Deoxycholic Acid
| 證據等級: L5 | 預測適應症: 10 個 |
目錄
Deoxycholic Acid: From Submental Fat Reduction to Rheumatoid Arthritis
One-Sentence Summary
Deoxycholic acid (DCA) is a secondary bile acid and FXR/TGR5 receptor agonist, clinically approved (as injectable ATX-101/Kybella) for submental fat reduction via adipocytolysis, and currently not registered in Singapore. The TxGNN model generated 10 predictions for DCA; Rheumatoid Arthritis (Rank 6) is the most evidence-supported repurposing candidate with 2 completed Phase 2 clinical trials and 20 publications, though most clinical evidence is for the related bile acid UDCA rather than DCA directly — warranting cautious progression.
Quick Overview
| Item | Content |
|---|---|
| Original Indication | Submental fat reduction (injectable ATX-101/Kybella; not registered in Singapore) |
| Predicted New Indication | Rheumatoid Arthritis (highest evidence level among 10 predictions) |
| TxGNN Prediction Score | 96.09% (Rank 24,059) |
| Evidence Level | L3 — Observational studies / Phase 2 trials (bile acid class) |
| Singapore Market Status | Not marketed |
| Number of Registrations | 0 |
| Recommended Decision | Proceed with Guardrails |
Multi-Indication Prediction Landscape
The TxGNN model identified 10 candidate indications. Rheumatoid arthritis is the only one with clinically actionable evidence; diabetic nephropathy warrants secondary investigation:
| Rank | Disease | TxGNN Score | Evidence Level | Recommendation |
|---|---|---|---|---|
| 1 | Autosomal dominant familial hematuria-retinal arteriolar tortuosity-contractures syndrome | 99.49% | L5 | Hold |
| 2 | Brain small vessel disease 1 with/without ocular anomalies | 99.49% | L5 | Hold |
| 3 | Diabetic nephropathy | 99.32% | L4 | Research Question |
| 4 | Prinzmetal angina | 97.35% | L5 | Hold |
| 5 | Hemoglobinopathy | 96.64% | L5 | Hold |
| 6 | Rheumatoid arthritis | 96.09% | L3 | Proceed with Guardrails |
| 7 | Partial deletion of chromosome 16p | 95.75% | L5 | Hold |
| 8 | Conjunctivitis | 95.65% | L5 | Hold |
| 9 | Beta-thalassemia with other manifestations | 95.40% | L5 | Hold |
| 10 | Hemolytic anemia due to GPI deficiency | 95.23% | L5 | Hold |
The remainder of this report focuses on Rheumatoid Arthritis. Ranks 1–2 and 4–10 are all rated Hold due to absent mechanistic links or zero supporting evidence. Diabetic nephropathy (Rank 3) is noted at the end as a secondary research signal.
Why is This Prediction Reasonable?
Deoxycholic acid is an endogenous secondary bile acid generated by intestinal bacterial 7α-dehydroxylation of primary bile acid cholic acid. As a natural FXR (Farnesoid X Receptor) and TGR5 agonist, DCA is positioned at the centre of the bile acid–immune axis. FXR activation suppresses NF-κB-driven transcription, reducing downstream pro-inflammatory cytokines IL-6 and TNF-α — which are the very same targets of approved biologics in rheumatoid arthritis (e.g., tocilizumab, adalimumab). TGR5 activation on macrophages and T cells further dampens innate inflammatory signalling.
The link to RA is reinforced by multiple mechanistic layers: bile acids modulate Th17/Treg balance, and Th17 suppression is a validated therapeutic strategy in RA; gut microbiome dysbiosis drives systemic inflammatory arthritis and is correctable via bile acid signalling; and fibroblast-like synoviocytes (FLS) — the key destructive cells in RA — can be targeted through apoptosis induction by bile acid derivatives (PMID 23510744). Strikingly, a 1975 clinical pilot (PMID 1166284) directly infused chenodeoxycholic acid (the primary bile acid that is converted to DCA in the gut) into 10 RA patients and documented clinical improvement, offering the earliest human proof-of-concept for this chemical class in RA.
The critical limitation is that modern clinical evidence is almost entirely for UDCA (ursodeoxycholic acid), a structurally related but pharmacologically distinct bile acid. UDCA is hydrophilic and hepatoprotective; DCA is more hydrophobic and can be pro-inflammatory at high concentrations. Whether UDCA findings translate to DCA requires direct experimental confirmation. An animal study (PMID 23041275) using an orally absorbable DCA conjugate (6ODS-LHbD) in collagen-induced arthritis provides the most direct evidence for DCA itself, showing inhibition of neovascularisation and bone destruction.
Clinical Trial Evidence
| Trial Number | Phase | Status | Enrollment | Key Findings |
|---|---|---|---|---|
| NCT05973370 | Phase 2 | Completed | 60 | UDCA as add-on to current DMARDs in Egyptian RA patients; assessed effects on synovial inflammation and disease activity score |
| NCT04834557 | Phase 2 | Completed | 90 | UDCA + Digoxin as add-on to DMARDs in RA patients; evaluated impact on synovial inflammation and disease activity across variant disease activity levels |
Important: Both trials investigated UDCA, not DCA. They provide class-level biological plausibility but are not direct evidence for DCA repurposing in RA.
