A procedure is described for preparing polydisperse polymer standards to validate the accuracy of any size-exclusion chromatography (SEC) method for aqueous or organic mobile phases. The prepared standard reflects the molecular weight distribution and detector response of an authentic sample. It is then analyzed by the SEC method to obtain number- and weight-average molecular weights. Percent accuracy of the SEC method is then calculated by comparing experimental results with actual data. This approach can be used for all calibration procedures with the exception of online light scattering and viscometric detection.
Accuracy validation in size-exclusion chromatography (SEC) depends on the type of analysis being performed. If it is used for the quantitation or assay of a single main or minor peak, validation protocol is exactly the same as for high performance liquid chromatography (HPLC), in which a reference standard is analyzed as if it were the sample (1–3). With this approach, the reference standard can be prepared directly in either the mobile phase, in the blank sample matrix, or in the sample itself by the method of standard addition (4,5). However, when SEC is used for determining the molecular weight distribution (MWD) and average molecular weights (MWs) of a sample, accuracy validation becomes more complicated.
Equations used to predict the accuracy of SEC columns have been developed by Yau and colleagues (6,7). These equations are not used for method validation, but rather to assess column performance based on peak broadening and the slope of the SEC calibration (log MW versus elution volume). Many other studies on calibration plots have been concerned with either peak broadening or local polydispersity corrections (8,9), or reproducibility of data points (10), not the accuracy of the method.
Many round-robin SEC studies have been reported with and without orthogonal validation (11–14). These approaches are used for inter- and intralaboratory reproducibility and accuracy measurements to appraise the potential performance of a given SEC method. However, none of the published work has addressed the problems of SEC accuracy validation.
Since accuracy is defined as the difference between the true or expected value and the experimentally determined result, we need a series of different molecular weight reference standards of well-characterized polymers that are chemically and structurally the same as the samples. Unfortunately, there are relatively few reference standards commercially available (see Table I) and most laboratories lack resources to prepare their own. In light of this situation, this article describes an alternative procedure consisting of preparing polydisperse reference standards from monodisperse standards that are dissimilar to the sample, but nonetheless mimic the MWD and detector response of actual samples. With this approach, the accuracy of any SEC procedure can be validated.
Table I: Commercially available polymers suitable for accuracy validation
To minimize errors, the proposed reference standard consists of a two-component mixture of monodisperse standards with certified MWs and known molecular weight distributions. Uncertainty is further reduced by using equal weights of standards; this approach precludes the need of moisture or purity corrections as long as detectable, low-MW impurities are separated from the polymer envelope.
Reference standards should be subjected to the same errors experienced by the sample, which implies
1. covering nearly the same MW calibration range (Figure 1)
2. having closely matched signal-to-noise ratios
3. duplicating sample peak broadening.
To satisfy the first criterion, the reference mixture should have approximately the same number-average M n and weight-average M w molecular weights as the sample based on the same calibration curve. In other words, we use the same slope of the calibration plot and similar MW range of the sample. The monodisperse standards used in the mixture are the same ones used to generate the calibration plot. Most importantly, sample MW averages can be either actual or apparent values.
Figure 1: A typical SEC calibration where V0 and Vt are the interstitial volume and total permeation volume, respectively. The ordinate can be the MW of primary or secondary monodisperse calibrants; computer-generated MWs of a broad-MW, primary calibrant; or the hydrodynamic volume of a secondary calibrant (see Table II). The red line encompasses the MW range of samples, and the three blue lines represent different reference mixtures of standards.
For the second criterion, the peak area of the injected mixtures should closely correspond to the sample peak area. Finally, the third criterion is satisfied by injecting the same volume and operating at conditions specified by the method. Needless to say, identical instrumentation and column sets must be used to validate the accuracy of the method.
This accuracy validation procedure is applicable for all SEC methods as summarized in Table II. As formulated in this article, the procedure cannot be applied to methods using online molecular-weight-sensitive detection methods such as light scattering, viscometry, and mass spectrometry, which require a different approach.
Table II: Conventional SEC calibration methods that are suitable for accuracy validation, as described in this paper*
Our goal is to prepare a reference standard that closely matches the actual or apparent sample MW values. This is accomplished by mixing together equal amounts of two monodisperse calibrants of known MW that cover the MW range of samples.
