Regorafenib
| 證據等級: L5 | 預測適應症: 10 個 |
目錄
Regorafenib: From Metastatic Colorectal Cancer to Liposarcoma
One-Sentence Summary
Regorafenib is an oral multikinase inhibitor originally developed for metastatic colorectal cancer, GIST and hepatocellular carcinoma. The TxGNN model predicts it may be effective for Liposarcoma, with 2 clinical trials and 9 publications available — however, the two completed Phase II randomised trials that specifically tested this indication both found no clinical benefit in liposarcoma patients, directly contradicting the model’s prediction.
Quick Overview
| Item | Content |
|---|---|
| Original Indication | Metastatic colorectal cancer, GIST, hepatocellular carcinoma (per international product labeling; no Danish marketing authorisation on file to confirm locally) |
| Predicted New Indication | Liposarcoma |
| TxGNN Prediction Score | 99.76% |
| Evidence Level | L2 (2 completed Phase II RCTs directly addressing this indication — but with negative results) |
| Denmark Market Status | Not marketed |
| Number of Marketing Authorisations | 0 |
| Recommended Decision | Hold |
Why is This Prediction Reasonable?
Regorafenib is an oral multikinase inhibitor targeting angiogenic (VEGFR1-3, TIE2), stromal (PDGFR-β, FGFR) and oncogenic receptor tyrosine kinases (KIT, RET, RAF) (per literature evidence, PMID 30069758). Its original approval was built on this broad antiangiogenic/antiproliferative profile in metastatic colorectal cancer, GIST and hepatocellular carcinoma.
Soft tissue sarcomas, including liposarcoma, are highly vascularised tumours in which angiogenesis signalling plays a key role in tumour biology (per REGOSARC trial protocol, PMID 25884155). This provided the mechanistic rationale for testing regorafenib across multiple sarcoma subtypes, and it is this class-level plausibility that the TxGNN model appears to be capturing.
However, mechanistic plausibility did not translate into clinical benefit for this specific subtype. The REGOSARC trial (PMID 27751846) showed regorafenib improved outcomes in leiomyosarcoma and synovial sarcoma but explicitly did not show benefit in the liposarcoma cohort. The independent SARC024 trial (PMID 32701199) confirmed this finding in a treatment-refractory liposarcoma population and concluded that “routine use of regorafenib in this patient population” is not supported. This is a case where a strong knowledge-graph-based mechanistic signal is directly contradicted by dedicated randomised clinical trial data.
Clinical Trial Evidence
| Trial Number | Phase | Status | Enrollment | Key Findings |
|---|---|---|---|---|
| NCT01900743 | Phase 2 | Completed | 219 | REGOSARC: international, randomised, double-blind, placebo-controlled trial of regorafenib in metastatic/unresectable soft tissue sarcoma after anthracycline failure; liposarcoma was one of five predefined cohorts |
| NCT02048371 | Phase 2 | Completed | 131 | SARC024: blanket protocol studying oral regorafenib across selected sarcoma subtypes, including a dedicated liposarcoma cohort |
Literature Evidence
| PMID | Year | Type | Journal | Key Findings |
|---|---|---|---|---|
| 27751846 | 2016 | RCT | The Lancet. Oncology | REGOSARC: regorafenib improved progression-free survival vs placebo in non-adipocytic soft tissue sarcoma, but not in liposarcoma |
| 32701199 | 2020 | RCT | The Oncologist | SARC024 liposarcoma cohort: results do not support routine use of regorafenib in treatment-refractory liposarcoma |
| 29902612 | 2018 | RCT (post-cross-over analysis) | European Journal of Cancer | Updated REGOSARC analysis confirms lack of efficacy in liposarcoma even after placebo-to-regorafenib cross-over |
| 28295221 | 2017 | RCT post-hoc analysis | Cancer | Q-TWiST analysis of REGOSARC; quality-adjusted clinical benefit concentrated in non-adipocytic cohorts |
| 25884155 | 2015 | Study protocol | BMC Cancer | REGOSARC trial design; rationale based on angiogenesis’s role in sarcoma biology |
| 29931504 | 2018 | Review | Targeted Oncology | Reviews regorafenib’s evolving role across soft tissue sarcoma subtypes, including liposarcoma |
| 33290314 | 2021 | Retrospective study | Anti-Cancer Drugs | Anlotinib in WDLS/DDLS; cites regorafenib as an approved TKI for non-adipocytic STS, not liposarcoma specifically |
| 40975452 | 2025 | Review | Critical Reviews in Oncology/Hematology | Reviews maintenance therapy strategies after first-line treatment of advanced STS |
| 26266019 | 2015 | Case report | Rare Tumors | Pazopanib case in Ewing sarcoma; cited as part of the rationale for including a liposarcoma arm in SARC024 |
Denmark Market Information
Regorafenib currently has no marketing authorisation on file in Denmark (0 authorisations recorded; market status: not marketed).
Cytotoxicity
| Item | Content |
|---|---|
| Cytotoxicity Classification | Targeted therapy (oral multikinase inhibitor: VEGFR1-3, TIE2, PDGFR-β, FGFR, KIT, RET, RAF) |
| Myelosuppression Risk | Low–Moderate — literature evidence for this drug class emphasises hand-foot skin reaction, hypertension and diarrhoea over myelosuppression (PMID 30069758); no direct haematologic toxicity data in this evidence pack |
| Emetogenicity Classification | Low (typical for oral multikinase TKIs) |
| Monitoring Items | Blood pressure (VEGFR-inhibition-related hypertension, PMID 36583425), liver function (hepatotoxicity risk across anti-angiogenic TKIs, PMID 23981115), skin examination for hand-foot skin reaction (PMID 23700287), CBC |
| Handling Protection | Oral targeted anticancer agent — follow institutional oral antineoplastic handling protocols; local TFDA/SmPC-specific handling requirements are a data gap pending label review |
Safety Considerations
Please refer to the approved Summary of Product Characteristics (SmPC) for safety information. TFDA-equivalent label warnings/contraindications and DDI data were not available in this evidence pack (flagged as a Blocking data gap — DG001 — preventing initial safety screening).
Conclusion and Next Steps
Decision: Hold
Rationale:
- Regorafenib has no existing marketing authorisation in Denmark, and local SmPC safety data is a Blocking gap that prevents even an initial safety screen.
- More importantly, the two completed Phase II RCTs that specifically tested regorafenib in liposarcoma (REGOSARC, SARC024) both found no clinical benefit in this subtype, directly contradicting the TxGNN model’s high prediction score. This is a case where the model’s knowledge-graph-level signal does not hold up against dedicated clinical trial evidence.
To proceed, the following is needed:
- TFDA/SmPC warnings, contraindications and drug interaction data (currently Blocking gap, DG001)
- Formal mechanism-of-action documentation (DrugBank MOA field, DG002)
- If pursued further, reframe as a research question around why TxGNN scores this indication highly despite negative trial data (e.g., biomarker-selected subpopulations, combination regimens) rather than pursuing liposarcoma monotherapy directly
- Confirm whether any EU/EMA centralised authorisation exists for regorafenib under its approved indications, which would affect off-label pathway feasibility in Denmark
Disclaimer
This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.