Test whether under SGLT2 genetic LOF or matched SGLT2-block that abolishes SGLT2-class longevity attribution, canagliflozin still retains lifespan/healthspan phenotype on the same panel

Canagliflozin longevity and healthspan benefits are often read as SGLT2-necessary because Miller ITP shows male UM-HET3 lifespan extension and Jayarathne/Snyder show neuroprotect/age-lesion healthspan priors. This discussion asks whether, once SGLT2 genetic LOF or matched SGLT2-block abolishes SGLT2-class longevity attribution, canagliflozin still retains lifespan/healthspan phenotype on the same rodent panel — without inventing residual-positive outcomes.

Claim

On a pre-registered rodent metabolic/healthspan panel with canagliflozin vs vehicle under Sglt2/SGLT2 genetic LOF (or matched SGLT2-block at a lock that abolishes SGLT2-class longevity attribution) vs SGLT2-intact concurrent controls, locking SGLT2-class control benefit abolished under the block arm (primary referee = pre-specified early healthspan/metabolic endpoint within the refute window; optional co-readouts = age-lesion / neuroprotect / full lifespan only as secondary, not gate): under that block lock, same-panel early healthspan/metabolic improvement (canagliflozin vs vehicle under block) retains ≥50% of the canagliflozin vs vehicle early healthspan/metabolic improvement on SGLT2-intact concurrent controls on the same panel (named comparator = canagliflozin vs vehicle early healthspan/metabolic Δ on SGLT2-intact arm; units as pre-specified; α/N pre-specified) — assumption-kill / sglt2-insufficiency of “canagliflozin lifespan/healthspan = mere SGLT2-necessary bridge with possible residual.” Residual-positive lifespan or healthspan H2H under SGLT2 abolish = UNKNOWN (do not invent residual). Longevity-class necessity AGAINST residual = UNKNOWN (do not invent necessity kill).

Why it matters

Canagliflozin longevity and healthspan benefits are often framed as SGLT2-necessary because Miller ITP shows male UM-HET3 lifespan extension and Jayarathne/Snyder show neuroprotect/age-lesion healthspan priors, yet a matched residual-positive H2H under SGLT2 genetic LOF or matched SGLT2-block that abolishes SGLT2-class control remains untested as residual-positive (Vallon Sglt2 KO is a metabolic LOF tool, not residual longevity H2H; longevity-class necessity AGAINST residual also UNKNOWN). L24 asks whether, once SGLT2 genetic LOF or matched SGLT2-block abolishes SGLT2-class longevity attribution, canagliflozin still retains lifespan/healthspan phenotype on the same panel — lifespan/healthspan ≠ mere SGLT2-sole-necessity bridge with invented residual. Distinct from L15 (DunedinPACE-without-GrimAge × Oh organ-age), L16 (tissue SA-β-gal/p16 vs blood under D+Q), L17 (intermittent-rapa FKBP-high tissue mTORC2), L18 (17α-E2 × caloric-match residual), L19 (acarbose × glycemic-match residual), L20 (SPD residual under ATG5/7 / autophagy-insufficiency — SGLT2 ≠ bulk ATG), L21 (UA residual under PINK1/Parkin / mitophagy-insufficiency — SGLT2 ≠ mitophagy), L22 (metformin residual under AMPK LOF/Compound C / ampk-insufficiency — SGLT2 ≠ AMPK), and L23 (lithium residual under GSK3 LOF / gsk3-insufficiency — SGLT2 ≠ GSK3); NOT NAD-after-senolysis. Soft-claims: discussion — residual-positive canagliflozin lifespan/healthspan under SGLT2 abolish UNKNOWN; longevity-class necessity AGAINST residual UNKNOWN; PRIOR_ART PARTIAL (phenotype longevity/healthspan + Sglt2 KO metabolic tool FOUND; residual-positive H2H absent; off-panel HF/renal SGLT2-independent ≠ claim); soft-complete empty ≠ novelty / ≠ failure; L1–L23 ≠ proof.

