Bioavailability by design Published equations and human PK data Educational simulations Proposed capability

Oral bioavailability is the product of release, absorption and survival, so a formulation should target the limiting step

How bioaccessibility, absorption and transformation combine into bioavailability, what the maximum absorbable dose says, and what tracer and curcumin studies measured.

Status. EUTECNIQ has no formulation, dissolution, permeability or pharmacokinetic data. Human results are those of the cited studies; a bioavailability gain is not evidence of efficacy.

Academy fundamentals: Course 8, Solubility, pH and the body

12.38%relative bioavailability of a ¹³C-labeled anthocyanin dose, from urine and breath

6.9–185×exposure reported for curcumin formulations against plain curcumin

5.3–136×one formulation's fold-change, depending on which curcuminoid was counted

105 mLsmall-intestinal fluid fasted, against 54 mL after a meal (mean, MRI)

Czank et al., Am J Clin Nutr 2013; Jamwal, J Integr Med 2018, with COT TOX/2025/09 Table 5; Jäger et al., Nutr J 2014, Table 1; Schiller et al., Aliment Pharmacol Ther 2005.

A published mass balance

Of a ¹³C-labeled anthocyanin dose, 12.38% left in urine and breath, 31.6% in feces, and 56.1% was not recovered in 48 hours

Where the ¹³C label from 500 mg of cyanidin-3-glucoside went in 8 men, as a share of the dosePublished data

Data table

Recovery of the 13C label after 500 mg of 13C-labeled cyanidin-3-glucoside, Czank et al. 2013
Fate of the ¹³C label within 48 h	% of dose	Basis
Urine	5.37 ± 0.67	Reported, mean ± SE
Exhaled breath (¹³CO₂)	6.91 ± 1.59	Reported, mean ± SE
Relative bioavailability (urine + breath)	12.38 ± 1.38	Reported; the two components above add to 12.28
Total recovered (urine, breath, feces)	43.9 ± 25.9	Reported; range 15.1–99.3 across participants
Feces	31.6	Derived here: 43.9 − 5.37 − 6.91
Not recovered in 48 h	56.1	Derived here: 100 − 43.9

Other reported values: labeled metabolites peaked at 5.97 ± 2.14 µmol/L in serum at 10.25 ± 4.14 h; their half-lives ranged from 12.44 ± 4.22 to 51.62 ± 22.55 h; metabolites were degradation products and phenolic, hippuric, phenylacetic and phenylpropenoic acids.

Czank C, Cassidy A, Zhang Q, Morrison DJ, Preston T, Kroon PA, Botting NP, Kay CD. Human metabolism and elimination of the anthocyanin, cyanidin-3-glucoside: a ¹³C-tracer study. Am J Clin Nutr. 2013;97:995–1003 (abstract values; NCT01106729). The Sankey's middle node is the sum of urine and breath (12.28%).

The 12.38% "relative bioavailability" counts labeled metabolites in urine and breath; the authors concluded that anthocyanins are more bioavailable than previously perceived.

Explanation, details and sources

A stable-isotope label lets a study follow the carbon of a compound wherever it goes, including into metabolites that a parent-compound assay would miss. What counts as "bioavailable" therefore depends on the analyte: the parent compound, its conjugates, or every labeled product. Recovery varied widely between participants (15.1–99.3%), and more than half of the label was not accounted for within 48 hours on average.

Four different results

Getting into solution, crossing the gut wall and reaching the blood unchanged are separate steps, and the smallest one sets the outcome

Each fraction is lost in a different place: the product and gut fluid, the gut wall, and the gut wall and liverSchematic, not to scale

Salvia-Trujillo L, Martín-Belloso O, McClements DJ. Nanomaterials. 2016;6(1):17, eq 1 and Table 2 (letters L, S, I, ML, TJ, MP, AT, ET, C, M as defined there); McClements DJ, Li F, Xiao H. Annu Rev Food Sci Technol. 2015;6:299–327.

