Baricitinib

證據等級: L5

預測適應症: 10 個

Baricitinib: From Rheumatoid Arthritis to Myeloid Leukemia (Multi-Indication Analysis)

One-Sentence Summary

Baricitinib is a selective JAK1/2 inhibitor approved globally for rheumatoid arthritis, atopic dermatitis, and COVID-19, but currently not registered in Taiwan. The TxGNN model generated 10 predicted indications; the top-ranked prediction is colobomatous microphthalmia-rhizomelic dysplasia syndrome (score 99.94%, L5, no evidence), while myeloid leukemia (rank 6) carries the strongest evidence base with 1 active Phase 2/3 clinical trial and 4 publications (Evidence Level: L3). Due to absent Taiwan regulatory data and gaps in MOA and safety documentation, the overall recommendation across all top predictions is Hold pending data remediation.

Quick Overview

Item	Content
Original Indication	Not registered in Taiwan; globally approved for rheumatoid arthritis, atopic dermatitis, COVID-19
Predicted New Indication (Rank 1)	Colobomatous microphthalmia-rhizomelic dysplasia syndrome
TxGNN Prediction Score (Rank 1)	99.94%
Evidence Level (Rank 1)	L5 — AI prediction only, no clinical studies
Taiwan Market Status	✗ Not marketed (0 authorizations)
Number of Authorizations	0
Recommended Decision	Hold

Why Is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available in this Evidence Pack (Data Gap DG002). Based on known pharmacological knowledge, Baricitinib is a selective JAK1 and JAK2 inhibitor that blocks downstream STAT phosphorylation triggered by cytokines including IL-6, IL-4, IL-13, and IFN-γ. This broad immunomodulatory effect underpins its approved use in rheumatoid arthritis and atopic dermatitis.

The rank 1 TxGNN prediction — colobomatous microphthalmia-rhizomelic dysplasia syndrome — is an ultra-rare congenital developmental disorder driven by genetic mutations affecting ocular and limb morphogenesis. While the JAK-STAT pathway has some known role in embryonic ocular patterning, the mechanistic connection to JAK1/2 inhibition for this condition is distant extrapolation. The exceptionally high model score (0.999) very likely reflects knowledge graph topological proximity rather than a direct biological or therapeutic relationship, and should be treated with caution.

Among all 10 predicted indications, myeloid leukemia (rank 6, score 91.01%) presents the most scientifically plausible repurposing rationale. JAK1/2 inhibition can interrupt IL-3, GM-CSF, and IFN-γ-driven myeloproliferative signaling, and in the allogeneic hematopoietic stem cell transplantation (allo-HSCT) setting, Baricitinib may reduce graft-versus-host disease (GvHD) by suppressing donor T-cell JAK-STAT activation. This is currently being tested in an active Phase 2/3 trial (n=150).

Predicted Indications — Summary Table

Rank	Disease	TxGNN Score	Evidence Level	Recommendation
1	Colobomatous microphthalmia-rhizomelic dysplasia syndrome	99.94%	L5	Hold
2	Brachydactyly-syndactyly syndrome	99.94%	L5	Hold
3	Indolent plasma cell myeloma	93.31%	L5	Hold
4	WHIM syndrome	93.12%	L5	Hold
5	Plasma cell myeloma	91.83%	L4	Hold
6	Myeloid leukemia	91.01%	L3	Research Question
7	Meester-Loeys syndrome	88.21%	L5	Hold
8	Ganglioneuroblastoma	87.59%	L5	Hold
9	Heparin cofactor 2 deficiency	86.31%	L5	Hold
10	Vertebral anomalies and variable endocrine and T-cell dysfunction	84.68%	L5	Hold

Clinical Trial Evidence

Rank 1 — Colobomatous Microphthalmia-Rhizomelic Dysplasia Syndrome

Currently no related clinical trials registered.

Rank 6 — Myeloid Leukemia (Highest-Evidence Prediction)

Trial Number	Phase	Status	Enrollment	Key Findings
NCT06475820	Phase 2/3	Active, Not Recruiting	150	GvHD prevention using post-transplantation cyclophosphamide combined with abatacept, vedolizumab, and baricitinib in children and young adults with hemoblastosis after HSCT from unrelated or haploidentical donors. Primary endpoint is GvHD prevention; myeloid leukemia is the dominant underlying diagnosis. Completion expected April 2027.

Note: This trial’s primary endpoint is GvHD prevention, not direct anti-leukemic efficacy. Baricitinib’s role here is as a HSCT adjunct rather than a primary oncology agent.

