Lifetime synthesis of the L1–L10 pathway — integrating the corpus does not produce a DAG; it produces a negative grammar and a classifier residual

1. Executive scientific abstract

Verified evidence does not support L1–L10 as a life-course mechanism. The roadmap ordered the laboratory — axes, timing, variation, metabolites, ovarian frontier — and that construction sequence is not a biological DAG. Of 17 feeds arrows, zero close identity, flux, compartment, independent function and non-common-cause at once (S1-EMAP-C1). The only object the audit left alive is not a cascade: it is a calendar residual whose axis matters. In SWAN, PdG/ELA fall during the transition (Santoro 2008, PMID 18285413: ELA 80.9%→64.7%; analytic n 834/700/624; PdG −6.6%/year in all collections vs −5.4%/year ovulatory) while DHEA-S falls with age (−2.81%/year cross-sectional) and rises +3.95% early→late when aligned to ovarian status (Crawford 2009, PMID 19470626, PMC2730879, reconfirmed this turn: n=2 886; 15 930 observations; P=0.003; 1 202/1 423=84.5% within-woman increase estimated from status random effects). A CYB5A territory that contracts with age (Parker 2005: 41.6%→30.1% of cortex in 20 vs 13 women; Tezuka 2021 r=−0.76, n=60) predicts a decline, not that stage rise. Opposite HSD3B2 sign (high in corpus luteum, low in zona reticularis) is textbook zonal identity, not the residual’s mechanism: Nanba 2019 already describes HSD3B2–CYB5A overlap as ZR involutes.

Adversarial review split the generation hypotheses. S1-HG-H1 v1 was a five-clause conjunction that survived if any clause lived; the five-bit audit makes any observational edge unobservable (S1-AR-C1, S1-AR-C2). S1-HG-H2 v1 read “calendars that do not copy” when Crawford copies status with inverted sign (S1-AR-C4). The operating hypotheses of this report are S1-AR-H1 v1 (non-substitution grammar and curriculum, not a rate-limiting node), S1-AR-H2 v1 split into (2a) bleeding-classifier circularity, (2b) extra-ovarian anti-phase inflection and (2c) frame asynchrony — not urine–serum identity — and parked S1-AR-HT1 v1. S1-HG-H0 v1 (L1–L10 cascade) remains superseded.

This turn reread the metrological anchor. Santoro 2020 (Menopause, not JCEM; PMID 31794501, PMC7050767): same-day serum P and uPdg give Pearson r=0.837 (n=170; β=1.10; R² 0.74). The r=0.26 (n=122; R²=0.09) is integrated luteal uPdg versus one timed P on days 16–24. Copying r=0.26 as an identity CCC kills the wrong discriminant (S1-ED-C1). The residual that can still be a meter is aligning an annual follicular serum (days 2–5) to a bleeding binary that is the luteal clock, while PdG lives on another calendar (S1-AR-C3; S1-ED-C2). Santoro 2017 (PMID 28368525: n=511, 1 987 cycles; ELA 87.9% at −5 years from FMP vs 22.8% at −1) already measured annual DHEA-S in the same DHS cohort and did not residualize the classifier.

Computation was omitted (decision: skip). GEO 2026-10-05: zero peri/STRAW × adrenal series; zero paired ovary–adrenal transcriptomes; the twelve ovary×adrenal×aging female hits are false friends (S1-CD-C1). Honda 2024 (PMID 38742994): n=115 enrolled, PPS 57/54; fecal GUS does not differ between arms (P=0.253); the abstract swaps E1/E2. Albert 2008: WAFACS RR 1.03 (0.90–1.19) in 5 442 women. Geng 2025: n=45 self-controlled PCOS, off-contrast. G0-COMBO=0. G8 closed. Lua data prohibited.

Scientific delta of this report (S1-REPORT-EN-C1): the English twin freezes the same bilingual factual base that REPORT_ES locked, correcting the recorded numeric-parity failure without changing a DOI, PMID, unit or conclusion. The three project hypotheses are those of adversarial review, not the generation conjunction and not SCOPE-H1 as a cascade. This turn reconfirms PMC7050767 (r=0.837 same-day vs r=0.26 integrated-vs-spot), PMC2730879 (+3.95%; 1 202/1 423) and Nair 2006 (PMID 17050889: 57 women, 27×50 mg DHEA vs 30 placebo, DHEA-S +3.8 µg/mL, no VO2, strength, composition, insulin or quality of life). The smallest discriminant is not a tissue n of HSD3B2 and not a FASTQ: it is residualizing the bleeding classifier in already-collected SWAN (annual ICPSR → Biobank FMP-time → DHS∩DHEA-S), without recruiting and without Lua data. Both drafts carry mechanism and experiment in headings for bilingual QA.

