resorcinol; 1,3-benzenediol

Specific investigations: other studies

Description of key information

Some early laboratory animal studies via dermal and oral routes under particular conditions along with human case reports (at high dermal exposures to damaged skin) have suggested that resorcinol may have a potential to affect mammalian thyroids. No thyroid effects were seen in numerous other studies, however, including animal studies conducted under clinically relevant conditions, medical case reports under the current therapeutic conditions, as well as occupational investigations of exposed worker populations, suggesting acceptable risk to humans based on these scientific and weight-of-evidence considerations. Recently, a well-conducted study (OECD TG 416 with detailed evaluation of thyroid endpoints) found no significant effects on the thyroid of rats given up to 233 mg/kg bw/day (males) or 304 mg/kg bw/day (females) in the drinking water through two generations. It is also generally accepted that humans are less susceptible than rats to thyroid disruption by chemicals. Thus, the risk that resorcinol presents to humans in terms of thyroid dysfunction resulting from occupational, environmental, or exaggerated use of proprietary preparation is negligible.

Additional information

Effect on the Thyroid and its relevance to the effect on development, especially for the brain:

Historically, there has been a focus on resorcinol’s potential to affect the thyroid, so a presentation of well-conducted studies/findings is warranted even though these studies do not meet current test guidelines.

(1)    The potency to inhibit TPO

Thyroid peroxidase (TPO) inhibitory activity of resorcinol has been previously reported using many models (Divi and Doerge, 1994; Fraser et al., 1955; Lindsay et al., 1992). TPO is a heme-containing multifunctional enzyme located in the apical membrane of follicular thyroid cells and is essential for the synthesis of thyroid hormones (THs) (Ehrenshaft and Mason, 2006). Recently, a study was reported which suggests that resorcinol could also inhibit TPO from the human thyroid follicular cell line (Jomaa et al., 2015). In this study, however, whole cell lysates are used as the TPO fraction, which means that many kinds of cytosolic soluble peroxidase were also included in addition to the TPO on the apical membrane. Therefore, contamination with such cytosolic soluble peroxidases is supposed and the true inhibitory activity of TPO itself may not have been evaluated accurately.

In the recent study (Paul et al., 2014) in which an optimised in vitro TPO activity assay was conducted with rat thyroid microsomes, the IC50 of resorcinol was reported to be 253 μM, indicating that resorcinol has very low potency to inhibit TPO. However, another later study by the same authors (Paul Friedman et al, 2016) reported a remarkably different value of IC50 for resorcinol to be 0.025μM under similar in vitro test conditions. The reason for this difference is not clearly explained in the report, but was considered by the authors to be likely resulted from differences in test chemical quality or purity. However, it is difficult to draw further conclusions on such a large difference in effect. 

Furthermore, the data (Paul et al., 2014) was employed in the investigation for developing an IVIVE (in vitro to in vivo extrapolation) approach (Leonard. et al., 2016). It was suggested that not only the potencies of TPO inhibition but also their pharmacokinetics as well as pharmacodynamic behaviours should be taken into account when assessing the risks of TPO inhibitors. In the analysis, low potency of blood T3/T4 reduction in humans was tentatively suggested for resorcinol.

(2)    Animal studies on thyroid dysfunction

According to the Adverse Outcome Pathway (AOP) proposed by Crofton (2012) and Paul et al. (2014), inhibition of TPO functions may cause a reduction of blood THs levels, resulting in the effects on the thyroid glands.

Effects on the thyroid gland have been reported in earlier studies following administration of resorcinol in a way that allowed for a continuous release of resorcinol to the systemic circulation (Arnott and Doniach, 1952; Doniach and Fraser, 1950; Doniach and Logothetopoulos, 1953; Samuel, 1955, Cooksey et al., 1985; Koppers Company, 1977; Seffneret et al., 1995). These studies may have been single dose, leading to limited conclusions with regards to a dose-depended effect and/or involved subcutaneous injections in oily solution as well as with low protein and low iodine conditions, which are not directly clinically relevant. Furthermore, some of the reports contain insufficient information to determine reliability (RTF, 2011).

