Pharmacovigilance of Drug–Drug Interactions with Nirmatrelvir/Ritonavir

Abstract

Introduction

Nirmatrelvir/ritonavir (NMV/r) is approved in the United States (US) and more than 70 other countries for the treatment of mild to moderate COVID-19 in nonhospitalized adults at high risk for severe disease. Because ritonavir inhibits several drug metabolizing enzymes, potential drug–drug interactions (DDIs) between ritonavir and concomitant medications are an important consideration for prescribers. Here, we conducted a real-world analysis of data from Pfizer’s global safety database regarding adverse events (AEs) reported during use of NMV/r concomitantly with potentially interacting drugs.

Methods

Data were extracted regarding DDI cases occurring from the start of NMV/r authorization through October 31, 2023. Results regarding concomitant treatment, specific AEs, and clinical outcomes are summarized. Overall NMV/r exposure was estimated based on packs of medication dispensed and was used to calculate reporting rates.

Results

Among 19,617,670 patients exposed globally to NMV/r, 966 cases of potential DDIs were reported. Of these, 594 occurred in the US against an estimated US exposure of 14,646,990 patients, representing a reporting rate of 0.004%. Globally and in the United States, 66.8% and 77.3% of cases, respectively, were nonserious. Simvastatin and tacrolimus were the most frequently reported drugs associated with potential DDIs, and the most frequently reported AE regarding a specific event or symptom was dysgeusia (altered sense of taste), an AE known to be associated with NMV/r.

Conclusions

Low reporting rates of DDIs support the potential for NMV/r treatment to be safely managed with careful use of available drug interaction resources to aid in risk mitigation.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40121-024-01050-w.

Key Summary Points

Why carry out this study?

Because ritonavir has potential drug–drug interactions (DDIs) with concomitant medications, we conducted a real-world analysis of data from Pfizer’s global safety database to determine the rate of DDIs occurring during real-world use.

What was learned from this study?

This study found that of the roughly 19.5 million patients exposed globally to nirmatrelvir/ritonavir (NMV/r) treatment during the study period, there was a considerably low reporting rate (0.005%) for potential DDIs.

The two most frequently reported drugs for potential DDIs were simvastatin and tacrolimus.

The most commonly reported adverse events was dysgeusia (altered sense of taste), which is known to be associated with NMV/r.

The low rates of reporting support the continued use of NMV/r in indicated populations with careful monitoring to reduce risk of any potential DDIs.

Introduction

The high hospitalization and death toll caused by the COVID-19 pandemic in 2020‒2021 necessitated the expedited development and authorization of safe and effective treatments [1, 2]. As a result, the US Food and Drug Administration (FDA) permitted newly developed antivirals to be made available under emergency use authorization (EUA), which is issued based on FDA determination that all known or potential risks of the authorized product are outweighed by its known or potential benefits [3, 4]. On December 22, 2021, nirmatrelvir coadministered with ritonavir (NMV/r; Paxlovid™; Pfizer, New York, NY, USA) became the first oral antiviral COVID-19 treatment to receive EUA [4], representing a critical milestone in advancing the clinical management of COVID-19.

The phase 2/3 Evaluation of Protease Inhibitor for COVID-19 in High-Risk Patients (EPIC-HR) clinical trial demonstrated that high-risk adults with mild to moderate COVID-19 who initiated NMV/r treatment within 5 days of symptom onset had an 86% relative risk reduction in hospitalizations for COVID-19 or all-cause deaths within 28 days compared with placebo [5–8]. Based on these results, NMV/r received EUA in the United States (US) for treatment of mild to moderate COVID-19 in adult and pediatric patients (≥ 12 years of age and ≥ 40 kg) with confirmed SARS-CoV-2 infection who were at high risk for progression to severe disease, including hospitalization or death. In May 2023, NMV/r received full FDA approval for treating mild to moderate COVID-19 in adults aged ≥ 18 years at high risk for progression to severe COVID-19, and use in adolescents continues to be supported under the original EUA [5, 9, 10]. In the European Union, NMV/r was authorized on January 28, 2022, for adults who are at increased risk of progressing to severe COVID-19 and who do not require supplemental oxygen [11]. As of May 2023, NMV/r had been authorized or approved across more than 70 countries [12]. Numerous real-world studies have demonstrated the effectiveness of NMV/r in significantly reducing the likelihood of progression to severe COVID-19, including hospitalization and death, as well as reducing healthcare resource use among high-risk patients irrespective of underlying high-risk condition, vaccination status, or SARS-CoV-2 variant [13–18].

