 initial dose followed by subsequent weekly doses of 250 mg/m2). Patients were randomly assigned to receive 3-week cycles of cytotoxic chemotherapy with either cisplatin 100 mg/m2 on the second day of each cycle or carboplatin area under the curve 5 on the first day of each cycle, each in combination with a continuous infusion of FU at escalating doses of 600 (low dose), 800 (medium dose), or 1,000 mg/m2/d (high dose) on days 1 to 5 of each cycle.
The study was divided into two treatment phases. All patients took part in phase A, which focused on the tolerability of treatment during the first two cycles (6 weeks) of therapy. At the end of this phase, patients who benefited from treatment, in terms of tumor regression or stable disease, carried on into phase B of the study. In phase B, patients received combination chemotherapy at the same dose as administered in phase A (or cetuximab monotherapy if patients had benefited from treatment but were unable to tolerate chemotherapy) until disease progression or the development of unacceptable adverse effects. Patients found not to benefit from treatment at the end of phase A left the study. A patient leaving the study for reasons other than toxicity before the end of phase A was replaced by another patient who was to receive the same platinum therapy and FU at the same dose as the patient being replaced. The study was approved by the independent ethics committees in each country. It was carried out in accordance with the Declaration of Helsinki (1996) and all patients provided written informed consent.
FU Dosing Levels
FU dose was escalated separately for the cisplatin and carboplatin arms in two steps for safety reasons. FU dose escalation was dependent on the incidence of patients with dose-limiting toxicities (DLTs) at each dose level. A maximum of 12 patients per dose level were to be treated for each platinum-based chemotherapy arm.
Dose Reductions/Delays
If grade 3 skin reactions occurred, cetuximab treatment could be delayed for up to two consecutive infusions until resolution of the reaction to grade ≤ 2. Cetuximab treatment was discontinued in the event of a delay of more than two consecutive infusions, a fourth occurrence of grade 3 skin reactions, or a first occurrence of grade 4 skin reactions.
Chemotherapy dose reductions/delays were allowed according to predefined toxicity criteria. Chemotherapy administration was not to be delayed due to cetuximab-related skin reactions.
Patient Selection
Patients were entered onto the trial if they fulfilled the eligibility criteria, which included ≥ 18 years of age; recurrent or metastatic SCCHN not suitable for local therapy; disease measurable by computed tomography or magnetic resonance imaging; life expectancy ≥ 3 months at study entry; tumor tissue available for immunohistochemistry staining for EGFR expression; Karnofsky performance status ≥ 70% at study entry; adequate hematologic, hepatic, and renal function; and recovery from relevant toxicities of previous treatment. Exclusion criteria included nasopharyngeal carcinoma; prior systemic chemotherapy, except if as part of multimodality treatment for locally advanced disease (completed > 3 months before inclusion in the study or ≥ 6 months if platinum based); previous treatment with cumulative cisplatin dose more than 300 mg/m2; or known grade 3/4 peripheral neuropathy.
End Points
The primary end point of the study was the safety and tolerability of the combination of cetuximab, platinum, and escalating doses of FU, measured by DLTs and adverse events. Secondary end points were the PK of cetuximab and cisplatin, the best overall response, the time to progression (TTP; defined as the time from random assignment until the first observation of disease progression, or death as a result of any cause within 60 days after last tumor assessment or from random assignment for all randomly assigned patients) and overall survival time (OS; defined as the time from the day of random assignment to death).
Safety and Tolerability Assessments
DLTs were assessed weekly in all patients during phase A, according to the National Cancer Institute Common Toxicity Criteria (NCI-CTC) version 2, and were defined as the occurrence of at least one of the following at least possibly treatment-related events: any more than grade 2 nonhematologic toxicity (excluding clinically nonrelevant adverse events, such as headache and grade 1 to 3 skin reactions); grade ≥ 2 hypercreatininemia; grade 3 neurotoxicity/ototoxicity; grade 4 nausea, vomiting not controlled by antiemetics, and skin reactions; grade 4 thrombocytopenia; grade 4 neutropenia lasting more than 7 days; grade 3/4 neutropenia with complications (eg, fever, infection); and drug-related adverse events requiring treatment discontinuation within the first two cycles of treatment.
