Pharmacokinetic-based drug–drug interactions (DDIs) significantly contribute to drug failure by decreasing safety or efficacy. Relevant DDIs result from a major change in plasma concentrations of the ‘victim’ drug exerted by the ‘perpetrator’ drug that may interfere at different pharmacokinetic (PK) steps; that is, absorption, distribution, metabolism and/or elimination. Many drugs can act as strong inhibitors or inducers of drug-metabolising enzymes, particularly of the cytochrome P450 (CYP) enzymes or compete for drugs’ transporters, namely P-glycoprotein (P-gp), breast cancer resistance protein (BCRP), organic anion transporter polypeptides (OATPs) 1B1 and 1B3, organic anion transporters 1 and 3, and organic cation transporter 2.1
Drugs are considered perpetrators when the clearance, maximal concentration (Cmax) or area under the curve (AUC) of the respective victim drug is significantly changed to an extent that can impact clinical outcomes, requiring dose adjustments or drug change. According to Food and Drug Administration (FDA)-defined boundaries, clinically relevant perpetrators can be: moderate or strong inhibitors if they cause at least a two- to five- or greater than fivefold increase, respectively, in the AUC of the victim substrate on a given CYP metabolic pathway; or moderate or strong inducers if they decrease the AUC of victim substrates by 50 to <80% or by ≥80%, respectively, on a given metabolic pathway.2 For a given transporter, moderate and strong substrates are drugs that increase in vitro inhibition or AUC ≥1.5-fold of the specific, co-administered probe drug.2 Weak CYP inhibitors, inducers or transporter substrates have no clinical impact, as the variation of drug concentration falls within its therapeutic window and requires no therapeutic changes.
The Lexi-Interact (UpToDate Lexidrug, formerly Lexicomp) and Micromedex online databases are the most sensitive in discriminating clinically relevant DDIs.3 In particular, Lexi-Interact grades DDIs from A (no clinical concern) to D (clinically significant interaction with a suggestion to modify regimen), and the X category identifies a contraindicated co-medication, since risks exceed benefits.3
Polypharmacy is a known risk factor for clinically relevant DDIs.4,5 While there is no validated consensus on a specific threshold, the most used is five or more medications, including over-the-counter, prescribed traditional and complementary medicines.4,6 However, this threshold varies with age and clinical settings.4,6 Polypharmacy is unavoidable in complex, chronic, comorbid and older patients, but it can be dangerous when it is inappropriate and includes interacting drugs that reduce safety and efficacy of the therapy.4
Patients with cardiovascular diseases are unavoidably exposed to multiple drugs, due to frequent comorbidities and, therefore, to potential, clinically relevant DDIs.7 Heart disease is associated with the highest prevalence of polypharmacy in US adults.8 Patients hospitalised for acute coronary syndrome (ACS) receive a mean of 9.9 ± 2.6 drugs/day, even in high LexiComp categories (C 75.3%, D 4.8 and X 0.3% of the patients).9 The most represented drugs in ACS patients are, in decreasing order, aspirin and a P2Y12 receptor inhibitor, statins, blood pressure-lowering drugs, glucose-lowering drugs and benzodiazepines.9 Dual antiplatelet therapy with low-dose aspirin and a P2Y12 inhibitor (ticagrelor, prasugrel or clopidogrel) is recommended post-ACS with or without percutaneous coronary intervention (PCI), and after elective PCI for variable duration.10,11 Moreover, cigarette smoking, which is strongly associated with cardiovascular diseases, and smoking cessation drugs can modulate some CYPs.12–15
This review reports evidence of PK-related DDIs with oral inhibitors of the P2Y12 platelet receptor; that is, clopidogrel, prasugrel and ticagrelor (see Supplementary Methods for methodology). It focuses only on clinically relevant DDIs, which may impact patient safety (mostly bleeding) and antithrombotic effectiveness. Pharmacodynamic interactions are beyond the scope of this review.
