Application of USP General Chapter <1220> in the Development of a Procedure for Quantitative NMR

Yang Liu- Scientific Liaison, General Chapters, U.S. Pharmacopeial Convention, Rockville, MD; Toru Miura- Principal Scientist, FUJIFILM Wako Pure Chemical Corporation, Kawagoe, Saitama, Japan; G. Joseph Ray- Adjunct Professor, University of Illinois at Chicago, IL

Introduction

NMR spectroscopy has an almost 100-year history since the first measurement of nuclear magnetic moments in the 1930s. It has historically been employed for structure elucidation; however, recent advances in instrumentation and data processing have prompted renewed interest in its quantitative applications. Quantitative NMR (qNMR) possesses numerous advantages, not the least of which are its universality and the molar proportionality of response. With the recent advent of affordable permanent-magnet benchtop systems, qNMR now exhibits compelling potential in quality control analysis.

For analytical procedures, Analytical Quality by Design (AQbD) principles can provide a basis for implementation of risk-based and systematic framework to ensure procedure fitness for use along the entire analytical procedure life cycle, robust assay as well as life cycle management. The present study applied the procedure life cycle approach to qNMR procedure validation, aligned with USP <1220>Analytical Procedure Life Cycle, that contains concepts such as analytical target profile (ATP)^1-3 and target measurement uncertainty (TMU).

Qualification of NMR Instruments

The qualification of an NMR instrument is a process that is documented according to pre-approved protocols as part of a quality management system and is usually performed in discrete stages: design qualification (DQ), installation qualification (IQ), operational qualification (OQ), and performance qualification (PQ). For example, to verify the operational capability of the instrument on a continuous basis, the OQ process should be repeated annually or as justified by a risk assessment. Some examples of OQ tests include:

• ¹H line shape
• Sensitivity
• Pulse width
• Sample temperature during acquisition
• Spectral uniformity
• Linearity
• Bias assessment

In practice, one of the major uncertainty components in the ATP is the instrument and tools used in a qNMR procedure. To ensure that the accuracy of a qNMR procedure is fit for its intended purpose, bias assessment must be performed before qNMR experiments, including qNMR method development and procedure validation using a life cycle approach. In many cases, bias assessment will also be part of the PQ tests.

To determine systematic bias that usually originates from the measuring instruments and tools, one can apply designed qNMR experimental measurements using two individual certified reference materials (CRMs), where one is used as a calibrant and the other is defined as a test analyte (CRM’). The bias assessment follows the decision rules of expanded uncertainty (U) possibilities. The acceptance criteria are shown in Equation 1.

Where p_CRM’ is labeled purity (or mass fraction) of the CRM’ (the measure and value can be found in the CRM’ certificate), u_CRM’ denotes labeled standard uncertainty of the CRM’ (the value can be found in the CRM’ certificate), p’_CRM’ represents measured purity (or mass fraction) of the CRM’ against the calibrant, tmu’_CRM’ means determined standard uncertainty of the CRM’, and k stands for coverage factor at a defined confidence level. As an analytical technique, a coverage factor k = 2 is most frequently used and corresponds to a confidence level of approximately 95%.

Appropriate limits for bias in ATP can be considered based on these and other factors as appropriate:

• Instrument qualification
• Sample preparation tools (e.g., readability and minimum weight of a balance)
• Quality of calibrant, solvent, and NMR tubes
• Scientist experience
• Data acquisition and processing parameters
• Software qualification
• Calculation tool

Metrological Traceability and Selection of Reference Material

In qNMR applications, CRMs are customarily used as calibrants due to the metrological traceability of measurement results (e.g., purity or potency) obtained for a given product and impurities. Metrological traceability is achieved through calibration. CRM calibration standards should deliver certified purity values, as well as expression of the associated measurement uncertainty, with documented traceability to the International System Units (SI) unit for mass, expressed as a mass fraction. Primary CRMs are available in national metrology institutes.^4,5 CRMs specified for use with qNMR are available from qualified producers that are traceable to the SI through primary CRMs. For establishing metrological traceability, records of the following documents should be kept:

• Evidence of metrological traceability of the chemical mass fraction value for the CRM used as a calibration standard;
• Evidence of calibration of balances used for qNMR sample solution preparation, using weights with certified mass values traceable to the SI unit for mass (kg);
• Measurement uncertainty budget for the qNMR measurement result.