In addition, NCI-CTC–graded adverse events were recorded and rated as unrelated, or possibly, probably, or definitely related to treatment. The severity of adverse events not included in the NCI-CTC version 2.0 was assessed by the investigator. Special adverse events, comprising a combination of certain COSTART-preferred terms pooled together, which were considered to be of particular clinical interest, were also presented.
Tumor Assessment
Tumor assessments were made at baseline and after every other cycle (ie, 6 weeks) until treatment completion. Evaluation of lesions was performed by the investigator at each center, based on computed tomography or magnetic resonance imaging scans; the same method was used both at baseline and follow-up for individual patients. Baseline and response evaluations were made according to Response Evaluation Criteria in Solid Tumors Group criteria, with tumor size defined as the sum of the longest diameter of all target lesions. Response based on target (and nontarget lesions) was defined as follows: complete response (CR), disappearance of all target (nontarget) lesions; partial response (PR), ≥ 30% reduction in size (or disappearance of one or more nontarget lesions); stable disease (SD), less than 30% decrease and less than 20% increase in size (or the persistence of one or more nontarget lesions); progressive disease, more than 20% increase in size (or the appearance of new nontarget lesions and/or progression of existing nontarget lesions).
The best overall response was defined as the best response recorded from the start of treatment until disease progression or recurrence, confirmed by repeated assessments performed no less than 4 weeks after the criteria for response were first met. Disease control was defined as CR + PR + SD.
Statistical Analyses
Statistical analyses were performed using the Statistical Analysis System software package, version 8.2 (SAS Institute, Cary, NC). Continuous variables were summarized using descriptive statistics. Qualitative variables were summarized by means of counts and percentages. Kaplan-Meier estimates were used for time-to-event end points.22 Two-sided CIs were calculated for efficacy data, according to Clopper-Pearson23 for response rates and according to Brookmeyer and Crowley for median time-to-event end points.24 For the purpose of clarity, percentages and CIs have been rounded off.
The following analysis populations are described: the DLT population (all patients who completed the first two cycles of treatment [phase A] or who stopped treatment during this time because of DLT); the intention-to-treat (ITT) population (all patients who were randomly assigned); the safety population (all patients receiving any dose of cetuximab); and the PK population (all patients with a complete cetuximab concentration-time profile).
PK parameters of cetuximab were calculated according to noncompartmental standard methods using the PK software program KINETICA, version 4.1.1 (InnaPhase Corp, Philadelphia, PA).
RESULTS
Fifty-three patients were enrolled onto the study: 27 were randomly assigned to cisplatin and 26 were randomly assigned to carboplatin (Fig 1). Eleven patients (41%) in the cisplatin group and four (15%) in the carboplatin group discontinued during phase A (including one in the carboplatin arm who was randomly assigned but who never received treatment and who was therefore included in the ITT population but excluded from the safety population), mainly due to adverse events (n = 11). Of the patients discontinuing for reasons other than toxicity, three patients were replaced in the cisplatin plus FU medium-dose group and one in the carboplatin plus FU high-dose group. An additional patient in the carboplatin plus FU medium-dose group was replaced, but this patient did not withdraw from the study. The remaining patients entered phase B (n = 16 for cisplatin and n = 22 for carboplatin); 19 patients (36%; 11 for cisplatin, eight for carboplatin) who discontinued chemotherapy due to unacceptable adverse events went on to receive cetuximab as monotherapy. Among the ITT population, there were no clinically relevant differences in the demographic and disease characteristics between the cisplatin and carboplatin treatment groups (Table 1).
Flow diagram showing the number of patients in each stage of the study (intention-to-treat population). FU, fluorouracil; LD, low dose; MD, medium dose; HD, high dose.
Baseline Demographic and Disease Characteristics: ITT Population*
DLTs
The frequency of DLTs experienced during phase A is summarized in Table 2. At the highest dose level of FU, four of 12 patients (33%) in the cisplatin group and two of 12 (17%) in the carboplatin group experienced a DLT, both of which are considered to be within the acceptable tolerability limit.