We start with the definitions of M n and M w:
Click here to view full-size graphic
where M i is the MW and w i is the corresponding weight of the ith component. For a two-component mixture, let M 1 and M 2 be the MWs of monodisperse standards with corresponding weights w 1 and w 2, then
Click here to view full-size graphic
If the amounts of the two standards are set equal to one another, equations 3 and 4 become
Click here to view full-size graphic
with a polydispersity of
Click here to view full-size graphic
The solutions to simultaneous equations 5 and 6 are simply
Click here to view full-size graphic
where M 1 and M 2 are the MWs of the monodisperse standards needed to give the sought after M n and M w values.
To ensure that the prepared standard has a detector response similar to that of the sample, the weight (w) of each standard in the mixture is
Click here to view full-size graphic
where w s is the typical sample weight specified by the method and R f is the detector response factor given by
Click here to view full-size graphic
in which dn/dc is the specific refractive index of the sample (subscript s) and standard (subscript std). If these values are not available or if a detector other than a refractometer is used, peak area A, normalized with respect to injected concentrations, can be used instead. A compilation of dn/dc values can be found in the literature (12,15).
The injection concentration c std of the prepared standard is
Click here to view full-size graphic
where v is the volume of solvent, typically the mobile phase, specified by the method for sample preparation. Note: If c std is greater than the injected sample concentration, the user must confirm that the concentration of the reference mixture is below the critical polymer concentration to avoid macromolecular crowding (1) or viscosity effects (1). This is done by injecting the standard at several lower concentrations to make certain that the elution volume is not a function of concentration.
Please note that primary standards are monodisperse samples of known MW used as calibrants. Secondary standards are monodisperse calibrants of known MW that are chemically different from samples.
1. Generate SEC calibration curve (log M versus V r) using either primary or secondary monodisperse standards (see Table I), or a broad-MW primary standard with defined average MWs (1).
2. Obtain M n and M w values of at least three representative samples using the SEC calibration specified by the method. Take the average of these results.
3. With equations 8 and 9, estimate M 1 and M 2 needed to generate the experimental M n and M w values of an average representative sample.
4. Select two monodisperse standards that most closely match M 1 and M 2 obtained in step 3 (see Table I). The polydispersity of the mixture should be equal to or greater than the average value calculated in step 2.
5. Recalculate to obtain the true number-average (M n)t and weight-average (M w)t of the mixture with equations 5 and 6, where subscript t represents the true or expected values.
6. Formulate two additional reference standards with MW averages greater and less than those in step 5.
7. Determine the response factor of the standard according to equation 11.
8. Weigh out the prescribed amounts of monodisperse standards and dilute to volume following equations 10 and 12.
9. Analyze the three prepared standards with triplicate injections (4–6). Determine (M n)exp and (M w)exp values using the calibration procedure specified for samples.
10. Calculate an average (M n)exp and (M w)exp for each reference mixtures.
11. Address accuracy by using equations 13a and 13b for absolute errors (M n)AE and (M w)AE and equations 14a and 14b for relative errors % (M n)RE and % (M w)RE (16):
Click here to view full-size graphic
A procedure was proposed for validating the accuracy of any SEC method that uses either primary or secondary monodisperse calibrants, including broad-MW and universal calibration. As written, the equations are not applicable for use with online molecular-weight-sensitive detection methods such as light scattering, viscometry, or mass spectrometry. This accuracy validation approach can be applied to the SEC analysis of any polydisperse sample. The method is used when representative well-characterized samples (that is, with known MW averages) are unavailable.
The preparation of a polymer standard is described that closely reflects the MWD and detector response factor of samples being analyzed. To ensure the highest accuracy, the standard is composed of just two secondary, monodisperse standards with known MWs. The two monodisperse standards are added together in equal amounts, eliminating errors related to their purity, typically moisture content.
The two-component mixture of standards will typically give a bimodal distribution, unless the MW values of the two standards are close to one another. As such, it approximates, not duplicates the actual MWD of samples. More-complicated MWDs can be prepared by adding together different amounts of monodisperse standards. To comply with accepted statistical analysis, we recommend using three different standard mixtures and triplicate injections (1,2,4,5).
