Why a FaSSIF–FeSSIF Comparison Needs Care
Fasted-state simulated intestinal fluid (FaSSIF) and fed-state simulated intestinal fluid (FeSSIF) are in vitro media designed to represent selected properties of intestinal contents in fasted and fed conditions. Their dissolution profiles can help reveal whether a drug’s apparent dissolution behavior changes in media containing different levels of bile components. But a direct difference between two profiles does not, by itself, identify which medium property caused the change.
The recipes can differ in bile-salt and phospholipid concentrations, pH, buffer composition, and other properties. Experimental conditions can add further differences: dose, vessel volume, agitation, and sampling can all affect the observed result. A useful interpretation therefore separates two questions: does the complete test medium change the measured profile? And, if it does, which change in composition or test conditions explains the difference?
FaSSIF and FeSSIF are recipe-defined media, not single universal chemical compositions. For example, commonly used formulations may have different nominal pH values and different bile-salt and phospholipid concentrations, but recipes and versions vary. Report the exact formulation, preparation, measured pH, and relevant concentrations rather than treating the media names as complete specifications.
What a Matched Comparison Can Establish
For a direct comparison, use the same drug lot and physical form, dose, vessel volume, apparatus, agitation, temperature, and sampling schedule in both media. Keep the analytical method and the definition of “dissolved” consistent. If pH is not deliberately being tested, either match it or explicitly recognize that the comparison includes a pH difference. When these conditions are matched, a reproducible profile difference supports the conclusion that the media as formulated affect the measured dissolution behavior.
A common summary at time \(t\) is the percentage of the administered dose measured as dissolved:
Here \(M_{\mathrm{diss}}(t)\) is the measured dissolved amount, in mass units, and \(D\) is the dose in the same units. The difference \(F_{\mathrm{FeSSIF}}(t)-F_{\mathrm{FaSSIF}}(t)\) is in percentage points. It is not a percentage increase unless it is divided by the FaSSIF value and expressed relative to that value.
The dissolved amount also needs a consistent operational definition. In bile-salt media, measured drug can include free drug and drug associated with micelles, depending on the separation and assay procedure. Such a total dissolved measurement is not automatically a measurement of free-drug concentration. If supersaturation and precipitation occur, a concentration may rise and later fall; the resulting curve cannot be interpreted as a simple record of dissolution alone. The earlier material on free versus micelle-bound drug and on supersaturation and precipitation provides useful context for these distinctions.
Worked Example: Describing a Matched Whole-Medium Difference
In a hypothetical experiment, the same 100 mg dose is tested in 500 mL of FaSSIF and 500 mL of FeSSIF. The vessel, agitation, temperature, sampling, and assay are matched. The respective mean percentages dissolved are 42% and 68% at 30 minutes, and 61% and 82% at 60 minutes. Calculate the profile differences and state what they show.
At 30 minutes, the difference is \(68\%-42\%=26\) percentage points. The corresponding amounts are \(0.42(100)=42\) mg in FaSSIF and \(0.68(100)=68\) mg in FeSSIF. The measured amount in FeSSIF is therefore 26 mg higher at that time.
At 60 minutes, the difference is \(82\%-61\%=21\) percentage points. The amounts are \(0.61(100)=61\) mg and \(0.82(100)=82\) mg, again a 21 mg difference.
Because the experimental conditions were matched, these results describe a higher measured dissolved fraction in the complete FeSSIF formulation at both times. They do not show that bile salts alone caused the difference: the media may also differ in pH or other recipe components. Nor do these data alone establish a statistical difference; replicate variability and an appropriate analysis would be needed for that claim.
Separating Bile-Salt and Phospholipid Effects
To distinguish components, use controlled media rather than attributing a whole-medium contrast to one ingredient. A simple factorial design holds pH, buffer, ionic conditions, dose, and hydrodynamics constant while varying bile salts and phospholipids independently. At minimum, compare a baseline with neither component added, bile salts alone, phospholipids alone, and both together. Use the intended concentrations and keep the remaining formulation as similar as possible.
This design makes it possible to describe component effects and their interaction. A bile-salt effect averaged across phospholipid levels is the mean result at high bile salt minus the mean result at low bile salt. An interaction asks whether the effect of bile salts changes when phospholipids are present. These are descriptive contrasts unless the experiment includes suitable replication and an analysis that accounts for experimental variation.
