Prasad · The British journal of nutrition 2026 · in vitro study and open-label parallel-arm pharmacokinetic trial · n=12

Liposomal glutathione outperforms plain glutathione in uptake, cell regeneration and systemic availability: evidence from cellular and human models.

Cited 0 times in the scientific literature.

Level 3 - non-randomized controlled study

Small, open-label parallel-arm human pharmacokinetic trial combined with in vitro cellular assays

PubMed 41559937 · doi:10.1017/S0007114526106254 · record verified 2026-08-26

What was done

Human embryonic kidney 293T cells were treated with a novel liposomal glutathione formulation (LipoDuo™) or plain glutathione to evaluate intracellular uptake, cell proliferation, and wound healing. In the clinical phase, a parallel-arm, open-label human trial evaluated 12 healthy subjects (n = 6 per group) receiving a single 1 g oral dose of either liposomal or plain glutathione, with plasma glutathione concentrations tracked over 24 hours via LC-MS/MS.

What was found

In vitro, liposomal glutathione showed ~1.9-fold higher cellular uptake at 6 hours (45% vs. 23%), up to 3-fold higher cell proliferation over control, and 100% wound closure at 24 hours compared to 59.8% for plain glutathione. In humans, liposomal glutathione reached a peak plasma concentration of ~1800 ng/ml (6-fold higher than plain), exhibited a bimodal absorption curve, and maintained plasma levels above 500 ng/ml at 24 hours.

Why it matters

Because standard oral glutathione is limited by gastrointestinal breakdown and poor cellular uptake, this study demonstrates that liposomal encapsulation substantially enhances oral bioavailability and cellular delivery in humans.

Limits

The clinical evaluation was small (n = 12 total, 6 per arm), open-label, non-randomized (or randomization not stated), and evaluated only a single acute dose without clinical endpoint validation. In vitro results in kidney cell lines may not reflect physiological antioxidant actions in intact human tissues, and statistical variance metrics were omitted from the abstract.

Cited by