The operating conclusion is negative and actionable. There is no master pathway to partner. If (2a) wins, the residual was the classifier. If (2b) survives, the next scientific euro is on-target function (L8-3 D0-A), not an NAD×senolytic combo and not a DHEA supplement. Restoring concentration is not function: Nair already measured that outside peri.

2. Scientific question and relevance

The question is what mechanisms, contradictions and opportunities emerge when integrating the entire L1–L10 pathway. Answering it requires separating objects the title fuses:

1. Is S1 a life-course mechanism, an inventory of contradictions, or a substitution audit?
2. Is the roadmap order a DAG or pedagogy (S1-SCOPE-C2)?
3. Does L1–L5 enter as causal base or as epistemic debt (S1-SCOPE-C3)?
4. Is the only shared human time STRAW+10 −2/−1, or are age, FMP, atlas young/old and product personas being used as if they were the same clock?
5. After the L6–L10 non-substitution grammar, does a residual edge remain, narrative/common-cause coupling, or a shared negative architecture?
6. Are Honda, Geng, urolithin, butyrate–ESR and “NAD×senolytic synergy” edges or program intention (L10-5-K1, L10-6-K1)?
7. Do age, adiposity, stage and inflammation explain covariation better than a shared node?
8. Does G8 open by stacking lines? No. An organ event with mediation and competing risks remains H0.
9. Does a multi-clock index or a product n fill the gap? Forbidden.
10. Only if a qualified edge survives: is there function independent of the proxy in a named tissue?

Relevance is high because “integrate L1–L10” is a cheap program slogan. A curriculum is not a mechanism. Blood is not tissue. PdG is not production. DHEA-S is not dose. Fecal GUS is not ovarian E2. CDKN1A+ is not senescent load. Blood NAD is not ovary. Later menopause is not net longevity (Ruth 2021, PMID 34349265). Observation is not do(A) (Greenland 1999, PMID 9888278; L7-5; L10-6). The correct unit is the woman classified by STRAW+10, not the subtopic, not the nucleus, not the end-user.

3. Scope, population and life stage

The primary biological estimand is the adult woman in natural reproductive aging, STRAW+10, not chronological age, not product personas. The transition (−2/−1) is the stratum where several lines name a change (luteal exposure, adrenopause vs menopause, historical allostasis) without those names sharing a qualified node. Late reproductive (−3b/−3a) is reference; early postmenopause (+1c) is replica, not mixture: Farr, KEEPS and extra-ovarian NAD trials live there and are not peri. The inferential unit is the woman. Cycles, visits, nuclei, stool, embeddings and analytes are not replicates. A Lua-user arm does not exist.

Preclinical material and historical L1–L5 enter as claims to classify, not as n of the synthesis protocol. S1 does not gavage, does not recruit, does not dose. PCOS, pregnancy, lactation, POI, ovarian cancer and transported male evidence sit outside the primary contrast. Iwahashi 2024 (PMID 38718896: one 27-year-old woman with pheochromocytoma vs two men aged 64 and 74) is not female peri n. Parker is age autopsy, not STRAW. Nanba is pre 20–40 vs post ≥60, a ~20-year gap.

Excluded: prescription, doses, stacks, current Lua data, PHI, PII, private genomes, food-logs, personas, hooks, features, “first LATAM dataset” as a finding, reopening local D0s as S1 tubes, concatenating Honda+Santoro+Farr+Jin, p>0.05 as equivalence, and calling AMH, symptoms, cycle or menopause healthspan. G8 remains closed. cpr::spdg is simulated PDG; using it as n of (2a) is a category fraud.

4. Background knowledge and mechanism map

4.1 Operational definitions

• L1–L10 pathway: curricular sequence of 10 lines × 6 subtopics. It is not a DAG. It is not a finding.
• Synthesis: an operation that can kill edges, not only join them.
• Non-substitution grammar: the unified K-rule set from L6-2 to L10-6. It is not negotiated per line.
• Program intention: a roadmap phrase or L1–L5 report phrase the laboratory did not consolidate as evidence (L10-5-K1, L10-6-K1).
• L1–L5 debt: a corpus closed in hallazgos.md without an autonomous contradiction-ledger. It enters for re-labeling, not as H4.
• STRAW+10 −2/−1: a bleeding calendar (Harlow 2012, PMID 22344196). It is not a senescent ceiling, not FMP, not Jin young/old.
• Admissible residual: an edge that survives identity, flux, compartment, independent function and non-common-cause. Today: a calendar object, not a master pathway.
• Bleeding classifier: STRAW/WHO-SWAN. It is the luteal clock. Entering it as a covariate of the status estimand nests the object (S1-AR-C3).
• Independent function: a readout that is not the proxy.
• Longevity / healthspan: an organ event, mediation and competing risks. Remains H0.