P2Y12 Inhibitors: in vitro Equipotency
Clopidogrel, prasugrel and ticagrelor inhibit platelet aggregation by preventing adenosine triphosphate (ADP) from binding to its P2Y12 receptor.
Clopidogrel and prasugrel are thienopyridines and prodrugs, their active metabolites (AM) covalently bind the P2Y12 receptor, thus permanently blocking the ADP-dependent activation pathway for the entire platelet lifespan (Figures 1 and 2). Thiol-containing clopidogrel AM (R-130964) and prasugrel (R-138727) AM irreversibly bind Cys97 in the first extracellular loop of the P2Y12 receptor, which permanently inactivates the receptor.16
Ticagrelor is a carbocyclic nucleoside analogue, with an adenosine-like (cyclopentyl triazolopyrimidine) structure that reversibly inhibits the P2Y12 receptor by binding an allosteric site distinct from the ADP-binding site. Ticagrelor also generates an AM (the AR-C124910XX), with a pharmacodynamic and PK, such as the parent compound (Figure 3). Based on chemical structure, ticagrelor increases extracellular adenosine levels in vitro by inhibiting the nucleoside transporter 1 that carries adenosine into the erythrocytes.17 Increased extracellular adenosine is the most plausible cause of some non-haemostatic, ticagrelor-specific adverse reactions in patients, namely dyspnoea, bradycardia, creatinine increase and thrombotic thrombocytopenic purpura.
Via the A1 and A2A receptors on the C fibres of the vagus nerve, adenosine provokes broncho-constriction and dyspnoea, while the A1 receptors in the heart trigger bradycardia and ventricular pauses.18 Moreover, adenosine interferes with the tubulo-glomerular system, increasing renal vascular tone and resistance, and decreasing renal blood flow.19 These mechanisms may account for the significant increase in serum creatinine in the ticagrelor as compared with clopidogrel groups, that reversed upon ticagrelor cessation in the PLATO trial.20 Those adverse events have not been reported for clopidogrel and prasugrel, ruling out a class effect.
The AMs of clopidogrel and prasugrel, ticagrelor and its AM added in vitro to platelet-rich plasma or washed platelets show similar receptor affinities (in the nM range) and potencies, since they inhibit platelets almost completely, with IC50s in the low µM range.21,22 However, when administered in vivo, clopidogrel achieves the lowest ADP-induced platelet inhibition ex vivo in healthy subjects and patients compared with the other two drugs. This apparent loss of potency reflects a less efficient clopidogrel AM generation in vivo, rather than a true difference in pharmacological potency.22
Pharmacokinetics: Drug Activation, Inactivation and Transporters
Clopidogrel
Intestinal absorption of clopidogrel is rapid, with AM plasma levels peaking within 1 hour after dosing.23 The P-gp binds and extrudes clopidogrel from enterocytes to the intestinal lumen, contributing to the amount of drug entering portal blood (Figure 1), as also suggested by pharmacogenomic studies on P-gp variants.24 The carboxylesterase, CES-1, in the intestine, plasma and liver degrades approximately 85% of clopidogrel into an inactive carboxylic acid derivative (SR-26334), further transformed into an acyl-β-D-glucuronide by the hepatic UDP-glucuronosyltransferases, UGT2B7 and 2B4, and the intestinal UGT2B17.25,26
Only approximately 15% of clopidogrel undergoes liver first-pass metabolism, and is activated through two sequential oxidative reactions (Figure 1) forming the 2-oxo-clopidogrel inactive intermediate and then the thiol R-130964 AM. Different CYPs mediate these steps (Figure 1) with different hierarchies. CYP2C19 participates in both reactions, and CYP3A4 appears relevant in the second step, depending on the residual clopidogrel concentration.27 In fact, CYP3A4 is the main isoenzyme for biotransformation of high clopidogrel concentrations (>10 µM), as clopidogrel inhibits the CYP2C19 at high concentrations, while below 10 µM, CYP2C19 is the major biotransforming enzyme.28 Intestinal CYP3A4 also contributes to clopidogrel bioactivation.29 CYP2B6, CYP1A2, CYP3A5 and CYP2C9 seem less relevant, while the role of hepatic paraoxonase-1 is debated.30 Clopidogrel metabolites are excreted in urine (approximately 46%) and faeces (39–59%).31