On the other hand, even with subcutaneous injections, no effect on thyroid was indicated when resorcinol was dissolved in aqueous solution. Twenty five male rats were given a total daily dose of 100 mg/kg bw/day of resorcinol in two separate doses (50 mg/kg) via subcutaneous injections over a 30-day period (Merker et al., 1982). Dosing did not result in any overt toxic signs or adverse changes in body weight gain, organ weight (liver, kidney, brain, spleen and testes), haematocrit, haemoglobin, red blood cell count and serum T3 and T4. Additional histopathology was judged to be within normal limits for the three organs examined: thyroid gland, spinal cord and brain. This report was considered as reliable with restriction (RTF, 2011).

A series of studies conducted by NTP (1992) which are considered as reliable showed no histopathological effects on the thyroid of rats or mice at doses via oral gavage of up to 520 mg/kg bw/day in rats and 450 mg/kg bw/day in mice for 13 weeks and 150-225 mg/kg bw/day for 5 days/week over 2 years in rats and mice. Although the T3/T4 ratio was not assessed at all doses in this study, T3 and T4 levels were assessed at the control and 130 mg/kg/day dose levels and no treatment-related effects on T3 and T4 were observed.

Furthermore, in a 2-generation reproductive toxicity study according to OECD Guidelines (WIL Research Laboratories, 2005), rats(30/sex/group) were exposed to resorcinol via drinking water at 0, 120, 360, 1000, and 3000 mg/l. For thyroid/pituitary hormone analysis (T3, T4 and TSH), the blood samples were collected via the vena cava from 15 randomly selected F0 or F1 parental animals or in the F1 or F2 pups (PND 4 or PND 21).The organs including thyroid gland were weighed from all F0 and F1 parental animals at the scheduled necropsies. For microscopic examination, selected tissues including the thyroid gland from all F0 and F1 parental animals in the control and high concentration level (3000 mg/l) groups and all animals found dead or euthanised in extremis, were examined. The thyroids of all F0 animals in the 1000 mg/l group were also examined. A peer review of the microscopic thyroid evaluation was conducted. In addition, the stereomicroscopy was done on 15 randomly selected F0 parent animals per sex in the control and 3000 mg/l groups and 15 randomly selected F0 parent males in the 1000 mg/l. The results showed no statistically significant test article-related changes in the mean concentrations of blood T3, T4 or TSH. The higher TSH values noted in the F0 males at the scheduled necropsy were not considered test article-related in the absence of effects on T3 or T4, organ weights or adverse macroscopic or microscopic findings. Decreased colloid in the thyroid histopathology was observed only in the 3000 mg/l group F0 males. However, this effect was not considered adverse due to the lack of associated functional effects. The NOAEL is considered to be 3000 mg/l for parental systemic and offspring toxicity (ca. 233 mg/kg/day (males), 304 mg/kg/day (females (premating and gestation)), and 660 mg/kg/day (females (lactation)) while the NOEL is 1000 mg resorcinol/l for decreases in water consumption (ca. 86 mg/kg/day (males), 126 mg/kg/day (females (premating and gestation)) and 225 mg/kg/day (females (lactation)). This study was considered as reliable without restriction (RTF, 2011).

Data on toxicokinetic analyses has revealed that resorcinol is readily absorbed upon oral administration, but was rapidly metabolised and excreted in the urine primarily as monoglucronide conjugate (EFSA, 2010). The data obtained by Merker et al., 1982 and by Kim and Matthews, 1987 showed no resorcinol accumulation in any organ or tissue, including the thyroid gland, when 14C- resorcinol was administered either subcutaneously or orally to rats. This can be reasoned by the properties of resorcinol of being metabolised and excreted rapidly from the body, and may explain the lack of thyroid effects in animal studies. These reports were considered as reliable with restriction.

Based on the above evidences, the effects of resorcinol on the thyroid gland are attained when the internal exposure is high, for instance, due to the manipulation of the toxicokinetics (e.g. subcutaneous injection in an oily solution). Without such particular manipulations, and thus, under clinically relevant exposure conditions, no effect on thyroid gland has been detected. This is very likely attributable to the rapid elimination of resorcinol as well as its low potency to inhibit TPO.