4.2 Favored chain the literature would have to prove — and does not

Title = "pathway L1–L10"
  requires: a shared node that is not age / adiposity / STRAW / inflammation

Audit (layer 0, closed)
  17 feeds → 0 five-bit edges
  29/29 autonomous governing questions = local
  G0-COMBO = 0
  Honda fecal GUS null between arms

Sources (not a node)
  luteal: HSD3B2 HIGH → P4   | calendar: PdG/ELA in MT
  ZR:     HSD3B2 LOW + CYB5A/SULT2A1 → DHEA-S | calendar: age; +3.95% if aligned to status
  FMP:    E2/FSH (Randolph) | parallel follicular topology
  SASP:   binary stage, 0/28 × AMH (Shin)
  ovarian NAD: ABSENT

FORBIDDEN as a bridge
  fecal GUS = ovarian E2     (Honda P=0.253)
  CDKN1A = senescent census  (Sugimoto; Coppé; Shin)
  NAD × senolytic            (G0-COMBO=0)
  PdG = production; DHEA-S = dose; blood = tissue

The only licit sum synthesis left open is occupancy of active ligand at a named receptor in a named extra-ovarian tissue (S1-MECH-C5). That is not L1 feeding L10. It remains INF. It does not open HT1.

5. Evidence method

Unit: the woman. Classes: H human, A animal, V in vitro/ex vivo, C computational/atlas, INF inferred across non-joint studies. Abstract does not beat the body. Binary pre/post is not STRAW. p>0.05 is not equivalence. “Feeds” is not an edge.

D0-CROSSLINE-SUBSTITUTION-AUDIT labeled each roadmap arrow by node, class, K non-substitution, non-common-cause, non-debt and ethics. EVIDENCE_VERIFICATION closed n/RR against primaries (Honda body, Albert PMC, Santoro Table 1, Geng n=45). MECHANISTIC_SYNTHESIS located the HSD3B2 fork. COMPUTE_DECISION re-audited GEO and skipped. HYPOTHESIS_GENERATION welded H2+H3; ADVERSARIAL_REVIEW split them. EXPERIMENT_DESIGN rewrote (2c) and froze ICPSR→Biobank. This report does not reopen local D0s, does not download microdata and does not sign a DUA.

Sources reconfirmed this turn (2026-10-06): PMC7050767, PMC2730879, PMID 17050889. Supplement hits, NutraIngredients and product personas do not vote.

6. Evidence map

6.1 Arrow inventory — G0-AUDIT cut

H2 wins the count. H3 wins the question mode (29/29 local; ≥70% threshold frozen). H1-cascade is not tested.

6.2 Human calendar residual

Three human clocks (luteal, FMP, DHEA-S) do not coincide if the axis is chosen. A fourth (SASP/AMH) does not couple reserve. A fifth (ovarian NAD) is unmeasured. That is desynchrony of meters and axes, not a master pathway.

6.3 L1–L5 debt reopened, not promoted

Honda 2024 (H, industry): Japanese women 45–55 y, peri and post mixed, not STRAW+10. 115 randomized; PPS 57/54. Body: E1 31.62±7.97 vs 25.12±8.17 pg/mL; E2 21.38±8.57 vs 13.18±8.77. Abstract swaps the labels. Between-group fecal GUS P=0.253; placebo raises GUS 0.23→0.46. mKMI improves in both, zero between-group difference. S03 prohibited with a mediator null.

Geng 2025 (H+A): n=45 PCOS, 29.5±3.1 y, inulin 10 g/d × 3 months, self-controlled. PCOS is not S1 n.

7. Contradictory evidence and null findings

O'Connor 2009 and Grub 2021 bound the luteal clock: anovulation and PDG change mainly in late peri; early vs late saliva did not predict E2 or P4. Waugh 2007 (mean luteal P4 null vs ovulatory status) is women 21–40 y: degraded as peri.

8. Multiscale mechanistic synthesis

8.1 Genes and the fork that kills the cascade

Luteal P4 requires high HSD3B2 after a serial HPO gate (competent follicle × LH × rupture × vascularized CL). ZR DHEA-S requires low HSD3B2 plus CYB5A/SULT2A1 territory. A shared steroidogenic blackout cannot lower both at once (S1-MECH-C1). That blocks an L7–L8 node and, a fortiori, an L1–L10 cascade.

Parker/Dharia 2005 (PMID 15870126): 20 women 21–40 y vs 13 women 50–91 y; cortical CytB5 41.6±6.4% → 30.1±5.9%, P<0.0001; intensity 2.2 vs 2.2. Territory, not cellular blackout. Sex-independent. Not STRAW. The authors note that ~25% territorial loss does not cover a 33–75% androgen decline.

Nanba 2019 (PMID 30753518): 100 pre 20–40 vs 100 post ≥60; DHEA-S 2.8×; 11-oxy C19 do not fall; 8+8 adrenals with ZR involution and HSD3B2–CYB5A overlap. Cortex does not fail as a block. Gap ~20 years: it is age, not FMP.