(1) G. Kateman and L. Buydens, Quality Control in Analytical Chemistry, 2nd ed. (Wiley Interscience, New York, New York, 1993).
(2) L. Huber, Validation and Qualification in Analytical Laboratories (Interpharm Press, Buffalo Grove, Illinois, 1999).
(3) S.S. Doss, N.P. Bhatt, and G. Jayaramen, J. Chromatogr. B 1060, 255–261 (2017).
(4) L.R. Snyder, J.J. Kirkland, and J.L. Glajch, Practical HPLC Method Development (Wiley-Interscience, New York, New York, 1997).
(5) L.R. Snyder, J.J. Kirkland and J.W. Dolan, Introduction to Modern Liquid Chromatography, 3rd ed. (Wiley-Interscience, New York, New York, 2009).
(6) W.W. Yau, J.J. Kirkland, D.D. Bly, and H.J. Stoklosa, J. Chromatogr. 125, 219–230 (1976).
(7) A.M. Striegel, W.W. Yau, J.J. Kirkland, and D.D. Bly, Modern Size Exclusion Liquid Chromatography (Wiley, New York, New York, 2009).
(8) T. Provder, Ed., Detection and Data Analysis in Size Exclusion Chromatography, ACS Symposium Ser. 352 (American Chemical Society, Washington, D.C., 1987).
(9) T.H. Mourey and S.T. Balke, J. Appl. Polym. Sci. 70, 831–835 (1998).
(10) S.T. Balke, Quantitative Column Liquid Chromatography (Elsevier, Amsterdam, The Netherlands, 1984).
(11) S. Mori, Int. J. Polym. Anal. Charact. 4, 531–546 (1998).
(12) S. Mori and H.G. Barth, Size Exclusion Chromatography (Springer Verlag, Berlin, Germany, 1999).
(13) U. Just., S. Weidner, P. Kiltz, and T. Hofe, Int. J. Polym. Anal. Charact. 10, 225–243 (2005).
(14) A. Ritter, M. Schmid, and S. Affolter, Polym. Test. 29, 945–952 (2010).
(15) J. Brandrup, E.H. Immergut, and E.A. Gurlke, Eds., Polymer Handbook, 4th ed. (Wiley-Interscience, New York, New York, 2003).
(16) D.A. Skoog, F.J. Holler, and S.R. Crouch, Instrumental Analysis (CENGATE Learning, Andover, Massachusetts, 2002).
ABOUT THE AUTHOR
Howard G. Barth is with Analytical Chemistry Consultants, Ltd., in Wilmington, Delaware. Direct correspondence to: [email protected]
In-house experts can help select the right systems and suppliers, making validation and compliance easy, says Siegfried Schmitt, principal consultant at PAREXEL.
Q. We are planning to upgrade several of our automated systems in production and in the laboratories. These upgrades are necessary so that we can implement functionality like audit trails, which are now required to achieve data integrity compliance. We contacted suitable vendors, and some have now offered to sell to us fully 21 Code of Federal Regulations (CFR) Part 11 and data integrity-compliant software packages. Such a package seems like a very good deal, but it is not offered by the majority of suppliers. Can you give some insight on how other companies address this situation?
A. You are in a very typical situation where not all your automated systems are of a technical standard that make compliance with the applicable regulations possible, unless you upgrade or replace certain systems. It is an unfortunate fact that there are still vendors out there who make misleading claims, either out of ignorance, or worse, knowingly. The only party that can legally commission and operate a computerized system in a fully compliant manner is the system owner (i.e., someone within your organization). Only you know how you are going to use the specific system and for what purpose. No vendor can do this for you. Therefore, automated system suppliers can merely offer to sell you systems that are designed and built in a manner that allows you, the customer, to operate them in a compliant manner (e.g., complying with 21 CFR Part 11 or other regulatory requirements).
Let me give you an example to clarify this: You may purchase a system upgrade that provides audit trail functionality. Although the vendor gives you an audit trail as you requested, you may choose not to activate it (perhaps because it slows the system down too much). Now you may be in a non-compliant situation. Or, you decide to activate the audit trail, but upon review you find that it is not in human readable form, or that it only captures a fraction of the transaction, or that the amount of data in the audit trail is so overwhelming that it becomes unmanageable.
