Ipilimumab

證據等級: L5

預測適應症: 4 個

Ipilimumab: From Melanoma to Non-Cutaneous Melanoma

One-Sentence Summary

Ipilimumab (Yervoy) is an anti-CTLA-4 immune checkpoint inhibitor originally developed and approved for the treatment of unresectable or metastatic cutaneous melanoma. The TxGNN model predicts it may be effective for non-cutaneous melanoma (including uveal and mucosal subtypes), with 50 clinical trials and 5 publications currently supporting this direction. The model also assigns its highest score to choroideremia (99.06%), though this finding is entirely unsupported by clinical or experimental evidence at this time.

Quick Overview

Item	Content
Original Indication	Unresectable or metastatic melanoma (cutaneous)
Predicted New Indication	Non-Cutaneous Melanoma (uveal and mucosal subtypes)
TxGNN Prediction Score	99.01%
Evidence Level	L1
Denmark Market Status	Not marketed (per data pack — see note below)
Number of Marketing Authorisations	0 (per data pack — see note below)
Recommended Decision	Proceed with Guardrails

Important note on Danish market status: The evidence pack records 0 marketing authorisations. This warrants verification: ipilimumab (Yervoy®) holds a centralised European Marketing Authorisation via the EMA, valid across all EU/EEA member states including Denmark. The discrepancy likely reflects a data gap in the current evidence pack. Clinicians should consult the EMA product page and the Danish Medicines Agency’s (Lægemiddelstyrelsen) product database directly before drawing any conclusions about availability.

Why is This Prediction Reasonable?

Melanoma — regardless of anatomical origin — is characterised by a tumour microenvironment in which T-cell activity is suppressed through CTLA-4-mediated immune tolerance. Ipilimumab binds and blocks CTLA-4, reversing this suppression, restoring tumour-infiltrating T-cell activity, and promoting the expansion of effector memory T cells. This mechanism is not inherently tissue-specific, which provides a biologically sound rationale for efficacy across all melanoma subtypes, including non-cutaneous variants.

Non-cutaneous melanoma subtypes — primarily uveal melanoma (arising from the choroid of the eye) and mucosal melanoma (arising from mucosal surfaces such as the oral cavity, GI tract, or genitourinary tract) — differ from cutaneous melanoma in important ways. They carry a lower tumour mutational burden (TMB), harbour distinct driver mutations (e.g., GNAQ/GNA11 in uveal melanoma), and show lower overall response rates to checkpoint inhibition compared with cutaneous disease. However, direct clinical evidence has confirmed ipilimumab activity in both uveal and mucosal subtypes (see PMID 24999899), and dual CTLA-4/PD-1 blockade with nivolumab + ipilimumab has become a recommended first-line option for these patients.

The TxGNN score of 99.01% for non-cutaneous melanoma is well grounded in this biological and clinical context. The prediction is corroborated by multiple completed Phase 3 randomised trials — including the pivotal NCT00324155 (ipilimumab + dacarbazine vs. placebo + dacarbazine, n=681) and NCT03068455 (nivolumab + ipilimumab adjuvant vs. nivolumab monotherapy, n=1,844) — providing the highest available level of clinical evidence (L1).