Adversarial review decouples this fork from the SWAN residual. Territory↓ predicts DHEA-S↓; the residual is stage-wise DHEA-S↑ (S1-AR-C5). The enzymatic sign survives as zonal identity, not as the calendar mechanism. A paired luteal–ZR n in −2/−1 does not exist (GEO=0) and is not fabricated from atlases.

CDKN1A/p21 is luteal-cycle exit (Sugimoto 2003, A: the double null preserves enzymes and receptors) and reversible QASP (Meinsohn 2025, A/V). Coppé: arrest without SASP. Shin: 0/28 SASP×AMH. Kill of R14. Luteal senolysis is not rehabilitated.

8.2 Cells, tissues and systems

Ovary — two clocks, distinct topologies (S1-MECH-C3). Follicular E2 does not require ovulation (parallel architecture of residual follicles). High cyclic P4 is a serial product. Randolph anchors FSH/E2 to FMP. Santoro anchors ELA/PdG to the MT, with ELA still high at −5 years. There is no paired P4–E2 change point (L7-4-K1).

Adrenal — branches that do not age together. ZF/CYP11B1 (11-oxy) persists; ZR contracts with age. Crawford: DHEA-S does not copy FMP if the axis is time; it copies status if the axis is WHO/SWAN. BSO (Lasley 2011 Menopause, 20/81 of 1 272 eligible; 14/20 with a 5–8% rise) excludes ovarian necessity of the inflection; it does not measure production; it is not inflated to prevalence.

Gut. Bacterial GUS can deconjugate estrogens in vitro. Honda does not move fecal GUS between arms. 21-dehydroxylation (biliary cortisol→fecal P4) is not luteal production.

Target. L_ER and L_AR can mix ovarian and intracrine ligand. A nonlinear f generates an E2 × DHEA-S term without the adrenal “knowing” the ovary. Guthrie: DHEA-S null vs E2/FSH for BMD. Adiol (Lasley 2012, PMID 22415563) is a circulating rival ligand, not measured occupancy. KEEPS/Faubion 2020 is +1, a precedent of do(ligand), not S1 n.

8.3 Direct versus inferred

9. Computational layer

decision: skip. NCBI GEO/PubMed, 2026-10-05: 0 peri/STRAW × adrenal series; 0 peri/STRAW × HSD3B2/CYB5A; 0 PMIDs of paired ovarian+adrenal transcriptome in women. The 12 ovary×adrenal×aging female UIDs are breast, Alzheimer, mouse ENCODE and Affymetrix platforms. The three “peri × ovary × RNA” series are ungulate. GSE329417 is +1 ovary 50–84 y, no adrenal. GTEx bulk would mask the fork (ZF-dominant = high HSD3B2). BioNeMo does not enter: the same protein does not demonstrate opposite zonal signs. Concatenating Jin+GSE253849+GSE329417 would fabricate the cascade H2/H3 already killed (S1-CD-C1–C3). COMPUTE_OPTIONAL archives the skip. It is not reopened.

10. Primary hypothesis

S1-AR-H1 v1 — integrating L1–L10 produces a non-substitution grammar and a curriculum, not a rate-limiting node

Role: operating primary, split from S1-HG-H1 v1 (weakened). Absorbs inventory, grammar and slogans. Does not absorb “zero edges” as a biological discovery nor common cause as closure of the residual.

Falsifiable statement: in adult women, (i′) the title “pathway L1–L10” names a laboratory order; re-labeling L1–L5 claim-by-claim does not add an edge that closes flux and independent function at once in two non-Lua sources; (iii) ≥70% of autonomous governing questions (already 29/29) remain local to flux, compartment, identity or function; (iv) NAD×senolytic synergy, fecal GUS = ovarian E2 and “LATAM dataset” remain intention or an internal null; (v) concatenating Honda+Santoro+Farr+Jin, or Iwahashi+Parker+Santoro, dies by category. The common-cause predicate on calendars is a separate estimand (AR-H2).

Lineage: S1-HG-H1 v1; S1-SCOPE-H3 v1; S1-EMAP-C1; S1-EVER-C1; S1-EVER-C6; S1-AR-C1; S1-AR-C2. Honda PMID 38742994. Albert PMID 18460663. Greenland PMID 9888278.

Mechanism: the recurrent error is not a missed gene. It is substituting blood for tissue, ratio for flux, observation for do(A), menopause for healthspan, and a serial slogan for a 2×2. Integrating the corpus makes that grammar visible; it does not fabricate a rate-limiting node.

Status: proposed_primary. Maturity: methodological H1 for grammar and inventory; H0 for the numeric DAG. Not computational H2. Confidence: 0.74 that L1–L10 is not a biological DAG of sources; 0.31 that common cause exhausts the residual (withdrawn from 0.48).