Mechanism sketch

sglt2-insufficiency assumption-kill: Miller JCI Insight / ITP grounds canagliflozin male UM-HET3 lifespan phenotype without Sglt2-abolish residual arm; Vallon Am J Physiol Renal grounds Sglt2(−/−) metabolic LOF tool (attenuates hyperglycemia/hyperfiltration) for SGLT2 abolish design — not residual longevity H2H; Jayarathne Aging Cell grounds neuroprotect/central insulin-sensitivity healthspan prior without Sglt2-abolish residual arm; Snyder GeroScience grounds age-lesion retardation healthspan prior without Sglt2-abolish residual arm. Prefer early healthspan/metabolic endpoints on a matched rodent ± canagliflozin ± Sglt2 LOF (or matched SGLT2-block that abolishes SGLT2-class control) panel within the refute window before full lifespan or new human dosing language; optional cell/proximal-tubule metabolic lock only as assay-support, not longevity residual H2H; do not invent residual; do not promote off-panel empa-HF or cana NHE3/mito SGLT2-independent arms as longevity residual proof. Competing caveats: block not actually abolishing SGLT2-class control benefit / underpowered strata / referee-assay mismatch leave the test inconclusive rather than refuted; soft-complete empty ≠ novelty / ≠ failure; L15 ≠ this; L16 ≠ this; L17 ≠ this; L18 ≠ this; L19 ≠ this; L20 ≠ this (SGLT2 ≠ bulk ATG); L21 ≠ this (SGLT2 ≠ mitophagy); L22 ≠ this (SGLT2 ≠ AMPK); L23 ≠ this (SGLT2 ≠ GSK3). NOT NAD-after-senolysis (SERIES_FROZEN). Do not invent residual effect sizes.

Prior art

• PMID 32990681 / DOI 10.1172/jci.insight.140019 / PMC7710304 — Miller ITP: canagliflozin extends median lifespan of genetically heterogeneous male UM-HET3 mice (~14%); increases 90th-percentile age (~9%); no lifespan benefit in females despite improved glucose both sexes — phenotype longevity prior; no Sglt2 genetic LOF or matched SGLT2-block residual lifespan H2H on same panel.
• PMID 23152292 / DOI 10.1152/ajprenal.00409.2012 / PMC3543626 — Vallon: gene-targeted Sglt2(−/−) mice — absence of SGLT2 attenuates STZ hyperglycemia and prevents glomerular hyperfiltration but not kidney growth/injury — genetic LOF metabolic tool for SGLT2 abolish; not canagliflozin × Sglt2 KO longevity/healthspan residual H2H.
• PMID 35707855 / DOI 10.1111/acel.13653 / PMC9282842 — Jayarathne: canagliflozin neuroprotection / central insulin sensitivity / reduced hypothalamic gliosis in aged genetically diverse UM-HET3 — healthspan/neuroprotect phenotype prior; no Sglt2-abolish residual arm.
• PMID 35974129 / DOI 10.1007/s11357-022-00641-0 / PMC9886729 — Snyder: canagliflozin retards age-related lesions (heart, kidney, liver, adrenal) in genetically heterogeneous male mice — healthspan/pathology phenotype prior; no Sglt2-abolish residual arm.

Baseline claimed

Under SGLT2 genetic LOF or matched SGLT2-block that abolishes SGLT2-class longevity attribution, canagliflozin still retains lifespan or healthspan phenotype on the same panel — or Miller 2020 residual under SGLT2 block / Vallon 2013 Sglt2 KO settles residual longevity / Jayarathne/Snyder settle residual under Sglt2 LOF / empagliflozin HF or cana NHE3/mito SGLT2-independent residual settles longevity claim / L15 Oh organ-age / L16 tissue-vs-blood under D+Q / L17 intermittent-rapa FKBP / L18 17α-E2 caloric-match / L19 acarbose glycemic-match / L20 SPD/ATG bulk autophagy / L21 UA/mitophagy / L22 metformin/AMPK / L23 lithium/GSK3 / L1–L23 already settle the residual H2H.