Literature Evidence

Rank 5 — Plasma Cell Myeloma

PMID	Year	Type	Journal	Key Findings
34856779	2021	Case Report	Minerva Medica	Simultaneous treatment of critical COVID-19 and newly diagnosed high-risk multiple myeloma with thalidomide, methylprednisolone, tocilizumab, and baricitinib. Myeloma outcomes were incidental observations during COVID-19 management, not a primary efficacy endpoint. Indirect and opportunistic evidence only.

Rank 6 — Myeloid Leukemia

PMID	Year	Type	Journal	Key Findings
34730109	2021	Preclinical / Mechanism	Journal of Clinical Investigation	ADC-based conditioning combined with JAK inhibitors enables MHC-mismatched allogeneic HSCT in preclinical models. Supports the concept that JAK inhibition — including baricitinib — reduces immune rejection in the HSCT setting relevant to acute leukemia treatment.
36569952	2022	Review / Hypothesis	Frontiers in Immunology	Review proposing repurposing of BCL-2 and JAK1/2 inhibitors for HIV-1 and viral infection by targeting long-lived infected cells. Indirect mechanistic relevance to immune modulation in hematologic contexts.
35442720	2022	Observational / Diagnostic	JCO Precision Oncology	Nanopore mRNA sequencing for acute leukemia classification in low-resource settings. No baricitinib efficacy data; relevant only to leukemia disease characterization.
31816725	2020	Pharmacokinetic / Analytical	Talanta	LC-MS/MS method for monitoring 11 TKIs including ruxolitinib in CML. Baricitinib is not among monitored drugs; relevance is limited to TDM methodology context.

Taiwan Market Information

Baricitinib (DB11817) has no registered products in Taiwan (0 licenses). Regulatory authorization data cannot be populated.

Globally, Baricitinib is marketed as Olumiant (Eli Lilly) and is approved by the US FDA and EMA for:

Moderate-to-severe rheumatoid arthritis (adults)
Moderate-to-severe atopic dermatitis (adults and adolescents ≥2 years)
COVID-19 requiring supplemental oxygen (adults)

Taiwan TFDA prescribing information is unavailable (Data Gap DG001 — Blocking severity).

Safety Considerations

Detailed Taiwan-specific warnings and contraindications are unavailable pending TFDA SmPC retrieval (Data Gap DG001). Based on the drug class profile:

Key Class-Level Risks: JAK inhibitors as a class carry boxed warnings in the US for serious infections, malignancies (including lymphoma), major adverse cardiovascular events (MACE), thrombosis (DVT/PE), and mortality. These are especially relevant when considering oncology repurposing in immunocompromised patients.
Heparin Cofactor 2 Deficiency (Rank 9): Specifically contraindicated conceptually — JAK inhibitors carry a known VTE risk, posing a compounding hazard in patients with pre-existing coagulation defects.
Drug Interactions: No DDI data was returned from the query (query_status: not_found). Please refer to the SmPC for full interaction profile.

Conclusion and Next Steps

Decision: Hold (all 10 indications)

Rationale: No predicted indication has reached sufficient evidence quality to advance. The rank 1 prediction (colobomatous microphthalmia-rhizomelic dysplasia syndrome) is almost certainly a knowledge graph artifact with no mechanistic or clinical basis. The most actionable candidate — myeloid leukemia (rank 6, L3) — has only indirect GvHD-context evidence and lacks direct anti-leukemic efficacy data. Combined with blocking data gaps in Taiwan safety documentation, no indication can pass Stage 1 safety screening.

To proceed, the following is needed:

[DG001 — Blocking] Download and parse TFDA prescribing information PDF to obtain Taiwan-specific warnings and contraindications, enabling entry into S1 safety screening
[DG002 — High] Retrieve MOA data from DrugBank API to enable proper mechanistic linkage analysis for all 10 predicted indications
For myeloid leukemia specifically: identify whether any Phase 1/2 studies directly test Baricitinib for anti-leukemic activity (not just GvHD prevention); monitor NCT06475820 results (completion: April 2027)
For plasma cell myeloma (rank 5): evaluate whether IL-6/JAK2/STAT3 pathway inhibition rationale merits a hypothesis-generating preclinical study, particularly in high-risk or refractory settings
Reassess all L5 predictions for mechanistic plausibility after MOA data is available — several (ranks 1, 2, 7, 9) appear to be structural false positives that may be deprioritized following manual review
Disclaimer

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