Figure 3 Sensitivity analysis of the effect of ABCG2
c.421C>A variant allele on MPA AUCτ,ss(GMR=1.40) to account for hypothetical bias arising from not accounting
for the SLCO1B3 c.334T>G and UGT1A9
c.98T>C SNPs. Effect of the TT/TG (vs. GG) SLCO1B3
c.334 genotype is estimated at ROM=1.136 (95%CI 0.949-1.361) (Figure
S2); effect of the TC (vs. TT) UGT1A9 c.98 genotype is estimated
at ROM=1.098 (0.548-2.198) (Figure S3); and prevalence of the TT/TG and
TC genotypes is estimated at 31% and 3.9%, respectively (Figure S4).
Shown are bias-adjusted effects (GMRs) of the ABCG2
c.421C>A variant allele assuming considerable imbalance
between variant carriers and wild type controls in prevalence of
confounders (i.e., SLCO1B3 c.334 TT/TG genotype and UGT1A9
c.98 TC genotype) and different effects of confounders (expressed as
ratio of means, ROM) – those estimated based on published studies
(Figure S2, Figure S3), and larger effects. A . Confounder to
adjust for is SLCO1B3 c.334T>G genotype TT/TG (vs.
GG). Effect to adjust is GMR=1.40. The confounder effect is estimated at
ROM=1.136, rounded-up to 1.15. The effect “higher than estimated”
(ROM=1.25) is effect somewhat higher than that reported in the largest
individual study with 110 TT/TG and 218 GG subjects [31].B. Confounder to adjust for is UGT1A9
c.98T>C genotype TC (vs. TT). Effect to adjust is 1.40.
The confounder effect is estimated at 1.098, rounded-up to 1.10. The
effect “higher than estimated” (ROM=1.50) is effect somewhat higher
than that reported in the largest individual study with 10 TC and 328 TT
subjects [32]. C . Assumed is simultaneous and maximum
imbalance between ABCG2 c.421 variant carriers and wild type
controls regarding SLCO1B3 c.334T>G TT/TG genotype
(50% vs. 20%) and regarding UGT1A9 c.98T>C TC
genotype (10% vs. 2%) with consecutive adjustment for their larger
effects (ROM=1.25, ROM=1.50): initial estimate (GMR=1.40) is first
adjusted for TT/TG to GMR=1.31, and GMR=1.31 is further adjusted forUGT1A9 SNP.
Dashed horizontal line depicts the conventional upper limit of
equivalent exposure (GMR=1.25).