DLTs in Phase A (first two cycles): DLT Population
Adverse Events
Each patient experienced at least one adverse event. The most common adverse events (ie, those seen in ≥ 30% of patients) in the cisplatin group were nausea (78%), asthenia and vomiting (both 74%), leucopenia (63%), acne-like rash (56%), and fever and diarrhea (both 41%). The most common adverse events in the carboplatin group were asthenia (72%), acne-like rash (56%), nausea and stomatitis (both 52%), thrombocytopenia (48%), mucositis (44%), rash (40%), and leucopenia and constipation (both 32%). The only adverse event seen in ≥ 30% of those patients discontinuing chemotherapy but continuing to receive cetuximab monotherapy was asthenia (10 patients experienced events; 53%), which may be explained by the more advanced stage of the underlying disease.
In the special adverse events category, nausea/vomiting, asthenia, and respiratory symptoms were the only grade 3/4 adverse events occurring in more than 15% of patients (Table 3). Despite the relatively high incidence of respiratory adverse events, only two (both in the cisplatin group) were considered possibly related to cetuximab (one grade 2 dyspnea and one grade 3 pleural effusion). As would be expected from the toxicity profiles of the two agents, there was a substantially higher incidence of serious adverse events in the cisplatin group (67%) than in the carboplatin group (32%).
Frequencies of Special AE Categories During Combination Therapy: Safety Population
Altogether, 15 patients (29%) discontinued cetuximab therapy during the combination phase due to treatment-related adverse events (only two of which were considered to be related to cetuximab only [hypersensitivity reactions]) and one patient discontinued during cetuximab monotherapy. Chemotherapy was discontinued due to adverse events in 14 (52%) and seven (28%) patients in the cisplatin and carboplatin groups, respectively.
There was one death as a result of septic shock that was considered to be related to cytotoxic chemotherapy within the cetuximab/cisplatin/FU low-dose group.
Efficacy
In the ITT population, for both treatment arms, 19 patients achieved a response (two CRs and 17 PRs), giving a best overall response rate of 36% (95% CI, 23% to 50%; Table 4). An additional 20 patients had SD, resulting in a disease control rate of 74% (95% CI, 60% to 85%). The best overall response rate in the cisplatin group was 33% and the disease control rate was 67%. For the carboplatin group, corresponding figures were 39% and 81%.
Response Rate: Overall and According to Platinum Group and FU Dose Level: ITT Population*
The median duration of response in the ITT population was 127 days (95% CI, 103 to 300 days); 122 days (95% CI, 103 to 143 days) and 202 days (95% CI, 92 to 300 days) in the cisplatin and carboplatin groups, respectively.
In the ITT population, the median TTP was 155 days (95% CI, 127 to 186 days) for the two groups, 183 days (95% CI, 86 to 189 days) in the cisplatin group, and 151 days (95% CI, 127 to 189 days) in the carboplatin group. In the same population, the median OS was 297 days (95% CI, 242 to 418 days) for the two groups, 324 days (95% CI, 229 to 418 days) in the cisplatin group and 260 days (95% CI, 238 days to not estimable) in the carboplatin group (Fig 2).
Overall survival.
PK Data
Sufficient cetuximab PK data were available for 39 patients (Fig 3). At 72 hours after cetuximab administration, the mean serum cetuximab concentration varied between 95.1 and 114.4 μg/mL across FU doses and the cisplatin/carboplatin groups. At 168 hours, corresponding values were 54.1 and 62.7 μg/mL. The mean values for all PK parameters investigated (area under the serum concentration-time curve during one dosing interval, mean residence time, half-life, clearance, and volume of distribution at steady-state) were similar across the three FU dose levels as well as between the two platinum arms, indicating that the different FU/platinum regimens had no relevant effect on the pharmacokinetics of cetuximab. Data for cisplatin pharmacokinetics are not available.
Plot of the mean serum cetuximab concentrations of cisplatin and carboplatin groups (fluorouracil dose levels combined), week 4.
