Thyroid disrupting effects of exposure to sunscreens: in vitro and in vivo evidence of the impact of organic UV-filters

Abstract

Background

Ultraviolet radiation (UVR), primarily from sunlight, is a natural component of environmental exposure and can cause genomic damage, photoaging, and skin cancer. Sunscreens containing UV-filters are widely used to prevent these effects and are also found in many personal care products. Despite their protective role, concerns have emerged regarding the accumulation of UV-filters in the environment and human matrices, along with their ability to act as endocrine disruptors, which main target include also thyroid hormone regulation.

Methodology

This narrative review was performed through a systematic search of PubMed (Medline), Embase, and the Cochrane Library. Global legislations were reviewed from institutional websites and official databases of major international agencies.

Main body

Various studies have shown that several UV-filters are detectable in human tissues, blood, urine, breast milk, and fetal circulation, suggesting systemic absorption and long-term exposure. Experimental evidence indicates that some UV-filters may affect hormone regulation by interacting with nuclear receptors, particularly those involved in thyroid function. Thyroid hormones are crucial for development, metabolism, and neurological function, mainly during critical life stages such as pregnancy and infancy. Disruption of thyroid hormone homeostasis may result in long-lasting adverse effects. While previous research has mainly focused on estrogenic and androgenic effects, increasing evidence highlights the need to assess thyroid-specific endpoints. This review summarizes current in vitro and in vivo findings on the potential thyroid-disrupting effects of UV-filters.

Conclusion

A better understanding of how UV-filters may interfere with thyroid hormone regulation is essential for evaluating their safety and informing regulatory decisions. Given the global variability in regulatory frameworks, this issue is particularly relevant for risk assessment and the development of safer and more sustainable alternatives in cosmetic formulations. Future research and awareness campaigns should focus on chronic exposure, mainly on vulnerable populations.

Background

Ultraviolet radiation (UVR) is defined as invisible electromagnetic radiation wavelength (100 nm to 400 nm) that is shorter than visible light but longer than X-rays. UVR is considered a natural exposome (environmental exposure that an individual encounter throughout life) for all humans. Indeed, the most common source of UVR is sunlight, but it can also be produced artificially. Exposure to UVR exerts biological effects, including damage to DNA and other cellular structures. UVR is divided into three bands defined as UV-A (315–400 nm), UV-B (290–315 nm) and UV-C (100–290 nm) [1, 2]. Among these three bands, UV-C radiation is not considered a significant source for human adverse effects. UV-A rays can penetrate deep into the skin and are primarily responsible for photoaging by activating cytokine cascades involved in inflammatory reactions and generating reactive oxygen species (ROS), which leads to melanocytes dysfunction and skin hyperpigmentation [3, 4]. UV-B radiation penetrates both the epidermis and dermis [5], causes sunburn, pigmentation, immunosuppression and skin carcinogenesis [6, 7]. Sunscreen products containing UV-filters are widely used in skincare with the aim of preventing damage from UVR, such as sunburn, production of reactive oxygen species and DNA damage, which can induce skin cancer [8, 9].

For this reason, UV-filters are found not only in sunscreens but also in personal care products such as shampoo, conditioners, shower gels, makeup products and perfumes [10]. Based on their physicochemical properties, UV-filters are divided into two main categories: inorganic (mineral compounds such as zinc oxide and titanium dioxide) and organic (synthetic, carbon-based molecules such as octinoxate, avobenzone, oxybenzone, homosalate) [11]. Inorganic filters offer advantages such as broad-spectrum UV-protection, which includes both UV-A and UV-B, and good photostability. Organic UV-filters absorb UVR and convert it into heat or into shorter wavelengths that are not dangerous for the skin [11]. Sunscreens based solely on inorganic filters (like zinc oxide and titanium dioxide) are mainly used in formulations for sensitive skin due to their low irritancy and broad-spectrum protection [12]. Most commercially available sunscreen products contain a combination of multiple UV-filters to ensure broad-spectrum prxxxxxAXaotection against both UV-A and UV-B radiation, as each compound typically absorbs only a specific portion of the UV spectrum. The synergistic use of different filters (aminobenzoic acids, ethyl cinnamates, salicylic acids and benzophenones) enhances photoprotective efficacy, formulation stability and increases the lipophilicity of the product, allow insufficient penetration into the skin [13]. Despite evidence that sunscreens can reduce photoaging and age-related skin diseases, several concerns have been raised from studies showing their accumulation in environmental (seawater, biota, and sediments) and human matrices (serum, urine, and breast milk) [14–17]. In particular, experimental studies, both in vivo and in vitro, have demonstrated that some UV-filters act as endocrine disruptors (EDs) compounds that can mimic, block, or alter the activity of hormones produced by the endocrine system [1, 18, 19]. At present, several UV-filters are recognised or suspected to be EDs, and their presence in different human matrices has been correlated with diabetes mellitus [20] and female infertility [21]. The global rise in thyroid diseases increased concerns about the endocrine-disrupting and, more specifically, thyroid-disrupting effects of various environmental contaminants [22]. Among endocrine disruptors, several classes of chemicals, including pesticides, plastics, food packaging materials, and industrial solvents, have been implicated in the disruption of thyroid function. Emerging hypotheses suggest that chronic exposure to such thyroid-disrupting chemicals may be a contributing factor to the increasing incidence of thyroid disorders [22]. In this view, concerns have emerged regarding the potential thyroid-disrupting effects of certain UV filters due to their widespread use and accumulation in the environment and in humans. The aim of this review is to summarize existing in vitro and in vivo studies investigating the effects of UV-filters on thyroid function. This will help clarify the potential risks and mechanisms by which UV-filters may impact thyroid health.

