👨‍💻 Quantitative Analysis of N-Nitrosodimethylamine in Metformin HCl by Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry

Analytical Hub Disclaimer

"This information is intended to serve as a resource for informational purposes only and not as an USP-NF compendial documentary standard. USP shall not be responsible to verify the accuracy, completeness, timeliness, quality, and correctness of the information developed and/or posted by its members. This document is on an “as-is”, “as available” basis and USP expressly disclaims all warranties, including the warranties of merchantability, fitness for a particular purpose, and non-infringement. This document does not reflect USP or USP’s Expert Body opinions. Parties relying on the information in this document bear independent responsibility for awareness of, and compliance with, any applicable federal, state, or local laws and requirements.”
“By participating in discussions and posting content within the Analytical Hub, you confirm your agreement to the original Terms of Service of the Nitrosamines Exchange Community. We remind all members to respect intellectual property rights when sharing information, data, or resources. This includes but is not limited to: 1) Avoiding unauthorized distribution. 2) Crediting original sources. 3) Protecting proprietary information.”

image

Quantitative Analysis of N-Nitrosodimethylamine in Metformin HCl by Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry

Carlo Joson and Therese Margaret Villar

  • This analytical method highlights a sensitive and robust procedure for the quantitation of NDMA in Metformin HCl by LCMS/MS. The procedure was validated according to ASEAN Harmonization Guidelines, USP General Chapters <1225> and <736>.

LOQ (Limit of Quantitation): 3 ng/mL with respect to 1 mg/mL sample concentration

Validated Range: 3 ng/mL to 200 ng/mL with respect to 50 ng/mL standard concentration.

Precision: %Relative standard deviation (RSD) of NDMA of the 6 recoveries at 100% levels is 2.3%

Accuracy: Mean % recoveries of NDMA at 3 ng/mL, 50 ng/mL, and 200 ng/mL are within 91.6 ~ 95.4%.


1. Experimental Information

  1. Objective/Scope
    In the Philippines, awareness of nitrosamine contamination risks continues to grow, prompting local pharmaceutical manufacturers and regulatory bodies to enhance safety measures in alignment with international standards. UNILAB, Inc. has developed an in-house quantitative testing method using ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-LCMS/MS) system to detect NDMA in Metformin HCl, ensuring the safety and quality of its pharmaceutical products. This initiative highlights the industry’s commitment to proactive risk assessment and mitigation strategies, prioritizing public health.

  2. Instrumentation

    • Waters ACQUITY UPLC I-Class PLUS System with Xevo TQ-S micro Triple Quadrupole Mass Spectrometer
  3. Method Principle

    • NDMA analysis using reverse-phase UPLC with LCMS/MS detection

2. Sample Preparation

  1. Sample Type and Matrix

    • Metformin Hydrochloride API
    • Note any unique considerations (e.g., sample stability, storage conditions)
  2. Diluent
    Methanol : Water (1:1) v/v

  3. Sample Preparation Procedure
    Weigh accurately about 100.0 mg of Metformin Hydrochloride. Transfer quantitatively into a 15-mL screw-capped test tube. Add 2 mL of Internal standard stock solution. Add 8 mL diluent. Mix the solution using a vortex mixer for 1 minute. Sonicate the solution for 40 minutes. Centrifuge at 4500 rpm for 15 minutes. Pipet 1 mL aliquot into a 15-mL screw-capped test tube. Add 9 mL diluent. Mix the solution using a vortex mixer for 1 minute. Filter through a 0.20-µm membrane filter, discarding the first few mL of the filtrate.

  4. Standard Solution Preparation

  • Standard stock solution (2500 ng/mL of NDMA): Transfer quantitatively the contents of 1 mL vial of NDMA primary reference standard into a 50-mL volumetric flask. Rinse the vial several times with methanol and transfer the washings into the same flask. Dilute to volume with methanol and mix well. Pipet 3.125 mL aliquot into a 25-mL volumetric flask. Dilute to volume with the diluent and mix well.

  • Internal standard stock solution (2500 ng/mL of NDMA-d6): Dissolve the contents of 25 mg vial of NDMA-d6 primary reference standard using methanol. Transfer quantitatively into a 50-mL volumetric flask. Rinse the vial several times and transfer the washings into the same flask. Dilute to volume with methanol and mix well. Pipet 125 µL into a 25-mL volumetric flask. Dilute to volume with the diluent and mix well.

  • Standard Solution (50 ng/mL of NDMA and 50 ng/mL NDMA-d6): Pipet 100 µL of Standard stock solution and 100 µL of Internal standard stock solution into a 15-mL screw-capped test tube. Add 4.8 mL of the diluent. Mix using a vortex mixer for 1 minute. Filter using 0.2 µm membrane filter, discarding the first few mL of the filtrate.

