Lipid-Based Nanocarriers for Oral Nutraceutical Delivery: Absorption Mechanisms, Manufacturing Routes and a Critical Appraisal of the Human Pharmacokinetic Evidence

1. Introduction

Demand for nutraceuticals has grown steadily, and with it the expectation that a capsule or sachet will deliver a measurable biological effect. The difficulty is that a large share of the compounds that perform well in cell culture and animal work are poorly absorbed in humans. Curcumin is the best documented case: it is almost insoluble in water, it is extensively conjugated in the intestinal wall and liver, and it is cleared quickly, so that conventional extracts produce plasma levels close to or below the limit of detection even at gram doses [1,2]. Similar barriers apply, in different combinations, to resveratrol, quercetin, catechins, coenzyme Q10, carotenoids and long-chain omega-3 fatty acids, and a different but related problem applies to water-soluble but chemically fragile actives such as ascorbic acid and glutathione [3,4].

Lipid-based carriers are attractive for food and supplement use because their building blocks (phospholipids, triglycerides, food-grade surfactants) are biocompatible and digestible and because several of them can be produced with equipment already used in food processing [5,6]. Commercial products now carry labels such as liposomal, micellar, phytosome, solid lipid particle or self-emulsifying, often with claims of many-fold higher bioavailability. Three recent reviews and trial reports, which together form the starting point of this article, illustrate the field from complementary angles: a review of curcumin bioavailability strategies and their clinical evidence [2], a review of lipid-based nanocarriers for nutraceuticals [5], and a randomised crossover trial of spray-dried liposomal vitamin C [7].

The aim of this review is not to repeat those accounts but to connect them: to explain why lipid carriers work when they do, to compare the carrier classes from a manufacturing as well as a scientific point of view, and to appraise critically how much of the reported human benefit survives scrutiny of trial design. The perspective is that of a manufacturer that formulates and produces nutraceutical products, and the review therefore gives particular weight to questions of scale, stability, powder conversion and claim substantiation.

2. Methods of Literature Selection

This is a narrative review. The three source publications above were read in full, and their reference lists were used to identify the primary pharmacokinetic and formulation studies they cite. Additional primary studies, reviews and regulatory documents published up to 2025 were identified through targeted searches of public bibliographic databases and journal websites using combinations of the terms liposome, niosome, solid lipid nanoparticle, nanostructured lipid carrier, self-emulsifying, micellar, pharmacokinetic, bioavailability, curcumin, vitamin C, glutathione and nanomaterial guidance. Human pharmacokinetic studies were prioritised over animal and in vitro work. Because the included studies differ widely in dose, analyte, sampling period and statistical reporting, no meta-analysis was attempted; ratios are reported as given by the original authors.

3. Why Oral Nutraceuticals Fail to Reach the Circulation

To reach the blood, an orally ingested compound must survive the stomach, dissolve in intestinal fluid, cross the mucus layer and the epithelium, avoid being pumped back by efflux transporters, and escape first-pass conjugation in the enterocyte and liver [3,8]. Each step can be rate-limiting, and different compounds fail at different steps.

For lipophilic molecules the first barrier is dissolution. Curcumin dissolves in water only to about 0.6 micrograms per millilitre, compared with about 1 milligram per millilitre in ethanol [2,9]. Undissolved crystals cannot be absorbed, so most of a conventional dose passes through the gut unused. The second barrier is biotransformation. Curcumin is a substrate of uridine diphosphate-glucuronosyltransferases and sulfotransferases, and of the efflux pump P-glycoprotein; most of what is absorbed appears in plasma as glucuronide and sulfate conjugates, which are generally regarded as far less active than the parent compound [2,10]. In rats given a 500 milligram per kilogram oral suspension, peak plasma curcumin reached only about 0.06 micrograms per millilitre [11].

Hydrophilic but unstable actives fail differently. Ascorbic acid is oxidised to dehydroascorbic acid and then irreversibly hydrolysed, a process accelerated by heat, light, oxygen, alkaline pH, water activity and transition metal ions such as iron and copper [4,7]. Its intestinal uptake is carrier-mediated and saturable, so higher doses yield proportionally less absorption [12]. For these molecules the role of a lipid carrier is less about dissolution and more about protection during manufacture, storage and transit, and about offering an additional, non-saturable uptake route.

