Cadmium Toxicology and Kidney and Lung Cancer Risk

RoseyFields evaluates the toxicology of cadmium using the following evidence base, drawn from the peer-reviewed occupational, environmental, and experimental animal toxicology literature.

A note on sourcing: This summary cites the underlying peer-reviewed studies rather than the regulatory compilations (e.g., EPA IRIS, ATSDR) in which many of these studies were originally identified and reviewed. Study details (species, dose, findings) are drawn from those publicly available compilations of the primary literature and have not been independently re-verified against the original journal articles by RoseyFields.

1. Chemical Identity and Physicochemical Properties Relevant to Toxicity

Cadmium (Cd; CAS No. 7440-43-9) is a soft, silver-white transition metal (atomic number 48, atomic weight 112.41) that occurs naturally in the earth’s crust, typically in association with zinc, lead, and copper ores. It has no known biological function and is toxic to humans and animals at low body burdens relative to many other metals. In toxicological studies, cadmium is administered or encountered as one of several inorganic salts or oxides — cadmium chloride, cadmium oxide, cadmium sulfate, cadmium sulfide, and cadmium carbonate are the forms most frequently used experimentally — and the choice of compound materially affects absorption, tissue delivery, and observed potency (Friberg et al. 1974).

The toxicologically active species is the divalent cadmium cation (Cd2+); the anion primarily determines water solubility and, for inhaled particles, deposition and clearance behavior rather than the intrinsic cellular mechanism of toxicity. Cadmium chloride and cadmium sulfate are highly water-soluble and are readily bioavailable by both the oral and inhalation routes. Cadmium oxide and cadmium carbonate are comparatively insoluble in water but dissolve appreciably at gastric or lysosomal pH and behave toxicologically much like the soluble salts once inside the body. Cadmium sulfide and cadmium selenium sulfide (used historically as yellow and red pigments) are markedly less soluble in biological fluids and are correspondingly less acutely toxic by inhalation (Glaser et al. 1986; Rusch et al. 1986).

Cadmium is not degraded environmentally or metabolically; it is a persistent, bioaccumulative element with an extraordinarily long biological half-life in humans, estimated at 10–30 years, driven primarily by slow renal clearance (Friberg et al. 1974). This persistence — rather than any single acute physicochemical property — is central to its chronic toxicity: repeated low-level exposures accumulate in the kidney and liver over decades, so that cumulative body burden, not any single exposure event, determines the risk of the chronic renal, skeletal, and pulmonary effects described below.

2. Toxicokinetics

2.1 Absorption

Gastrointestinal absorption of ingested cadmium in humans and experimental animals is low, typically on the order of 3–10% of an oral dose, though absorption is increased substantially by iron, calcium, or protein deficiency and is generally higher in females than males, an observation attributed to lower iron stores (Flanagan et al. 1978). Cadmium bound to metallothionein in food (e.g., in liver, kidney meats, and shellfish) may be handled somewhat differently in the gut than the free ionic form, and a small number of specific foods (e.g., certain shellfish) show unusually efficient cadmium uptake (European Chemicals Bureau data as reflected in the compiled toxicokinetic literature).

RoseyFields — Cadmium Toxicology Summary

Pulmonary absorption is considerably more efficient than gastrointestinal absorption and depends heavily on particle size and solubility in respiratory fluids. Fume-sized particles of cadmium oxide, generated by heating or welding cadmium-containing metal, are deposited deep in the alveolar region and are substantially absorbed into the systemic circulation — occupational studies estimate pulmonary absorption in the range of 15–50% (Friberg 1950; Adamsson et al. 1979). Larger dust particles are subject to greater mucociliary clearance and partial swallowing, which adds a secondary gastrointestinal exposure component even in inhalation-dominant occupational settings.

2.2 Distribution

Absorbed cadmium enters the blood bound predominantly to albumin and other high-molecular-weight proteins and is transported to the liver, where it induces synthesis of the low-molecular-weight, cysteine-rich protein metallothionein. Hepatic cadmium becomes progressively sequestered as cadmium-metallothionein (Cd-MT) complex. This complex is released into the circulation (e.g., following hepatocyte turnover) in small amounts, is freely filtered at the renal glomerulus, and is reabsorbed by the proximal tubular epithelium via endocytosis (Foulkes 1978; Dorian et al. 1992a). As a result, the kidney — particularly the renal cortex — becomes the principal long-term repository of the body cadmium burden, typically accumulating roughly one-third of total body cadmium in chronically exposed humans, with the liver holding a comparable or somewhat smaller fraction and the remainder distributed among other soft tissues.

