Simultaneous Determination of 80 Unapproved Compounds using HPLC and LC-MS/MS in Dietary Supplements

Abstract

We developed analytical methods using high performance chromatography (HPLC) and liquid chromatography tandem mass spectrometry (LC-MS/MS) for the simultaneous determination of 80 unapproved compounds in dietary supplements. The target compounds for analysis were unapproved ingredients (e.g., pharmaceuticals) that have potential adverse effects on consumers owing to accidental misuse, overuse, and interaction with other medication in dietary supplement. Two analytical methods were tested to identify the optimal validation results according to AOAC guideline. As a result, limit of quantification (LOQ) was 0.14-0.5 ㎍ mL^-1; linearity (r²) was ≥ 0.99; accuracy (expressed as recovery) was 78.9-114%; precision (relative standard deviation) was ≤ 4.28% in the HPLC method. In the LC-MS/MS method, LOQ was 0.01-2 ng mL^-1, linearity (r²) was ≥0.98, accuracy was 71.7-119%; precision was ≤ 12.5%. The developed methods were applied to 51 dietary supplements collected from 2019 to 2021 through MFDS alert system. Based on our previous monitoring study, major compounds were icariin, sibutramine, yohimbine, sildenafil, tadalafil, sennosides (A, B), cascarosides (A, B, C, D), and phenolphthalein. In this study, we re-analyzed samples of detected compounds, and evaluated the statistical difference using Bland-Altman analysis to compare two analytical approaches between HPLC and LC-MS/MS. These results showed a good agreement between two methods that can be used to monitor the unapproved ingredients in dietary supplements. The developed two methods are complementarily suitable for monitoring the adulteration of 80 unapproved compounds in dietary supplements.

Introduction

Lately, dietary supplements containing unapproved (hidden, undeclared, and unauthorized) ingredients (e.g., pharmaceuticals) that could be unsafe have been largely sold in global market. These unapproved compounds have potential adverse health effects on consumers owing to accidental misuse, overuse, interaction with other medications, underlying health conditions, or other pharmaceuticals within the supplements.¹ The consumers seeking optimum health or well-being have inflated the opportunities for exposure to unapproved ingredients compared to those in the past owing to the widespread distribution of illegal dietary supplements. Therefore, continuous monitoring study needs to be proceeded in order to protect consumer against illegal dietary supplements.

Due to the potential health risks associated with unapproved ingredients, regulatory authorities of each country have warning alert systems and exchange information in adulterated foods and dietary supplements.^1,2 Currently, the Korean ‘Ministry of Food and Drug Safety (MFDS)’ takes charge of safety management of adulterated dietary supplements according to the ‘Food Sanitation Act’ and ‘Custom Law’. In these regulations, illegal compounds are defined as pharmaceutical ingredients and erectile dysfunction drugs, anti-obesity drugs, anti-diabetes, etc., and their analogues. These compounds should not be detected in food and dietary supplements and foods containing these compounds should not be sold, manufactured and imported. Illegal dietary supplements containing these compounds are monitored and blocked through MFDS alert system.¹

Due to the increase in the adulteration of illegal compounds in dietary supplements, several analytical methods have been previously reported using high-performance liquid chromatography (HPLC) and liquid chromatography-tandem mass spectrometry (LC-MS/MS).^3–14 The multi-class analytical method using LC-ESI-MS/MS with MRM (multiple reaction monitoring) mode is the compound-specific method by the MS fragmentation patterns. LC-MS/MS method has been applied for reliable quantitation and confirmation of illicit compounds and suitable for simultaneously detecting rapidly target compounds, while HPLC assay is suitable for screening a wide variety of hidden compounds.^7,15,16 Thus, both analytical instruments can be complementarily used for monitoring of unapproved compounds.

In this study, we selected 80 illicit compounds, including erectile dysfunction drugs, anti-obesity drugs, anti-diabetes, anti-thyroid drugs, anti-psychotic drugs, laxatives, and botanical ingredients. Next, we developed high-performance liquid chromatography-photodiode array detector (HPLC-PDA) and LC-MS/MS analytical methods for simultaneous determination of 80 illicit compounds in dietary supplements. Furthermore, we reviewed these methods suitable for monitoring and detecting illicit compounds in adulterated dietary supplements.

