# Development and Validation of an HPLC-UV Method for the Dissolution Studies of 3D-Printed Paracetamol Formulations in Milk-Containing Simulated Gastrointestinal Media

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

^{−1}. Following a protein precipitation-based sample clean-up, a thorough investigation of the effect of the precipitation reagent (methanol, acetonitrile, 10% v/v trifluoroacetic acid solution) on the analyte recovery was performed. The matrix effect was assessed in each biorelevant medium by comparing the slopes of the calibration curves of aqueous and matrix-matched calibration curves. The method was comprehensively validated using the accuracy profiles. The β-expectation tolerance intervals did not exceed the acceptance criteria of ±15%, meaning that 95% of future results will be included in the defined bias limits. The relative bias ranged between −4.5 and +3.9% for all analytes, while the RSD values for repeatability and intermediate precision were less than 2.7% and 3.0%, respectively. The achieved limit of detection (LOD) was 0.02 μg mL

^{−1}and the lower limits of quantitation (LLOQ) were established as 10 μg mL

^{−1}, which corresponded to 2% of the highest expected concentration of paracetamol. The proposed scheme was utilized for the determination of paracetamol in dissolution studies of its 3D-printed formulation in milk-containing biorelevant media.

## 1. Introduction

## 2. Results and Discussion

#### 2.1. Method Development

_{3}PO

_{4}: CH

_{3}OH (80:20, v/v) using a flow rate of 1 mL min

^{−1}. Under the optimum experimental conditions, a symmetric PAR peak was achieved (T

_{f}= 1.1), while the number of theoretical plates was N > 13,000. The runtime was set to 12 min. Extended analysis time up to 60 min evidenced no additional peaks, suggesting that PAR did not decompose in these media. Furthermore, the certain runtime was adequate for the elution of all matrix components. The retention time for PAR was 5.9 min.

#### 2.2. Optimization of Sample Preparation

^{−1}in order to assess the method’s efficiency in the entire determination range.

_{FaSSGF}= 1.6 and pH

_{FaSSIF/FeSSIF}= 6.5) as well.

_{a}of PAR of 9.5, its molecule is expected to be fully protonated at the sample pH of both SGF and SIF. Figure 1 shows representative chromatograms of the blank FaSSGF: milk 4:1 medium at different milk fat content levels and spiked with (c = 10 μg mL

^{−1}).

#### 2.3. Method Validation

#### 2.3.1. System Suitability

^{−1}PAR solution. The %RSD values for the peak area and retention time were 1.2 and 0.8%, respectively, which are within the acceptance criterion of ±2%. Under the optimum conditions, the peak tailing was found to be less than 1.1 and the number of theoretical plates was more than 5000 (ca. 13000), indicating the acceptability of the peak characteristics.

#### 2.3.2. Selectivity and Carry-Over

^{−1}). The HPLC method showed good chromatographic separation of PAR and milk-based biorelevant media. No interferences were observed in the tested matrices. As depicted in Figure 2, PAR was eluted as a single peak and good resolution between PAR and the other peaks of the sample matrix was observed.

^{−1}. No “ghost peaks” were recorded and no effect on the response of PAR was observed, suggesting that the washing procedure that was followed between the injections was sufficient.

#### 2.3.3. Selection of the Response Function

_{r}, %) and the upper and the lower β-ETI were calculated by using the back-calculated concentrations of the validation standards through each regression model. The suitability of these models was evaluated by plotting the accuracy profiles at a probability of 95%, while the acceptance limits of λ at ±15% level were considered [19].

#### 2.3.4. Linearity

^{−1}).

^{2}value was found to be 0.9999, indicating the acceptable linearity of the developed analytical method.

#### 2.3.5. Accuracy, Precision and Trueness

_{R}, %) and of repeatability (s

_{r}, %). As shown in Table 2, the s

_{r}and s

_{R}values were less than 2.7 and 3.1%, respectively, showing the good precision of the proposed method.

#### 2.3.6. LODs and LOQs

^{−1}), respectively. The LOD was estimated to be 0.02 μg mL

^{−1}based on the S/N = 3 criterion.

