Prospective Cost-effectiveness Analysis of Cetuximab in Metastatic Colorectal Cancer: Evaluation of National Cancer Institute of Canada Clinical Trials Group CO.17 Trial

Evaluation of National Cancer Institute of Canada Clinical Trials Group CO.17 Trial

Nicole Mittmann; Heather-Jane Au; Dongsheng Tu; Christopher J. O'Callaghan; Pierre K. Isogai; Christos S. Karapetis; John R. Zalcberg; William K. Evans; Malcolm J. Moore; Jehan Siddiqui; Brian Findlay; Bruce Colwell; John Simes; Peter Gibbs; Matthew Links; Niall C. Tebbutt; Derek J. Jonker


J Natl Cancer Inst. 2009;101(17):1182-1192. 

In This Article


Study Parameters

CO.17 was conducted by the NCIC CTG in collaboration with the Australasian Gastro-Intestinal Trials Group (AGITG) (December 2003 to August 2005). Funding for the trial was provided by the NCIC CTG, the AGITG, Bristol-Myers Squibb, and ImClone Systems Incorporated. The trial database was maintained by NCIC CTG. Approval for the conduct of the trial, including the collection of resource utilization data and health preference utilities, was granted by the research ethics board of each participating institution. All patients provided written informed consent. The results of all study patients (Canadian and Australian) were used in this analysis.

A total of 572 patients with chemorefractory colorectal cancer were randomly assigned to receive cetuximab (an initial intravenous dose of 400 mg/m2 of body surface area, followed by a weekly infusion of 250 mg/m2) plus best supportive care (N = 287) or best supportive care alone (N = 285). Treatment was continued until the occurrence of unacceptable adverse events, tumor progression, worsening symptoms of the cancer, patient request (with or without the withdrawal of consent for continued follow-up), or death. The median duration of cetuximab treatment was 8.1 weeks (range = 1–60 weeks). The primary endpoint was overall survival and the secondary endpoint was progression-free survival. The median age of patients in the cetuximab group was 63.0 years (range = 28.6–88.1 years) and in the best supportive care group, 63.6 years (range = 28.7–85.9 years). A total of 368 patients (64%) were male. The majority of patients had an Eastern Cooperative Oncology Group (ECOG) performance status of 1. Nearly 60% of the patients had been diagnosed with colon cancer only. All patients who underwent random assignment were included in the efficacy analyses.

Economic Parameters

This economic evaluation was conducted from the perspective of the payer (ie, the Canadian government) as required by decision makers for cost-effectiveness analyses. Cetuximab was compared with best supportive care because, at the time of the trial, there were no other standard chemotherapy options available for these patients. The choice of best supportive care is consistent with economic evaluation guidelines that recommend using the standard intervention as the comparator agent.[18]

Determination of Direct Medical Costs

Costs are presented in 2007 Canadian dollars (Can $1 = US $1). Costs not available in 2007 Canadian dollars were adjusted for inflation by using the consumer price index (

Cost of Drugs. The cost of cetuximab in 2005 ranged from $2.94 to $6.73/mg in countries that were reviewed by the Patented Medicine Prices Review Board.[19] The base case cost used for this study was $3.24/mg and represented the median cost of the drug among these countries.

Cost of Outpatient Visits. The number of outpatient clinic visits was recorded. The cost of an outpatient physician assessment was based on the Ontario Health Insurance Plan (OHIP) fee schedule[20] and recorded according to the specific type of assessment performed (consultation, general assessment, or partial assessment). The cost of an outpatient chemotherapy visit was determined from the amount of time spent by a nurse for patient education and counseling and infusion chemotherapy administration time and the average 2007 salary levels for nursing and supportive care staff at one study site (Sunnybrook Health Sciences Centre, Ontario, Canada).[21]

Cost of Hospitalization and Surgical Procedures. The location(s) where care was provided (general hospital ward, oncology ward, intensive care unit, and/or rehabilitation facility) was recorded. A per diem cost for the appropriate care unit was obtained from one study site (Sunnybrook Health Sciences Centre) and multiplied by length of stay for each patient.[21]