Prasugrel
Prasugrel is rapidly absorbed and almost immediately hydrolysed to an inactive intermediate (thiolactone 2-oxo-prasugrel, R-95913) in the intestine and plasma, mostly by CES-2 (Figure 2).32 The R-95913 is then converted into the sulfhydryl-containing R-138727 AM, which accounts for approximately 70% of the absorbed dose. Prasugrel AM is already detectable in blood after 15 minutes and peaks approximately 30 minutes from dosing.33 in vitro studies indicate that the AM is generated by parallel, rather than hierarchical, activities of CYP3A, 2B6, 2C9 and 2C19 (minor). Consistent with parallel biotransformation by multiple CYPs, CYP2C19 polymorphisms do not affect prasugrel biotransformation, which is different to clopidogrel (Figures 1 and 2, and Tables 1 and 2).34 The rapid appearance of the plasma AM is also consistent with a significant, mostly CYP3A4-mediated, conversion of the intermediate R-95913 already in the intestine at a rate comparable to the liver.33,35 Drug transporters do not participate in prasugrel PK.34 Prasugrel metabolites are excreted in urine (approximately 70%) and faeces (approximately 25%).36
Ticagrelor
The absorption of ticagrelor is rapid and peaks 1.5 hours post-dosing.37 Ticagrelor is both a substrate and weak inhibitor of the P-gp, OATPs 1B1, 1B3 and 2B1 at concentrations in the low μM range, consistent with those reached in plasma and intestinal mucosa.38–40 Ticagrelor oxidation via the CYP3A4 and 3A5 in the intestine and liver generates a major, equipotent AM, the AR-C124910XX, which is also a weak inhibitor of the P-gp (Figure 3).41 This AM contributes to approximately 30% of the total antiplatelet effect.37 Based on polymorphisms, the OATP1B1, CYP3A4 and UGT2B7 can influence ticagrelor and its AM concentrations, but at levels with no clinical consequences.42,43 Ticagrelor and AR-C124910XX are mainly excreted in the faeces (approximately 50%). Ticagrelor-inactive metabolite AR-C133913XX, resulting from its N-dealkylation, and its glucuronide conjugate are excreted in urine; other minor inactive metabolites have been detected in urine and faeces.37
Drug–Drug Interactions
Clopidogrel
Victim
Strong CYP2C19/3A4 inducers (e.g. carbamazepine, phenytoin, rifampin) substantially increase clopidogrel AM concentrations and platelet inhibition (Table 1). For example, rifampin increases the AUC of clopidogrel AM by approximately fourfold, and platelet inhibition by approximately 23%.44,45 Considering that CYP2C19 is the main path for clopidogrel bioactivation (Figure 1), the concomitant use of strong CYP2C19 inducers is ‘discouraged’ or should be avoided according to regulatory agencies, to avoid an increased bleeding risk (Table 1).46,47 Fluoxetine and fluvoxamine, the CYP2C19-inhibiting selective serotonin reuptake inhibitors, decrease clopidogrel AM plasma levels (Table 1).48 In a population study, patients co-treated with clopidogrel and fluoxetine or fluvoxamine showed increased ischaemic events (HR 1.12; 95% CI [1.01–1.24]), as compared with non-CYP2C19-inhibiting selective serotonin reuptake inhibitors. The risk was particularly high in patients aged ≥65 years (HR 1.22; 95% CI [1.00–1.48]).49
Proton pump inhibitors, except rabeprazole and pantoprazole, are extensively metabolised by and competitively inhibit the CYP2C19 and 3A4, which results in a large, clinically relevant reduction of clopidogrel AM concentration and antiplatelet effect. A recent network meta-analysis of 16 studies showed an increase in major adverse cardiovascular events post-PCI in patients co-treated with clopidogrel and esomeprazole (effect size 1.28; 95% CI [1.09–1.51]), omeprazole (effect size 1.23; 95% CI [1.07–1.43]), pantoprazole (effect size 1.38; 95% CI [1.18–1.60]) and lansoprazole (effect size 1.48; 95% CI [1.22–1.80]), as compared with control groups, while rabeprazole has no impact.50 Thus, regulatory agencies recommend avoiding omeprazole and esomeprazole with clopidogrel (Table 1).47,50,51 Pantoprazole, dexlansoprazole, rabeprazole or H2 receptor inhibitors (ranitidine and famotidine) are appropriate options.