Materials and methods

Standards and reagents

All chemical standards and reagents were HPLC-grade. Two psychotropic compounds [β-methylphenylethylamine (BMPEA), 95.0%; and fenfluramine hydrochloride, 99.2%] were taken from the MFDS according to the Narcotics Control Act. Most of the target compounds including acetaminotadalafil, acetil acid, acetylvardenafil, aminotadalafil, avanafil, benzylsildenafil, carbodenafil, chlorodenafil, chloropretadalafil, cinnamyldenafil, cis-cyclopentyltadalafil, transcyclopentyltadalafil cyclopentynafil, demethylhongdenafil, demethyltadalafil, descarbonsildenafil, desmethylpiperazinylpropoxysildenafil, desulfonylchlorosildenafil, desulfovardenafil, dichlorodenafil, dimethylsildenafil, dimethylthiosildenafil, dithiopropylcarbodenafil, gendenafil, homosildenafil, homotadalafil, hongdenafil, hydroxychlorodenafil, hydroxyhomosildenafil, hydroxyhongdenafil, hydroxythiohomosildenafil, hydroxyvardenafil, imidazosagatriazinone, isopropylnortadalafil, methylhydroxyhomosildenafil, nitrodenafil, norneosildenafil, norneovardenafil, N-octylnortadalafil, oxohongdenafil, piperidinohongdenafil, propoxyphenylthioaildenafil, propoxyphenylthiohomosildenafil, propoxyphenylthiohydroxyhomosildenafil, propoxyphenylthiosildenafil, pseudovardenafil, thiohomosildenafil, thiosildenafil, thioquinapiperifil, udenafil, vardenafil, xanthoanthrafil, cascarosides (A, B, C, D), chlorosibutramine, chlorosipentramine, desmehtylsibutramine, didesmethylsibutramine, icaritin, and N-nitrosofenfluramine were synthesized or isolated by the MFDS. Commercial standard compounds [mirodenafil, sildenafil citrate, tadalafil, yohimbine hydrochloride, rauwolscine hydrochloride (α-yohimbine), icariin, ephedrine hydrochloride, fluoxetine, glibenclamide, gliclazide, glimepiride, glipizide, levothyroxine (T4), liothyronine (T3), orlistat, phenolphthalein, β-phenylethylamine hydrochloride (β-PEA), sennosides (A, B), and sibutramine] were purchased from Sigma-Aldrich (St. Louis, MO, USA), Cayman Chemical (Ann Arbor, MI, USA), Pfizer (New York, USA), Eli Lilly and company (Indianapolis, IN, USA), SK Chemical (Gyeonggi-do, Korea) and Hanmi (Seoul, Korea). HPLC-grade water, acetonitrile and methanol were purchased from Merck (Darmstadt, Germany). Sodium-1-hexane sulfonate was obtained from Tokyo Chemical Industry (Tokyo, Japan); phosphoric acid was purchased from Wako (Osaka, Japan); and formic acid (≥ 95%) was purchased from Sigma-Aldrich (St. Louis, MO, USA). The polytetrafluoroethylene (PTFE) membrane filters (0.22 μm pore size) were obtained by Teknokroma (Barcelona, Spain).

Preparation of standard solutions

The stock standard solutions were individually produced at 1,000 μg mL^–1 by weighing and dissolving each standard in methanol. The mixed working solutions were prepared by performing a dilution of aliquots of the stock solutions with methanol. For calibration curve and validation, working solutions were diluted to obtain seven serial dilutions that cover the range of 0.5–50 μg mL^–1 (0.5, 1, 2, 5, 10, 20, and 50 μg mL^–1) for HPLC; and six serial dilutions that cover the range of 1–20 ng mL^–1 (1, 2, 5, 8, 10, and 20 ng mL^–1) for LC-MS/MS. All solutions were stored at 4℃ in amber vials.

Sample preparation

A total of 51 samples advertised as sexual enhancement, weight-loss, muscular strengthening were collected between 2019 and 2021 through the MFDS alert system that monitors unapproved ingredients in dietary supplement. The sample types were capsule, tablet, powder, and liquid. The inner powder of capsule form was homogenized after removing a shell of capsule. Tablet form was grinded into powder using a mortar and pestle. In brief, homogenized sample weighed at 1 g was transferred into a 50 mL conical tube. The sample was mixed with 15 mL of water for 1 min. Then, 25 mL of methanol was added to sample solution and the solution was sonicated for 10 min. After the solution was set to the final volume up to 50 mL, the supernatant was filtered through a 0.22 μm PTFE syringe filter. Final extract of the sample was directly applied or appropriately diluted with 70% methanol before injecting into HPLC or LC-MS/MS.

