Assessment of the Regulatory Methods for the Comparison of Highly Variable Dissolution Profiles

Development of Quantitative Comparative Approaches to Support Complex Generic Drug Development

On October 27-28, 2022, the US Food and Drug Administration (FDA) and the Center for Research on Complex Generics (CRCG) hosted a virtual public workshop titled "Best Practices for Utilizing Modeling Approaches to Support Generic Product Development." This report summarizes the presentations and panel discussions for a session titled "Development of Quantitative Comparative Approaches to Support Complex Generic Drug Development." This session featured speakers and panelists from both the generic industry and the FDA who described applications of advanced quantitative approaches for generic drug development and regulatory assessment within three main topics of interest: (1) API sameness assessment for complex generics, (2) particle size distribution assessment, and (3) dissolution profile similarity comparison. The key takeaways were that the analysis of complex data poses significant challenges to the application of conventional statistical bioequivalence methods, and there are various opportunities for using data analytics approaches for developing and applying suitable equivalence assessment method.

Dissolution Profiles Comparison Using Conventional and Bias Corrected and Accelerated f2 Bootstrap Approaches with Different Software's: Impact of Variability, Sample Size and Number of Bootstraps

Dissolution profiles comparison is an important element in order to support biowaivers, scale-up and post approval changes and site transfers. Highly variable dissolution can possess significant challenges for comparison and f2 bootstrap approach can be utilized in such cases. However, availability of different types of f2 and confidence intervals (CI) methods indicates necessity to understand each type of calculation thoroughly. Among all approaches, bias corrected and accelerated (BCa) can be an attractive choice as it corrects the bias and skewness of the distribution. In this manuscript, we have performed comparison of highly variable dissolution data using various software's by adopting percentile and BCa CI approaches. Diverse data with different variability's, number of samples and bootstraps were evaluated with JMP, DDSolver, R-software, SAS and PhEq. While all software's yielded similar observed f2 values, differences in lower percentile CI was observed. BCa with R-software and JMP provided superior lower percentile as compared to other computations. Expected f2 recommended by EMA has resulted as stringent criteria as compared to estimated f2. No impact of number of bootstraps on similarity analysis was observed whereas number of samples increased chance of acceptance. Variability has impacted similarity outcome with estimated f2 but chance of acceptance enhanced with BCa approach. Further, freely available R-software can be of attractive choice due to computation of various types of f2, percentile and BCa intervals. Overall, this work can enable regulatory submissions to enhance probability of similarity through appropriate selection of number of samples, technique based on variability of dissolution data.

Discriminative Dissolution Method Using the Open-Loop Configuration of the USP IV Apparatus to Compare Dissolution Profiles of Metoprolol Tartrate Immediate-Release Tablets: Use of Kinetic Parameters

The use of the USP IV apparatus (flow-through cell) has gained acceptance in recent years due to its versatility and ability to discriminate due to its hydrodynamic conditions. Therefore, the objective of the present study was to develop a discriminative dissolution method in the USP IV apparatus using the open-loop configuration, as well as to propose a method to compare non-cumulative dissolution profiles obtained in the open-loop configuration considering kinetic parameters and validate its predictive power through its comparison with independent and dependent methods using five commercial immediate-release tablet drugs (one reference drug and four generic drugs) of metoprolol tartrate as a model drug. The comparison of the non-accumulated dissolution profiles consisted of determining the geometric ratio of Cmax, AUC0∞, AUC0Cmax, and Tmax (kinetic parameters) of the generic/reference drugs, whereby generic drugs "C" and "D" presented the highest probability of similarity since their 90% confidence intervals were included, or they were very close to the acceptance interval (80.00-125.00%). These results were consistent with the f2, bootstrap f2, and dissolution efficiency approaches (independent models). In conclusion, the proposed comparison method can be an important tool to establish similarity in dissolution profiles and to facilitate the development/selection of new formulations and positively ensure bioequivalence in clinical studies.

