Lua Labs Report — Thyroid autoimmunity (anti-TPO) and premature ovarian insufficiency/failure (POI/POF): association and mechanism

Date: 2026-06-01 Researcher: Lua Labs Classification: Neuroendocrine + Immuno-endocrine Line: L3 — Thyroid axis and reproductive function Subtopic: 3.3 — Thyroid autoimmunity (anti-TPO) and premature ovarian insufficiency/failure (POI/POF): association and mechanism

External sources

Hsiao S-M et al. (2021). "Thyroid autoimmunity is associated with higher risk of premature ovarian insufficiency — a nationwide Health Insurance Research Database study." Human Reproduction 36(6):1621–1629. DOI: 10.1093/humrep/deab025. PMID 33569594.
Jin Y et al. (2024). "Prevalence of thyroid autoantibody positivity in women with infertility: a systematic review and meta-analysis." BMC Women's Health 24:622. DOI: 10.1186/s12905-024-03473-6. PMC11600930. PMID 39604908.
Arlıer S, Kükrer S (2025). "Autoimmune Thyroid Disease and Female Fertility: Does Anti-TPO Accelerate Ovarian Aging?" Journal of Clinical Medicine 14(22):8024. DOI: 10.3390/jcm14228024. PMC12653286. Published 2025-11-12.
Kelkar RL et al. (2005). "Circulating auto-antibodies against the zona pellucida and thyroid microsomal antigen in women with premature ovarian failure." Journal of Reproductive Immunology 66(1):53–67. PMID 15949562.
Sharif K, Watad A, Bridgewood C, Kanduc D, Amital H, Shoenfeld Y (2019). "Insights into the autoimmune aspect of premature ovarian insufficiency." Best Practice & Research Clinical Endocrinology & Metabolism 33(6):101323. DOI: 10.1016/j.beem.2019.101323.
Bagheri A et al. (2023). "The association of Treg and Th17 cells development factors and anti-TPO autoantibodies in patients with recurrent pregnancy loss." BMC Immunology 24:40. PMC10619307.
Komorowska B (2016/Update Naz et al. 2010). "Multiplicity of molecular and cellular targets in human ovarian autoimmunity: an update." Journal of Reproductive Immunology. PMC2965339.
Korevaar TIM et al. (2022). "Effect of hypothyroidism and thyroid autoimmunity on the ovarian reserve: A systematic review and meta-analysis." Reproductive Medicine and Biology 21(1):e12427. DOI: 10.1002/rmb2.12427. PMC8656199.
Magri F et al. (2019). "Overtime trend of thyroid hormones and thyroid autoimmunity and ovarian reserve: a longitudinal population study with 12-year follow-up." BMC Endocrine Disorders 19:47. DOI: 10.1186/s12902-019-0370-7. PMC6505305. PMID 31064360.
Szeliga A et al. (2021). "Autoimmune Diseases in Patients with Premature Ovarian Insufficiency — Our Current State of Knowledge." International Journal of Molecular Sciences 22(5):2594. PMC7961833.
Sullivan SD et al. (2024/2025). "Autoimmune Disease is Increased in Women With Primary Ovarian Insufficiency." Journal of Clinical Endocrinology & Metabolism 110(8):e2614–e2620. DOI: 10.1210/clinem/dgae828. PMC12261096.
Ocelo-Mora C et al. (2014). "Thyroid function and thyroid autoimmunity in apparently healthy pregnant and non-pregnant Mexican women." (Mérida, Yucatán cohort). PMID 24781677.
Gaberšček S, Zaletel K, Schwetz V, Pieber T, Obermayer-Pietsch B, Lerchbaum E (2015). "Mechanisms in endocrinology: thyroid and polycystic ovary syndrome." European Journal of Endocrinology 172(1):R9–R21 (mechanistic base reference for immune-ovarian crosstalk).

Base knowledge (what I know before searching)

Thyroperoxidase (TPO) is an apical heme-enzyme in the follicular thyrocytes of the thyroid gland; it catalyzes iodine organification (iodination of tyrosyl residues in thyroglobulin) and the coupling of iodotyrosines to produce T4/T3. Anti-TPO antibodies (TPOAb) are IgG autoantibodies directed against conformational epitopes of TPO's extracellular domain. They are the most sensitive serological marker of autoimmune thyroiditis (Hashimoto's, present in 90-95% of cases; also in 70% of Graves'). In the general population of reproductive-age women, their prevalence is ~10-14% (NHANES). Their presence defines TAI (thyroid autoimmunity) regardless of functional status (euthyroid, SCH, overt hypothyroidism).

