Sabroxy

Oroxylum indicum is a small deciduous tree of South and Southeast Asia. It is used in Ayurvedic medicine and is an ingredient of the multi-herb formulas Chyawanprash and Dashamoolarishta [1]. More than 100 compounds have been identified in the plant, and flavonoids make up most of them [3]. Sabroxy is Sabinsa's methanolic extract of the bark. It is standardised to 10% oroxylin A, 6% chrysin and 15% baicalein, so a 500 mg capsule holds about 50 mg oroxylin A [1][4].

The human evidence is one trial [2]. In it, 82 Australians aged 60 to 85 with self-reported memory decline took 500 mg Sabroxy or placebo twice daily for 12 weeks; 73 completed [1]. The pre-specified episodic memory composite improved more on Sabroxy (p = 0.015). Of its seven parts, only immediate word recall (+5.4 vs −1.0 percentage points, d = 0.52) and numeric working memory accuracy differed from placebo. Delayed word recall, word and picture recognition, location recall and the MoCA memory index did not. Sabroxy participants also learned the location task faster over its five trials. Speed of performance, MoCA total score, the Cognitive Failures Questionnaire and quality of life did not differ between groups [1].

The trial was funded by Sabinsa, whose founder is a co-author. The authors called it exploratory and did not correct for multiple comparisons [1].

In mice given chemotherapy, the extract protected memory and brain antioxidant status. Oddly, the lower dose (250 mg/kg) worked better than the higher one, which mostly missed significance [5]. A fruit extract of the same plant reduced memory loss and preserved hippocampal neurogenesis in D-galactose-aged rats [6].

How Sabroxy might work in people is unknown. Most mechanistic data concern oroxylin A, studied on its own in rodents.

• GABA-A antagonism. Oroxylin A binds the benzodiazepine site of the GABA-A receptor (IC50 about 1.1 µM) as an antagonist. In mice it blocked diazepam's anti-anxiety and muscle-relaxant effects but not its sedative or anticonvulsant ones [7]. It also reversed scopolamine-induced amnesia, an effect that GABA-A agonists cancelled [8].
• Dopamine reuptake. In vitro, oroxylin A inhibited dopamine uptake, as methylphenidate does, but not noradrenaline uptake. It improved inattention and impulsivity in spontaneously hypertensive rats, a model of ADHD, and the dopamine blocker haloperidol weakened that effect [9].
• BDNF and neurogenesis. Oroxylin A raised mature BDNF in the mouse hippocampus and helped memory consolidation; blocking the BDNF receptor TrkB stopped that effect [10]. It also increased new neurons in the dentate gyrus [11] and protected memory after reduced blood flow to the brain [12]. Sabroxy itself raised BDNF expression about 8-fold in neuronal cells [4].

None of this has been confirmed in humans. In the trial, plasma BDNF rose on both Sabroxy and placebo. The authors suggest seasonal changes as a cause, and BDNF changes did not track cognitive changes [1]. Oroxylin A's oral bioavailability is low in rats [13], though it enters the mouse brain more readily than related flavones [14].

Sabroxy is made by a patented process: the bark is extracted with aqueous alcohol, purified and standardised to its three flavones by HPLC [4]. Generic Oroxylum indicum products may use leaves, fruit, stem or root and other solvents. Their flavone content will differ, and the Sabroxy trial does not apply to them [15].

Pure oroxylin A, made synthetically, is also in development in China as a drug tablet. It is in phase Ib/IIa trials for liver cancer [16].

In the human trial, 10 people on Sabroxy and 6 on placebo reported adverse events, mainly loose stools, nausea, bloating and headache. There were no serious events. Nine people withdrew, three for nausea or digestive complaints and one for persistent headaches on Sabroxy [1]. Blood pressure did not change. Weight fell 1.1 kg on Sabroxy versus 0.13 kg on placebo, a difference that missed significance (p = 0.058) [1].

In rodents, the extract caused no liver injury at 500 mg/kg a day for 4 weeks in mice [15]. A manufacturer-run programme of acute, subacute, subchronic and reproductive studies set a no-observed-adverse-effect level of 400 mg/kg, and the extract was not mutagenic [19]. In healthy volunteers, pure oroxylin A at up to 2,400 mg a day caused no serious adverse events, though digestive complaints were more common with repeated dosing [18].

Oroxylin A inhibited the drug transporters OAT1, OAT3 and BCRP in vitro at sub-micromolar concentrations [20]. Whether a supplement dose does this in people has not been studied.