Text version of the three fractions

Oral bioavailability as a product of three fractionsF = F_B × F_A × F_T

F_B, bioaccessibility: The fraction released from the food or product matrix and solubilized in gastrointestinal fluids. Limited by liberation (L), solubility (S) and interactions that make it insoluble (I).
F_A, absorption: The fraction of what is bioaccessible that crosses the mucus layer (ML) and the epithelium, through tight junctions (TJ), the membrane (MP) or active transporters (AT), and is not pumped back by efflux transporters (ET).
F_T, transformation: The fraction that reaches circulation without chemical (C) or metabolic (M) transformation into a less active or inactive form.

Salvia-Trujillo L, Martín-Belloso O, McClements DJ. Nanomaterials. 2016;6(1):17, eq 1 (written BA = B* × A* × T*) and Table 2; nutraceutical bioavailability classification scheme (NuBACS): McClements DJ, Li F, Xiao H. Annu Rev Food Sci Technol. 2015;6:299–327.

Bioaccessibility F_B Absorption F_A Transformation F_T (fraction surviving)

Oral bioavailability F

Limiting step

F if only the limiting step doubled

Where the dose is lostIllustrative fractions

F = F_B × F_A × F_T (Salvia-Trujillo et al. 2016, eq 1). Fractions are illustrative inputs, not values for any compound.

Explanation, details and sources

McClements and colleagues split oral bioavailability into three fractions that multiply. A formulation that triples one of them helps only if that one is the bottleneck. Evidence for each step comes from different experiments, described on Evidence and pharmacokinetics.

The regulatory definition, for drugs

"Bioavailability is the rate and extent to which the active ingredient or active moiety is absorbed from a drug product and becomes available at the site of drug action." For products not meant to reach the bloodstream, it is assessed by measurements that reflect availability at the site of action.

21 CFR 314.3(b), which 21 CFR 320.1 applies to bioavailability and bioequivalence requirements.

Bioaccessibility≠systemic bioavailability
Dispersibility≠solubility
A stable nanoemulsion≠demonstrated bioavailability
A bioavailability gain≠efficacy
Higher exposure≠a wider safety margin

Biopharmaceutics numbers

When solubility limits absorption, the absorbed amount stops rising with dose, and the maximum absorbable dose estimates where that ceiling is

Dose Solubility in intestinal fluid S Effective permeability P_eff Small-intestinal water volume Small-intestinal transit time Particle radius (for Dn)

Maximum absorbable dose

Ceiling on fraction absorbed

Absorption rate constant K_a

Dose number Do

Absorption number An

Dissolution number Dn

Maximum absorbable dose over solubility and permeability (log axes)Educational simulation

Drag to rotate, or use the buttons. On log axes the MAD is a plane, because it is a product; amber marks where it exceeds the dose.

At the defaults (10 µg/mL, 1 × 10⁻⁴ cm/s, 250 mL, 199 min) the MAD is about 4.8 mg, so at most about 1% of a 500 mg dose could be absorbed if solubility were the only limit.

MAD = S × K_a × SIWV × SITT with K_a = 2P_eff/R, R = 1.25 cm. Defaults: SIWV 250 mL (the BCS volume convention), SITT 199 min (3.32 h).

Schematic of a solid surface with a thin stagnant liquid layer: the concentration falls from the saturation value at the surface to the bulk value across the layer. Beside it, the Noyes-Whitney and Nernst-Brunner equations and a computed example in which 5-micrometer particles approach saturation ten times faster than 50-micrometer particles, while in both cases only 2.5 milligrams of a 100-milligram dose can dissolve in 250 milliliters because solubility caps the amount. — Noyes AA, Whitney WR. J Am Chem Soc 1897;19(12):930–934. Nernst W. Z Phys Chem 1904;47:52–55. Brunner E. Z Phys Chem 1904;47:56–102. Illustrative inputs as stated on the figure; constant-area approximation. Generated illustration, EUTECNIQ. Full-size SVG