Strong CYP3A4 inhibitors variably reduce clopidogrel AM; ketoconazole and clarithromycin reduce AM AUC by approximately 30% and ritonavir by approximately 50%.35,52 Residual platelet aggregation was 30% higher in patients co-treated with calcium channel blockers versus those not on calcium channel blockers, likely through a competition on the CYP3A4.53 However, a large registry of patients post-PCI showed no impact of the co-administration on serious vascular events.54 The CYP3A4 substrates, atorvastatin and simvastatin, inhibit clopidogrel biotransformation and antiplatelet activity less potently than ketoconazole.28 Thus, although not contraindicated, caution should be exerted when co-prescribing strong CYP3A4 inhibitors or substrates with clopidogrel (Table 1).
Angiotensin-converting enzyme inhibitors may compete to displace clopidogrel from CES-1 increasing clopidogrel AM levels; however, co-administration of clopidogrel and angiotensin-converting enzyme inhibitor was not associated with increased bleeding (adjusted HR 1.10; 95% CI [0.97–1.25]).55
Morphine reduced the AUC of clopidogrel AM by approximately 34% and halved platelet inhibition in a placebo-controlled study on healthy individuals.56 A post-hoc analysis of the EARLY-ACS trial showed that patients receiving clopidogrel and morphine had a significant increase in ischaemic events at 96 hours, and mortality or MI at 30 days post-ACS.57
An analysis of the FDA Adverse Event Reporting System on 187,919 reports, including 165,487 of clopidogrel, 22,432 of ticagrelor, and 2,504 with morphine and a P2Y12 inhibitor co-medication, showed a significant increase in cardiovascular events (adjusted OR 1.55; 95% CI [1.14–2.11]) and renal impairment (adjusted OR 1.65; 95% CI [1.20–2.26]) when morphine is co-administered with clopidogrel or ticagrelor.58 A validation cohort of 5,240 ACS patients receiving either clopidogrel or ticagrelor also showed increased MI in patients using morphine (adjusted OR after propensity score matching 1.50; 95% CI [1.07–2.10]).58 The most plausible explanation is that by slowing gastric emptying and inducing vomiting, morphine and other opioids reduce the absorption of clopidogrel.59 However, it has also been hypothesised that high creatinine, rather than a DDI, may mediate worse cardiovascular outcomes.58 It is plausible that fentanyl affects clopidogrel similarly to morphine, based on a class effect, as indicated by regulatory agencies (Table 1).46,47 However, further studies on clinical outcomes are needed.59
Cigarette smoking in clopidogrel-treated patients is associated with the so-called ‘smoker’s paradox’. Tobacco smoking, likely by increasing CYP1A2 activity, is associated with 1.2-fold higher clopidogrel AM concentrations versus non-smokers and increased platelet inhibition.60 However, the increased haematocrit associated with smoking may explain the apparent higher degree of platelet inhibition with whole-blood assays in clopidogrel-treated smokers.61 A large meta-analysis of observational and randomised studies showed an increased efficacy of clopidogrel in indirect comparisons between smokers and non-smokers.62 An in vitro study showed an induction of P2Y12 receptors on megakaryocytic cell lines induced by nicotine, whose clinical impact remains unknown.63
No PK-based interactions have been described for the co-administration of clopidogrel with either warfarin or direct oral anticoagulants.58