Nirmatrelvir functions as a potent and highly selective inhibitor of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main protease (M^pro) [19]. By inhibiting SARS-CoV-2 M^pro, nirmatrelvir renders the virus incapable of processing polyprotein precursors and therefore prevents viral replication [5]. As NMV/r, nirmatrelvir is coadministered with ritonavir, a cytochrome P450 (CYP) 3A4 inhibitor [20] which serves as a pharmacokinetic enhancer. By inhibiting nirmatrelvir’s CYP3A4-mediated metabolism, ritonavir maintains nirmatrelvir plasma concentrations above the minimum effective concentration required for antiviral activity [5, 11, 19]. In addition to CYP3A4, ritonavir is also an established inhibitor of other enzymes, including P-glycoprotein (P-gp), CYP2D6, and organic anion-transporting polypeptide 1B1 (OATP1B1), and is an inducer of drug-metabolizing enzymes, including CYP3A, CYP1A2, CYP2B6, CYP2C9, CYP2C19, and uridine 5'-diphospho-glucuronosyltransferase (UGT) [21–24]. Ritonavir may therefore have significant drug–drug interactions (DDIs) with drugs that are substrates of these enzymes [22]. The profile of drugs that can potentially interact with ritonavir has been well characterized over decades of ritonavir use, both as an HIV treatment and a pharmacokinetic enhancer [25].

Several DDI studies were conducted as part of the NMV/r clinical development program [5, 26, 27]. Coadministration of midazolam (used as a model CYP3A4 substrate) with NMV/r substantially increased midazolam exposure compared with midazolam alone [1430% for the area under the concentration–time curve (AUC) from time 0 extrapolated to infinity (AUC_inf) and 368% for the maximum plasma concentration (C_max)]; increases were also observed, but to a much lesser extent, for dabigatran, which was used as a model P-gp substrate (194% and 233%, respectively) [5, 26]. Importantly, midazolam and dabigatran exposures were generally similar when coadministered with NMV/r or ritonavir alone, suggesting that increases in exposure were predominantly mediated by ritonavir rather than by nirmatrelvir [26].

In further DDI studies evaluating NMV/r as a victim drug, coadministration of NMV/r with itraconazole, an index CYP3A4 inhibitor, yielded a small increase in nirmatrelvir exposure (AUC from time 0 to the end of a 12-h dosing interval, 139%; C_max, 119%) [5, 27]. This suggests that other coadministered CYP3A4 inhibitors similar to itraconazole may not exert additional inhibitory effects beyond those mediated by ritonavir. However, coadministration with the index CYP3A4 inducer carbamazepine led to a substantial decrease in nirmatrelvir exposure (AUC_inf, 45%; C_max, 57%) [5, 27] highlighting the DDI risk of coadministering NMV/r with strong CYP3A4 inducers. In summary, results from the DDI studies suggest that clinical drug interactions of NMV/r are primarily attributable to inhibition of CYP3A4 metabolism by ritonavir, and, to a lesser extent, ritonavir-mediated inhibition of P-gp and OATP1B1. Additionally, coadministration of NMV/r with strong CYP3A4 inducers reduces nirmatrelvir concentrations [5, 27], which may affect efficacy.

Here, we present real-world data from Pfizer’s global safety database (SDB), a pharmacovigilance system that collects information from numerous sources including published literature, nonclinical study programs, clinical studies, business partners, healthcare providers (HCPs), patients, and health authorities [28]. This system allows for potential new safety signals to emerge that can then be appropriately reviewed and mitigated through communications with regulatory agencies and continuing information dissemination for HCPs [28]. For the current study, we specifically collected information regarding cases of potential interactions between NMV/r and contraindicated drugs or other interacting drugs designated as “can be used with caution.” The goal of this report is to expand upon previous analyses and references regarding the potential for DDIs by investigating DDIs occurring in real-world use. In turn, we aim to raise awareness among HCPs of NMV/r DDIs, including specific drug classes that pose a safety risk, to support the safe prescribing and administration of NMV/r.