Method, assumptions and limits

What it calculates: The maximum absorbable dose, the resulting ceiling on the fraction of a dose absorbed when solubility is limiting, the absorption rate constant, and the dose, absorption and dissolution numbers.
Equations: MAD = S·K_a·SIWV·SITT; K_a = 2P_eff/R; ceiling = min(1, MAD/dose); Do = dose/(250 mL × S); An = P_eff·T_si/R; Dn = 3D S T_si/(r²ρ), with T_si = SITT.
Variables and units: S in µg/mL (mg/mL inside MAD); P_eff in cm/s; K_a in min⁻¹; SIWV in mL; SITT in min; R = 1.25 cm; D = 5 × 10⁻⁶ cm²/s; ρ = 1.2 g/cm³; r in µm. Sliders for dose, S, P_eff and r are logarithmic.
Assumptions: Saturated luminal concentration throughout transit; passive permeation with a sink on the blood side; one well-mixed tube; no degradation or metabolism; no supersaturation, micellar solubilization or lipid digestion.
Valid range: A screening estimate only. Fasted small-intestinal water measured by MRI averaged 105 ± 72 mL in 12 volunteers and 54 ± 41 mL after a meal (Schiller et al. 2005), well below the 250 mL convention. Ding et al. (2012) found the equation gave conservative estimates of the absorbable dose compared with phase I dose-escalation results.
What it does not establish: Absorption of any real ingredient, the effect of food or formulation, first-pass metabolism (F_T), or efficacy. ICH M9 criteria apply to drug biowaivers, not to supplements.
Verification: An independent Python reference recomputes MAD, the ceiling, K_a, Do, An and Dn from the slider values with explicit unit conversions (tests/numerics/ref/technologies-bioavailability-design.py).

Explanation, details and sources

Amidon and colleagues showed that "the key parameters controlling drug absorption are three dimensionless numbers": an absorption number, a dissolution number and a dose number. If the gut fluid stays saturated, the most that can be absorbed during transit is solubility × absorption rate constant × fluid volume × transit time. Above that dose, more material mostly passes through.

Definitions

Dose, absorption and dissolution numbersDo = M₀V₀S, V₀ = 250 mL
An = P_effT_siR Dn = 3D S T_sir²ρ

M₀: dose; S: solubility; P_eff: effective intestinal permeability; T_si: small-intestinal transit time (3.32 h); R: intestinal radius (1.25 cm); D: diffusion coefficient (typically 5 × 10⁻⁶ cm²/s); r: particle radius; ρ: particle density (typically 1.2 g/cm³).

Amidon GL, et al. Pharm Res. 1995;12(3):413–420; values and forms as given by Abuhassan Q, et al. Eur J Pharm Biopharm. 2022;170:160–169 (open access), eq 6 and Table 3 notes. ICH M9 (2019): a drug substance is highly soluble if the highest single therapeutic dose dissolves in 250 mL or less over pH 1.2–6.8 at 37 ± 1 °C. The dose-number calculator is on Delivery.

Maximum absorbable dose

Curatolo's MAD, and the absorption rate constant of a tubeMAD = S × K_a × SIWV × SITT
K_a = 2 P_effR

SIWV: small-intestinal water volume; SITT: small-intestinal transit time. The K_a relation follows from Amidon's absorption-rate integral, dm/dt = ∬P_wC_wdA: with the wall concentration held at saturation, absorption rate = P_effS A, and a tube has A/V = 2/R. With V = SIWV the MAD equals 2·An·S·SIWV.

MAD equation: Curatolo W. Pharm Sci Technol Today. 1998;1(9):387–393, as given by Abuhassan et al. 2022, eq 2. Absorption-rate integral: Amidon et al. 1995, eq 2. The K_a step is our derivation derived.

Curcumin as a case study

Curcumin formulations report 6.9- to 185-fold higher exposure than plain curcumin, and the fold-change depends on the analyte, dose and comparator

Evidence levels: Traditional useIn vitroAnimalHuman PK/absorptionHuman RCTMeta-analysis of RCTsNot assessedDocumented safety concern

Relative bioavailability versus unformulated curcumin: review value and each study's own headlinePublished data Human PK/absorption