Cost of Adverse Events. Only grades 3 and 4 adverse events were included because less severe adverse events would not be associated with substantial use of health-care resources. Adverse events included in this analysis were rash, non-neutropenic infections, and pain. These events differed statistically significantly between treatment groups. Confusion was not included even though it differed statistically significantly between treatment groups because resources for its management were unclear. Treatment protocols for each adverse event were standardized based on expert opinion. Hospital costs, based on the Ontario Case Costing Initiative,[22] were applied to the corresponding adverse event International Classification of Diseases, Tenth Revision diagnostic code.[23]

Cost of Laboratory Tests and Diagnostic Procedures. The number and type of laboratory and radiologic tests and procedures conducted during the trial were extracted from the trial database. Laboratory costs and diagnostic procedure costs were based on the 2007 OHIP fee schedule.[20]

Other Costs. All emergency department visits and blood transfusions during the trial were recorded. Institutional costs were applied to emergency department visits and blood transfusion costs (calculated as the number of units of blood product transfused multiplied by the per-unit cost of the blood product).[24] Nonmedical costs (both direct and indirect) were not included in this analysis.

Cost of Radiation Treatment. The use of palliative radiotherapy did not differ between the two treatment arms and therefore its cost was not included in this analysis.

Cost of KRAS Mutation Screening. All patients with available paraffin-embedded tumor tissue samples were screened for the KRAS mutation status of their tumor. Thus, the cost of KRAS mutation screening did not differ between the treatment arms and was not included in this analysis.

Determination of Outcomes

The overall direct medical costs for each patient in each study arm were estimated as the sum of the medical costs listed above. The treatment benefit was defined as the mean survival gain after random assignment. We defined treatment benefit in terms of mean survival rather than in terms of median survival because it allowed us to perform the required arithmetic manipulation of incremental survival duration to derive the incremental cost-effectiveness ratio. We used two methods to calculate overall survival: the restricted mean survival method, which restricts calculation of mean survival to the longest observed survival time (ie, the horizon of the trial), and the Kaplan–Meier method, which takes into account the survival times of censored patients (including those still alive at the end of the study).[25]

Determination of QALYs

The self-reported Health Utility Index Mark 3 (HUI3) was prospectively collected to assess preference-based measures of health status throughout the study (ie, at random assignment [baseline] and at 4 weeks, 8 weeks, 16 weeks, and 24 weeks after random assignment).[26] Questions in the HUI3 cover the following eight health attributes: vision, hearing, speech, ambulation, dexterity, emotion, cognition, and pain. At each assessment time, patients rated their health attributes. An overall utility score was calculated based on the answers to these questions and weighted using the HUI formula according to values derived from the general Canadian population. Utility values in the HUI3 range from 0 (death) to 1 (perfect health). The number of QALYs for each patient was calculated based on the utility scores at each assessment time multiplied by follow-up time interval. The area under the curve was used to determine the mean survival.[27] For patients who did not have a utility score (ie, censored patients), the last utility observation was carried forward.

Time Horizon

The time horizon of the analysis was the duration of the clinical trial (ie, 18–19 months) because more than 77% of the patients on cetuximab and 82% of those on best supportive care alone had died by the end of the data collection period. Because the median survival time in the study was less than 1 year, discounting was not used.

Base Case Analysis

In the base case analysis, each unit cost was multiplied by the resources used on a per-patient basis. The incremental ratios of the differences in mean costs, survival, and QALYs between the cetuximab arm and the best supportive care arm were determined for the entire study population and for the patients whose tumors harbored wild-type KRAS. Results are presented as the cost per life-year gained and the cost per QALY gained.

Sensitivity Analyses

We performed one-way deterministic sensitivity analyses to test the robustness of the incremental ratios. In these analyses, the cost of cetuximab was varied from $2.94 (Switzerland) to $6.73 (United States) per mg to reflect the international cost of the drug.[19] Survival was varied from +20% to −20% of the original base case value. In addition, between-country heterogeneity was examined. The 95% confidence intervals (CIs) for the incremental cost-effectiveness ratios and the incremental cost–utility ratios were estimated by use of a nonparametric bootstrapping method (with 1000 iterations). Uncertainty surrounding the estimates of cost-effectiveness was illustrated by means of cost-effectiveness acceptability curves, which depict the probability that an intervention is cost-effective when compared with an alternative (either another intervention or no intervention) for a range of incremental cost-effectiveness ratios.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.