Xiong Huang

Multicentre open-label non-inferiority randomised phase 3 trial at 14 Chinese centres comparing oral realgar-Indigo naturalis formula plus all-trans retinoic acid with intravenous arsenic trioxide plus ATRA in non-high-risk acute promyelocytic leukaemia. The oral realgar-based regimen was non-inferior for two-year event-free survival. This is the strongest human evidence for realgar in any indication, and it concerns a specific standardised manufactured formula, dosed by weight, used as a hospital-supervised substitute for intravenous arsenic in one diagnosed leukaemia; it does not support crude realgar for any traditional indication.

Mechanistic dissection of the three-component realgar-Indigo naturalis formula in APL cell lines and a mouse model, assigning roles to tetraarsenic tetrasulfide from realgar, indirubin from Indigo naturalis and tanshinone IIA from Salvia miltiorrhiza. Realgar arsenic was identified as the principal agent, degrading the PML-RARA oncoprotein, with the other two components enhancing differentiation and increasing arsenic uptake through upregulation of the aquaglyceroporin AQP9. The corollary is that co-prescribed herbs can raise arsenic absorption rather than temper it.

Very small single-centre randomised trial (19 children, 8 given oral realgar-Indigo naturalis formula and 11 intravenous arsenic trioxide, both with ATRA and conventional chemotherapy) reporting 100 percent complete remission and 100 percent two-year disease-free survival in both arms with no significant differences. The sample is far too small to establish comparative safety, and the finding applies only to a standardised manufactured formula given inside a paediatric oncology protocol.

Forensic case report with a 30-year Chinese literature review. A 35-year-old man with severe psoriasis died of acute arsenic poisoning after applying a folk ointment containing mainly realgar to affected skin for about four days; the ointment assayed at about 6 percent arsenic, blood arsenic was 1.76 micrograms per mL and skin arsenic 4.71 micrograms per g, and autopsy showed severe renal tubular necrosis. The authors identify seven further similar cases in the preceding 30 years. This is the direct refutation of the common assumption that topical realgar cannot kill, and it matters because damaged skin absorbs far more than intact skin.

Rats given the realgar-containing paediatric patent medicine Xiao-Er-Zhi-Bao-Wan orally for 7 and 14 days accumulated dimethylarsinic acid in kidney from 18.57 to 22.74 ng/g, about tenfold above controls, with total kidney arsenic roughly double control values, and showed mild histopathological injury in liver and kidney by day 14. The authors conclude that dimethylarsinic acid rather than total arsenic is the appropriate toxicity marker for realgar-containing medicines. Preclinical, but it is a preclinical study of a product actually given to children.

Simulated-gastrointestinal extraction of the widely sold realgar-containing patent medicine Niu Huang Jie Du Pian, combined with human urinary excretion measurement, quantifying how much of the very large total arsenic content is actually released and absorbed. The study establishes that the arsenic in these pills is not inert mineral ballast, while also showing that total arsenic on its own overstates the delivered dose.

Mouse acute toxicity comparison showing that realgar and the realgar-containing pill Liu-Shen-Wan have markedly higher lethal doses than soluble sodium arsenite or arsenate, consistent with the low solubility of the sulfide. The result is often quoted as reassurance; read correctly it says only that realgar is less acutely toxic than one of the most acutely toxic arsenicals known, and it does not address chronic accumulation, carcinogenicity, or what happens when the mineral is heated.

Repeat-dose rodent comparison of realgar, cinnabar and the compound pill An-Gong-Niu-Huang Wan against sodium arsenite and mercuric chloride, finding substantially less renal injury and less renal metal accumulation from the mineral drugs. Again this is a relative statement against highly soluble reference toxicants; nephrotoxicity was reduced, not absent, and the comparison does not license unsupervised chronic use.

Cell-culture and mouse evaluation of the realgar-containing patent medicine Niu-Huang-Jie-Du Pian alongside reference arsenicals, showing lower cytotoxicity and lower acute lethality than soluble arsenite while confirming that the preparation delivers genuine arsenical exposure rather than being biologically inert.

Two toxicologically confirmed cases of systemic heavy-metal poisoning from mucocutaneous application of Chinese mineral medicines, one of them from a hong-dan mixture containing lead tetraoxide, arsenic and mercury applied into an anal fistula, producing perianal gangrene, anaemia, hair loss and a peripheral neuropathy that persisted four years after DMPS chelation. The authors conclude that short-term application of such preparations to damaged or infected tissue causes serious systemic poisoning.