Main text

In parallel, regulatory information on UV-filters was retrieved from institutional websites and official databases of major international agencies. Global legislations were reviewed to identify substances currently authorized, restricted, or banned, with specific attention to regulations addressing endocrine and thyroid-disrupting potential. This comparison aimed to highlight differences across jurisdictions and assess the consistency of current regulatory frameworks with emerging scientific evidence.

Methodology

A comprehensive narrative review was conducted to provide an overview of the current knowledge on the potential thyroid-disrupting effects of UV-filters. Relevant literature was identified through a systematic search of PubMed (Medline, 1805 papers), Embase (1085 papers), and the Cochrane Library (12 papers). The search strategy included the following terms and, if applicable, the trade names: (“thyroid” [MeSH Terms] OR “thyroid” [All Fields]) AND (“UV filters” 43 papers [All Fields] OR “sunscreens” 56 papers [All Fields] OR “benzophenone” 73 papers [All Fields] OR “octinoxate” 17 papers [All Fields] OR “oxybenzone” 26 papers [All Fields] OR “4-MBC” 13 papers [All Fields] OR “homosalate” 10 papers [All Fields] OR “Endocrine Disruptors” 681 papers [All Fields]). No language restrictions were applied. Articles published from 1970 to 2024 were considered, including original in vitro and in vivo studies, relevant toxicological evaluations, and regulatory reports. Reference lists of selected articles were also screened to identify additional pertinent literature. Only studies investigating the effects of UV-filters on thyroid function, hormone levels, receptor interactions, gene expression, or developmental outcomes were included in the synthesis.

Sunscreens as endocrine disruptors

The reasons for which several UV-filters are considered EDs include structural similarity to hormones, hormone receptor binding, enzyme effects, and bioaccumulation [23].

In particular many studies highlighted the ability of some organic UV-filters to interfere with sex hormones. Just to give a few examples, benzophenones (BP) showed multiple ED effects mediated by an interference with estrogen receptor (ER), androgen receptor (AR), progesterone receptor (PR) and other nuclear receptors [24–28] in vitro, in vivo and in silico. Indeed both in vitro and in vivo studies showed that BPs could cause developmental and reproductive toxicity as well as multiple estrogenic and anti-androgenic effects [4]. UV-filters were demonstrated to be involved in estrogenic and androgenic effects, negatively affecting reproduction and development in many organisms. Table 1 summarises the effects of the most studied UV-filters in terms of interference with sex hormones.

Table 1.

UV-filters effects on sex hormones as endocrine disruptors

COMPOUND	ESTROGENIC EFFECT	ANDROGENIC EFFECT	REFERENCES
Benzophenones	Inhibition of the activity of 17β-Estradiol; Induction of VTG in fathead minnows; Activation of ERα, ERβ; Induction of proliferation of MCF-7 cell; Reduction of the uterine weight in immature Long-Evans rats	Antagonists of human AR transactivation; Repression of 4,5-dihydrotestosterone-induced transactivational activity; Inhibition of testosterone formation in mice and rats	[29–33]
4-Methylbenzylidene Camphor	Activation of ERα in a concentration-dependent manner; Increase of mRNA expression level of ERα and VTG in male Japanese medaka;	Inhibition of AR activity; AR antagonist and inhibit luciferase activity Reduction concentration-dependent of testosterone formation by inhibition of androgen-metabolizing 17β-HSD3	[27, 31, 34, 35]
3-Benzylidene Camphor	Increase in rats’ uterine weight; Induction of VTG in juvenile fathead minnow	Inhibition of AR activity; Reduction concentration-dependent of testosterone formation by inhibition of androgen-metabolizing 17β-HSD3	[27, 36, 37]
Octinoxate	Moderately activate ERα with no effect on ERβ; Effect on the uterus of female Sprague–Dawley rats; Increase of plasma concentration of VTG; Up-regulation of mRNA expression levels of ERα	Reduction prostate and testis weights, dose-dependent decrease in testosterone levels	[35, 38, 39]
Homosalate	Reduction of levels of luteinizing hormone and estradiol; Reduction thyroid gland weight in Wistar Hannover rats.	Antagonist of the transactivation of human androgen receptors	[24, 40]

UV-filters as thyroid disruptors

In vitro and in vivo data suggest that UV-filters may disrupt thyroid hormone (TH) synthesis, metabolism, and receptor function, providing further demonstration of their hazard for human health. Indeed, these compounds can reach the thyroid gland by systemic circulation principally mediated by skin absorption and by other environmental sources (such as seawater, foods and sediments). Once absorbed, they can interact with TH transport proteins, like transthyretin (TTR), and potentially alter circulating TH levels [41].

Here follows a description of the thyroid-disrupting effect of benzophenones (BP-1, BP-2, BP-3, BP-4), camphor derivatives (4-MBC, 3-BC), octinoxate and homosalate, the main studied and regulated organic UV-filters. A summary of observed results is provided in Tables 2 and 3, including also the concentration of the UV-filters used in each study.

Table 2.

