Non-HIV Antivirals in FAERS: 508,024 Adverse-Event Reports by the Numbers

The FDA Adverse Event Reporting System (FAERS) contains 508,024 individual safety reports that name at least one non-HIV antiviral agent—spanning anti-herpes/cytomegalovirus (CMV) therapies, anti-influenza treatments, hepatitis C virus (HCV) direct-acting antivirals (DAAs), COVID-19 agents, and hepatitis B virus (HBV) agents—in any reported role (suspect, concomitant, or interacting). Filed between 1998 and mid-2026, this cohort represents the primary safety registry for non-HIV antiviral pharmacology in the real world. Of these 508,024 reports, 312,149 (61.4 percent) carry an FDA seriousness flag, and 46,956 reports (9.2 percent) record a fatal outcome. Within the database, the peak reporting year was 2022 (55,293 reports), driven by the mass distribution of oral COVID-19 antivirals like Paxlovid (nirmatrelvir/ritonavir) under Emergency Use Authorization and subsequent full approval, superimposed on the historical baseline of the high-volume HCV DAA launch era.

This article provides a class-level, descriptive pharmacovigilance analysis of non-HIV antivirals in FAERS. It is tailored for safety professionals, transplant and oncology pharmacists, infectious-disease reviewers, and formulary managers who evaluate real-world safety data, drug-interaction profiles, and cost-containment frameworks. Every statistic presented is computed from the public openFDA FAERS extract containing 20,328,575 total reports with an export date of June 8, 2026.

Before diving into the numbers, it is critical to state the standard pharmacovigilance caveats: FAERS is a spontaneous-reporting system that collects suspected associations. It does not establish clinical causality, and it lacks a population-exposure denominator. Many of these agents are administered as prophylaxis to highly vulnerable, immunocompromised cohorts (such as solid organ transplant recipients, hematopoietic stem cell transplant patients, and oncology patients) or to hospitalized patients with severe acute infections. Consequently, the high rates of seriousness and death reflect the underlying medical fragility and comorbidities of these patient populations rather than the intrinsic lethality of the antiviral medications themselves. For an overview of the registry's broader structure and methods, see the analysis of 20 million FAERS reports; for the complementary half of the antiviral landscape, see the HIV antiretroviral FAERS analysis.

Methodology, in one paragraph

A report is included in the non-HIV antiviral cohort if the generic or brand name of an approved non-HIV antiviral agent appears in the report’s drug substances or brand names list, using a case-insensitive substring match. The search covers meganames and combination brands to ensure complete capture. To prevent double-counting and maintain cohort hygiene, all dedicated HIV antiretrovirals (such as dolutegravir, efavirenz, tenofovir alafenamide, and lamivudine) are excluded, except when ritonavir is co-mentioned with nirmatrelvir (the active component of Paxlovid). Tenofovir disoproxil fumarate (TDF) and lamivudine are excluded entirely because their dual-use footprint in HIV and HBV cannot be separated at the database level; these agents are assigned to the HIV antiretroviral post. "Serious" status corresponds to the FDA-coded seriousness flags (hospitalization, life-threatening, disabling, or other medically significant events). Outcome categories represent the report-level resolution fields, which are separate from seriousness flags and may sum to more than 100 percent because a single case can record multiple clinical resolutions. MedDRA Preferred Terms (PTs) are grouped into clinically relevant adverse-event (AE) families to capture class-wide toxicities.

How many FAERS reports involve non-HIV antivirals, and how serious are they?

Across the 20,328,575 reports in the openFDA database, non-HIV antivirals are cited in 508,024 cases. The overall seriousness rate of 61.4 percent is relatively high compared to standard primary care drug classes, reflecting the clinical contexts in which these drugs are prescribed. A total of 312,149 reports are flagged as serious, and 46,956 reports carry a death flag.

The table below breaks down the report-level clinical resolutions. Because a report can carry multiple outcome flags (e.g., a patient who was hospitalized and subsequently recovered), these categories overlap.