Mechanistically, anti-TPO is not only a marker — it has documented pathogenic function: it fixes complement (classical pathway), opsonizes thyrocytes, and participates in antibody-dependent cellular cytotoxicity (ADCC) via FcγRIII on NK cells. This contradicts the old dogma that "anti-TPO is only a marker, not pathogenic." Autoimmune "spillover" occurs when: (a) there is cross-reactivity through molecular mimicry against shared antigens in other tissues, (b) there is systemic complement activation with consumption and bystander damage, (c) there is Th17 polarization / Treg deficiency that extends broken tolerance to other endocrine organs. Autoimmune Polyglandular Syndromes (APS-1, APS-2, APS-3, APS-4) are the clinical manifestation of this spillover.

POI/POF (premature ovarian insufficiency/failure) is defined as amenorrhea ≥4 months with FSH ≥25 IU/L (ESHRE 2016 criteria) or ≥40 IU/L (classic criteria) before age 40. DOR (Diminished Ovarian Reserve) is the preceding functional continuum — low AMH for age, reduced AFC, suboptimal response to gonadotropic stimulation, without necessarily meeting FSH/amenorrhea criteria. Mechanistically, autoimmune POI occurs through: lymphocytic oophoritis (mononuclear infiltrate of the theca of developing follicles + corpus luteum, sparing primordial and primary follicles in early stages — this is critical because it defines the therapeutic window before reserve depletion), anti-ovarian antibodies (anti-StCA against steroidogenic cells, anti-NALP5, anti-ZP3), and paracrine damage by Th1/Th17 cytokines.

The expected mechanistic connection between TPO and ovary includes at least four plausible a priori pathways: (1) ZP3↔TPO molecular mimicry (partial homology in T-cell epitopes), (2) systemic complement consumption → bystander damage in granulosa cells expressing CD55/CD59 at intermediate levels, (3) Th17 polarization/Treg deficiency → pro-inflammatory follicular environment, (4) functional subclinical hypothyroidism (inherited L3.2) that reduces ovarian tissue T3 → affects StAR/CYP19A1/AMH-Sertoli-like signaling in granulosa cells. These four mechanisms are NOT mutually exclusive — they may act in series.

Findings from recent papers

Robust epidemiological association — established magnitude and direction:

Hsiao 2021 (Taiwan NHIRD, n=21,325, population cohort with 12-year follow-up): women with Hashimoto's disease have a 2.40× higher risk of infertility due to ovarian failure (adjusted HR) and 89% higher risk of amenorrhea vs. 1:4 matched controls. Graves' disease: 68% higher risk of amenorrhea. This is the strongest published epidemiological evidence for TAI → POI causality. It does not measure AMH directly — it uses validated diagnostic codes.
Magri 2019 (12-year population cohort): women in the lowest quartile of ovarian reserve have higher baseline TPOAb levels + an upward longitudinal trend in TPOAb (vs. flat high quartiles). This suggests that TAI is not only associated with cross-sectional DOR but also accelerates the decline. Effect size is modest but directionally consistent.
Korevaar/Hasegawa 2022 SCH+TAI meta-analysis: AMH was lower in TAI+ vs TAI− with mean difference −0.12 ng/mL (95% CI −0.18 to −0.06). Small effect on population average but significant in younger strata (<35).
Arlıer & Kükrer 2025 (1,460 euthyroid infertile women, 2022-2025): anti-TPO+ prevalence 17.6%. Anti-TPO+ vs anti-TPO−: AMH 1.47 ± 1.52 vs 3.33 ± 3.03 ng/mL (clinically substantial difference, ~2.3× lower); reduced AFC; elevated FSH. Stronger association in <35 years, non-obese, PCOS-negative — rules out metabolic confounder. Effect size larger than previous meta-analyses.
Jin 2024 BMC Women's Health meta-analysis (search through Feb 2024): pooled worldwide prevalence of TAI in infertile women = 20%, RR vs healthy women = 1.51. Stratified by POI: in overt POI with TSH >2.5 µIU/mL, TPOAb and TgAb are associated (all p<0.001); with TSH ≤2.5 there is no association. This implies that the TAI-POI association is modified by thyroid functional status — it is not "anti-TPO only."
Sullivan 2024 (JCEM): women with POI have significantly higher incidence of any autoimmune disease vs controls, with thyroid and celiac disease dominating. Supports the logic that "POI is a window into the diagnosis of underlying APS."