UV-filters effect on the HPT axis

Class	Compound	Effect			References
		In vitro	Animal model	Human studies
Benzophenones	Benzophenone-1	↓ TSHβ, TRβ, ↓ TPO, ↑ NIS	↓ T3, T4	Interference with IGF system	[42–44]
	Benzophenone-2	Inhibition of TPO	↓ T4, T3; ↑ TSH	No data	[45, 46]
	Benzophenone-3	↓ TSHβ, TRHR, TRβ ↑ NIS, TG, ↓ TPO	↑ T3 and T4 in rats ↓ T3 and T4 in zebrafish; Interference with hormone synthesis	↓ serum T3 and FT4; ↓ T3 and FT4 in maternal urine; Accumulation in the organism	[42, 43, 47–51]
	Benzophenone-4	↑ NIS, TG genes; ↓ TRβ; TPO genes	Interference with development of the thyroid; ↑ HHEX and PAX8	Detected in placenta	[42, 52–54]
Camphor derivatives	4-methylbenzylidene camphor	—	Thyroid hormone pathway alteration ↓ iodine uptake, serum T4; ↑ TSH	No disturb in thyroid hormone homeostasis	[51, 55, 56]
	3- benzylidene camphor	—	No negative effect on thyroid system	Detected in placenta	[34, 36, 57]
Cinnamate derivatives	Octinoxate	Interference with HPT axis ↓ T3, T4, TSH	Interference with HPT axis; ↑ T4; ↓ thyroid hormone regulation genes in rainbow trout ↓ T3, T4, TSH in pregnant female rats	Detected in urine, plasma, breast milk and placenta	[1, 44, 51, 58–63]
Salicylic	Homosalate	Cytotoxicity and genotoxicity at high concentrations; upregulation of TG and TPO mRNA	↓ TSH, TSHβ, TRHR and TRβ ↑ TRΑα, TPO, NIS, and DIO2 imbalance alteration of thyroid weights and T4 levels in rats	Found at higher concentration in human samples (placenta, breast milk, plasma, urine) than in the environment	[64–68]

Table 3.

UV-filters doses tested in vitro, in animal studies and in humans (general population and concentrations measured in controlled exposure studies)

UV-filter	In vitro (cell lines)	Animal studies	Human biomonitoring (general population)	Human controlled exposure (max absorption)	Reference
BP-1	GH3: 1–32 µM FRTL-5: 10–320 µM	Zebrafish: 100–1000 µg/L	Urine: 0.47 ng/mL (pregnant women) Placenta: 0.4–0.5 µg/Kg.	—	[42–44]
BP-2	FTC-238: 300 nmol/L	Rats: 100–1000 mg/Kg	—	—	[45, 46]
BP-3 (Oxybenzone)	GH3: 3.2–100 µM FRTL-5: 10–320 µM	Rats: 33–1000 mg/Kg; Zebrafish: 32–320 µg/L; 0.04–1.40 µM	Urine: 2.57 ng/mL; 3.5-92000 ng/mL (median: 11.3 ng/mL) Women: plasma 187 ng/mL, urine 44 ng/mL Men: plasma 238 ng/mL, urine 81 ng/mL Cord blood: 56 ng/mL	Topical application of 2 mg/cm2 of basic cream (week 1) and with the three sunscreens at 10% (wt/wt) of each (week 2)	[42–44, 48, 50, 69]
BP-4	FRTL-5: 10–320 µM	Zebrafish: 30–3000 µg/L	Placenta: 0.25–5.41 ng/g	—	[42–44, 48, 50, 52, 53, 69]
4-MBC	MCF-7: EC50: 3.9 µM	Zebrafish:100 mg/Kg; 0.08–0.77 mg/L; Rats: 7–47 mg/Kg; 47, 24, 7.0, 0.7 mg/Kg/day per 10 weeks.	Women: plasma 16 ng/mL, urine 4 ng/mL Men: plasma 18 ng/mL, urine 4 ng/mL	Topical application: 2 mg/cm2 of basic cream in week 1 and with a mixture of the three sunscreens at 10% (w/w) each in week 2.	[36, 44, 55, 63, 70–73]
3-BC	MCF-7: EC50: 0.68 µM)	Frogs: 1.5–50 µg/L; Rats: 0.04 mL/10 g; 7.0, 2.4, 0.7, 0.24, 0.07 mg/Kg/day per 10 weeks.	—	—	[36, 57, 63]
Octinoxate (OMC)	—	Rats: 33–1000 mg/Kg; Rainbow trout: 6.9–395.6 µg/Kg; Zebrafish larvae: 0.3–30 µM	Women: plasma 7 ng/mL, urine 6 ng/mL Men: plasma 16 ng/mL, urine 4 ng/mL	Topical application of 2 mg/cm2 of basic cream (week 1) and mixture of three sunscreens at 10% (wt/wt) of each. (week 2)	[58, 60, 61, 69, 72, 74–76]
Homosalate (HMS)	FRTL-5, human thyrocytes: 250–2000 µM; MCF-7: 0.01-1 µM	Zebrafish: 0.03–30 µg/L; 0.026–262 µg/L; Rats: 1–5 mg/cm²	Plasma: 13.9–23.1 µg/L.	2 mg/cm2 to 75% of body surface area during 4 days of the study, with a total of 13 applications.	[64–67]

Benzophenones (BPs)

Benzophenones (BPs) are key components of countless personal care products, from skincare to haircare. Their great versatility made them a popular component in a wide variety of beauty products, including sunscreens, body lotions, shampoos, nail polish, lipsticks and hair sprays [77–79]. Recent studies raised concern regarding the endocrine-disrupting effects and environmental impacts of these substances, including the damage to coral reefs [80, 81]. Here follows a summary of the effects of BPs (BP-1, BP-2, BP-3, BP-4) on the TH axis.

Benzophenone-1 (BP-1)

Current regulations allow the presence of BP-1 in sunscreens in Japan and Australia only as an impurity. Other major jurisdictions have prohibited its use (Table 4).

Table 4.

