Checkpoint inhibitors in FAERS: 266,209 adverse-event reports by the numbers

The FDA Adverse Event Reporting System (FAERS) holds 266,209 individual safety reports that name at least one of the ten marketed immune checkpoint inhibitors — pembrolizumab (Keytruda), nivolumab (Opdivo), ipilimumab (Yervoy), atezolizumab (Tecentriq), durvalumab (Imfinzi), avelumab (Bavencio), cemiplimab (Libtayo), tremelimumab (Imjudo), dostarlimab (Jemperli), or relatlimab (Opdualag) — in any role, filed between 2011, when ipilimumab became the first approved checkpoint inhibitor, and early 2026. Pembrolizumab and nivolumab anchor the class with 97,737 and 93,396 reports respectively, reflecting more than a decade of frontline use across melanoma, lung, and genitourinary cancers; ipilimumab contributes 42,010 and atezolizumab 37,644. The class is the most serious-skewed of any FAERS cohort this publication has analyzed: 237,357 of the reports — 89.2 percent — carry an FDA seriousness flag, and 60,942 (22.9 percent) record a fatal outcome. Across the class the most coded reaction term after "death" (26,580) is "malignant neoplasm progression" (23,605), a reminder that these reports come from an advanced-cancer population, but the immune-related adverse-event (irAE) footprint is unmistakable: 23,369 reports in the colitis/diarrhoea family, 16,726 endocrinopathies, 12,223 pneumonitis/interstitial lung disease reports, and 5,067 myocarditis/pericarditis reports.

This article is a class-level descriptive read of checkpoint-inhibitor reporting in FAERS, written for pharmacovigilance, medical-affairs, oncology market-access, and payer teams who encounter checkpoint-inhibitor safety numbers in REMS design, irAE-management policy, indication-expansion reviews, and combination-protocol briefings. Every figure here is computed from the public openFDA FAERS extract (20,328,575 total reports, export dated 2026-06-08). It is not a disproportionality signal analysis, it is not a safety claim, and it is not incidence data. FAERS is a spontaneous-reporting system: it captures what was reported, never what was caused, it has no exposure denominator, and this population is skewed toward advanced malignancy with high baseline mortality. Those caveats shape every number below. For the class-wide FAERS methodology and the database aggregate picture, see the companion analysis of 20 million FAERS reports; for the other class cuts, see GLP-1 receptor agonists, SGLT2 inhibitors, JAK inhibitors, and DOACs in FAERS. This article stays on the ten checkpoint inhibitors.

Methodology, in one paragraph

A report is counted for a substance when the substance's generic name appears anywhere in that report's drug list — as a suspect, concomitant, or interacting drug — using a substring match across the drug_substances, drug_generics, and drug_brands fields. Because a single report can name several drugs (combination ipilimumab/nivolumab or nivolumab/relatlimab regimens are common), the per-substance totals sum to more than the 266,209 class-level (union, deduplicated) reports. Relatlimab is approved only as a fixed combination with nivolumab (Opdualag), so nearly all relatlimab reports also count toward nivolumab. A "serious" report is any report the FDA coded as serious. A report counts toward death when either the seriousness field carries a death flag or the outcome field carries a Fatal outcome. Reactions are coded to MedDRA Preferred Terms; reaction counts are case-normalized. The immune-related adverse-event families aggregate related Preferred Terms (for example, "pneumonitis/interstitial lung disease" combines pneumonitis and interstitial lung disease; "colitis/diarrhoea" combines colitis and diarrhoea), and a report contributes once to a family if any term in the family appears. The "any role" counts reported here are intentionally inclusive, so they will exceed suspect-only or differently normalized counts cited elsewhere for the same agent; they are not directly comparable to single-substance figures in the companion database-wide analysis.

The reporting-volume picture, by drug

Three patterns matter for access and medical-affairs teams. First, volume is indication breadth and combination use, not risk: pembrolizumab and nivolumab together account for nearly three quarters of class reporting because they hold the largest label footprints — Keytruda alone carries approvals across more than a dozen tumor types — and sit at the center of the most-used combination regimens. Ipilimumab's 42,010 reports are elevated for a drug used largely in combination with nivolumab, because any-role counting pulls the CTLA-4 agent into every ipilimumab-plus-nivolumab report. Second, the serious-share floor is extraordinarily high — every agent sits between 75 and 97 percent serious — because this is an advanced-cancer population in which hospitalization and death are common regardless of attribution. Third, the per-drug death-flag percentages move with indication mix and combination exposure, not with a clean drug-safety differential: durvalumab's 30.9 percent and tremelimumab's 29.0 percent exceed pembrolizumab's 18.1 percent, a gap far better explained by durvalumab's heavy use in locally advanced and metastatic lung cancer and tremelimumab's near-exclusive use in combination with durvalumab in metastatic disease than by any claim that those agents are intrinsically more lethal. Quoting any single drug's death-flag rate without that caveat is quoting an indication-confounded number.