Bootstrap Statistics and Its Application in Disintegration and Dissolution Data Analysis

Disintegration time (DT) and rate of drug dissolution in different media are among the most widely studied crucial parameters for various types of drug products. In the ever-evolving landscape of generic formulation development, dissolution comparison of reference and test products is the major reliable in vitro method of establishing product similarity. This is one of the most widely accepted methods of proving pharma equivalency between two drug products. A well-studied match between the disintegration and dissolution profile of the test and reference formulations can ensure in vitro product similarity. Various statistical approaches have been employed to establish product performance similarity; among them, the similarity factor (f2) calculation based approach is the most widely accepted and explored method to date. However, the f2 statistics fail to predict the similarity of batches with unit-to-unit variability. Bootstrap statistical analysis of dissolution data between the test and reference products was introduced to overcome the problems associated with batches with unit variability. Bootstrap can also be applied to extract statistically significant results by treating a series of data from different batches, which can further help to understand the trend. The current review depicts different case study based approaches to show the applications of bootstrap statistics in disintegration and dissolution similarity evaluation for both conventional and additively manufactured solid dosage forms. It is concluded that bootstrap statistics can be a very promising and reliable data analytical tool for establishing in vitro product similarity for both conventional and additively manufactured formulations with a high level of intraunit variability.

Estimators and confidence intervals of f2 using bootstrap methodology for the comparison of dissolution profiles

BACKGROUND AND OBJECTIVES

The most widely used method to compare dissolution profiles is the similarity factor f₂. When this method is not applicable, the confidence interval of f₂ using bootstrap methodology has been recommended instead. As neither details of the estimator nor the types of confidence intervals are described in the guidelines, the suitability of five estimators and fourteen types of confidence intervals were investigated in this study by simulation.

METHODS

One million individual dissolution profiles were simulated for the reference and test populations with predefined target population f₂ values, where random samples of different sizes were drawn without replacement. From each pair of random samples, five f₂ estimators were calculated, and fourteen types of confidence intervals were obtained using 5000 bootstrap samples. The whole process was repeated 10000 times and the percentage of the similarity conclusions was measured. In addition, the uncertainty associated with the current practice of using f^₂ point estimate alone for the statistical inference was evaluated.

RESULTS

When combined with different types of confidence intervals, the estimated f₂ (f^₂), the bias-corrected f₂ (f^_2,bc), and the variance- and bias-corrected f₂ (f^_2,vcbc) are not suitable estimators due to higher-than-acceptable type I errors. The estimator f^_2,exp, calculated based on the mathematical expectation of f^₂, and f^_2,vcexp, the variance-corrected f^_2,exp, showed acceptable type I errors when combined with any of the ten percentile intervals. However, they have the drawback of low power, which might be addressed by increasing the sample size. To properly control the type I error, samples with at least 12 units should be used.

CONCLUSION

The best combinations of estimator and type of confidence interval are f^_2,exp and f^_2,vcexp combined with any of the ten types of percentile intervals. When the sample f₂ value is close to 50, the use of the confidence interval of f₂ is recommended even when the variability of the dissolution profiles is low and the prerequisites defined in the regulatory guidelines for using the conventional f₂ method are fulfilled in order to control the type I error rate.

Comparison of free software platforms for the calculation of the 90% confidence interval of f2 similarity factor by bootstrap analysis

INTRODUCTION

The calculation of the 90% confidence interval of f₂ based on the bootstrap methodology has been proposed and accepted by the main regulatory authorities when the dissolution data shows excessive variability. Different free software platforms allow the calculation of the 90% CI of f₂ by means of bootstrapping. Their use in regulatory submissions is growing, but divergent results have been observed between the available software platforms. Therefore, the objective of this work is to analyze the characteristics of these software platforms and evaluate their results.

METHODS AND MATERIALS

Highly variable in vitro dissolution data from two products were selected. Three different similarity factors, f₂, E(f₂) and bc-f₂, and their corresponding 90% confidence intervals were calculated with three free software platforms, Pheq_bootstrap, Bootf2bca and DDSolver, and computed by four different approaches, normal approximation, bootstrap-t-CI, percentile CI, and bias corrected and accelerated CI.

RESULTS

All three platforms report the same f₂ value, 49.66 upon comparison of products 1 and 3 and 54.87 for products 2 and 3 (no truncation rule). Bootf2bca and Pheq_bootstrap provided the same f₂ and E(f₂) also under other truncation rules (EMA or FDA), which are not implemented in DDSolver. Pheq_bootstrap allowed the calculation of bc-f₂. The conclusion of similarity is based on a bootstrap percentile CI of E(f₂) and f₂ in Pheq_bootstrap and DDSolver, respectively. Bootf2bca provides all four 90% CI.