Comparative regulatory status of selected UV-Filters

UV-FILTER	EU	USA	Canada	Japan	Australia
Benzophenone‑1 (BP‑1)	Not authorized (impurities tolerated)	Not approved	Not permitted	Permitted as impurity only	Permitted as impurity only
Benzophenone‑2 (BP‑2)	Not authorized	Not approved	Not permitted	Permitted as impurity only	Permitted as impurity only
Benzophenone‑3 (BP‑3/Oxybenzone)	Authorized: max 6% (2.2% children)	Approved: max 6%	Permitted: ~6%	Approved: ≤10%	Approved: max 10%
Benzophenone‑4 (BP‑4/Sulisobenzone)	Authorized: max 5%	Approved: max 10%	Permitted: max 10%	Approved: ≤10%	Approved: max 10%
4‑Methylbenzylidene Camphor (4‑MBC)	Prohibited	Not approved	Prohibited	Approved: max 4%	Approved: max 4%
3‑Benzylidene Camphor (3‑BC)	Not authorized	Not approved	Prohibited	Not approved	Not approved
Octinoxate (OMC)	Authorized: max 10%	Approved: max 7.5%	Permitted: max 10%	Approved: max 10%	Approved: max 10%
Homosalate	Authorized: max 0.5%	Approved: max 15%	Permitted: max 15%	Approved: max 10%	Approved: max 10%
REFERENCES	[76, 82]	[83]	[84]	[85]	[86]

In vitro and animal model studies

Lee et al. investigated the endocrine-disrupting potential of BPs, particularly BP-1’s, and their impact on TH regulation. GH3 (a pituitary cell line) and FRTL-5 (Fisher Rat Thyroid Low-serum; a rat thyroid cell line) were used for the in vitro evaluation, while the zebrafish model (Danio rerio) was employed for an in vivo study.

BP-1 interfered with TH signalling in both GH3 and FRTL-5 cell lines. In GH3 cells, exposure to BP-1, in a dose dependent manner, led to decreased expression of the beta subunit of thyroid stimulating hormone gene (TSHβ) and also of the thyroid receptor beta gene (TRβ) at doses of 1–32 µM and 3.2–32 µM, respectively. In FRTL-5 cells, 320 µM BP-1 exposure caused down-regulation of the TPO gene and an increase of the NIS gene. In the embryo-larval zebrafish model, exposure to BP-1 induced a significantly decrease of thyroid hormones production (T3 and T4) in a dose-dependent manner, starting from a concentration of 320 and 1000 µg/L, respectively). These results highlight the potential of BP-1 to disrupt thyroid function during early development, reinforcing the notion that it should be classified as a thyroid disruptor [42].

Epidemiological and human studies

Krause et al. evaluated, in 183 pregnant women in the 2nd trimester, the effects of maternal exposure to BP-1, in terms of changes in thyroid function parameters, insulin-like growth factor (IGF) system, and birth outcomes. The authors showed that, despite BP-1 being detected in a significant percentage (70%) of urine samples from pregnant women (with a mean concentration of 0.47 ng/mL), no significant correlation was observed between BP-1 concentrations and alterations in maternal free thyroxine (FT4) and TSH [43].

Benzophenone-2 (BP-2)

Current regulations allow the presence of BP-2 in sunscreens in Japan and Australia only as an impurity. Other major jurisdictions have prohibited its use (Table 4).

In vitro and animal model studies

Schmutzler et al. demonstrated an impact of BP-2 on thyroid function both in vitro and in vivo. In vitro assays on FTC-238 (Follicular Thyroid Carcinoma − 238) lines showed that BP-2 at a concentration of 300 nmol/L is a strong inhibitor of TPO (thyroid peroxidase), interfering with TH production [45]. In addition, in vivo experiments on rats exposed to BP-2 (at 333 mg/Kg and1000 mg/Kg concentrations) revealed significant alterations in TH homeostasis. In particular, exposure to BP-2 caused a decreasein serum T4 levels in a dose-dependent manner. An increase in TSH levels was observed after exposure to both BP-2 concentrations, with similar effects for the two concentrations [45]. On the other hand, TPO activity in the thyroid of treated animals was unchanged. As a last finding, BP-2 did not inhibit iodide uptake by FRTL-5 cells [45].

Broniowska et al. demonstrated that exposure to BP-2 (at a concentration of 100 mg/Kg) through the skin can have a substantial impact on HPT (hypothalamic-pituitary-thyroid) axis function of Wistar rats, resulting in reduced serum levels of T4 and T3, accompanied by high levels of TSH [46].

Epidemiological and human studies

No data from epidemiological and human studies regarding BP-2 potential thyroid-disrupting effect are available in the literature.

Benzophenone-3 (BP-3, Oxybenzone)

Major regulations allow the use of BP-3 in sunscreens but different concentrations are permitted across countries (Table 4).

In vitro and animal model studies

Pregnant female rats treated with BP-3 showed increased levels of T3 and T4, and reduced TSH, indicating thyreotoxicosis [47]. BP-3 also interfered with thyroid function through an agonistic effect on the thyroid receptor [87] and decreased expression of the ERR1 gene in the thyroid of ovariectomized rats [88].

A study by Lee et al. showed that exposure to BP-3 (at a concentration of 32 µM), led to a decreased expression of thyroid-stimulating hormone beta subunit (TSHβ), thyroid hormone receptor (TRHR), and thyroid hormone receptor beta (TRβ) genes in GH3 cells. Furthermore, in FRTL-5 cells, BP-3 (at a concentration of 320 µM) significantly increased NIS (sodium iodide symporter) and TG (thyroglobulin) expression, while reducing TPO expression. In addition, exposure of zebrafish embryos to BP-3 (at concentrations of 32, 100, 320 µg/L) resulted in a significant decrease in T3 levels six days after fertilisation, in a dose-dependent manner. These findings suggest a possible disruption of HPT axis due to an impact on both central regulation and peripheral hormone metabolism [42]. Reum Kwon et al. showed that BP-3 disrupts in dose dependent manner (0.04, 0.14, 0.44, 1.40 µM concentrations) the thyroid function in zebrafish. These effects included a decrease in circulating TH levels, which was attributed to their enhanced metabolism and excretion. Furthermore, the authors suggested that BP-3 may interfere with the synthesis, transport, or signalling of thyroid hormones due to changes in transcription of several thyroid-related genes. Such disruptions in thyroid homeostasis during early development induced alterations ofphysiological processes such as growth and metabolism [48].