(3)    Human studies on thyroid dysfunction

Medical cases:

In humans, resorcinol has been used therapeutically in dermatology since the late 19th century based on its reported properties to promote the healing process and to treat conditions such as ulcerating skin lesions. At that time well-controlled, standardised safety and efficacy assessments of chemicals intended for therapeutic use in humans were not conducted. Clinical case reports were often the common means to communicate trial-and- error observations in the medical literature and beneficial or adverse outcomes in human patients. The over-exuberance of early, unregulated pharmacists during the period prior to 1950 led to the use of topical ointments containing up to 50% free resorcinol. Resorcinol at these high doses is associated with reversible hypothyroidism based on an established mechanism related to the inhibition of thyroid peroxidase enzymes.

Extended exposure to high levels of free resorcinol, when applied to open wounds, led to several medical case reports reporting goitres (Boeck, 1915; Klem, 1930; Strakosch, 1943; Bull and Fraser, 1950). Bull and Fraser (1950) reported the clinical signs (enlarged thyroid glands, hypoactivity) in three case reports, where ointments containing resorcinol (up to 12 %) were applied onto the skin (leg ulcers) over long time periods.

Thyroid gland effects, particularly reversible goiters, were typically observed only upon extended periods of application to skin with compromised barrier function. A common denominator is that the subjects applied ointments containing up to 12.5% resorcinol in copious quantities to large areas of ulcerated and thus barrier-impaired, skin over extended periods (Katin et al., 1977; RTF, 2014). Katin et al. (1977) reported one case of hypothyroidism in a 70-year-old male patient caused by long-term dermal application (ca. 3 months) of large amounts of a paste containing 2 % resorcinol to dry and coarse skin containing multiple senile keratoses. After stopping the use of resorcinol, free thyroxine and TSH were within normal limits within 2 weeks. While the administered dermal resorcinol concentrations in these historic case reports were typically high, the actual delivered (systemic) doses or the quality of the raw material specification cannot be verified.

Welsch, 2008 (Routes and Modes of Administration of Resorcinol and their Relationship to Potential Manifestations of Thyroid Gland Toxicity in Animals and Man) provides documentation to explain compromised human skin barrier function is a likely cause of drastic increases in resorcinol absorption in human clinical case reports. Benfeldt et al. (1999) demonstrated that penetration/absorption of a topically applied drug is increased when the human skin barrier function is experimentally impaired based on the data collected on salicylic acid (SA) in 18 human subjects. SA is actually judged as being comparable to resorcinol in its keratolytic properties. This may be generally applicable to hydrophilic agents. Skin absorption differences may thus be profound and penetration may be up to 150-fold higher in severely barrier function impaired human skin. One can conclude that the human myxedema and hypothyroidism occurring in patients with skin ulcers were caused by prolonged high rates of absorption of resorcinol through barrier impaired skin. The disease conditions allowed sufficiently high concentrations of free resorcinol to reach the human thyroid gland and exert thyrotoxic effects.

The RTF report (RTF, 2014) concluded that it was difficult to assess the significance of the hypothyroidal effects to resorcinol sequelae in these case reports. What can be unequivocally stated is that clear warning signals about high dose resorcinol applications and thyroid toxicity in human patients of both genders had been posted during this time period. When tighter self-regulation of the topical ointment sector was implemented, these case histories, however, declined and there was only a very small sample of patient clinical cases reported over a period of approximately 25 years from 1950 to 1977 (Berthezene et al., 1973, and Pascher, 1978). To date there are no known recurrences of the medical case reports of the earlier period and the risks associated with the inclusion of resorcinol within topical ointments appear to be fully controlled with limits of <5% or more typically <2% resorcinol within currently marketed dermatological products. All raw material specifications for medicinal applications in modern times are manufactured to the highest quality and grade. It is evident that as the dose and practices and conditions of use were modified the case reports were not observed.

