Carbidopa
| Evidence Level: L5 | Predicted Indications: 10 |
Table of Contents
Carbidopa: From Parkinson Disease to Rasmussen Subacute Encephalitis
One-Sentence Summary
Carbidopa is a peripheral decarboxylase inhibitor, originally used together with levodopa to treat Parkinson disease and related parkinsonian disorders. The TxGNN model predicts it may be effective for Rasmussen subacute encephalitis, but no clinical trials and no publications currently support this specific prediction. The evidence is model prediction only, so the top-ranked prediction should be treated as a hypothesis rather than a lead.
Quick Overview
| Item | Content |
|---|---|
| Original Indication | Parkinson disease (and related parkinsonian disorders, per the knowledge graph) |
| Predicted New Indication | Rasmussen subacute encephalitis |
| TxGNN Prediction Score | 98.43% |
| Evidence Level | L5 |
| Singapore Market Status | ✓ Marketed |
| Number of Registrations | 9 |
| Recommended Decision | Hold |
Note: HSA records compiled for this report contain no approved-indication text, so the original indication comes from the knowledge graph.
Why is This Prediction Reasonable?
Detailed mechanism of action data is not available in the data compiled for this report. Based on known information, carbidopa inhibits L-aromatic amino acid decarboxylase (DDC) outside the brain. It has no antiparkinson action by itself, but when combined with levodopa it raises the amount of levodopa that reaches the brain.
For Rasmussen subacute encephalitis, the link is weak. Rasmussen encephalitis is an immune-mediated, chronic focal encephalitis with epilepsy, and there is no established dopaminergic or decarboxylase-inhibition rationale connecting carbidopa to it. The prediction most likely reflects network proximity to parkinsonism and encephalitis nodes in the knowledge graph (for example, postencephalitic parkinsonism). The high score therefore should not be read as evidence of efficacy.
Other predictions in the top 10 have more plausible mechanistic stories, and they are listed below as alternatives:
| Rank | Predicted Disease | Score | Evidence Level | Comment |
|---|---|---|---|---|
| 2 | PLA2G6-associated neurodegeneration | 97.60% | L4 | Often presents with dopa-responsive parkinsonism; only indirect case-level literature |
| 4 | Myelitis | 97.40% | L4 | Only secondary parkinsonism case reports; no direct link |
| 6 | Lewy body dementia | 95.99% | L3 | Shared nigrostriatal deficit with Parkinson disease; no RCT in this condition |
| 8 | Progressive supranuclear palsy-corticobasal syndrome | 95.36% | L4 | Levodopa response usually poor and transient |
| 9 | X-linked intellectual disability-ataxia-apraxia syndrome | 94.87% | L4 | Literature on MCT8 deficiency with levodopa/carbidopa response; disease-label match uncertain |
| 10 | X-linked intellectual disability-cerebellar hypoplasia syndrome | 93.72% | L4 | One observational paper; indirect mapping |
Ranks 3, 5 and 7 have no evidence. Ranks 3 and 5 (transaldolase deficiency, fructose-1,6-bisphosphatase deficiency) have no plausible mechanism. Rank 7 (juvenile paralysis agitans of Hunt) is a historical synonym for juvenile parkinsonism and overlaps with the existing indications, so it is likely not true repurposing.
Clinical Trial Evidence
For the top-ranked prediction (Rasmussen subacute encephalitis): Currently no related clinical trials registered.
Trials linked to other predictions are shown below for reference. Both are non-therapeutic.
| Trial Number | Phase | Status | Enrollment | Key Findings |
|---|---|---|---|---|
| NCT04246437 | Phase 1 | Recruiting | 40 | [18F]F-DOPA imaging in autonomic failure (synucleinopathies); diagnostic study, not a test of carbidopa in Lewy body dementia |
| NCT02994719 | N/A | Active, not recruiting | 120 | Gait pattern analysis in neurological disease including parkinsonian disorders; observational, no carbidopa intervention |
Literature Evidence
For the top-ranked prediction (Rasmussen subacute encephalitis): Currently no related literature available.
