Janvier et al. 2018 — Elemental impurities in falsified research peptides

Summary

This is a comprehensive impurity-profiling study of falsified injectable research peptides, and for the Heavy Metal Index it is the highest-toxicological-significance occurrence finding in the June 8 research-chemical set. The authors acquired 27 unregulated polypeptide preparations sold as “research chemicals” through three illegal internet pharmacies and measured elemental impurities by ICP-MS against the ICH Q3D parenteral limits. Six of the preparations exceeded the class-1 arsenic limit of 1500 ppb, ranging from 1660 to 12,890 ppb — more than ten times the allowed concentration — and a follow-up HPLC-ICP-MS speciation confirmed that the arsenic was present entirely as inorganic arsenic, the most toxic and carcinogenic form. One preparation exceeded the lead limit, and single-sample exceedances were also found for copper, silver, and barium. Because these products are self-injected, an inorganic-arsenic load at ten times the parenteral limit is a serious, previously-undocumented exposure route. The finding establishes arsenic (specifically inorganic arsenic) as the priority elemental contaminant for the unregulated injectable-peptide category.

Key numbers

ICP-MS elemental impurities versus ICH Q3D parenteral limits. Concentrations in ppb (µg/kg); 1500 ppb = 1.5 mg/kg.

• The decisive finding: arsenic speciation (HPLC-ICP-MS) confirmed that all the elevated arsenic was inorganic arsenic (iAs), not the less-toxic organic forms.
• Maximum arsenic 12,890 ppb (≈12.9 mg/kg), roughly 8.6× the 1500 ppb ICH parenteral limit; the lowest exceeding sample (1660 ppb) was just over the limit.
• Silver and barium also showed exceedances; no platinum-group metal-catalyst residues were detected.
• Residual solvents were present in every sample (20–900 ppb) but below ICH Q3C limits; small-molecule contaminants (mannitol, sorbitol, PEG, residual reagents) were common.

Methods (brief)

Elemental impurity analysis used three platforms: an Agilent 8800 triple-quadrupole ICP-MS (octopole collision/reaction cell, He as collision gas, O₂ in the octopole for As and Se) for most elements; a Varian 820 ICP-MS with H₂ collision gas for lead (to remove interferences); and an Advanced Mercury Analyzer AMA 245 (PS Analytical) for mercury on 100 µL of digest. Arsenic speciation was carried out on a second batch by HPLC-ICP-MS (Varian, PRP-X-100 anion-exchange column, 40 mM ammonium carbonate eluent, pH 9.4), able to distinguish inorganic As(III)/As(V) from methylarsonate, dimethylarsinate and arsenobetaine. Sample prep: ≥2.0 mg of each peptide acid-digested (0.5 mL HNO₃ + 0.125 mL HCl + 0.25 mL H₂O₂, ramped to 90 °C and held 7 h), then diluted to 25 mL. Reference materials: NIST 1573a tomato leaves (trueness 88–98 % for Cd, Co, Cu) for the elemental panel and NMIJ 7532a brown rice flour (inorganic As 0.298 ± 0.008 mg/kg) for the speciation run. API identity by LC-IT-MSⁿ on a Dionex Ultimate 3000 / Bruker amaZon system with tryptic digestion; small-molecule contaminants by GC-MS (Agilent 7890A / 5975C) and LC-MS² (Waters Synapt G2Si); residual solvents by HS-GC-MS. All elemental results benchmarked against ICH Q3D parenteral limits and solvents against ICH Q3C. Vanadium trueness was 130 % (interference not fully removable), so V results are not reliable; for Hg and Ni the per-sample material was below LOQ in many vials, so the absence of exceedances for those two elements is a detection-limit gap, not a clean result.

Implications

• Establishes inorganic arsenic as the priority elemental contaminant for Research Chemicals, Peptides & SARMs (HMTc Category 26 master) (injectable research peptides), based on speciation-confirmed iAs at 1660–12,890 ppb in 6 of 27 preparations.
• The iAs speciation result is the load-bearing toxicological point: it elevates the concern well beyond what total-arsenic screening alone would imply, and it routes to Arsenic as confirmed inorganic-As occurrence in a self-injected product.
• Supports a certification/screening case for elemental-impurity testing of unregulated injectable peptides against ICH Q3D parenteral limits.

Wiki pages this source may touch

• Research Chemicals, Peptides & SARMs (HMTc Category 26 master) — confirmed iAs exceedances (>10× ICH) in injectable research peptides.
• Arsenic — inorganic-arsenic occurrence (1660–12,890 ppb) in a self-injected product; speciation-confirmed.
• Lead — single-sample ICH parenteral exceedance.

Verification notes

• Concentrations and exceedance counts from Section 3.3 (“Screening for elemental impurities”) and Table 2 (ICH Q3D element panel, ppb). The 1660–12,890 ppb arsenic range and “six samples exceed, >10× limit” are stated explicitly in the text; the speciation result (“all arsenic was present as inorganic arsenic”) is reported with the supporting HPLC-ICP-MS data noted as not shown.
• Speciation: this is one of the few sources with confirmed inorganic-As (iAs), not just total As — recorded in metals as both iAs and tAs, with iAs the operative finding.
• Vendors are anonymized (“vendor X/Y/Z”, “three illegal internet pharmacies”); no brand names are reproduced (Part 12).
• Cadmium was measured with no exceedance; mercury and nickel could not be assessed for exceedance due to insufficient sample material, so their absence here is a detection-limit gap, not a clean result.
• metals: was expanded on the 2026-06-08 merge-enhance pass to include Cu, Ag, and Ba — each of which the paper reports as having one or more samples above the ICH Q3D parenteral limit (Section 3.3). The prior list ([iAs, tAs, Pb, Cd]) omitted the three non-Class-1 exceedances. V was excluded because the authors flagged 130 % trueness from unresolved interference, so V results are not reliable.
• matrices: values (research-peptide, injectable-peptide, falsified-medicine) are outside the standard food-matrix vocabulary in docs/gpt-collaboration/system-prompt.md. They are kept because this paper sits outside the food/supplement scope the vocabulary covers — these are falsified injectable medicines acquired from illegal internet pharmacies — and the descriptors are the most accurate available. Flagging here for vocabulary-list review when peptide/injectable-medicine coverage is consolidated.
• 2026-06-08 merge-enhance pass also expanded the Methods section to specify the three analytical platforms used (Agilent 8800 triple-quad ICP-MS, Varian 820 ICP-MS with H₂ for Pb, AMA 245 mercury analyzer for Hg) and the second reference material (NMIJ 7532a brown rice flour for As speciation). The prior text described all elemental work as generic “ICP-MS” and named only NIST 1573a.
• 2026-06-08 autonomous-audit application: (a) removed a cross-source synthesis bullet under ## Implications that compared this paper’s findings to craven2025-aas-heavy-metals-australia — cross-paper synthesis belongs in Part 9 output, not on a source page; (b) replaced an unfilled “(Cu limit)” placeholder in the Key numbers table with “ICH Q3D Class 3 parenteral limit” (specific µg/kg value not verified in the body text and Table 2 only renders the threshold row at low fidelity in this PDF, so the table descriptor is kept qualitative).
• Product slug research-chemicals-peptides-sarms exists in wiki/products/ as of 2026-06-08 but is not in the 2026-05-18 taxonomy snapshot (snapshot is stale). The slug is real; flagging for snapshot refresh, not for slug correction.

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.