Molecular mechanism — partial but cumulative evidence:

Kelkar 2005 (POF n=82): sera from POF show anti-ZP3 (35-45% prevalence) + thyroid anti-microsomal (anti-TPO) + strong immunoreactivity on granulosa cells. POF antibodies cross-react ONLY with thyroid and not with uterus, spleen, kidney, liver, adrenal, pancreas, or pituitary. Direct evidence of TPO↔ZP3 cross-reactivity. This is the paper that mechanistically anchors molecular mimicry.
Naz et al. (review PMC2965339) + Sharif/Shoenfeld 2019: at least 6 antigenic targets identified in autoimmune oophoritis: ZP3, granulosa cells, theca cells, steroidogenic cells (StCA — includes p450scc, p450c17, p450c21, 3β-HSD), FSH receptor, oocyte. ZP3 shows T-cell epitopes with substantial homology to multiple non-ovarian antigens (~44% experimental cross-reaction rate).
Bagheri 2023 (Treg/Th17 in RPL): women with RPL + anti-TPO+ have TGFβ↓, Foxp3↓, RORγT↑, IL-17↑ vs RPL anti-TPO−. The molecular signature of Treg/Th17 imbalance is activated specifically by thyroid autoimmunity. This is the cleanest bridge between systemic TAI and endometrial-ovarian damage mediated by cellular immunity.
Histological lymphocytic oophoritis: predominantly CD4+ T-cell infiltrate (with CD8+, NK, B-cells, plasma cells IgG+) in the theca of developing follicles + corpus luteum, sparing primordial/primary follicles in early stages. The diagnostic gold standard is ovarian biopsy (not performed due to risk/cost). Preservation of the primordial reserve in early phases is the conceptual therapeutic window of opportunity.
Complement C3a/C5a in systemic autoimmunity: robust evidence in other autoimmune systems (lupus, RA) that C3a/C5a accelerate bystander tissue damage via anaphylatoxin receptors C3aR/C5aR1. Specific ovarian literature is scarce, but granulosa cells express C3aR/C5aR1 (human transcriptomics). Reasonable hypothesis but not directly verified in the human TAI+ ovary.

LATAM data:

Ocelo-Mora 2014 (Mérida, Yucatán): anti-TPO+ and/or anti-Tg+ in 14.4% non-pregnant, 13.5% pregnant (n=200+). Consistent with global prevalence, NOT higher in LATAM.
Carrillo-Lozano 2021 (inherited L3.1, n=1,496 infertile Mexican women): SCH 40.7% with TSH>2.5, 14.7% with TSH>4.1. Crossing with anti-TPO not reported by stratum — gap.
There is no published LATAM cohort stratifying POI/DOR by anti-TPO with quantified AMH. Real literature gap.

