Nivolumab
Opdivo · PD-1 checkpoint inhibitor
PD-1 inhibitor; ICI acute interstitial nephritis is the prototype.
Keytruda · Pembro
PD-1 checkpoint inhibitor · approved 2014 · 15 citations · FAERS AKI reporting ROR 2.59 (95% CI 2.49–2.69, 2,572 AKI reports)
Describes how this page is sourced, not how dangerous the drug is. Thinly sourced means fewer of the sourcing signals are met — not that the agent is kidney-safe. A rule-based summary, not a formal certainty appraisal.
The archetypal checkpoint-inhibitor kidney injury: a delayed, often PPI-associated acute interstitial nephritis that usually responds to drug withdrawal and steroids.
Atlas fields gathered in one place, each linked to the section with its full text and sources. Educational, not medical advice.
Signature lesion
In real-world cohorts of patients receiving immune checkpoint inhibitors, any AKI is common (roughly 16-18%), but AKI attributable to the checkpoint inhibitor itself (ICPi-AKI) is less frequent. A single-center cohort reported AKI in 16.5% of ICI-treated patients with checkpoint-attributable nephrotoxicity in a minority, while a larger real-world study found ICPi-AKI in about 3.6%. Acute interstitial nephritis is the dominant biopsy lesion (>80% in the largest multicenter series). These figures are pooled across PD-1/PD-L1/CTLA-4 agents rather than pembrolizumab-specific.Source: Lumlertgul, Eur J Cancer 2023
Typically weeks to months; multicenter cohort medians ~14–16 weeks; can follow a single dose or many months out and may recur on rechallenge.
Distilled from: “Delayed and highly variable, typically weeks to months after initiation; multicenter cohorts reported median onsets around 14-16 weeks. Can occur after a single dose or after many months, and may recur on rechallenge.”
In the multicenter ICI-AKI cohort ~22% were rechallenged and ~23% of those developed recurrent AKI — feasible but an individualized, co-managed decision. PMID 31896554 (opens PubMed in a new tab)
About two-thirds of patients recover kidney function after checkpoint-inhibitor AKI; corticosteroids are the mainstay for moderate-to-severe disease, and earlier initiation (within ~3 days of AKI) is associated with higher odds of renal recovery.PMID 34625513 (opens PubMed in a new tab)
Distilled from the agent's cited literature — educational, not a substitute for the primary sources.
Recovery across agentsThis agent's kidney lesions ordered by prominence — the #1 signature lesion first, then secondary and rare patterns. Cited incidence is shown where a citable figure exists; otherwise the tier stands qualitatively.
Kidney toxicity in 12 of 676 pembrolizumab-treated patients (1.77%) referred to one nephrology center; on biopsy, acute tubular injury alone in 5, acute interstitial nephritis in 4, minimal change disease (with or without ATI) in 2.
Not the same figure as §01's 3.6%. §01 is AKI judged to be caused by any checkpoint inhibitor in a 1,037-patient cohort; this row is pembrolizumab kidney toxicity referred to one nephrology center, of any lesion type.
Renal electrolyte derangement — magnesium/potassium/calcium wasting (cisplatin, anti-EGFR antibodies) or retention (FGFR-inhibitor hyperphosphatemia, tumor-lysis hyperkalemia/hyperphosphatemia).
Glomerular lesions (minimal change disease, membranous nephropathy) reported in biopsy series of ICI-related AKI, secondary to the predominant interstitial nephritis. PMID 32601079 (opens PubMed in a new tab)
Thrombotic microangiopathy is a rare, case-level ICI kidney lesion (2 of 12 biopsied in one ICI-AKI cohort). PMID 32601079 (opens PubMed in a new tab)
Slow, cumulative tubulointerstitial scarring — fibrosis, tubular atrophy and glomerulosclerosis with no discrete acute phase. The nitrosourea (carmustine/lomustine) lesion and delayed radioligand (radiation) nephropathy; often irreversible and detected only as a creeping creatinine months to years later.
Case-level immune-related (irAE) cystitis - immune-mediated, not acrolein toxicity.
Tap a signature to trace where it strikes the nephron.
Representative incidence
Acute Interstitial Nephritis strikes: Interstitium.
Acute Interstitial Nephritis
Immune-mediated inflammation of the renal interstitium — the signature kidney injury of checkpoint inhibitors.
Humanized IgG4 monoclonal antibody against PD-1 that blocks the PD-1/PD-L1 inhibitory axis, releasing the brakes on T cells and restoring antitumor immunity. The same loss of peripheral tolerance that drives efficacy also underlies its immune-related adverse events, including the kidney.
27 grade-indexed syndromes across 11 organ systems, written for checkpoint blockade as a class rather than per agent — where a rate differs between anti-PD-1, anti-PD-L1, anti-CTLA-4 and combination therapy, the card names the class it was measured in.
Class-level context for the major non-renal toxicities of the PD-1 checkpoint inhibitor class.
Endocrine
Thyroiditis, hypophysitis, diabetes
Gastrointestinal
Diarrhea, colitis, mucositis, perforation
Hepatic / Liver
Transaminitis, hepatitis, VOD/SOS
Pulmonary
Pneumonitis, ILD, effusions, hypertension
Dermatologic
Rash, HFS, SJS/TEN, vitiligo
12 primary references — trials, cohorts, mechanism, and reviews. Single-patient case reports are listed separately below, graded by strength. Citation metadata via PubMed / NLM.
