Despite the advancements in non-oral drug delivery, most products in the commercial circulation are meant only for oral intake owing to apparent benefits of better patient compliance, reduced cost of therapy, better safety, and ease in manufacturing.
Development of such oral drug products and their subsequent regulatory approval is a fastidious process. Amongst multiple quality traits demanded by the federal agencies to testify their product performance and therapeutic capability, the bioavailability-bioequivalence (BA-BE) studies remain as the most vital prerequisite and requisite parameters.
Drugs administered through oral route encounter several hiccups during their absorption into systemic circulation including differences in pH and surface area across different segments of gastrointestinal (GI) tract, first-pass effect, metabolism by gut enterocytes, intestinal motility, and membrane transport by structure-specific transporters.
Figure 1 portrays the sequential events associated with oral administration of solid dosage forms including drug dissolution, absorption into systemic circulation, and transit of unabsorbed fraction.
Generic drugs are considered as the customized preparations comparable to their branded counter parts in dosage form, strength, quality, efficacy, performance, intended use, and above all, in BA-BE studies (biostudies). In order to demonstrate therapeutic equivalence for ensuring prescribability, switchability and interchangeability, the generic products must scientifically exhibit pharmaceutical equivalence and bioequivalence to the reference listed drug (RLD) products. Pharmacokinetic metrics like AUC and Cmax are embarked upon to establish metrics like average, individual and population bioequivalence in terms of rate and extent of drug absorption. Significance of rate as well as extent of drug absorption in enunciating the concept of biostudies are enunciated in Figure 2.
Such biostudies, carried out in human subjects, tend to involve myriad hassles, encompassing phenomenal expenditure of resources, expertise in biopharmaceutics, pharmacokinetics and clinical research, coupled with the strict adherence to the ethical principles. As lion’s share of product development cost is expended only on biostudies, every pharma house dreams and aspires for circumnavigating the circuitous path of BA-BE studies by getting waived off (i.e., biowaivers) by federal agencies. The regulatory agencies have been pharmaceutical firms for biowaivers by submitting the pertinent documentary evidence for the purpose. To accomplish the said task, a biopharmaceutical scientist invariably draws the maximum out of the most valuable tool in his armamentarium, i.e., In Vitro/In Vivo Correlations (IVIVC).
IVIVC
USFDA describes IVIVC as “a predictive mathematical model with relationship between an in vitro property of a dosage form and a relevant in vivo response”. Generally, the in vitro property is the rate or extent of drug dissolution, while the in vivo response parameter is amount of drug absorbed. A pictographic depiction of IVIVC is portrayed in Figure 3. Table 1 enlists a chronological account on evolution of IVIVC and biowaiver by USFDA.
Understanding and controlling the relationship between in vitro and in vivo parameters help in establishing meaningful IVIVC not only for obtaining biowaivers, but for developing robust immediate release (IR) and extended release (ER) drug products, fixed-dose combinations and drug delivery systems of diverse kinds too. Today, key regulatory agencies across the world have either developed their own guidances or have endorsed USFDA to grant biowaivers using IVIVC or beyond (Figure 4).
A typical IVIVC model tends to involve three
cardinal elements, viz. model development, model validation and model application.
IVIVC model development
Developing a predictable IVIVC model depends upon the physicochemical properties of drug, complexity of drug product, its formulation composition, method of manufacture, and dissolution methodology. There could be four different types of federally acceptable IVIVC models, viz. level A, level B, level C and multiple level C.
Level A is the highest correlation, and extensively employed for obtaining federal biowaiver during ANDA approval. It is a point-to-point correlation between entire in vitro dissolution time course and entire in vivo absorption time course. The values of fraction of drug absorbed can be estimated from plasma level profiles using model-dependent two-stage deconvolution approach employing Wagner-Nelson or Loo-Riegelman method or model-independent numerical deconvolution approach. Subsequently, correlation between the in vivo absorption profile and in vitro dissolution is evaluated as level A. An instance of establishing “level A” IVIVC is illustrated in Figure 5.
Level B is based upon mean time parameters computed using statistical moment analysis. Typically, mean residence time (MRT) in vivo or mean dissolution time (MDT) in vivo is compared with the MDT in vitro. Level C represents single-point correlation between one dissolution time point (t50, t90, Q4h) with one pharmacokinetic parameter (Cmax, tmax, AUC). Multiple level C relates one or more pharmacokinetic parameters to the amount of drug dissolved at various time points. Level B and C models are considered as less predictive than level A.
Multiple level C, however, is taken almost equivalent to level A, and thus can be employed as a surrogate to level A to justify biowaiver. A pictorial account of different IVIVC models is depicted in Figure 6.