Placental transfer of cadmium is limited relative to many other toxic metals; the placenta itself accumulates cadmium and induces its own metallothionein, which is thought to act as a partial barrier that limits fetal exposure relative to maternal blood levels, though it is not a complete barrier (data compiled from occupational and environmental cohort studies, e.g., Huel et al. 1984).

2.3 Metabolism

Cadmium is not metabolized in the classical biotransformation sense (it is not oxidized, conjugated, or enzymatically degraded), but its principal ‘metabolic’ fate is sequestration by metallothionein induction, a cellular defense mechanism common to liver, kidney, and other tissues. Within renal proximal tubular cells, filtered Cd-MT complex is taken up by endocytosis and degraded within lysosomes, releasing free cadmium ion intracellularly; this free cadmium re induces local metallothionein synthesis, and so long as this induction keeps pace with the free cadmium released, the metal remains sequestered in a relatively non-toxic bound form (Foulkes 1978; Goyer et al. 1989).

2.4 Excretion

Unabsorbed ingested cadmium is excreted in the feces. Absorbed cadmium is excreted very slowly, predominantly in urine, at a rate on the order of 0.005–0.01% of body burden per day under normal conditions — the toxicokinetic basis for cadmium’s characterization as a cumulative toxicant. Urinary cadmium excretion increases once tubular reabsorptive capacity is exceeded or once frank tubular damage has occurred, at which point urinary cadmium reflects a combination of filtered load and leakage from damaged, previously cadmium-laden tubular cells rather than current exposure alone. This nonlinearity complicates the use of urinary cadmium as a simple biomarker of either exposure or cumulative body burden once renal injury has begun (data compiled from occupational biomonitoring literature, e.g., Roels et al. 1989, 1991, 1993; Järup et al. 1995b).

2.5 Species and Sex Differences Relevant to Interpreting Animal Data

Toxicokinetic handling of cadmium is broadly similar across the species used experimentally (principally rats, mice, and rabbits) and humans, with target organs (lung on inhalation exposure; kidney on both routes; testis, in some species, on parenteral or high oral exposure) largely conserved (relevance-to-public-health synthesis in the compiled ATSDR toxicological review). Notable differences that affect interpretation of animal bioassays include: (1) female rodents have shown greater susceptibility to lethality from inhaled cadmium oxide and cadmium chloride than males

RoseyFields — Cadmium Toxicology Summary

in several chronic inhalation studies (Oldiges et al. 1989); (2) mice have shown a greater pulmonary inflammatory response than rats at matched inhaled cadmium concentrations, though lung cadmium burden also differed between species at equal exposure concentrations, complicating direct extrapolation (Oberdörster et al. 1994); and (3) the intramuscular/subcutaneous route, historically used to demonstrate cadmium’s carcinogenic potential in rodents (e.g., testicular and injection-site tumors), is not physiologically comparable to human occupational or dietary exposure routes and its relevance to human risk is correspondingly limited.

3. Mechanism(s) of Action

Cadmium’s toxicity is not attributable to a single unifying mechanism but to several interacting pathways that converge most destructively in the renal proximal tubule, the primary target organ for chronic exposure by both the oral and inhalation routes.

3.1 Metallothionein Saturation and Free Cadmium Ion Toxicity

As summarized above, hepatic and renal metallothionein induction sequesters cadmium in a form with markedly reduced acute toxicity. Renal proximal tubular injury is understood to occur once the total cadmium content of the renal cortex exceeds a threshold in the range of roughly 50–300 µg/g wet weight, at which point the local capacity to synthesize metallothionein and bind incoming cadmium is exceeded and the pool of unbound, catalytically active free cadmium ion rises sufficiently to damage tubular cells directly (mechanistic synthesis compiled from Foulkes 1978; Dorian et al. 1992a, 1992b; Goyer et al. 1989).