4. How Lipid Carriers Increase Absorption

4.1. Three Mechanisms

Lipid excipients raise the absorption of poorly soluble compounds through three main mechanisms [5,8]. First, digestion of the carrier lipids by gastric and pancreatic lipases produces monoglycerides and free fatty acids that combine with bile salts and phospholipids to form mixed micelles and vesicles. These colloidal structures hold the active in solution at the epithelial surface and prevent it from precipitating as the formulation is diluted. Second, long-chain lipids are re-assembled in the enterocyte into chylomicrons, which leave the cell through the intestinal lymphatics rather than the portal vein; a lipophilic active carried in this route reaches the systemic circulation via the thoracic duct without first passing through the liver. Third, lipids and some surfactants alter the activity of intracellular lipid-binding proteins and transporters, and certain surfactants inhibit P-glycoprotein, which changes how the enterocyte handles the co-administered compound [8,13].

4.2. Routes of Particle Uptake

Whether intact nanocarriers cross the epithelium has been debated for decades. Paracellular passage between cells is restricted by tight junctions and is relevant mainly for small hydrophilic molecules, although chitosan coatings can transiently open these junctions [14]. Specialised microfold cells overlying Peyer’s patches can transcytose particles, but they represent only about one per cent of the epithelial surface [15], which limits their quantitative contribution. Endocytosis by ordinary enterocytes appears more important: in Caco-2 monolayers, solid lipid nanoparticles entered cells through macropinocytosis and clathrin- and caveolae-dependent routes and were transported across without damaging tight junctions [16]. For most commercial products, however, the dominant effect is probably not uptake of intact particles but improved solubilisation and protection in the lumen. This interpretation is supported by a crossover trial in which the best-performing formulations were those with the highest post-digestive solubility and micellisation, while those aimed at blocking metabolism added little [17].

4.3. Implications for Formulation Design

These mechanisms lead to practical design rules. A carrier that is digested to mixed micelles quickly will tend to raise peak concentration, whereas one that resists digestion or adheres to mucus will tend to prolong exposure. Long-chain triglycerides favour lymphatic transport more than medium-chain triglycerides, although medium-chain oils often dissolve more active per gram [18]. Surface charge and coating determine interaction with mucus: positively charged chitosan derivatives adhere to the mucosa and can extend residence time, while dense hydrophilic coatings can help particles diffuse through the mucus layer [19,20]. The Delivery by Design framework formalises these choices into a staged process, beginning with the physicochemical properties of the active and of the final product and ending with performance testing and life-cycle optimisation [6,18].

5. Carrier Classes

Table 1 summarises the main lipid-based carriers used for nutraceuticals. The sections below discuss their strengths and limitations.

5.1. Liposomes

Liposomes are closed vesicles of one or more phospholipid bilayers surrounding an aqueous core. Their main attraction is versatility: hydrophilic actives sit in the core and lipophilic actives in the bilayer [5,21]. Their main weakness is physical and chemical fragility. Unsaturated phospholipids oxidise, ester bonds hydrolyse, and vesicles aggregate, fuse and leak their contents during storage and in the stomach [5,22]. Several strategies address this. Polysaccharide coatings such as pectin, chitosan and cationic guar gum add electrostatic and steric protection; cationic guar gum, for example, improved curcumin retention at 70 degrees Celsius [23]. The phospholipid source matters: liposomes made from bovine milk phospholipids, which are more saturated and rich in sphingomyelin, were smaller (about 163 nanometres) and more strongly charged (about −26.7 millivolts) than those made from krill phospholipids (about 212 nanometres and −15.2 millivolts), and retained curcumin better [24]. Block copolymer (Pluronic) modification raised in vitro bioaccessibility of curcumin from 26.9 to 43.3 per cent [25]. Chitosan-coated catechin liposomes gave 1.37-fold higher bioavailability than uncoated liposomes and 2.12-fold higher than catechin solution in rats [14].

Two variants deserve mention. Proliposomes are dry phospholipid-coated granules that form liposomes on hydration; a quercetin proliposome raised peak concentration about 6.5-fold and area under the curve about 12-fold in an animal study [26]. Bilosomes incorporate bile salts into the bilayer, which improves resistance to intestinal bile; a mangiferin bilosome increased aqueous solubility about seven-fold [27].

Conventional preparation by thin-film hydration uses chloroform or other organic solvents whose residues are undesirable in food products, and scale-up is not straightforward [28,29]. Solvent-free alternatives include heating methods in which phospholipids, active and a polyol are stirred at 40 to 60 degrees Celsius under nitrogen and then annealed above the phase transition temperature [30,31], ethanol injection with subsequent removal of ethanol, high-pressure homogenisation and microfluidisation, and supercritical carbon dioxide processes [5,28].

