🧬 Aging Wiki

canakinumab

Properties1
aliasesIlaris, ACZ885

16 min read

Canakinumab

A fully human IgG1κ monoclonal antibody that neutralizes interleukin-1β (IL-1β) with sub-picomolar affinity. Developed by Novartis (development code ACZ885) and marketed as Ilaris. FDA-approved since 2009 for rare autoinflammatory genetic diseases driven by NLRP3 or other IL-1β-axis gain-of-function mutations. Not approved for cardiovascular prevention, despite being the intervention tested in CANTOS — the largest and most influential prospective trial of anti-inflammatory therapy in atherosclerosis. Canakinumab is the wiki’s principal clinical anchor for the chronic-inflammation hallmark.

Identity

  • Class: fully human monoclonal antibody, IgG1Îş isotype
  • Molecular weight: ~145 kDa (heavy + light chains)
  • Development code: ACZ885 (Novartis)
  • Trade name: Ilaris
  • ChEMBL ID: CHEMBL1201834 (confirmed against ChEMBL API)
  • DrugBank ID: DB06168
  • PubChem CID: null — biologics are not assigned small-molecule PubChem CIDs; the compound frontmatter field is inapplicable for monoclonal antibodies. needs-canonical-id — if a PubChem protein databank or biologics entry exists, link here on next lint pass.

Target specificity

Canakinumab binds IL-1β exclusively — it does not bind IL-1α or the IL-1 receptor antagonist (IL-1Ra). This narrow specificity is a critical pharmacological distinction:

  • anakinra — a recombinant IL-1Ra that competitively blocks the IL-1 receptor (IL-1R1); therefore blocks signaling from both IL-1α and IL-1β. Broader suppression but shorter half-life (4–6 hours; daily SC injection).
  • Canakinumab — binds free IL-1β directly, prevents its engagement of IL-1R1, leaving IL-1α-mediated signaling intact. This selectivity may explain its tolerability profile relative to broad IL-1 blockade and means IL-1α-driven inflammation is not suppressed. needs-replication — the functional consequence of IL-1α sparing in the context of chronic atherosclerotic inflammation has not been independently characterized.

The target is the upstream cytokine driving the nlrp3-inflammasome → IL-1β → nf-kb axis prominent in chronic low-grade inflammaging. See chronic-inflammation for the pathway context.

Mechanism and pharmacokinetics

Mechanism: Canakinumab binds IL-1β with a reported Kd in the sub-picomolar range, effectively stoichiometrically sequestering circulating IL-1β. By blocking IL-1R1 engagement it interrupts the downstream MyD88–IRAK–TRAF6–IKK–NF-κB signaling cascade responsible for pro-inflammatory gene transcription (cytokines, adhesion molecules, acute-phase proteins). needs-replication — the precise measured Kd and its comparison to endogenous IL-1β concentrations have not been independently verified against the primary binding characterization papers for this page; verify on next verifier pass.

Pharmacokinetics:

  • Half-life: ~26 days — an exceptionally long half-life attributable to IgG FcRn-mediated recycling and the large molecular weight. This enables once-every-3-months (q3mo) subcutaneous dosing — the schedule used in CANTOS.
  • Route: subcutaneous injection; no oral form exists for a ~145 kDa protein.
  • Clearance: predominantly via proteolysis (typical for IgG antibodies); not hepatically or renally metabolized in a conventional sense — contrast with small-molecule drugs.
  • Dose used in CANTOS: 50, 150, or 300 mg SC every 3 months 1.

The long half-life is a double-edged feature: it supports infrequent dosing but also means immune suppression persists for months after the last dose, relevant to the infection signal (see below).

FDA approvals

Canakinumab is approved for rare IL-1β-driven autoinflammatory monogenic diseases, not for common-population indications:

YearIndicationGenetic basis
2009CAPS: FCAS (familial cold autoinflammatory syndrome), MWS (Muckle-Wells syndrome), NOMID/CINCA (neonatal-onset multisystem inflammatory disease)NLRP3 gain-of-function variants
2013Systemic juvenile idiopathic arthritis (sJIA)Polygenic; IL-1β-driven flare mechanism
2016TRAPS (TNF receptor-associated periodic syndrome), HIDS/MKD (hyperimmunoglobulin D syndrome / mevalonate kinase deficiency), FMF (familial Mediterranean fever)Rare monogenic periodic fever syndromes

Off-label use: adult-onset Still’s disease (AOSD), sharing pathophysiology with sJIA.