3.2 Enzyme Inhibition and Interference with Essential Metal Homeostasis

Free cadmium ion can displace zinc and other essential divalent cations from metalloenzymes and zinc-finger structural motifs, inactivating or altering the function of a range of metal-dependent enzymes and DNA-binding/DNA repair proteins. Cadmium also activates calmodulin-dependent pathways, and its ionic mimicry of calcium and zinc is believed to underlie several of its downstream disruptive effects on cellular signaling and gene expression (mechanistic discussion compiled in the toxicokinetic/mechanistic literature review; see also functional consequences in Girolami et al. 1989, renal kallikrein excretion as a marker of distal nephrotoxicity).

3.3 Oxidative Stress

Although cadmium is not directly redox-active in the manner of iron or copper, it depletes cellular glutathione and inhibits antioxidant enzymes, indirectly promoting formation of reactive oxygen species and lipid peroxidation; this oxidative burden has been demonstrated in both renal and hepatic tissue and in the developing central nervous system (Gill et al. 1989a, 1989b).

3.4 Genotoxic and Epigenetic Mechanisms Relevant to Carcinogenesis

Cadmium is not a potent direct-acting mutagen in standard bacterial reverse-mutation assays, but it has produced positive responses in several mammalian test systems, including chromosomal aberrations and mutation induction in Chinese hamster and mouse lymphoma cells (data compiled in IRIS 2012 carcinogenicity review, citing e.g. Casto 1976; Ochi and Ohsawa 1983; Oberly et al. 1982) and DNA damage with concurrent metallothionein induction in human hepatoma cells (Fatur et al. 2002). Proposed mechanisms include inhibition of DNA mismatch repair and other repair pathways (rendering cadmium an indirect mutagen that potentiates the mutagenicity of other agents rather than a strong mutagen in its own right), induction of oxidative DNA damage, and interference with apoptotic clearance of damaged cells — the latter illustrated experimentally by the observation that cadmium-adapted alveolar epithelial cells show both reduced capacity to repair DNA damage and attenuated apoptosis, a combination proposed to increase susceptibility to tumor formation with repeated low-level pulmonary exposure (Hart et al. 2001).

RoseyFields — Cadmium Toxicology Summary

4. Animal Toxicological Evidence by Organ System

The following sections summarize representative animal toxicology findings by organ system and route of exposure. Study details (species, dose, duration, and endpoint) are drawn from the compiled experimental literature; exposure concentrations for inhalation studies are expressed as mg Cd/m3 and for oral studies as mg Cd/kg-day unless otherwise noted.

4.1 Respiratory System (Inhalation Route)

The respiratory tract is the primary target of acute and chronic inhalation exposure. Single acute exposures of rats to 5–10 mg Cd/m3 as cadmium oxide dust, cadmium oxide fume, or cadmium chloride for 1–5 hours produced moderate to severe multifocal interstitial pneumonitis, diffuse hemorrhagic alveolitis, macrophage inhibition, and necrosis of alveolar type 1 cells with subsequent type 2 cell hyperplasia (Bus et al. 1978; Grose et al. 1987). Comparable pneumonitis was produced in hamsters exposed to cadmium chloride at 10 mg Cd/m3 for 30 minutes (Henderson et al. 1979) and in rabbits exposed to cadmium oxide dust at 4.5 mg Cd/m3 for 2 hours (Grose et al. 1987).

At intermediate durations, a 13-week inhalation bioassay in rats and mice across five cadmium oxide concentrations (0.022–0.88 mg Cd/m3, 6.33 hours/day, 5 days/week) produced concentration-related alveolar histiocytic infiltration, alveolar epithelial hyperplasia, and tracheal epithelial hyperplasia or squamous metaplasia, with lesion severity increasing with dose; minimal lesions were also identified in the larynx and nasal cavity at the lower end of this range (National Toxicology Program bioassay data as compiled in the toxicological literature). In chronic-duration studies, Takenaka et al. (1983) exposed Wistar rats to cadmium chloride at 12.5, 25, and 50 µg/m3 for 18 months (23 hours/day, 7 days/week) followed by a 13-month observation period and observed adenomatous hyperplasia in the bronchoalveolar region as well as a significant, concentration-related increase in lung tumors — findings discussed further in Section 5 below. Oldiges et al. (studies compiled with Takenaka et al. 1983) found that chronic lethality across cadmium compounds followed the order cadmium chloride > cadmium sulfate ≈ cadmium oxide dust > cadmium sulfide, correlating more closely with long-term pulmonary retention of bioavailable cadmium than with water solubility alone.