5.2. Niosomes

Niosomes replace phospholipids with non-ionic surfactants such as sorbitan esters and polysorbates, usually with cholesterol. They are cheaper and chemically more stable than liposomes and can carry both hydrophilic and lipophilic actives [5]. A niosome carrying vitamin D3 and ferrous sulfate had a zeta potential of about −58 millivolts and remained stable at 20 degrees Celsius, but lost encapsulated material on refrigeration [32]. Niosomes of polysorbate 60 and cholesterol protected epigallocatechin gallate under simulated intestinal conditions, although particle size rose from about 60 to 178 nanometres once bile salts and pancreatin were added [33]. The main limitation for nutraceutical use is the surfactant load, which affects taste, regulatory acceptability and consumer perception; human pharmacokinetic data for oral nutraceutical niosomes are lacking.

5.3. Solid Lipid Nanoparticles and Nanostructured Lipid Carriers

Solid lipid nanoparticles are emulsions in which the liquid oil has been replaced by a lipid that is solid at body temperature, typically 0.1 to 30 per cent lipid stabilised by 0.5 to 5 per cent surfactant, with particle sizes of about 40 to 1000 nanometres [34,35]. The solid matrix slows release and protects the active, but on storage the lipid tends to recrystallise into a more ordered form that expels the payload. Nanostructured lipid carriers, the second generation, blend solid and liquid lipids (typically from 70:30 to 99.9:0.1) to create an imperfect matrix with higher loading and less expulsion [34,36]. Both can be made by hot or cold high-pressure homogenisation at 100 to 2000 bar, a mature and scalable food-industry process that does not require organic solvents [35].

Animal data are encouraging. Curcumin solid lipid nanoparticles containing a P-glycoprotein modulator gave a relative bioavailability of about 940 per cent compared with a suspension in rats [13]. Coating with N-trimethyl chitosan raised peak curcumin in mice from about 0.24 micrograms per millilitre (solution) to 1.21 micrograms per millilitre, compared with 0.58 for uncoated particles [20], and an N-carboxymethyl chitosan coating roughly doubled peak concentration and delayed peak time [37]. In mice, curcuminoid solid lipid nanoparticles and nanostructured lipid carriers increased peak concentration about 4.5-fold and 3.4-fold respectively [38]. Controlled human pharmacokinetic studies of food-grade solid lipid nanoparticles are, however, scarce; the best known human data concern a solid lipid curcumin particle product, which produced measurable unconjugated curcumin where native curcumin did not, in a cohort of six volunteers [39].

5.4. Nanoemulsions, Micelles and Self-Emulsifying Systems

Oil-in-water nanoemulsions are the most widely recommended system for food applications because they can be made from inexpensive plant oils and proteins by homogenisation or microfluidisation, can be dried into powders and show high in vitro bioaccessibility [2,18,40]. Micelles are smaller (about 5 to 20 nanometres), optically clear and thermodynamically stable, but they require high surfactant concentrations; one commercial micellar curcumin consists of about 7 per cent curcumin and 93 per cent polysorbate 80 [2].

Self-emulsifying delivery systems are anhydrous mixtures of oil, surfactant and co-solvent that form a fine emulsion spontaneously on contact with gastrointestinal fluid. They avoid the storage instability of pre-formed emulsions, can be filled into capsules or adsorbed onto solid carriers by spray drying, and have improved the absorption of curcumin, tocotrienols, coenzyme Q10 and omega-3 esters [41,42,43]. Adding mucoadhesive polymers can prolong residence time [19]. A caution comes from metabolite profiling in rats: after a curcumin self-emulsifying formulation, most of the curcumin that crossed the intestine had already been converted to curcumin glucuronide in the intestinal wall [44]. Faster dissolution therefore does not by itself protect against presystemic conjugation.

6. From Liquid Dispersion to Stable Powder

Most nutraceutical formats used in Asia and elsewhere, including capsules, tablets, sachets, effervescent granules and premixes, require a dry ingredient. Liquid liposomal dispersions are prone to oxidation, aggregation and leakage, and need preservatives and often cold storage; drying removes these constraints and reduces transport costs [22,45]. Freeze drying is gentle on heat-sensitive actives but is slow and expensive, and ice crystals can rupture bilayers unless cryoprotectants are added. Spray drying is continuous and cheaper and produces a powder directly, but exposes the dispersion to heat and shear [22,46].