Complete molecular/endocrine mechanism

HLA-DR3/DR5/DR4 predisposition + ACEs (inherited L2)
         ↓
Broken thymic central tolerance → autoreactive T clones escape
         ↓
Peripheral activation (inherited L1 LPS, viral mimicry, high cortisol stress L2)
         ↓
Th1/Th17 polarization + deficient Treg (Foxp3↓, RORγT↑, IL-17↑) [Bagheri 2023]
         ↓
B-cells produce anti-TPO IgG (conformational epitopes extracellular domain)
         ↓
┌────────────────────┬─────────────────────┬─────────────────────┐
│ Pathway 1 — Direct │ Pathway 2 — TPO↔ZP3 │ Pathway 3 — Systemic│
│ TPO thyrocyte      │ molecular mimicry   │ Th17 + complement   │
│ damage             │                     │                     │
│                    │                     │                     │
│ ADCC + complement  │ Anti-ZP3 cross-     │ Systemic C3a/C5a    │
│ on follicular cell │ reactive antibodies │ + IL-17 + TNFα      │
│                    │ bind ovarian        │                     │
│ ↓                  │ granulosa+oocyte    │ ↓                   │
│ Hashimoto's        │ [Kelkar 2005]       │ Granulosa C3aR/     │
│ (T4↓, tissue T3↓)  │                     │ C5aR1 activated     │
│                    │ ↓                   │                     │
│ Inherits L3.1+L3.2:│ Lymphocytic         │ ↓                   │
│ TSH↑ → ovarian TSHR│ oophoritis          │ Granulosa oxidative │
│ + DIO2 ubiq +      │ (CD4+CD8+ infiltrate│ stress +            │
│ tissue T3 ↓ +      │ theca + luteum,     │ mitochondrial       │
│ HOXA10/LIF↓        │ spares primordial)  │ dysfunction         │
│                    │                     │ (inherited L2.2     │
│                    │ ↓                   │ ER stress; future   │
│                    │ Granulosa apoptosis │ L9 NAD+)            │
│                    │ + accelerated atresia│                    │
└────────────────────┴─────────────────────┴─────────────────────┘
                                ↓
                Convergence in granulosa + corpus luteum:
                AMH↓ (compromised Sertoli-like AMH secretion) +
                CYP19A1↓ (aromatase) → follicular E2 ↓ +
                StAR/CYP11A1 ↓ (inherited L3.2) → luteal P4 ↓ +
                endometrial HOXA10/LIF ↓ → implantation window
                                ↓
              DOR (low AMH, reduced AFC) → POI/POF
                                ↓
                Recurrent biochemical pregnancy loss (subclinical LATAM)

Three parallel mechanisms, one clinical signature: the anti-TPO+ woman with POI/DOR does not have ONE dominant mechanism — she has all three partially active. The relative weight of each pathway depends on: current TSH (pathway 1 dominant if SCH coexists, inherited L3.2), ZP3 reactivity (pathway 2 dominant if POF with detectable anti-ZP3), and Th17/Treg signature (pathway 3 dominant if RPL coexists, Bagheri 2023). This explains why T4LIFE (L3.2, LT4 in euthyroid TPO+ women) failed: it covers pathway 1 but leaves pathway 2 and pathway 3 intact.

Cross-synthesis with previous findings

L3.1 → L3.3 (functional hypothyroid symptoms in euthyroid women): in euthyroid anti-TPO+ women (the ASRM 2024 gray zone, TSH 2.5-4.12), a sustained burden of functional hypothyroid symptoms —without overt biochemical abnormality— can coexist with low ovarian reserve through the three pathways simultaneously. The symptomatic signature of "functional ovarian damage associated with thyroid autoimmunity" is a composite pattern, not attributable to any single isolated symptom.

L3.2 → L3.3 (luteal-thyroid phenotype + luteal pentagon): "front 4" of the luteal pentagon (low endometrial T3 → HOXA10/LIF↓) in anti-TPO+ is amplified by the three pathways of autoimmune damage. Natural sub-hypothesis: in anti-TPO+ with biochemical SCH, biochemical pregnancy loss (O1 variable inherited L3.2) is a combined marker of luteal pentagon + subclinical lymphocytic oophoritis + endometrial Th17 — not clinically differentiable among the three causes but predictable through longitudinal pattern recognition.

L2.4 / L2.5 (diurnal cortisol phenotype + composite HPA-load phenotype): TAI has a Th17 component that is directly modulated by chronic cortisol (inherited L2.1-L2.4). Mechanistic prediction: diurnal cortisol phenotype=A (hyperreactive HPA) → higher cortisol production → Treg/Th17 imbalance → higher probability of progressive anti-TPO+ + faster AMH decline. However diurnal cortisol phenotype=B (collapsed allostatic load, chronic low-grade inflammation) → high baseline IL-17/TNF-α environment → greater susceptibility to autoimmune ovarian damage even with lower cortisol. Two different routes to the same POI destination via chronic stress — testable clinical bifurcation.

L2.3 (GR phase sensor + FKBP51 → functional P4 withdrawal): autoimmune POI with anti-TPO+ enriches the luteal failure pentagon (inherited L3.2) — it adds a sixth arm: loss of granulosa mass through chronic autoimmune damage, not compensable with T4 or cortisol management. Pentagon → Hexagon of luteal failure in women with active TAI.

