1. The efficacy and safety of paxlovid in pregnant women
A descriptive study based on real-world data demonstrated that out of 11 pregnant women who received a short 5-day course of paxlovid, 7 successfully completed the treatment. All patients who received paxlovid experienced symptom relief without requiring additional treatment. Throughout the treatment period, all patients tolerated the medication well, with no immediate adverse reactions reported, except for one patient who experienced nausea. Furthermore, there were no observed adverse effects on fetal or neonatal health [11]. Another study involved 12 pregnant women who underwent a short 5-day course of paxlovid. Adverse effects included dysgeusia (n=11, 91.7%), diarrhea (n=2, 16.7%), and mild abdominal pain (n=1, 8.3%). Eleven patients delivered successfully with favourable outcomes for both the mother and fetus, while one patient was still in her pregnancy stage [12]. In a study conducted at the Johns Hopkins Institute for Clinical and Translational Research involving 47 pregnant women aged between 22 and 43 years, among the group of patients who initiated paxlovid within days 0-5 after symptom onset (30 patients), two individuals (4.3%) discontinued the medication before completing their prescribed course due to adverse effects; specifically, excessive fetal growth and polyhydramnios were observed in one patient (2.1%), while oligohydramnios developed in another patient (which also accounted for approximately 2%). The remaining participants tolerated the medication well throughout its administration period. In total, twenty-five patients (53.2%) were discharged following the completion of paxlovid treatment [13] (Table 3).
In our study, the two pregnant women demonstrated favourable short-term pregnancy and neonatal outcomes, consistent with previous research findings [14,15,16]. However, the woman who underwent kidney transplantation delivered a preterm infant at 26+ GW, resulting in unfavourable neonatal outcomes. Currently, there is extensive documentation indicating that pregnancies in kidney transplant recipients present significant risks to the mother, fetus, and transplanted kidney. The primary risk factors include abortion, infection, proteinuria, preeclampsia, premature delivery, foetal growth restriction, stillbirths and neonatal malformations [14]. Based on current evidence-based research, it is probable that the adverse outcomes of neonates delivered by this woman can be primarily attributed to her mother's medical history rather than her use of the paxlovid device. The present study represents the initial report on short-term pregnancy outcomes and fetal effects in three pregnant women who were concurrently exposed to both paxlovid and immunosuppressive agents. Despite these encouraging findings, further investigations are necessary to evaluate the effectiveness and safety of oral antiviral therapies against various COVID-19 variants during pregnancy, as well as their long-term impact on maternal and neonatal outcomes.
Therefore, based on the evaluation of paxlovid's mechanism of action, real-world study data, and recommendations from domestic and international authorities, it is recommended that a risk-benefit assessment be conducted for pregnant women with COVID-19. This assessment may encompass various factors, such as comorbid medical conditions, body mass index ≥35, unvaccinated status, and prolonged use of immunosuppressive agents. Based on the risk-benefit assessment, paxlovid is recommended for eligible pregnant women [15]. Paxlovid is a relatively safe option during pregnancy.
2. Interactions between paxlovid and calcineurin inhibitors
Ritonavir, a protease inhibitor for human immunodeficiency virus type 1 in paxlovid, does not exhibit activity against the major protease of SARS-CoV-2. Instead, it increases the plasma concentration of nematavir by inhibiting its CYP3A-mediated metabolism [4-5]. As a potent and irreversible inhibitor of CYP3A, ritonavir may interact with drugs metabolized by this hepatic enzyme or inhibitors/inducers, leading to either enhancement or attenuation of the combined drugs' effects [16].
Both cyclosporine and tacrolimus are calcineurin inhibitors (CNIs) and exhibit significant interactions with ritonavir through CYP3A metabolism and P-glycoprotein transport. Pharmacokinetic studies have demonstrated that coadministration of CNIs with ritonavir (100 mg) resulted in a 57-fold increase in total exposure to tacrolimus and a 5.8-fold increase in total exposure to cyclosporine [17]. According to the "Clinical Pharmacy Guidelines for Antiviral Treatment of Novel Coronavirus Pneumonia" issued by the Guangdong Pharmaceutical Association of China [16], it is recommended that the daily dose of cyclosporine be reduced by 80% during days 1-5 of paxlovid treatment combined with CNIs; tacrolimus administration should be suspended, and the blood concentration of CNIs should be reevaluated as soon as possible after the completion of paxlovid therapy to guide dose adjustments for restarting treatment. Studies have indicated that when tacrolimus is combined with ritonavir, its inhibitory effect reaches peak levels shortly after exposure, resulting in a sudden increase in tacrolimus trough concentration within 48 hours following ritonavir use [18,19]. After ritonavir is discontinued, its inhibitory effect decreases significantly by 46%-61% within the first 24 hours and by 70%-90% from the second to fifth days; however, it may take up to three weeks for enzyme function to fully recover [20].