#### 2.3.7. Solution Stability

#### 2.4. Method Application—Dissolution Study

## 3. Materials and Methods

#### 3.1. Chemicals and Solutions

^{®}) were utilized for the aqueous mobile phase filtration.

_{2}O. Accordingly, the pH value of FaSSGF was adjusted to 1.6 with hydrochloric acid, and water was added to obtain a final volume of 500 mL. Then, an amount of 0.030 g of SIF Powder Original was added in 250 mL of the hydrochloric acid/NaCl solution and the volume was made up to 500 mL with the hydrochloric acid/NaCl solution.

_{2}PO

_{4}and 6.186 g NaCl in 475 mL water, followed by pH adjustment to 6.8 using NaOH in order to achieve a final pH value of 6.5 after (1:1) mixing of FaSSGF: 2 × FaSSIF media. The final volume of this solution was made up to 500 mL. Accordingly, SIF Powder (2.24 g) was dissolved in 375 mL of the FaSSIF and the volume was made up to 500 mL with the same medium. The obtained medium was left to stand for 2 h prior to its use.

_{3}COOH in 450 mL of H

_{2}O, followed by pH adjustment to 5.2 in order to achieve a final pH value of 5.0 after (1:1) mixing of FaSSGF: 2 × FeSSIF media. Then, the final volume of this solution was made up to 500 mL and an amount of 11.2 g SIF Powder was dissolved in 250 mL of the FeSSIF medium, followed by volume adjustment to 500 mL with the FeSSIF. The obtained medium was left to stand for 2 h prior to its use.

#### 3.2. Instrumentation and Chromatographic Conditions

_{18}(250 × 4.6 mm, 5 μm, Thermo Fisher Scientific, Waltham, MA, USA). Data acquisition and instrument control were performed using Lab Solutions software (version 5.42 SP3).

^{−1}and the chromatographic run lasted 12 min. An injection volume of 10 μL was chosen to reduce the exposure of the stationary phase to potential fat and protein residues. The column was maintained at 30 °C. The analyte was detected spectrophotometrically at its maximum wavelength of 243 nm to reduce absorbance interferences from the dissolution medium. The samples were maintained at 4 °C in the autosampler tray. Between the injection of different samples, the autosampler was sequentially washed with 500 μL of water: CH

_{3}OH (50:50, v/v) mixture to remove any residual sample.

#### 3.3. Preparation of Calibration (CS) and Validation Standards (VS)

^{−1}) were prepared and stored at 4 °C. Working solutions were used on the same working day. Validation standards (VS) and calibration standards (CS) were made by serial dilution of the stock solutions on three successive days (n = 3). Five concentration levels (m = 5), namely 10, 50, 100, 300 and 600 μg mL

^{−1}(which corresponded to 2, 10, 20, 60 and 120% of the highest expected PAR concentrations after the complete dissolution of a 3D-printed formulation), were prepared in three replicates (n =3) for each experiment series (k = 3). Calibration standards were used to determine the most appropriate response function, allowing the quantification of samples.

#### 3.4. Sample Preparation

#### 3.5. Method Validation

_{r}%) and the mean relative bias were evaluated through the calculation of the back-calculated concentrations that corresponded to the validation standards through the linear model. Moreover, accuracy profiles at the probability of 95% were plotted to assess its suitability. In this way, the results that might fall outside the acceptance limits λ are expected to be less than 5% [18,20,21,22]. In order to fulfil the conditions of a valid analytical method, the β-ETI must be included inside the range −λ and +λ. The mathematical expression that describes the analytical profile is the following:

_{j}= $\frac{{\widehat{\mu}}_{j}-{\mu}_{Tj}}{{\mu}_{Tj}}\times 100$, s

_{r,j}= $\frac{{\widehat{\sigma}}_{W,j}^{2}+{\widehat{\sigma}}_{B,j}^{2}}{{\widehat{\mu}}_{j}}\times 100$, B