11. Competing hypothesis

S1-AR-H2 v1 — the residual is not HSD3B2 independence: it is (2a) classifier circularity, or (2b) extra-ovarian anti-phase inflection, or (2c) frame asynchrony

Role: competing contrary to reading AR-H1 as cross-biology silence, and contrary to HG-H2 v1 as the enzymatic mechanism of the calendar. Does not rehabilitate the cascade.

Falsifiable statement: in already-measured SWAN women (DHS ∩ annual DHEA-S; woman=unit; not Lua), (2a) the +3.95% DHEA-S early→late inflection disappears on replacing the stage/bleeding classifier with continuous FMP-time and with PdG/ELA as continuous exposure, age+BMI without STRAW; (2b) the inflection survives (2a) with opposite sign to PdG; (2c) same-day P–PdG CCC passes (≥0.70) and the residual dies on requiring DHEA-S and PdG from the same cycle, or the residual is the integrated-vs-spot failure. Exactly one of (2a)(2b)(2c) must win the first cut. A shared HSD3B2 blackout is not a prediction: DHEA-S↑ with PdG↓ is what that blackout does not produce.

Lineage: S1-HG-H2 v1 (weakened); S1-SCOPE-H1 v1 restricted to residual; Crawford PMID 19470626; Lasley PMID 21961714 and 21178790; Santoro PMID 18285413, 28368525, 31794501; Tezuka PMID 33524149; Allshouse PMID 25714237; S1-AR-C3–C5; S1-ED-C1–C2.

Internal ranking, not probability: (2a) > (2b) > (2c), because the classifier is the luteal clock, 84.5% is a model, and same-day r=0.837 makes (2c-i) improbable. Confidence 0.41 in that order. Confidence 0.64 that “they do not copy” was an axis error. The 0.38 HG-H2 confidence that the fork explains the uncoupling is withdrawn.

Status: proposed_competing. Maturity: observational H1 for the status inflection; H0 for circularity vs anti-phase vs frame.

12. Translational hypothesis

S1-AR-HT1 v1 — a multi-clock index does not enter; the only licit sum is occupancy at a named receptor in a named extra-ovarian tissue; collapsing by equating ligand is tautological until occupancy is measured

Role: translational, rewritten, parked. HUMAN_QA_REQUIRED. Not called a biomarker. Does not prescribe. Does not recruit. G8 closed.

Falsifiable statement: (i) a parameter concatenating PdG, DHEA-S, SASP/p16, blood NAD, fecal GUS or L1–L5 digital scores will not out-of-sample predict an event of one organ better than STRAW+10 + age + adiposity + one inflammatory marker, with competing risks; (ii) if functional coupling exists, it is read as an E2 × DHEA-S term (or homolog) on extra-ovarian function that collapses on equating occupancy — and that collapse is not called a result until a prespecified occupancy proxy; Adiol is named as a rival ligand; (iii) NAD×senolytic, Honda, food-log and Lua dataset do not enter; (iv) menopause, AMH, symptoms and extra-ovarian null CTx are not the event; (v) marking H5, prescribing P4/DHEA/NMN/senolytic, or calling a proxy a biomarker, dies by category.

Lineage: S1-SCOPE-HT1 v1 (not promoted); S1-HG-HT1 v1 (weakened); S1-MECH-C5; Guthrie PMID 15042284; Farr PMID 38956196; Albert PMID 18460663; Nair PMID 17050889; Faubion PMID 32857481; Lasley PMID 22415563.

Status: parked_translational. Maturity: H0. Not H5. Confidence 0.62 that the index does not beat STRAW+adiposity+inflammation; the occupancy-collapse prediction carries no numeric confidence.

13. Falsifiable predictions and kill criteria

Global ethical kill: Lua n, prescription, non-indicated ovary, cpr::spdg as n, STRAW in the adjustment of the status estimand, P>0.05 as equivalence, 84.5% as a census, BSO 14/20 as prevalence, marking H5.

14. Discriminating experiment

Does not recruit. Does not use Lua data. Does not reopen E0, ED3-FMP, Q0-R2-IDENTITY, G0-COMBO, concatenated GEO or Iwahashi as an S1 tube.

Layer 0  D0-AUDIT          already closed (17 vs 0; 29/29)
   ↓
Layer 1a D0-STATUS-AGE     annual ICPSR: status vs age? Crawford replica, no PdG
   ├─ not comparable (>30% visits missing) → Biobank without calling a null
   └─ replicates +3.95% by status and −age
        ↓
Layer 1b D0-FMP-TIME       annual Biobank: Y ~ f(t_FMP) + age + BMI. No status. No STRAW
   ├─ inflection dies (TOST residual ∈ ±1.5 pp) → (2a) wins; do not spend PdG
   └─ inflection survives
        ↓
Layer 1c D0-CLASSIFIER     DHS ∩ DHEA-S: continuous cycle PdG. Unit = woman
   ├─ dies on entering PdG → (2a)
   ├─ survives with opposite sign → (2b)
   └─ n < floor or frames unpaired → not estimable, not a null
        ↓
Layer 2  D0-FRAME          same-day CCC (luteal pilot already measured) and d2–5 vs cycle asynchrony
   ├─ CCC <0.40 → (2c-i) identity; rare given r=0.837
   ├─ CCC passes and residual dies in the same cycle → (2c-asynch)
   └─ CCC passes, same cycle, residual lives → (2c) dies; (2b) remains
        ↓
Layer 3  D0-OCCUPANCY      names L8-3 D0-A. Not executed. G8 closed.