Savvy companies have in-house experts with a sound understanding of the regulations covering automated systems, how to perform computerized systems validation, and how to optimally harness the vendors’ expertise. These experts will put together the user requirement specifications (URS) for the various systems. The URS is the document that will steer how the system (or system upgrade) will help you to operate in compliance with the regulations (i.e., what it takes to make sure that data are trustworthy and your system is fully validated). In the URS, companies will specify what they expect from the audit trail (e.g., it must be human readable, sortable, exportable, searchable, etc.).
The URS also forms the basis for the testing requirements, namely the testing by the users. Users may be quality unit personnel who need to verify that on an analytical instrument the series of injections for an analysis tally with the method, or that there were no rogue injections. Only these people will know what they are looking for and how they want to perform their review. Your system vendors are now tasked with providing you with a system that meets your needs, and not just a ‘one-size-fits-all’ solution.
Don’t be lulled into a false sense of security by sales promises; instead, make sure you have experts at hand who can help you select the systems and suppliers who best meet your needs. Once you do this, you will find that validation and compliance even with the most demanding regulations become not only possible, but exciting.
Vol. 42, No. 3
When referring to this article, please cite it as S. Schmitt, “Computerized Systems Validation,” Pharmaceutical Technology 42 (3) 2018.
The main regulatory standard for ensuring pharmaceutical quality is 21 Code of Federal Regulations (CFR) Parts 210 and 211, collectively referred to as the current good manufacturing practice (CGMPs) regulation for human pharmaceuticals. It’s not only the CGMP regulations that matter; the approaches biopharmaceutical companies take to interpret and embrace these regulations are of equal, if not more, importance. CGMPs place emphasis on product quality and compliance with the regulations. So how do companies embrace, and even embody, quality and compliance with CGMPs? One way is to operate under a quality culture.
A company should have a pharmaceutical quality system as described in FDA Guidance for Industry, Q10 Pharmaceutical Quality System (1). A quality culture is created when managers believe a company has a duty to create a mutually beneficial relationship between itself, its employees, and its customers. Culture is the shared beliefs, values, attitudes, and behavior patterns that characterize a family, a community, or an organization. A healthy organizational culture is rooted in the understanding that quality is good for the company and its customers. Thus, its existence is a driving force behind how employees act and behave regardless of level, title, or decision-making authority. A quality culture begins with company leaders who believe in the necessity of serving customers in order for their organizations to succeed. In defining a set of desirable values for a corporate culture, some of the primary core values should include integrity, customer focus, and people.
FDA conducts several types of inspections to help protect consumers from unsafe products: pre-approval inspection, routine inspections of a registered facility, and “for-cause” inspections. After FDA completes an inspection, company management may receive an FDA Form 483 (2) when an investigator(s) has observed any conditions that, in their judgment, may constitute violations of the Food Drug and Cosmetic (FD&C) Act, related acts, and applicable sections of 21 CFR 210 and 211. Observations are made when, in the investigator’s judgment, conditions or practices observed would indicate that product has been adulterated or is being prepared, packed, or held under conditions whereby it may become adulterated or rendered injurious to health. FDA Form 483 notifies the company’s management of objectionable conditions. Companies respond to the 483 in writing with their corrective action plan and implement schedule. The 483 is closed when the company receives their establishment inspection report (EIR). Unfortunately, there are companies that either do not follow through on their commitments or they do so too slowly. If circumstances merit, FDA can choose to escalate the situation by serving the company with a warning letter.
Typically, FDA gives individuals and companies an opportunity to take voluntary and prompt corrective action before the agency initiates an enforcement action. A warning letter is FDA’s principal means of notifying regulated companies of violations and achieving prompt voluntary correction. The following factors are used to determine whether to issue a warning letter (3):
- The firm’s compliance history (e.g., a history of serious violations, or repeated failure to prevent the recurrence of violations)
- The nature of the violation (e.g., a violation that the firm was aware of [was evident or discovered] but failed to correct
- The risk associated with the product and the impact of the violations on such risk
- The overall adequacy of the firm’s corrective action and whether the corrective action addresses the specific violations, related violations, related products or facilities, and contains provisions for monitoring and review to ensure effectiveness and prevent recurrence
- Whether documentation of the corrective action was provided to enable the agency to undertake an informed evaluation
- Whether the timeframe for the corrective action is appropriate and whether actual progress has been made in accordance with the timeframe.
