Roflumilast
| Evidence Level: L5 | Predicted Indications: 10 |
Table of Contents
Roflumilast: From COPD to Pulmonary Hypertension
One-Sentence Summary
Roflumilast is a selective phosphodiesterase-4 (PDE4) inhibitor used to reduce exacerbations in patients with severe COPD. The TxGNN model predicts it may be effective for Pulmonary Hypertension, but this prediction is currently supported by no clinical trials and no literature, and the underlying mechanistic link is considered weak.
Quick Overview
| Item | Content |
|---|---|
| Original Indication | Chronic Obstructive Pulmonary Disease (COPD), per pharmacovigilance literature in this evidence pack |
| Predicted New Indication | Pulmonary Hypertension |
| TxGNN Prediction Score | 97.24% |
| Evidence Level | L5 (AI prediction only, no supporting studies) |
| EU Market Status | Not marketed |
| Number of Authorizations | 0 |
| Recommended Decision | Hold |
Why is This Prediction Reasonable?
Currently, detailed mechanism of action data is not available (blocking data gap). Based on known information, Roflumilast is a selective PDE4 inhibitor that reduces cAMP breakdown, producing anti-inflammatory effects that have proven efficacy in severe COPD.
The proposed link to pulmonary hypertension is mechanistically weak. PDE4 inhibition has some theoretical anti-inflammatory and vascular-remodeling-modulating effects, which could plausibly influence pulmonary vascular remodeling. However, the dominant pharmacology for approved PH therapies (e.g., sildenafil, tadalafil) acts through PDE5 inhibition and the cGMP pathway — a mechanistically distinct target from PDE4/cAMP. This is a pure AI-predicted association with no clinical or preclinical data currently supporting it.
Given this gap between the predicted mechanism and the established PH treatment pathway, this candidate should be treated as a hypothesis-generation signal only, not as a validated repurposing opportunity.
Clinical Trial Evidence
Currently no related clinical trials registered.
Literature Evidence
Currently no related literature available.
EU Market Information
Roflumilast currently has no marketing authorization on record in this system (market status: not marketed, 0 authorizations). No product-level data is available for review.
Safety Considerations
No formal SmPC-level warnings, contraindications, or drug-drug interaction data are available for this drug (blocking data gap — TFDA/EMA label not yet retrieved).
However, evidence collected under a separate predicted-indication entry (“headache disorder,” rank 3) flags an important safety signal: clinical trial and real-world pharmacovigilance data consistently identify headache, weight loss, decreased appetite, insomnia, diarrhea, vomiting, and nausea as common adverse effects of Roflumilast, occurring mainly early in treatment (trial NCT02018432; pharmacovigilance analyses PMID 41458197 and PMID 41602587). This is a known class effect of PDE4 inhibitors and should be treated as a safety consideration for the drug itself, not as a repurposing opportunity — the TxGNN model’s prediction of “headache disorder” as a treatable indication is directionally inconsistent with the actual evidence and should not be pursued.
Conclusion and Next Steps
Decision: Hold
Rationale: The predicted indication (pulmonary hypertension) has no clinical trial or literature support and rests on a mechanistically weak link (PDE4 vs. the PDE5-dominant PH treatment pathway). Combined with the absence of basic label-level safety data and the drug’s current unmarketed status, there is insufficient evidence to advance this candidate beyond model prediction.
To proceed, the following is needed:
- Retrieve TFDA/EMA label safety data (warnings, contraindications) — currently a blocking gap
- Obtain detailed MOA data from DrugBank to properly assess mechanistic plausibility
- Preclinical or mechanistic studies specifically testing PDE4 inhibition in pulmonary vascular remodeling/PH models
- Clarify and separately track the headache/adverse-effect signal so it is not conflated with genuine repurposing candidates
- Consider evaluating higher-evidence candidates in this pack instead — e.g., “female breast carcinoma” (rank 4, L4, in vitro mechanistic data available) offers a more substantive starting hypothesis than the top-ranked PH prediction
Disclaimer
This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.