Floors. 1b: 400 women with FMP and at least one pre/early visit and one late or early post. 1c confirmatory: 200 women with both status windows and PdG. Strict pairing: 80 women with DHEA-S on days 2–5 of the same DHS cycle. Below that: not estimable. Allshouse 2015 (PMID 25714237) already warned that DHS and the annual visit are not simultaneous; pairing is an estimand. Access 2026-10-06: ICPSR V00–V10 without DHS; NIA Aging Research Biobank for DHS+FMP. Layer 3 opens only if (2b) wins and a human reviewer authorizes L8-3.

This experiment does not demonstrate menopause, fertility, senolysis or longevity.

15. Biomarkers and stratification

No analyte in this report is called a biomarker.

Prespecified stratification (heterogeneity, not rescue): SWAN ethnicity, BMI <25 vs ≥25, smoking, ±60 d window vs strict. LATAM ancestry / Lua dataset does not enter. Inflammatory marker (hs-CRP/IL-6) is sensitivity, not primary: it may be a stage mediator.

16. Individual variability

Crawford: status×ethnicity interaction P=0.31; ethnicities are not pooled to fabricate n. Randolph: obesity attenuates FSH and delays onset. Santoro 2008: obesity OR 4.68 (1.33–16.52) for bleed-with-anovulation vs ovulatory — common cause on two ovarian clocks, not the PdG–DHEA-S edge; wide CI. Attrition biased toward anovulatory women can fabricate less luteal change. Tepper: four E2 trajectories; obesity favors flat E2. The 84.5% does not authorize saying “almost all women compensate.” Completers with observed FMP are not the enrollment. Honda mixes peri+post; baseline E2 variance (80.9±114 vs 103±148 pg/mL) is the signature of that mix, not a phenotype.

17. Pharma relevance and maturity

Integrating L1–L10 does not deliver a cascade target. The opportunity is negative if (2a) and conditional if (2b). There is no readiness. There is no indication, compound or dose. HUMAN_QA_REQUIRED before any partnering. Maximum S1 maturity: observational H1 for the status inflection; H0 for circularity vs anti-phase vs frame; methodological H1 for the grammar. Not H5.

S1 Pharma kill: (2a) wins; Nair-like functional null; tautological occupancy; effective exposure not traceable to peri; any longevity claim. The useful reading for a committee: there is no master pathway to partner; if (2b) survives, the next euro is L8-3 (on-target function), not a combo and not a DHEA supplement.

18. Limitations

• Numeric G3-CONF was not run. Layers 1a–1c are designed, not executed. Strict pairable DHS∩DHEA-S n is unknown until a DUA; 511/1 987 is a ceiling, not analytic n.
• FMP-time and late peri are collinear near 0; 1b power lives in −5 to −2 years (ELA still high).
• ELA is circular with PdG. L7-4 ultrasound is not in SWAN.
• 84.5% remains random effects. P=0.06 of the −0.33%/year is not equivalence.
• BSO n=20 selected. Ovariectomized macaque is primate coupling, not peri n.
• Parker is autopsy; Nanba 8+8; Tezuka does not identify STRAW; Iwahashi is 1 woman vs 2 men; Sugimoto/Meinsohn are mouse; Pignatti is adrenarche in vitro.
• CCC will be computed on historical ACS-180, not LC–MS/MS. Pearson r=0.837 is not CCC.
• Honda is industry, exploratory, peri+post mixed. Geng is self-controlled PCOS.
• L1–L5 is not re-labeled claim-by-claim. Partial G1-DEBT.
• Nair is ≥60 years. Replacement polarity, not peri.
• GEO=0 does not prove the fork is absent in the same woman; it proves the paired object is not deposited.
• It is not inferred that a woman should receive progesterone, DHEA, NMN, a senolytic or a probiotic. Longevity is not inferred. Product n is not used.

19. Conclusions

Integrating L1–L10 yields a negative architecture, not a new mechanism. Zero of seventeen roadmap arrows close as an edge. Twenty-nine of twenty-nine autonomous governing questions are local. Synergy, fecal estrobolome and LATAM-dataset slogans die by internal null, program intention or guardrail. The HSD3B2 fork blocks a shared steroidogenic node and does not explain the SWAN residual: ZR territory predicts a fall and Crawford measures a stage rise.