When a company repeatedly violates cGMP requirements, FDA can force it, through legal channels, to make specific changes. Under this severe form of escalation by FDA, the objective is no longer a discussion about responses to 483s or warning letter observations, it’s about a forced company make-over. This process, known as consent decree, exposes all broken systems within a company. FDA does not care about any efforts and expenses undertaken by the company to redesign and implement a robust quality management system; a common element of a consent decree is demonstrating sustainability.
Sustainability is the capability of an organization to know when it is veering off course and to make the correct decisions, take the appropriate actions, and maintain a state of control without external intervention. Sustainability embraces the core quality culture values and expected behaviors of integrity, empowerment, and accountability. A consent decree mandates a series of annual inspections performed by a third-party to monitor sustainability. FDA recommends companies hire external experts and invest time and money to inspect and certify compliance, often for many years.
Operating under a consent decree is a dire situation for the company and one where there is no certain predictability of the outcome. To understand the full magnitude of a consent decree’s impact, one needs to take into consideration the many and various modes in which the negative consequences can be realized (see Figure 1).
The cost of non-compliance
This discussion focuses on three ways to look at the costs of non-compliance: quantifiable costs, difficult-to-quantify costs, and invisible costs with hidden impacts.
Quantifiable costs. Quantifiable costs are those that can have a value assigned to them with a reasonable degree of accuracy. Quantifiable costs are usually more obvious, but they manifest in several ways as shown in Figure 1.
Often termination and replacement of specified employees with retraining of remaining legacy employees is an immediate step associated with a consent decree. This company action is meant to set the tone that the old way of doing business will not be tolerated anymore. A major requisite term of a consent decree is for the company to retain, at its own expense, an independent third-party for ongoing certification and oversight of the implementation of agreed to corrective action. Another requirement may be a commitment to have every released batch certified to be CGMP.
FDA can levy significant fines for not meeting action dates: Examples include $15,000 per day for missed dates, royalty payments up to 24.6% (4) per product not revalidated on time, and costs for FDA inspections. Furthermore, the US Treasury can garnish profit from sales through fines. Such was the case for Wyeth in October 2000. Wyeth agreed to a consent decree regarding its Marietta, PA and Pearl River, NY. Inspections in 1995, 1996, and 1998 found several GMP deviations and resulted in warning letters (5).
In 2013, one of the largest drug safety settlements occurred when generic-drug manufacturer, Ranbaxy USA Inc., a subsidiary of Indian generic-pharmaceutical manufacturer Ranbaxy Laboratories Limited, pleaded guilty to felony charges relating to the manufacture and distribution of certain adulterated drugs made at two of Ranbaxy’s manufacturing facilities in India (3). Ranbaxy paid a criminal fine and forfeiture totaling $150 million and settled civil claims under the False Claims Act and related state laws for $350 million.
Entering into consent decree can also expose a company to civil penalties. Depending on the circumstances, shareholders, patients, and sometimes even company employees may be able to sue for damages. One such law suit arose from an unsuccessful effort by Baxter International to fix problems with its Colleague Infusion Pump. Westmoreland County Employee Retirement System (Westmoreland) alleged that Baxter’s directors and officers breached their fiduciary duties by “consciously disregarding their responsibility to bring Baxter into compliance with the 2006 consent decree and related health and safety laws” (6).
The breach was alleged to have caused Baxter to lose more than $550 million after FDA mandated a recall of the Colleague Infusion Pumps in 2010. Baxter invested time and money trying to fix the pumps, but the problems persisted into early 2010. FDA invoked its power under the 2006 consent decree by ordering Baxter to recall and destroy all Colleague Infusion Pumps then in use in the US, reimburse customers for the value of the recalled device, and to assist in finding replacement devices for those customers. The company’s stock price fell by more than 4% after the announcement and the company later recorded a pre-tax charge of $588 million to account for the estimated costs of the recall.