_{j}=$\sqrt{\frac{\frac{{\widehat{\sigma}}_{{\rm B},j}^{2}}{{\widehat{\sigma}}_{W,j}^{2}}+1}{n\frac{{\widehat{\sigma}}_{{\rm B},j}^{2}}{{\widehat{\sigma}}_{W,j}^{2}}+1}}$, ν = $\frac{{\left(R+1\right)}^{2}}{R+\frac{1}{n\left(p-1\right)}+1-\frac{1}{p{n}^{2}}}$,

- -
- ${\widehat{\mu}}_{j}$ is the estimate of the mean results of the j
^{th}concentration level; - -
- ${\widehat{\mu}}_{T}$ is the unknown “true value”;
- -
- p is the number of series;
- -
- n is the number of independent replicates per series;
- -
- Q
_{t}(ν;$\frac{1+\beta}{2})$ is the β quantile of the t-Student distribution with ν degrees of freedom; - -
- ${\widehat{\sigma}}_{W,j}^{2}$ is the within-series variance;
- -
- ${\widehat{\sigma}}_{B,j}^{2}$ is the between-series variance.

#### 3.6. System Suitability Test

^{−1}were performed to determine the column efficiency, N (in terms of number of theoretical plates), retention time, injection repeatability and tailing factor (T

_{f}). The acceptance criteria were considered as N > 5000, RSD ≤ 2% for both retention time and peak area and T

_{f}not more than 1.5.