Methods

Data Sources and Search Strategy

Data were extracted from Pfizer’s global SDB, which includes information regarding adverse events (AEs) occurring in association with Pfizer product use [28]. A cumulative review of the SDB was performed to include cases reported through October 31, 2023. Included cases were all valid post-marketing NMV/r cases that reported any drug listed in (1) Sections 4.3 (contraindications) and 4.5 (drug interactions) of Pfizer’s NMV/r Core Data Sheet (CDS), for global cases, which included US cases, or (2) Sections 4 (contraindications) and 7 (drug interactions) of the 2023 US prescribing information [5] for US cases only, as a co-suspect or concomitant medication. We note that the drugs listed in both the CDS and the US prescribing information were identical. Cases also had to include any of the following Medical Dictionary for Regulatory Activities (MedDRA) preferred terms: “contraindicated product administered,” “contraindicated product prescribed,” “drug interaction,” “inhibitory drug interaction,” “labelled drug–drug interaction issue,” “labelled drug–drug interaction medication error,” or “potentiating drug interaction.”

Ethics committee approval was not required. This article is based on a retrospective database review and does not contain any new studies with human participants or animals performed by any of the authors.

Data Analysis

For analysis, drugs were categorized as “contraindicated” if they were listed in Section 4.3 of the NMV/r CDS (global cases) or Section 4 of the US prescribing information (US cases), whereas drugs were considered to be “drugs that can be used with caution” if they were listed only in Sections 4.5 (global) or 7 (US) of these documents, respectively. Co-suspect drugs were defined as any medications taken in addition to NMV/r and reported as being possibly associated with the AE, whereas concomitant drugs were defined as any medications used or given at, or almost at, the same time as NMV/r.

Specific AEs and serious AEs (SAEs) co-reported with cases involving contraindicated drugs or drugs that can be used with caution were summarized. For this analysis, AEs were defined as “any untoward medical occurrence in a patient administered a pharmaceutical product,” per International Conference on Harmonization (ICH) Good Clinical Practice (GCP) guidelines. SAEs, also defined according to ICH GCP guidelines, were any untoward medical occurrences or serious medically important event that resulted in death or were life-threatening, required hospitalization or prolonged existing hospitalization, resulted in persistent or significant disability/incapacity, or were a congenital anomaly or birth defect; all other AEs were categorized as nonserious AEs. AEs and SAEs were categorized using MedDRA preferred terms (version 26.1).

Estimates of Patient Exposure

For US-specific data, cumulative patient exposure to NMV/r was estimated based on Pfizer wholesale shipment data and availability at US points of contact (POCs) according to the COVID-19 Public Therapeutics Locator [29]. The difference between numbers of wholesale shipment packs and packs remaining at POCs was presumed to be the number of packs dispensed, and an estimated weekly growth percentage by US state was applied to this value to better estimate the actual number of patients exposed.

Cumulative exposure to NMV/r in countries outside of the US was estimated using available sales data from IQVIA Health Prescribing Insights Medical. Data were extracted through June 30, 2023, and exposure from July 1 through October 31, 2023, was estimated by extrapolating the average value from the previous 4 quarters. Number of patients was estimated by assuming that each patient used the medication as directed (2 nirmatrelvir tablets and 1 ritonavir tablet twice daily for 5 days).

Results

Overview of Cases and Exposure

From the first temporary authorization for emergency supply up to October 30, 2023, the US wholesaler shipped 17,771,654 packs of NMV/r to US POCs. On October 30, 2023, a total of 1,304,445 packs remained available across all POCs, indicating that 16,467,209 packs had been dispensed to US patients. By applying a weekly growth percentage, estimated actual exposure was calculated to be approximately 14,646,990 patients as of the end of October 2023. Global exposure during the same period was estimated to be 19,617,670 patients.

As of October 31, 2023, a total of 966 global cases included in the SDB for NMV/r reported drug interactions with contraindicated drugs or other interacting drugs that can be used with caution, representing a reporting rate of 0.005% globally. Of these, 594 cases occurred in the US, representing a US reporting rate of 0.004%. By month, reporting rate peaked at 0.011% in June 2022 and thereafter remained ≤ 0.005% (Online Resource 1).

Demographic and clinical characteristics of global and US cases are detailed in Table 1. The majority of global cases were nonserious (n = 645; 66.8%); of the 321 serious cases (33.2%), 169 involved hospitalizations (52.6%). Of the 594 US cases, 459 were nonserious (77.3%); of the 135 serious cases (22.7%), 76 involved hospitalizations (56.3%).

Table 1.