Perpetrator
The acyl-β-D-glucuronide metabolite of clopidogrel is a strong CYP2C8 inhibitor and inhibits the metabolism of montelukast, repaglinide, selexipag and dasabuvir to a clinically significant extent (Table 2).52,64,65 The association of cerivastatin–clopidogrel was associated with a high rate of rhabdomyolysis (OR 29.6; 95% CI [6.1–143]), likely through clopidogrel-induced strong inhibition of CYP2C8, the main metabolic clearance path for cerivastatin that contributed to its marketing withdrawal.66,67
Clopidogrel inhibits CYP3A4 in vitro, but this seems of no clinical relevance since the concentration of the CYP3A4 substrate simvastatin is not affected.68,69 Moreover, being a strong CYP2B6 inhibitor, clopidogrel increases plasma levels of the reverse transcriptase inhibitor, efavirenz, and halves the concentrations of bupropion AM (hydroxybupropion), an antidepressant and smoking-cessation agent.67,70
Clopidogrel does not inhibit the OATP1B1 and 1B3 to a clinically relevant extent.68 in vitro, clopidogrel inhibits P-gp; however, in vivo it has no effect on digoxin, a P-gp probe drug.71,72
Prasugrel
Victim
Strong CYP3A4 inhibitors minimally affect prasugrel AM concentrations, thus coadministration of these drugs is not contraindicated (Table 1).35,67,73 A recent study on HIV patients further confirms the absence of DDIs with prasugrel, differently from clopidogrel or ticagrelor.74 Consistent with a lack of CYP3A4-based DDIs, at variance with ticagrelor, the CYP3A4*22 does not affect prasugrel AM concentrations.75 Omeprazole and pantoprazole also do not interact with prasugrel.75
The CYP2C19/3A4 inducer rifampin does not affect prasugrel AM concentrations nor platelet inhibition.76 Data of morphine on prasugrel are heterogeneous. Prasugrel AUC and platelet inhibition were not significantly reduced by morphine in a crossover trial of healthy volunteers, while the effect on platelet function seemed more pronounced in patients, albeit in a non-randomised study.77,78 Clinical outcome studies on prasugrel and morphine are too limited and largely on platelet function, to assess whether these changes have relevance on clinical outcomes.59,79,80 However, regulatory agencies warn about prasugrel–morphine interaction based on the reduced motility of the gastrointestinal tract caused by morphine (Table 1). Cigarette smoking has no effect on prasugrel AM, and CYP1A2 is not involved in prasugrel biotransformation.60
Perpetrator
Prasugrel does not interact with CYP2C9, 2C19 and P-gp, thus has no effect on quinidine, verapamil and digoxin (Table 2).81 It is a weak inhibitor of CYP2B6, but this effect is considered relevant only with drugs with a narrow therapeutic window and having CYP2B6 as the only metabolic pathway (e.g. cyclophosphamide, efavirenz).36,67,73
Prasugrel does not inhibit CYP2C9, 2C19 and 2C8.67,81,82 The inactive and active metabolites of prasugrel (R-95913 and R-138727) do not inhibit the CYP1A2, 2C9, 2C19, 2D6 or 3As. Unlike ticagrelor, prasugrel is not associated with rhabdomyolysis when co-administered with some statins.83
Ticagrelor
Victim