In a clinical study, resorcinol was applied topically to the face, shoulders, upper chest and upper back of 3 healthy men, twice a day, 6 days/week over a period of 4 weeks. A 20 ml hydroalcoholic vehicle containing 2% resorcinol (800 mg resorcinol per day corresponding to a daily dose 12 mg/kg/bw and 0.30 mg/cm²) was applied. No detectable levels of free resorcinol or its conjugates were found in blood (detection limit of the method applied was 0.1 µg/ml). In 24-hour urine samples collected after 14 days of continuous treatment, a maximum of 0.47 to 2.87% (up to 23 mg resorcinol) of the applied daily dose was excreted and detected as the glucuronide and sulphate conjugates. Blood chemistry and thyroid function (T3, T4, T7 and TSH) were normal throughout the study. Topical exposure to resorcinol at 2% concentration was considered to be safe, even under exaggerated conditions of use at this concentration (Yeung et al., 1983). Thus, it was suggested when applied on healthy skin, percutaneous absorption is limited and rapid metabolism in animals precludes resorcinol from reaching thyroid gland toxic concentrations.

Topical applications of resorcinol are still prescribed for the treatment of acne, seborrheic dermatitis, eczema, psoriasis, and other skin disorders representing a significant patient exposure to this ingredient. It is reasoned based on the significant exposure to resorcinol at current therapeutic exposures and conditions of use that there is an absence of an alert for clinical safety as no case reports can be found.

As in the case with animal studies, the reported effects on thyroid in humans are due to anomalously high and continued internal exposure by unregulated applications to patients in the past. Under the current therapeutic conditions, such effects should be never observed.

Occupational exposure:

In 1978, medical examinations, chest X-rays and pulmonary function, haematology and clinical chemistry were performed with 281 of 329 persons actively employed at a resorcinol manufacturing plant having occupational exposures to resorcinol, benzene, by-products of benzene sulfonation, and formaldehyde. About 60 % were under 40 years of age and about 50 % had worked at this plant for at least 10 years. The jobs of these workers at the time of medical examination entailed no or minimal exposure to resorcinol (dry fusion operator, dry fusion material handler, process engineer, storekeeper, labourer, and utility operator). The prevalence of medical findings possibly consistent with subclinical hypothyroidism (low T4 and/or high TSH) was 5/280 (1.8 %) and the prevalence of possible goiter was 2/280 (0.7 %). One person showed a palpable thyroid with normal T4 and TSH values (TOMA, 1978). The investigators concluded that the thyroid evaluations “did not reflect any resorcinol hazards from the work environment”. In 1980, medical examinations (see above) and thyroid assessments were performed with 247 of 387 presumably active plant workers (214 men and 33 women). About 60 % were under 40 years of age and 153 of these subjects were tested for total T4 and TSH. 5/153 (3.3 %) showed signs of clinical / subclinical hypothyroidism, but in 3 of these 5 cases other reasons such as treatment with radioiodine were given as cause for the thyroid abnormalities (observed findings (TOMA, 1981)). The report did not provide any evidence that resorcinol exposure was responsible for the thyroid effects in the remaining two workers. The 3rd study performed in 1984 included 192 of 312 active workers. In 188 subjects (175 men, 13 women) with a mean age of 37 years both laboratory and other tests incl. medical examination were done and in none of the subjects abnormal thyroid glands or changes in T4 values were found when compared with normal values (Bauer, 1985). These data are of limited reliability due to small study sizes, lack of comparison groups, missing current and historical control data and missing information concerning potential exposure categories.