The CAPS approval is mechanistically coherent with the nlrp3-inflammasome page: NLRP3 gain-of-function mutations constitutively activate the inflammasome → excess IL-1β secretion → autoinflammatory episodes. Canakinumab neutralizes that excess IL-1β.

CANTOS — aging-relevant trial evidence

The trial-result tables and full statistical narrative live on indication-specific pages to avoid duplication. The brief synopsis below links to the canonical locations.

CANTOS (Canakinumab Anti-inflammatory Thrombosis Outcomes Study) — Phase 3 RCT, n=10,061 post-MI patients with hsCRP ≥2 mg/L at baseline. Canakinumab 50/150/300 mg q3mo SC vs placebo. Pre-specified primary endpoint: first MACE (CV death, non-fatal MI, or non-fatal stroke) 1.

Key results (canonical home: atherosclerosis and chronic-inflammation):

  • Primary endpoint (MACE) incidence rates (per 100 person-years): placebo 4.50 · 50 mg 4.11 (HR 0.93, 95% CI 0.80–1.07; P=0.30, ns) · 150 mg 3.86 (HR 0.85, 95% CI 0.74–0.98; P=0.021, met pre-specified threshold p<0.02115) · 300 mg 3.90 (HR 0.86, 95% CI 0.75–0.99; P=0.031, did NOT meet pre-specified threshold p<0.01058).
  • The pre-specified closed-testing procedure set different thresholds per dose: p<0.02115 for both 50 mg and 150 mg; p<0.01058 for 300 mg. Only the 150 mg arm met its threshold.
  • hsCRP reduction at 48 months: median reductions greater than the placebo group were 26 percentage points (50 mg), 37 percentage points (150 mg), and 41 percentage points (300 mg) — all P<0.001 for comparison to placebo. No significant reduction in LDL cholesterol — establishing that the CV benefit, if real, is attributable to inflammation reduction rather than lipid lowering. This was the first prospective demonstration of that principle at population scale.
  • All-cause mortality: neutral across all doses combined (HR 0.94, 95% CI 0.83–1.06; P=0.31).

Class-level meta-analytic confirmation (2025): A 25-RCT systematic review and meta-analysis of immunomodulatory drugs in coronary heart disease (He 2025 BMC Cardiovasc Disord) found that overall, the immunomodulator class did not significantly reduce MACE (RR 0.92, 95% CI 0.84–1.01, p=0.09; I²=60%), but subgroup analyses recovered the CANTOS-aligned signal: interleukin-pathway inhibitors (which here cluster canakinumab with IL-6 inhibitors and IL-1Ra agents) reduced MACE with RR 0.86 (95% CI 0.75–0.97, p=0.02), and NLRP3 inflammasome inhibitors (a category that captures the canakinumab/colchicine mechanism with different drug examples) reduced MACE with RR 0.75 (95% CI 0.65–0.86, p<0.0001). Benefits were evident only in trials with follow-up >6 months. This is the first meta-analysis to formally aggregate canakinumab + adjacent IL-1/NLRP3-pathway interventions and confirms that the CANTOS effect is mechanistically robust across the class, not idiosyncratic to canakinumab 2.

CANTOS lung cancer pre-specified secondary analysis 3 — exploratory analysis within CANTOS:

  • 300 mg arm showed a striking reduction in incident lung cancer mortality (HR 0.49, 95% CI 0.31–0.75) and in new lung cancer incidence.
  • The result was hypothesis-generating for anti-inflammatory cancer prevention but was pre-specified as exploratory.
  • Canonical home: cancer page.