Epidemiological and human studies

Aker et al. [49] showed that increased maternal urinary BP-3 resulted in a 3% decrease in T3 without changes in circulating TSH and T4 levels in pregnant women. Kim et al, in the 2007–2008 National Health and Nutrition Examination Survey (NHANES) from the US general population, showed BP-3 was significantly and negatively correlated with serum total T3 and FT4 levels, indicating thyroid-disrupting effects. The authors also report gender-specific differences, with stronger associations found in males. The above evidence raised concerns since the exposure to BP-3 could contribute to disruptions in thyroid hormone homeostasis within the general population [50].

Indeed, Krause et al. observed a correlation between exposure to BP-3 and fluctuations in thyroid hormone levels during pregnancy. BP-3 (2.12–3.60 ng/mL, average 2.57 ng/mL) has been detected in a significant proportion of urine samples from pregnant women, suggesting widespread exposure. Elevated BP-3 concentrations have been associated with diminished FT4 levels, indicating a possible disruption of thyroid hormone regulation. Nonetheless, a definitive association with TSH has not been established, implying a complex and potentially indirect effect on thyroid function [43].

Janjua et al. observed that BP-3 is rapidly absorbed into the bloodstream. Following the application of sunscreen containing 10% BP-3 (2 mg/cm²), BP-3 was detectable in plasma within a time frame of 1–2 h (women: 187 ng/mL; men: 238 ng/mL) and in cord blood (56 ng/mL) [89]. In addition, BP-3 was also detected in urine samples (women: 44 ng/mL; men: 81 ng/mL) (Table 3). It is noteworthy that BP-3 persisted in the plasma even 96 h after the final application, suggesting the potential for the compound to accumulate in the body with repeated use [58, 69].

Benzophenone-4 (BP-4)

BP-4 use in sunscreens is permitted in all major jurisditiations but at different final concentrations (Table 4).

In vitro and animal model studies

Lee et al., reported that BP-4 (10–320 µM) exerts no effect on the regulation of TSHβ and TRHR genes in GH3 cells but induced the downregulation of the TRβ gene. In FRTL-5 cells, BP-4 (10–320 µM) was found to significantly upregulate NIS and TG genes, while concomitantly downregulating the TPO gene [42].

Zucchi et al. demonstrated that exposure of zebrafish to BP-4 (3000 µg/L and 3 µg/L, respectively) upregulated HHEX and PAX8, two pivotal genes implicated in the development of the thyroid gland. Their increased expression suggests the possibility of effects on thyroid gland formation during the early stages of development [52].

Epidemiological and human studies

The effects of BP-4 exposure have not been specifically investigated in human and epidemiological studies, but BP-4 has been detected in human placenta with concentrations ranging between 0.25 ng/g and 5.41 ng/g. However, its role as a potential thyroid disruptor is still not known [53].

Overall summary

Overall, the above results identified BPs as potential thyroid disruptors. The main mechanisms sustaining their interference with HPT axis include inhibition of TPO activity, accelerated TH metabolism and alterations in the expression of several thyroid-related genes [4].

4-Methylbenzylidene Camphor (4-MBC)

4-methylbenzylidene camphor (4-MBC), is a camphor derivative used as UV-B filter in sunscreen and cosmetic products [90]. Significant concentrations of 4-MBC were found in several human tissues, including the placenta [34]. The European Commission, amending its Cosmetics Regulation (EC) No 1223/2009, has prohibited the use of 4-MCB in cosmetic products, from 1 st May 2025. Starting from 1 st May 2026 cosmetic products containing this substance will not be allowed in the EU [91]. Only Japan and Australia approve the use of 4-MBC, even if at low concentration (Table 4).

In vitro and animal model studies

No in vitro data are available regarding the effects of 4-MBC on thyroid cells. However, in zebrafish embryos, exposure to 4-MBC (100 mg/Kg) resulted in alterations of the expression of several genes implicated in TH synthesis [70]. Another study showed that the expression of genes related to TH synthesis and metabolism was altered by 4-MBC (at a concentration of 0.165 mg/L), indicating a possible disruption of normal thyroid regulation during early development [55]. Schmutzler and Brent demonstrated that adult ovariectomized rats exposed to 4-MBC exhibited decreased iodine uptake, increased TSH levels, and decreased serum T4 levels [56, 87]. Furthermore, Maerkel et al. found that this compound led to an increase in both the relative and absolute thyroid weights in adult mice, when both the mice and their parent generation were treated with 4-MBC at increasing concentrations (7, 24, 47 mg/Kg) [71, 92] (Table 3).

Epidemiological and human studies

Janjua et al. demonstrated a weak correlation between 4-MBC levels in plasma and urine of human subjects (females: plasma 16 ng/mL, urine 4 ng/mL; males: plasma 18 ng/mL, urine 4 ng/mL) and T4, T3 and thyroxine-binding globulin (TBG) levels. Even if a significant correlation could be observed, the authors concluded that this filter did not significantly disturb human thyroid homeostasis [72] (Table 3).

Overall summary

Currently available data indicate that 4-MBC poses potential risks to both the environment and human health, mainly due to its persistence, bioaccumulation and endocrine disrupting properties. However, comprehensive epidemiological studies in human populations are currently lacking, making it difficult to fully assess the public health implications of the exposure to 4-MBC. Further research is needed to elucidate 4-MBC effects and to inform regulatory decisions regarding its use in consumer products [70].

3-benzylidene camphor (3-BC)

3-benzylidene camphor (3-BC) is a derivative of camphor. The use of this compound is not approved by any major jurisditiation [93] (Table 4).

In vitro and animal studies

Very few data are available on the thyroid disrupting effects of 3-BC. In particular, the study by Kunz et al. showed that the exposure of tadpoles of the frog Xenopus laevis at environmental concentrations of 3-BC of 1.5 and 50 µg/L does not negatively affect the thyroid system during frog metamorphosis [57].