Demographic and clinical characteristics of global and US cases

	Globala		United States
	Cases, n (n = 966)	Percentage	Cases, n (n = 594)	Percentage
Sex
Male	443	45.9	292	49.2
Female	397	41.1	251	42.3
No data	126	13.0	51	8.6
Age range, years
≤ 17	10	1.0	5	0.8
18‒30	17	1.8	10	1.7
31‒50	97	10.0	47	7.9
51‒64	158	16.4	110	18.5
65‒74	249	25.8	171	28.8
≥ 75	217	22.5	136	22.9
Unknown	218	22.6	115	19.4
Case outcome
Fatal	14	1.4	6	1.0
Not recovered/not resolved	137	14.2	107	18.0
Recovered/resolved	225	23.3	121	20.4
Recovered/resolved with sequelae	1	0.1	0	0
Recovering/resolving	147	15.2	95	16.0
Unknown	442	45.8	265	44.6
Country of occurrence
United States	594	61.5	594	100
Japan	86	8.9	NA	NA
France	83	8.6	NA	NA
Canada	24	2.5	NA	NA
China	23	2.4	NA	NA
Germany	21	2.2	NA	NA
United Kingdom	21	2.2	NA	NA
Italy	14	1.4	NA	NA
Australia	10	1.0	NA	NA
Austria	9	0.9	NA	NA
Source
Clinical study	1	0.1	0	0
Literature, nonstudy	83	8.6	46	7.7
Literature, study	27	2.8	2	0.3
Solicited	12	1.2	10	1.7
Spontaneous	843	87.3	536	90.2
Case seriousness
Serious	321	33.2	135	22.7
Fatal	14	1.4	6	1.0
Hospitalizationb	169	17.5	76	12.8
Nonserious	645	66.8	459	77.3

NA not applicable

^aIncludes US cases

^bIncludes cases with a fatal outcome

Most Commonly Reported Drugs Across all Cases

The most commonly implicated contraindicated drugs or drugs that can be used with caution, along with additional case details, are summarized in Table 2. Globally, simvastatin was the most commonly reported contraindicated drug (234 cases), followed by alfuzosin (45 cases) and flecainide (32 cases). Tacrolimus was the most commonly reported drug that can be used with caution (212 cases), followed by atorvastatin (75 cases) and amlodipine (72 cases). Results in the US were similar regarding contraindicated drugs, with simvastatin (195 cases) and alfuzosin (36 cases) being the most commonly reported; in the US, however, the third most commonly reported contraindicated drug was lovastatin (31 cases). Similar to global data, most commonly reported drugs that can be used with caution in the US were tacrolimus (67 cases), atorvastatin (45 cases), and amlodipine (38 cases). For most of these drugs, fewer than half of cases were serious; however, 66.5% (141/212) of global cases and 76.1% (51/67) of US cases associated with tacrolimus were considered serious.

Table 2.

Cases concerning most commonly contraindicated drugs or drugs to be used with caution

Drug	Co-suspect	Concomitant	Total	Serious	Nonserious
Globala
Contraindicated drugs
Simvastatin	143	91	234	25	209
Alfuzosin	31b	15	45	5	40
Flecainide	27	5	32	1	20
Lovastatin	18	13	31	1	30
Amiodarone	24	6	30	9	21
Drugs to be used with caution
Tacrolimus	205	11	212c	141	71
Atorvastatin	24	52	75d	32	43
Amlodipine	23	50	72e	32	40
Rosuvastatin	10	26	36	14	22
Apixaban	16	21	37	11	26
United States
Contraindicated drugs
Simvastatin	125	70	195	17	178
Alfuzosin	25b	11	36	2	34
Lovastatin	18	13	31	1	30
Flecainide	16	5	21	1	20
Carbamazepine	9	7	16	2	14
Drugs to be used with caution
Tacrolimus	64	3	67	51	16
Atorvastatin	18	28	45d	13	32
Amlodipine	10	29	38e	9	29
Rosuvastatin	6	21	27	9	18
Apixaban	13	11	24	7	17
Bupropion	4	20	24	3	21

^aIncludes US cases

^bIn 1 case, alfuzosin and alfuzosin hydrochloride were reported as suspect drugs

^cIn 4 cases, tacrolimus or tacrolimus monohydrate was reported as both a suspect and concomitant drug

^dIn 1 case, atorvastatin was reported as both co-suspect (atorvastatin calcium) and concomitant (atorvastatin)