Clinical Trial Evidence

Trial Number	Phase	Status	Enrollment	Key Findings
NCT00324155	Phase 3	Completed	681	Pivotal RCT: ipilimumab 10 mg/kg + dacarbazine vs. dacarbazine + placebo in previously untreated Stage III/IV melanoma; formed a key basis for regulatory approval of ipilimumab in melanoma
NCT03068455	Phase 3	Completed	1,844	Adjuvant nivolumab + ipilimumab vs. nivolumab monotherapy after complete resection of Stage IIIb/c/d or IV melanoma; the largest trial in this data set
NCT02339571	Phase 2/3	Active, not recruiting	600	Nivolumab + ipilimumab ± sargramostim in unresectable Stage III/IV melanoma; long-term follow-up data ongoing until 2033
NCT02506153	Phase 3	Active, not recruiting	1,301	Pembrolizumab vs. physician/patient choice of high-dose interferon or ipilimumab in high-risk resected Stage III-IV melanoma; positions ipilimumab as a comparator in the adjuvant setting
NCT01783938	Phase 2	Completed	138	Randomised: sequential nivolumab then ipilimumab in advanced/metastatic melanoma; directly supports dual checkpoint dosing strategy design
NCT02905266	Phase 3	Completed	106	Phase IIIb: multiple administration regimens of nivolumab + ipilimumab in previously untreated unresectable/metastatic melanoma; informs dosing schedule optimisation
NCT02320058	Phase 2	Completed	119	Nivolumab + ipilimumab in melanoma metastatic to the brain; directly relevant to advanced non-cutaneous melanoma with CNS involvement
NCT01950390	Phase 2	Completed	169	Randomised: ipilimumab ± bevacizumab in unresectable Stage III/IV melanoma; evaluates VEGF/immune checkpoint combination benefit
NCT00972933	Early Phase 1	Completed	59	Neoadjuvant ipilimumab in Stage IIIB-C melanoma; important immunogenicity and biomarker data in the neoadjuvant setting
NCT07230613	Phase 2	Recruiting	50	Neoadjuvant intratumoral anti-CTLA-4 + anti-PD-1 in localised melanoma; novel delivery route exploring reduced systemic toxicity with maintained local efficacy

Literature Evidence

PMID	Year	Type	Journal	Key Findings
24999899	2014	Retrospective cohort	The Medical Journal of Australia	Efficacy and tolerability of ipilimumab across cutaneous, uveal, and mucosal melanoma subtypes in an Australian clinical setting; the most directly relevant publication for non-cutaneous melanoma activity
37887546	2023	Real-world evidence	Current Oncology	Multi-centre retrospective cohort comparing anti-PD-1 monotherapy vs. combination with ipilimumab across age groups in advanced melanoma; real-world comparative effectiveness data
29466692	2018	Clinical review	Discovery Medicine	Comprehensive update on anti-PD-1 mAbs as monotherapy or combined with ipilimumab; summarises mature Phase 3 survival data and current treatment landscape
28183255	2018	Systematic review	Current Cancer Drug Targets	Systematic review of Phase I–III adjuvant melanoma trials (2000–2015) including ipilimumab; reviews evidence base at the time of ipilimumab’s introduction to adjuvant setting
40236344	2025	Case report	Cureus	Case of colonic metastasis from melanoma treated with systemic immunotherapy; highlights immune-related gastrointestinal adverse events (perforation) relevant to checkpoint inhibitor safety monitoring

Cytotoxicity

Ipilimumab is an antineoplastic agent in the immune checkpoint inhibitor class. It is used for the treatment of malignant melanoma and is classified as an immunotherapy rather than a conventional cytotoxic drug.

Item	Content
Cytotoxicity Classification	Immunotherapy — immune checkpoint inhibitor (anti-CTLA-4 IgG1 monoclonal antibody); not a conventional cytotoxic agent
Myelosuppression Risk	Low — ipilimumab is not associated with haematological toxicity in the conventional sense; immune-related adverse events (irAEs) are the primary toxicity concern
Emetogenicity Classification	Minimal — not classified as emetogenic; nausea may occur as part of an immune-related adverse event rather than direct chemotherapy-related emetogenicity
Monitoring Items	Liver function (ALT, AST, bilirubin), thyroid function (TSH, fT4), adrenal function (morning cortisol, ACTH), full blood count, renal function, blood glucose; clinical monitoring for irAEs: dermatitis, colitis, hepatitis, hypophysitis, pneumonitis, uveitis
Handling Protection	Standard biohazard precautions for monoclonal antibody preparations; does not require the full cytotoxic drug handling precautions (closed-system drug transfer devices, chemotherapy personal protective equipment) mandated for conventional cytotoxic chemotherapy

Safety Considerations

Detailed warnings and contraindications are not available in this evidence pack. Please refer to the approved Summary of Product Characteristics (SmPC) for Yervoy® for complete safety information.

