Carfilzomib

證據等級: L5

預測適應症: 5 個

Carfilzomib: From Multiple Myeloma to Melanoma

One-Sentence Summary

Carfilzomib is an irreversible proteasome inhibitor approved by the US FDA for relapsed/refractory multiple myeloma, but currently not marketed in Taiwan. The TxGNN model’s top-5 predictions all cluster around melanoma and its subtypes — with CMM7 (a melanoma susceptibility locus) scoring highest at 99.37%, and melanoma (rank 5) carrying the only actual evidence: 5 preclinical publications and 0 clinical trials. Overall evidence strength is at the preclinical stage only (L4); the top-4 predictions (CMM7, pediatric leptomeningeal, uveal, vulvar melanoma) are all L5 with no supporting data.

Quick Overview

Item	Content
Original Indication	Multiple myeloma (relapsed/refractory) — US FDA-approved; no Taiwan license
Predicted New Indication	Melanoma (all 5 TxGNN predictions cluster around melanoma entities)
TxGNN Prediction Score	99.37% (CMM7, rank 1) → 99.03% (Melanoma, rank 5)
Evidence Level	L4 (melanoma, rank 5); L5 (ranks 1–4, no evidence)
Taiwan Market Status	✗ Not marketed in Taiwan
Number of Taiwan Licenses	0
Recommended Decision	Hold

Why Is This Prediction Reasonable?

Detailed mechanism of action data was not available in the Evidence Pack. Based on established medical knowledge, Carfilzomib (Kyprolis®) is a selective, irreversible inhibitor of the 20S proteasome’s chymotrypsin-like activity (β5 subunit). By blocking the ubiquitin-proteasome system (UPS), it causes accumulation of pro-apoptotic proteins (BIM, NOXA, PUMA), prevents IκB degradation thereby suppressing NF-κB survival signaling, and triggers ER stress → unfolded protein response (UPR) → CHOP-mediated apoptosis. Its irreversible binding distinguishes it from bortezomib and provides more sustained proteasome inhibition.

The mechanistic bridge between multiple myeloma and melanoma is UPS dependency: both are cancers with high protein synthesis and turnover rates, making them vulnerable to proteasome blockade. PMID 33671902 directly demonstrated that Carfilzomib combined with Bortezomib enhances apoptotic cell death in B16-F1 melanoma cells in vitro through caspase 3, 8, 9, and 12 activation — confirming the rationale is not purely theoretical.

However, the prediction carries significant caveats. The top-ranked entity, CMM7 (Cutaneous Malignant Melanoma susceptibility locus 7), is a genetic risk locus — not an actionable disease entity or treatment target. Predictions 2–4 (pediatric leptomeningeal, uveal, and vulvar melanoma) are rare subtypes with distinct molecular drivers (GNAQ/GNA11, KIT mutations) and specific barriers: poor blood-brain barrier penetration (high MW, P-gp substrate) limits CNS activity, and mucosal/ocular subtypes have not been studied preclinically with proteasome inhibitors. Only the rank-5 prediction (melanoma broadly) is supported by any evidence.

Clinical Trial Evidence

Currently no related clinical trials registered for Carfilzomib in melanoma or any of its predicted subtypes.

Literature Evidence

All publications are preclinical; no RCTs or clinical studies identified.

PMID	Year	Type	Journal	Key Findings
33671902	2021	In vitro preclinical	Biology	Carfilzomib + Bortezomib combination directly enhances apoptosis in B16-F1 melanoma cells via caspase 3/8/9/12 activation; Annexin V flow cytometry confirmed cell death
36134605	2023	Computational / in silico	J Biomol Struct Dyn	Molecular docking and MD simulations evaluated Carfilzomib against 18 kinase targets across 10 cancer types including melanoma; identifies potential off-target activity
27016342	2016	Preclinical mechanistic	Matrix Biology	Carfilzomib and Bortezomib activate NF-κB to upregulate heparanase in myeloma cells; relevant mechanistic caveat: proteasome inhibitors may paradoxically promote pro-tumorigenic heparanase in some contexts
31540997	2019	Mechanistic / molecular biology	Mol Cancer Res	AIRAP/AIRAPL (UPS regulators) govern proteasome-mediated cell survival in human melanoma; implicates proteasome pathway as a regulatory axis in melanoma biology
29581547	2018	Chemical biology / PROTAC	Leukemia	BRD4-targeting PROTACs require functional proteasome for degradation; Carfilzomib co-treatment data contextualize UPS biology across cancer types including beyond myeloma

Taiwan Market Information

Carfilzomib is not approved or marketed in Taiwan. No TFDA licenses found.

Note: Carfilzomib (Kyprolis®) holds US FDA approval for relapsed/refractory multiple myeloma in combination regimens (e.g., with lenalidomide/dexamethasone or daratumumab). Taiwan market entry would require a separate TFDA regulatory submission.

Cytotoxicity

Carfilzomib is an antineoplastic agent (proteasome inhibitor class — targeted oncology).

Item	Content
Cytotoxicity Classification	Targeted therapy — irreversible proteasome inhibitor (non-classical cytotoxic; not a DNA-damaging agent)
Myelosuppression Risk	Moderate — thrombocytopenia and anemia reported in myeloma trials; less pronounced neutropenia compared to classical cytotoxics
Emetogenicity Classification	Low
Monitoring Items	CBC with differential; serum creatinine and eGFR (renal); cardiac function (LVEF by ECHO before and during treatment); blood pressure monitoring each cycle; liver enzymes (ALT/AST)
Handling Protection	Please refer to the package insert warnings and precautions; cardiac toxicity (cardiomyopathy, heart failure) is the primary safety concern and requires cardiac screening prior to use

Safety Considerations

Please refer to the package insert for safety information.

(Package insert warnings, contraindications, and drug interaction data were not available in the current Evidence Pack. Resolution of data gaps DG001 and DG002 is required before any formal safety assessment can be completed.)

Conclusion and Next Steps

Decision: Hold

Rationale: All five TxGNN predictions center on melanoma entities, but the evidence base is limited to in vitro preclinical studies only (L4 at best for general melanoma; L5 for all specific subtypes). CMM7 — the highest-scored prediction — is a genetic susceptibility locus rather than a disease indication, and cannot serve as a clinical target. No animal model data or clinical trials exist to support advancement at this time.

To proceed, the following is needed:

Resolve data gaps first: Retrieve TFDA package insert (DG001) for warnings/contraindications; query DrugBank API for full MOA data (DG002)
In vivo validation: Xenograft or syngeneic mouse melanoma models with Carfilzomib monotherapy or combination to confirm anti-tumor activity beyond cell culture
PK/PD assessment: Tumor penetration studies in cutaneous melanoma; separate CNS penetration assessment before considering leptomeningeal indications
Subtype prioritization: Determine which melanoma molecular subtype (BRAF-mutant, BRAF-WT, NRAS-mutant) shows greatest sensitivity to proteasome inhibition — this will define the actionable patient population
Mechanistic safety clarification: PMID 27016342 raises a concern that Carfilzomib may upregulate heparanase via NF-κB in some tumor contexts; this potential pro-metastatic effect should be evaluated in melanoma models before clinical development
Regulatory pathway: Given zero Taiwan licensing, any clinical development would require de novo TFDA IND application; US Phase 1 basket trial including melanoma cohort would be the recommended entry point
Disclaimer

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