Primary (non–case-report) references per year — a proxy for how actively the agent's renal literature is accruing. Recent years are highlighted. Reflects curation depth, not a systematic bibliometric count.
Single-patient and small-series reports, graded by evidentiary strength — A Strong (biopsy-proven plus a series and/or positive rechallenge), B Moderate, and C Limited (a single clinically-diagnosed case). Strongest first. Grades are inferred automatically from each report's abstract and journal — a heuristic ranking aid, not a formal quality appraisal.
Kidney rows of the label's dosage tables
From the KEYTRUDA label (Oct 2026) · DailyMed
The rows that name proteinuria, nephrotic syndrome, creatinine, creatinine clearance, renal or nephritis together with a withhold, reduce, resume or discontinue instruction, reproduced cell for cell. A merged cell is written out on each row it covers. The tables' footnotes are not reproduced, and they can hold the conditions for resuming: read the full table on DailyMed. A cell marked fn carries one of those footnotes on the label. Educational only, not medical advice.
| Adverse Reaction | Severityfn | Dosage Modification |
|---|---|---|
| Nephritis with Renal Dysfunction | Grade 2 or 3 increased blood creatinine | Withholdfn |
| Nephritis with Renal Dysfunction | Grade 4 increased blood creatinine | Permanently discontinue |
Immune-related nephritis is managed grade by grade on the atlas's kidney irAE card; this block does not restate it.
Everything below is FAERS — adverse events someone chose to report, about 109,044 of them for this agent. Nobody counts the patients who were fine, so none of these numbers is an incidence, a risk, or a rate: they describe what gets reported, shaped by a drug's fame, its indication, and who was watching. How these numbers work.
Only significant signals appear (95% CI lower bound above 1) — a phenotype missing here was tested and did not reach significance, except Prerenal / Hemodynamic AKI, Pseudo-AKI, Renal Cysts, Chronic Interstitial Nephropathy — outside the clinician-reviewed MedDRA term map, never queried — and ATN and AIN, queried but biopsy-bound: real cases are filed as generic “acute kidney injury”, so their absence is not a negative. As of 2026-11-15.
What reporting says about this profile's documented lesions
Reported with a death outcome
19,411 of 109,044 reports
Reported with hospitalization
46,002 of 109,044 reports
Reports per year
Reports received per year, 2015–2026 · most recent year is partial · 623 of the 109,044 reports predate 2015 and are not drawn.
Bars rank systems by summed reaction-term mentions (a report counts once per term it names) — an ordinal “more vs less reported” cue, not a tally of distinct reports. Renal & urinary first. Showing the top 10 of 12 classified systems; 2 lower-ranked systems are not drawn. As of 2026-11-15.
Each recommendation below is this atlas's faithful summary of the source, not a quotation from it — follow the PubMed link for the wording the society published. Summaries may be superseded; consult the current full text and individualize to the patient.
General onco-nephrology references
Where Pembrolizumab sits in nephrotoxicity space — each dot is an anti-cancer agent, positioned so neighbors share a kidney-injury phenotype. Its 6 closest are filled and lead to a numbered marker, matching the numbered cards below.
Opdivo · PD-1 checkpoint inhibitor
PD-1 inhibitor; ICI acute interstitial nephritis is the prototype.
Loqtorzi · PD-1 immune checkpoint inhibitor
First FDA-approved drug for nasopharyngeal carcinoma; renal risk is class-typical immune-mediated interstitial nephritis.
Unloxcyt · PD-L1 immune checkpoint inhibitor
PD-L1 blockade for cutaneous SCC — watch for late immune interstitial nephritis
Cejemly · PD-L1 immune checkpoint inhibitor
A full-length anti-PD-L1 antibody whose kidney is its immune system: rare but real autoimmune interstitial nephritis.
Penpulimab (AK105) · PD-1 immune checkpoint inhibitor
Fc-silent PD-1 blocker for nasopharyngeal carcinoma — class-typical, infrequent immune interstitial nephritis
Tevimbra · PD-1 immune checkpoint inhibitor
A PD-1 blocker whose kidney risk is the class-typical rare immune-mediated interstitial nephritis.
Nearest agents by kidney-injury phenotype (shared injuries, nephron target, severity, class) — a similarity approximation, not a claim of shared drug identity or mechanism.
Same-class agents ordered by their documented kidney-injury profile — atlas severity, an acute-kidney-injury FAERS signal, and how many injury types each is documented to cause. Agents nearer the top carry the lighter documented renal profile. Agents sharing a number are tied on all three. An agent marked FAERS n/a has too few reports to test, so its place rests on severity and injury breadth alone.
A comparison of documented kidney-injury data within one drug class — not a substitution recommendation. Efficacy, indication, and non-renal toxicity differ between these agents and are out of scope here. Educational only, not medical advice.
The leading contributors to Pembrolizumab’s clinical kidney literature on PubMed, ranked by a blend of publication volume and citation impact — filtered toward clinical work via the PubMed Humans heading and clinical publication types (trials, cohorts, case reports, guidelines, reviews). Names link to that author’s work on Pembrolizumab; the PMIDs beside each name are up to three of their most recent papers on it, not the full count.
Ranked by a 50/50 blend of publication volume and a position-weighted, capped Relative Citation Ratio (NIH iCite) on this agent’s renal literature; the citation count shown is the raw total, not the ranking score — counted over the 164 clinical records among all 187 PubMed matches, so counts are within-sample — bibliometric context, not an endorsement or a measure of clinical authority.