Microbiome-vagal axis → immune modulation: the Th17 immune environment is bidirectionally modulated by the microbiome. Strains such as Lactobacillus brevis (GABA producers) reduce IL-17 and increase systemic Foxp3+ Tregs in the 2023-2024 microbiome literature. This raises a plausible non-pharmacological bridge —via diet and microbiome— over the Th17/Treg signature that sustains thyroid-ovarian autoimmunity.

TAI ↔ POI ↔ biochemical pregnancy loss LATAM triangle: L3.3 closes the conceptual triangle TSH (modulator) ↔ SCH (function) ↔ TPO (autoimmunity). The 3 consecutive sessions L3.1+L3.2+L3.3 now produce a unified theory of the suboptimal thyroid-reproductive axis: point-in-time TSH does not capture the pattern; functional SCH + active autoimmunity + immuno-inflammatory burden compose the "Thyroid-Reproductive Dysfunction Continuum" — the first Lua Labs model in this line.

Lua Labs hypotheses

Hypothesis 21 — "Continuum of functional ovarian aging accelerated by thyroid autoimmunity"

Statement: In women aged 28-42, the simultaneous coexistence of (i) a sustained burden of functional hypothyroid symptoms, (ii) a systemic-autoimmune symptomatic signature (migratory arthralgias, persistent unexplained fatigue, diffuse alopecia, episodic urticaria, dry eye, family history of Hashimoto/POI/celiac), (iii) an obstetric history of biochemical pregnancy loss ≥1, and (iv) early menstrual markers of diminished ovarian reserve (shortened cycle ≤24 days, recurrent peri-menstrual spotting, luteal phase ≤10 days) defines a phenotype of "ovarian aging accelerated by thyroid autoimmunity" with anti-TPO+ probability ≥40% and a descending AMH trajectory ≥0.15 ng/mL/year — significantly greater than the expected population decline (~0.05-0.10 ng/mL/year). The statement is falsifiable: the predicted association between the composite symptomatic signature and serological (anti-TPO) and reserve (AMH/AFC) outcomes can be confirmed or refuted in a longitudinal cohort.

Proposed mechanism: the three TAI→ovary damage pathways (direct TPO, ZP3 mimicry, systemic Th17) produce a composite symptomatic signature combining (a) peripheral functional hypothyroid symptoms (thyroid-symptom phenotype) through DIO2 ubiquitination + reduced thyrocyte function, (b) nonspecific systemic autoimmune symptoms through Th17 activation + chronic complement, (c) early menstrual symptoms of DOR through granulosa-corpus luteum damage. Specificity emerges from the combined pattern, not from any individual item. This specificity is often invisible to conventional clinical practice because each symptom is assigned to its silo; it requires longitudinal pattern analysis before any clinical inference.

Confidence level:

High for the TAI ↔ DOR/POI association (robust epidemiology — Hsiao 2021, Magri 2019, Arlıer 2025, Jin 2024 meta).
Medium-High for the three parallel mechanisms (Kelkar 2005 + Bagheri 2023 + systemic complement + histological oophoritis).
Medium for the digital discriminability of the phenotype without blood — this is Lua Labs' original component and requires cohort validation.
Low for the absolute predictive AUC against serum anti-TPO (preliminary estimate 0.65-0.72).

How to validate:

With formal nested substudy: n=80, prospective 12 months, AMH + anti-TPO + anti-Tg + TSH + fT4 at baseline + 12m. Validate thyroid-autoimmune phenotype ↔ serum anti-TPO correlation (target r ≥ 0.30) and thyroid-autoimmune phenotype ↔ 12-month AMH slope (target r ≥ 0.25).

Limitations:

Weakest link: the specificity of the systemic-autoimmune symptomatic signature (fatigue, arthralgias, alopecia) — these are nonspecific symptoms. Specificity improves with longitudinal composition but is not eliminated.
The ZP3↔TPO cross-reactivity from Kelkar 2005 has small n and has not been replicated in large cohorts with modern methodology. Plausible mechanism but limited post-2015 literature.
T4LIFE 2022 (inherited L3.2) refutes the "TPO+ → LT4 → cure" logic in euthyroid women — it implies that the effective intervention is NOT only thyroid-focused but multi-front (microbiome + stress + nutrigenomics + thyroid). Lua Labs must embrace this complexity without simplistic claims.