The dose of cyclosporine was not adjusted when paxlovid was initiated in Patient 1, leading to increased blood concentrations of cyclosporine and subsequent incidents of high blood pressure, abdominal pain, and other adverse reactions caused by drug overdose. The dose of tacrolimus was appropriately adjusted when paxlovid was initiated in patient 2, resulting in the maintenance of tacrolimus blood concentrations within the therapeutic range even after the discontinuation of paxlovid. Our study further confirmed that paxlovid significantly affects the plasma concentration of CNIs. Previous studies have demonstrated a notable interaction between paxlovidand immunosuppressants in nonpregnant individuals [21,22]. Our study is the first to report that paxlovid can significantly increase the blood concentrations of cyclosporin and tacrolimus in pregnant women. Therefore, when paxlovid is combined with CNIs, the dose should be adjusted over time, and the blood concentration should be closely monitored to avoid drug toxicity.
3. The pharmacogenomic factors influencing CNI therapy
The genomics of cyclosporine strains that have been extensively investigated include the activity of P-glycoprotein in ABCB1 and the metabolic enzyme activity of CYP3A4 [23,24]. Studies conducted by Crettol, Lee et al. have demonstrated that individuals with the AA genotype of ABCB1 exhibit reduced P-glycoprotein activity, leading to increased concentrations of cyclosporine and necessitating lower doses to achieve target drug levels [25,26]. In our study, the patients 1 who received cyclosporine had a low risk for hepatotoxicity due to their AA genotype for CYP3A4*18. Despite a significant increase in cyclosporine concentration during treatment resulting from drug interactions, no hepatotoxicity was observed. The patient's ABCB1 genotype was AA, indicating slow metabolism. Even though they were administered a dose equivalent to 2.2 mg/kg/day (50 mg bid), which did not meet the recommended therapeutic dose for SLE (3-5 mg/kg/day), the concentration reached a stable level of 127.3 ng/ml (>100 ng/ml), indicating that patients with an AA genotype for ABCB1 require lower doses of cyclosporine than those with a normal genotype.
Several studies have reported that the blood concentration of tacrolimus can be influenced by CYP3A5 gene polymorphisms [27,28]. For instance, Provenzani et al. reported that renal transplant patients with the CYP3A5*3 *1/*1 and *1/*3 genotypes required higher doses of tacrolimus to achieve blood drug concentrations comparable to those of patients with the CYP3A5*3 *3/*3 genotype [27]. In our study, the patients in Patient 2 who carried the GG genotype for CYP3A5*3 of tacrolimus were considered slow metabolizers. Therefore, the initial dose of cyclosporine in this patient should be cautiously initiated from the minimum dosage, as it is recommended that the required dosage is lower than that of the normal genotype population[29].
The implementation of TDM has played a crucial role in mitigating the adverse reactions associated with CNIs to some extent. However, it is important to note that TDM exhibits a certain degree of delay and can only be conducted after medication administration. Prior to initiating CNI therapy, it is imperative to employ genetic detection technology to determine the genotypes of CYP3A5*3, CYP3A4*18B, and ABCB1 to ascertain patients' metabolic profiles and predict the initial dosage requirements for CNIs. Subsequently, based on TDM results, individualized medication dosages can be adjusted accordingly to optimize clinical efficacy while minimizing the occurrence of adverse events. The field of pharmacogenomics serves as a prospective tool for determining the initial dosage of medication for individual patients. Given the unique physiological state during pregnancy, genotype polymorphisms and individual variations in drug response, it is crucial to integrate genetic testing with TDM for therapeutic drug monitoring to guide precision medicine specifically tailored for pregnant women.