DISCUSSION
The results from this phase I/II study showed that cetuximab in combination with cisplatin/carboplatin and FU was fairly well tolerated and showed encouraging activity in the first-line treatment of patients with recurrent and/or metastatic SCCHN.
The adverse event profile was in line with what was expected due to the underlying disease and the known safety profile of cetuximab, cisplatin/carboplatin, and FU, and there was no evidence that cetuximab aggravated the known toxicities of cisplatin, carboplatin, or FU. There was no indication of any FU dose-dependent trends in adverse events. There was only one death as a result of septic shock that was considered to be related to cytotoxic chemotherapy. The majority of patients were able to receive the full dose of cetuximab, further demonstrating that treatment with cetuximab in combination with cisplatin or carboplatin was well tolerated. As expected, there was a lower incidence of grade 3/4 adverse events in the carboplatin arm compared with the cisplatin arm, although this was not subject to statistical analysis. This aspect of treatment tolerability needs to be considered alongside the relative efficacy of the investigative regimens once this has been determined in a larger randomized trial. In addition, in keeping with earlier clinical studies,14 there was no evidence that concomitant chemotherapy administration had any impact on cetuximab PK.
Efficacy was a secondary end point in this study. A best overall response rate of 36% and a disease control rate of 74% were observed for the ITT population. One patient in each of the cisplatin and carboplatin FU high-dose arms achieved a CR. The median TTP of 155 days (5.1 months) and the median OS of 297 days (9.8 months) were encouraging for this group of patients. Although the sample size is too small to allow definite conclusions to be drawn, there seemed to be no clinical difference in the activity of the cisplatin and carboplatin arms, and no trend toward a higher efficacy at higher doses of FU.
The efficacy of cetuximab as part of first-line therapy for recurrent/metastatic SCCHN was demonstrated by Burtness et al16 in a randomized phase III study in which cetuximab was combined with cisplatin 100 mg/m2 for 28 days. In this trial of 117 assessable patients, the combination of cetuximab and cisplatin gave a significantly higher response rate than placebo and cisplatin (26% v 10%; P = .03). Differences in progression-free survival (4.2 v 3.1 months) and OS (9.2 v 7.9 months) between the arms were not statistically significant, possibly because the study was not powered to show this. This study clearly showed that cetuximab increased the antitumor activity of cytotoxic chemotherapy. However, a drawback was that the study used cisplatin alone rather than what is now a currently recognized standard of treatment for recurrent/metastatic disease, namely the combination of cisplatin (or carboplatin) and FU. Within the limits of the design and the small patient numbers, the findings from our study suggest that the triple combination regimen (cetuximab/cisplatin [carboplatin]/FU) may provide a useful therapeutic approach. A randomized trial investigating this triple combination is now warranted.
The findings from this study can be added to the growing bank of evidence for the usefulness of cetuximab in the treatment of SCCHN. In addition to its use in the first-line setting in recurrent/metastatic disease, cetuximab has also shown activity in SCCHN, for which platinum-based therapy has failed in combination with cisplatin or carboplatin in SCCHN18,17 and in recurrent/metastatic nasopharyngeal carcinoma.25 In addition, single-agent activity has been reported in platinum-resistant SCCHN, with a response rate of 12%.19
In conclusion, we found that the combinations of cetuximab, cisplatin/carboplatin, and FU were reasonably well tolerated and active in the first-line treatment of recurrent/metastatic SCCHN and merit additional investigation. In view of our findings, the recommended dose level of FU in combination with cisplatin/carboplatin for use in future studies can be 1,000 mg/m2/d during 5 days every 3 weeks.
Authors’ Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Author Contributions
Manuscript writing: Jean Bourhis, Fernando Rivera, Ricard Mesia, Ahmad Awada, Lionel Geoffrois, Christian Borel, Yves Humblet, Antonio Lopez-Pousa, Ricardo Hitt, Eugenia Vega Villegas, Lionel Duck, Nadia Amellal, Armin Schueler, Andreas Harstrick
Footnotes
-
Presented as preliminary results at the 40th Annual Meeting of the American Society of Clinical Oncology, New Orleans, LA, June 5-8, 2004.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
- Received October 24, 2005.
- Accepted March 28, 2006.
