e-Poster Presentation Clinical Oncology Society of Australia Annual Scientific Meeting 2021

Comparison of capecitabine concentration determined by capillary sampling versus venous blood sampling for therapeutic drug monitoring: a pilot study (#204)

Mohsen Shafiei 1 , Peter Galettis 2 , Philip Beale 1 , Jennifer Martin 2 , Andrew McLachlan 3 , Prunella Blinman 1
  1. Concord Clinical School, Faculty of Medicine and Health, University of Sydney, Concord, NSW, Australia
  2. School of Medicine & Public Health, The University of Newcastle, Centre for Drug Repurposing and Medicines Research, Callaghan, NSW, Australia
  3. Sydney Pharmacy School, University of Sydney, Sydney, NSW, Australia

Purpose Capecitabine is a convenient oral fluoropyrimidine chemotherapy agent that can cause severe toxicity especially in older adults. Therapeutic drug monitoring (TDM) allows personalised dosing of chemotherapy to reduce toxicity but is not well established for capecitabine. The purpose of this study was to compare the concentrations of capecitabine and its metabolites obtained by capillary sampling and venous blood sampling for the purpose of TDM, and the acceptability of both sampling methods.

Methods Paired (duplicate) capillary volumetric absorptive microsampling (VAMS) using MITRA® devices and venous whole blood samples were collected in participants taking capecitabine for colorectal cancer.  Samples were analysed by liquid chromatography with tandem mass spectrometry (LC-MS/MS). Correlation between the two sampling methods was determined by Bland-Altman bias plot and Passing-Bablok analyses. Patients’ preferences were elicited by Likert numeric rating scale and pain by a Visual Analog Scale (range, 0-10).

Results Capecitabine concentrations by capillary sampling and venous blood sampling were highly correlated (Pearson correlation: 0.97, Coefficients of determination: 0.94, p<0.0001). Capecitabine concentrations in VAMS were consistently lower than the paired plasma concentration (mean capecitabine VAMS/Plasma=2774/3709=75%).  Bland-Altman analysis indicated poor agreement between sampling matrices with 28% (95% Cl, 4.02 – 52.00) bias. Participant ratings showed VAMS was the preferred method by all 10 patients (finger prick strongly preferred: 9/10, finger prick slightly preferred: 1/10). Most patients reported no pain (median 0, range 0 to 1).

Conclusion Capecitabine concentration measured by capillary and venous blood sampling were highly correlated, but consistently lower in capillary sampling. High bias was likely due to small number of samples and great variation of differences. VAMS was the preferred method with minimal pain. Further research with higher number of participants is required to determine the effectiveness of capillary sampling as a substitute for venous sampling.