Before the start of sample collection questionnaire A on paper will be filled in by all study subjects. The order of sample collection will be: automated capillary sampling, lancet capillary sampling and venipuncture. Herein automated and lancet capillary sampling will be performed by the study subjects themselves whereas the venipuncture will be performed by the study personnel. After all sampling has been completed, the subject is asked to complete questionnaire B which will evaluate pain, burden, ease of use and preference.

For subjects in arm A and C this entails the end of the study

The subjects of arm B are requested to perform automated capillary and lancet capillary sampling at home following their next two outpatient visits. During these outpatient visits, a reference value blood CEA measurement will be obtained using venipuncture by the personnel of the clinical laboratory of Erasmus MC. The required materials will be sent to the home address of the patient. Sampling will be performed at home and by the subjects themselves. Subjects will have access to the tutorial videos for automated and lancet capillary sampling.

Before the start of sample collection questionnaire A on paper will be filled in by all study subjects. The order of sample collection will be: automated capillary sampling, lancet capillary sampling and venipuncture. Herein automated and lancet capillary sampling will be performed by the study subjects themselves whereas the venipuncture will be performed by the study personnel. After all sampling has been completed, the subject is asked to complete questionnaire B which will evaluate pain, burden, ease of use and preference.

For subjects in arm A and C this entails the end of the study

The TAP-II device will be compared to lancet capillary sampling and the venipuncture

Home based (automated) capillary sampling will be considered feasible if a success rate of 85% or greater has been reached. Herein a successful (automated) capillary sampling at home is defined as a sampling of blood by the patient that reached the clinical laboratory of the hospital via post and in which a CEA level could be determined reliably. Both capillary sampling methods will be analysed and compared to venepuncture separately.

Reliability will be assessed by a Bland-Altman analysis to determine mean bias and the 95% limits of agreement of measurements from (automated) capillary samples compared to venipuncture samples. These will be compared to predefined clinically relevant cut-off values for the mean bias and the limits of agreement. A mean bias of greater or equal to +/- 5% and or 95% limits of agreement equal to or greater than +/- 10% will be considered clinically relevant and thereby unreliable. These cut-off values are defined based on previously found 95% limits of agreement of the automated capillary sampling device and the precision of the Cobas 8000 analyzer which will be used to perform the CEA analyses.