^eIn 1 case, amlodipine besylate was reported as both co-suspect and concomitant

Most Frequently Reported Adverse Events

The most frequently co-reported AEs are described overall and by top drugs among US cases involving contraindicated drugs or drugs that can be used with caution in Tables 3 and 4, respectively. Dysgeusia (an altered sense of taste, such as metallic or bitter taste) was the most frequently co-reported AE associated with contraindicated drugs [n = 33 (8.4%)] and drugs that can be used with caution (n = 30 (9.2%)] with the exception of AEs concerning potential DDIs (contraindicated product administered, contraindicated product prescribed, drug interaction, labelled drug–drug interaction medical error), COVID-19, or disease recurrence. For contraindicated drugs, the other most frequently co-reported AEs overall were diarrhea [n = 26 (6.6%)] and cough [n = 18 (4.6%)]. For drugs that can be used with caution, the other most commonly co-reported AEs overall were acute kidney injury [n = 28 (8.6%)], toxicity to various agents [n = 28 (8.6%)], diarrhea [n = 25 (7.7%)], and nausea [n = 25 (7.7%)]. Examination of co-reported AEs by top contraindicated drugs (Table 3) and drugs that can be used with caution (Table 4) indicated that AEs were generally consistent with the known safety profile of NMV/r or with adverse drug reactions reported for the interacting drugs [5, 30–39]. Specifically regarding tacrolimus DDIs in the US, the number of patients concomitantly using both NMV/r and tacrolimus ranged from 55 to 1260 per month from January 2022 through October 2023; nevertheless, the monthly DDI reporting rates remained < 2% during this period and were < 1% for 14 of the 22 months (Online Resource 2).

Table 3.

Co-reported AEs among US cases involving contraindicated drugs^a

System organ class	Preferred term	Cases, n (%)
		Overall (n = 392)	Simvastatin (n = 195)	Alfuzosin (n = 36)	Lovastatin (n = 31)	Flecainide (n = 21)	Carbamazepine (n = 16)
Injury, poisoning, and procedural complications	Contraindicated product administered	350 (89.3)	183 (93.9)	30 (83.3)	29 (93.6)	15 (71.4)	15 (93.8)
	Contraindicated product prescribed	33 (8.4)	9 (4.6)	5 (13.9)	–	6 (28.6)	–
	Off-label use	11 (2.8)	–	–	–	–	–
	Labelled drug–drug interaction medication error	11 (2.8)	5 (2.6)	–	–	–	–
Infections and infestations	COVID-19b	138 (35.2)	87 (44.6)	12 (33.3)	14 (45.2)	–	–
General disorders and administration site conditions	Disease recurrence	133 (33.9)	84 (43.1)	12 (33.3)	14 (45.2)	–	–
	Drug interaction	20 (5.1)	8 (4.1)	–	–	–	–
	Fatigue	17 (4.3)	8 (4.1)	–	–	–	–
	Pyrexia	10 (2.6)	6 (3.1)	–	–	–	–
Nervous system disorders	Dysgeusia	33 (8.4)	22 (11.3)	–	–	–	–
	Dizziness	13 (3.3)	5 (2.6)	–		–	–
	Headache	15 (3.8)	8 (4.1)	–	–	–	–
Gastrointestinal disorders	Diarrhea	26 (6.6)	15 (7.7)	–	–	–	–
	Nausea	12 (3.1)	–	–	–	–	–
	Vomiting	8 (2.0)	6 (3.1)	–	–	–	–
Respiratory, thoracic, and mediastinal disorders	Cough	18 (4.6)	8 (4.1)	–	–	–	–
	Rhinorrhea	9 (2.3)	5 (2.6)	–	–	–	–
	Nasal congestion	9 (2.3)	6 (3.1)	–	–	–	–
	Oropharyngeal pain	8 (2.0)	7 (3.6)	–	–	–	–
Cardiac disorders	Bradycardia	8 (2.0)	–	–	–	–	–

AE adverse event

^aData are shown for all AEs occurring in ≥ 5 total cases

^bOf the 138 AEs coded under the preferred term COVID-19, 129 were coded to lowest level terms indicative of COVID-19 rebound (COVID-19 recurrent, n = 126; COVID-19 reinfection, n = 3)

Table 4.