The most relevant publications for the other predicted indications are below.
| PMID | Year | Type | Journal | Key Findings |
|---|---|---|---|---|
| 36402160 | 2022 | Phase 3 RCT (advanced Parkinson disease) | Lancet Neurol | Continuous subcutaneous foslevodopa-foscarbidopa infusion in advanced PD; supports the original indication, not a new one |
| 41144879 | 2026 | Clinical/observational | Mov Disord | Altered dopamine metabolism and response to levodopa/carbidopa in MCT8 deficiency |
| 40088079 | 2025 | Observational/case series | Mov Disord | Childhood parkinsonism in Allan-Herndon-Dudley syndrome responds to levodopa/carbidopa |
| 41031701 | 2025 | Pharmacodynamic study | Pharmacotherapy | Acute blood pressure effects of oral levodopa/carbidopa in Parkinson disease |
| 14594099 | 2003 | Review | Can Fam Physician | Diagnosis and drug management of dementia with Lewy bodies |
| 11581527 | 2001 | Review | Curr Treat Options Neurol | Levodopa/carbidopa used to improve parkinsonism in diffuse Lewy body disease |
| 37016564 | 2023 | Review | Am J Case Rep | Dopamine deficiency evaluation in non-Alzheimer dementias |
| 41769496 | 2026 | Case report | Cureus | Adult-onset PLA2G6-associated parkinsonism with claval hypertrophy |
| 40585751 | 2025 | Case report | Cureus | Sublingual crushed levodopa/carbidopa in a Lewy body dementia patient with dysphagia |
| 32372748 | 2020 | Case report | Am J Trop Med Hyg | Dengue virus-triggered parkinsonism in an adolescent |
Singapore Market Information
All nine authorizations are valid single-ingredient registrations. No approved-indication text is available for them. Several are fixed-dose products combining carbidopa with levodopa (and entacapone for Stalevo and Antipar).
| Authorization Number | Product Name | Dosage Form | Approved Indication |
|---|---|---|---|
| SIN10286P | CREDANIL 25/250 TABLET | Tablet | Not stated |
| SIN12575P | TIDOMET FORTE TABLET | Tablet | Not stated |
| SIN10285P | CREDANIL 25/100 TABLET | Tablet | Not stated |
| SIN13142P | Stalevo Film Coated Tablets 100/25/200 mg | Film-coated tablet | Not stated |
| SIN16957P | ANTIPAR FILM COATED TABLET 150 mg/37.5 mg/200 mg | Film-coated tablet | Not stated |
Four more registrations exist (SIN13143P, SIN13566P, SIN13141P, SIN16956P): three Stalevo strengths and one Antipar strength. All dosage forms are oral.
Safety Considerations
Package insert warnings and contraindications were not available in the data compiled for this report. Please refer to the package insert for safety information.
Knowledge-graph contraindication signals (not label text, so they need verification against the insert):
- Neurological and psychiatric: epilepsy, drug-induced dyskinesia, multiple system atrophy, psychotic disorders (including schizoaffective and schizophreniform disorders), anxiety disorder, neurotic disorder, dysthymic disorder, mental disorder
- Cardiovascular: orthostatic hypotension, postural orthostatic tachycardia syndrome, myocardial infarction, coronary thrombosis, sudden cardiac arrest, heart conduction disease
- Hepatic, renal and gastrointestinal: liver failure, liver disease, kidney disease, gallbladder disease, gastric ulcer, peptic ulcer disease, diarrheal disease, pyoureter
- Metabolic and other: diabetes mellitus, diabetic ketoacidosis, hemoglobinopathy, melanoma, open-angle glaucoma
The epilepsy signal is worth noting for the top-ranked prediction, since Rasmussen encephalitis involves epilepsy.
Drug interactions: The only interaction record is pharmacological. Carbidopa inhibits its target, L-aromatic amino acid decarboxylase (DDC). No drug-drug interactions with severity levels were reported.
Conclusion and Next Steps
Decision: Hold
Rationale: The top-ranked prediction, Rasmussen subacute encephalitis, rests on a model score alone (L5). It has no trials, no literature, and no plausible mechanistic link to decarboxylase inhibition. The better-supported candidates (Lewy body dementia, MCT8-related syndromes, PLA2G6-associated neurodegeneration) are still limited to reviews, case reports and small observational studies, and they are best treated as research questions.
To proceed, the following is needed:
- Package insert warnings and contraindications from HSA (currently a blocking gap for safety screening)
- Mechanism of action data for carbidopa from DrugBank
- A reconciliation of the prediction list with the existing parkinsonism indications (for example, rank 7 is a synonym of an existing indication)
- For the most plausible alternatives (Lewy body dementia, MCT8 deficiency, PLA2G6-associated neurodegeneration), a targeted review for controlled evidence and a check that the disease labels map correctly
- A route and formulation compatibility check, which has not yet been performed
Disclaimer
This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.