The study by Schlumpf et al. showed that 3-BC and 4-MBC stimulated MCF-7 cell proliferation (EC50: 0.68 and 3.9 µM), but no thyroid disturbing-effects were observed in rats exposed to 3-BC (0.04 mL/10 g) during pregnancy [36] (Table 3).

Epidemiological studies

Although this compound was found in human placenta (1.7–150 ng/g) [34], no epidemiological studies are available regarding 3-BC effects on thyroid system in humans.

Overall summary

Animal and in vitro studies report no thyroid effects, despite some estrogenic activity, and although 3-BC has been detected in human placenta, current evidence does not support thyroid disruption and epidemiological data in humans are lacking.

Octinoxate

Octinoxate, also known as octyl methoxycinnamate (OMC), is a UV-B filter found in hair products, lipsticks, makeup, and skin care items. It is also present in pools, tap and drinking water, leading to increased human exposure. Banned in Hawaii for its potentially dangerous effects on marine ecosystems, OMC is suspected to exert estrogenic activities [94]. On the other hand, the use of octinoxate is anyway permitted, even if at different concentration, by most jurisditiations (Table 4).

In vitro and animal model studies

Regarding thyroid disruption, in vitro and animal studies revealed that OMC interferes with the HPT axis at different levels, but results are controversial [1].

Several studies indicate that OMC reduces the levels of T3, T4 and TSH hormone levels in animals.

Klammer et al. reported a decrease in TSH concentrations, suggesting that the simultaneous dose-dependent reduction of TSH, T4, and T3 levels in female rats could result from a direct action of OMC (33, 100, 333, 1000 mg/Kg) on pituitary cells, leading to reduced TSH release [58]. Similarly, exposure to OMC (0.3, 1, 3, 10, 30 µM) in zebrafish larvae reduced T3 and T4 levels in a dose-response manner, and downregulated TH-related genes [59]. In rats, Axelstad et al., found that T4 levels of maternal and male offspring were reduced in dose-dependent manner after OMC exposure (500, 750, 1000 mg/Kg) [60]. On the other hand, Cahová et al., observed an increase in T4 levels and the downregulation of TH regulatory genes in rainbow trout exposed to OMC (6.9, 96.0, 395.6 µg/Kg), in a dose-dependent manner, contradicting previous reports of TH reduction [61]. Despite these discrepancies, there is a general consensus that OMC can interfere with thyroid function through multiple mechanisms. In vitro, OMC reduced iodine uptake in thyroid follicular cells but did not affect TPO activity [95]. OMC had also an agonistic activity on thyroid receptors (TR) in human hepatocarcinoma cells. Additionally, exposure to OMC a together with its transformation products, as well as mixtures with other endocrine-disrupting chemicals (EDCs), could enhance its endocrine-disrupting effects.

Epidemiological and human studies

OMC has been detected in human urine, plasma [51], breast milk samples [62, 63] and placenta [44], raising concerns about its possible impact on fetal development [58, 96].

Limited data exists on the effects of OMC on human physiology [94]. In a study by Janjua et al., 32 participants applied a cream with 10% of OMC (also containing BP-3 and 4-MBC) for one week. While OMC could be detected in plasma and urine samples (females: plasma 7 ng/mL, urine 6 ng/mL; males: plasma 16 ng/mL, urine 4 ng/mL) no significant changes in T3, T4 or TBG could be observed. Reproductive hormone levels were also unaffected, suggesting that short-term topical application of OMC does not alter hormone regulation [58, 69] (Table 3).

Overall summary

Many in vitro studies and studies on animal models support the hypothesis that OMC disrupts thyroid function by reducing TSH, T4, and T3 and iodine uptake. However, human studies reported no significant interference of this compound on thyroid homeostasis. The variability in findings, especially across different species and experimental conditions, highlights the complexity of this issue. Further research is needed to clarify its mechanisms and potential risk for humans.

Homosalate

Homosalate (HMS) is a UV-B filter used in sunscreens and personal care products. It has been shown to have estrogenic and anti-androgenic effects in animal models (REF), but there is limited data on its effects on thyroid cells. The use of homosalate is employed in major jurisdictions but with varying concentrations (Table 4).

In vitro and animal studies

Coperchini et al., evaluated the in vitro effects of homosalate exposure (250, 500, 750, 1000, 2000 µM) on FRTL-5 and human thyroid cells. The results of this study showed that homosalate induced: (i) cytotoxicity at high concentrations and/or after long time of incubation; (ii) genotoxic effects exclusively in rat thyroid cells, but only at the highest concentration tested; (iii) increased TG mRNA in thyroid cells and TPO mRNA in FRTL-5 cells; (iv) no changes in cell proliferation or oxidative stress [64].

Ka et al., reported that homosalate (30 µg/L) exposure increased T3 and T4 levels, in zebrafish. In the same study, a significant downregulation of TSH, TSHβ and TSHR genes was observed, suggesting that homosalate may affect TSH directly or indirectly through a feedback mechanism of increased T4 secretion [65]. On the other hand, exposure to homosalate upregulated the expression of TRΑα, TPO, NIS, and DIO2 genes [65].

The study by Erol et al., investigated the potential association between dermal exposure to homosalate (1–5 mg/cm²) and pubertal development and thyroid function in both male and female rats. The results showed intriguing insights into growth and hormonal regulation during the prenatal, lactation, and early infancy periods. In the prenatal period, T4 levels increased in the exposed female rats, while in males the thyroid weights and T4 levels were unchanged. During lactation, female rats showed increased thyroid gland weights, while male had unchanged thyroid weights and T4 levels. During infancy, the weight of the thyroid gland increased in female rats but the T4 levels, were not affected, while in the male, the T4 levels were increased, without affecting thyroid weight [66].

