Fluoxetine

證據等級: L5

預測適應症: 10 個

One-Sentence Summary

Fluoxetine (DrugBank DB00472) is a selective serotonin reuptake inhibitor (SSRI) with well-established use in major depressive disorder internationally. The TxGNN model returned 10 candidate indications for this drug; the top-ranked hits by raw score are Cluster A personality disorders (weak, indirect evidence), but the best-supported repurposing signal is Agoraphobia, backed by 19 publications, including three tier-1 RCTs, though no registered clinical trials were found for this specific indication. Critically, fluoxetine currently has zero SAHPRA registrations in South Africa and a blocking data gap on Professional Information (PI) warnings, so no formal safety assessment can be completed yet.

Quick Overview

Item	Content
Original Indication	Major depressive disorder (general pharmacological knowledge — not present in this evidence pack; South Africa registry data is empty)
Predicted New Indication	Agoraphobia
TxGNN Prediction Score	99.86%
Evidence Level	L2
South Africa Market Status	Not marketed / not SAHPRA-registered
Number of SAHPRA Registrations	0
Recommended Decision	Proceed with Guardrails (indication-level) — see Conclusion for drug-level caveat

TxGNN Predicted Indications — Full Ranking

This is a multi-indication candidate pack (10 predictions). Evidence quality varies substantially and does not track raw TxGNN score — the highest-scoring predictions (Cluster A personality disorders) are the weakest-supported, while several mid-ranked predictions have solid RCT/meta-analytic backing:

Rank	Predicted Indication	TxGNN Score	Evidence Level	Decision Stage	Recommendation
1	Schizoid personality disorder	99.92%	L4	S0	Hold
2	Schizotypal personality disorder	99.92%	L3	S1	Research Question
3	Paranoid personality disorder	99.92%	L4	S0	Hold
4	Histrionic personality disorder	99.92%	L4	S0	Hold
5	Benign paroxysmal torticollis of infancy	99.89%	L5	S0	Hold
6	Agoraphobia	99.86%	L2	S2	Proceed with Guardrails
7	Manic bipolar affective disorder	99.64%	L4	S0	Hold (mechanistic concern: SSRIs can precipitate manic switch)
8	Phobic disorder	99.63%	L2	S2	Proceed with Guardrails
9	Ohdo syndrome and variants	99.63%	L5	S0	Hold
10	Melancholia	99.57%	L2	S2	Proceed with Guardrails

The three “Proceed with Guardrails” indications (agoraphobia, phobic disorder, melancholia) are all serotonergic/anxiety-depression spectrum conditions consistent with fluoxetine’s known pharmacology. Agoraphobia is presented as the primary candidate below because it has the highest score among this group and the most direct literature base.

Why is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available for this candidate (DG002, High severity). Based on known pharmacological class information, fluoxetine is a selective serotonin reuptake inhibitor (SSRI); its efficacy in major depressive disorder and panic disorder has been established internationally, and mechanistically this profile may extend to agoraphobia.

Agoraphobia most commonly presents as a comorbid or secondary feature of panic disorder, and serotonergic modulation of amygdala/limbic circuit hyperactivation is the accepted mechanism underlying SSRI efficacy in panic-spectrum anxiety. Several SSRIs in this class (fluvoxamine, fluoxetine, paroxetine, sertraline, citalopram) have documented superiority over placebo for panic disorder and associated agoraphobic avoidance.

The TxGNN signal for agoraphobia is corroborated by a Cochrane network meta-analysis and multiple randomized, double-blind trials evaluating fluoxetine specifically in panic disorder with agoraphobia, which strengthens confidence in this prediction relative to the higher-scoring but mechanistically unsupported personality-disorder predictions.

Clinical Trial Evidence

Currently no related clinical trials registered for agoraphobia specifically (0 results in ClinicalTrials.gov and ICTRP queries). Evidence for this indication comes from published literature (below), not registry-indexed trials.

Literature Evidence

PMID	Year	Type	Journal	Key Findings
7786880	1995	RCT	J Psychiatry Neurosci	Fluoxetine (mean 20mg/day) demonstrated anti-panic efficacy in patients with panic disorder and agoraphobia
11001241	2000	RCT (double-blind, placebo-controlled)	J Clin Psychopharmacol	Pindolol augmentation of fluoxetine studied in treatment-resistant panic disorder with/without agoraphobia
10362436	1999	RCT	J Clin Psychiatry	Once-weekly fluoxetine maintained remission in panic disorder, leveraging its long half-life
38014714	2023	Review (network meta-analysis, Cochrane)	Cochrane Database Syst Rev	Comparative efficacy of pharmacological treatments for panic disorder in adults
11110016	2000	Review	Int Clin Psychopharmacol	SSRIs (incl. fluoxetine) superior to placebo for panic disorder, agoraphobia, and associated depression
14967551	2004	Cohort	Psychiatry Research	Factors influencing psychiatrists’ treatment choices (CBT ± benzodiazepine ± fluoxetine) in panic disorder with agoraphobia
22090798	2011	Review	Neuropsychiatr Dis Treat	Pharmacological interventions, including SSRIs, for complex agoraphobia
18090457	2007	Naturalistic comparison study	Clin Neuropharmacol	Long-term comparison of SSRIs (incl. fluoxetine) in panic disorder tolerability and outcome
3500189	1987	Open-label pilot study	J Clin Psychopharmacol	Early open trial: 7/16 patients with panic disorder/agoraphobia had complete cessation of panic attacks on fluoxetine
1884341	1991	Case series	Can J Psychiatry	Two treatment-refractory panic disorder with agoraphobia cases responded to fluoxetine after failing other agents

South Africa Market Information

Fluoxetine currently has no registered products in South Africa (market_status: 未上市, total_licenses: 0). No dosage forms, brand names, or approved indication texts are available from the evidence pack. Any repurposing pathway would first require a SAHPRA marketing authorization (new application or import route) before local prescribing could occur.

Safety Considerations

Please refer to the SAHPRA-approved Professional Information (PI) for safety information. Report adverse drug reactions to SAHPRA.

(Key warnings, contraindications, and drug-interaction data were queried but are not currently available in this evidence pack — DG001, Blocking severity: TFDA/SAHPRA PI warnings and contraindications could not be retrieved, which prevents a formal S1 safety pre-assessment for this candidate.)

Conclusion and Next Steps

Decision: Proceed with Guardrails (indication-level, contingent on resolving Blocking data gap)

Rationale: Agoraphobia (and the related phobic disorder / melancholia predictions) are mechanistically plausible and supported by L2-level literature evidence, including tier-1 RCTs, unlike the top-scoring but weakly-evidenced personality-disorder predictions. However, fluoxetine is not currently registered in South Africa and a Blocking data gap (DG001) prevents completion of the mandatory S1 safety pre-assessment — so this candidate cannot advance past guarded research status until that gap is closed.

To proceed, the following is needed:

Retrieve SAHPRA/TFDA-approved Professional Information (PI) — warnings, contraindications, DDI — to close DG001 (Blocking)
Retrieve fluoxetine’s mechanism of action via DrugBank API to close DG002 (High) and strengthen the mechanistic rationale
Confirm a SAHPRA registration or import pathway, since the product currently has zero local licenses
If pursuing agoraphobia specifically, commission or identify a registered clinical trial (none currently exist) rather than relying solely on published literature
Clarify whether “manic bipolar affective disorder” (rank 7) should be down-weighted or excluded given the known risk of SSRI-induced manic switch — this is a case where a high TxGNN score does not track clinical appropriateness
Disclaimer

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