A distinct adaptive phenomenon has also been demonstrated: repeated low-concentration cadmium acetate exposure in rats (1.6 mg Cd/m3, 3 hours/day) produced an initial inflammatory response that resolved by 3–6 weeks of continued exposure, and pre-exposed animals showed reduced pulmonary damage from a subsequent high concentration challenge relative to non-adapted animals (Hart 1986; Hart et al. 1989a). This tolerance appears to be mediated by induction of metallothionein, glutathione, and γ-glutamylcysteine synthetase, but is accompanied by reduced DNA-repair capacity and blunted apoptosis in the adapted epithelial cells — a combination proposed to increase, rather than decrease, long-term carcinogenic risk despite the reduced acute inflammatory injury (Hart et al. 2001).

4.2 Renal System (Inhalation and Oral Routes)

The kidney is the most sensitive target organ for chronic cadmium exposure by either route, and the animal evidence broadly parallels the extensive human occupational and environmental findings. Early inhalation studies demonstrated that rabbits exposed to cadmium metal dust at 4 mg Cd/m3 (3 hours/day, 21 days/month) developed proteinuria after approximately 4 months, with histologic renal lesions appearing after an additional 3–4 months of exposure (Friberg 1950). Subsequent inhalation studies at lower concentrations and comparatively shorter durations generally did not reproduce proteinuria, indicating that the exposure–duration products used in most later bioassays (chosen to demonstrate pulmonary endpoints) were insufficient to reach the cadmium body burden required for detectable renal injury.

Mechanistically, renal tubular injury is understood to follow accumulation of unbound cadmium in the proximal tubular epithelium once local metallothionein-binding capacity is exceeded (Section 3.1). This is consistent with the

RoseyFields — Cadmium Toxicology Summary

finding that cadmium-metallothionein complex administered directly is markedly more nephrotoxic, on a molar cadmium basis, than an equivalent dose of ionic cadmium chloride, because it delivers cadmium directly to the proximal tubule via receptor-mediated endocytosis rather than requiring prior hepatic sequestration and slow release (Dorian et al. 1992a, 1992b, 1995). Renal kallikrein excretion has also been used experimentally as a marker of distal nephron injury following cadmium exposure in rats, providing evidence that cadmium nephrotoxicity is not confined to the proximal tubule alone (Girolami et al. 1989).

4.3 Hepatic Effects

Hepatic effects of cadmium in animals are generally less pronounced than renal or pulmonary effects and typically require higher doses or acute/intermediate exposure durations to elicit. Cats examined shortly after inhalation exposure to cadmium oxide fume showed a spectrum of hepatic lesions ranging from cellular granulation at low doses to fatty infiltration at higher doses (Prodan 1932, as compiled in the toxicological literature). Rats exposed to cadmium at 0.1 mg Cd/m3 for 30 days showed increased serum alanine aminotransferase activity, indicative of hepatocellular injury, that returned to normal within two months after exposure ended (Glaser et al. 1986). Increased relative liver weight was observed in rats following cadmium chloride inhalation at 1.06 mg Cd/m3 (6 hours/day, 5 days/week, 62 days), while several other chronic-duration studies at somewhat lower concentrations did not reproduce this finding, suggesting a threshold-dependent relationship (Kutzman et al. 1986). The liver’s relative resistance compared to the kidney is generally attributed to its higher capacity for local metallothionein synthesis, which more effectively buffers free cadmium ion at the hepatocyte level (Goering and Klaassen 1984a, 1984b, 1984c, describing tolerance mechanisms to cadmium hepatotoxicity).

4.4 Cardiovascular Effects

The animal cardiovascular database for cadmium is comparatively sparse and, in the inhalation studies available, largely negative: rats exposed to cadmium oxide up to 0.88 mg Cd/m3 for 13 weeks showed no significant alteration in systolic blood pressure or cardiac histopathology (National Toxicology Program 13-week bioassay data as compiled in the toxicological literature). Where cardiovascular effects have been reported experimentally, findings are inconsistent across laboratories, and some investigators specifically failed to reproduce cadmium-induced hypertension in chronically exposed rats under carefully controlled conditions (Fingerle et al. 1982), while others have reported vascular functional changes in chronically exposed rabbits (Boscolo and Carmignani 1986) or hypertension in rats under differing exposure protocols (Balaraman et al. 1989). Taken together, the experimental cardiovascular evidence for cadmium is considerably weaker and more equivocal than the renal, pulmonary, or skeletal evidence.