A company under consent decree loses future revenue in different ways. One way, which is particularly difficult to quantitate, is the price due to “lost innovation”. The company is redirecting revenue into compliance, oversight, and remediation instead of reinvesting it into research and development of new products.
When a drug manufacturer subject to enforcement action is the sole supplier of an important medicine, drug shortages become a concern. Short-term solutions may include doctors substituting medications that may have lesser efficacy. Pharmaceutical companies with potential manufacturing capability may be incentivized by FDA to manufacture identical or equivalent drug products; however, medium-to-long lead times can delay product availability. When supplies dwindle, patients must pay higher prices for the same drug.
One illustration of this scenario is the FDA consent decree with Genzyme in 2010 (7) regarding repeated manufacturing issues at the company’s Allston, MA facility, which included an up-front disgorgement of past profits of $175 million and the requirement to move fill/finish operations out of the Allston plant by a specific date. Had Genzyme not met those deadlines, FDA could have required the company to disgorge 18.5% of revenue for the affected products.
Difficult-to-quantify costs. Other financial ramifications from consent decrees may be difficult to quantify. For example, some employees may lose their confidence in the commitment or ability of the company’s CEO and other executives to manage the situation. Employee attrition is expected, but if the exodus includes long-tenured employees the company may be drained of valuable knowledge and talent. The longer the situation exists, the more difficult it becomes to retain the best employees.
A company under a consent decree is subject to reputation damage, which can be accentuated by the ease of publicly available negative media coverage about how the company operated. Negative media coverage can result in public fear. In the Ranbaxy case, an import ban had been in place since 2008 for 30 drugs manufactured at two of its Indian manufacturing plants resulting from alleged data falsification (7). This type of information can result in public concerns that medicines could be adulterated.
Lost revenue due to the company’s inability to sell product can also be an issue. Companies under consent decree experience long delays in releasing product due to the intensive oversight required, when a batch fails release testing, and when a product is recalled. The cost for the logistics of product recall and destruction can be high, especially if the API is expensive or has a long lead time.
Another potential cost is the inability to sell a product. Group purchasing organizations (GPOs) use the power of collective purchasing to buy pharmaceuticals at discounts (8). An underlying premise is continuity of supply; if a pharmaceutical manufacturer is unable to deliver the product contracted, the GPO must obtain replacement product from the open market, often at a higher price. GPOs wary of a potential inability of the company to supply its products may consider other sources.
In one of the largest settlements to date, Ranbaxy pled guilty to felony charges (9). The charges were manufacturing and distributing adulterated drugs made at the Indian manufacturing sites. The criminal fine and forfeiture totaled $150 million and another $350 million to settle civil claims under the False Claims Act.
Invisible costs and hidden impacts. If the cost to bring a facility under consent decree into CGMP compliance is determined to exceed the company’s financial ability or it makes no financial sense to continue operations, the company’s management may decide to close the facility. Shuttering a facility, can have devastating impact on the local community, particularly in rural areas where the manufacturer is a primary employer. The community’s tax base will also suffer.
The company can lose its competitive edge because its busy focusing inside rather than externally. This can be a competitive advantage to your closest competition looking for a way in which to leverage your situation for their benefit. The potential opens for the company to become alienated and therefore lose rank compared to its direct competitors.
A company can also be denied approval of new drug while non-compliance exists (4). Because of its consent decree, approval of two new drugs by Eli Lilly was delayed. This was the result of CGMP issues being found during pre-approval inspection in the fall of 2001, which was six years after its consent decree.
The collective cost of remediation of non-compliance far exceeds the cost to remain in compliance. The number of consent decrees issued per year has remained consistent during the past decade. However, companies have found it difficult to extricate themselves from the agreements. As a result, the number of companies under consent decrees at any given time has increased. Generally, it takes many years for a company to demonstrate that it is in full compliance. Only one company that has received a decree in the past 10 years has met all requirements and had the decree lifted.
Considering the fines and the payments to the third-party consultants, the costs associated with a consent decree can become very high and have a significant effect on a company’s profit. It is estimated that the costs incurred by Warner-Lambert from 1993 to 2002 for a 1993 consent decree–in terms of product terminations, delays in approvals, and bringing facilities and systems into compliance–was nearly $1 billion (4). The Warner Lambert fine was only $10 million, a small percentage of the total cost. Schering-Plough’s initial fine was $500 million. Abbott Laboratories has spent almost $1 billion resulting from a consent decree issued in 1999, including a fine of approximately $100 million.