HPLC-PDA analysis

A Shiseido Nanospace S1-2 HPLC system (Osaka soda Co., Ltd., Tokyo, Japan) accompanied with a PDA detector and a Capcell Pak C₁₈ column (MG II, 4.6 × 250 mm, 5.0 μm) was used. The UV detection was performed at 210 nm, 220 nm and 291 nm. The oven temperature was held at 40℃. The injection volume was 5 μL and the flow rate was 1.2 mL min^–1. The binary mobile phase consisted of a 0.5 mM sodium-1-hexane sulfonate in 0.1% (v/v) phosphoric acid solution (A) and 95% acetonitrile (B). The gradient elution program was as follows: 0–6 min (A 85%, B 15%); 6–21 min (A 70%, B 30%); 21–31 min(A 60%, B 40%); 31–35 min (A 60%, B 40%); 35–43 min (A 0%, B 100%); 43–50 min (A 0%, B 100%); 50–52 min (A 85%, B 15%); and 52–60 min (A 85%, B 15%).

LC-MS/MS analysis

A Waters AQUITY ultra-pressure liquid chromatography (UPLC) equipped with Waters Xevo TQ-S (Waters, Milford, MA, USA) and AQUITY UPLC^® HSS C₁₈ column (2.1 × 150 mm, 1.8 μm) was used. The column temperature was held at 40℃. The injection volume was 5 μL. The binary mobile phases were 0.1% (v/v) formic acid in water (A) and 0.1% (v/v) formic acid in acetonitrile (B). The flow rate was 0.3 mL min^–1. The mobile phase gradient was as follows: 0–2 min (A 95%, B 5%); 2–19 min (A 30%, B 70%), 19–20.1 min (A 0%, B 100%); 20.1–21.9 min (A 0%, B 100%); 21.9–22 min (A 95%, B 5%); and 22–25 min (A 95%, B 5%). The mass spectrometer was operated in electrospray ionization (ESI) positive or negative mode, and data acquisition was practiced in the MRM mode using the MassLynx v4.1 software (Waters, Milford, MA, USA). The source settings were as follows: capillary voltage of 3.5 kV and -2.8 kV in ESI positive mode and negative mode, respectively; source temperature of 150℃; desolvation temperature of 600℃; cone nitrogen gas flow rate of 60 L h^–1; and desolvation gas flow rate of 650 L h^–1. Collisioninduced dissociation was performed using argon as the collision gas at a pressure of 4 × 10^–3 mbar in the collision cell. The [M + H]⁺ and [M - H]^- ions were selected as precursor ions; and the two or three intense product ions were used as product ions. The most intense ion was selected for quantification, and the other ions for confirmation.

Method validation

The validation on analytical method was carried out in accordance with AOAC guideline (AOAC, 2002). The validation parameters were linearity, limit of detection (LOD), limit of quantitation (LOQ), accuracy and precision. Blank samples were confirmed to be free of the target analytes. The linearity was gained from the correlation coefficient (r²) of the calibration curve. The calibration curves were based on a spiked standard calibration. The LOD and LOQ were calculated by measuring the equation, 3 × σ/s and 10 × σ/s of the standard calibration curve, respectively. The accuracy (recovery, %) and precision [relative standard deviation (RSD), %] were determined through the analysis of spiked blank sample in five replicates for three different concentrations. The intra-day tests were analyzed on single day by performing five replicates at each level, whereas the inter-day tests were performed once a day during 3 days at each level.

HPLC-PDA and LC-MS/MS method comparison

To compare HPLC-PDA and LC-MS/MS methods, concentrations of detected compounds were determined in 51 dietary supplements by the two different analytical approaches. For method comparison, the linear correlation of the relationship between the two methods was assessed by the Passing and Bablok regression analysis.¹⁷ The results were also compared by means of Bland-Altman plot.¹⁸ Statistical analysis was carried out using MedCalc (Windows) version 20.1 (MedCalc Software, Ostend, Belgium)

Results and Discussion

Optimization of HPLC parameter

HPLC-PDA was used for screening and confirming the presence of 80 illicit compounds in dietary supplements. We optimized the basic chromatographic conditions and HPLC parameters, including mobile phase and detection wavelength, after testing the different conditions that affect HPLC analysis.^{19– 21} The parameters offered the best separation, apparent peak shape, and maximum signal. The chromatographic division of target compounds was also optimized by adjusting the retention time of each compound. Figure 1 represents HPLC chromatograms of 80 illicit compounds.