Co-reported AEs among US cases involving drugs that can be used with caution

System organ class	Preferred term	Cases, n (%)
		Overall (n = 327)	Tacrolimus (n = 67)	Atorvastatin (n = 45)	Amlodipine (n = 38)	Rosuvastatin (n = 27)	Apixaban (n = 24)	Bupropion (n = 24)
General disorders and administration site conditions	Drug interaction	171 (52.3)	62 (92.5)	23 (51.1)	9 (23.7)	11 (40.7)	10 (41.7)	–
	Disease recurrence	66 (20.2)	–	7 (15.6)	18 (47.4)	10 (37.0)	–	12 (50.0)
	Fatigue	16 (4.9)	–	–	–	–	–	–
	Asthenia	11 (3.4)	–	–	–	–	–	–
	Feeling abnormal	9 (2.8)	–	–	–	–	–	–
	Malaise	8 (2.5)	–	–	–	–	–	–
	Pain	7 (2.1)	–	–	–	–	–	–
Injury, poisoning, and procedural complications	Contraindicated product administered	122 (37.3)	–	17 (37.8)	24 (63.2)	14 (51.9)	9 (37.5)	19 (79.2)
	Labelled drug–drug interaction medication error	42 (12.8)	11 (16.4)	6 (13.3)	–	–	–	–
	Toxicity to various agents	28 (8.6)	28 (41.8)	–	–	–	–	–
	Contraindicated product prescribed	8 (2.5)	–	–	–	–	–	–
	Off-label use	7 (2.1)	–	–	–	–	–	–
Infections and infestations	COVID-19b	73 (22.3)	–	9 (20.0)	19 (50.0)	10 (37.0)	8 (33.3)	12 (50.0)
Nervous system disorders	Dysgeusia	30 (9.2)	–	6 (13.3)	–	6 (22.2)	–	6 (25.0)
	Headache	14 (4.3)	–	–	–	–	–	–
	Dizziness	12 (3.7)	–	–	–	–	–	–
	Taste disorder	7 (2.1)		–	–	–	–	–
Renal and urinary disorders	Acute kidney injury	28 (8.6)	25 (37.3)	–	–	–	–	–
Gastrointestinal disorders	Diarrhea	25 (7.7)	5 (7.5)	–	–	–	–	–
	Nausea	25 (7.7)	8 (11.9)	–	–	–	–	–
	Vomiting	16 (4.9)	6 (9.0)	–	–	–	–	–
Investigations	Immunosuppressant drug level increased	20 (6.1)	20 (29.9)	–	–	–	–	–
	Drug level increased	18 (5.5)	15 (22.4)	–	–	–	–	–
	Blood creatinine increased	10 (3.1)	10 (14.9)	–	–	–	–	–
	Blood pressure increased	7 (2.1)	–	–	–	–	–	–
	Drug level above therapeutic	7 (2.1)	7 (10.5)	–	–	–	–	–
Vascular disorders	Hypotension	12 (3.7)	–	–	–	–	–	–
	Hypertension	7 (2.1)	–	–	–	–	–	–
Cardiac disorders	Bradycardia	9 (2.8)	–	–	–	–	–	–
Respiratory, thoracic, and mediastinal disorders	Cough	8 (2.5)	–	–	–	–	–	–

AE adverse event

^aData are shown for all AEs occurring in ≥ 5 total cases

^bOf the 73 AEs coded under the preferred term COVID-19, 62 were coded to lowest level terms indicative of COVID-19 rebound (COVID-19 recurrent, n = 61; COVID-19 reinfection, n = 1)

Fatal Cases and Associated Drugs

Globally, 14 of 966 reported cases (1.4%) were associated with a fatal outcome, including 6 of the 594 cases (1.0%) that occurred in the US. In 8 of the 14 cases, the drug interaction was specifically reported as fatal, although in all cases these interactions occurred within a context of multiple confounding factors including underlying comorbidities, additional concomitant medications, COVID-19 infection, or indeterminate factors due to insufficient information. Many of the 8 cases were associated with ≥ 1 fatal event, but no single event was listed in ≥ 1 case with the exception of events describing the drug interaction itself (e.g., “labelled drug–drug interaction error”). Other fatal events included acute kidney injury, asthenia, dyspnoea, fall, gait disturbance, respiratory distress, respiratory failure, hemorrhage, seizure, COVID-19, drug ineffective, shock, multiple organ dysfunction syndrome, toxicity to various agents, cardiogenic shock, lactic acidosis, bradycardia, overdose, ejection fraction decreased, right ventricular dysfunction, and hypotension.