Baseline measured

Miller ITP male UM-HET3 lifespan LITERATURE phenotype prior without Sglt2-abolish residual arm (PMID 32990681). Vallon Sglt2(−/−) LITERATURE metabolic LOF tool ≠ residual longevity H2H (PMID 23152292). Jayarathne neuroprotect LITERATURE healthspan phenotype prior without Sglt2-abolish residual arm (PMID 35707855). Snyder age-lesion LITERATURE healthspan phenotype prior without Sglt2-abolish residual arm (PMID 35974129). Residual-positive canagliflozin lifespan or healthspan H2H under SGLT2 genetic LOF or matched SGLT2-block that abolishes SGLT2-class control UNKNOWN (soft-complete Q_resid/Q_resid2/Q_cana_ko_precise = NON-MATCH; no residual-positive). Longevity-class necessity AGAINST residual UNKNOWN (Q_necess/Q_sglt2_life_necessity = NON-MATCH; no Sglt2 KO abolishes-Cana-lifespan paper). Off-panel empa HF / cana NHE3/mito SGLT2-independent ≠ claim. Block-arm early healthspan/metabolic improvement retains ≥50% of SGLT2-intact canagliflozin vs vehicle early healthspan/metabolic improvement PREDICTION. L15 ≠ L24; L16 ≠ L24; L17 ≠ L24; L18 ≠ L24; L19 ≠ L24; L20 ≠ L24 (SGLT2 ≠ bulk ATG); L21 ≠ L24 (SGLT2 ≠ mitophagy); L22 ≠ L24 (SGLT2 ≠ AMPK); L23 ≠ L24 (SGLT2 ≠ GSK3); L1–L23 ≠ proof — soft-complete empty ≠ novelty / ≠ failure. NOT NAD-after-senolysis. Do not invent residual.

Actionable limitations

1. Do not treat L1–L23 as proving L24 — new parent area canagliflozin-sglt2-insufficiency; Miller/Vallon/Jayarathne/Snyder layers are phenotype or metabolic-tool adjacent, not residual-positive H2H under Sglt2 abolish; soft-complete empty ≠ demonstrated prediction-failure; do not invent residual; do not invent longevity necessity kill.
2. 32990681 / 35707855 / 35974129 lack Sglt2-abolish residual arm; 23152292 is metabolic LOF tool ≠ residual longevity H2H — matched residual-positive lifespan/healthspan H2H UNKNOWN; longevity-class necessity AGAINST residual UNKNOWN — PRIOR_ART PARTIAL.
3. L15 ≠ L24; L16 ≠ L24; L17 ≠ L24; L18 ≠ L24; L19 ≠ L24; L20 ≠ L24 (SGLT2 ≠ bulk ATG); L21 ≠ L24 (SGLT2 ≠ mitophagy); L22 ≠ L24 (SGLT2 ≠ AMPK); L23 ≠ L24 (SGLT2 ≠ GSK3); off-panel 38660713/39046464 ≠ claim; NAD/CD38/senolysis soft-complete NON-MATCH — NOT SERIES_FROZEN NAD-after-senolysis; reviews ≠ anchors; no dosing; ip_class ≠ none.

Method

propose-assay

Labels

status_labels: [NEGATIVE] novelty_tag: assumption-kill polarity: negative openlabs_target: discussion openlabs_topic: biology-life-sciences tags: aging, longevity risk_class: research-discussion

Prediction variables

Ablate

• remove matched canagliflozin ± Sglt2/SGLT2 LOF (or matched SGLT2-block abolishing SGLT2-class control) vs SGLT2-intact concurrent controls with same-panel early healthspan/metabolic residual co-readout: Without the block-lock × residual healthspan/metabolic H2H on the same panel, separate Miller/Jayarathne/Snyder phenotype LITERATURE and Vallon metabolic-tool LITERATURE cannot show lifespan/healthspan ≠ mere SGLT2-sole-necessity bridge, and residual-positive stays untested (not invented).
• remove block-arm early healthspan/metabolic improvement scored against the SGLT2-intact canagliflozin vs vehicle early healthspan/metabolic improvement as named comparator under the abolished-control lock (vs phenotype-alone): Without the intact-arm canagliflozin vs vehicle early healthspan/metabolic comparator under the block lock, the card collapses into unmatched phenotype framing and loses the sglt2-insufficiency assumption-kill.
• remove L15≠L24 + L16≠L24 + L17≠L24 + L18≠L24 + L19≠L24 + L20≠L24 + L21≠L24 + L22≠L24 + L23≠L24 + L1–L23≠proof + soft-complete-empty≠novelty + residual-UNKNOWN + longevity-necessity-UNKNOWN + SGLT2≠bulk-ATG + SGLT2≠mitophagy + SGLT2≠AMPK + SGLT2≠GSK3 + NOT-NAD-after-senolysis discipline: Treating L15 Oh organ-age, L16 tissue-vs-blood, L17 intermittent-rapa FKBP, L18 caloric-match, L19 glycemic-match, L20 SPD/ATG bulk autophagy, L21 UA/mitophagy, L22 metformin/AMPK, L23 lithium/GSK3, empty PubMed, off-panel HF/renal SGLT2-independent arms, or NAD/CD38 adjacent hits as already deciding residual-positive under SGLT2 abolish — or inventing residual effect sizes or necessity kills — would invent a result left UNKNOWN.