Status of cancer-signal replication as of 2026-05 — the CANTOS lung cancer signal has not replicated in any of four prospective NSCLC trials in therapeutic settings. The Novartis CANOPY oncology program (named to evoke CANTOS) directly tested whether IL-1β neutralization improves NSCLC outcomes when given alongside standard cancer therapy:

  • CANOPY-A (Garon 2024 J Clin Oncol) — Phase 3 adjuvant trial in completely-resected stage II–IIIB NSCLC after platinum-based chemotherapy, n=1,382. Primary endpoint disease-free survival HR 0.94 (95% CI 0.78–1.14, one-sided p=0.258). Negative. CRP and IL-6 dropped as expected, confirming target engagement, but no clinical benefit. OS not formally tested per the closed-testing hierarchy 4.
  • CANOPY-1 (Tan 2024 J Clin Oncol) — Phase 3, first-line metastatic NSCLC, canakinumab vs placebo each combined with pembrolizumab + platinum-based chemotherapy. Did not improve survival vs placebo arm 5.
  • CANOPY-2 (Paz-Ares 2024 Lung Cancer) — Phase 3, advanced NSCLC after platinum-doublet chemotherapy + immunotherapy progression, canakinumab + docetaxel vs placebo + docetaxel, n=237. Did not meet primary OS endpoint 6.
  • CANOPY-N (Lee 2025 JTO Clin Res Rep) — Phase 2 neoadjuvant, n=88; major pathologic response rate 2.9% (canakinumab alone) vs 17.1% (canakinumab + pembrolizumab) vs 16.7% (pembrolizumab alone). Canakinumab alone or combined with pembrolizumab did not improve MPR vs pembrolizumab alone 7.

The four-trial replication failure suggests the CANTOS lung cancer signal does not extend to therapeutic NSCLC settings, where established disease confounds the intended anti-inflammatory mechanism. Whether the CANTOS signal reflects a real cancer-prevention effect — operating in early/precancerous biology rather than established tumors — remains the open question the prevention-setting trials are designed to test:

  • Can-Prevent-Lung (NCT04789681) — Phase 2 single-arm, canakinumab in patients with indeterminate pulmonary nodules; primary endpoint = nodule regression; recruiting (n=50; estimated completion 2027). Tests the prevention hypothesis directly.
  • CANIFS (NCT06038526) — Phase 2 in high-risk former smokers; molecular endpoints (ASC speck, caspase-1, IL-1β); active, not recruiting (n=41).
  • CCUS prevention (NCT05641831) — Phase 2 RCT canakinumab vs placebo in clonal cytopenias of unknown significance; primary endpoint = time to overt myeloid malignancy; recruiting (n=110). Tests cancer-prevention from a CHIP-related premalignant context.

Exploratory biomarker analyses from CANOPY-1 + CANOPY-N (Solomon 2025 Cancer Res Commun) reported a possible benefit of canakinumab + ICI in the subset with low T-cell tumor infiltration (“immunosuppressive TME”) 8 — hypothesis-generating but underpowered.

needs-replication — the prevention-setting hypothesis has not yet been tested in a Phase 3-powered trial; until Can-Prevent-Lung and CCUS-prevention report, the cancer signal must be considered unconfirmed in any indication.

CANTOS infection signal

A critical safety finding from CANTOS: fatal infection or sepsis rate was higher in canakinumab-treated arms vs placebo — 0.31 vs 0.18 events per 100 person-years (P=0.02) 1. Sepsis accounted for a significant fraction of excess fatal infections. This signal is expected mechanistically: IL-1β is a central mediator of innate immune responses to bacterial pathogens, and its sustained neutralization over the long 26-day half-life impairs early infection response. Canonical safety summary lives on senomorphics.

The infection signal was a key factor in Novartis’s decision not to pursue FDA approval for a cardiovascular prevention indication.

Real-world post-marketing confirmation (2026): A FAERS pharmacovigilance analysis of 9,262 canakinumab-related adverse-event reports from 2010 onward (Wang 2026 Pharmacol Res Perspect) confirmed and refined the CANTOS infection signal — infections and infestations showed a strongly elevated reporting odds ratio (ROR 2.54; n=3,532 reports), and age-stratified analyses found that older patients (≥65 years) most frequently reported pneumonia, sepsis, and cellulitis, with additional strong signals for neoplasms in the same age stratum. ~40% of adverse events occurred after >1 year of exposure, indicating both early- and long-term surveillance is required 9. This is the first large-scale real-world confirmation that the infection-and-malignancy signals observed in CANTOS persist in routine clinical use across the CAPS / sJIA / periodic-fever indications, and that risk is age-stratified — directly relevant to any future aging-prevention use case.