Literature Evidence
| PMID | Year | Type | Journal | Key Findings |
|---|---|---|---|---|
| 1166284 | 1975 | Clinical Pilot | Scand J Rheumatol | Chenodeoxycholic acid IV infusion in 10 RA patients; clinical improvement observed — earliest direct human evidence for a bile acid in RA |
| 73677 | 1976 | Case Series | Lancet | Oral chenodeoxycholate (DCA's primary bile acid precursor) in RA patients; therapeutic benefit reported |
| 23041275 | 2012 | Animal Study | J Controlled Release | DCA conjugate (6ODS-LHbD) suppressed neovascularisation and bone destruction in collagen-induced arthritis — the closest direct DCA evidence in an arthritis model |
| 23510744 | 2013 | In Vitro | Eur J Pharmacol | Taurochenodeoxycholic acid induced apoptosis of fibroblast-like synoviocytes (FLS), key destructive cells in RA joint pathology |
| 38313166 | 2023 | Animal Study | J Medicine Life | UDCA (50–100 mg/kg) compared to dexamethasone and diclofenac in Freund's adjuvant RA model; reduced paw swelling, IL-17, and pro-inflammatory cytokines |
| 28539269 | 2017 | Animal Study | Immunol Lett | UDCA attenuated experimental autoimmune arthritis by targeting Th17 cells via pAMPK/SMILE pathway — mechanistic support for bile acid Th17 suppression |
| 36604114 | 2023 | Animal Study | Gut | Gut commensal Parabacteroides distasonis alleviates inflammatory arthritis via bile acid metabolite modulation — establishes gut-bile acid-joint axis |
| 949914 | 1976 | Observational | Dan Med Bull | Abnormal bile acid metabolism documented in RA patients, suggesting bile acid dysregulation as part of RA pathophysiology |
| 8291509 | 1994 | Review | Am J Med Sci | Comprehensive review of chenodeoxycholate (DCA precursor) biology, pharmacology, and clinical indications including early RA evidence |
| 14508047 | 2003 | Case Report | Rheumatology | UDCA normalised MTX-induced transaminase elevation in an RA patient — demonstrates practical utility of bile acid adjuncts in RA treatment settings |
Singapore Market Information
Deoxycholic acid (DB03619) is not currently registered in Singapore and has no approved product licences on record.
Global reference: DCA is marketed as Kybella (USA, Allergan) and Belkyra (Europe) as an injectable formulation for non-surgical submental fat reduction in adults. These approvals are for a local-injection, adipocytolytic mechanism and are not directly applicable to systemic anti-inflammatory use in RA.
Safety Considerations
Detailed package insert safety data were not available for this report. Please refer to the manufacturer's package insert for complete warnings, contraindications, and precautions.
Critical safety signals relevant to RA repurposing:
- DCA ≠ UDCA: DCA is significantly more hydrophobic than UDCA. At high systemic concentrations, DCA can be pro-inflammatory and hepatotoxic — in contrast to UDCA's hepatoprotective profile. The therapeutic window for systemic anti-inflammatory DCA dosing has not been established.
- Hepatotoxicity risk: Oral or systemic DCA at doses relevant for RA treatment would require dedicated hepatotoxicity monitoring (liver function tests at baseline and regular intervals).
- Route-of-administration gap: The only approved DCA formulation is a local subcutaneous injection. Developing oral or IV formulations for RA would require separate pharmaceutical development, stability testing, and bioavailability data.
- No DDI data: Drug-drug interaction profile for DCA is not available; caution is warranted when combining with DMARDs (methotrexate, leflunomide) due to potential shared hepatic metabolism.
Secondary Signal: Diabetic Nephropathy (Rank 3, L4)
The diabetic nephropathy prediction (TxGNN score 99.32%) carries more biological plausibility than ranks 1–2 and is supported by 20 preclinical and clinical cohort publications. The mechanistic rationale is coherent: FXR activation is renoprotective in diabetic models; bile acids reduce SGLT2 expression (PMID 22429686), podocyte ER stress, and renal oxidative damage. However, all evidence is preclinical (animal studies, in vitro), and a clinical cohort study (PMID 39384774) shows only that bile acid metabolism is altered in diabetic kidney disease, not that DCA treatment is beneficial. The core concern — that existing evidence is almost entirely for UDCA/TUDCA, not DCA, and DCA may be nephrotoxic at high doses — places this at an exploratory research stage only.
Recommendation for diabetic nephropathy: Research Question — fund a focused mechanistic study to determine whether DCA specifically (not UDCA) is safe and active in diabetic kidney models before clinical consideration.
Conclusion and Next Steps
Decision: Proceed with Guardrails (Rheumatoid Arthritis)
Rationale: Two completed Phase 2 trials demonstrate that the bile acid class has clinically meaningful anti-inflammatory effects in RA patients when added to standard DMARDs. Historical human pilots with chenodeoxycholic acid (DCA's direct metabolic precursor) dating to 1975, combined with a direct animal study using a DCA conjugate in arthritis, provide multi-layer convergent evidence. The FXR/TGR5 → NF-κB → Th17 suppression pathway is mechanistically coherent with RA pathology. However, all modern clinical data are for UDCA, and DCA carries a meaningfully different and less favourable safety profile that must be directly characterised before patient exposure.
To proceed, the following is needed:
- DCA-specific safety window study: Determine the maximum tolerated dose and hepatotoxicity threshold for systemic/oral DCA in an appropriate animal model before any human dosing
- MOA confirmation: Obtain and review the full DrugBank mechanistic profile (FXR/TGR5 binding affinity, receptor selectivity relative to UDCA) to confirm or refute class-effect assumptions
- Formulation development: Design and characterise an oral or parenteral DCA formulation suitable for systemic anti-inflammatory use (distinct from the approved local-injection product)
- Bridging Phase 1/2 study in RA patients: Test DCA (not UDCA) in a small RA cohort with intensive safety monitoring, using the two completed UDCA trials (NCT05973370, NCT04834557) as benchmark comparators
- Singapore regulatory pathway assessment: Since no local registration exists, a full de novo application would be required; early engagement with HSA regarding bridging requirements from existing global bile acid safety data is advisable
Disclaimer
This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.