4.5 Immune System Effects

Acute inhalation exposure of mice to cadmium chloride at 0.190 mg Cd/m3 for 2 hours suppressed the primary humoral immune response, with a no-observed-adverse-effect level of 0.11 mg Cd/m3 in the same study (Graham et al. 1978). Spleen lymphocyte cytotoxicity was observed in mice following a 1-hour exposure to 0.88 mg Cd/m3 (Krzystyniak et al. 1987, as compiled in the toxicological literature). At intermediate durations, increased relative spleen weight and lymphoid hyperplasia were observed in rats following cadmium chloride inhalation at 1.06 mg Cd/m3 for 62 days (Kutzman et al. 1986), and enlarged thoracic lymph nodes were noted in animals that died during chronic-duration cadmium sulfate and cadmium oxide inhalation studies (Oldiges and colleagues, as compiled in the toxicological literature). The functional consequences of cadmium immunomodulation appear complex and route/dose dependent: the same exposure paradigm has been reported to decrease resistance to bacterial infection while simultaneously increasing resistance to viral infection in mice (Bouley et al. 1982), and cadmium chloride has been shown to affect chemical- and viral-induced tumor production in separate murine models (Blakley 1985, 1986, 1988).

4.6 Musculoskeletal Effects

RoseyFields — Cadmium Toxicology Summary

No animal studies of musculoskeletal effects following inhalation exposure to cadmium were identified in the compiled literature. The oral/dietary database, by contrast, is substantial: young, growing rats given cadmium orally show increases in bone turnover and decreases in mineralization consistent with the skeletal pathology (osteomalacia, osteoporosis, and increased fracture risk) documented in heavily exposed human populations, most notably the itai itai disease cohort in Japan (mechanistic and dose-response synthesis compiled in the toxicological literature; see also the extensive rodent skeletal dose-response series by Brzóska and Moniuszko-Jakoniuk, e.g. Brzóska and Moniuszko Jakoniuk 2004a, 2004b, 2005a–d, examining bone mineral density, biomechanical strength, and fracture susceptibility across chronic low-to-moderate oral cadmium dose ranges in rats). These skeletal effects are generally understood to be secondary to cadmium-induced renal tubular injury and the resulting disturbance of calcium, phosphorus, and vitamin D metabolism, rather than a direct primary action of cadmium on bone (compiled mechanistic discussion; Blainey et al. 1980 in the human occupational literature draws the same secondary-injury conclusion).

4.7 Neurological Effects

Cadmium is not generally regarded as a potent adult neurotoxicant following inhalation exposure, though acute high concentration exposure has produced tremors (cadmium carbonate, 132 mg Cd/m3, 2 hours) and reduced locomotor activity (cadmium fume, 112 mg Cd/m3, 2 hours) in rats (Rusch et al. 1986). No neurological effects were observed in rats continuously exposed for 30 days to cadmium chloride at 0.105 mg Cd/m3, cadmium dust at 0.098 mg Cd/m3, or cadmium sulfide at 1.034 mg Cd/m3 (Glaser et al. 1986). By contrast, developmental exposure has produced more consistent neurobehavioral findings: rats exposed to cadmium oxide by inhalation prior to and during gestation at 0.02 or 0.16 mg Cd/m3 showed decreased exploratory motor activity and impaired avoidance-acquisition performance in offspring assessed as juveniles (Baranski 1984, 1985). Independent oral developmental studies have similarly reported neurobehavioral alterations in offspring following low-level in utero cadmium exposure (Ali et al. 1986).

