HMG: Dosing Protocols

Research Dosing Disclaimer#

The dosing information below is derived from research studies and is provided for educational purposes only. HMG is not approved for human use, and no official dosing guidelines exist.

Dose-Response Data#

Summary of dose–response findings. A 5‑day rat study comparing 25 versus 50 mg/kg intraperitoneal (i.p.) HMG demonstrated dose‑dependent lipid lowering: at 25 mg/kg, modest reductions were seen in serum triglycerides and phospholipids (~24% and ~20%, respectively), whereas 50 mg/kg produced larger effects including ~38% reduction in serum cholesterol, ~40% in triglycerides, and ~25% in phospholipids; hepatic and aortic cholesterol were also reduced (about 30% and 22%, respectively). No overt toxicity was reported in this regimen. These data indicate a clear dose–response for hypolipidemic effects over 25–50 mg/kg i.p. administered daily for 5 days in a vitamin D2/cholesterol–induced atherosclerosis rat model.

Acute brain exposure models in developing rats used either direct intrastriatal injections or high systemic doses normalized by body weight. A single intrastriatal injection of 4 μmol HMG per striatum (1 μL over 3 minutes) produced acute increases in reactive oxygen species (DCFH oxidation), lipid peroxidation (MDA), protein carbonyls, and decreases in reduced glutathione with alterations in antioxidant enzymes (decreased SOD and GR; increased GPx; HMG‑specific increased catalase; reduced G6PD). Pharmacologic modulation showed MK‑801 (NMDA receptor antagonist) and L‑NAME (NOS inhibitor) attenuated these effects; antioxidant pretreatments (melatonin 100 mg/kg, NAC 150 mg/kg, vitamin E 40 mg/kg + vitamin C 100 mg/kg) were protective. While this paradigm reports a single dose rather than a graded series, it establishes sensitivity of the developing brain to micromole‑range local exposures.

Systemic i.p. dosing in developing rats at 10 μmol/g body weight (equivalent to 10 mmol/kg) provided age‑dependent brain exposure: high striatal concentrations were achieved in 7‑day‑old rats (HMG ~348 μM) but were minimal in 30‑day‑old animals, consistent with greater permeability of the immature blood–brain barrier. Associated oxidative stress markers aligned with the attained brain levels in the younger animals. This high, body‑weight–normalized systemic dose thus reveals a strong age effect on exposure and outcomes, though formal dose grading was not performed in this model.

Key details are organized in the table below.

Interpretation. Across studies, HMG shows a dose‑responsive hypolipidemic effect in adult rats with repeated i.p. administration at 25–50 mg/kg over 5 days, with larger effects at 50 mg/kg. In developing rat brain models, single high local (4 μmol/striatum) or systemic (10 μmol/g i.p.) exposures provoke oxidative stress–related endpoints, with systemic exposure producing substantial brain levels predominantly in immature animals. Together, these data provide specific, body‑weight–normalized doses, routes, regimens, and outcomes relevant to dose–response for HMG in rodents.

Administration Routes#

We compared human menopausal gonadotropin (hMG; menotropins) pharmacokinetics across subcutaneous (SC), intramuscular (IM), oral, and topical/transdermal routes using primary PK studies and clinical data.

Key quantitative differences and clinical equivalence

• SC vs IM: In phase I crossover studies of highly purified urinary hMG/FSH, SC and IM administration produced equivalent FSH exposure. After single 225–445 IU doses, typical FSH Cmax was ~5–7.5 IU/L, AUCt ~410–486 IU·h/L, with slow absorption (Tmax ≈ 20 h) and a terminal half-life ~39–54 h; steady state occurred within ~5 days of daily dosing. Extent of absorption (AUC) met bioequivalence limits for SC vs IM, while Cmax showed more variability but remained similar. These data indicate comparable bioavailability and route-independent elimination kinetics for hMG/FSH injections.
• Clinical equivalence of SC and IM hMG: In a randomized multicenter IVF trial (Merional, hMG-IBSA), SC self-injection was clinically equivalent to IM for the primary outcome (oocytes retrieved) and pregnancy outcomes, with better local tolerability (injection-site pain reported only with IM).
• hCG as a comparator for gonadotropin PK by route: Prior IM vs SC comparisons of hCG have demonstrated bioequivalent extent of absorption and similar half-life (~32–39 h), with SC median Tmax around 16–24 h. Product source/formulation can shift Cmax and Tmax without meaningfully altering half-life, underscoring that parenteral route primarily affects absorption rate rather than elimination. Reviews also note SC dosing supports once-daily administration for gonadotropins, with hCG’s longer half-life driving greater accumulation than LH-like activity, relevant because urinary hMG contains hCG-derived LH activity.