Based on the established clinical profile of ipilimumab, clinicians should be particularly aware of:

Immune-related adverse events (irAEs): colitis, hepatitis, dermatitis, hypophysitis, adrenal insufficiency, and pneumonitis — managed with high-dose corticosteroids and, in severe cases, permanent discontinuation

Ocular irAEs: uveitis and other inflammatory eye conditions have been reported — this is especially relevant given the choroideremia prediction discussed below, where pre-existing retinal pathology would represent a significant safety concern

No drug-drug interaction data is available in this evidence pack (DDI query returned no results)

Secondary TxGNN Prediction: Choroideremia — Hold

The TxGNN model assigns its highest score in this evidence pack to choroideremia (99.06%), placing it above non-cutaneous melanoma by a small margin. Choroideremia is a rare X-linked inherited retinal dystrophy caused by mutations in the CHM gene (Xq21.2), leading to progressive degeneration of retinal pigment epithelium (RPE) and photoreceptors due to Rab Escort Protein-1 (REP-1) deficiency.

Evidence level: L5 — model prediction only; no clinical or experimental evidence exists.

There is no established biological link between ipilimumab’s CTLA-4 blockade mechanism and the primary monogenic pathology of choroideremia. The TxGNN score likely reflects indirect connections within the knowledge graph (e.g., shared disease–protein interaction networks relating to the ocular immune microenvironment or the eye’s immune-privileged status), rather than a direct mechanistic relationship. No hypothesis-generating publications, animal studies, or exploratory clinical studies support this application.

Importantly, ipilimumab can itself cause immune-related uveitis, indicating a dual — and potentially harmful — immune impact on ocular tissue. This raises significant safety concerns about its use in patients with pre-existing retinal disease.

Recommendation: Hold — not appropriate for clinical pursuit without substantial preclinical mechanistic investigation.

Conclusion and Next Steps

Decision: Proceed with Guardrails (for non-cutaneous melanoma)

Rationale: Multiple completed Phase 3 randomised trials provide Level 1 evidence supporting ipilimumab — particularly in combination with nivolumab — for the treatment of advanced melanoma including non-cutaneous subtypes. Direct evidence from published literature confirms activity in uveal and mucosal melanoma specifically.

To proceed, the following is needed:

Verify Danish market authorisation status: Cross-check Yervoy® registration with the EMA and Lægemiddelstyrelsen databases; the 0 authorisations recorded in this evidence pack appears inconsistent with the known EMA centralised authorisation
Obtain the approved SmPC: Review EU-approved indications to clarify whether non-cutaneous melanoma subtypes (uveal, mucosal) are explicitly included or are off-label use in Denmark
Document the mechanism of action formally: MOA data is listed as a data gap in this evidence pack; obtain from DrugBank API or the SmPC pharmacodynamics section
Retrieve key warnings and contraindications: Download and parse the Yervoy® SmPC for the safety section; currently blocking (Data Gap severity: Blocking)
Subtype-specific response rate data: Uveal melanoma response rates to ipilimumab monotherapy are substantially lower than cutaneous melanoma; formal patient-selection criteria differentiating uveal vs. mucosal vs. other non-cutaneous subtypes should be defined before treatment initiation
Immune-related adverse event management protocol: Establish a clinical pathway for irAE monitoring and management, with particular attention to ocular irAEs given the uveal melanoma patient population
For choroideremia prediction: Preclinical mechanistic studies are required before any clinical consideration; this prediction should be classified as Hold pending further investigation

This report is intended for research reference purposes only and does not constitute medical advice. Drug repurposing candidates require clinical validation before application. This report should be read in conjunction with the approved Summary of Product Characteristics and current clinical guidelines.

Disclaimer

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