Worked Example: Interpreting a Two-Component Comparison
A hypothetical, replicated experiment measures the percentage dissolved at 30 minutes. All four media have matched pH and buffer, and use the same dose, volume, apparatus, and agitation. The mean results are: neither added component, 35%; bile salts only, 49%; phospholipids only, 40%; and both components, 62%. Calculate the average component contrasts and the interaction contrast.
The average bile-salt contrast is the mean at high bile salt minus the mean at low bile salt:
The average phospholipid contrast is:
The interaction contrast is the result with both components, minus the bile-salt-only and phospholipid-only results, plus the baseline:
The positive interaction means that, on this percentage-point scale, the combined result is 8 points above the result predicted by adding the two separate increments to the baseline. It is consistent with a combined effect beyond simple additivity in this experiment. It does not establish a molecular mechanism, and statistical evidence for an interaction depends on the replicate data and their variability.
These contrasts are interpretable only if the intended component changes are not accompanied by uncontrolled changes in other properties. Adding a bile-salt stock solution, for example, may alter volume, ionic strength, or pH. Prepare matched control additions, verify pH, and report the final concentrations. A factorial design is more informative than a sequence of unrelated formulations, but it cannot eliminate confounding introduced during preparation.
Check Dose, Solubilization Capacity, and Hydrodynamics
A medium comparison can be distorted if the dose approaches or exceeds the medium’s solubilization capacity. A useful screening quantity is the ratio of the nominal initial dose concentration to apparent solubility:
Here \(V\) is the vessel volume and \(S_{\mathrm{app}}\) is the apparent solubility in the specified medium and at the relevant pH and temperature. A larger ratio indicates a greater challenge to maintaining sink conditions. It is a screening comparison, not a universal pass/fail threshold: the study should state its chosen sink criterion and how apparent solubility was measured. In micellar media, the relevant solubility definition must match the experimental measurement; total apparent solubility and free-drug solubility are not interchangeable.
Hydrodynamics also matter. Agitation and vessel geometry affect particle motion, boundary-layer transport, and the rate at which dissolved drug is carried away from the dissolving surface. Matching nominal rpm is helpful only when apparatus and geometry are also matched. A change in apparatus or agitation can therefore confound a medium comparison even if the chemical recipes are identical.
Worked Example: Identifying Dose and Hydrodynamic Confounding
In a hypothetical comparison, 100 mg is tested in 500 mL of FaSSIF, where measured apparent solubility is 0.12 mg/mL. The FeSSIF test uses the same dose in 900 mL, where apparent solubility is 0.30 mg/mL. Agitation is 50 rpm in FaSSIF and 100 rpm in FeSSIF. At 30 minutes, 48% and 70% of the doses are measured dissolved. Calculate the dose-to-solubility ratios and assess whether the profile contrast isolates a medium effect.
For FaSSIF, \(D/V=100/500=0.200\) mg/mL, so:
For FeSSIF, \(D/V=100/900=0.111\) mg/mL, so:
Both ratios exceed a stringent sink criterion of \(R_{\mathrm{dose}}\leq 0.10\), if that is the criterion prespecified for this hypothetical study. The FeSSIF condition has a lower ratio and a higher agitation speed. Its 70% result (70 mg) versus 48% (48 mg) cannot be attributed to medium composition alone: volume and agitation also differ. A follow-up comparison should match volume and hydrodynamics, and should characterize the medium-specific apparent solubility under the chosen measurement conditions.
Sampling, Diagnostics, and Alternative Comparisons
Repeated sampling can lower vessel volume and remove dissolved drug. If each sample of volume \(v\) is replaced with drug-free medium, the cumulative amount measured at time \(t_n\), under a well-mixed, constant-volume assumption, is:
Here \(V\) is the maintained vessel volume, \(C_n\) is the measured concentration at \(t_n\), and each earlier sample \(i\) had volume \(v_i\) and concentration \(C_i\). The first term is the amount currently in the vessel; the sum adds the amounts removed in earlier samples. This correction assumes replacement volumes are accounted for and that the sample represents the well-mixed vessel. Without replacement, or with a changing vessel volume, use the corresponding volume history rather than this constant-volume expression.