IVIVC in the light of BCS
Biopharmaceutical Classification System (BCS), based on the solubility and permeability characteristics, classifies the drug molecules into four categories, as indicated in Figure 7.
Oral bioavailability of a drug product is a composite function of two vital steps of drug dissolution from dosage form and permeation from gastrointestinal milieu. Oral drug absorption, therefore, can be considered as a consecutive process limited by drug dissolution rate, membrane permeability or solubility (Figure 8).
BCS identifies key parameters controlling drug absorption as dimensionless numbers, viz. absorption number (An), dissolution number (Dn) and dose number (Do). An is the ratio of MRT to the MAT, Dn is the ratio of MRT to MDT and Do is the mass divided by an uptake volume of 250 mL and solubility of the drug. Invariably, drugs with high An, low Dn and low Do hold good for establishing IVIVC.
BCS class I compounds formulated as IR products, generally do not exhibit good IVIVC as absorption rate depends only on gastric emptying. However, formulating these as ER products yields good IVIVC, as absorption becomes dissolution-dependent. Class II drugs formulated as ER products show good IVIVC when solubility and permeability are site-independent. Class III drugs depict limited or no IVIVC, as permeability is a rate-controlling factor. No IVIVC is usually expected in class IV drugs, as these possess significant problems in oral drug absorption and generally formulated in parenteral dosage forms. Table 2 compiles a succinct account on IVIVC establishment potential of various BCS compounds.
Recent modifications in BCS proposed by FDA, i.e., Biopharmaceutics Drug Disposition Classification System (BDDCS), also holds good in regulating the role of solubility and metabolism on the plausibility of IVIVC establishment and potential biowaivers.
IVIVC model validation
To successfully predict the outcome (in vivo profile) from a given model and test condition (in vitro profile), an IVIVC model needs to demonstrate predictability of in vivo performance of drug product(s) from their in vitro dissolution characteristics. Model predictability is validated by calculating prediction error (%PE):
IVIVC models are validated by internal and/or external prediction. Internal validation is conducted for at least three formulations with at least differing release rates (slow, medium and fast) for drugs with wide therapeutic index (TI). If the acceptable criteria of mean %PE of 10% or less with none greater than 15% are not met, external validation is carried out wherein mean %PE should be 10% or less with none greater than 20%. External validation is preferred for narrow TI drugs.
IVIVC model application
The main objective of establishing IVIVC is to seek biowaiver for products containing wide TI drugs. IVIVC acts as a surrogate for approved drug products in reducing biostudies during level 3 post-approval changes (SUPAC-MR). It also helps in the cases when drug products already on the market which need some modifications, for instance, in changes in manufacturing site, equipment, process or formulation excipients.
If an IVIVC is developed with the highest strength, leverages for changes in any lower strength may be granted, provided these strengths are compositionally proportional or qualitatively identical. IVIVC model can facilitate setting-up of the multi-point dissolution specifications for product development batches.
IVIVC makes it easier to establish the specifications of multipoint dissolution particularly w.r.t. biorelevant media and discriminating dissolution apparatus. Once IVIVC is successfully developed, the manufacturer seeking biowaiver needs to simply demonstrate the equivalence with in vitro dissolution profile (say using f2 factor), with that of the previous approved batch with proven bioequivalence.
Epilogue
In the era of swelling drug development costs, the pharmaceutical industry has been striving hard to find tangible ways to save precious resources. Tools evolved through naïve scientific probing like BCS and IVIVC have tremendous promise to act as magic-wands. Besides obtaining biowaivers, IVIVC can also be applied to develop robust products during formulation screening, setting dissolution specifications and product development. Relatively newer paradigm of in vitro/in vivo relationship (IVIVR) is very frequently being used for product development esp. for IR products. IVIVR is a qualitative or semi-quantitative association between in vivo data and in vitro metrics to predict a rank order relationship. In the era of Quality by Design (QbD), IVIVR has been found to be an excellent tool for demonstrating product understanding particularly during scale-up studies. Newer paradigms like in vitro/in vivo matching (IVIVM) and in vitro/in vivo profiling (IVIVP) are also being deliberated and practiced at industrial level. While the principles of IVIVC have been mostly applied to oral products, there exists a rational need to develop effectual and cost-effectual methodologies and standards for non-oral DDS too.
(Dr Bhupinder Singh Bhoop is Professor (Pharmaceutics & Biopharmaceutics) University Institute of Pharmaceutical Sciences and Sarwar Beg is UGC- Meritorious Research Fellow, University Institute of Pharmaceutical Sciences Panjab University, Chandigarh)