US EPA 1997 — Toxicological review of tributyltin oxide (IRIS)

This July 1997 toxicological review by Robert Benson (USEPA Region 8) is the supporting document for the Integrated Risk Information System (IRIS) summary on tributyltin oxide (TBTO, bis-(tri-n-butyltin)-oxide; CAS 56-35-9). It is the document that establishes EPA’s chronic oral Reference Dose (RfD) for TBTO at 3 × 10⁻⁴ mg/kg-day (0.3 µg/kg-day), grounded in the Vos et al. 1990 chronic immunotoxicity study in rats with a benchmark dose 10% (BMD10) of 0.03 mg/kg-day, an uncertainty factor of 100, and a modifying factor of 1. EPA assigned high overall confidence to the RfD. No inhalation Reference Concentration (RfC) and no oral cancer slope factor were derived because available data did not meet the minimum requirements. TBTO was assigned to cancer category D (1986 guidelines) / “cannot be determined” (1996 proposed guidelines). The review went through agency-wide consensus approval and external peer review (R. Baggs, M. Holsapple, A. Silverstone).

Scope note (metals classification). TBTO is an organotin compound, toxicologically distinct from inorganic tin (Sn²⁺, Sn⁴⁺) by approximately three orders of magnitude in derived reference values. The critical effect for TBTO is immunotoxicity (thymus-dependent immune suppression); the critical effect for inorganic tin is local gastrointestinal irritation. These two classes share an element (Sn) but no toxicological pathway. The frontmatter carries metals: [Sn] for element-level routing, but the substantive content routes to organotins and must not be conflated with inorganic-tin reference values. The 0.3 µg/kg-day RfD reported here is the basis for the ATSDR 2005 chronic oral MRL of 0.3 µg/kg-day for TBTO (see atsdr2005-tin-toxicological-profile).

Key numbers

Oral RfD derivation (Section 5.1)

• Oral RfD: 3 × 10⁻⁴ mg/kg-day (0.3 µg/kg-day, equivalent to 0.0003 mg/kg-day)
• Principal study: Vos et al. 1990 (chronic 18-month feeding study, weanling SPF-derived Riv:TOX Wistar rats, dietary TBTO 0/0.5/5/50 ppm = 0/0.025/0.25/2.5 mg/kg-day; males only)
• Critical effect: immunosuppression — depressed IgE titers and increased Trichinella spiralis larvae in muscle at 16.5 months exposure
• NOAEL: 0.025 mg/kg-day
• LOAEL: 0.25 mg/kg-day
• Benchmark dose (BMD10): 0.03 mg/kg-day (10% relative change benchmark response; lower 95% confidence bound on dose corresponding to BMR; polynomial and Weibull power mean response models, THC/THCW software, I.C.F. Kaiser 1990a,b)
• Uncertainty factor (UF): 100 (10× animal-to-human extrapolation; 10× sensitive human subpopulation protection)
• Modifying factor (MF): 1
• Overall confidence: high
• Database adequacy: no additional UF for data-base limitations applied

Inhalation RfC (Section 5.2)

• Inhalation RfC: not derived. Available inhalation data (Schweinfurth and Gunzel 1987 — 4-hour LC50 77 mg/m³ total particles, 65 mg/m³ < 10 µm; 29-32 day “nose-only” rat study at 0/0.03/0.16 vapor or 2.8 aerosol mg/m³, 4 h/d, 5 d/wk) did not meet the minimum 90-day inhalation bioassay requirement. No pharmacokinetic studies available for route-to-route extrapolation.

Cancer classification (Sections 4.6, 5.3)

• Classification: Category D (1986 cancer guidelines) / “cannot be determined” (1996 proposed cancer guidelines)
• Oral slope factor: not derived
• Basis: Wester et al. 1987-1990 two-year rat bioassay produced increased benign pituitary tumors, pheochromocytomas, and parathyroid adenomas at the high dose only; results inconclusive because of increased high-dose mortality, reduced statistical power from dose spacing (0/0.5/5/50 ppm), and high/variable background tumor rates for this rat strain. Daly 1992 18-month CD-1 mouse bioassay showed no tumor increase. Genotoxicity assays were overwhelmingly negative across multiple endpoints.