The spray-dried liposomal vitamin C described by Żmuda and colleagues illustrates what a well-characterised powder looks like [7]. Ascorbic acid solution was mixed under reduced pressure with a 1:1 blend of sunflower phosphatidylcholine and glycerol, which formed liposomes spontaneously without organic solvents. Maltodextrin was added and the suspension was dried in a disc spray dryer at temperatures not exceeding 160 degrees Celsius, giving a powder of at least 25 per cent ascorbic acid. Scanning electron microscopy showed intact spherical microcapsules with a median diameter of about 23 micrometres. Cryogenic transmission electron microscopy of the rehydrated powder showed mainly large unilamellar vesicles with continuous bilayers, together with some multivesicular structures. Centrifugal sedimentation gave a median vesicle size of 262 nanometres with a narrow distribution (span 0.500), and the zeta potential was about −41 millivolts, indicating a low tendency to aggregate. The authors also report a 22 per cent higher transport across Caco-2 monolayers relative to free ascorbic acid.

Two points from this case study generalise. First, the carbohydrate shell that forms around the liposomes during spray drying acts as a second barrier, protecting both bilayer and active; maltodextrin and similar carriers are therefore functional, not merely fillers. Second, the authors deliberately did not use cryogenic microscopy to size vesicles, because the technique samples only a few microlitres and sorts vesicles by size across the grid; they used an ensemble method instead [7,47]. Size claims for commercial products should likewise be based on ensemble methods and reported with distribution width, not only a single mean.

Table 2 lists the critical quality attributes that should be specified and controlled for a lipid-based nutraceutical ingredient.

7. Human Pharmacokinetic Evidence

Table 3 collects human pharmacokinetic comparisons for lipid-based and related formulations of curcumin and vitamin C, the two actives with the most published data, together with selected examples from other nutraceuticals.

7.1. Curcumin

Curcumin is the most intensively studied case. Early work combined curcumin with piperine, which inhibits glucuronidation and P-glycoprotein; 2 grams of curcumin with 20 milligrams of piperine increased area under the curve about 20-fold in a small crossover study [51], yet a later study using 12 grams of curcumin with 5 milligrams of piperine detected no curcumin in any of ten volunteers [52]. Formulations combining curcuminoids with turmeric essential oil increased unconjugated curcumin exposure about seven-fold [53]. Lipid and surfactant systems produced the largest reported increases: micellar curcumin in polysorbate 80 raised total curcumin exposure 185-fold at 410 milligrams and 88-fold at 80 milligrams, with absorption differing between women and men [54,55]; a lipid-based formulation containing polyethylene glycol and caprylocaproyl macrogolglycerides raised total exposure 94-fold [56]; a phosphatidylcholine complex raised it about 18-fold [57]; and a polar-nonpolar natural turmeric matrix gave 5.6 to 7.3-fold more unconjugated curcumin than an essential-oil product and a phospholipid product [58]. Non-lipid technologies such as cyclodextrin inclusion and hydrophilic polymer carriers produced comparable apparent gains of 85 to 136-fold [59,60].

These numbers cannot be ranked against each other, for reasons discussed in Section 8. The most informative single study is the nine-arm, double-blind crossover trial of Flory and colleagues, in which twelve adults received each formulation normalised to 207 milligrams of curcumin [17]. Only micellar curcumin (57-fold) and the gamma-cyclodextrin complex (30-fold) significantly increased area under the curve compared with native curcumin; the liposomal, phytosomal, turmeric-oil, piperine and submicron-particle products did not differ significantly from native curcumin. In every participant and after every product, only conjugated curcumin was detected. In vitro digestion of the same products showed that micellar and cyclodextrin curcumin were the only ones with substantial post-digestive solubility and micellisation, which supports solubilisation in the lumen as the dominant mechanism.