Roberts et al. (1990) investigated 4 cases of clinical overt hypothyroidism over a 6 year time period among 539 subjects working in a textile factory. In the finishing departments both thiourea and resorcinol were used and measurements taken at the inlet of the local exhaust ventilation of stenters gave concentrations of 5 µg/m³for thiourea and < 20 µg/m³for resorcinol. About 44 % of the total workforce (189 men and 48 women) participated. 115 persons were process workers and 122 worked in the management, office and laboratory jobs. Thyroid function tests included and antimicrosomal / antithyroglobulin antibodies and participants also filled in a questionnaire. The study found 15 new cases of thyroid abnormalities: one case of thyroid hyperactivity and 14 cases of hypothyroidism. Of these 14, one had inherited pituitary hypothyroidism and one partial thyroidectomy. In the remaining 12 cases (7 males [age distribution 26 – 60 years; 5 females [age distribution 18 – 58 years]), 2 males and 2 females (3 of them had minor symptoms of hypothyroidism) showed slightly increased TSH values. The other 8 persons had normal TSH values and no symptoms but increased circulating thyroid antibodies. Seven of the twelve individuals with elevated TSH levels or elevated thyroid antibodies were in the control group and designated as “unexposed” in the study because they were management, administration, and office staff. The authors say in the report they have not been able to demonstrate a statistically significant occurrence of biochemically and immunologically detectable disturbances in thyroid function among this workforce. Even if these findings are of significance, the interpretation of the data is difficult since thiourea is also a goitrogenic agent and because the rates of potential thyroid effects are higher in the control group compared to the exposed group.

It is important to note that numerous findings are available in humans which do not report thyroid effects after exposures to resorcinol.

(4) Animal studies on development, especially for the brain

According to the Adverse Outcome Pathway (AOP) proposed by Crofton (2012) and Paul et al. (2014), inhibition of TPO functions may cause effects not only on the thyroid gland, but on development, especially for the brain.

Maintenance of maternal thyroid function during pregnancy is important for normal development, in particular during neurological development of the foetus. In the first trimester of human gestation, the embryo/foetus depends entirely on the maternal thyroid hormones (THs): thyroxine (T4) and/or triiodothyronine (T3). Later in pregnancy and during lactation, maternal THs still contribute significantly to foetal and neonatal thyroid homeostasis (Hartoft- Nielsen et al., 2011).

The potential developmental neurotoxicity of resorcinol was explored in Sprague-Dawley rats exposed at 0, 10, 40, 120 and 360 mg/l in drinking water. No test article related changes in the mean concentrations of blood THs and TSH were detectable in the F0 and F1 generations. Furthermore, functional observational battery (FOB) evaluations, acoustic startle response and Biel maze swimming trials in F1 pups were unaffected at all resorcinol dose levels. There were no qualitative microscopic changes in the brain (forebrain, midbrain or hindbrain) and locomotor activity was also unaffected on PND21. Higher locomotor activity values among F1 males and females were observed on PND 61. These increases, however, were not considered conclusive evidence for a change in CNS function because no concurrent correlating histopathological changes in the brain were detectable and there was a lack of clear dose-response relationships. No indication of developmental delay or other changes in CNS function were detected (WIL Research Laboratories, 2003). This study was considered as reliable with restriction.

Exposure of rats to resorcinol under clinically relevant conditions did not cause reduction of serum THs levels and thus, any developmental neurotoxicity. This confirms that the potency of a chemical to inhibit TPO in vitro is not indicative of its potential to affect THs levels in vivo in maternal rats and their offspring, and thus, does not necessarily disturb normal development, especially for the brain.

(5)    Species Differences

It is known that the rat is more sensitive than humans to disruption in thyroid function (Hard, 1998; Capen, 2000) as the result of a substantially shorter half-life for circulating thyroxine (12-24 hours in rats versus 5-9 days in humans). The fact that rodents are more sensitive to the effect of thyroid disrupting agents than humans has been thoroughly discussed in WRc-NSF (2002). Specifically, WRc-NSF (2002) stated in the chapter on resorcinol: “It needs to be recognised that rodents, especially rats, have been reported to be particularly susceptible to goitrogens, primarily due to the lack of thyroid binding protein (TBP) which is the primary protein for thyroid hormone binding and transport. In the rat, the absence of TBP results in a much shorter half-life of T4 and much higher levels of TSH. These differences suggest that the activity of the thyroid gland in rats is considerably higher than that of other species, including humans, and this increased activity should correlate to a greater susceptibility to hormonally-induced thyroid effects. Thus, there are considerable differences in thyroid physiology and biochemistry between rats and humans, the rat being more susceptible to thyroid disrupting chemicals. This higher sensitivity in rats may provide an additional margin of safety in humans.