#### 3.7. Application of the Proposed Method—Dissolution Test

## 4. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Dressman, J.B.; Amidon, G.L.; Reppas, C.; Shah, V.P. Dissolution testing as a prognostic tool for oral drug absorption: Immediate release dosage forms. Pharm. Res.
**1998**, 15, 11–22. [Google Scholar] [CrossRef] [Green Version] - US Food and Drug Administration Guidance for Industry: Extended Release Oral Dosage Forms, Development Evaluation and Application of In Vitro/In Vivo Correlations. Cent. Drug Eval. Res.
**1997**. - Jantratid, E.; Janssen, N.; Reppas, C.; Dressman, J.B. Dissolution Media Simulating Conditions in the Proximal Human Gastrointestinal Tract: An Update. Pharm. Res.
**2008**, 25, 1663–1676. [Google Scholar] [CrossRef] - Wang, Q.; Fotaki, N.; Mao, Y. Biorelevant Dissolution: Methodology and Application in Drug Development. Dissolution Technol.
**2009**, 16, 6–12. [Google Scholar] [CrossRef] - Elkasabgy, N.A.; Mahmoud, A.A.; Maged, A. 3D printing: An appealing route for customized drug delivery systems. Int. J. Pharm.
**2020**, 588, 119732. [Google Scholar] [CrossRef] - Perrot, S.; Cittée, J.; Louis, P.; Quentin, B.; Robert, C.; Milon, J.; Bismut, H.; Baumelou, A. Self-medication in pain management: The state of the art of pharmacists’ role for optimal Over-The-Counter analgesic use. Eur. J. Pain
**2019**, 23, 1747–1762. [Google Scholar] [CrossRef] - Karavasili, C.; Zgouro, P.; Manousi, N.; Lazaridou, A.; Zacharis, C.K.; Bouropoulos, N.; Moschakis, T.; Fatouros, D.G. Cereal-Based 3D Printed Dosage Forms for Drug Administration During Breakfast in Pediatric Patients within a Hospital Setting. J. Pharm. Sci.
**2022**. [Google Scholar] [CrossRef] - Baxevanis, F.; Kuiper, J.; Fotaki, N. Strategic drug analysis in fed-state gastric biorelevant media based on drug physicochemical properties. Eur. J. Pharm. Biopharm.
**2018**, 127, 326–341. [Google Scholar] [CrossRef] - Koesukwiwat, U.; Jayanta, S.; Leepipatpiboon, N. Solid-phase extraction for multiresidue determination of sulfonamides, tetracyclines, and pyrimethamine in Bovine’s milk. J. Chromatogr. A
**2007**, 1149, 102–111. [Google Scholar] [CrossRef] - Nász, S.; Debreczeni, L.; Rikker, T.; Eke, Z. Development and validation of a liquid chromatographic-tandem mass spectrometric method for determination of eleven coccidiostats in milk. Food Chem.
**2012**, 133, 536–543. [Google Scholar] [CrossRef] [PubMed] - Hamajima, M.; Ishida, S.; Hirata, Y.; Hirata, K.; Watanabe, K.; Seno, H.; Ishii, A. Simple analysis of acetaminophen in human plasma by solid-phase microextraction and gas chromatography. Jpn. J. Forensic Toxicol.
**2004**, 22, 22–26. [Google Scholar] - Kishida, K. Simplified extraction of tetracycline antibiotics from milk using a centrifugal ultrafiltration device. Food Chem.
**2011**, 126, 687–690. [Google Scholar] [CrossRef] - Franek, F.; Holm, P.; Larsen, F.; Steffansen, B. Interaction between fed gastric media (Ensure Plus
^{®}) and different hypromellose based caffeine controlled release tablets: Comparison and mechanistic study of caffeine release in fed and fasted media versus water using the USP dissolution apparatus 3. Int. J. Pharm.**2014**, 461, 419–426. [Google Scholar] [CrossRef] [PubMed] - Baxevanis, F.; Kuiper, J.; Fotaki, N. Fed-state gastric media and drug analysis techniques: Current status and points to consider. Eur. J. Pharm. Biopharm.
**2016**, 107, 234–248. [Google Scholar] [CrossRef] [Green Version] - ICH Harmonised Tripartite Guideline, Validation of Analytical Procedures: Text and Methodology, Q2 (R1). 2005. Available online: https://database.ich.org/sites/default/files/Q2%28R1%29%20Guideline.pdf (accessed on 10 June 2022).
- Valko, K.; Nunhuck, S.; Bevan, C.; Abraham, M.H.; Reynolds, D.P. Fast Gradient HPLC Method to Determine Compounds Binding to Human Serum Albumin. Relationships with Octanol/Water and Immobilized Artificial Membrane Lipophilicity. J. Pharm. Sci.
**2003**, 92, 2236–2248. [Google Scholar] [CrossRef] - Hubert, P.; Nguyen-Huu, J.J.; Boulanger, B.; Chapuzet, E.; Chiap, P.; Cohen, N.; Compagnon, P.A.; Dewe, W.; Feinberg, M.; Lallier, M.; et al. Validation of quantitative analytical procedure, harmonization of approaches. STP Pharma Prat.
**2003**, 13, 101–138. [Google Scholar] - Hubert, P.; Nguyen-Huu, J.J.; Boulanger, B.; Chapuzet, E.; Chiap, P.; Cohen, N.; Compagnon, P.A.; Dewe, W.; Feinberg, M.; Lallier, M.; et al. Harmonization of strategies for the validation of quantitative analytical procedures: A SFSTP proposal—Part I. J. Pharm. Biomed. Anal.
**2004**, 36, 579–586. [Google Scholar] [CrossRef] [PubMed] - Bouabidi, A.; Talbi, M.; Bourichi, H.; Bouklouze, A.; El Karbane, M.; Boulanger, B.; Brik, Y.; Hubert, P.; Rozet, E. Flexibility and applicability of β-expectation tolerance interval approach to assess the fitness of purpose of pharmaceutical analytical methods. Drug Test. Anal.
**2012**, 4, 1014–1027. [Google Scholar] [CrossRef] - Hubert, P.; Nguyen-Huu, J.-J.; Boulanger, B.; Chapuzet, E.; Cohen, N.; Compagnon, P.-A.; Dewé, W.; Feinberg, M.; Laurentie, M.; Mercier, N.; et al. Harmonization of strategies for the validation of quantitative analytical procedures: A SFSTP proposal—Part III. J. Pharm. Biomed. Anal.
**2007**, 45, 82–96. [Google Scholar] [CrossRef] [PubMed] - Hubert, P.; Nguyen-Huu, J.-J.; Boulanger, B.; Chapuzet, E.; Cohen, N.; Compagnon, P.-A.; Dewé, W.; Feinberg, M.; Laurentie, M.; Mercier, N.; et al. Harmonization of strategies for the validation of quantitative analytical procedures: A SFSTP proposal—Part IV. J. Pharm. Biomed. Anal.
**2008**, 48, 760–771. [Google Scholar] [CrossRef] [PubMed] - Feinberg, M. Validation of analytical methods based on accuracy profiles. J. Chromatogr. A
**2007**, 1158, 174–183. [Google Scholar] [CrossRef] [PubMed]