7.2. Vitamin C

For ascorbic acid, four human trials compared liposomal and conventional forms (Table 3). Ratios of area under the curve ranged from about 1.3 to 1.8 [7,61,63,64]. A 2025 scoping review identified ten eligible human studies, nine of which reported higher peak and total plasma exposure for liposomal forms, but noted that elimination, cellular uptake and biological effects had rarely been measured and that participants with low baseline vitamin C status had not been studied [67]. In the spray-dried powder study, 1000 milligrams of liposomal vitamin C produced an area under the curve of 445 compared with 342 milligram-hours per litre for the conventional form (p = 0.013), a 30 per cent higher concentration at 24 hours, and a numerically but not significantly higher peak concentration (28.0 versus 22.5 milligrams per litre, p = 0.057) [7]. Time to peak did not differ, which was also the finding of a smaller study using a phospholipid-fibre powder [65]. The modest size of these gains is consistent with what would be expected for a well-absorbed, transporter-dependent molecule: the liposome mainly adds protection and prolonged release rather than overcoming a solubility barrier.

7.3. Other Actives

Liposomal glutathione at 500 or 1000 milligrams per day for one month raised glutathione stores in several blood compartments within one to two weeks and increased natural killer cell cytotoxicity, in an uncontrolled pilot study of twelve adults [66]. Self-emulsifying formulations of coenzyme Q10 and of omega-3 ethyl esters, and a dispersion technology for trans-resveratrol, have each shown higher plasma exposure than standard forms in healthy volunteers [42,43,68]. Sucrosomial iron, ferric pyrophosphate carried in a phospholipid and sucrose ester matrix, is one of the few lipid-carried products with a substantial body of clinical efficacy data rather than pharmacokinetic data alone, and has been reported to be absorbed partly as intact particles through enterocytes and microfold cells with better gastrointestinal tolerance than iron salts [69].

8. Critical Appraisal of the Evidence

8.1. Total Versus Unconjugated Analyte

Most curcumin pharmacokinetic studies treat plasma with glucuronidase and sulfatase before analysis and therefore report total curcumin, the sum of parent compound and conjugates. Because the conjugates dominate, this can overstate the unconjugated, presumably active, form by ten-fold or more [2,10]. A formulation that increases total curcumin may simply be delivering more glucuronide. Claims of bioavailability should state which analyte was measured, and preference should be given to unconjugated concentrations.

8.2. Unequal Comparator Doses and Dose Normalisation

Several frequently quoted studies compared a few hundred milligrams of formulated curcumin with 1.3 to 1.8 grams of native curcumin and then normalised the result to dose [57,59,60]. Because absorption of native curcumin is not linear with dose, this procedure can inflate the apparent fold increase. Trials that administer equal doses of active, as in Flory and colleagues [17] and the vitamin C trials [7,64], give more defensible estimates.

8.3. Small Cohorts, Inter-Individual Variability and Funding

Most pharmacokinetic studies in Table 3 enrolled twenty-four or fewer participants, one vitamin C study used separate groups of ten rather than a crossover, and several analysed fewer participants than they enrolled. Absorption varies widely between individuals, including between sexes [12,54]. Crossover designs reduce but do not remove this problem. Many studies were sponsored by, or conducted with, the owner of the formulation; in the spray-dried vitamin C trial, eight of the twelve authors were employees of the funding company or of a second commercial company, a fact disclosed transparently [7]. Industry sponsorship does not invalidate data, but it raises the importance of independent replication and of pre-registration, which that trial did provide.

8.4. The Label Does Not Define the Product

The term liposomal is applied to products ranging from carefully characterised vesicles to simple mixtures of lecithin and active. In the eight-way curcumin comparison the liposomal product performed no better than native curcumin [17], and comparable formulations made by different methods can behave differently in vivo [7]. Analyses of marketed curcumin products have also highlighted gaps between the technology claimed and the performance achieved [70,71]. A carrier claim should therefore be supported by product-specific characterisation (Table 2) and, ideally, by product-specific human data.

8.5. Exposure Is Not Efficacy

Higher plasma levels are a necessary but not sufficient condition for benefit. Micellar curcumin taken at 241 milligrams per day for six weeks raised plasma curcumin markedly but did not change blood lipids, inflammatory markers or glucose in moderately hyperlipidaemic adults [72], and in glioblastoma patients it produced the highest intratumoral curcumin then reported, yet still far below concentrations active in cell culture [73]. Clinical trials with positive outcomes exist for several formulations, including in knee osteoarthritis and non-alcoholic fatty liver disease [2], but formulation-specific efficacy evidence remains much thinner than pharmacokinetic evidence.