Chemical and physical information (Section 2.0)

• IUPAC name: bis-(tri-n-butyltin)-oxide
• Primary synonym: tri-n-butyltin oxide (TBTO)
• CAS number: 56-35-9
• Molecular formula: C₂₄H₅₄OSn₂
• Structural formula: (CH₃CH₂CH₂CH₂)₃Sn-O-Sn(CH₂CH₂CH₂CH₂)₃
• Molecular weight: 596.07 g
• Boiling point: 220-230 °C
• Melting point: < 45 °C
• Density: 1.17 g/cc (20 °C)
• Vapor pressure: 1 × 10⁻³ Pa (20 °C)
• Henry’s constant: 2 × 10⁻⁵ kPa·m³/mol (20 °C)
• Air conversion factor: 1 ppb = 26.6 µg/m³

Synthesis table (Section 4.5 — LOAELs and NOAELs across the database)

The bolded row is the principal study supporting the RfD.

Mechanism / mode of action (Section 3.0)

The mechanism of TBTO immunotoxicity is induction of apoptosis (programmed cell death) within the thymus. Raffray and Cohen 1991 showed thymocytes in culture undergo apoptotic changes at TBTO concentrations that did not affect cell viability; Raffray et al. 1993 showed these effects occur independently of protein-synthesis requirements and persist despite ATP depression below 20% of control values. Raffray and Cohen 1993 correlated reduced thymus weight in vivo with DNA fragmentation in thymocytes after a single oral dose. Dibutyltin (the major metabolite) is less effective than tributyltin at inducing apoptosis in vitro, indicating the in vivo toxicity is directly attributable to tributyltin. Alternative or parallel mechanisms include mitochondrial structural alteration and ATP synthesis depression (Hara et al. 1994; Yoshizuka et al. 1992a,b). No data are available on toxicokinetics relevant to risk assessment.

Childhood susceptibility (Section 4.7.1)

EPA noted some evidence that children may be more sensitive than adults: Smialowicz et al. 1989 showed pre-weanling rats more sensitive than adult rats in immune-function endpoints; the principal Vos et al. 1990 study itself dosed weanling rats for the remainder of their lives, so EPA considered any potential childhood sensitivity already accounted for in the RfD derivation.

Methods (review structure)

The review is structured per the 1983 National Research Council framework and the EPA 1987-1996 Risk Assessment Guidelines (Developmental, Reproductive, Neurotoxicity, Carcinogen, Inhalation Dosimetry, Benchmark Dose, Risk Characterization). The literature search drew on RTECS, HSDB, TSCATS, CCRIS, GENETOX, EMIC, EMICBACK, DART, ETICBACK, TOXLINE, CANCERLINE, MEDLINE and MEDLINE backfiles, plus public IRIS Submission Desk material. Internal peer review was performed by Cogliano, Henningsen, Smialowicz, and Vu (EPA); external peer review by Baggs (Rochester), Holsapple (Dow Chemical), and Silverstone (SUNY Syracuse). The benchmark-dose modeling used the THC (polynomial mean) and THCW (Weibull power mean) regression programs from K. S. Crump Division (I.C.F. Kaiser 1990a,b), with a 10% relative change BMR applied to IgE titer, T. spiralis larvae in muscle (digestion method), and T. spiralis larvae in muscle (histology method) from Vos et al. 1990. The lowest BMD10 (0.03 mg/kg-day) was selected for RfD derivation.

Provenance Notes

EPA IRIS toxicological review supporting the IRIS Summary record for TBTO (CAS 56-35-9). Document is a public-domain US government work; no DOI assigned. The canonical access URL is the IRIS PDF endpoint at iris.epa.gov/static/pdfs/0349tr.pdf. Local copy resides in the Kimi-agent June 3 manual-fetch folder. The numerical anchor (RfD = 3 × 10⁻⁴ mg/kg-day) has remained on the IRIS record since 1997 and is the reference value cited downstream by ATSDR 2005, EFSA 2004 (TDI of 0.25 µg/kg bw/day as TBTO, equivalent to 0.1 µg/kg bw/day as Sn for the group TBT+DBT+TPT+DOT), and Hong Kong CFS 2019.