The strong CYP3A4 inhibitor, ketoconazole, increases ticagrelor AUC by 7.3-fold, prolongs ticagrelor effect and increases bleeding (Table 1).84–86 Consistent with the relevance of CYP3A4 on ticagrelor PK, recent pharmacogenomic data associate the loss-of-function CYP3A4*22 polymorphism to an approximately twofold increase in ticagrelor AUC, higher platelet inhibition in healthy individuals, and approximately threefold increase in major bleeding in ticagrelor-treated patients, although these clinical data remain controversial.75,87,88 Ticagrelor is contraindicated by regulatory agencies with strong CYP3A4 inhibitors, such as lopinavir/ritonavir, also used in COVID-19 infections, or ritonavir alone.85,86,89,90
The strong induction of CYP3A4 activity by rifampicin halves the Cmax, and considerably shortens the half-life of ticagrelor.91 A registry study in patients co-treated with ticagrelor and anti-epileptic CYP3A inducers (i.e. phenytoin, carbamazepine and phenobarbital) showed a residual platelet aggregation approximately sevenfold higher than in patients without concomitant anti-epileptic drugs.92 The European Medicine Agency (EMA) discourages ticagrelor in association with strong inducers of the CYP3A4 (phenytoin, carbamazepine and phenobarbital), while this association is not discouraged for prasugrel (Table 1).84
Cyclosporine, a probe drug and potent P-gp inhibitor, increases the AUC and Cmax of ticagrelor greater than twofold, and the concentration of its AM by approximately 30%.93 Hence, the EMA discourages ticagrelor with potent P-gp and moderate CYP3A4 inhibitors (e.g. verapamil, quinidine) that may increase ticagrelor concentrations (Table 1).90
Several studies show that morphine significantly reduces the antiplatelet effect of ticagrelor (Table 2).79,94 Recent population PK/pharmacodynamic model studies confirmed the significant inhibiting effect of morphine on ticagrelor antiplatelet effect.95,96 A subanalysis of the TREAT trial in morphine-treated versus -untreated patients showed that morphine was associated with an approximately fivefold increase of early and approximately twofold increase of late reinfarctions.97 However, another study did not confirm this clinical consequence.98 Whether the morphine effect can be extended to other opioids, such as fentanyl, is plausible, as also indicated by the FDA and the EMA.84,99–101
There are no specific studies on the effect of cigarette smoking on ticagrelor.
Perpetrator
Ticagrelor generates false negative results interfering with diagnostic assays of heparin-induced thrombocytopenia (HIT; Figure 3), a rare, severe and fatal complication if misdiagnosed.102,103 This laboratory effect appears related to the mode of action, chemical structure and plasma concentrations of ticagrelor.103 The regulatory labelling of ticagrelor has been updated to include a warning for this clinically relevant, drug–laboratory interaction.84,101 Given the possibility of HIT recurrence, especially in the months after a clinical episode, avoiding ticagrelor in patients with a recent history of HIT may be advised. The EMA states that in patients who have developed HIT, the benefit–risk of continued treatment with ticagrelor should be assessed, taking both the prothrombotic state of HIT and the increased risk of bleeding with concomitant anticoagulant and ticagrelor treatment into consideration.84
Ticagrelor is a weak inhibitor of P-gp and CYP3A4.104 More frequent drug monitoring is required when digoxin and cyclosporin are co-administered (Table 2).40,105 Ticagrelor and its AM have minimal impact on CYP1A1, 1A2, 2C9, 2B6, 2C8, 2C19, 2D6 and 2E1.