9. Safety and Regulatory Considerations

Lipid carriers built from dietary phospholipids and triglycerides are generally regarded as low risk. Attention is needed for three components. Synthetic surfactants such as polysorbates may be present at high ratios to the active in micellar products [2]; their chronic intake at those ratios has been less studied than that of the actives. Residual organic solvents from conventional liposome preparation are avoidable through solvent-free processes and should be specified and tested [28,29]. Finally, particles in the nanometre range may fall within nanomaterial definitions. The European Food Safety Authority updated its guidance on risk assessment of nanomaterials in the food and feed chain in 2021, setting out physicochemical characterisation, in vitro degradation, toxicokinetic and genotoxicity testing [74]; carriers that are digested quickly to their molecular constituents may qualify for a reduced testing burden, but this must be demonstrated. For pharmaceutical liposomes, the United States Food and Drug Administration has issued guidance covering chemistry, manufacturing and controls and human bioavailability studies [75], which provides a useful model for the voluntary characterisation of nutraceutical liposomes. In the United Kingdom, the Committee on Toxicity has begun reviewing supplement formulations designed to increase oral bioavailability, using liposomal vitamin C as a case study and noting that only a handful of small controlled human studies exist [62].

The hormetic behaviour of polyphenols, with stimulatory effects at low exposure and inhibitory or toxic effects at high exposure, adds a further consideration: a technology that raises systemic exposure many-fold also changes the safety margin established with conventional forms [2]. Safety data generated with native extracts should not be transferred automatically to highly bioavailable versions.

10. A Minimum Reporting Standard for Formulation Trials

Based on the weaknesses identified above, the following minimum standard is proposed for future human pharmacokinetic studies of lipid-based nutraceutical formulations.

• Full disclosure of the formulation: carrier composition and ratio to active, method of manufacture, and verified active content of the batch tested.
• Characterisation of the batch used in the trial, including particle or vesicle size by an ensemble method with distribution width, zeta potential, encapsulation efficiency and, for dried products, evidence that vesicles reform on rehydration.
• Equal doses of active in test and comparator arms, with any dose normalisation justified and reported alongside raw values.
• Measurement of the unconjugated active, with conjugates reported separately where relevant, using a validated method.
• Randomised crossover design with adequate washout, a sample size justified by a power calculation, and pre-registration in a public trial registry.
• Reporting of area under the curve, peak concentration, time to peak and late time-point concentrations with variability and confidence intervals for the ratio, not only p-values.
• Standardised dietary control, including the fat content of meals around dosing, and analysis by sex where numbers allow.
• Transparent statement of funding and of any employment or ownership links between authors and the formulation owner.

Adoption of these points would make it possible to compare technologies on a common basis, a comparison that the current literature does not allow.

11. Future Directions

Several developments are likely to shape the field. First, independent multi-arm comparisons at equal dose, of the kind reported by Flory and colleagues, are needed for other actives such as coenzyme Q10, resveratrol, quercetin and carotenoids. Second, standardised in vitro digestion models, such as the international consensus static protocol [49], coupled with Caco-2 permeability [50], could be used as a screening tier to select formulations before human testing, provided that their predictive value is validated against human data. Third, solvent-free, continuous manufacturing processes and dry formats, including spray-dried liposomes and solid self-emulsifying systems, are likely to replace liquid dispersions for most supplement applications. Fourth, co-formulation of lipid carriers with mucoadhesive or permeation-modifying biopolymers, and with matrix components of the source plant such as turmeric oils and fibres, deserves systematic study. Finally, the interaction between poorly absorbed polyphenols and the gut microbiota means that unabsorbed fractions may not be wasted; formulations designed for systemic exposure may alter colonic delivery and microbial metabolism, an effect that should be measured rather than assumed [2].

12. Conclusions

Lipid-based nanocarriers can substantially increase the systemic exposure of poorly soluble and labile nutraceuticals, mainly by keeping them in solution in the intestinal lumen, protecting them from degradation and, for some carriers, engaging lymphatic transport. The size of the benefit depends strongly on the active: large for lipophilic, poorly soluble molecules such as curcumin, modest for well-absorbed but fragile molecules such as ascorbic acid. Published fold increases are, however, highly sensitive to trial design, and the most rigorous equal-dose comparison found that only two of eight curcumin technologies outperformed native curcumin, with no unconjugated curcumin detectable after any of them. For manufacturers, the practical conclusions are to select the carrier according to the specific absorption barrier of the active, to convert liquid carriers into well-characterised powders using solvent-free and food-compatible processes, to specify and control critical quality attributes, and to substantiate bioavailability claims with product-specific, equal-dose, independently replicated human data reported to a common standard.