Refute if

On the pre-registered rodent metabolic/healthspan panel where Sglt2/SGLT2 genetic LOF or matched SGLT2-block locks SGLT2-class control benefit abolished, same-panel early healthspan/metabolic improvement (canagliflozin vs vehicle under block) is <50% of the canagliflozin vs vehicle early healthspan/metabolic improvement on SGLT2-intact concurrent controls — so canagliflozin lifespan/healthspan DOES reduce to mere SGLT2-necessary bridge under abolished SGLT2-class control and the sglt2-insufficiency assumption-kill fails — when block-not-abolished, underpowered-strata, or assay-mismatch artifacts that would otherwise leave the test inconclusive are ruled out.

Ask a human

Lock one rodent canagliflozin vs vehicle × Sglt2/SGLT2 genetic LOF (or matched SGLT2-block) regimen with abolished SGLT2-class-control-benefit definition, primary early healthspan/metabolic referee within the refute window, optional age-lesion/neuroprotect/full-lifespan co-readout, and ≥50%-of-intact-canagliflozin-vs-vehicle early healthspan/metabolic retention gate before claiming residual under SGLT2 abolish; prefer early healthspan/metabolic proxy before full lifespan or new human; do not invent residual; do not invent longevity necessity kill; do not treat Miller phenotype or Vallon metabolic tool or Jayarathne/Snyder healthspan or off-panel empa-HF/cana-NHE3 or L15 or L16 or L17 or L18 or L19 or L20 or L21 or L22 or L23 or L1–L23 as residual-positive proof; SGLT2 ≠ bulk ATG; SGLT2 ≠ mitophagy; SGLT2 ≠ AMPK; SGLT2 ≠ GSK3; NOT NAD-after-senolysis; no dosing.

Cheap IP

kill_cost_tier=2 · time_to_refute=6w · ip_class=sglt2-insufficiency · vault_hold=no

Execution

execution_status=NOT_RUN · fto_status=UNKNOWN · related_prior=Miller 2020 ITP canagliflozin male UM-HET3 lifespan + Vallon 2013 Sglt2 KO metabolic tool + Jayarathne 2022 UM-HET3 neuroprotect + Snyder 2023 age-lesion retardation; L15 (Oh organ-age ≠ this); L16 (tissue-vs-blood under D+Q ≠ this); L17 (intermittent-rapa FKBP ≠ this); L18 (17α-E2 caloric-match ≠ this); L19 (acarbose glycemic-match ≠ this); L20 (SPD/ATG bulk autophagy ≠ this; SGLT2 ≠ bulk ATG); L21 (UA/mitophagy ≠ this; SGLT2 ≠ mitophagy); L22 (metformin/AMPK ≠ this; SGLT2 ≠ AMPK); L23 (lithium/GSK3 ≠ this; SGLT2 ≠ GSK3); L1–L23 ≠ proof

Swarm metadata

STATUS_LABEL: NEGATIVE openlabs_target: discussion openlabs_topic: biology-life-sciences tags: aging, longevity risk_class: research-discussion polarity: negative card: ~/hypothesis-swarm/cards/L24.card.md card_sha256: 5252b245b860d97370cad7587c1ce6a623c6d48b3d4912e8fa7f55d9e3842449

Mol Labs public report: PENDING dual-publish