Bioequivalence and OSD CMC — BCS Classification, In Vivo Data, and the NDA CMC Interface

The bioequivalence study and the CMC dossier are written by different teams, reviewed by different FDA divisions, and filed in different parts of the NDA — but they share one number that both teams must agree on before filing: the dissolution Q-value.

The BE-CMC interface is where NDA dossiers fail — not because the data is wrong, but because the two sides of the dossier were never made to speak the same language.

BCS Classification Methodology — pH-Dependent Solubility, Dose:Solubility Volume Calculation, and Caco-2 Permeability Threshold That Determines Your Dissolution Method Requirement

A defensible BCS classification requires solubility measurement across all three physiologically relevant pH points, gastric and two points spanning the intestinal range, run at body temperature rather than ambient conditions, because using a single pH point or the wrong temperature can assign a classification that doesn’t actually reflect the compound’s real behavior. The dose-to-solubility volume calculation itself uses the minimum solubility observed across those three points, not whichever value happens to be most convenient, and an API dosed at 40 mg with a minimum solubility around 0.09 mg/mL at pH 6.8 produces a dose-to-solubility volume near 444 mL, comfortably above the 250 mL threshold that separates high from low solubility and placing the compound in BCS Class II territory alongside a permeability result from a Caco-2 monolayer assay clearing the high-permeability threshold. A classification that calculates the dose-to-solubility volume using only the pH 6.8 result, without checking whether solubility at pH 1.0 might actually be higher for an acidic compound and shift the classification toward Class I, has skipped a step FDA’s BCS biowaiver guidance requires, and getting this classification wrong doesn’t just create a labeling inconsistency, it determines whether the entire downstream dissolution method needs a biorelevant medium modification at all. A BCS Class II classification specifically flags that standard aqueous dissolution media likely won’t discriminate meaningfully between acceptable and unacceptable formulations, meaning the classification itself is what triggers the need for a modified medium in the first place.

Discriminating Dissolution Method Development — SDS Concentration Optimization, 15-Point Discrimination Criterion, and the BE Batch Dissolution Profile That Anchors the Q-Value

Once a BCS Class II classification establishes that a surfactant-modified medium is likely necessary, the actual concentration selected has to be optimized rather than chosen by convention, comparing dissolution profiles of a target formulation against a deliberately suboptimal comparator across a range of surfactant concentrations and identifying the lowest concentration that produces a meaningfully large gap between the two, commonly set at a minimum 15 percentage point difference as the threshold for adequate discrimination. A medium producing only a modest single-digit percentage point gap between an optimal and clearly suboptimal formulation has failed to discriminate for anything clinically meaningful, while a medium producing a difference well above the 15-point threshold, in the range of thirty percentage points or more, has demonstrated real sensitivity to the formulation variable that actually matters. Once that discriminating medium is settled, the specification itself only becomes defensible once the pivotal bioequivalence batch’s own dissolution profile is measured in that exact same medium, not a different medium used during earlier development work, because a specification set at 80% dissolved in a surfactant medium means nothing to a reviewer if the only dissolution data on file for the BE batch was generated in plain buffer. The margin between the BE batch’s actual dissolution result in the specification medium and the proposed Q-value is what anchors the entire specification to real clinical performance, and a mismatch between the medium used for the BE batch and the medium named in the 3.2.P.5 specification is one of the most direct and easily identified gaps a reviewer can spot.

IVIVC Level A Validation, SUPAC Dissolution Comparison Requirements, and the Post-Approval CMC Planning That Avoids New BE Studies for Manufacturing Changes

A Level A IVIVC model, correlating in vitro dissolution fraction against in vivo fraction absorbed derived from deconvoluting the BE study’s own plasma concentration data, only earns regulatory usefulness once it passes external validation against formulations that weren’t part of building the model in the first place, with prediction error on AUC held within 10% and prediction error on Cmax held within 15% as the acceptance standard. The practical payoff of building this model during the original NDA program, rather than treating it as optional, shows up specifically when a post-approval manufacturing change comes along: SUPAC Level 1 and Level 2 changes typically clear with a straightforward multi-point dissolution profile comparison against a similarity factor threshold, but a change large enough to be classified as Level 3, or a Level 2 change whose dissolution comparison falls just short of that similarity threshold, would otherwise require a brand-new clinical bioequivalence study, adding many months and a substantial direct cost to the change timeline. A validated Level A IVIVC model built during the original development program can instead predict the new formulation’s in vivo performance directly from its dissolution profile, allowing the change to proceed without repeating the clinical study entirely — but only for sponsors who invested in building and validating that model when the original BE data was collected, rather than discovering after the fact that no such model exists and now facing the full cost of a new BE study to support what might otherwise have been a routine manufacturing change.

The XGene OSD BE-CMC Integration Architecture — BCS Classification, Discriminating Method, BE Batch Anchoring, IVIVC Development, and the Complete FDA NDA Bioequivalence-CMC Interface Package

The XGene OSD BE-CMC Integration Architecture is a structured bioequivalence-CMC specification integration strategy built around the recognition that the BE team and the CMC team must produce data that speaks the same regulatory language before an NDA is filed.

1. BCS Classification Package — Establish solubility across all three physiological pH points at body temperature and calculate the dose-to-solubility volume using the true minimum solubility value. 2. Discriminating Dissolution Method Development — Optimize surfactant or biorelevant medium concentration against a documented 15-percentage-point discrimination criterion between optimal and suboptimal formulations. 3. BE Batch Dissolution Anchoring — Measure the pivotal BE batch’s dissolution profile in the exact specification medium, not a different development-stage medium, to establish the Q-value margin. 4. IVIVC Level A Model Development for SUPAC Readiness — Build and externally validate the correlation during the original NDA program, not after a post-approval change is already underway. 5. Post-Approval Dissolution Comparison Strategy — Plan explicitly for f2-based comparison at Level 1 and 2 changes and IVIVC-based bioequivalence prediction at Level 3, avoiding a new BE study by design.

The output is the BE-CMC interface package that gives FDA OPQ and clinical pharmacology reviewers the explicit connective evidence between bioequivalence data and dissolution specification that an NDA dossier otherwise leaves for the reviewer to question.

FDA’s Guidance for Industry: Waiver of In Vivo Bioavailability and Bioequivalence Studies… Based on a Biopharmaceutics Classification System (2017) establishes the BCS classification methodology this article’s analysis is built around, while FDA’s Guidance for Industry: Extended Release Oral Dosage Forms… In Vitro/In Vivo Correlations (1997) establishes the Level A IVIVC validation criteria applied to post-approval SUPAC planning. FDA’s Guidance for Industry: Dissolution Testing of Immediate Release Solid Oral Dosage Forms (1997) and the FDA SUPAC-IR (1995) guidance establish the dissolution specification derivation and post-approval change comparison framework applied throughout.

For your OSD NDA program, can you confirm today that your 3.2.P.2 section documents the BCS classification of your API with solubility measurements at all three pH points at 37°C, and that your pivotal BE batch’s dissolution profile was measured in the exact same medium named in your 3.2.P.5 specification?