Sang Zhi

The pivotal registration trial for the Ramulus Mori alkaloid tablet, and the strongest human evidence attached to any part of Morus alba. Six hundred patients with type 2 diabetes were randomised to the mulberry twig alkaloid tablet (n = 360) or acarbose (n = 240) for 24 weeks in a double-blind, double-dummy, noninferiority design. HbA1c fell 0.93 percent (95 percent CI -1.03 to -0.83) on the twig alkaloid and 0.87 percent (-0.99 to -0.76) on acarbose, a least-squares mean difference of -0.05 percent that met noninferiority. Treatment-related adverse events and gastrointestinal disorders were significantly less frequent on the twig alkaloid (p less than 0.01), with no difference in serious adverse events. Note that this is a purified alkaloid tablet, not a decoction of the crude twig.

Placebo-controlled counterpart to the acarbose trial: 200 patients randomised 1:1 to the Sangzhi alkaloid tablet or placebo for 16 weeks. HbA1c changed by -0.80 percent (95 percent CI -0.98 to -0.62) on active treatment against -0.09 percent (-0.27 to 0.09) on placebo, with 46.8 percent versus 21.6 percent reaching HbA1c below 7 percent. Fasting glucose, 1-hour and 2-hour postprandial glucose and postprandial area under the curve all separated significantly, but body weight and BMI did not. Adverse events, treatment-related adverse events and gastrointestinal disorders were more frequent on active treatment; serious adverse events were not.

Thirty patients on premixed insulin were randomised to insulin alone, insulin plus metformin, or insulin plus Sangzhi alkaloids for 12 weeks with continuous glucose monitoring. Both combination arms improved mean glucose, glycaemic variability, time in range, HbA1c, fasting and 1-hour postprandial glucose and lipid measures over insulin alone. The Sangzhi alkaloid arm was significantly better than both other arms for largest amplitude of glycaemic excursion, postprandial glucose excursion, mean amplitude of glycaemic excursion, dinner postprandial area under the curve and 2-hour postprandial glucose. The trial is small and open in design, so it supports the additive-hypoglycaemia interaction rather than proving superiority over metformin.

The earlier, much smaller phase 2 comparison that preceded the registration trial: 38 patients randomised to the twig alkaloid tablet (n = 23) or acarbose (n = 15) for 24 weeks. HbA1c fell significantly from baseline on the twig alkaloid (p less than 0.001) with no significant difference from acarbose (p = 0.652), and 1-hour and 2-hour postprandial glucose fell while fasting glucose did not change significantly in either arm. Gastrointestinal adverse events occurred in 1 of 23 on the twig alkaloid against 5 of 15 on acarbose. The sample is too small to support conclusions on its own and it should be read as supporting evidence to the 600-patient trial.

Sixty patients with type 2 diabetes inadequately controlled on oral agents were randomised to add-on Sangzhi alkaloids or canagliflozin for 12 weeks, assessed by flash glucose monitoring. Both arms cut HbA1c, fasting and postprandial glucose, 24-hour mean glucose and time above range, with comparable time in range (72.1 versus 73.9 percent) and no increase in time below range. The twig alkaloid arm gave significantly greater reduction in postprandial glucose excursion after breakfast and dinner, consistent with its alpha-glucosidase inhibitory mechanism. Registered as NCT05856578.

Pharmacokinetic and drug-interaction characterisation of the standardised twig alkaloid fraction. Absorption was rapid and linear over 25 to 200 mg/kg, with highest tissue concentrations in kidney, liver and aorta, then brown and subcutaneous fat. Apart from trace oxidation products of fagomine, no phase I or phase II metabolites were detected, and the fraction neither inhibited nor activated the major cytochrome P450 enzymes in human liver microsomes. The paper also documents the composition of the licensed extract: deoxynojirimycin 36.5 to 36.9 percent, fagomine 8.8 to 9.6 percent and 1,4-dideoxy-1,4-imino-D-arabinitol 5.8 to 7.6 percent.

Fifty-seven authenticated Morus alba samples (15 root barks, 11 twigs, 12 fruits, 19 leaves) were assayed for twelve marker compounds and analysed by principal component and heatmap methods. The four medicinal parts separated cleanly, with mulberroside A, taxifolin, kuwanon G and morusin marking root bark, chlorogenic acid, rutin, isoquercitrin, astragalin, quercitrin and skimmin marking leaf, and 4-hydroxycinnamic acid and oxyresveratrol marking twig. This is the direct analytical basis for treating Sang Zhi, Sang Bai Pi, Sang Ye and Sang Shen as chemically distinct drugs rather than as parts of one interchangeable herb.