CANTOS-CHIP subgroup (Svensson 2022)

A pre-specified exploratory genomic substudy of CANTOS examining whether cardiovascular benefit from canakinumab is modified by clonal hematopoiesis of indeterminate potential (CHIP) 10. Targeted deep sequencing of 74 CHIP-associated genes was performed on baseline peripheral blood DNA from n=3,923 CANTOS participants (of whom 338, or 8.6%, met CHIP criteria). TET2 was the most common CHIP variant gene in this population (in contrast to general population cohorts where DNMT3A predominates — a difference attributed to the high-hsCRP selection criterion of CANTOS). TET2-CHIP carriers showed a 62% relative reduction in MACE on canakinumab vs placebo (HR 0.38, 95% CI 0.15–0.96; P=0.04 in stratified analysis), compared with no significant benefit in non-TET2-CHIP patients (HR 1.08, 95% CI 0.58–2.00; P=0.82). However, the interaction test was not statistically significant (P for interaction = 0.14). The mechanistic interpretation: TET2-deficient myeloid cells exhibit hyper-inflammatory IL-1β production in response to inflammatory stimuli (NLRP3 inflammasome pathway), so IL-1β neutralization corrects a specifically elevated substrate in TET2-CHIP carriers.

This result is exploratory and underpowered (26% power for incident MACE per the paper’s own limitations statement) — differential benefit is hypothesis-generating rather than established. Canonical home: clonal-hematopoiesis.

Prospective replication of CANTOS-CHIP underway (R34 update, 2026-05): the post-hoc TET2-CHIP signal is now being directly tested in a Phase 2 RCT — NCT06691217 (“Effects of IL-1 Beta Inhibition on Vascular Inflammation in TET2 Clonal Hematopoiesis”), a Massachusetts General Hospital–led trial randomizing TET2-CHIP carriers to canakinumab vs placebo with primary endpoint = change in coronary perivascular fat attenuation index (a coronary CT measure of vascular inflammation), n=120, recruiting since April 2026. This is the first prospective genotype-stratified canakinumab trial — the experiment Svensson et al. flagged as essential. The 2026 Front Cardiovasc Med CHIP–CVD review (Wu 2026) frames mutation-guided IL-1β-pathway prevention as the emerging “cardio-hematology” precision-medicine direction, and identifies the lack of prospective genotype-stratified trials as the principal knowledge gap NCT06691217 is designed to close 11.

DimensionStatus
Pathway conserved in humans?yes (human RCT)
Phenotype conserved in humans?yes (MACE endpoints, CANTOS)
Replicated in humans?in-progress / limited (CANTOS-CHIP subgroup not independently replicated)

Translation gap — why canakinumab is not used for aging prevention

Despite CANTOS providing the strongest prospective human evidence that chronic inflammation drives atherosclerotic events and possibly cancer, canakinumab has not been adopted for aging-prevention or cardiovascular-prevention indications. Several reasons:

  1. Infection signal: The fatal infection or sepsis excess (0.31 vs 0.18 per 100 PY; P=0.02) raises the risk-benefit threshold for use in the general post-MI population, let alone healthy individuals. For rare genetic diseases where the alternative is severe recurrent autoinflammatory episodes, the trade-off is favorable; for broad cardiovascular prevention, it is not. long-term-unknown — long-term infectious risk with decades of use (relevant to aging-prevention use case) is unknown.

  2. Cost: Canakinumab is priced at approximately $150,000–200,000 USD per year for approved indications (orphan-disease pricing). This is economically non-viable for a chronic prevention drug in a general population, even at lower doses.

  3. Regulatory pathway: Novartis submitted but then withdrew a supplemental BLA for a cardiovascular indication. The drug is currently not FDA-approved for any cardiovascular or anti-inflammatory aging-related use outside the rare disease indications listed above.

  4. Competing biology: IL-1β is one node in a multi-cytokine chronic-inflammation network. Other interventions — statins, colchicine (shown in COLCOT/LoDoCo2 to reduce MACE), low-dose methotrexate (neutral in CIRT), and lifestyle interventions — offer overlapping benefit at far lower cost and risk. Colchicine in particular now has a CANTOS-comparable MACE signal at ~$5/month. See senomorphics for the comparative landscape.

