Benzoyl Peroxide Drug Products Form Benzene

Introduction

Benzoyl peroxide (BPO) is a diacyl peroxide with bactericidal activity used in topical drug products up to 10% concentration and is available through prescription or over-the-counter (OTC) for treatment of acne vulgaris (acne).¹ Acne drug treatments commonly include topical creams, gels, or other drug product formulations that are left on the skin daily. Wash-away cleansers are also available, as are microencapsulated formulations that aim to stabilize benzoyl peroxide.² Most products are applied daily and consistent use ranges from months to years where patients are chronically exposed to drug products with BPO.

BPO, produced as a crystalline powder, is known to thermally decompose to form two molecules of benzoyloxy radicals that can further decompose to benzoic acid or phenyl radicals with liberation of carbon dioxide. Depending on the chemical environment, the phenyl radicals produce end products that include benzene, phenyl benzoate, and biphenyl.³ A degradation mechanism from BPO $({\left({\mathrm{C}}_6{\mathrm{H}}_5\mathrm{CO}\right)}_2{\mathrm{O}}_2)$ to benzene $\left({\mathrm{C}}_6{\mathrm{H}}_6\right)$ is shown below:

$$\begin{array}{l}{\left({\mathrm{C}}_6{\mathrm{H}}_5\mathrm{CO}\right)}_2{\mathrm{O}}_2\rightleftharpoons 2{\mathrm{C}}_6{\mathrm{H}}_5{\mathrm{CO}}_2•\\ {\mathrm{C}}_6{\mathrm{H}}_5{\mathrm{CO}}_2•\to {\mathrm{C}}_6{\mathrm{H}}_5•+{\mathrm{CO}}_2\\ {\mathrm{C}}_6{\mathrm{H}}_5•+{\mathrm{H}}^{+}\to {\mathrm{C}}_6{\mathrm{H}}_6\end{array}$$

There is no safe level of benzene exposure to humans and US Food and Drug Administration (FDA) guidance specifies that drug products should not contain benzene because of its unacceptable toxicity and benzene is permissible up to 2 ppm only for products where benzene is required for manufacturing.⁴ Since 2021, independent studies identifying benzene contamination have led to global recalls of consumer drug products, including hand sanitizers, sunscreens, antifungal sprays, and antiperspirants.⁵

To investigate both benzene concentrations produced within the topical drug products and gaseous benzene released into the surrounding air, orthogonal analytical techniques of gas chromatography–mass spectrometry (GC-MS) and selected-ion flow-tube MS (SIFT-MS) were used. Incubation temperature of $37°\mathrm{C}$ ($98.6°\mathrm{F}$) was used to evaluate effect of standard body temperature, $50°\mathrm{C}$ ($122°\mathrm{F}$) was used to evaluate shelf-life performance as an accelerated stability testing temperature for pharmaceuticals,⁶ and $70°\mathrm{C}$ ($158°\mathrm{F}$) was used to evaluate transportation/passenger vehicle excursion temperature.⁷

Materials and Methods

Single lots of seven differently formulated and branded BPO products were selected based on availability from national retailers. The products used are listed with their packaged volume and unique product codes (UPCs) for a) GC-MS studies: Equate Beauty 2.5% BPO cleanser 4 oz ($118\text{\hspace{0.17em}mL}$) UPC 681131200066, Neutrogena 10% BPO cleanser 5 oz ($148\text{\hspace{0.17em}mL}$) UPC 070501024638, CVS Health 10% BPO face wash 6.6 oz ($195\text{\hspace{0.17em}mL}$) UPC 050428326930, Walgreens 10% BPO cream 1.12 oz ($33\text{\hspace{0.17em}mL}$) UPC 049022657021, Clean & Clear 10% BPO cleanser 5 oz ($148\text{\hspace{0.17em}mL}$) UPC 38137-0032892; and b) SIFT-MS: Equate Beauty 10% BPO acne wash 5.5 oz ($163\text{\hspace{0.17em}mL}$) UPC 681131436496, and Proactiv 2.5% BPO cleanser 4 oz ($118\text{\hspace{0.17em}mL}$) UPC 842944102170.

GC-MS

GC-MS methodology in this investigation followed that previously used by Valisure’s laboratory for the analysis of benzene in consumer products, including the use of solvent blanks and isotopically labeled benzene in representative samples of consumer products.⁸ ${\text{Benzene-}}^{13}{\mathrm{C}}_6$ (99.8% Sigma-Aldrich) was used for BPO studies. Dimethyl sulfoxide (DMSO; GC headspace grade, Thermo Fisher Scientific) was used for sample preparation and was separately tested with up to 10% BPO (98% Sigma-Aldrich) or benzoic acid (99.5% Sigma-Aldrich), and no benzene was observed following incubation up to 10 d at $37°\mathrm{C}$. The lower limit of detection (LLOD) was $10\text{\hspace{0.17em}ng}$ (equivalent to $0.02\text{\hspace{0.17em}ppm}$ in products), and the lower limit of quantitation (LLOQ) was $25\text{\hspace{0.17em}ng}$. The measurement uncertainty was determined to be 20%.