Figure 1. HPLC Chromatograms of 80 unapproved compounds: (A) 19 compounds were detected at 291 nm; (B) 20 compounds were detected at 291 nm; (C) 18 compounds were detected at 291 nm; (D) 2 compounds (yohimbine and icariin) were detected at 291 nm; (E) 4 compounds (cascaroside A, cascaroside B, cascaroside C, and cascaroside D) were detected at 291 nm; (F) 2 compounds (β-PEA and BMPEA) were detected at 210 nm; (G) 6 compounds (ephedrine, liothyronin, phenolphthalein, fluoxetine, levothyroxine and orlistat) were detected at 210 nm; 13 compounds (sennoside A, sennoside B, fenfluramine, didesmethylsibutramine, desmethylsibutramine, sibutramine, glipizide, chlorosibutramie, chlorosipentramine, gliclazide, glibenclamide, N-nitrosofenfluramine, and glimepiride) were detected at 220 nm.

Using an organic mobile phase with lower absorbance and a proper buffer solution results in less noise and fewer ghost peaks for baseline in reversed-phase chromatography and UV detection, leading to high-sensitivity analysis. The elution capacity is also higher when aqueous and organic solvents are mixed together.^22,23 The separation of target compounds was achieved using 0.5 mM sodium-1-hexane sulfonate in 0.1% (v/v) phosphoric acid solution and 95% acetonitrile as organic phase on a C₁₈ column. The proportion of the mobile phase components was optimized by adding variety to mobile phases and analyzing repeatedly the standard mixture. The optimized gradient condition enabled good peak resolution and narrow retention time within 50 min. When we scanned the standard working solutions for 80 illicit compounds in the range from 190 to 400 nm using HPLC-PDA detection method for adulteration provided by the “Food code”, target compounds were integrated at UV wavelengths 210 nm, 220 nm, and 291 nm. A total of 60 peaks were detected at the wavelength of 291 nm, while 12 peaks and 8 peaks were separated at 220 nm and 210 nm, respectively. Moreover, the sensitivity of certain compounds decreased when applying these method. In order to achieve the optimum sensitivity with HPLCP-DA, we improved analyzing wavelengths by adjusting from 195 nm to 210 nm for orlistat, and from 291 nm to 210 nm for phenolphthalein, which represent good response and maximum peak intensities.

Optimization of LC-MS/MS parameter

Simultaneous LC-MS/MS method was developed for identifying 80 target compounds in dietary supplements (Figure 2). We adjusted the LC parameters, and optimized the chromatographic separation on a basis of previous study.¹⁹ The MS signal intensity depends on multiple factor such as mobile phase pH and organic percentage, type and concentration of mobile phase electrolytes, and LC separation efficiency.²⁴

Figure 2. LC-MS/MS Chromatograms of 80 unapproved compounds.

Reversed-phase solvents (water, acetonitrile, methanol, etc.) are suitable for ESI application as they transfer ions from the liquid phase to the gas phase. They can support ions in solution, sensitivity results are better than normal-phase solvents (hexane, toluene, dichloromethane, etc.). In addition, adjusting pH is an effective strategy of facilitating analyte ionization. It can be good for making the analyte into charged form, and increasing a signal intensity. Ion pair reagent should be used, such as acidic additives (formic acid, acetic acid, etc.) and basic additives (ammonium formate, ammonium acetate, etc.).^24,25

When we compared peak intensities using formic acid, ammonium formate, and ammonium acetate, formic acid increased MS signal intensities and offered a good peak separation (data not shown). Therefore, 0.1% formic acid was chosen to improve chromatographic resolution. As a result, the optimal separation of target compounds was achieved using 0.1% formic acid in aqueous mobile phase and 0.1% formic acid in acetonitrile as organic phase on a C₁₈ column. The gradient elution condition was also optimized to yield good separation over 25 min by the variation of mobile phase with the repeated analysis of the standard solution.