Show the data as a table

Relative oral bioavailability (fold) of curcumin formulations in healthy volunteers
Formulation (product studied)	Study	Jamwal 2018, dose-normalized (as reproduced by COT 2025)	Study's own headline
Liquid micelles (NovaSol)	Schiborr et al. 2014	185	185-fold (curcuminoid AUC, all subjects; 277 in women, 114 in men)
Water-dispersible complex, polyvinylpyrrolidone and cellulose based (CurcuWin)	Jäger et al. 2014	136.3	45.9-fold (total curcuminoids)
Solid lipid curcumin particles (LongVida)	Gota et al. 2010	100	Not a ratio: unformulated curcumin was not detected at the same dose (formulation Cmax 22.43 ng/mL)
γ-Cyclodextrin formulation (CW8)	Purpura et al. 2018	85	39-fold (total curcuminoids, AUC_{0–12 h})
Phytosome: curcumin, soy lecithin, microcrystalline cellulose (Meriva)	Cuomo et al. 2011	48	About 29-fold (total curcuminoids)
Colloidal nanoparticles (Theracurmin)	Sasaki et al. 2011	15.9	27-fold (AUC, 30 mg of each)
Fenugreek soluble-fiber delivery system	Im et al. 2012	15.8	Not checked against the primary paper
Dispersed micronized curcuminoids, sustained release (MicroActive Curcumin)	Madhavi & Kagan 2014	9.7	Abstract states no fold-change
Micronized curcumin powder	Schiborr et al. 2014	9	9-fold (all subjects; 14 in women, 5 in men)
Turmeric essential-oil formulation (BCM-95CG)	Antony et al. 2008	6.9 (corrected value)	About 6.93-fold

Review values: Jamwal R. J Integr Med. 2018;16(6):367–374, with corrigendum J Integr Med. 2019;17(4):310 (the turmeric essential-oil value is the corrected one), as reproduced in UK Committee on Toxicity discussion paper TOX/2025/09, Annex A case study 2, Table 5 (last updated March 19, 2025). The COT table as published online shows ten values; Jamwal's abstract reports eleven formulations, and we could not access the full text to identify the eleventh. COT's table shows a "Phytosomal" label without a value directly above the 48 entry for Cuomo et al.; we read them as one row. Product names are those used in the primary papers or in Jamwal's abstract. Own headlines: abstracts of Schiborr et al. 2014; Jäger et al. 2014; Gota et al. 2010; Purpura et al. 2018; Cuomo et al. 2011; Sasaki et al. 2011; Madhavi & Kagan 2014; Antony et al. 2008.

Five methodological choices move a fold-change more than most formulation differences doHuman PK data Illustrative curves

COT TOX/2025/09, paragraphs 95 and 98; Gota VS, et al. 2010; Shoba G, et al. 1998; Cuomo J, et al. 2011; Stohs SJ, et al. 2020; Schiborr C, et al. 2014 (277-, 185- and 114-fold); Purpura M, et al. 2018. Curves in the first two panels are shapes, not data (one-compartment oral profile, absorption rate 1 h⁻¹, elimination 0.3 h⁻¹).

One study, three formulations, five analytes: the same formulation's fold-change ran from 5.3 to 136 depending on which curcuminoid was countedPublished data

Data table (Jäger et al. 2014, Table 1)

AUC 0–12 h (ng·h/mL, geometric mean) and relative absorption F by analyte and formulation
Analyte	CS AUC (1,800 mg)	CTR AUC	CTR F	CP AUC	CP F	CHC AUC	CHC F
Curcumin	10.8	5.8	2.6	28.7	12.7	307.6	136.3
Demethoxycurcumin (DMC)	18.4	2.2	0.6	28.0	7.3	54.5	14.2
Bisdemethoxycurcumin (BDMC)	9.3	2.6	1.3	6.7	3.5	10.2	5.3
Tetrahydrocurcumin (THC)	1.1	0.3	1.3	1.9	8.3	7.7	33.5
Total curcuminoids	39.6	10.9	1.3	65.3	7.9	380.0	45.9

CTR, CP and CHC were each given at 376 mg of curcuminoids. Check: total curcuminoids, CHC: (380.0 ÷ 39.6) × (1,800 ÷ 376) = 45.9.