Epidemiological and human studies

Homosalate is frequently detected in human samples, including placental tissue, breast milk, plasma and urine. Notably, homosalate levels in human plasma (13.9–23.1 µg/L) [67] exceed those found in the environment, indicating a potential accumulation in the human body following chronic low-dose exposure [68]. However, at present no studies were specifically designed to evaluate its potential thyroid-disrupting effect on humans.

Overall summary

Data on potential thyroid-disruting effects of Homosalate are limited to only one study in vitro and few data on animal models, which however suggest a potential effect on pivotal genes involved in thyroid homeostasis. The lack of epidemiological studies suggests the need to encourage further studies.

Regulations of sunscreens

Sunscreens made their debut in 1928, marking the beginning of a journey towards sun protection that would evolve significantly over the decades. By the 1940 s, the Food and Drug Administration (FDA) stepped in to ensure that these vital compounds were regulated, paving the way for safer and more effective sun care products. Current regulations on sunscreen regarding UV-filters differ from one country to another (Fig. 1). Indeed, EU countries and Australia, Canada, and Korea, consider UV-filters as cosmetic or therapeutics; while, other countries, such as the US, consider these products as non-prescription therapeutic medicines (over-the-counter (OTC) products) [23, 97, 98]. Moreover, in Asia regulatory processes have a different approach compared to Western standards: for example, Japan categorises sunscreens as quasi-drugs, which are subject to specific ingredient and labelling guidelines under the Japanese Cosmetic Standards. South Korea, similarly, regulates sunscreens under its functional cosmetics category, with a focus on both safety and efficacy [99, 100]. Currently, EU Regulation CE/1223/09 replaces the previous system based on Directive 76/768, ensuring that all Member States have the same text and legal framework based on both positive and negative ingredient lists. The European system is continually updated as the technical progress evolves, based on opinions from the Scientific Committee on Consumer Safety (SCCS) [101]. In 2019, the FDA reclassified 12 organic UV-filters [102] including the most commonly used ones, namely octosalate, homosalate, octocrylene, benzophenone-3 (BP-3), octinoxate, and avobenzone, highlighting the need for additional safety data (FDA, 2019) [103]. Due to inadequate evidence, in 2021, the US FDA reclassified a number of UV-filters as “not generally recognized as safe and effective” (non-GRASE). This designation was applied to filters like BP-3 and avobenzone, which had been components of sunscreens for decades.

Fig. 1.

Schematic worldwide distribution of sunscreen regulation. Analysis of the data in according with the literature. The order considers the different stringency of regulation in the world: purple countries banned few specific UV-filters; red countries defined sunscreens as quasi-drugs; yellow countries defined the difference between cosmetic and therapeutics; blue countries established a simple classification as OTC product

In order to keep these substances on the market, at present, manufacturers must provide more safety information. In Europe, standard quality requirements are equally elevated. The allowed UV-filters are regulated by Annex VI of the Cosmetics Regulation (EC N. 1223/2009), and the parameters established for labelling are: Sun Protection Factor (SPF) with a maximum value of 50 + and in vitro UV-A protection factor (UV-APF) with a value of at least one-third of the UV-B protection [104–106]. In the US, only 16 UV-filters have been approved by the FDA in the sunscreen monograph. In order to avoid environmental contamination, in 2017, the EU reduced the maximum concentration of BP-3 allowed in PCPs (personal care products) from 10 to 6% [107]. More significantly, in January 2020, the Republic of Palau was the first country to ban the use of BP-3, octinoxate, octocrylene, and 4-methylbenzylidene camphor (4-MBC) in PCPs [108]. The island of Aruba quickly followed suit and banned the use, distribution, and import of BP-3 in July 2020 [109]. A number of similar bills banning UV-filter use in PCPs have been passed in Hawaii, Bonaire, and in some areas of Mexico [74, 75]. Recently, an amendment to the European Cosmetics Directive limited homosalate to facial products only, with a maximum concentration of 7.34%. Its use will be restricted to non-spray face products. This restriction started from January 1, 2025 [80]. Regulatory systems for UV-filters differ from country to country (Fig. 1) and each molecule has a maximum allowed concentration. Moreover, to provide an overview of the heterogeneous global regulatory status of these commonly used organic UV-filters, a comparative table was compiled summarizing their approval status, concentration limits, and regulatory references across five major jurisdictions (EU, USA, Canada, Japan, and Australia) [76, 82–86] (Table 4). Overall, although some UV filters have already been classified as EDs, clarifying their thyroid-disrupting potential is essential for a comprehensive assessment of chemical risks and to provide useful information for regulatory decisions aimed at protecting vulnerable populations.

To address this need, several international agencies have developed dedicated guidelines and test methods. The Organisation for Economic Co-operation and Development (OECD) has established a battery of in vitro and in vivo assays, designed to detect alterations in TH synthesis, metabolism and signaling pathways [110, 111]. The European Food Safety Authority (EFSA) and the European Chemicals Agency (ECHA) have established a weight-of-evidence methodology to evaluate the effects of EDs [112] including changes in serum levels of T3, T4, and TSH, thyroid histopathology, and neurodevelopmental fetal outcomes [113].

Conclusion and outlook

UV filters are essential for protecting human health by shielding the skin from harmful UVR, the main risk factor for photoaging and skin cancer. The growing demand for cosmetic products has driven an increase in their production and use. However, recent studies have raised concerns due to the bioaccumulative properties of UV filters, particularly in environmental matrices such as water, sediment, and marine organisms, pointing to their persistence and possible long-term effects. The detection of several UV filters in human matrices (Table 3) further indicates systemic absorption and bioaccumulation, highlighting the risk of chronic low-dose exposure given their widespread use in personal care products [27, 114].