The ideal MS/MS parameters was specifically set up for each compound with the evaluation of sensitivity and abundance. The determination was performed by the direct infusion of individual solution (100–500 ng mL^–1) using ESI positive or negative mode. We got the maximum of intensity for the fragment ions while adding cone voltages (10–70 V) and collision energies (5–60 eV). Both the singly protonated ([M + H]⁺) and deprotonated ([M - H]^-) molecular ions were selected as the precursor ions for 74 compounds and 6 compounds, respectively. The MRM transitions were produced with the most abundant ion for quantification and the other ions for confirmation. As seen from Table 1, there are the optimized MRM conditions for 80 illicit compounds. When applying developed this method, the chromatographic condition provided the best peak shape, separation, and resolution. Furthermore, this method can be achieved simultaneous detection of 80 compounds with high quality separation despite a large number of target analytes.

Table 1. Optimized multiple reaction monitoring (MRM) conditions for 80 unapproved compounds.

1) Values in underline denote quantification ion.

Method validation

Table 2 summarized linearity, LOQ, accuracy and precision for target compounds in dietary supplements using HPLC. The calibration curves were made by sevenpoint calibrations of standards in blank matrices at 0.5–50 μg mL^–1. The correlation coefficients (r²) were above 0.99 in all compounds. The LOQs were 0.14–0.50 μg mL^–1. The accuracy and precision were evaluated in spiked blank samples at three target concentrations of 2, 10, and 20 μg mL^–1. The accuracy (expressed as recovery) was in ranges of 78.9–114% for intra-day, and 83.9–109% for inter-day. The precision (expressed as RSD) was below 4.28% for intra-day, and below 2.21% for inter-day.

Table 2. Linearity, limit of quantification (LOQ), accuracy, and precision of 80 unapproved compounds using HPLC-PDA

1) RSD represents relative standard deviation.

Table 3 describes linearity, LOQ, accuracy and precision for target compounds in dietary supplements using LC-MS/MS. The calibration curves were made by six-point standard calibration in blank solid matrices at 1–20 ng mL^–1. All correlation coefficients (r²) were higher than 0.98, which showed a good linear relationship. The LOQs were ranged from 0.01 to 2 ng mL^–1 in solid-type blank samples. The accuracy and precision were evaluated in spiked blank samples at concentrations of 1, 5, and 10 ng mL^–1. The LOD was below 1 ng mL^–1. The target testing level of three compounds (cascarosides, sennosides, and β-PEA) were 10, 50, and 100 ng mL^–1 due to the lower sensitivity. The accuracy (expressed as recovery) was in ranges of 71.7–119% for intra-day, and 78.3–114% for inter-day. The precision (expressed as RSD) below 12.5% for intra-day, and below 12% for inter-day. These analytical methods showed satisfactory values for all method validation parameters (linearity, LOD, LOQ, accuracy, and precision) according to the requirements of AOAC guidelines.

Table 3. Linearity, limit of quantification (LOQ), accuracy, and precision of 80 unapproved compounds using LC-MS/MS.

1) RSD represents relative standard deviation.

Analysis of samples using HPLC-PDA and LC-MS/MS methods

The developed methods were applied to 51 dietary supplements collected in 2019–2021 through MFDS’s alert system. The product category of samples was classified into sexual enhancement product (39.2%), weight-loss product (49%), and muscular strengthening product (11.8%). Table 4 provides the number and concentration (range, mg g^–1) of illicit compounds detected in dietary supplements by HPLC-PDA and LC-MS/MS. Of all monitored products, the most detected compound in dietary supplements was icariin (9 products), sibutramine (9 products), and yohimbine (9 products), followed by sildenafil (8 products), tadalafil (7 products), sennosides (A, B) (6 products), cascarosides (A, B, C, D) (4 products), and phenolphthalein (4 products).

Table 4. Analytical results for illicit compounds in 51 sample of dietary supplements collected between 2019 and 2021 through MFDS alert system.