Non-injectable routes

• Oral: Intact hMG/FSH/LH are large glycoprotein hormones with negligible oral bioavailability due to gastric/intestinal proteolysis and first-pass metabolism; no human data show systemic exposure of intact gonadotropins after oral dosing. Thus, oral administration is not clinically viable for hMG.
• Topical/transdermal: There are no approved or credible systemic topical/transdermal hMG products. The stratum corneum severely restricts permeation of macromolecules like FSH/LH/hCG, in contrast to small lipophilic steroids, so transdermal delivery is not feasible for hMG.

Route-specific pharmacokinetic features and practical implications

• SC: Equivalent exposure to IM with convenient self-administration. Typical single-dose PK for FSH derived from hp-hMG/hp-FSH: Cmax ~5–7.5 IU/L; Tmax ≈ 20 h; AUCt ~410–486 IU·h/L; t1/2 ~39–54 h; steady state ~5 days. Absorption can vary by injection site (e.g., vaginal-wall SC yields higher Cmax and AUC, earlier Tmax, and slower clearance than abdominal SC with rhFSH), reflecting depot and lymphatic uptake effects.
• IM: Bioequivalent to SC in extent with similar half-life and Tmax in the ~20 h range for FSH; more local pain and less practical for repeated dosing. Clinical outcomes in IVF are comparable to SC.
• Oral: Not feasible—no measurable systemic PK for intact hMG because of degradation and first-pass loss; experimental animal data do not demonstrate intact hormone exposure.
• Topical/transdermal: Not feasible for gonadotropins; no measurable systemic PK reported or clinical products; macromolecular size and hydrophilicity preclude skin permeation.

Embedded summary table

Conclusion For hMG, parenteral SC and IM routes provide equivalent systemic exposure and similar PK (slow absorption, ~1.5–2-day half-life), with SC preferred due to convenience and tolerability. Oral and topical routes are not clinically viable because of negligible bioavailability for large glycoprotein hormones.

Human-Equivalent Dosing#

We summarize how animal study doses labeled “HMG” are translated to human-equivalent doses (HED) and the allometric methods used, noting that HMG is used in three distinct modalities: (i) small-molecule HMG‑CoA reductase inhibitors (statins), (ii) human menopausal gonadotropin (hMG, a biologic), and (iii) 3‑hydroxy‑3‑methylglutaric acid (3‑HMG, an endogenous small acid administered in rodent models). The appropriate scaling method is modality- and mechanism-dependent.

Overview of commonly used methods

• BSA/Km-based HED (NOAEL→MRSD workflow). Animal NOAELs (mg/kg) are converted to human-equivalent mg/kg using body-surface-area normalization via species-specific Km factors; the human-equivalent dose (HED) is typically further divided by a safety factor to set a maximum recommended starting dose (MRSD). This approach is widely used for small molecules and is conservative versus direct body-weight scaling. It is not recommended as the sole approach for biologics with species-specific pharmacodynamics (PD). Equations and the regulatory context are described in reviews of FIH dose estimation; using body weight instead of BSA would markedly increase HEDs for mice, rats, and dogs, underscoring why BSA/Km is preferred as a conservative default for small molecules (e.g., statins).
• Allometric scaling by body weight (power law). Pharmacokinetic parameters scale with body weight using Y = a·W^b. Common fixed exponents are: clearance (CL) ~ W^0.75, volume of distribution (Vss) ~ W^1.0, and half‑life ~ W^0.25, often used when only one preclinical species is available. Better accuracy is obtained by multi-species log–log regression to estimate b, ideally using three or more species. Reported exponents vary widely; therefore fixed-exponent scaling should be applied with caution, and prediction error increases with fewer species.
• Rule of Exponents (ROE) and corrections. When simple allometry gives exponents signaling poor fit, corrections using maximum lifespan potential (MLP) or brain weight can improve clearance predictions. A practical ROE: exponents ~0.55–0.70 often favor simple allometry; 0.71–0.99 may favor CL×MLP; ~1.0 may favor CL×brain weight; values outside these ranges warn of poor predictivity. These corrections are most discussed for proteins/antibodies; their applicability is case-dependent.
• PK-guided exposure matching (AUC/CL-based HED) and MABEL. When animal NOAEL exposures (AUC) and predicted human CL and F are available, a human starting dose can be set to match or stay below the animal exposure: Dosehuman = AUCtarget × CLhuman / F. For biologics or agonists where PD drives risk, minimal anticipated biological effect level (MABEL) or receptor occupancy-based approaches are recommended, as NOAEL-based HEDs can overestimate safe doses (e.g., TGN1412). These approaches require human target potency/occupancy modeling and interspecies PD comparison.
• Mechanistic PBPK/IVIVE. Clearance and tissue exposure can be predicted mechanistically using in vitro intrinsic clearance and physiological scalars (e.g., MPPGL, HPGL) integrated into PBPK models. BW^0.75 is often a default when first-order systemic clearance is the key dose metric, but portal-of-entry effects, highly reactive metabolites, or acute saturable kinetics may invalidate such defaults. In vitro systems frequently underpredict in vivo CL; validation and sensitivity analysis are needed (kenyon2012interspeciesextrapolation. pages 16-18, kenyon2012interspeciesextrapolation. pages 11-14).
• Renal clearance allometry. Renal drug clearance often scales near W^0.75 across species, supporting renal-CL-based extrapolation when renal excretion dominates, although transporter differences can limit accuracy.