Three of the 14 cases involved patients receiving contraindicated drugs while taking NMV/r (simvastatin, n = 1; dronedarone hydrochloride, n = 1; primidone, n = 1). Simvastatin and dronedarone were reported as suspect drugs, whereas primidone was reported as concomitant. Regarding drugs that can be used with caution, suspect drugs associated with fatal cases included tacrolimus (n = 5), apixaban (n = 1), atorvastatin calcium (n = 1), nifedipine (n = 2), and verapamil hydrochloride (n = 1); some cases were associated with > 1 suspect drug. Other drugs reported as suspect included edoxaban tosylate, heparin, prednisolone acetate, sacubitril, valsartan, baricitinib, and methotrexate sodium (n = 1 each); all such cases also involved ≥ 1 contraindicated drug or drugs that can be used with caution that was reported as co-suspect or concomitant.

Discussion

A total of 966 global cases were reported in the SDB regarding drug interactions between NMV/r and either contraindicated drugs or drugs that can be used with caution. Against the backdrop of an estimated 19,617,670 patients exposed to NMV/r before data cutoff, reporting rates were considerably low (0.005%). Of all reported cases, 66.8% were nonserious. The most frequently reported contraindicated drug was simvastatin, and the most frequently reported drug that can be used with caution was tacrolimus. Across all drugs, the most frequently reported types of AEs were altered sense of taste, an AE known to be associated with NMV/r, and gastrointestinal issues, such as diarrhea, which are associated with COVID-19 as well as with ritonavir-boosted protease inhibitors as a class [5, 40–42]. Other observed AEs were consistent with the known safety profiles of each concomitant drug [5, 30–39]. Fourteen reported interaction events were associated with fatal outcomes.

Consistent with the results reported here, NMV/r prescribing information notes that calcineurin inhibitors such as tacrolimus were the most commonly reported concomitant medications resulting in serious adverse reactions [5]. Tacrolimus, a drug used to prevent organ rejection among liver, kidney, and heart transplant recipients, is primarily metabolized by CYP3A4 [37]. Thus, when coadministered along with CYP3A4 inhibitors such as ritonavir, increased concentrations are observed and must be carefully monitored [5, 37]. Because tacrolimus has a narrow therapeutic window, maintaining tacrolimus concentrations within the appropriate range through careful drug monitoring is critical to prevent complications including neurotoxicity, nephrotoxicity, infections, malignancies, diabetes, gastrointestinal symptoms, elevated serum creatinine levels, and acute kidney injury [43]. Multiple case reports in the literature have demonstrated drastic increases in tacrolimus concentrations, often leading to serious AEs, among patients who have undergone organ transplants and received concomitant NMV/r administration [43–49]. Because of the potential for serious complications, the NMV/r label advises to avoid concomitant treatment with tacrolimus or other calcineurin inhibitors unless concentrations can be closely monitored and consistently adjusted as needed; expert consultation is also advised to appropriately manage proper dosing and potential AEs. The importance of these recommendations is once again underscored by results from the current study.

Several resources are available to assist HCPs with prescribing NMV/r and to assist pharmacists with their systems of identifying medications that may interact to safely manage their patients. These include the US and EU labels for NMV/r (each listing > 35 drug classes and > 150 drug examples that are likely to cause DDIs and including a boxed warning) [5, 11], two letters disseminated to HCPs in 2022 with additional information regarding DDIs [50, 51], a website maintained by the US National Institutes of Health that lists medications without clinically significant DDIs as well as additional resources [52], the Pfizer-hosted interactive Drug Interaction Checker (accessible on the NMV/r medical information website) [53], the Drug Interaction Resource describing potentially significant DDIs with NMV/r [54], and a 2023 Pfizer publication of results from a real-world evidence study assessing the potential for DDIs with NMV/r across the top 100 most prescribed drugs among patients at high risk of developing severe COVID-19 complications [55]. The study identified low potential for clinically meaningful DDIs associated with 70 of the top 100 drugs; the remaining 30 most commonly prescribed drugs were primarily corticosteroids, narcotic analgesics, HMG Co-A reductase inhibitors (statins), and sedative hypnotics [55]. In addition to resources specifically related to NMV/r, more general resources available to HCPs include the US FDA’s Patient Eligibility Screening Checklist Tool [56], and an extensive drug interaction checker maintained by the University of Liverpool that is accessible as a website or smartphone application [57]. The analysis presented here supplements these DDI resources, which focus largely on potential DDI risk across a broad range of drugs, by highlighting which of these drugs specifically are associated with actual DDIs in clinical practice. Although this analysis did not identify any new safety information, the sponsor will continue to monitor the risk of NMV/r DDIs as a component of routine pharmacovigilance activities, with the potential to add specific recommendations for individual populations beyond what is currently included in the label.