Sub-hypothesis H21a — "diurnal cortisol phenotype bifurcation determines which TAI pathway dominates"

Statement: In women with thyroid-autoimmune phenotype ≥ 60, the diurnal cortisol phenotype=A phenotype (hyperreactive HPA) has a higher proportion of pathway 1 + pathway 3 (chronic cortisol → DIO2 ubiquitination + cortisol-mediated Th17 polarization) → clinical phenotype "TSH 2.5-4.12 + low AMH + hypo symptoms + short cycle." diurnal cortisol phenotype=B (collapsed allostatic load) has a higher proportion of pathway 2 + pathway 3 from low-grade inflammation → phenotype "normal TSH + very low AMH + heavy systemic autoimmune symptoms + recurrent biochemical pregnancy loss." Same background autoimmunity, two distinct clinical routes through differences in the HPA axis.

Confidence level: Low-Medium (speculative but mechanistically coherent). Testable: diurnal cortisol phenotype × thyroid-autoimmune phenotype interaction term in biochemical pregnancy loss vs short cycle outcome should be significant p<0.10.

Sub-hypothesis H21b — "Progesterobolome + vagal neurobolome modulate TAI progression speed"

Statement: In the thyroid-autoimmune phenotype ≥ 60 cohort, the top tertile dietary-diversity phenotype + vagal-tone phenotype will have 40% lower incidence of new thyroid diagnosis at 12 months vs the bottom tertile, mediated by: (a) Treg-inducing bacteria (inherited L1.6, L. brevis + butyrate producers) → restoration of systemic Foxp3+ Tregs → Th17 reduction, (b) reduced systemic LPS → lower complement activation → less bystander granulosa damage, (c) GABA + enterochromaffin serotonin buffer → reduced chronic cortisol → lower DIO2 ubiquitination. Significant thyroid-autoimmune phenotype × dietary-diversity phenotype interaction term p<0.10 → confirms. Significant thyroid-autoimmune phenotype × vagal-tone phenotype interaction term p<0.10 → confirms vagal pathway. Both null → progesterobolome/neurobolome are NOT relevant modulators when TAI dominates (refutes sub-hypothesis).

Confidence level: Medium (robust microbiome-autoimmunity literature, specific TAI application less verified).

Candidate formulation

Name: "Thyroid-Ovarian Coherence Foundation" (extension of Luteal-Thyroid Coherence Foundation L3.2 with autoimmune vertex).

Compounds (100% food and behavioral — GRAS, zero medical claims):

Calibrated dietary selenium — 1-2 Brazil nuts 4-5x/week (~80-150 µg/d), tuna/sardines 2x/week, organic egg. Mechanism: glutathione peroxidase + iodothyronine deiodinase cofactor; reduces serum TPOAb ~20-40% in RCTs (Hu 2021 selenomethionine 200 µg/d 6-month meta-analysis). Food only — supplementation >200 µg/d carries type 2 diabetes risk.
Dietary vitamin D + morning sun exposure — oily fish, egg, UV-treated mushrooms, fortified milk; sun 7-9 AM 15-20 min without sunscreen. Mechanism: VDR in T-cells modulates Treg/Th17; VDR deficiency is a strong predictor of TAI progression.
Dietary omega-3 EPA/DHA — sardine, light tuna, chia, ground flaxseed (ALA precursor). Reduce systemic IL-17 and TNF-α; improve Treg/Th17 ratio.
Flavonoid polyphenols — green tea EGCG, curcumin (with fat + black pepper), pomegranate, raspberry, 70%+ cacao. Mechanism: systemic NF-κB inhibition + anti-TPO reduction in observational studies (modest effect, causality not proven).
Functional DIO2 triad inherited L3.2: zinc (pumpkin seeds, beans, shellfish), controlled iodine (~150 µg/d iodized salt NOM, do not exceed 300), iron (legumes + vit C).
Fermentable fiber ≥30 g/day (inherited L1.5+L1.6) — preserves Lactobacillus brevis + butyrate producers + Treg-inducing bacteria → IL-17 reduction. LATAM sources: black beans, lentils, nopal, oats, chia, green plantain, nixtamal resistant starch.
Gluten restriction in women with thyroid-autoimmune phenotype ≥ 60 + digestive symptoms — Hashimoto/celiac co-occur ~5-10% (APS-3 subtype); observational evidence of TPOAb improvement with gluten elimination (not causal, not to apply generically, only if symptoms).
Behavioral component (inherited L2.6): pre-sleep dietary glycine (bone broth, natural gelatin), 4-6 cpm breathing 5 min/day, regular sleep 22:30-06:30, evening alcohol restriction, management of care burden (LATAM-dominant variable inherited L2.5).