What remains is a small falsifiable object. In the same women who already have daily urine and annual DHEA-S, is the +3.95% inflection the shadow of the bleeding classifier, an extra-ovarian anti-phase calendar, or a frame artifact (annual follicular serum vs cycle PdG)? Exactly one of those three must win. The laboratory does not need to recruit, does not need a FASTQ and does not need Lua data. It needs to residualize the classifier.

Until that is split, “the L1–L10 pathway” has no crossed biological object to partner. It has a grammar, a curriculum and a residual. That is a result, not a wait.

20. References

1. Santoro N, Crawford SL, Lasley WL, et al. Factors related to declining luteal function in women during the menopausal transition. J Clin Endocrinol Metab. 2008;93:1711–1721. PMID 18285413. PMC2386686. DOI 10.1210/jc.2007-2165. (H; ELA/PdG; body)
2. Santoro N, Crawford SL, El Khoudary SR, et al. Menstrual cycle hormone changes in women traversing menopause: SWAN DHS. J Clin Endocrinol Metab. 2017;102:2218–2229. PMID 28368525. PMC5505186. DOI 10.1210/jc.2016-4017. (H; n=511; annual DHEA-S in methods)
3. Santoro N, El Khoudary SR, Nasr A, et al. Daily luteal serum and urinary hormone profiles in the menopausal transition: SWAN. Menopause. 2020;27:127–133. PMID 31794501. PMC7050767. DOI 10.1097/GME.0000000000001453. (H; reconfirmed 2026-10-06)
4. Crawford S, Santoro N, Laughlin GA, Sowers MF, et al. Circulating dehydroepiandrosterone sulfate concentrations during the menopausal transition. J Clin Endocrinol Metab. 2009;94:2945–2951. PMID 19470626. PMC2730879. DOI 10.1210/jc.2009-0386. (H; reconfirmed 2026-10-06)
5. Randolph JF Jr, Zheng H, Sowers MR, et al. Change in follicle-stimulating hormone and estradiol across the menopausal transition: SWAN. J Clin Endocrinol Metab. 2011;96:746–754. PMID 21159842. DOI 10.1210/jc.2010-1746. (H)
6. Lasley BL, Crawford S, McConnell DS. Adrenal androgens and the menopausal transition. Obstet Gynecol Clin North Am. 2011;38:467–475. PMID 21961714. PMC3185242. (R/H)
7. Lasley BL, Crawford SL, Laughlin GA, et al. Menopause. 2011. PMID 21178790. PMC3123411. (H; BSO 20/81)
8. Lasley BL, Chen J, Stanczyk FZ, et al. Androstenediol complements estrogenic bioactivity during the menopausal transition. Menopause. 2012;19:650–657. PMID 22415563. (H; Adiol; no dose)
9. Dharia S, Slane A, Jian M, Conner M, Conley AJ, Brissie RM, Parker CR Jr. J Clin Endocrinol Metab. 2005;90:4357–4361. PMID 15870126. DOI 10.1210/jc.2005-0017. (H tissue)
10. Nanba AT, Rege J, Ren J, Auchus RJ, Rainey WE, Turcu AF. J Clin Endocrinol Metab. 2019;104:2615–2622. PMID 30753518. DOI 10.1210/jc.2018-02527. (H)
11. Tezuka Y, Atsumi N, et al. J Clin Endocrinol Metab. 2021;106:1389–1397. PMID 33524149. PMC8502483. (H tissue; n=60)
12. Shin JW, Park E, Choe S, Jeon YJ. Circulating senescence-associated secretory phenotype proteins and menopause. Menopause. 2023;30:1053–1057. PMID 37610717. DOI 10.1097/GME.0000000000002238. (H)
13. El Khoudary SR, et al. J Clin Endocrinol Metab. 2025. PMID 40123296. DOI 10.1210/clinem/dgaf175. (H)
14. Honda S, Tominaga Y, Espadaler-Mazo J, et al. J Med Food. 2024;27:720–727. PMID 38742994. DOI 10.1089/jmf.2023.k.0320. (H; body; GUS null)
15. Albert CM, Cook NR, Gaziano JM, et al. Effect of folic acid and B vitamins on risk of cardiovascular events. JAMA. 2008;299:2027–2036. PMID 18460663. PMC2684623. DOI 10.1001/jama.299.17.2027. (H extra-ovarian)
16. Song Y, Cook NR, Albert CM, Van Denburgh M, Manson JE. Diabetes. 2009. PMID 19491213. (H; WAFACS T2D)
17. van Dijk MM, Vissenberg R, Goddijn M, et al. T4LIFE. Lancet Diabetes Endocrinol. 2022;10:322–329. PMID 35298917. (H)
18. Farr JN, et al. Nat Med. 2024. PMID 38956196. DOI 10.1038/s41591-024-03096-2. (H extra-ovarian; not reopened as D0)
19. Ruth KS, et al. Nature. 2021. PMID 34349265. DOI 10.1038/s41586-021-03779-7. (H; G8)
20. Nair KS, Rizza RA, O'Brien P, et al. DHEA in elderly women and DHEA or testosterone in elderly men. N Engl J Med. 2006;355:1647–1659. PMID 17050889. NCT00254371. (H extra-ovarian interventional; reconfirmed 2026-10-06)
21. Guthrie JR, Dennerstein L, Taffe JR, et al. Osteoporos Int. 2004. PMID 15042284. (H; n=159)
22. Harlow SD, Gass M, Hall JE, et al. STRAW+10. J Clin Endocrinol Metab. 2012. PMID 22344196. DOI 10.1210/jc.2011-3362. (R/consensus)
23. Coppé JP, Rodier F, Patil CK, et al. J Biol Chem. 2011. PMID 21880712. DOI 10.1074/jbc.M111.257071. (V)
24. Sugimoto M, Martin N, et al. Mol Endocrinol. 2003. PMID 12609976. (A)
25. Meinsohn MC, et al. Biol Reprod. 2025. PMID 41074767. (A/V)
26. Geng L, Yang X, Sun J, et al. Adv Sci. 2025;12:e2412558. PMID 40192074. PMC12120758. (H PCOS + A; off-contrast)
27. O'Connor KA, Ferrell R, Brindle E, et al. Menopause. 2009;16:1178–1187. PMID 19568209. (H)
28. Grub J, Süss H, Willi J, Ehlert U. Front Glob Womens Health. 2021;2:774308. PMID 34970652. (H)
29. Allshouse AA, Santoro N, Green R, et al. Menopause. 2015;22:267–274. PMID 25714237. PMC4341112. (H; DHS not simultaneous)
30. Wildman RP, Tepper PG, Crawford S, et al. J Clin Endocrinol Metab. 2012. PMID 22723312. (H; adiposity)
31. Waugh EJ, Lamontagne E, et al. Am J Clin Nutr. 2007;86:1791–1801. PMID 18065600. (H young; degraded as peri)
32. Iwahashi N, Umakoshi H, Fujita M, et al. Mol Metab. 2024;84:101954. PMID 38718896. PMC11101872. (C; sex×age confounded; not peri n)
33. Pignatti E, et al. Endocrinology. 2022;163:bqac076. DOI 10.1210/endocr/bqac076. (V; adrenarche)
34. Greenland S, Pearl J, Robins JM. Causal diagrams for epidemiologic research. Epidemiology. 1999. PMID 9888278. (method)
35. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics. 1989;45:255–268. PMID 2720055. (method)
36. Faubion SS, et al. Physiol Rep. 2020. PMID 32857481. (precedent do(ligand) at +1, not S1 n)
37. Human Protein Atlas v24. ENSG00000203859 HSD3B2, adrenal gland. https://v24.proteinatlas.org/ENSG00000203859-HSD3B2/tissue/adrenal+gland (C/IHC; zonal identity, not calendar)
38. SWAN Data Access. https://www.swanstudy.org/swan-research/data-access/process/ (accessed 2026-10-06)