Stability testing in a laboratory oven (Precision Freas Mechanical Convection Oven 605, Thermo Electron Corporation) was performed on five BPO products at $37°\mathrm{C}$ for 4 wk and at $50°\mathrm{C}$ for 3 wk.

Selected-Ion Flow-Tube MS

A Syft Technologies Voice200ultra mass spectrometer equipped with a high-performance inlet was used for measurement of benzene in real time from an environmental chamber.⁸ Two common BPO products were separately incubated at $70°\mathrm{C}$ for 16.7 h using a $2\text{-L}$ DuoBath (Benchmark Scientific) that was placed in the chamber and filled with metallic Lab Armor Beads. The BPO products were removed from their external carboard packaging, if present, and caps on the primary BPO product containers were not removed.

The two product ions, ${\mathrm{C}}_6{\mathrm{H}}_6·{\mathrm{NO}}^{+}$ and ${\mathrm{C}}_6{\mathrm{H}}_6^{+}$, with the respective masses of 108 and 78, were used to measure the concentration of benzene before and during the incubation of BPO products in the environmental chamber with 3-s measurement cycles.

System sensitivity was determined using two separate reference gasses (Don Wolf & Associates and GASCO) where benzene was contained at a known concentration of $5\text{\hspace{0.17em}ppm}$. The gasses were injected at known volumes into the chamber generating a 7-point curve with $R^2$ values $>0.995$ and absolute concentrations within the expected variability of the reference standards. The LLOD and the LLOQ of benzene were calculated from the two separate reference gas calibration curves and resulted in Don Wolf & Associates LLOD of 1.55 ppb and LLOQ of $4.70\text{\hspace{0.17em}ppb}$ and GASCO LLOD of $1.64\text{\hspace{0.17em}ppb}$ and LLOG of $4.97\text{\hspace{0.17em}ppb}$.

Results

Screening for the presence of benzene via GC-MS analysis was performed on a representative selection of BPO acne treatment products. Product formulation matrix effects were accounted for using carbon-13 benzene, which has the same characteristics of benzene in recovery tests but can be differentiated by mass. All BPO products could form benzene, whereas no such evidence was observed in acne treatment products without BPO (detection below LLOD). The results of stability testing at $37°\mathrm{C}$ and $50°\mathrm{C}$ are shown in Figure 1. Testing products incubated at $70°\mathrm{C}$ was discontinued due to package ruptures in the first week of incubation.

Figure 1.

Five common benzoyl peroxide (BPO) drug products incubated at (A) $37°\mathrm{C}$ for 28 d and at (B) $50°\mathrm{C}$ for 21 d and periodically sampled for benzene analysis. Results are displayed in parts per million. Depending on the pattern of benzene formation, either an exponential (Expon) or linear fit was applied.

Two common OTC BPO products were separately incubated at $70°\mathrm{C}$ for 16.7 h and monitored in real time using SIFT-MS for the presence of benzene in the surrounding environment. No visible product container ruptures were observed in these tests. Results are shown in Figure 2.

Figure 2.

Benzene levels detected in real-time by selected-ion flow-tube mass spectrometry (SIFT-MS) during $70°\mathrm{C}$ incubation of two benzoyl peroxide (BPO) drug products. Benzene concentration is displayed in parts per billion and as a moving average of 1-min detection intervals. Both products were incubated without modification from primary packaging; notably caps were not opened.

Discussion

This study raises substantial concerns about the safety of BPO products currently on the market, which appear to form benzene, a known human carcinogen and environmental hazard. Benzene was detected in all BPO product samples tested and levels increased during incubation at body and shelf-life performance temperatures to $>2\text{\hspace{0.17em}ppm}$, which is the conditionally restricted FDA limit for benzene for drug products.³ Furthermore, BPO was detected in the environment surrounding two unopened primary product containers, which suggests that BPO products could emit substantial amounts of benzene from their original packaging prior to use. The US Environmental Protection Agency (EPA) has determined that a lifetime exposure to $0.4\text{\hspace{0.17em}ppb}$ of benzene in air can lead to one additional cancer case in 100,000 exposed persons.⁹ If the benzene concentration detected in this study in a $528\text{-L}$ environmental chamber were formed in the $\sim \mathrm{3,000}\mathrm{L}$ of air in a hot passenger vehicle, benzene concentration could exceed $400\text{\hspace{0.17em}ppb}$, or over 1,000 times this US EPA limit.

Solid BPO at purities up to 75% is known to be thermally stable to $98°\mathrm{C}$¹⁰ and various pharmaceutical additives are known to destabilize BPO.³ Therefore, reformulation with attention to benzene formation, could lead to BPO drug products that will not form benzene over time. Further investigation of BPO stability should be undertaken and regulatory actions should be considered in the interest of protecting public health.

Conclusions and opinions are those of the individual authors and do not necessarily reflect the policies or views of EHP Publishing or the National Institute of Environmental Health Sciences.