4.8 Reproductive Effects

In rats, inhalation exposure to cadmium oxide dust at 1 mg Cd/m3 (5 hours/day, 5 days/week, 20 weeks) increased the duration of the estrous cycle (Baranski and Sitarek 1987). A comparable lengthening of estrous cycle length was reported in rats exposed to cadmium sulfate aerosol at 2.8 mg Cd/m3 before and during pregnancy, and a 13-week cadmium oxide bioassay at 0.88 mg Cd/m3 reported both a significant increase in estrous cycle length and a significant decrease in spermatid counts in exposed males (National Toxicology Program data as compiled in the toxicological literature). By contrast, male and female rats exposed to cadmium chloride at 1.06 mg Cd/m3 (6 hours/day, 5 days/week, 62 days) and subsequently mated with unexposed partners showed no reduction in reproductive success as measured by viable embryos or preimplantation loss, although males in this study did show increased relative testes weight (Kutzman et al. 1986); similarly, no alteration in fertility was observed in female rats exposed to cadmium oxide at 0.16 mg Cd/m3 for five months prior to mating (Baranski 1984). Testicular injury has also been demonstrated experimentally via non-inhalation routes: cadmium disrupts tight-junction-associated microfilaments in rat Sertoli cells and can cause failure of spermiation after a single low parenteral dose (Hew et al. 1993a, 1993b), and zinc, cysteine, and selenium have each been shown to provide partial protection against cadmium-induced vascular injury to the mouse testis (Gunn et al. 1968a, 1968b), underscoring that testicular vascular damage is a well-established, if route-dependent, effect of cadmium in rodents.

4.9 Developmental (Prenatal/Postnatal) Effects

In utero cadmium exposure produces a consistent constellation of developmental effects in rodents: decreased pup viability, decreased fetal and postnatal body weight, delayed skeletal ossification, and neurobehavioral impairment. Decreased pup viability (survival to postnatal day 4) was observed in offspring of rats exposed to cadmium oxide at 0.16 mg Cd/m3 by inhalation for five months before mating and through gestation day 20 (Baranski 1984). Decreased fetal body weight was observed at ≥0.581 mg Cd/m3 (cadmium chloride) and in comparable cadmium oxide inhalation studies in both rats and mice, generally accompanied by maternal toxicity (reduced maternal weight gain, hypoactivity,

RoseyFields — Cadmium Toxicology Summary

dyspnea) at the same exposure levels, complicating attribution of the fetal effect to a direct versus maternally mediated mechanism (Prigge, cadmium chloride gestational studies as compiled in the toxicological literature; National Toxicology Program cadmium oxide developmental data). Delays in skeletal ossification of the sternum and ribs were reported in offspring of rats given cadmium by gavage at 2 mg Cd/kg-day on gestation days 7–16, with fused lower limbs, decreased numbers of live fetuses, and increased resorptions at 40 mg Cd/kg-day (Baranski 1985).

Frank teratogenicity (structural malformation) appears to require substantially higher oral doses than those producing growth or skeletal delay: a significant increase in malformations was observed in offspring of rats given 18.39 mg Cd/kg-day by gavage on gestation days 6–15, whereas no developmental effects were observed at 12.5 mg Cd/kg-day administered via drinking water over the same gestational window (Machemer and Lorke 1981). This roughly ten fold-to-order-of-magnitude gap between the lowest dose associated with growth/ossification delay (on the order of 2 mg Cd/kg-day) and the lowest dose producing frank malformation illustrates a graded, dose-dependent developmental toxicity profile rather than a single threshold for all developmental endpoints. Neurobehavioral alterations in offspring — decreased exploratory motor activity, impaired avoidance-acquisition, altered open-field ambulation, and prolonged latency on negative geotaxis testing — were reported following both low (0.02 mg Cd/m3) and higher (0.16 mg Cd/m3) maternal inhalation exposures in the Baranski (1984, 1985) study series, with some inconsistency between the two reports as to which effects occurred at which exposure level.

5. Animal Carcinogenicity Bioassay Findings

The animal carcinogenicity database for cadmium is route-dependent in a way that is central to interpreting its overall weight of evidence. By the inhalation route, Takenaka et al. (1983) exposed Wistar rats to cadmium chloride aerosol at 12.5, 25, and 50 µg Cd/m3 for 18 months (23 hours/day, 7 days/week), followed by a 13-month observation period, and reported statistically significant, concentration-related increases in primary lung tumors, alongside the adenomatous bronchoalveolar hyperplasia noted in Section 4.1. This study is widely regarded as the pivotal demonstration that inhaled cadmium is a complete pulmonary carcinogen in rats and forms the quantitative basis for an inhalation-route cancer potency estimate that is roughly fifty-fold higher than the potency estimate derived independently from human occupational cohort data (see Section 6).