DISCUSSION

Bootf2bca or Pheq_bootstrap should be considered for the estimation of the 90% CI of the f₂ similarity factor when dissolution profiles show high variability.

A Novel Simulation-Based Approach for Comparing the Population Against Average Bioequivalence Statistical Test for the Evaluation of Nasal Spray Products on Spray Pattern and Droplet Size Distribution Parameters

The aim of this work is to evaluate average bioequivalence (ABE) and population bioequivalence (PBE) statistical approaches so as to identify which approach is most suitable for in vitro bioequivalence (IVBE) testing of nasal spray products. For droplet size distribution (DSD) and spray pattern (SP), in vitro data were collected using a well-established nasal spray on the market (Nasonex®, manufactured by Merck Sharp & Dohme Limited). Simulations were performed using in vitro data to comparatively investigate ABE and PBE tests. For highly variable parameters such as SP area, this study clearly demonstrates that the level of agreement between ABE and PBE test conclusions is much smaller as compared with that of DSD Dv(50), which was found to have moderate variability. PBE approach dictates equivalence for both means and variances, and was found to handle both SP and DSD parameters with similar passing rates compared to the passing rates from the ABE approach. However, pronounced asymmetric behavior of PBE empirical power curves for highly variable SP area was observed. A modified PBE statistical approach is proposed for DSD span and Dv(50) in vitro parameters, where acceptance criteria would be based on comparison of reference/branded product to itself as part of "pre-IVBE study" via innovative statistical bootstrap simulations. Due to inherent high variability of the SP area parameter driving pronounced asymmetric behavior of PBE power curves, and due to unclear in vivo relevance for SP area and ovality, authors propose that SP parameters be used as development and quality control tools rather than for demonstration of IVBE.

Ion-pair approach coupled with nanoparticle formation to increase bioavailability of a low permeability charged drug

Atenolol is a drug widely used for the treatment of hypertension. However, the great drawback it presents is a low bioavailability after oral administration. To obtain formulations that allow to improve the bioavailability of this drug is a challenge for the pharmaceutical technology. The objective of this work was to increase the rate and extent of intestinal absorption of atenolol as model of a low permeability drug, developing a double technology strategy. To increase atenolol permeability an ion pair with brilliant blue was designed and the sustained release achieved through encapsulation in polymeric nanoparticles (NPs). The in vitro release studies showed a pH-dependent release from NPs, (particle size 437.30 ± 8.92) with a suitable release profile of drug (atenolol) and counter ion (brilliant blue) under intestinal conditions. Moreover, with the in vivo assays, a significant increase (2-fold) of atenolol bioavailability after administering the ion-pair NPs by oral route was observed. In conclusion, the combination of ion-pair plus polymeric NPs have proved to be a simple and very useful approach to achieve a controlled release and to increase the bioavailability of a low permeability charged drugs.

Equivalence analyses of dissolution profiles with the Mahalanobis distance

For some postapproval changes, the manufacturer has to demonstrate that the dissolution profile of the drug product before the change is statistically equivalent to the dissolution profile after the change. Guidelines suggest the so-called similarity factor f₂ as standard approach for the equivalence analysis. f₂ is a statistically questionable transformation of the Euclidean distance between both profile means and does not allow a control of the type I error rate. An alternative multivariate distance measure for quantifying the dissimilarity between both profile groups is the Mahalanobis distance. Current equivalence procedures based on the Mahalanobis distance implicate some practical problems in the dissolution context: either one chooses an exact method but the determination of a product independent equivalence margin will not be practically feasible or one chooses an approximate alternative that suffers from the bias of the Mahalanobis distance point estimate. This paper suggests the T2EQ approach for dissolution profile comparisons. T2EQ is a practically feasible equivalence procedure based on the Mahalanobis distance with an internal equivalence margin for comparing dissolution profiles. The equivalence margin is compliant with current dissolution guidelines. The operating characteristics (size, robustness, and power) are investigated via simulation: T2EQ meets the needs of both authorities and industry: not affected by the bias of the point estimate the type I error rate can be reliably controlled for various distribution assumptions and the power of T2EQ exceeds the power of methods recently discussed in the literature. These results were presented for the first time at CEN-ISBS 2017.