Claims of this stage

• S1-REPORT-EN-C1 — The English twin freezes the bilingual factual base. Operating hypotheses = S1-AR-H1/H2/HT1 v1. HG-H0 superseded. This turn reconfirms PMID 31794501, 19470626 and 17050889.
• S1-REPORT-EN-C2 — (2c) remains frame asynchrony, not PdG ≠ P4. r=0.837 (n=170) vs r=0.26 (n=122).
• S1-REPORT-EN-C3 — The HSD3B2 fork kills the cascade and does not explain the +3.95% stage rise. Territory predicts a fall.
• S1-REPORT-EN-C4 — Discriminant = ICPSR → FMP-time → DHS∩DHEA-S. No recruitment. No Lua. No cpr::spdg.
• S1-REPORT-EN-C5 — Pharma: negative opportunity if (2a); conditional and H0 if (2b). Nair closes restoring DHEA-S as function. Not H5.
• S1-REPORT-EN-C6 — Compute skip. G8 closed. HT1 parked. HUMAN_QA_REQUIRED.

Hypotheses touched: S1-AR-H1 v1; S1-AR-H2 v1 (2a)/(2b)/(2c); S1-AR-HT1 v1; S1-HG-H0 v1 (superseded); S1-HG-H1/H2/HT1 v1 (not reopened as primaries); S1-SCOPE-H1/H2/H3/HT1 v1 (lineage, not operating).

Nothing is written into memoria agentes/Lua Labs/reportes/. Laboratory memory is not consolidated at this stage. The artifact is ready for BILINGUAL_QA without converting the calendar residual into a prescription or a ready target.