This gap is foundational: CANTOS validated the mechanistic hypothesis (IL-1β drives vascular inflammation that causes clinical events) but the specific drug cannot be the translational vehicle due to cost and safety. The search for cost-effective, safer IL-1β pathway modulators remains active.

Cross-references

Limitations and gaps

  • No aging or longevity endpoint in any trial. CANTOS used MACE (cardiovascular events) and exploratory cancer as endpoints — not healthspan, frailty, or lifespan. Whether IL-1β neutralization slows biological aging per se is untested. needs-human-replication

  • Population specificity. CANTOS enrolled post-MI patients with elevated hsCRP — a selected high-inflammation population. Extrapolation to normative aging in otherwise healthy individuals is unestablished. needs-replication

  • Optimal dosing for non-autoinflammatory use unclear. CANTOS tested 50/150/300 mg q3mo. Whether lower doses (reducing infection risk) would preserve benefit is unknown. dose-response-unclear

  • CHIP subgroup finding not independently validated. The TET2-CHIP differential benefit is a single exploratory analysis. Prospective Phase 2 RCT NCT06691217 began recruiting April 2026 to test the hypothesis directly with vascular-inflammation imaging endpoint. needs-replication

  • CANTOS lung cancer signal failed to replicate in NSCLC therapeutic settings. As of 2026-05, all four prospective Novartis-sponsored CANOPY oncology trials (CANOPY-A adjuvant n=1,382; CANOPY-1 first-line metastatic; CANOPY-2 advanced post-platinum; CANOPY-N neoadjuvant) reported negative primary endpoints. Whether the original CANTOS signal reflects a prevention-specific effect (operating before clinical disease) remains untested in adequately-powered Phase 3 trials; ongoing Phase 2 prevention-setting trials (Can-Prevent-Lung NCT04789681, CANIFS NCT06038526, CCUS-prevention NCT05641831) may resolve this question. contradictory-evidence

  • No biomarker-guided patient selection strategy exists for identifying which aging individuals with elevated hsCRP would have a favorable risk-benefit profile. unsourced

  • PubChem CID: null — not applicable for this biologic; verify whether a biologics cross-reference entry exists in PubChem or another canonical registry. needs-canonical-id

Footnotes

Footnotes

  1. ridker-2017-cantos-primary · n=10,061 randomized (placebo=3,344; 50 mg=2,170; 150 mg=2,284; 300 mg=2,263) · rct · primary endpoint MACE: 150 mg HR 0.85 (95% CI 0.74–0.98; p=0.021, pre-specified threshold p<0.02115 met); 300 mg HR 0.86 (p=0.031, threshold p<0.01058 not met); 50 mg HR 0.93 (p=0.30, ns) · median follow-up 3.7 years · model: post-MI humans with hsCRP ≥2 mg/L · doi:10.1056/NEJMoa1707914 · PDF locally available ↩ ↩2 ↩3

  2. He D et al. · BMC Cardiovasc Disord 2025 Nov 10;25(1):798 · systematic review and meta-analysis of 25 RCTs (Jan 2014–Oct 2024) of immunomodulatory drugs in coronary heart disease; PROSPERO CRD42024597008 · primary endpoint MACE: overall RR 0.92 (95% CI 0.84–1.01, p=0.09; I²=60%, ns); IL-pathway inhibitor subgroup RR 0.86 (0.75–0.97, p=0.02); NLRP3-inhibitor subgroup RR 0.75 (0.65–0.86, p<0.0001); benefit only in trials with follow-up >6 months · GRADE quality of evidence: moderate · model: humans with coronary heart disease · doi:10.1186/s12872-025-05250-1 · PMID:41214504 · PMC:12604377 · PDF not yet downloaded ↩

  3. ridker-2017-cantos-lung-cancer · n=10,061 (sub-analysis) · rct (pre-specified exploratory) · p=0.0009 (300 mg lung cancer mortality) · model: post-MI humans · doi:10.1016/s0140-6736(17)32247-x · PDF not yet downloaded no-fulltext-access ↩