This study provides the first comprehensive evaluation of DDIs reported after prescribing NMV/r in real-world clinical practice. However, it did have some important limitations. The first of these is in regard to estimates of global patient exposure retrieved from IQVIA. IQVIA does not include data from all countries where NMV/r is dispensed, particularly those countries without sizeable sales. IQVIA also does not capture retail sales and hospital dispensing data in all countries. Together, these limitations likely led to a substantial underestimate of the actual distributed product. Furthermore, data were retrieved from the Pfizer Global SDB, which includes voluntarily submitted cases and therefore is subject to an unknown magnitude of underreporting. Because several external factors can influence the likelihood of AE reporting, incidence rates cannot be accurately calculated. The reporting rates calculated in our study are not analogous to incidence rates, but they provide an important measure of observed trends in DDIs, and periodic reviews of these rates over time can provide useful insight into the stability of incident reporting. Additionally, cases were retrieved from the database only if they involved drugs currently listed in the NMV/r prescribing information as either contraindicated drugs or drugs that can be used with caution. Because drugs listed in the prescribing information are not comprehensive, this may not represent an exhaustive list of drugs associated with potential DDIs. Cases were also only retrieved if they were coded as an adverse event, and our data therefore did not capture the potential for loss of effectiveness (of either NMV/r or the coadministered medication) as a result of concomitant administration. Finally, it is critical to note that, in many instances, the case outcome was not necessarily a direct result of the DDI but instead could be attributed to multiple confounding factors, including underlying comorbidities, concomitant medications, and COVID-19 infection; furthermore, contributing factors may not have been completely elucidated for all cases because of missing information. These factors, particularly disease state and inflammation, can potentially alter CYP3A expression [58–60], which may affect the likelihood and severity of DDIs between NMV/r and concomitant medications.

Conclusion

In this review of relevant cases in Pfizer’s Global SDB, the reporting rate of drug interactions globally was low (0.005%) and primarily represented by nonserious events consistent with the known safety profile of NMV/r [5]. Overall, 33.2% of global cases and 22.7% of US cases were serious, including 1.4% and 1.0% with fatal outcomes, respectively, and 17.5% and 12.8% involving hospitalization. Importantly, the case level outcome may have been due to an event not related to the drug interaction. In addition, it is important to note that NMV/r is administered over 5 days [5]; thus, while it is well known that ritonavir may interact with medications that are dependent on CYP3A4 for drug clearance, the short duration of use aids in the risk mitigation regarding NMV/r interactions compared with other uses of ritonavir.

As the recommended first-line treatment for nonhospitalized adults with mild to moderate COVID-19 who are at high risk of progression to severe disease [61], NMV/r continues to represent a critical tool in reducing COVID-19 morbidity and mortality worldwide. The most important risk to weigh against the significant benefit of NMV/r treatment is potential DDIs, but DDI risk can be effectively mitigated by appropriate patient selection and management of concomitant medications. To enhance awareness of potential DDIs and to aid in the proper management of concomitantly administered drugs that may incur increased concentrations due to ritonavir, Pfizer has developed and implemented a robust program of risk mitigation activities and education that supports prescribers and patients in the safe use of NMV/r.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Victoria Hendrick, Erast Pohorylo, Lubna Merchant, Jackie Gerhart, Iqra Nz Arham, Florin Draica, Romina Quercia, Ayman Ayoub, and Reeme Mehta all contributed sufficiently to the manuscript (concept and planning of the work described; acquisition, analysis and interpretation of the data; drafting and/or critical revision of the manuscript; and approved the final submitted version of the manuscript) and, therefore, share collective responsibility and accountability for the manuscript.

Funding

This study, and the journal’s publication fees, was funded by Pfizer Inc.

Data Availability

Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions, and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.

Declarations

Conflict of Interest

All named authors are employees of Pfizer Inc and may hold stock or stock options.

Ethical Approval

Ethics committee approval was not required. This article is based on a retrospective database review and does not contain any new studies with human participants or animals performed by any of the authors.