The Problem

Glucocorticoids remain the backbone of acute management in SLE, vasculitis, ILD, and inflammatory myopathies. Their well-known complications — osteoporosis, osteonecrosis, diabetes, dyslipidemia, adrenal suppression — are treated as universal risks. But they are NOT universal. The genetic architecture of glucocorticoid toxicity varies dramatically by ancestry, and Indigenous Mexican populations sit at a uniquely dangerous intersection.

The Pharmacogenomic Triple Threat

1. NR3C1 — The Glucocorticoid Receptor

The glucocorticoid receptor (GR) gene NR3C1 has well-characterized polymorphisms that alter sensitivity to corticosteroids:

• BclI (rs41423247): Increases GC sensitivity → more metabolic side effects at standard doses
• N363S (rs6195): Hypersensitivity variant → obesity, insulin resistance, reduced bone density
• ER22/23EK (rs6189/rs6190): Relative RESISTANCE → may need higher doses (fewer side effects but risk of undertreating)

In Mexican populations: ZERO published data on NR3C1 variant frequencies in Indigenous or Mestizo populations. We are dosing prednisone blind to whether our patients are hypersensitive or resistant.

2. SLC16A11 — The Diabetes Accelerator

SLC16A11 carries the strongest population-specific diabetes risk variant in the Americas:

• Frequency: ~33% in Yucatán Maya, ~25% in general Mexican mestizo, ~0% in Europeans
• Origin: Neanderthal introgression variant that reached fixation in pre-Columbian American populations
• Effect: Disrupts lipid metabolism in hepatocytes → insulin resistance → T2D (SIGMA Consortium, PMID 25053765)

The convergence: Glucocorticoids cause diabetes through insulin resistance. SLC16A11 ALREADY primes 33% of Maya patients for insulin resistance. The combination is multiplicative, not additive:

• European patient on prednisone 10mg: baseline DM2 risk + GC-induced IR
• Maya patient with SLC16A11 on prednisone 10mg: baseline DM2 risk + GENETIC IR + GC-induced IR

This may explain why DM2 prevalence in our 321-patient autoimmune cohort from IMSS Mérida was 14.3% — and why patients with DM2 were almost EXCLUSIVELY on rituximab (avoiding metabolic drugs like anti-TNF).

3. VDR — Vitamin D Receptor and Bone

Vitamin D receptor polymorphisms (BsmI, ApaI, TaqI, FokI) determine:

• Calcium absorption efficiency
• Bone mineral density response to vitamin D supplementation
• Susceptibility to glucocorticoid-induced osteoporosis (GIO)

In Mexican populations: Limited data, but Indigenous populations in southern Mexico show distinct VDR haplotype frequencies compared to Europeans (partially studied in osteoporosis contexts). The combination of:

• Low dairy intake (lactose intolerance ~80% in Indigenous Mexicans)
• Low vitamin D despite tropical latitude (melanin, indoor work)
• VDR variants that may reduce calcium absorption
• Glucocorticoid-induced bone loss

...creates a PERFECT STORM for fractures.

4. ABCB1/MDR1 — The Drug Efflux Pump

P-glycoprotein (ABCB1 C3435T) determines how much glucocorticoid actually reaches tissues:

• TT genotype: Reduced efflux → MORE intracellular GC → more toxicity at same dose
• CC genotype: Efficient efflux → less intracellular GC → may need higher doses

In Addison disease patients, ABCB1 polymorphisms were associated with reduced bone mass on GC replacement (Løvås et al., PMID 19282465). Frequency in Mexican Indigenous: NOT STUDIED.

5. HSD11B1 — 11β-Hydroxysteroid Dehydrogenase Type 1

Converts inactive cortisone → active cortisol in tissues (especially liver, fat, bone):

• Gain-of-function variants → MORE local cortisol → amplified metabolic/bone toxicity
• Frequency in Indigenous populations: UNKNOWN

The Combined Risk Model

For a Maya patient with SLE on chronic prednisone:

| Risk Factor | European | Maya-Mestizo | |---|---|---| | SLC16A11 (DM2 priming) | 0% carrier | 33% carrier | | NR3C1 sensitivity | Known frequencies | UNKNOWN | | VDR bone protection | Characterized | Poorly characterized | | ABCB1 efflux | Studied | NOT STUDIED | | HSD11B1 local activation | Some data | ZERO data | | Lactose intolerance | ~15% | ~80% | | Baseline DM2 prevalence | ~8% | ~14-18% |

The result: Maya patients may develop GC complications at LOWER doses and EARLIER than European patients — but we have no pharmacogenomic data to predict who is at highest risk.

Testable Predictions

1. Maya-mestizo patients on equivalent prednisone doses should develop DM2 at HIGHER rates than European patients (SLC16A11 effect)
2. Glucocorticoid-induced osteoporosis should be MORE severe in Indigenous patients with specific VDR haplotypes
3. NR3C1 BclI/N363S frequencies in Maya may explain inter-individual variation in steroid toxicity
4. ABCB1 TT genotype carriers should show more GC toxicity at standard doses
5. A pharmacogenomic GC-risk score incorporating these 5 genes could identify patients needing aggressive steroid-sparing strategies from DAY ONE

Clinical Implication: Steroid-Sparing as Pharmacogenomic Imperative

If confirmed, this hypothesis reframes steroid-sparing therapy in Maya-mestizo populations from a nice-to-have to a pharmacogenomic necessity:

• Faster escalation to biologics (rituximab, anifrolumab, belimumab) in SLE
• Aggressive calcium/vitamin D with VDR-guided dosing
• Lower prednisone thresholds for DM2 screening in SLC16A11 carriers
• DEXA monitoring starting EARLIER than current guidelines suggest

Proposed Genes for STORM Panel (Tier 4: GC Toxicity)

| Gene | Variant | Complication | Mexican Data | |---|---|---|---| | NR3C1 | BclI, N363S, ER22/23EK | GC sensitivity/resistance | ❌ ZERO | | SLC16A11 | Multiple | DM2 acceleration | ✅ ~33% Maya | | VDR | BsmI, ApaI, TaqI, FokI | Osteoporosis | Partial | | ABCB1 | C3435T | GC tissue exposure | ❌ ZERO | | HSD11B1 | Regulatory variants | Local cortisol amplification | ❌ ZERO | | CYP3A4 | Multiple | GC metabolism | Partial |

Population: 607 patients (321 autoimmune + 89 SAAF + 95 ILD + 102 SpA/PsA), IMSS HGR1 Mérida, Yucatán Key observation: DM2 14.3% in autoimmune cohort, 52% of DM2 patients on rituximab (steroid-sparing by necessity) References: PMID 25053765 (SIGMA SLC16A11), 19282465 (ABCB1 bone), 17376011 (steroid receptor pharmacogenomics), 12127038 (VDR osteoporosis), 26812836 (RIBEF Mexican pharmacogenomics)