Page 245 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Supplemental Figure 3. Different genotyping techniques used in PGx
Overview of 32 answers from 27 responders, shown as percentages.
Abbreviations: PCR: Polymerase Chain Reaction; RFLP: Restriction Fragment Length Polymorphism; Melting Curve: Melting curve analyses, including High Resolution Melt, ‘LightCycler’, ‘Rotorgene Melt Curve’ and ‘LightSnip assay’; ASA: allele-specific amplification.
Supplement
3% 3%
16%
3%
6%
35%
34%
PCR-RFLP
Ta qM an
KASPA
Strip assay (blot)
Melting Curve
P yro se qu e nc in g PCR-ASA
12% 23%
12%
two independent techniques,
two technicians using the same technique,
one technician using the same technique on two independent days,
23%
15%
15% using two independent samples per patient, none
other
Supplemental Figure 4. Confirmation practice in PGx
Overview of 34 answers from 27 responders, shown as percentages. ‘Other’ includes: “two technicians judge the results”, “new sample is requested with divergent result”, “two persons judge”, “authorized by Clinical Chemistry”, “phenotyping if necessary”, “use of synthetic controls”, “use of internal sequenced controls”, “use of heterozygous positive control DNA”. In this study these are not considered a (cost inducing) confirmation or replication method.
Two independent genotyping methods as confirmation practice
Since 2005 (Erasmus MC) and 2010 (LUMC) PGx tests are executed as a part of routine diagnostics in clinical care. Digitally stored results were included in the analysis. For LUMC and Erasmus MC data on 16,932 and 72,910 SNP tests were available, making the sample size a total of 89,842 tests. Results are shown for each SNP and separated per laboratory in Supplemental Table 1. Results of CYP2D6 are shown in Supplemental Table 2.
At LUMC, in 2011 a single discrepancy was observed as previously described.11,12 Briefly; the TaqMan assay identified a patient as CYP3A5*3/*3, while results from PSQ showed that the patient was CYP3A5*1/*3. A second blood sample was obtained and genotyped. Results from the second analysis (both Taqman as PSQ) classified the patient as CYP3A5*1/*3. However, the PSQ results showed some inconsistencies, making the results questionable.
243
9