Using AI to predict melanoma treatment response based on skin damage

Generation of an Artificial Intelligence Algorithm Based on the Analysis of Melanoma Peri-scar Dermatoheliosis, as a Predictive Factor of Response to Anti-PD-1

This study is testing if an AI can help predict how well people with advanced melanoma will respond to a specific treatment by looking at their skin damage from sun exposure.

Quick facts

• Study type: Observational
• Enrollment: 700 (estimated)
• Ages: 18 Years and up
• Sex: All
• Sponsor: Nantes University Hospital (Academic / other)
• Drugs / interventions: Chemotherapy, immunotherapy
• Locations: 20 sites (Besançon, Bourgogne-Franche-Comté and 19 other locations)
• Trial ID: NCT05856565 on ClinicalTrials.gov

What this trial studies

This observational study aims to develop an artificial intelligence algorithm that analyzes visible skin damage, known as dermatoheliosis, to predict patient responses to anti-PD-1 therapy in metastatic melanoma. By leveraging the correlation between high tumor mutational burden (TMB) and UV-induced skin changes, the study seeks to identify patients who are most likely to benefit from immunotherapy. The research will involve both retrospective and prospective cohorts of patients with inoperable melanoma or skin carcinoma who have undergone systemic treatment. The goal is to enhance treatment personalization and minimize unnecessary toxicity for non-responding patients.

Who should consider this trial

Good fit: Ideal candidates include adult patients with inoperable stage III or IV melanoma or inoperable skin carcinoma who have received systemic treatment for at least 3 months.

Not a fit: Patients who have received adjuvant immunotherapy within the last 6 months or those whose treatment duration is less than 90 days may not benefit from this study.

Why it matters

Potential benefit: If successful, this approach could significantly improve treatment outcomes for patients with metastatic melanoma by identifying those most likely to respond to immunotherapy.

How similar studies have performed: While the approach of using AI for predictive analytics in melanoma treatment is innovative, previous studies have shown the importance of TMB in treatment response, indicating potential for success in this novel application.

Eligibility criteria

Inclusion Criteria:

▪ Adult patients with inoperable stage III or IV melanoma, or inoperable skin carcinoma (squamous cell carcinoma or basal cell carcinoma).

Retrospective cohort: patients who have received curative treatment with anti-PD1, +/- anti-CTLA-4 or anti-LAG-3 for their skin cancer for at least 90 days, with at least 6 months of follow-up, without immunosuppression and whose primary tumour site is not altered by concomitant dermatosis. Adjuvant immunotherapy is tolerated if it was stopped at least 6 months before the start of curative treatment. Interferon is also tolerated if it was stopped at least 6 months before the start of curative treatment. Radiotherapy is tolerated if it did not take place at the site of the primary melanoma scar. For squamous cell carcinomas, prior radiotherapy on the scar is acceptable (it must simply not have been administered during the period of anti-PD-1 treatment in order to be able to reliably assess the response).

Inoperable primary tumours are eligible. Targeted therapy is accepted before the start of curative treatment with immunotherapy.

Chemotherapy is not accepted prior to the initiation of curative treatment with immunotherapy.

▪ Prospective cohort: Patients who have not received immunotherapy for the management of their skin cancer at the start of curative treatment with anti-PD-1, +/- anti-CTLA-4 or anti-LAG-3.

Adjuvant immunotherapy is tolerated if it has been discontinued for at least 6 months prior to the start of curative treatment. Interferon is also tolerated if it has been discontinued for at least 6 months prior to the start of curative treatment. Radiotherapy is tolerated if it has not been administered at the site of the primary cancer scar.

Targeted therapy is accepted before starting curative treatment with anti-PD-1, +/- anti-CTLA-4 or anti-LAG-3.

Chemotherapy is not accepted before the start of curative treatment with anti-PD-1, +/- anti-CTLA-4 or anti-LAG-3.

▪ Patients who have agreed to participate in the research and have signed an image rights authorisation form.

Exclusion criteria:

* Retrospective cohort: Patients who have received curative treatment with anti-PD-1, +/- anti-CTLA-4 or anti-LAG-3 for skin cancer for less than 90 days
* Patients who received adjuvant immunotherapy within 6 months prior to curative treatment
* Patients who received chemotherapy prior to curative treatment
* Patients whose primary skin cancer site cannot be photographed (e.g. choroidal melanomas, mucosal melanomas, with the exception of vulvar or penile melanomas, etc.).
* Patients treated with systemic corticosteroids (dose greater than 10 mg/day) at the start of the immunotherapy in question,
* Patients who are immunocompromised (associated blood disorder, human immunodeficiency virus infection, transplant patients, etc.) at the start of immunotherapy,
* Patients with iatrogenic peri-scar vitiligo,
* Patients who have refused to participate in the research,
* Adults protected by law.

Where this trial is running

Besançon, Bourgogne-Franche-Comté and 19 other locations

• Besancon University Hospital — Besançon, Bourgogne-Franche-Comté, France (Recruiting)
• Brest University Hospital — Brest, Finistère, France (Recruiting)
• Angers University Hospital — Angers, Maine-et-Loire, France (Recruiting)
• Blois Hospital site — Blois, France (Not_yet_recruiting)
• Bordeaux University Hospital — Bordeaux, France (Not_yet_recruiting)
• Dijon University Hospital — Dijon, France (Not_yet_recruiting)
• Grenoble University Hospital — Grenoble, France (Not_yet_recruiting)
• CHU de La Rochelle — La Rochelle, France (Not_yet_recruiting)
• CH du Mans — Le Mans, France (Not_yet_recruiting)
• Léon Site Bérard in Lyon — Lyon, France (Not_yet_recruiting)
• Nantes University Hospital — Nantes, France (Recruiting)
• Ambroise Paré Hospital - APHP — Paris, France (Not_yet_recruiting)
• Avicenne Hospital - APHP — Paris, France (Not_yet_recruiting)
• Bichat Hospital - APHP — Paris, France (Not_yet_recruiting)
• Saint-Louis Hospital - APHP — Paris, France (Not_yet_recruiting)
• CHU de Rennes — Rennes, France (Not_yet_recruiting)
• Eugène Marquis site - Rennes — Rennes, France (Not_yet_recruiting)
• Rouen University Hospital — Rouen, France (Not_yet_recruiting)
• Ico — Saint-Herblain, France (Not_yet_recruiting)
• Valence Hospital Site — Valence, France (Not_yet_recruiting)

Study contacts

• Study coordinator: Lise BOUSSEMART, PU-PH

Conditions Metastatic Melanoma