Jäger R, Lowery RP, Calvanese AV, Joy JM, Purpura M, Wilson JM. Comparative absorption of curcumin formulations. Nutr J. 2014;13:11, Table 1 (open access). Randomized, double-blind crossover in 12 healthy volunteers. CS, standardized curcumin; CTR, with turmeric volatile oils; CP, phytosome; CHC, hydrophilic carrier with cellulosic derivatives and antioxidants. Plasma was treated with glucuronidase/sulfatase, so values include conjugated metabolites. F = (AUC test ÷ AUC CS) × (1,800 ÷ 376), which assumes exposure proportional to dose.

Text version: why the numbers differ

Why the numbers differ

Five methodological choices move a fold-change more than most formulation differences do

Dose normalization. Relative bioavailability = (AUC_formulation × dose_control)/(AUC_control × dose_formulation); most studies gave different doses, so values had to be dose-normalized (COT 2025), which assumes exposure is proportional to dose. Gota et al. observed nonlinear dose dependency.
An undetectable comparator. When plain curcumin is below the assay's detection limit (Gota et al. 2010; Shoba et al. 1998, "undetectable or very low"), any ratio depends on how the missing value is handled. Course 8 sets Shoba 1998 beside a 30-person crossover.
What is measured. Free curcumin, total curcuminoids and conjugated metabolites are different analytes. After Meriva only phase-2 metabolites were detected (Cuomo et al. 2011), and Stohs et al. (2020) call hydrolyzing plasma samples before quantifying curcumin "widely used but inappropriate".
Who is studied. Micellar curcumin was 277-fold better absorbed in women and 114-fold in men (Schiborr et al. 2014).
How many. Crossover studies of 12 to 23 healthy volunteers (Purpura et al. 2018; Schiborr et al. 2014) estimate averages with wide individual variation.

COT TOX/2025/09, paragraphs 95 and 98; Gota VS, et al. J Agric Food Chem. 2010;58:2095–2099; Shoba G, et al. Planta Med. 1998;64(4):353–356; Cuomo J, et al. J Nat Prod. 2011;74:664–669; Stohs SJ, et al. Molecules. 2020;25(6):1397; Schiborr C, et al. Mol Nutr Food Res. 2014;58:516–527; Purpura M, et al. Eur J Nutr. 2018;57:929–938.

Higher plasma curcuminoids are not clinical benefit, and higher exposure is also a safety question: 3 of 7 tested products in a turmeric liver-injury series contained piperine.

Explanation, details and sources

Jamwal's 2018 review collected human pharmacokinetic studies of eleven formulations. The UK Committee on Toxicity reproduced its dose-normalized values in 2025. The same products' own papers often report a different headline number, because they measured total curcuminoids rather than curcumin, did not normalize dose, or could not detect the comparator.

What the gain does not show

Higher plasma curcuminoids are not clinical benefit, and evidence reviews rarely separate formulations

Cuomo and colleagues noted that even after the phytosome formulation, plasma concentrations "were still significantly lower than those required for the inhibition of most anti-inflammatory targets of curcumin".

Of 171 systematic reviews of oral curcumin published from 2003 to 2022, subgroup or sensitivity analyses addressing formulation bioavailability appeared in the methods of only 3.5% and the results of only 6.4%, although 57% mentioned bioavailability in their discussion. The authors conclude that most such reviews "should be taken with caution".

Exposure is also a safety question. Among ten turmeric-associated liver-injury cases in the U.S. Drug-Induced Liver Injury Network, 3 of the 7 products tested also contained piperine, and 7 patients carried HLA-B*35:01 (Halegoua-DeMarzio et al. 2023). Documented safety concern More on Evidence and pharmacokinetics.

Cuomo J, et al. 2011; Bučević Popović V, et al. Pharmaceuticals. 2024;17(2):164 (PMC10891944); Halegoua-DeMarzio D, et al. Am J Med. 2023;136(2):200–206.