**Figure 1.**Representative overlaid chromatograms of blank FaSSGF:milk, 4:1, at different milk fat content levels (0.1, 1.5 and 3.6%) extracted with (

**A**) 10% v/v aq. solution of TFA, (

**B**) MeOH and (

**C**) ACN. The chromatographic conditions are described in Section 2.2.

**Figure 2.**Representative HPLC chromatograms of the analysis of blank and spiked samples (with 10 μg mL

^{−1}PAR). (

**A**) FaSSGF: milk (4:1); (

**B**) FaSSIF: milk (4:1); (

**C**) FeSSIF: milk (4:1). Full-fat (3.6%) milk was used in all cases.

**Figure 3.**Accuracy profile for the determination of PAR using (

**A**) linear unweighted regression model; (

**B**) linear regression fitted only at the highest concentration (120% level); (

**C**) weighted (1/X) linear regression; (

**D**) linear regression after square root transformation. The relative error (%), the accuracy profile and the acceptance limits λ (±15%) are described by the red plain, blue dashed and blank dotted lines, respectively. The dotted curves show the acceptance limits λ ± 15%, the plain blank line shows the Y = X identity line, and the blue dashed line shows the accuracy profile (β-ΕΤΙ).

**Figure 4.**Linearity profile for PAR. The identity line (Y = X) corresponds to the plain blank line, the accuracy profile (β-ΕΤΙ) corresponds to the blue dashed line, and the acceptance limits λ ±15% correspond to the dotted curves.

**Figure 5.**Dissolution profile of paracetamol-loaded cereal following mixing with full-fat (3.5%, blue and red lines) or low-fat (1.5%, green and black lines) milk under fasted-state gastric conditions (FaSSGF) followed by fasted-state (FaSSIF) or fed-state (FeSSIF) intestinal conditions at 37 °C (n = 3, ±S.D.). The red vertical line shows the time-point of medium change (60 min) (adapted from [7]).

**Figure 6.**An overlay of chromatograms of the HPLC analysis of the dissolution samples taken at (

**A**–

**I**) 5–240 min, and (

**J**) PAR standard solution (10% level). The bottom figure is a magnified overlaid chromatogram (t

_{R}, PAR = 5.8 min).

Calibration Curve ^{1} | Slope ± SD ^{2} (×10^{3}) | |||||
---|---|---|---|---|---|---|

Aqueous | 21.1 ± 0.10 | |||||

Precipitation reagent/fat content | MeOH | ACN | 10% TFA | |||

Slope ± SD (×10^{3}) | ME ^{3} (%) | Slope ± SD (×10^{3}) | ME% | Slope ± SD (×10^{3}) | ME% | |

[FaSSGF:milk, 4:1] | ||||||

0.1% | 20.7 ± 0.31 | 98.1 | 21.4 ± 0.26 | 101.4 | 20.8 ± 0.31 | 98.6 |

1.5% | 20.7 ± 0.10 | 98.1 | 20.9 ± 0.07 | 99.1 | 20.4 ± 0.07 | 96.7 |

3.6% | 21.0 ± 0.08 | 99.5 | 22.0 ± 0.28 | 104.2 | 21.4 ± 0.16 | 101.4 |

[FaSSIF:milk, 4:1] | ||||||

0.1% | 19.4 ± 0.28 | 91.9 | 20.8 ± 0.36 | 98.6 | 20.2 ± 0.17 | 95.7 |

1.5% | 20.3 ± 0.06 | 96.2 | 19.0 ± 0.22 | 90.0 | 19.6 ± 0.22 | 92.9 |

3.6% | 20.9 ± 0.17 | 99.1 | 21.5 ± 0.12 | 101.9 | 19.7 ± 0.14 | 93.4 |

[FeSSIF:milk, 4:1] | ||||||

0.1% | 19.5 ± 0.31 | 92.4 | 20.8 ± 0.45 | 98.6 | 20.1 ± 0.26 | 95.3 |

1.5% | 20.6 ± 0.28 | 97.6 | 19.7 ± 0.17 | 93.3 | 19.2 ± 0.43 | 91.0 |

3.6% | 20.5 ± 0.23 | 97.2 | 21.3 ± 0.20 | 101.0 | 20.1 ± 0.35 | 95.3 |

^{1}Constructed using five calibration levels of 2–120%.