Importantly, ticagrelor can increase concentrations and toxicity of statins not only through the interaction with the CYP3A4 (high-dose simvastatin and lovastatin; Table 2), but also through drug transporters (rosuvastatin and atorvastatin), as indicated by recent data.42,84,101,106,107 Rhabdomyolysis has been repeatedly reported in patients on ticagrelor and rosuvastatin or atorvastatin.108,109
In the WHO pharmacovigilance database (VigiBase), among 2,464 rhabdomyolysis reports in patients taking a statin and antiplatelet drugs (i.e. aspirin, prasugrel, ticagrelor and clopidogrel), rhabdomyolysis significantly increased versus statin-only when atorvastatin (adjusted reporting OR [ROR] 1.30; 95% CI [1.02–1.65]) or rosuvastatin (adjusted ROR 1.90; 95% CI [1.42–2.54]) were co-administered with ticagrelor. There was a similar trend for simvastatin (adjusted ROR 1.42; 95% CI [0.92–2.18]). At variance with ticagrelor, there was no increase in rhabdomyolysis reports when statins were given with aspirin, clopidogrel or prasugrel compared to statins alone.83 Rhabdomyolysis in ticagrelor-treated patients was further increased by age ≥75 years, chronic kidney disease and statin dose.83
Ticagrelor in vitro inhibits the BCRP, OATP 1B1, -1B3 and -2B1 in the low μM range, consistent with concentrations in the intestinal mucosa, and this mechanism reduces the extrusion of rosuvastatin. An in silico model of ticagrelor predicted a greater than twofold increase in rosuvastatin concentration through BCRP inhibition.38 In a placebo-controlled, crossover, randomised trial, participants randomised to rosuvastatin plus ticagrelor showed an increase in the AUC and Cmax of rosuvastatin of >2.5-fold (90% CI [1.8–3.8] and [1.7–4.0], respectively) with a doubled half-life as compared with rosuvastatin plus placebo.38 Thus, the intestinal BCRP inhibition of ticagrelor accounts for the interaction with rosuvastatin.38 In the PLATO trial, ticagrelor was associated with increased serum creatinine as compared with clopidogrel, mostly in older adults.20 Although the kidney is not a major excretion route for statins, the ticagrelor–statin interaction may worsen toxicity by a decreased renal clearance.42 Regulatory agencies warn of the co-administration of some statins and ticagrelor (Table 2).84
Finally, the inhibition of URAT1 and BCRP transporters at the renal tubular levels may account for the increase in uric acid that is associated with ticagrelor intake (Figure 3).18,110,111 The incidence of gout was nearly doubled in patients on ticagrelor 90 mg twice daily in the PEGASUS-TIMI 54 trial (HR 1.77; 95% CI [1.32–2.37]).112 Thus, measuring uric acid is recommended after starting ticagrelor.90 None of the other P2Y12 inhibitors displace uric acid from its transporters in the kidney.
Conclusion
P2Y12 inhibitors, either as part of a dual antiplatelet therapy or as monotherapy, as for clopidogrel in aspirin-intolerant patients, are used in the secondary prevention of cardiovascular diseases. Within this drug class, clopidogrel and ticagrelor have the highest potential of generating PK-based, clinically relevant DDIs. Thus, caution and consideration are needed in multimorbid cardiovascular patients exposed to appropriate polypharmacy, as post-ACS.113,114 More data on heavy smoking, smoking cessation drugs or the newer nicotine-based products are needed regarding pharmacological interactions and clinical impact.
Low-dose aspirin is the reference single antiplatelet therapy, unless there is an aspirin allergy. Importantly, aspirin biotransformation does not involve CYPs or drug transporters, thus offering the advantage of lack of PK-based drug interactions as compared with clopidogrel or ticagrelor.115 Among P2Y12 inhibitors, prasugrel has the lowest PK-based DDI potential, since neither strong inhibitors nor inducers of the CYPs affect its bioactivation. Prasugrel does not interact with transporters and has a minimal perpetrator potential. Notably, one randomised trial comparing head-to-head prasugrel versus ticagrelor as part of dual antiplatelet therapy post-ACS and two large observational studies, consistently showed a superior efficacy of prasugrel and similar or even better safety as compared to ticagrelor.90,116,117 Especially in observational studies, which reflect real-world settings, the low potential of prasugrel to generate clinically relevant DDIs in multimorbid and poly-treated patients with ACS may have contributed to the better outcomes of prasugrel.
Clinically relevant DDIs increase the variability in response to drugs, and worsen the efficacy and safety balance observed in randomised trials that usually include specific exclusion criteria to avoid DDIs.118 Among P2Y12 inhibitors, prasugrel is the agent with the lowest potential to generate clinically relevant DDIs.119 In the daily exercise of precision medicine, preventing significant DDIs, and choosing the right drug within a drug class, at the right time, in the multimorbid patients on unavoidable polypharmacy is central to improving treatment quality, avoiding failures and reducing healthcare costs.