Worked Example: Correcting for Drug Removed in Samples
A hypothetical vessel is maintained at 500 mL by replacing each 10 mL sample with drug-free medium. Concentrations are 0.080 mg/mL at 10 minutes and 0.120 mg/mL at 20 minutes. Calculate the cumulative dissolved amount at 20 minutes and the percentage of a 100 mg dose.
At 20 minutes, the vessel contains \(500(0.120)=60.0\) mg. The 10-minute sample previously removed \(10(0.080)=0.800\) mg. Therefore:
The dose-normalized fraction is \(100(60.8/100)=60.8\%\). Reporting only the 60.0 mg currently in the vessel would undercount cumulative measured drug by 0.800 mg, or 0.800 percentage points of the dose.
Useful diagnostic checks include verifying the final pH before and after dissolution, confirming the media recipe and concentrations, checking for undissolved material or visible precipitation, and examining replicate profiles for unusual variability. A falling concentration should prompt checks for precipitation, adsorption, degradation, sampling artifacts, or analytical recovery; it should not automatically be called reverse dissolution. An apparent plateau can reflect limited solubilization, slow dissolution, or a measurement limitation, so a profile alone may not identify the cause.
A matched FaSSIF-versus-FeSSIF comparison is appropriate when the question concerns the net difference between specified fasted- and fed-state recipes. A component-controlled factorial comparison is preferable when the question concerns the separate effects of bile salts and phospholipids. A pH-matched comparison helps isolate non-pH medium differences, while a deliberate pH series is needed to study pH dependence. These alternatives answer different questions; none should be interpreted as an interchangeable substitute for the others.
Common Pitfalls and Limits
- Attributing a whole-recipe contrast to bile salts. A higher FeSSIF profile does not isolate bile salts if pH, phospholipids, or other formulation properties also differ. Use component-controlled comparisons to support component-specific claims.
- Calling percentage points a percent increase. A change from 42% to 68% is 26 percentage points. The relative increase is \(100(68-42)/42=61.9\%\), a different quantity.
- Ignoring dose and hydrodynamics. A different dose-to-volume ratio or agitation condition can change the profile independently of medium composition. Match these conditions or state the resulting confounding.
- Treating measured dissolved drug as free drug. The assay may include micelle-associated drug. Interpret the result according to the sampling, separation, and analytical procedure actually used.
- Equating an in vitro difference with an in vivo outcome. A dissolution profile is evidence about behavior under its test conditions. It does not alone establish the magnitude of an absorption or exposure difference in people.
Check Your Understanding
Answer each question, then compare your reasoning with the brief solutions.
- A matched experiment reports 55% dissolved in FaSSIF and 70% in FeSSIF at 45 minutes. What is the difference in percentage points, and what can be concluded if the recipes differ in pH as well as bile-salt concentration?
- In a controlled two-factor experiment, the four results are 30% with neither component, 44% with bile salts only, 36% with phospholipids only, and 54% with both. Calculate the interaction contrast.
- A 120 mg dose is tested in 600 mL, with measured apparent solubility \(0.20\) mg/mL. Calculate \(R_{\mathrm{dose}}\) and state what it indicates relative to a prespecified sink criterion of 0.10.
- Why can a measured total dissolved concentration in FaSSIF or FeSSIF not automatically be interpreted as free-drug concentration?
- In a replaced-sample experiment, \(V=400\) mL, \(C_2=0.050\) mg/mL, and the earlier 8 mL sample had concentration \(0.040\) mg/mL. What cumulative amount is measured at the second time point?
Solutions. (1) The difference is \(70-55=15\) percentage points. With pH also differing, the result describes the whole-medium contrast and does not isolate bile salts. (2) The interaction is \(54-44-36+30=4\) percentage points, a positive interaction contrast on this scale. (3) \(D/V=120/600=0.200\) mg/mL, so \(R_{\mathrm{dose}}=0.200/0.20=1.00\). This exceeds 0.10 and does not meet that stated sink criterion. (4) The assay can measure micelle-associated drug as well as free drug, depending on the separation and analytical method. (5) The cumulative amount is \(400(0.050)+8(0.040)=20.0+0.320=20.32\) mg, assuming the sample was replaced and vessel volume maintained.