  • LOQ Solution: Pipet 6 µL of Standard stock solution and 100 µL of Internal standard stock solution into a 15-mL screw-capped test tube. Add 4.894 mL of the diluent. Mix using a vortex mixer for 1 minute. Filter using 0.2 µm membrane filter, discarding the first few mL of the filtrate.


3. Instrument Conditions

  1. Chromatographic Conditions
    • HSS T3, 1.8 µm, 2.1 x 100 mm (or equivalent)
    • Mobile phase A: 2 mM Ammonium Formate with 0.1% Formic Acid in Water
    • Mobile Phase B: 2 mM Ammonium Formate with 0.1% Formic Acid in Methanol
    • Column Oven Temperature: 40 °C
    • Autosampler Temperature: 5 °C
    • Seal Wash and Needle Wash: Methanol : Water (10:90) v/v
    • Flow rate: 0.41 mL/min
    • Injection volume: 3 µL
    • Gradient Program

  1. MS Detector Settings

  1. System Suitability Criteria
    a. Inject the LOQ solution. The test is not valid unless the signal-to-noise ratio is NLT 10.
    b. Determine the area ratio of five (5) replicate readings of the standard solution. The test is not valid unless the %RSD of the five (5) replicate readings is not more than 10.0%

4. Analysis and Results

  1. Data Processing

  2. Calculation

  1. Representative Data
    • NDMA Retention Time: 1.33 mins ± 0.05 mins
    • NDMA-d6 Retention Time: 1.33 mins ± 0.05 mins

  1. Observations
    • Active is light-sensitive. Perform analysis under subdued light and use low-actinic glassware.

5. Performance Characteristics

  1. Specificity
    • No interference from the blank is observed. The method is able to confirm that a positive or increase in peak area is not due to the structurally-related substance or placebo but due to the analyte of interest. (See Figure 2, Figure 3, Figure 4, and Figure 5)

  1. Linearity
    • Standard linearity was demonstrated by preparing five (5) standard solutions within the range 3 ppb to 200 ppb. Each solution was prepared by serial dilution from a single stock. Linear regression analysis was performed, excluding the origin as a point. Results are presented in Table 2. All acceptance criteria were met. A graph of area versus concentration is shown in Figure 6. The y-intercept did not show significant departure from zero.

  1. LOQ
    • The LOQ was established through determining and comparing the signal-to-noise ratio of six injections each of NDMA standards in low concentrations. Table 3 shows the results for 3 ppb NDMA standard.
    • From the linearity studies, this analytical procedure is found to be linear, accurate, and precise at a range of about 3 ppb to 200 ppb standard concentration.

  1. Accuracy, Repeatability, & Intermediate Precision

    • Accuracy of the method for the Test for N-Nitrosodimethylamine (NDMA) of Metformin Hydrochloride, BP/EP was demonstrated by analyzing data obtained from spiked solutions across the specified range of the analytical procedure. The percent recoveries of each individual sample and the average at each concentration level were determined and presented in Table 4. All acceptance criteria were met.
    • Method precision results are presented in Table 5. Results show that the method is precise for the determination of N-Nitrosodimethylamine (NDMA) in Metformin Hydrochloride, BP/EP

  2. Robustness

  • Robustness tests were performed on sample preparation variation, particularly in difference in sonication time. The overall RSD for n=12 spiked samples composing of 6 trials under the method condition and 6 trials under each variation, are at 4.5% for sample preparation with decreased sonication time, while 3.6% with increased sonication time.
  • Robustness tests for instrument parameter variation were also done through change in column oven temperature and change in column serial number. System suitability were met by obtaining a RSD below 10.0% for 6 replicate readings of the standard solution and a S/N ratio of NLT 10 from the LOQ solution.

References


Thank you for contributing to the Nitrosamines Exchange!
By sharing your method, you help the community enhance scientific knowledge and practical approaches for nitrosamines analysis. If you have any questions or need clarifications, please reach out in the discussion thread below.

Big shout-out to Carlo Joson and Therese Margaret Villar from Unilab, Philippines for making the very first analytical-method donation to the Hub! :raised_hands: Your detailed submission sets a high bar and perfectly captures the spirit of open science we’re striving for.

Let’s keep the momentum going—if you’ve developed methods, tweaks, or troubleshooting notes that could help others, hit the “Share” button and add your voice. Every contribution strengthens our collective toolbox and accelerates solutions for nitrosamine challenges worldwide. :rocket:

Thank you, Carlo @cajoson_unilab.ph, Therese @theresevillar

6 Likes

Great work, everyone! Thanks for sharing with the community

3 Likes

I just wanted to say how great it was to see @cajoson_unilab.ph and @theresevillar share their method. That kind of openness really makes a difference — it helps move everyone forward and creates a space where others feel more comfortable doing the same.

I’ll take a closer look at what you posted, and if I have any thoughts or ideas to add, I’ll reach out later.

For now, I just wanted to acknowledge and thank you for putting it out there.

All the best,

Lucas Maciel

4 Likes