Page 262 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Chapter 10
Oral uracil loading dose
The oral uracil loading dose assay was performed in 92 patients. Results of 82 wild-type patients were compared between study centres. The mean U/DHU ratio was significantly lower in one centre (0.622) compared to the reference centre (1.03, p=0.046). It appeared that baseline BSA was significantly associated with the outcome of the U/DHU ratio (p=0.008), a higher baseline BSA was related to a lower U/DHU ratio. Baseline BSA was not differently distributed between the centres (p=0.637). The different mean U/DHU ratio of one centre was far from the cut-off U/DHU ratio (2.4)21 of DPD deficient patients, therefore no patients were excluded from further analyses.
2-13C-uracil breath test
The 2-13C-uracil breath test was determined in 82 patients.23 On average, 488 mg 2-13C-uracil was administered, ranging from 312 to 840 mg (6 mg/kg dose). Results of 74 wild-type patients were compared between study centres. The mean delta-over-baseline ratio at t=50 minutes (DOB50) was significantly lower in one centre (137.7 ‰) compared to the other two centres (173.5 ‰ and 168.4 ‰, p<0.009). It appeared that gender was significantly associated with the outcome of the DOB50 (p=0.003), with higher DOB50 values in females. Males and females were not significant differently distributed between the centres (p=0.263). The significantly different mean DOB50 was not as low as the DOB50 cut-off value (128.9 ‰)25 for DPD deficient patients, therefore no patients were excluded. A significant correlation between the DOB50 determined in breath samples and the 13C-dihydrouracil plasma levels (r2=0.178, p<0.001) could be demonstrated, not for the 13C-DHU/U ratio (r2=0.014, p=0.29). Results are shown in Supplementary Figure 2.
Association with onset of severe toxicity
Clinical validity parameters, i.e. sensitivity, specificity, NPV, PPV and F1-score of the endogenous DHU/U ratio and the endogenous uracil levels for their association with the onset of severe fluoropyrimidine-induced toxicity were calculated and shown in Table 3. The endogenous uracil levels have the highest F1-score of 24%. No significant difference was identified between the median endogenous DHU/U ratio or endogenous uracil level between patients who experienced severe toxicity or not (Figure 2). For the oral uracil loading dose and 2-13C-uracil breath test too few patients were enrolled, therefore the association with the onset of severe toxicity was investigated in an explorative way only for these two phenotyping assays (Supplementary Table 2, Supplementary Figure 3). Yet, the data show similar results in clinical validity parameters and also no difference between medians of patients who experienced severe toxicity or not.
Association with DPD deficiency
DPD deficiency, defined as low DPD activity levels in PBMCs (≤5.9 nmol/[mg*h]),6 was identified in 7 out of 73 patients (9.6%) or 6 out of 64 patients (9.4%). Clinical validity parameters for association with DPD deficiency are shown in Table 4. High specificity and NPV values were identified, but low sensitivity and PPV values. The oral uracil loading dose has the highest F1-score of 40%. The endogenous uracil levels have the highest sensitivity of 43%.
260