^{2}Standard deviation of the slopes.

^{3}Matrix effect was calculated as ratio of the slopes of the matrix-matched (individual sample) to the aqueous calibration curves.

**Table 2.**Validation results for the determination of PAR in milk-containing biorelevant dissolution samples.

Validation Criteria | |||
---|---|---|---|

Response function (linear regression) | Slope (×10^{3}) | Intercept (×10^{3}) | r^{2} |

(k ^{1} = 3; m = 5; n = 3) (2–120%) | |||

Day 1 | 21.07 | −2.483 | 0.9997 |

Day 2 | 21.13 | −0.238 | 0.9999 |

Day 3 | 21.18 | −3.135 | 1.0000 |

Precision (k = 3; n = 3) | |||

C (μg mL^{−1}) | s_{r} (%) ^{2} | s_{R} (%) ^{3} | |

10 | 2.1 | 2.6 | |

50 | 2.7 | 3.0 | |

100 | 1.9 | 3.1 | |

300 | 1.5 | 2.4 | |

600 | 1.4 | 2.5 | |

Trueness (k = 3; n = 3) | |||

C (μg mL^{−1}) | Relative bias (%) | ||

10 | +1.1 | ||

50 | +3.9 | ||

100 | −1.2 | ||

300 | −3.0 | ||

600 | −4.5 | ||

Accuracy (k = 3; n = 3) | |||

C (μg mL^{−1}) | Relative β-ΕΤΙ (%) | ||

10 | [−6.13, 8.25] | ||

50 | [−11.67, 3.81] | ||

100 | [−11.75, 9.28] | ||

300 | [−11.00, 5.10] | ||

600 | [−13.34, 4.29] | ||

Linearity (k = 3; n = 3; m = 5) (2–120%) | |||

Slope | 1.000 | ||

Intercept | −0.0004 | ||

r^{2} | 0.9999 | ||

LOD (μg mL^{−1}) | 0.02 | ||

LLOQ/ULOQ (%) | 2/120 |

^{1}k, m and n correspond to the number of experiments, calibration levels and replicates, respectively.

^{2}s

_{r}(%): relative standard deviation under repeatability conditions.

^{3}s

_{R}(%): relative standard deviation under intermediate precision.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Manousi, N.; Karavasili, C.; Fatouros, D.G.; Tzanavaras, P.D.; Zacharis, C.K.
Development and Validation of an HPLC-UV Method for the Dissolution Studies of 3D-Printed Paracetamol Formulations in Milk-Containing Simulated Gastrointestinal Media. *Pharmaceuticals* **2022**, *15*, 755.
https://doi.org/10.3390/ph15060755

**AMA Style**

Manousi N, Karavasili C, Fatouros DG, Tzanavaras PD, Zacharis CK.
Development and Validation of an HPLC-UV Method for the Dissolution Studies of 3D-Printed Paracetamol Formulations in Milk-Containing Simulated Gastrointestinal Media. *Pharmaceuticals*. 2022; 15(6):755.
https://doi.org/10.3390/ph15060755

**Chicago/Turabian Style**

Manousi, Natalia, Christina Karavasili, Dimitrios G. Fatouros, Paraskevas D. Tzanavaras, and Constantinos K. Zacharis.
2022. "Development and Validation of an HPLC-UV Method for the Dissolution Studies of 3D-Printed Paracetamol Formulations in Milk-Containing Simulated Gastrointestinal Media" *Pharmaceuticals* 15, no. 6: 755.
https://doi.org/10.3390/ph15060755