The discrepancy between the 46,956 death-flagged reports and the 36,258 Fatal outcome records arises because the seriousness flag represents the reporter's initial severity grading, whereas the outcome field is a separate, structured resolution flag that is frequently left blank or updated later. Hospitalization is flagged in 184,337 reports (36.3 percent of the cohort), representing a massive clinical and economic burden. Clinicians should interpret this high hospitalization share as a reflection of the acute infectious states (like severe influenza or COVID-19) or the critical prophylaxis windows (post-transplant CMV prevention) where these drugs are utilized.

Which antiviral agents and subclasses drive the report volume?

The non-HIV antiviral class is highly heterogeneous, divided into several key therapeutic subclasses. Herpesvirus and CMV therapies represent the largest share of historical reporting, followed by HCV direct-acting antivirals and ribavirin. COVID-19 therapeutics have accumulated a massive volume in a short period.

The table below aggregates report volume, seriousness rates, and death-flag rates by subclass. (Note: Reports may name multiple antivirals across subclasses, so subclass totals do not sum to the deduplicated cohort union).

To understand these distributions, safety teams must look at the specific substances driving the volume. The table below lists the top 16 individual substances by reporting volume in any role.

Crucial Data Note on Acyclovir Overlap: Acyclovir (111,729 reports) and acyclovir sodium (103,674 reports) represent overlapping entries. Because reporters write brand names or generic variations, the same patient case often names both strings or gets mapped to both. Clinicians must never add these two numbers together to claim a simple total: doing so would double-count every case that lists both the free-acid and the sodium-salt formulation. The figures above are per-string counts, not mutually exclusive cohorts. This string-overlap phenomenon is a known characteristic of the openFDA system and also affects valganciclovir and amantadine.

What are the signature adverse-event signals, and how do they map to FDA boxed warnings?

When we filter out non-specific systemic symptoms such as fatigue (41,521 reports), nausea (28,432 reports), and headache (26,663 reports), the remaining clinical reports align closely with the established toxicities in the FDA-approved labels.

The table below maps the structured MedDRA Preferred Term (PT) groupings (adverse-event families) to their corresponding reporting volume and the primary drugs or classes responsible for the signal.

Analyzing the Nephrotoxicity Signal

The 24,758 nephrotoxicity reports are a major point of interest for hospital pharmacists. Intravenous acyclovir sodium is known to precipitate in the renal tubules, leading to obstructive nephropathy. The FAERS profile confirms that "acute kidney injury" and "blood creatinine increased" are highly prevalent. Furthermore, foscarnet and cidofovir—used for CMV retinitis in HIV patients or acyclovir-resistant herpes—have severe, dose-limiting nephrotoxicity boxed warnings that restrict their use to secondary lines of therapy.

Analyzing the Neuropsychiatric Signal

The 31,768 neuropsychiatric reports highlight another significant safety concern. Oseltamivir (Tamiflu) prescribing information contains a warning regarding postmarketing neuropsychiatric events (hallucinations, delirium, abnormal behavior), particularly in pediatric patients. This signal is prominent in the influenza subclass (46,648 total reports). Additionally, high-dose acyclovir or valganciclovir can cause confusion, tremors, and encephalopathy, particularly in patients with pre-existing renal impairment who fail to receive proper dose adjustments.

Analyzing the Hematologic Signal

With 49,911 reports, hematologic toxicity is the largest non-infectious AE family. Ribavirin is notorious for causing dose-dependent hemolytic anemia, which can worsen cardiac disease and lead to myocardial infarction (resulting in a prominent boxed warning). Ganciclovir and its oral prodrug valganciclovir are bone marrow suppressants, carrying boxed warnings for severe granulocytopenia, thrombocytopenia, and aplastic anemia. This requires frequent complete blood count (CBC) monitoring.

Why is ritonavir excluded, and how is the Paxlovid and HIV-contamination gotcha handled?