Wiki pages this source may touch

• tin
• organotins

Implications

Certification: TBTO and the wider butyltin/organotin class fall outside the HMTc 10-analyte certification scope (Pb, tAs, Cd, MeHg, tHg, iAs, Ni, Al, Cr-VI, Sn). Standards reasoning under standards-methodology cannot fold this RfD into inorganic-tin or total-tin certification limits; the toxicological pathway is different by ~3 orders of magnitude. If a future HMTc scope expansion contemplates organotins (driven by seafood, PVC food-contact, or wood-preservative residue), this is the load-bearing US-EPA reference value that anchors the chronic oral side, with the EFSA 2004 group TDI as the European equivalent.

Courses: Illustrative case of how an immune-system endpoint can drive a reference dose below endpoints that look more clinically severe (chronic toxicity LOAEL 2.1 mg/kg-day in the same animal model vs. immunotoxicity LOAEL 0.25 mg/kg-day — a 10-fold gap). Also illustrates the difference between organotin and inorganic-tin reference values and why species speciation matters in tin literature.

App: Not directly relevant to the ingredient-level contamination estimator — this is a chronic toxicology reference, not a food-occurrence dataset. Consumers’ organotin exposure is best estimated via the seafood occurrence pathway covered in cfs2019-organotin-aquatic-hongkong and the PVC food-contact pathway covered in yamada1993-organotin-household-commodities.

Verification notes

• Cite-key follows the existing epa1997- agency-year prefix used by epa1997-arsenic-fish-shellfish.
• The PDF carries no DOI; access URL is the canonical IRIS endpoint.
• metals: [Sn] is the element-level routing tag; substantive routing target is organotins. Routing must not pool this RfD with inorganic-tin reference values — flagged in the Scope note above and in the ATSDR 2005 near-duplicate page.
• ingredients: [], products: [], matrices: [] are deliberately empty: this is an animal-toxicology / dose-response reference document, not a food-occurrence survey. The routing layer should route on metals only, not invent ingredient/product/matrix scopes.
• No regulation slug is referenced. The EPA IRIS TBTO RfD is conceptually parallel to the existing epa-iris-cadmium-rfd / epa-iris-lead-rfd regulation pages and may warrant a dedicated epa-iris-tbto-rfd regulation slug at a future Step 0 review; not created in this ingest per the no-new-regulations-without-Karen rule.
• Brand-firewall (Part 12): the review names a vendor reference material (Bio/dynamics, Inc. — Daly 1992 unpublished study sponsor for the TBTO Consortium) only as the corporate-author of a regulatory submission, not as a contamination-attribution; this is within the scientific-method-vendor exception. No consumer-brand attribution appears in this document.
• HMTc firewall (Part 2): no synthesis claims, threshold proposals, or consumer-audience translations included. The reported RfD is what EPA established, not what HMTc certifies to.
• 2026-06-03 fresh-context audit subagent (general-purpose) flagged label typo “Oral RfC: not derived” under the Section 5.2 sub-heading; independently verified against PDF p. 27 §5.2 “Inhalation Reference Concentration” — corrected to “Inhalation RfC: not derived.” (Oral RfC is not a defined quantity; the two endpoints are oral RfD and inhalation RfC.) Subagent also flagged that the “Critical effect” bullet lists two endpoints (depressed IgE titers + increased T. spiralis larvae in muscle) while BMD10 modeling in §5.1.2 used three endpoints (adds T. spiralis larvae by histology method); verified against PDF — the LOAEL itself on p. 22 is defined by the two endpoints named, while the third endpoint is named in the Methods section as one of the BMD-modeling inputs. Not corrected: the bullet describes the LOAEL definition, not the BMD-modeling input set, and is faithful to PDF p. 22 as written. Subagent also flagged the Provenance Notes / Implications cross-source assertions linking this RfD downstream to ATSDR 2005 MRL, EFSA 2004 TDI, and CFS 2019 — accepted as provenance-framing context, with a note to revisit during the Part 9 organotin synthesis pass.

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.