Detecting recurrence of colorectal cancer following five years of scheduled CEA and CT – An economics analysis from the Follow-up after Colorectal Surgery (FACS) Randomized Clinical Trial

The problem

Computed tomography (CT) imaging of the chest, abdomen and pelvis, and regular blood carcinoembryonicantigen (CEA) measurement can be used to detect recurrence of colorectal cancer after surgery. The FACS trial is investigating their benefits and costs. An earlier analysis of 3 to 5 year follow-up trial data suggested that intensive CT or CEA screening increased the rate of surgical treatment of recurrence with curative intent compared with minimal follow-up, but found no advantage in combining intensive CEA and CT. The survival advantage of any strategy tended to be small. We conducted a cost effectiveness analysis alongside the FACS trial using five year follow-up data to examine the effect of the strategies on patient’s health related quality of life and utilization of health services, and to assess whether the strategies are cost-effective to the NHS.

The approach

The analysis adapted a NHS perspective, therefore only NHS related health care costs were included in the analysis. The total cost of colorectal cancer related health care over five years was estimated, using data from annual follow-up questionnaires self-reported by the patients, supplemented by test and treatment information collected in trial case report forms. Unit costs for resource use were obtained from standard national sources.The effectiveness of the follow-up strategies was measured using QALYs, estimated from EQ-5D data using the ‘area under the curve’ approach. Multiple imputation techniques were used to handle missing resource use and EQ-5D data. Between-group differences in mean costs and QALYs were estimated using datasets after multiple imputations. Seemingly unrelated regression and bootstrapping were used to simultaneously estimate differences in costs and QALYs between follow up strategies, adjusting for patient age, sex, EQ-5D scores at baseline, cancer stage, and smoking status. Sensitivity analysis was conducted to examine the impact on the results of different assumptions and scenarios.

Findings

A total of 1202 patients were randomized into four groups: CEA only (n=300), CT only (n=299), CEA+CT (N=302), or minimum follow-up (n=301). Mean EQ-5D scores decreased similarly for all groups, from 0.84(SD=0.19) to 0.65(SD=0.39) for the minimum group, from 0.83 (SD=0.22) to 0.68 (SD=0.39) for the CEA group, from 0.83 (SD=0.22) to 0.67 (SD=0.40) for the CT group, and from 0.83 (SD=0.2) to 0.67 (SD=0.38). Surgical treatment of recurrence occurred 97 times during the 5 year follow-up. Note that these are interim results, and are not to be quoted without permission. We only recently obtained the final 5 year follow up data so other health economics analysis is currently ongoing. Full results of health economics analysis will be presented at the conference in July.

Consequences

Will be presented in the conference in due course.

Credits

• Yaling Yang, Southamton University, Southampton, UK
• Helen Campell, Southamton University, Southampton, UK
• Alastair Gray, Southamton University, Southampton, UK
• David Mant, Southamton University, Southampton, UK
• John Primrose