1. FDA, Q10 Pharmaceutical Quality System, Guidance for Industry (ICH, April 2009).
2. FDA, FDA Form 483 Frequently Asked Questions.
3. FDA, Warning Letter Procedures, accessed Sept. 29, 2017.
4. S. Chrai and M. Burd, BioPharm International 17 (6) (June 2004).
5. Wyeth, “Wyeth-Ayerst Labs Enters into Consent Decree with FDA,” Press Release, Oct. 2, 2000.
6. R. Kreisman, “Shareholder’s Right to Sue for Breach of Fiduciary Duty by Baxter International Directors and Officers Allowed,” Kreisman Law Offices, Chicago Injury Lawyer Blog, September 24, 2013.
7. Sanofi Genzyme, “Genzyme Announces Final Terms of FDA Consent Decree,” Press Release, May 24, 2010.
8. AmerisourceBergen “The Evolution of GPO Contracting,” Knowledgedriven.com, Nov 25, 2014, accessed Sept. 29, 2017.
9. US Department of Justice, “Generic Drug Manufacturer Ranbaxy Pleads Guilty and Agrees to Pay $500 Million to Resolve False Claims Allegations, CGMP Violations and False Statements to the FDA,” Press Release, May 13, 2013.
When referring to this article, please cite it as S. Ayd, “Managing the Cost of Non-Compliance,” Pharmaceutical Technology 41 (11) 2017.
Regulatory guidance documents are written by committees, resulting in statements that are both exact and generic. Meeting regulatory requirements involves not only interpreting these documents correctly but also addressing their omissions. This article provides practical guidance on issues that are not thoroughly covered by current guidance documents regarding validation of analytical methods for biopharmaceuticals.
Four of the existing regulatory guidance documents on methods validation state, “Methods validation is the process of demonstrating that analytical procedures are suitable for their intended use.”1–4 We have all read, and likely used, this phrase many times when summarizing method-validation results. According to Muire-Sluis, development scientists often point out that “validated methods may not be valid.”5 The question therefore arises, what exactly makes a validated method valid? According to the Center for Biological Evaluation and Research (CBER), “the acceptability of analytical data corresponds directly to the criteria used to validate the method.”4
We can generate evidence for the validity of analytical data in the formal method-validation program where all critical parameters are extensively tested under a detailed protocol that includes scientifically justified and logical step-by-step experimental approaches. All planned data sets must fall within pre-established protocol acceptance criteria limits. These criteria should be derived from and justified in relation to historical data and product specifications. Once evidence for all critical elements is provided, the validated method will become the official, licensed procedure for that particular product and process step, and it will then support production and product release. The relationship between “valid” or “suitable and validated” is often overlooked, but there is a high price when “validated” test systems are simply inappropriate.
Incentives to replace existing licensed test procedures may come from regulatory agencies, or they could be motivated by potential cost savings, ease of use (automation), and the opportunity to generate more accurate and reliable results.
The International Conference on Harmonization (ICH)’s Q2(R1),1 should be used for basic guidance. However, following just these guidelines will not necessarily produce a “valid” method and may not provide sufficient evidence that this method is suitable for product release. The intent of USP 30 <1225> is to provide guidance only on validation requirements for test methods for inclusion into USP with the expectation that validated USP methods still require verification from users.6–7 The US Food and Drug Administration (FDA) and European Medicines Agency (EMEA) provide guidance on some of the scientific issues that are not covered by Q2(R1).
Process Map. A process map showing the recommended steps for the selection, development, validation, and potential transfer of analytical methods, illustrating all proposed functional responsibilities was developed. Frequently, larger companies have separate functional units for method development, validation, and testing. The process flow in Figure 1 describes an ideal sequence of steps for better analytical method validation (AMV).
The rigorous standards suggested here are ideal but they are not necessarily required or followed during method development. Methods can be developed without strict adherence to good manufacturing practices (GMP) regulations if adequate documentation systems are used.