Target population:

Primary: Sofía (28, active cycle) with thyroid-autoimmune phenotype ≥ 60 — active therapeutic window, maximum preventive potential, before irreversible damage to the primordial reserve.
Secondary: Valentina (19, possible PCOS) — PCOS-Hashimoto co-occurrence ~25-30% (inherited L2.2 + future L3.4 bridge); early intervention modulates trajectory.
Tertiary: Carmen (47, perimenopause) with active TAI — ovarian decline is already present, focus on preservation of any residual reserve + symptomatic management.
NOT indicated for Rosa (55, postmenopause) — ovarian window closed; different focus (future L8 DHEA-adrenopause).

Complementary mechanisms:

Selenium + VDR → TPOAb reduction (pathway 1)
Omega-3 + polyphenols + fiber → Th17/IL-17 reduction (pathway 3)
Glycine + breathing + sleep → cortisol reduction → reduced DIO2 ubiquitination (indirect pathway 1)
L. brevis + butyrate → Treg induction (microbiome-mediated pathway 3)
Zinc + iodine + iron → thyroid substrate function (functional DIO2)

Requires validation: formal substudy n=80, 12 months, anti-TPO + AMH baseline + 12m, TOCF intervention arm vs standard nutrition control arm. Primary outcome: Δ TPOAb. Secondary outcome: Δ AMH + Δ thyroid-autoimmune phenotype. Sufficient power to detect Δ TPOAb 20%.

Individual variability

Genetics:

HLA-DR3, HLA-DR5 (DRB10301, DRB11101) — strong predisposition to Hashimoto's + APS; requires genotyping for assessment.
CTLA-4 polymorphisms (rs231775 G/G) — immune checkpoint; G/G higher risk of TAI + autoimmune POI.
PTPN22 (rs2476601) — T variant (T/T or C/T) reduces TCR threshold signaling → increased risk of TAI + lupus + RA.
FOXP3 polymorphisms — Treg function; X-linked, modulates severity of Th17 imbalance.
DIO2 Thr92Ala (inherited L3.1) Ala/Ala — interaction with TAI: reduced tissue T3 amplifies functional ovarian damage under autoimmune burden; more vulnerable combination.
VDR FokI (rs2228570) — TT variant reduces response to vitamin D → lower Treg/Th17 modulation.
STAT4 variants — modulate response to IL-12/IL-23 → Th17 polarization.

Epigenetics / environment:

ACEs ≥ 4 (inherited L2.1) — HPA + immune system programming toward lifetime Th17-dominance. Predisposition to active adult TAI + early POI.
Early microbiome (inherited L1.3) — C-section, short breastfeeding, antibiotics 0-12 years → programmed dysbiosis → lower Treg-induction.
Subclinical regional iodine deficiency in LATAM (variable, not homogeneous in Mexico) — cyclical iodine exposure programs TPO autoimmunity in susceptible individuals.
Exposure to competing halogens (fluoride, bromide, perchlorate — municipal water, industrial products) — displace iodine, chronic thyrocyte damage.
Smoking + dietary acculturation (inherited L1.4) — increases Graves' risk AND reduces Treg.
Sustained chronic stress (diurnal cortisol phenotype=A or B, inherited L2.4) — potent modulator of Th17 polarization.

Critical LATAM gap:

No HLA-DR3/DR5 characterization in Mexican women with POI (real gap).
No LATAM cohort combining anti-TPO + AMH + thyroid-symptom phenotype + autoimmune symptoms longitudinally.
Prevalence of APS-3 (Hashimoto + another autoimmune disease) in LATAM women not characterized — possibly underdiagnosed.