Arsenic Trioxide

證據等級: L5

預測適應症: 10 個

Arsenic Trioxide: From Acute Promyelocytic Leukemia to Myelodysplastic Syndrome

One-Sentence Summary

Arsenic trioxide (ATO) is a well-established treatment for Acute Promyelocytic Leukemia (APL), exerting its effect through apoptosis induction and forced differentiation of malignant promyelocytes. The TxGNN model predicts it may be effective across the Myelodysplastic Syndrome (MDS) spectrum — the single most evidence-rich predicted indication — supported by 24 clinical trials and 20 publications, including a 2025 randomized controlled trial and a 2023 systematic meta-analysis directly evaluating ATO in MDS. Among all 10 predicted indications, MDS (rank 6) carries the strongest clinical evidence base and an actionable “Proceed with Guardrails” recommendation.

Quick Overview

Item	Content
Original Indication	Acute Promyelocytic Leukemia (APL)
Predicted New Indication	Myelodysplastic Syndrome (MDS)
TxGNN Prediction Score	99.91%
Evidence Level	L2
Taiwan Market Status	✗ Not Marketed (TFDA)
Number of Authorizations	0
Recommended Decision	Proceed with Guardrails

Why Is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available from the structured source (DrugBank query pending). Based on established pharmacology, arsenic trioxide (As₂O₃) is an inorganic cytotoxic compound with multi-target activity originally proven in APL, where it degrades the PML-RARα fusion protein, generates reactive oxygen species (ROS) to trigger mitochondrial apoptosis, and forces malignant promyelocytes to mature into normal white cells.

APL and MDS share a common origin: both arise from clonal dysfunction of bone marrow myeloid progenitor cells. ATO’s mechanisms extend naturally into the MDS context through at least four pathways: (1) Apoptosis induction — downregulation of Bcl-2/NF-κB anti-apoptotic signalling in dysplastic myeloid clones; (2) Differentiation promotion — aberrant myeloid precursors are pushed toward maturation, reducing blast burden; (3) Immune modulation — correction of abnormal Treg proportions and re-balancing of the IFN-γ/IL-17/TGF-β1 cytokine axis implicated in immune-mediated MDS variants; (4) Epigenetic synergy — cooperative DNA demethylation when combined with hypomethylating agents (decitabine or azacitidine), validated both in vitro (PMID 30898879) and in a 2025 RCT (PMID 40167011).

It is also notable that TxGNN simultaneously predicted multiple MDS-spectrum conditions among its top-10 outputs (unclassified MDS, refractory cytopenia of childhood, aregenerative anemia, 5q− syndrome), suggesting the model identifies a coherent mechanistic signal across the entire bone marrow failure landscape rather than an isolated statistical artefact. This convergent pattern strengthens confidence in the biological plausibility of ATO repurposing for MDS.

Clinical Trial Evidence

Trial Number	Phase	Status	Enrollment	Key Findings
NCT00093366	Phase 1/2	Completed	32	ATO + etanercept (TNF-α inhibitor) for advanced MDS; completed with efficacy and safety readouts, directly addressing ATO in MDS
NCT00454480	Phase 2/3	Completed	2,000	Largest trial in this data set; elderly AML and high-risk MDS patients, testing gemtuzumab ozogamicin, tipifarnib, and ATO-based combinations; provides Phase 3-level context
NCT02190695	Phase 2	Completed	92	Randomized three-arm study: decitabine vs. decitabine + carboplatin vs. decitabine + ATO for relapsed/refractory and elderly AML/MDS; provides direct comparative evidence for ATO add-on benefit
NCT00195104	Phase 1/2	Completed	87	ATO + low-dose cytarabine in high-risk MDS and poor-prognosis AML; 17% complete remission rate, tolerable toxicity profile
NCT00671697	Phase 1	Completed	13	Decitabine + ATO + ascorbic acid triple combination in MDS and AML; safety and feasibility confirmed
NCT00274781	Phase 2	Completed	30	ATO + gemtuzumab ozogamicin for advanced MDS; completed with safety and response data
NCT00621023	Phase 2	Completed	7	Pilot study of decitabine + ATO + ascorbic acid in MDS; triple combination safety confirmed
NCT00020969	Phase 2	Terminated	N/A	ATO monotherapy MDS-specific trial; terminated early, limiting conclusions but confirming biological interest
NCT00803530	Phase 2	Terminated	55	Multicenter ATO + ascorbic acid in MDS; terminated with partial accrual, provides safety signal data
NCT06778187	Phase 2	Recruiting	30	Oral ATO (Arsenol®) + ascorbic acid for previously untreated or relapsed/refractory TP53-mutated AML/MDS/CMML; active 2025 trial