  4. Garon EB et al. · J Clin Oncol 2024 Jan 10;42(2):180–191 · n=1,382 (canakinumab 200 mg q3w SC × 18 cycles n=693; placebo n=689) · randomized double-blind phase 3 · primary endpoint disease-free survival HR 0.94 (95% CI 0.78–1.14; one-sided p=0.258; negative) · model: completely-resected stage II–IIIA/IIIB NSCLC after adjuvant cisplatin-based chemotherapy · doi:10.1200/JCO.23.00910 · PMID:37788412 · PDF not yet downloaded (R34 abstract-only verification) no-fulltext-access ↩

  5. Tan DSW et al. · J Clin Oncol 2024 Jan 10;42(2):192–204 · randomized double-blind phase 3 · canakinumab 200 mg q3w SC vs placebo each combined with pembrolizumab + platinum-based doublet chemotherapy as first-line therapy · primary OS endpoint not met (negative) · model: advanced/metastatic NSCLC, treatment-naive · doi:10.1200/JCO.23.00980 · PMID:38039427 · PDF not yet downloaded (R34 abstract-only verification) no-fulltext-access ↩

  6. Paz-Ares L et al. · Lung Cancer 2024 Mar;189:107451 · n=237 (canakinumab+docetaxel n=120; placebo+docetaxel n=117) · randomized double-blind phase 3 · primary endpoint OS not met (negative) · model: advanced NSCLC after platinum-doublet chemotherapy + anti-PD-1/PD-L1 progression · doi:10.1016/j.lungcan.2023.107451 · PMID:38354535 · PDF not yet downloaded (R34 abstract-only verification) no-fulltext-access ↩

  7. Lee JM et al. · JTO Clin Res Rep 2025 Jun 13;6(8):100859 · n=88 (canakinumab alone n=35; canakinumab + pembrolizumab n=35; pembrolizumab alone n=18) · randomized phase 2 neoadjuvant · primary endpoint major pathologic response (≤10% residual tumor cells): canakinumab alone 2.9% (1/35), canakinumab + pembrolizumab 17.1% (6/35), pembrolizumab alone 16.7% (3/18) — canakinumab did not improve MPR · model: resectable stage Ib–IIIa NSCLC · doi:10.1016/j.jtocrr.2025.100859 · PMID:40810132 · PMC: open-access · PDF not yet downloaded ↩

  8. Solomon BJ et al. · Cancer Res Commun 2025 Apr 1;5(4):632–646 · exploratory biomarker analyses across CANOPY-1 + CANOPY-N tumor samples (IHC, transcriptomics, multiplex IF) · low T-cell tumor infiltration (immunosuppressive TME) associated with PFS+OS improvements when canakinumab was added to pembrolizumab-based regimens · hypothesis-generating for biomarker-stratified IL-1β + ICI combinations · doi:10.1158/2767-9764.CRC-24-0490 · PMID:40116353 · PDF not yet downloaded ↩

  9. Wang Y et al. · Pharmacol Res Perspect 2026 Jun;14(3):e70251 · pharmacovigilance disproportionality analysis of FAERS database (2010 onward); canakinumab as primary suspected drug · n=9,262 adverse-event reports · age-stratified findings: <12 yr — pyrexia, GI/hepatobiliary disorders dominant; 12–17 yr — leukopenia; ≥65 yr — pneumonia, sepsis, cellulitis, strong neoplasm signals · ROR for infections and infestations 2.54 (n=3,532); for general/admin-site disorders 1.47 (n=6,765) · ~40% of AEs after >1 year of exposure · model: real-world post-marketing humans · doi:10.1002/prp2.70251 · PMID:41998856 · PMC:13090527 · PDF not yet downloaded ↩

  10. svensson-2022-chip-cantos · n=3,923 (genomic substudy); 338 (8.6%) with CHIP; TET2-CHIP subgroup: placebo n=31, canakinumab n=71 · pre-specified exploratory genomic substudy within rct · model: post-MI humans with hsCRP ≥0.20 mg/dL · doi:10.1001/jamacardio.2022.0386 · PDF locally available (PMC) ↩

  11. Wu H et al. · Front Cardiovasc Med 2026 Apr 10;13:1796328 · narrative review · CHIP–CVD mechanistic synthesis advocating “cardio-hematology” precision-medicine framework; identifies absence of prospective genotype-stratified IL-1β trials as the principal knowledge gap · doi:10.3389/fcvm.2026.1796328 · PMID:42039360 · PMC:13105890 · PDF not yet downloaded ↩

Graph View

Backlinks