The CRM used in qNMR (or calibrant) is best selected as a stable crystalline solid, which facilitates accurate and precise qNMR measurement. Normally, the calibrant must meet the following criteria:

• The calibrant should be a high-purity chemical with a minimum of impurities;
• The calibrant should be non-hygroscopic, non-volatile, and non-sublimable and should have an absence of interaction with static electricity;
• The calibrant must be soluble in multiple deuterated solvents;
• The internal calibrant signal cannot interfere with the integrated regions of the analyte quantitative signal response and impurities’ response.

Life Cycle Approach for qNMR Validation

For the validation of qNMR analytical procedures, one of two approaches can be adopted: the life cycle approach and the traditional approach (also called traditional minimal approach in ICH Q14). The challenge of the traditional approach is limited to the initial validation experiments, which usually does not incorporate a holistic approach to ensure continuous procedure fitness for use, because quality risk and knowledge management are not extensively explored. Thus, there is an absence of continued procedure performance monitoring and establishment of proper analytical control strategies that can support an analytical procedure life cycle management and promote continuous quality improvement. Under the traditional approach, if there are changes in a qNMR procedure, it is not possible to determine whether the quality of the modified analytical procedure continues to be fit for the intended purpose. The adoption of the analytical procedure life cycle approach helps address this challenging issue, which does not conflict with the traditional validation methodology, but expands the understanding of how a procedure should be validated throughout its life cycle.

In the life cycle approach, the procedure is comprised of the Analytical Target Profile (ATP) which involves three life cycle stages, as shown in Figure 1. The stages are:

• Stage 1 – Procedure Design
• Stage 2 – Analytical Procedure Performance Qualification (APPQ)
• Stage 3 – Ongoing Procedure Performance Verification (OPPV)

The ATP defines the procedure intended purpose and stipulates the acceptance criteria for the reportable results produced by the analytical procedure. The analytical procedure performance is monitored during subsequent use to ensure that it continues to meet the ATP criteria. An example of ATP in an NMR analytical procedure is shown below. The qNMR analytical procedure enables determination of the mass fraction of a sample in the presence of not more than (NMT) 0.5% of impurity(ies) with an expanded measurement uncertainty of NMT 2.0% (coverage factor, k = 2, 95% confidence) within a range of 80.0– 120.0% of the nominal test sample concentration.

Procedure Design

This stage includes knowledge gathering, analytical procedure development, risk assessments, and establishment of replication strategy and initial analytical control strategy (ACS). The purpose of the quality risk management (QRM) process is to assess the identified analytical procedure conditions and appropriate controls on the analytical procedure parameters that will ensure that the analytical procedure meets the ATP criteria. The QRM process includes:

• Identification of risk factors using risk assessment methodologies (e.g., cause-and-effect diagram or Ishikawa diagram), as shown in Figure 2;
• Quantitative assessment of risk factors;
• Identification of method operatable design region (MODR);
• Establishment of replication strategy;
• Establishment of analytical control strategy (ACS).

Specifically, once potential risks have been identified using an Ishikawa diagram (Figure 2), the associated hazards can be evaluated and ranked. This assessment is driven by prior knowledge and scientific expertise, but some risks may warrant a conservative treatment if such knowledge is limited. A heat map can be used to support a quantitative assessment of the risk and provide a visual indication of which variables should be considered, as shown in Table 1.

The method operable design region (MODR) is the multivariate space of analytical procedure parameters that ensure the analytical procedure meets the ATP criteria. In this example (MODR, Figure 3), among the data acquisition parameters illustrated in the heat map, two parameters, the tip angle and the relaxation delay were examined and found to have a high impact level for accuracy and precision of reportable value. The detected purity (mass fraction) became very much lower within a region of a combined high tip angle and short relaxation delays, showing a large bias relative to the mass purity (91%– 98%, mass fraction).

On the other hand, longer delays provided a relatively wide region (purity, 99%-100%, mass fraction) the could provide a mass fraction with negligible bias at all tip angles. The ACS is a planned set of controls to ensure that the NMR analytical procedure continues to meet the ATP criteria. The analytical procedure and preliminary ACS are established in Stage 1, Procedure Design, and the initial development activities are concluded. In this example, a control strategy was determined by establishing workable ranges for critical parameters: (1) the relaxation delay shall be no less than five times the longest T1 value; and (2) the tip angle shall use 90° pulse in order to achieve a mass fraction with minimized bias and dispersion.

Analytical Procedure Performance Qualification (APPQ)

Subsequently, the performance of the analytical procedure is ready to be evaluated. The goal of the entire analytical procedure qualification, including sample preparation, NMR measurement, and data processing, is to confirm that the analytical procedure generates reportable values that meet the ATP criteria and is suitable for its intended purpose in the laboratory. By comparing the measurement uncertainty calculated by implementing an established qNMR analytical procedure with the ATP criteria, the performance of the qNMR analytical procedure will be qualified. As an example of evaluation of measurement uncertainty, the determination of the mass fraction of a butyl p-hydroxybenzoate by ¹ H qNMR with the internal calibration method with 1,4-BTMSB-d₄ as internal calibrant was demonstrated in the present study. The raw data were displayed in Table 2.

