The data behind the atlas
Ten ways of looking at anti-cancer nephrotoxicity — the same citation-grounded catalog, re-seen as similarity space, as a growing evidence base, as pharmacovigilance signal, and as renal anatomy. Every figure is drawn from the profiled data and needs no login.
Every anti-cancer agent placed so that nearness means a similar kidney-injury phenotype — computed from its injury signature, the nephron segments it strikes, severity, and drug class. Neighborhoods emerge on their own: platinum tubular injury, anti-VEGF glomerular disease, the checkpoint-inhibitor interstitial cluster. Dot size scales with FAERS reporting volume. Filter by injury; tap a profiled agent to open it.
263 agents · dot size = FAERS reporting volume · position = clinical-toxicity similarity (2-D MDS). Some catalog agents share an identical fingerprint and are gently fanned out to stay legible.
Every citation in the atlas placed by its publication year and its drug's signature kidney injury. The river's changing composition traces how the field's focus moved — from platinum tubular injury to anti-VEGF vascular disease to the checkpoint-inhibitor interstitial-nephritis era. Hover a year; click a band to isolate an injury.
50 years · citations attributed to each drug's signature injury · band thickness = citations that year
The kidney-injury combinations that strike the same agent, ranked by how many agents carry each exact set. The dot column under each bar shows which injuries are in that combination — the multi-way overlaps a pairwise chord can only hint at.
Top 14 injury combinations among profiled agents · bar = number of agents with exactly that set · connected dots = injuries in the combination.
How long each agent's kidney signal took to reach the literature — from approval (hollow) to its first nephrotoxicity citation (filled). The bar is the lag.
90 agents with both an approval year and a dated citation · hollow = approval · filled = first renal citation · bar = years to first documented signal.
Every profiled agent by evidence depth (citations, horizontal) and recency (newest reference, vertical). The crosshairs are the medians — the upper-left quadrant is the frontier (fresh but thin), the lower-right is deep but aging.
263 profiled agents · x = citation count (log) · y = newest citation year · crosshair = medians.
Each agent's acute-kidney-injury reporting odds ratio (horizontal, log) against its AKI report count (vertical, log). Dots to the right of the ROR = 1 line report AKI disproportionately; those high and right are the loudest, best-populated renal signals. Size = total report volume.
80 agents with a computed AKI ROR · filled = 95% CI lower bound > 1 (a disproportionality signal) · size = total FAERS reports. Reporting, not incidence.
Every profiled agent as a row; the 14 kidney injuries as columns. A colored tick means the agent carries that injury. Grouped by drug class, the class signatures read at a glance — whole families lighting up the same columns. Tap a column header to isolate an injury.
263 profiled agents × 14 injuries · tick = injury present · outlined tick = the agent's signature injury.
Where each drug class strikes the nephron. Rows are class families, columns are nephron segments; a cell's depth of color is how many agents in that class target that segment. Read down a column for the agents converging on one structure, across a row for a class's anatomical footprint.
Glomerulus22 | Vasculature / Endothelium129 | Proximal Tubule58 | Distal Tubule / Collecting Duct23 | Interstitium22 | Tubular Lumen8 | Bladder / Urothelium1 | |
|---|---|---|---|---|---|---|---|
| Kinase inhibitors (TKIs) | 4 | 34 | 22 | 10 | |||
| Other targeted agents | 2 | 35 | 9 | 3 | 2 | ||
| Alkylating agents | 4 | 7 | 5 | 4 | 1 | 1 | |
| Antimetabolites | 11 | 4 | 2 | ||||
| Checkpoint inhibitors | 16 | ||||||
| Anti-angiogenic (VEGF) | 11 | 3 | |||||
| Bispecifics / T-cell engagers | 1 | 12 | 1 | ||||
| Antibody-drug conjugates | 10 | 4 | |||||
| Monoclonal antibodies (other) | 1 | 6 | 2 | ||||
| Hormonal / endocrine | 3 | 4 | 1 | ||||
| Antitumor antibiotics | 1 | 3 | 2 | 1 | |||
| Radiopharmaceuticals | 1 | 4 | |||||
| Platinum agents | 4 | ||||||
| CAR-T cell therapy | 4 | ||||||
| Cytokines & enzymes | 1 | 3 | |||||
| Bisphosphonates & bone | 1 | 2 | 1 |
Agent counts per class × nephron segment · color depth scales within the grid · column totals above each segment.
Assemble a regimen and see where kidney risk compounds. Each bar is a kidney injury; its length is how many of the chosen agents carry it. Injuries struck by two or more drugs are the additive hazards to watch.
2 agents selected · bar segments = contributing drugs · ⚠ marks injuries hit by ≥2 agents (compounding risk). Educational — not a dosing tool.
How the signature kidney injury shifts across drug generations. Each ribbon carries agents from an era (left) to their signature injury (right); width = number of agents. The established backbone sits in tubular injury while the frontier fans into thrombotic microangiopathy, interstitial nephritis, and glomerular disease.
Ribbon width = agents flowing from each era to their signature injury · numbers = agents per era.
Figures are derived from the static, citation-grounded catalog. FAERS is a spontaneous-reporting system — reporting odds ratios reflect disproportionate reporting, not incidence or proven causation. Medical-education content only — not medical advice.