By contrast, intratracheal instillation of cadmium oxide in Fischer 344 rats did not produce lung tumors but instead produced mammary tumors in male rats and tumors at multiple other sites, indicating that the tumor-site profile obtained from bolus intratracheal instillation does not simply replicate the inhalation-route (aerosol) result and should be interpreted as a distinct, less physiologically representative exposure paradigm (Sanders and Mahaffey 1984). Injection-site sarcomas and distant-site tumors, most notably testicular tumors, have been produced reproducibly in rats and mice following intramuscular or subcutaneous administration of cadmium metal or cadmium chloride, sulfate, oxide, and sulfide salts across numerous independent studies; a carcinogenic effect of cadmium on the rat prostate following parenteral administration has also been reported specifically (Hoffmann et al. 1985). These parenteral-route tumors are informative as to cadmium’s intrinsic carcinogenic capacity at a cellular level but are of limited direct relevance to human inhalation or dietary exposure, since the injection route bypasses normal absorption, first-pass hepatic sequestration, and metallothionein-mediated detoxification.

Notably, seven independent oral carcinogenicity bioassays in rats and mice, using cadmium salts (acetate, sulfate, chloride) administered in the diet or drinking water, have shown no evidence of a carcinogenic response by the oral route (weight-of-evidence synthesis reflected in both the EPA and ATSDR compilations of this literature). This oral/inhalation discordance — positive, concentration-related pulmonary carcinogenicity by inhalation; no demonstrated carcinogenicity by the oral route despite a substantial testing effort — is one of the more consistent and toxicologically important features of the cadmium animal database, and it parallels the pattern seen in the human epidemiological literature (Section 6), where lung cancer risk has been associated with occupational inhalation cohorts but a comparable dietary/oral cancer signal has not been established.

RoseyFields — Cadmium Toxicology Summary

Additional inhalation carcinogenicity data in a second species (hamster) and in mice have examined multiple cadmium compounds, generally supporting the conclusion that the pulmonary carcinogenic response is a class effect of inhaled cadmium rather than being restricted to a single salt, though comparative potency across compounds and species has not been fully characterized (Heinrich et al. 1989).

6. Comparison of Animal-Derived Findings to Human Epidemiological Findings

6.1 Renal Toxicity

Animal and human renal findings are highly concordant in character (proximal tubular dysfunction manifesting as low-molecular-weight proteinuria, progressing at higher body burdens to glomerular filtration rate decline) and in the underlying mechanism (accumulation of unbound cadmium once renal cortical metallothionein-binding capacity is exceeded). The principal divergence is one of exposure-response characterization rather than qualitative disagreement: human data derive overwhelmingly from occupational cohorts and environmentally exposed populations in which individual cumulative dose is reconstructed indirectly (from urinary or blood cadmium, or from estimated air/dietary concentrations over years to decades), for example in the large Cadmibel population study of Belgian residents (Staessen and Lauwerys 1993) and its subsequent five-year follow-up (Hotz et al. 1999), whereas animal bioassays permit direct, controlled dosing but rarely extend to the multi-decade cumulative exposures characteristic of human occupational cadmium nephropathy. This difference in achievable exposure duration, rather than any species difference in mechanism, is the principal source of quantitative uncertainty when extrapolating animal renal NOAELs/LOAELs to chronic human risk.

6.2 Pulmonary Toxicity and Carcinogenicity

The qualitative pattern of pulmonary injury (interstitial pneumonitis progressing to fibrosis, emphysema, and reduced diffusing capacity with chronic exposure) is well conserved between occupationally exposed workers (e.g., Davison et al. 1988; Chan et al. 1988; Cortona et al. 1992) and inhalation-exposed rodents. For carcinogenicity, however, animal and human potency estimates diverge substantially: the rat inhalation unit risk derived from the Takenaka et al. (1983) bioassay (approximately 9.2 × 10⁻² per µg/m3) is roughly fifty-fold higher than the unit risk derived from the human occupational cohort data of Thun et al. (1985) (approximately 1.8 × 10⁻³ per µg/m3). Regulatory risk assessments have generally favored the human-derived estimate as more directly relevant despite its lower statistical power, reasoning that species differences in respiratory tract dosimetry, particle deposition and clearance, and the confounding presence of arsenic and tobacco smoke in some human cohorts (addressed analytically in Thun et al. 1985) make the rodent bioassay a conservative but less physiologically representative basis for human risk quantification.