A common pitfall in analyzing antiviral safety data is the "ritonavir gotcha." Ritonavir is a potent CYP3A inhibitor. Historically, it was developed as an HIV protease inhibitor, but today it is primarily used at low doses (100 mg) as a pharmacokinetic "booster" to increase the exposure of other protease inhibitors, such as lopinavir or darunavir. Crucially, ritonavir is also co-packaged with nirmatrelvir to form Paxlovid, the standard oral therapy for mild-to-moderate COVID-19 in high-risk patients.

Because ritonavir is used in both HIV and COVID-19 regimens, including all ritonavir reports in a non-HIV antiviral cohort would cause significant contamination. Of the 508,024 reports in this cohort, 58,140 (11.4 percent) mention ritonavir. However, 66,213 reports (13.0 percent) also contain at least one HIV-specific antiretroviral drug.

To address this, our analysis applies a strict filter:

1. Ritonavir is excluded from the cohort definition as a primary search term.
2. Reports containing ritonavir are allowed into the cohort only when they also name nirmatrelvir (Paxlovid).
3. If a report names ritonavir alongside HIV drugs (such as dolutegravir or tenofovir) but does not name nirmatrelvir, it is excluded.

This logic ensures that Paxlovid's safety profile is fully captured via nirmatrelvir (71,095 COVID subclass reports) without allowing HIV therapy reports to inflate the non-HIV antiviral cohort. The residual non-ritonavir HIV co-mention rate of approximately 1.6 percent represents patients with HIV who were treated for non-HIV viral infections, such as shingles (acyclovir) or hepatitis C (sofosbuvir). This confirms that the cohort is clean and free of HIV therapy contamination.

How does the 9.2% death share reflect population rather than intrinsic lethality?

The overall cohort death-flag rate of 9.2 percent (46,956 reports) is high. For example, it is higher than the death-flag rates seen in cardiovascular or endocrinology drug classes. However, this figure is heavily influenced by the patient populations receiving these drugs.

To understand this confounding factor, safety teams must look at the reporter mix and the top reported indications. The reporter mix for the cohort is detailed in the table below:

Unlike many other classes where consumer self-reports make up over 60 percent of the volume, the non-HIV antiviral cohort is heavily driven by healthcare professionals (over 62 percent combined). Healthcare professionals are more likely to report severe, life-threatening, or fatal events. The lawyer reporting share is very low (0.3 percent), confirming that these reports represent clinical cases rather than litigation-driven reporting.

Furthermore, the indication profile explains the high mortality. The table below lists the top indications cited for these drugs:

Consider these patient profiles:

• Oncology and Transplant Patients: Multiple myeloma (65,539 reports) and AML (7,310 reports) are leading indications because patients undergoing chemotherapy or stem cell transplants receive mandatory acyclovir or valganciclovir prophylaxis to prevent herpes zoster or CMV reactivation. These patients have high baseline mortality rates due to their primary disease and opportunistic infections.
• Severe COVID-19: Remdesivir (Veklury) is administered intravenously to hospitalized patients with severe COVID-19 who require supplemental oxygen. Paxlovid is prescribed to outpatients at high risk of progressing to severe disease. The 12.9 percent death-flag rate in the COVID subclass reflects the severity of the pandemic and the high-risk nature of the target population, not the toxicity of the antiviral drugs.
• Hepatitis C and Cirrhosis: HCV DAAs are often prescribed to patients with advanced liver cirrhosis or hepatocellular carcinoma. While DAAs are highly effective at curing the virus, these patients remain at risk for hepatic encephalopathy, variceal bleeding, and liver failure.

Because of these confounding factors, the 9.2 percent death-flag rate must be interpreted as a population-level marker of patient fragility rather than a drug-safety signal.

FAQs

Why are HCV DAAs and COVID antivirals grouped with acyclovir and oseltamivir?

While these drugs treat different viruses, they make up the non-HIV antiviral class. Grouping them allows safety teams to compare the safety profiles of acute, short-course therapies (such as oseltamivir for influenza or Paxlovid for COVID-19) against chronic or prophylaxis regimens (such as acyclovir for shingles suppression or valganciclovir for CMV prevention). This comparison highlights the differences between drug-induced toxicities and patient-related risks.