The class-level outcome profile

A report can carry more than one outcome, so these shares sum to more than 100 percent. The dominant "unknown outcome" share is typical of FAERS follow-up gaps; the Fatal outcome count (59,127) sits close to the death-flag measure (60,942) because most death-flagged reports also carry a Fatal outcome. As with every FAERS outcome table, none of these percentages are incidence — the denominator is reports, not exposed patients.

The immune-related adverse-event footprint

Stripping out the non-specific terms (death, malignant neoplasm progression, fatigue, pyrexia) that dominate the raw reaction list for an advanced-cancer cohort, the immune-related adverse-event families that distinguish checkpoint inhibitors from cytotoxic oncology leave a clear footprint:

The pattern is consistent with what the labels and the published disproportionality literature describe. Colitis clusters toward the CTLA-4 agent (ipilimumab) and the combinations — a 2025 FAERS disproportionality study of ipilimumab/nivolumab combination therapy identified colitis, pneumonitis, hepatitis, hypophysitis, and myocarditis among the most-reported events, and clinical-trial data have long shown higher colitis rates with CTLA-4 inhibition than with PD-1/PD-L1 monotherapy. Pneumonitis and endocrinopathy concentrate in the two PD-1 workhorses: hypothyroidism is the single most-reported endocrine irAE across the class at 5,951 reports, and a 2024 FAERS analysis of PD-1, PD-L1, CTLA-4, and LAG-3 inhibitors (Frey and Etminan, Antibodies) found pembrolizumab carried the strongest hypothyroidism signal (reporting odds ratio ~25.9) and nivolumab the strongest colitis signal (~29.5).

Myocarditis — rare in count, dominant in severity

Myocarditis is the signal that disproportionately shapes the regulatory and access posture of the class even though it is rare in raw count. The 5,067 myocarditis/pericarditis reports here are an order of magnitude below the colitis or endocrinopathy volumes, but myocarditis carries the highest reporting odds ratios of any checkpoint-inhibitor event in the published pharmacovigilance literature — dostarlimab and combination ipilimumab/nivolumab generate the largest signals — and a case fatality that is exceptional for an irAE. A 2025 FAERS study of checkpoint-inhibitor myocarditis and pericarditis (Yao et al., BMC Cancer) counted 8,882 myocarditis/pericarditis reports under nivolumab-plus-ipilimumab combination therapy with a reported mortality of 27.6 percent, and clinical meta-analyses place all-grade myocarditis incidence near 0.2–1.0 percent with a disproportionate share of irAE deaths. Combination therapy is the dominant risk amplifier: dual CTLA-4 plus PD-1 blockade has been associated with roughly a four- to five-fold increase in myocarditis reporting relative to PD-1 monotherapy.

The access and label consequence is concrete. Myocarditis, pneumonitis, colitis, hepatitis, endocrinopathies, and nephritis sit in the Warnings and Precautions section of every checkpoint-inhibitor label, and the combination regimens carry the most intensive monitoring and steroid-escalation expectations. Prior-authorization criteria and oncology medical-policy bulletins across the class now routinely require baseline and interval cardiac, thyroid, hepatic, and renal monitoring and document irAE-management capability, and combination-protocol approvals carry explicit irAE-risk language. The descriptive counts here are the reporting backdrop to that policy layer; the disproportionality literature is what established the risk-ranking.

The reporting trajectory, by year

The trajectory tracks the class's commercial uptake, not a safety trend. Reporting begins with ipilimumab's March 2011 approval (568 reports that year), jumps when pembrolizumab (September 2014) and nivolumab (December 2014) launch, accelerates through 2016–2019 as the two PD-1 agents expanded across lung, renal, bladder, and head-and-neck cancer, and plateaus above 30,000 reports per year from 2022. The 2026 figure (8,220) is a partial year through the extract cut-off. The flattening at ~33,000 reports per year is the reporting steady-state of a class now used in the adjuvant and first-line metastatic setting across most solid tumors.