Using an approach based on the Guide to the expression of uncertainty in measurement (GUM),³ the combined standard uncertainty of the mass fraction of butyl p-hydroxybenzoate, u_c (P_s ), was calculated by combining the relative standard uncertainties of each proportionality for mass fraction determination described in Equation 2:

P is purity; S denotes peak area; N represents number of resonating hydrogen being measured; M means molar mass; m stands for mass.

The measurement uncertainty budget includes qNMR experiments, weighting best practices, molar mass, number of resonating hydrogens, and purity of internal calibrant. Relative contributions of each component to the combined standard uncertainty were mainly from variability in qNMR experiments and purity of internal calibrant. The standard uncertainties were calculated as:

• qNMR experiments

» replicate qNMR sample solution preparations, 0.052%;

» variability among the selected peak integrals, 0.075%;

» repeatability of qNMR measurement, 0.036%;

• uncertainty associated with the purity of the internal calibrant in the CoA document, 0.250%.

All other factors mentioned above were treated as practically negligible. Accordingly, the combined standard uncertainty that was calculated using the values in the uncertainty budget was 0.268%. The result of purity determination (mass fraction, %) of butyl p-hydroxybenzoate by 1 H qNMR can be labeled as 99.6% ± 0.6% (k = 2). These results show that the specifications of the ATP are met (measurement uncertainty, NMT 2.0%) and the analytical procedure is fit for purpose. A qualified qNMR procedure was obtained.

Ongoing Procedure Performance Verification (OPPV)

The analytical procedure performance is monitored during subsequent use, including validation, verification, and transfer, to ensure that it continues to meet the ATP criteria.

Conclusion

With utilization of certified reference materials, the life cycle approach has been successfully adopted in a measurement of uncertainty budget for the validation of a qNMR procedure. This included the ATP establishment considering a maximum allowable measurement uncertainty, followed by a system bias assessment to ensure that an NMR facility is able to achieve the intended accuracy of a qNMR procedure. These findings showed the benefits, incorporating the life cycle approach and AQbD principles described in USP <1220>into the USP NMR General Chapters <761>and <1761>. This has been recommended by qNMR Expert Panel, in order for a proposed revision of USP <761>and <1761>.

References

1. PF 43 (1) 2016 STIMULI TO THE REVISION PROCESS Proposed New USP General Chapter: The Analytical Procedure Lifecycle <1220>. Available at: https://www.uspnf.com/notices/ stimuli-revision-process-proposed-new-usp-general-chapter-analytical-procedure[1]lifecycle. Accessed January 5, 2023
2. PF 46 (5) 2020 <1220>Analytical Procedure Lifecycle. Available at: https://www.uspnf. com/sites/default/fi les/usp_pdf/EN/USPNF/usp-nf-notices/gc-1220-pre-post-20210924. pdf. Accessed January 5, 2023
3. PF 46 (6) 2020 STIMULI TO THE REVISION PROCESS: Quantitative Nuclear Magnetic Resonance (qNMR), a Metrological Method: Proposed Revisions to the USP General Chapters on NMR <761>and <1761>. Available at: https://www.usp.org/sites/default/ fi les/usp/document/workshops/stimuli-article-qnmr.pdf. Accessed January 5, 2023
4. USA (NIST) – PS1 Benzoic Acid. Available at: https://www.nist.gov/programs-projects/ nist-ps1-primary-standard-quantitative-nmr-benzoic-acid. Accessed January 5, 2023
5. JP (NMIJ) – 4601, 4602, 4603. Available at: https://unit.aist.go.jp/nmij/english/refmate/ crm/46.html. Accessed January 5, 2023

Author Biographies

Yang Liu is a Scientific Liaison in General Chapters at USP, and he coordinates the proposed revisions of USP NMR General Chapters <761><761> and <1761>

Toru Miura is a member of the qNMR Expert Panel at USP and is highly experienced in the development of Certified Reference Materials.

Joe Ray is the chair of qNMR Expert Panel at USP. His interests include the use of multinuclear, multidimensional, high-resolution, and solid-state NMR techniques to characterize organics, fuels, polyesters, polyolefi ns, catalysts, zeolites, and biotech-related materials.

Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special
offers from American Pharmaceutical Review delivered to your inbox!
Sign up now!