Strategy by limiting factor

Each limiting factor has its own tools, test and risk, and a metabolism inhibitor adds drug-interaction risk

Diagnose the limiting step from its sign, then pick only the tools, tests and risks that belong to itPublished evidence

Human examples: micronized and micellar curcumin (Schiborr 2014), γ-cyclodextrin curcumin (Purpura 2018), piperine with curcumin (Shoba 1998). In vitro: piperine inhibited P-glycoprotein and CYP3A4 (Bhardwaj 2002). Safety: Halegoua-DeMarzio 2023. INFOGEST 2.0 is "not suitable for simulating digestion kinetics" (Brodkorb 2019).

Text version with examples and evidence levels

Formulation strategies by the factor that limits oral bioavailability, with evidence and risks
Limiting factor	Signs	Formulation options	Evidence that it worked	Risks and limits
Bioaccessibility: solubility and dissolution	High dose number, low dissolution number, MAD below the dose	Particle-size reduction; amorphous dispersions; micelles and lipid systems; cyclodextrin complexes. Human examples: micronized and micellar curcumin (Schiborr 2014), γ-cyclodextrin curcumin (Purpura 2018) Human PK/absorption	Solubility in biorelevant media; static in vitro digestion (INFOGEST 2.0, "not suitable for simulating digestion kinetics"); then human PK	Recrystallization and precipitation; excipient legal status for supplements; dispersibility mistaken for solubility
Absorption: permeability and efflux	Low absorption number despite dissolved drug; efflux in cell models	Lipid systems that use lipid absorption pathways; efflux inhibition. Piperine inhibited P-glycoprotein transport of digoxin (IC₅₀ 15.5 µM) and CYP3A4 (K_i 36–77 µM) in vitro (Bhardwaj 2002) In vitro	Caco-2 and similar models, then human PK	The same inhibition can raise exposure to co-administered drugs that are P-glycoprotein or CYP3A4 substrates
Transformation: first-pass metabolism	Mostly conjugated metabolites in plasma; low parent compound despite good absorption	Metabolism inhibitors such as piperine: 20 mg with 2 g curcumin raised curcumin bioavailability by 2000% in volunteers (Shoba 1998) Human PK/absorption; structural modification creates a different ingredient	PK measuring parent compound and metabolites with a validated, non-hydrolyzing assay	Drug interactions; piperine-containing turmeric products in liver-injury cases Documented safety concern; a modified molecule may need an NDI notification (NDIN)
Stability in the product or the gut	Loss on storage or in simulated gastric or intestinal fluid	Encapsulation, antioxidants, pH and oxygen control (see Emulsions and Drying)	Stability-indicating assays (ICH Q2(R2) section 2.4); stability studies under defined conditions	Stability testing is not a validated shelf life until real-time data support it

Schiborr C, et al. 2014; Purpura M, et al. 2018; Brodkorb A, et al. Nat Protoc. 2019;14:991–1014; Bhardwaj RK, et al. J Pharmacol Exp Ther. 2002;302(2):645–650; Shoba G, et al. 1998; Halegoua-DeMarzio D, et al. 2023; ICH Q2(R2) (2023).

Explanation, details and sources

The table pairs the step that limits F with formulation families (described on Delivery), the experiment that would show the step was fixed, and what can go wrong. Examples cite human pharmacokinetic studies where they exist.

Proposed capability

EUTECNIQ would diagnose the limiting step before choosing a technology, and would test exposure and interaction risk together

None of the following exists yet. Human studies would be run by qualified clinical partners under ethics approval.

Find the bottleneck first, then the technology, the assay and the interaction check, before any human studyProposed sequence

Technology families: Delivery. Assay development and validation: Method development.

Text version of the six steps

Characterize the moleculeSolubility, log P, stability, metabolism pathways from the literature
Locate the bottleneckDo, An, Dn and MAD screening; bioaccessibility in vitro
Match the technologyOnly families that act on that step (Delivery)
Develop the assayParent and metabolites, validated (Method development)
Check interactions and safetyCYP and transporter liabilities; exposure versus safety margin
Partner a human PK studyDose-matched comparator, defined analyte, adequate sample size

Planning a bioavailability claim or a formulation?

Share the ingredient and the target. EUTECNIQ would identify the likely limiting step and the evidence a credible claim would need.

Discuss bioavailability