What this means for pharmacovigilance, medical-affairs, and access teams

• For pharmacovigilance teams, the class's 89.2 percent serious share and 22.9 percent death-flag rate are population effects, not signal. Read serious and death proportions against an advanced-cancer baseline; the signals that matter are the disproportionality-documented irAE families — colitis, pneumonitis, hepatitis, endocrinopathy, and especially myocarditis — not the headline death counts.
• For medical-affairs and label teams, the myocarditis count (5,067) is small but the severity and combination-therapy amplification are the durable regulatory pressure point. Combination CTLA-4/PD-1 regimens carry the densest monitoring requirements, and any label or risk-management change to the class will continue to center cardiac and pulmonary irAEs.
• For oncology market-access and payer teams, irAE-monitoring and management capability is now standard medical-policy language across the class. Prior-authorization criteria that require baseline cardiac, thyroid, hepatic, and renal assessment, and documentation of irAE-management infrastructure, are defensible against the reporting picture here; combination regimens justify the stricter monitoring tier.
• For dossier and evidence teams, combination-regimen reporting (ipilimumab/nivolumab, nivolumab/relatlimab) inflates the per-drug counts through any-role counting. When quoting a per-agent FAERS figure, anchor it with the regimen context — a nivolumab or relatlimab number in a combination-protocol population is not comparable to a monotherapy number.
• For 340B and buy-and-bill oncology programs, the class's reporting volume and serious-outcome mix are consistent with high-acuity medical-benefit administration. The irAE burden reinforces why these agents route through oncology specialty and hospital channels with intensive clinical support, not retail pharmacy.

Sources

• US FDA, openFDA FAERS (Adverse Event Reporting System) extract, export dated 2026-06-08 (20,328,575 total reports). Per-drug, class-union, outcome, irAE-family, and trajectory aggregates computed by PharmaDossier from the public FAERS extract. https://open.fda.gov/data/faers/
• US FDA, Drugs@FDA approval records for checkpoint inhibitors — ipilimumab (Yervoy, approved March 25, 2011); pembrolizumab (Keytruda, approved September 4, 2014); nivolumab (Opdivo, accelerated approval December 22, 2014); atezolizumab (Tecentriq, 2016); durvalumab (Imfinzi, 2017); avelumab (Bavencio, 2017); cemiplimab (Libtayo, 2018); dostarlimab (Jemperli, 2021); relatlimab (Opdualag, nivolumab and relatlimab, 2022); tremelimumab (Imjudo, 2022). https://www.accessdata.fda.gov/scripts/cder/daf/
• Frey C, Etminan M. "Adverse Events of PD-1, PD-L1, CTLA-4, and LAG-3 Immune Checkpoint Inhibitors: An Analysis of the FDA Adverse Events Database." Antibodies 2024;13(3):59 — hypothyroidism, pneumonitis, colitis, and myocarditis reporting odds ratios by agent across the four checkpoint-inhibitor classes. https://www.mdpi.com/2073-4468/13/3/59
• Yao G, Zhang Y, Zhang H, et al. "Immune checkpoint inhibitor-associated myocarditis and pericarditis: a pharmacovigilance study based on the FAERS database." BMC Cancer 2025;25:1294 — nivolumab plus ipilimumab myocarditis/pericarditis reports and 27.6 percent mortality. https://link.springer.com/article/10.1186/s12885-025-14668-x
• Dai X, Cao H, Li J, et al. "Disproportionality analysis of adverse events associated with ipilimumab and nivolumab combination therapy based on FAERS database." Scientific Reports 2025;15:21055 — colitis, pneumonitis, hepatitis, hypophysitis, and myocarditis as dominant combination-regimen irAEs. https://www.nature.com/articles/s41598-025-07287-w
• Mahmoudi M, Mercado C, J noskuti L, et al. "Immune Checkpoint Inhibitor–Induced Cardiotoxicity: A Systematic Review and Meta-Analysis." JAMA Network Open 2024 (PD-1 all-grade cardiovascular adverse-event incidence ~1.07 percent; all-grade myocarditis ~0.24 percent). https://pmc.ncbi.nlm.nih.gov/articles/PMC11342217
• "Myocarditis occurrence with cancer immunotherapy across indications in clinical trial and post-marketing data." Scientific Reports 2021 — checkpoint-inhibitor myocarditis reporting odds ratios by agent and combination regimen. https://www.nature.com/articles/s41598-021-96467-5
• US FDA, Prescribing Information (Warnings and Precautions: immune-mediated pneumonitis, colitis, hepatitis, endocrinopathies, nephritis, and myocarditis) — Keytruda (pembrolizumab), Opdivo (nivolumab), Yervoy (ipilimumab), Tecentriq (atezolizumab). https://www.accessdata.fda.gov/scripts/cder/daf/