Adulterated product containing two illicit compounds

Icariin is an active flavonoid extracted from horn goat weed (Epimedium koreanum) and used in traditional Chinese medicine for the treatment of erectile dysfunction (ED). Icariin as an adulterant in dietary supplements has been used for the purpose of enhancing sexual and sporting performance. However, the toxicology of icariin has not been investigated and established for safe intake of icariin in food supplements.^1,2 Icariin was detected in sexual enhancement, weight-loss, and muscular strengthening products in this study. Yohimbine is a natural tryptamine alkaloid extracted from yohimbe bark (Rausinystalia yohimbe). Yohimbe bark is traditionally used in Africa as an aphrodisiac or yohimbine is a drug in veterinary medicine in livestock products in Korea. Due to a significant risk to consumers, yohimbine is banned in dietary supplement in several countries. However, so far, yohimbine is consumed to improve sexual satisfaction, athletic performance and weight loss as an adulterant in dietary supplements.^1,26 In this study, yohimbine was detected in weight-loss and muscular strengthening products. Sibutramine is an approved drug for long-term use in obesity treatment. But no longer available in dietary supplements because this drug may pose cardiovascular risks, such as heart attack, stroke and cardiac arrest.^1,2 In this study, sibutramine was detected in weight-loss products.

Comparison of HPLC-PDA and LC-MS/MS method

In this study, illicit compounds detected in 51 dietary supplements were determined by two different analytical approaches such as HPLC-PDA and LC-MS/MS. Similar concentrations were found by the two methods, showing icariin values in sexual enhancement products of 0.12–8.80 mg g^–1 and 0.10–8.68 mg g^–1 by HPLC-PDA and LC-MS/MS, respectively. Also, other compounds, including sibutramine, yohimbine, sildenafil, tadalafil, sennosides, cascarosides, and phenolphthalein, showed similar result values (Table 4).

The Passing and Bablok regression analysis indicated a high degree of linear correlation between two methodologies giving correlation coefficient (r²>0.9). These results were also confirmed by the Bland-Altman analysis (Figure 3). The statistical results showed good agreement between two methods with a mean bias of 6.1 mg g^-1 for sildenafil (95% limits of agreement: -12~24), 1.0 mg g^-1 for tadalafil (95% limits of agreement: -12~14), 6.9 mg g^-1 for sibutramine (95% limits of agreement: -5~19), 1.1 mg g^-1 for phenolphthalein (95% limits of agreement: 0~2), 0 mg g^-1 for icariin (95% limits of agreement: -0.3~0.3), -0.9 mg g^-1 for yohimbine (95% limits of agreement: -3~1), 0.5 mg g^-1 for sennosides (95% limits of agreement: -1~2), -1.9 mg g^-1 for cascarosides (95% limits of agreement: -8~4). Both methods can be effectively used to monitor the unapproved ingredients in dietary supplements.

Figure 3. Bland-Altman plots showing agreement between the HPLC and the LC-MS/MS methods. (A) Agreement for sildenafil determination showed a mean bias of 6.1 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: -12 to 24. (B) Agreement for tadalafil determination showed a mean bias of 1.0 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: -12 to 14. (C) Agreement for sibutramine determination showed a mean bias of 6.9 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: -5 to 19. (D) Agreement for phenolphthalein determination showed a mean bias of 1.1 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: 0 to 2. (E) Agreement for icariin determination showed a mean bias of 0 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: -0.3 to 0.3. (F) Agreement for yohimbine determination showed a mean bias of -0.9 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: -3 to 1. (G) Agreement for sennosides determination showed a mean bias of 0.5 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: -1 to 2. (H) Agreement for cascarosides determination showed a mean bias of -1.9 mg g^–1 (LC-MS/MS–HPLC-PDA), 95% limits of agreement: -8 to 4.

Conclusion

The objective of this study was to develop as reliable, easy and fast method for the simultaneous determination of 80 illicit compounds in dietary supplements using HPLC-PDA and LC-MS/MS. The unapproved compounds in 51 dietary supplements collected between 2019 and 2021 were successfully determined using two analytical methods. The validation results of developed method were satisfactory with the AOAC guidelines, and showed acceptable performances for accuracy and precision. The concentrations of detected compounds were re-analyzed using two different analytical approaches such as HPLC-PDA and LC-MS/MS. In addition, these values were compared using the Passing and Bablok analysis and Bland-Altman analysis, and we confirmed a good agreement between two methods. Therefore, it is suggested that developed methods can be complementarily used to monitor simultaneously the unapproved 80 ingredients in dietary supplements.