Discussion
We report a DDI between crizotinib and sofosbuvir/velpatasvir associated with severe cardiac toxic effects leading to crizotinib discontinuation. Overexposure to crizotinib could have accounted for these effects, which was confirmed by the clinical improvement with crizotinib dose reduction (3). Drug causality was probable according to the Naranjo scale (score = 8) (9). Factors that could explain the variability in crizotinib overexposure could be the patient being slightly non-compliant (Girerd score) and pharmacogenetic analyses showing no cytochrome 3A5 (CYP3A5) and CYP3A4 functional polymorphism (10,11). The pharmacokinetics mechanism of this DDI may have been a reciprocal interaction between crizotinib and velpatasvir (Figure 2). C rizotinib is a CYP3A4/5 and P-glycoprotein (P-gp) moderate inhibitor in vitro and could lead to a supratherapeutic plasma level of velpatasvir, a substrate of CYP3A4 and P-gp (Table I) (3,12). Furthermore, velpatasvir is a P-gp weak inhibitor (Table I) (12). Velpatasvir may have increased the plasma concentration of crizotinib, a P-gp substrate (Table I) (3). We tested this hypothesis with a disproportionality analysis using VigiBase (the World Health Organization global database of individual case safety reports (https://www.who-umc.org/vigibase/vigibase/)) and observed a significant association between the concomitant use of crizotinib and P-gp inhibitors and cardiac arrhythmia (odds ratio of reporting 2.5, 95% confidence interval 1.2-5.2) as compared with the use of crizotinib without a P-gp inhibitor (13). This pharmacovigilance statistical approach supports a DDI between crizotinib and P-gp inhibitors resulting in higher cardiac adverse-drug-reaction reporting.