Does the cohort include tenofovir or lamivudine, which treat both HBV and HIV?

No. Tenofovir disoproxil fumarate (TDF), tenofovir alafenamide (TAF), and lamivudine are excluded from this cohort. Although they are FDA-approved to treat chronic hepatitis B, they are also used as the backbone of HIV antiretroviral therapy. Because FAERS reports often do not distinguish the indication at the substance level, including them would introduce significant HIV safety data into the non-HIV antiviral cohort. To maintain cohort hygiene, these dual-use agents are assigned to the HIV antiretroviral FAERS analysis.

Which generic oral antivirals are lowest-cost by NADAC, and does low cost mean low FAERS burden?

According to the National Average Drug Acquisition Cost (NADAC) database as of July 8, 2026, generic oral antivirals are highly affordable:

• Acyclovir 400 mg tablet: $0.0998 per unit ($2.99 for a standard 30-tablet supply).
• Acyclovir 800 mg tablet: $0.1881 per unit.
• Valacyclovir 1 g tablet: $0.4105 per unit.
• Entecavir 0.5 mg tablet: $0.2117 per unit.
• Famciclovir 500 mg tablet: $0.7654 per unit.
• Oseltamivir 75 mg capsule: $0.8273 per unit.
• Valganciclovir 450 mg tablet: $1.8828 per unit (reflecting its role as a specialty transplant agent).

Low cost does not correlate with a low FAERS burden. Acyclovir and valacyclovir are among the least expensive drugs, yet they drive the largest share of reports (over 43 percent of the cohort). This is due to their high prescribing volume and their use in vulnerable transplant and oncology patient populations. Conversely, valganciclovir is more expensive and has a smaller reporting footprint, but it carries a higher rate of serious reports (over 64 percent) due to its bone marrow toxicity and use in high-risk transplant settings.

Sources

• US FDA, FDA Adverse Event Reporting System (FAERS) / openFDA drug adverse event data, public extract (20,328,575 reports, export dated 2026-06-08). Aggregates computed by PharmaDossier from the openFDA FAERS full extract. https://open.fda.gov/data/faers/
• US Centers for Medicare & Medicaid Services, National Average Drug Acquisition Cost (NADAC) database, weekly pricing files (pricing effective July 8, 2026). https://www.medicaid.gov/medicaid/prescription-drugs/pharmacy-pricing/index.html
• US FDA, Paxlovid (nirmatrelvir and ritonavir co-packaged) prescribing information (NDA 217188, full approval May 25, 2023). https://www.accessdata.fda.gov/drugsatfda_docs/label/2023/217188s000lbl.pdf
• US FDA, Veklury (remdesivir) prescribing information (NDA 214787, initial US approval October 22, 2020). https://www.accessdata.fda.gov/drugsatfda_docs/label/2023/214787s019lbl.pdf
• US FDA, Drug Safety Communication: FDA warns about risk of hepatitis B reactivating in patients treated with direct-acting antivirals for hepatitis C (October 4, 2016). https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/fda-warns-risk-hepatitis-b-reactivating-patients-treated-direct-acting-antivirals-hepatitis-c
• JAMA. 2023;329(24):2118. FDA Grants Full Approval to Paxlovid for Adults at High Risk of Progression to Severe COVID-19. https://pubmed.ncbi.nlm.nih.gov/37285173
• Perazella MA. Drug-induced acute kidney injury: diverse mechanisms of tubular injury. Curr Opin Pediatr. 2019;31(2):211-220. (PMID: 30724817) https://pubmed.ncbi.nlm.nih.gov/30724817
• Jefferson T, et al. Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children. Cochrane Database Syst Rev. 2014;(4):CD008965. (PMID: 24718923) https://pubmed.ncbi.nlm.nih.gov/24718923
• US FDA, MedWatch: The FDA Safety Information and Adverse Event Reporting Program. https://www.fda.gov/safety/medwatch-fda-safety-information-and-adverse-event-reporting-program