Literature Evidence

PMID	Year	Type	Journal	Key Findings
37908176	2023	Meta-analysis	Hematology (Amsterdam)	Systematic review and component network meta-analysis of ATO-containing regimens in MDS; evaluates overall efficacy and adverse event profiles across multiple combination strategies
40167011	2025	RCT	Hematology (Amsterdam)	Randomized controlled trial of decitabine + ATO versus decitabine monotherapy in elderly high-risk MDS; demonstrates superior outcomes with the combination approach
20956016	2011	Phase 1/2 clinical	Leukemia Research	ATO + low-dose cytarabine in 49 previously untreated intermediate-2/high-risk MDS patients; 17% CR including patients with unfavorable cytogenetics
20425329	2006	Review	Curr Hematol Malignancy Rep	Comprehensive review of ATO as MDS treatment covering pro-apoptotic, antiproliferative, and anti-angiogenic mechanisms and clinical trial landscape
18282365	2007	Review	Clin Lymphoma Myeloma	Summary of ATO clinical data in leukemia and MDS; >80% CR in relapsed APL, extrapolation rationale to MDS
38816179	2024	Comparative clinical	Immunopharmacol Immunotoxicol	Comparative immunological characterization of oral vs. IV arsenic in MDS murine model; delineates immune reprogramming differences relevant to treatment optimization
30898879	2019	In vitro mechanistic	J Investig Med	Decitabine + ATO induces synergistic apoptosis in MUTZ-1 and SKM-1 MDS cells via endoplasmic reticulum stress; provides combination mechanistic rationale
22964015	2012	Mechanistic/Clinical	J Hematol Oncol	ATO + ascorbic acid modulates BCL2 family apoptotic gene expression in primary MDS bone marrow samples (ex vivo, n=12); validates in vivo apoptotic mechanism
16105982	2005	Mechanistic	Blood	NF-κB and FLIP upregulation in advanced MDS cells; ATO-induced apoptosis via NF-κB suppression validated in primary patient CD34+ cells
17920679	2008	Clinical	Leukemia Research	ATO + retinoic acid + thalidomide combination in higher-risk MDS; clinical efficacy and safety evaluation

Taiwan Regulatory Status

Arsenic trioxide is currently not marketed in Taiwan (TFDA). No domestic marketing authorizations were identified.

International Reference: ATO is approved as Trisenox® by the US FDA and EMA for the treatment of Acute Promyelocytic Leukemia (APL) — both relapsed/refractory and, in combination with ATRA, as front-line therapy for newly diagnosed low- to intermediate-risk APL.

Cytotoxicity

Item	Content
Cytotoxicity Classification	Conventional cytotoxic with targeted properties (inorganic arsenic compound; induces ROS-mediated apoptosis and differentiation in myeloid malignancies)
Myelosuppression Risk	Moderate — leukocytosis is characteristic during APL treatment (differentiation syndrome risk); in MDS, cytopenia worsening may occur early in treatment; CBC monitoring mandatory
Emetogenicity Classification	Low to moderate
Monitoring Items	CBC with differential (weekly during induction), liver function (ALT/AST/bilirubin), renal function (creatinine/BUN), electrolytes (K⁺, Mg²⁺, Ca²⁺), QTc interval via ECG before initiation and during treatment, blood glucose
Handling Protection	Must follow cytotoxic drug handling regulations; arsenic-containing compounds require dedicated containment, PPE, and waste disposal per hazardous drug protocols

Safety Considerations

Please refer to the SmPC for safety information.

Note: TFDA label warnings and contraindications could not be retrieved (Data Gap DG001 — blocking). Based on established pharmacology, the following safety signals are well-characterised for ATO and should be verified against the formal label before any clinical application: QTc prolongation (potentially fatal arrhythmia risk; avoid with other QT-prolonging agents), electrolyte disturbances (hypokalemia and hypomagnesemia increase arrhythmia risk), hepatotoxicity (transaminase elevation), and APL differentiation syndrome (relevant to APL indication; monitor for respiratory distress in MDS patients with high blast counts who may develop a similar syndrome).

Conclusion and Next Steps

Decision: Proceed with Guardrails

Rationale: ATO has demonstrated coherent multi-mechanism biological plausibility in the MDS setting, backed by a 2025 RCT and a 2023 network meta-analysis directly supporting its clinical utility — particularly in high-risk and elderly MDS in combination with decitabine. Multiple completed Phase 1/2 trials confirm a manageable safety profile in this population.

To proceed, the following is needed:

Resolve safety data gap first (blocking): Obtain and parse the TFDA drug label (SmPC) to complete the mandatory S1 safety gate assessment (DG001)
Clarify MOA data: Query DrugBank API for structured mechanism of action data to strengthen the mechanistic narrative (DG002)
Define patient population: Specify MDS subtype (IPSS-R low/intermediate vs. high-risk), TP53 mutation status, and prior treatment history as stratification variables — recent trials suggest TP53-mutated MDS as a particularly compelling subgroup
Select dosing strategy: Evaluate IV (conventional) vs. emerging oral formulation (Arsenol®, currently recruiting in NCT06778187 and NCT06670222) based on Taiwan patient access and safety preferences
Establish cardiac safety protocol: Pre-treatment ECG, electrolyte correction plan, and monitoring schedule (QTc) must be defined before initiating any Taiwan clinical investigation
Design a focused Taiwan feasibility study: Given the existing Phase 2 evidence base globally, a prospective observational registry or a small randomized Phase 2 in Taiwan-eligible elderly high-risk MDS patients (decitabine ± ATO) would be appropriate next steps
Disclaimer

This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.