A Simple Approach for Comparing the In Vitro Dissolution Profiles of Highly Variable Drug Products: a Proposal

When in vitro dissolution profile variability prohibits the use of the F2 metric, there currently is no satisfactory alternative available. Published reports evaluating alternative approaches such as Multivariate Statistical Distance and use of a bootstrap F2 identify sources of bias that can limit the utility of these alternatives. Within veterinary medicine, an additional complication is the potential magnitude of interlot variability associated with dosage forms containing "natural" ingredients. In situations when both interlot and intralot variability need to be factored in the test and reference profile comparison, we designed a method that integrates such concepts as F2, USP S1 and S2 criteria and statistical tolerance limits. Unlike F2, this alternative approach integrates a statistical confidence into the determination through the use of tolerance limits about the reference product profile. Moreover, while differences in product variability, along with differences in mean profiles, will influence the comparability assessment, this method does not impose the need to confirm homogeneity of variances: there is not direct statistical comparison of test versus reference dissolution data. For more typical situations when interlot variability is not a concern, the F2 component can be omitted from the profile comparison. Lastly, by being a model-independent approach, we avoid the potential for introducing error into the comparability determination due either to model misspecification or problems associated with a lack of collinearity. This manuscript details this alternative approach and the results of performance characterization efforts to illustrate its behavior under a range of potential situations.

Regulatory assessment of drug dissolution profiles comparability via maximum deviation: Regulatory assessment of drug dissolution profiles' comparability via maximum deviation

In drug development, comparability of dissolution profiles of 2 different formulations is usually assessed using the similarity factor f₂ . In practice, the drug dissolution profiles are deemed similar if the f₂ exceeds 50, which occurs when a 10% maximum difference in the mean percentage of the dissolved drug at each time point between test and reference formulation is obtained. According to the Guideline on the Investigation of Bioequivalence (CPMP/EWP/QWP/1401/98 Rev. 1/ Corr **) use of the f₂ is however restricted by a set of validity conditions. If some of these conditions are not satisfied, the f₂ is not considered suitable, and alternative statistical methods are needed. In this article, we propose an inferential framework based on the maximum deviation between curves to test the comparability of drug dissolution profiles. The new methodology is applicable regardless whether the validity criteria of the f₂ are met or not. Contrary to the f₂ , this approach also integrates the variability of the measurements over time and not only their average. To benchmark our method, we performed simulations informed by 3 real case studies provided by the European Medicines Agency and extracted from dossiers submitted to the Centralised Procedure for Marketing Authorisation Application. In the scenarios of the simulation study, the new method controlled its type I error rate when the maximum deviation was greater than the similarity acceptance limit of 10%. The power exceeded 80% for small values of the maximum deviation, while the test was more conservative for intermediate ones. Our results were also very robust to sampling variations. Based on these positive findings, we encourage applicants to consider the new maximum deviation-based method as a valid alternative to the f₂ , especially when the validity criteria of the latter are not met.

Comparison of Dissolution Similarity Assessment Methods for Products with Large Variations: f2 Statistics and Model-Independent Multivariate Confidence Region Procedure for Dissolution Profiles of Multiple Oral Products

The current Japanese Ministry of Health Labour and Welfare (MHLW)'s Guideline for Bioequivalence Studies of Generic Products uses averaged dissolution rates for the assessment of dissolution similarity between test and reference formulations. This study clarifies how the application of model-independent multivariate confidence region procedure (Method B), described in the European Medical Agency and U.S. Food and Drug Administration guidelines, affects similarity outcomes obtained empirically from dissolution profiles with large variations in individual dissolution rates. Sixty-one datasets of dissolution profiles for immediate release, oral generic, and corresponding innovator products that showed large variation in individual dissolution rates in generic products were assessed on their similarity by using the f₂ statistics defined in the MHLW guidelines (MHLW f₂ method) and two different Method B procedures, including a bootstrap method applied with f₂ statistics (BS method) and a multivariate analysis method using the Mahalanobis distance (MV method). The MHLW f₂ and BS methods provided similar dissolution similarities between reference and generic products. Although a small difference in the similarity assessment may be due to the decrease in the lower confidence interval for expected f₂ values derived from the large variation in individual dissolution rates, the MV method provided results different from those obtained through MHLW f₂ and BS methods. Analysis of actual dissolution data for products with large individual variations would provide valuable information towards an enhanced understanding of these methods and their possible incorporation in the MHLW guidelines.