Modality-specific guidance for “HMG” contexts

• Statins (HMG‑CoA reductase inhibitors; small molecules). BSA/Km conversion of NOAELs is common and conservative for first-in-human considerations. Because statins can show extensive hepatic first-pass extraction and tissue selectivity, PK-guided scaling using predicted human CL and F is advisable, especially when presystemic extraction saturates nonlinearly at high doses. Cross-species data for fluvastatin and related statins document moderate-rapid absorption, high hepatic extraction, biliary excretion, and species differences in Vd and t1/2; these properties argue for PK-based scaling or PBPK rather than dose-only scaling where feasible.
• Human menopausal gonadotropin (hMG; biologic). For gonadotropins and other biologics with species-specific PD, MABEL or PK/PD model–based approaches are preferred over BSA-only conversion. Starting doses should target low receptor occupancy for agonists and incorporate safety factors. Fixed-exponent or BSA scaling can be misleading due to interspecies differences in receptor expression, binding, and downstream signaling.
• 3‑hydroxy‑3‑methylglutaric acid (3‑HMG; endogenous small acid). If administered systemically in rodents, HED translation should be based on the relevant dose metric: systemic exposure (AUC) or site-of-action concentrations. For small acids with significant renal elimination, renal-CL allometry or mechanistic renal PBPK can support human CL prediction and exposure matching. If portal-of-entry or local formation drives effect, scaling by delivered dose per unit surface area or mechanistic PBPK is preferable to BW^0.75. Renal-CL scaling near W^0.75 is a reasonable first approximation, with caveats for transporter differences.

Key equations and practical notes

• Power-law allometry: Y = a·W^b; fit log Y = log a + b·log W across species; common fixed exponents: CL 0.75, Vss 1.0, t1/2 0.25 (single-species heuristic). Multi-species fits reduce uncertainty.
• BSA/Km HED: HED (mg/kg) = Animal dose (mg/kg) × (Km_animal / Km_human); after HED, apply safety factor for MRSD. BSA scaling yields lower HEDs than BW scaling for small-animal species and is considered conservative for small molecules.
• Exposure-matching: Dosehuman = AUCtarget × CLhuman / F; or Dose = CLhuman × Css × τ / F for steady-state targets. Use species with the lowest NOAEL AUC as the index; apply safety factors. For biologics, use MABEL/RO thresholds rather than NOAEL exposure alone.
• ROE and corrections: If allometric b is ~0.55–0.70, simple allometry may suffice; ~0.71–0.99 suggests MLP correction; ~1.0 suggests brain-weight correction; exponents outside this range warn of poor predictivity. Applicability is strongest in protein therapeutics literature; validate case by case.
• PBPK/IVIVE parameters: Use MPPGL and HPGL scalars to translate in vitro CLint to hepatic CL; consider plasma protein binding and hepatic blood flow limitations. Compare PBPK projections against allometric estimates; IVIVE often underpredicts CL without empirical scaling factors (kenyon2012interspeciesextrapolation. pages 16-18).

Embedded summary of methods

Limitations and caveats

• Allometry assumes comparable ADME mechanisms; violations (e.g., transporter polymorphisms, target-mediated drug disposition, nonlinear first-pass extraction) reduce accuracy. Biologics require PD-aware methods (MABEL/RO); BSA-only conversions can be unsafe for agonists. PBPK/IVIVE accuracy depends on input quality; portal-of-entry and local site-of-action effects may require alternative dose metrics (delivered dose per area, tissue exposure).

Conclusion

• For animal “HMG” doses, select the scaling method by modality and mechanism. Use BSA/Km as a conservative default for small molecules, but prefer PK-guided or PBPK approaches when presystemic extraction or nonlinearities are expected (statins). For hMG, use MABEL/PK–PD with receptor occupancy constraints. For 3‑HMG, consider exposure-based scaling with renal-CL allometry or mechanistic renal PBPK if renal elimination dominates. In all cases, corroborate predictions across methods and apply safety factors appropriate to the uncertainty.

Evidence Gaps#

• No human dose-finding studies have been completed
• Allometric scaling from animal models has inherent limitations
• Route-specific bioavailability data in humans is absent
• Optimal treatment duration has not been established

Related Reading#

• HMG overview and research guide
• HMG side effects profile
• HMG research evidence