Studies on the ED properties of UV-filters have traditionally focused on their interaction with sex hormones, demonstrating their potential side effects on reproduction, development and metabolic processes in humans and animal models [115–117]. More recently the attention is being directed toward UV-filters capacity to act as thyroid disruptors. In vitro and in vivo studies have shown some degree of interference with HPT axis after exposure to almost all UV-filters considered in the present review (Tables 2 and 3). While the evidence from basic studies is compelling, human epidemiological data remain limited and often inconclusive. The widespread detection of UV filters in human tissues, breast milk and fetal circulation, underscores concerns about long-term exposure during sensitive developmental windows, with vulnerable populations such as children, pregnant women, and the elderly being at particular risk.

Current regulations on the use of major UV filters remain highly heterogeneous worldwide, with significant differences not only between continents but even among countries within the same continent (Fig. 1; Table 4). Although some of the most hazardous compounds have been banned and others are regulated, the lack of global harmonization creates regulatory uncertainty. In many countries, UV filters are still not subject to specific safety assessments or legislative controls. Recent guidance from high-level authorities (OECD, ECHA, EPA) emphasizes the need to evaluate chemicals not only for general endocrine-disrupting properties but also specifically for their ability to interfere with thyroid hormone signaling. In this context, evidence of thyroid-disrupting activity may warrant a re-evaluation of their regulatory status. This could include stricter limits on permitted concentrations, additional usage restrictions, inclusion of a warning label on each formulation or, in some cases, complete removal from consumer products. A thorough understanding of the thyroid-disrupting potential of UV filters is therefore essential to ensure a science-based regulation and to determine the minimal effective dosages that balance photoprotection with health safety. While further research is needed to establish definitive causal links between UV-filters and thyroid disruption in humans, current evidences strongly support the hypothesis that several UV-filters can act as thyroid disruptors. The alarming effects include alterations in TH levels, interference with TH receptors and potential neurodevelopmental impairments. Future research should concentrate on chronical exposure, low-dose in vivo studies and comprehensive epidemiological data involving vulnerable populations. Additionally, mechanistic studies need to identify the specific pathways through which thyroid disruption occurs. It’s essential to implement standardised testing for thyroid effects in regulatory assessments to ensure adequate protection of public health.

Beyond endocrine-disrupting activity, UV filters may also affect health through other mechanisms. For example, some compounds, while providing strong UV protection, can paradoxically become more toxic after sunlight exposure [118].

Despite these concerns, there is no doubt that sunscreens remain effective tools against UV-related skin pathologies. To address safety issues, the cosmetic industry is adopting different strategies, including:

• i)The introduction of updated product labels with clear scientific information on potential exposure-related risks;
• ii)The use of technological alternatives such as nanotechnology-based delivery systems (polymeric nanoparticles, lipid nanoparticles, liposomes, nanogels, inclusion complexes) [119, 120] which offer benefits such as minimal skin absorption, photostability, reduced irritability, and improved UV filtering [121, 122].
• iii)The incorporation of enzymes encapsulated in liposomes (e.g., T4 endonuclease V, UV endonuclease, photolyase) derived from UV-resistant microorganisms, which can penetrate sun-damaged skin and support DNA repair as a complementary photoprotective strategy [123–127].
• iv)The use of alternative more reliable tests to assess product safety. For example, the current SPF test methods has been improved with the use of chromametry to evaluate erythema intensity, in order to reduce sources of variability (FDA 2011, ISO 24444:210). Nevertheless, this method still has some limitations in estimating erythema.
• v)The implementation of awareness campaigns in order to inform the population regarding the harm caused by chronic exposure to UV filters, paying special attention to vulnerable groups.

The future sunscreen should meet the individual needs of consumers regarding age, skin, physiology, sun exposure and disease risks, with the development of novel personalized photoprotection in the near future. In conclusion, given the widespread use of UV-filters, regulatory measures, information campaigns and further toxicological evaluations are crucial to better assess their safety and protect public health.

Abbreviations

3-BC

3-Benzylidene camphor

4-MBC

4-Methylbenzylidene camphor

AR

Androgen receptor

BP

Benzophenone

ED

Endocrine disruptor

EDCs

Endocrine-disrupting chemicals

ER

Estrogen receptor

ECHA

European Chemicals Agency

EFSA

European Food Safety Authority

FDA

Food and Drug Administration

FTC-238

Follicular Thyroid Carcinoma − 238

FRTL-5

Fisher Rat Thyroid Low-serum

FT4

Free thyroxine

HMS

Homosalate

HPT

Hypothalamic-pituitary-thyroid

IGF

Insulin-like growth factor

OECD

Organisation for Economic Co-operation and Development

OMC

Octinoxate

OTC

Over-the-counter

PCPs

Personal care products

PFAS

Perfluoroalkyl substances

PCBs

Polychlorinated biphenyls

PR

Progesterone receptor

SCCS

Scientific Committee on Consumer Safety

SPF

Sun Protection Factor

T3

Triiodothyronine

T4

Thyroxine

TBG

Thyroxine-binding globulin

TG

Thyroglobulin

TH

Thyroid hormones

TPO

Thyroid peroxidase

TRβ

Thyroid hormone receptor beta

TRHR

Thyroid hormone receptor

TSH

Thyroid-stimulating hormone

TSHβ

Thyroid-stimulating hormone beta

TTR

Transthyretin

UV

Ultraviolet

UVR

Ultraviolet radiation

Authors’ contributions

A.G, E.F, M.D. AND S.A conceptualized and wrote the original draft of the manuscript, F.C., F.M. prepared figures and tables and revised all the text, M.R. revised original draft, tables and figures.

Funding

No funding.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Consent for publication

All authors agreed with the content and that all gave explicit consent to submit.

Competing interests

The authors declare no competing interests.