A further divergence concerns exposure route: the rodent evidence for cadmium carcinogenicity is strongest by inhalation and by parenteral injection, while seven independent oral bioassays have been uniformly negative. Human epidemiological evidence for cancer risk is similarly concentrated in inhalation-exposed occupational cohorts (smelting, battery manufacture, pigment production), with case-control and cohort evidence for prostate cancer specifically showing an inconsistent pattern across studies — an initial signal in several early cohorts (Kipling and Waterhouse 1967; Lemen et al. 1976) that was not confirmed, and in some analyses was specifically contradicted, by later, larger, or updated cohorts (Thun et al. 1985, an update of the Lemen et al. 1976 cohort, found no excess prostate cancer risk).

6.3 Reasons for Divergence Between Animal and Human Data

Several factors plausibly explain the quantitative (though generally not qualitative) divergence between the animal and human cadmium toxicology databases:

RoseyFields — Cadmium Toxicology Summary

• Toxicokinetic differences in particle deposition, mucociliary clearance, and pulmonary retention half-life of inhaled cadmium particles across species, which affect the cadmium dose actually delivered to target lung tissue at a given ambient air concentration (Oberdörster et al. 1994).
• Differences in achievable study duration — lifetime rodent bioassays span roughly two years, while human chronic occupational and environmental exposures commonly span two to five decades, so that cumulative renal and pulmonary cadmium burdens reached in the most heavily exposed human cohorts may exceed those achievable in a standard rodent bioassay design.
• Confounding exposures in human occupational cohorts (arsenic co-exposure in smelting operations, tobacco smoke) that complicate isolation of the cadmium-specific contribution to cancer risk, an issue explicitly analyzed and, in the Thun et al. (1985) cohort, judged not to fully explain the observed excess lung cancer risk.
• Route-specific differences in first-pass handling: inhaled cadmium reaches the lung directly and bypasses the substantial first-pass hepatic metallothionein sequestration that limits the fraction of an oral dose reaching target organs in a toxicologically active (unbound) form, plausibly contributing to the oral/inhalation discordance in carcinogenicity findings described above.
• Adaptive pulmonary responses demonstrated experimentally (Hart et al. 2001) — reduced acute inflammatory injury but impaired DNA repair and apoptosis in cadmium-adapted lung epithelium — illustrate a mechanism by which chronic low-level human occupational exposure could carry a disproportionate long-term carcinogenic risk relative to what would be predicted from acute or subchronic rodent injury endpoints alone.

7. Data Gaps and Areas of Ongoing Research

• Quantitative reconciliation of the animal (inhalation unit risk, Takenaka et al. 1983) and human (Thun et al. 1985) cadmium lung cancer potency estimates remains incomplete; the roughly fifty-fold difference between them has not been fully resolved mechanistically or toxicokinetically.
• The oral carcinogenicity database, while consistently negative across seven independent rodent bioassays, predates much of the modern mechanistic literature on cadmium genotoxicity and DNA-repair inhibition; whether contemporary bioassay designs (e.g., longer duration, inclusion of susceptible genetic backgrounds, or co-exposure paradigms) would alter this conclusion has not been systematically tested.
• Cardiovascular effects of cadmium remain the least consistent endpoint across both the animal and human literature, with some rodent and human studies reporting positive associations with blood pressure or vascular disease and others, including large population-based studies such as the Cadmibel cohort, finding none; the basis for this inconsistency (dose range studied, co-exposure to lead, duration of follow-up) warrants further characterization.
• The relative contributions of direct fetal cadmium toxicity versus maternally mediated (nutritional/toxic) effects to the developmental toxicity observed in rodent inhalation and gavage studies have not been fully disentangled, since maternal toxicity (reduced weight gain, respiratory distress) frequently co-occurs with the exposure levels producing fetal effects.
• Mechanistic work on cadmium-induced pulmonary adaptation (tolerance) and its implication for long-term carcinogenic risk in chronically, intermittently, or low-level exposed human populations remains an active area of investigation, with the balance between protective (anti-inflammatory) and adverse (DNA-repair impairing) consequences of this adaptive response not yet fully characterized in human tissue.
• Species and sex differences in susceptibility (e.g., the greater lethality observed in female versus male rodents in several chronic inhalation studies) are noted empirically in the literature but their mechanistic basis, and their implications for interpreting human sex-specific epidemiological findings, have not been comprehensively addressed.

References