1105P - Predicting the outcomes of advanced cutaneous melanoma cases following discontinuation of immune checkpoint inhibition

Background

Cutaneous melanoma (CM) is among the most aggressive skin cancers. Immune checkpoint inhibition (ICI) has transformed the treatment of advanced melanoma, notably improving overall and progression-free survival rates. However, biomarkers to identify patients that can safely stop ICI are still lacking.

Methods

This multicentre study involved 83 patients diagnosed with unresectable or metastatic CM who electively stopped treatment with anti-PD-1 monotherapy or combination with anti-CTLA-4 in the absence of disease progression (Relapse vs relapse-free: N=27 vs 56). Patients with a minimum of 2 years of follow-up data were included in the study. Subsequent analysis utilized targeted-transcriptome profiling and image analysis. A total of 770 genes related to cancer-immune pathways were assessed and 5 image parameters were measured on tumor tissues using H&E images: tumor cell, lymphocyte, neutrophil, plasma cell, and eosinophil densities.

Results

Out of the five parameters assessed through image analysis, tumor cell density, plasma cell density, and eosinophilic granulocyte density significantly stratified CM patients based on their progression-free survival (PFS) after discontinuation of ICI (p = 0.04, 0.04, 0.017, respectively). To enhance stratification, a cumulative score combining lymphocyte, plasma cell, and tumor cell density parameters was generated, effectively stratifying patients based on PFS (p < 0.0001). Cox regression analysis indicated significant associations between 8 genes and PFS in our cohort. Multivariate Cox regression analysis revealed an interaction between TGFBR1, LOXL2, and the image parameter, where high expression of TGFBR1 and a higher score in the image parameter were significantly associated with relapse after discontinuation of ICI. Using these parameters, a model was trained and achieved 84.6% accuracy in predicting outcomes in the test cohort.

Conclusions

Through our study, we propose a novel approach to patient risk stratification for tumor progression after ICI treatment. Our investigation into predictive biomarkers for relapse post-ICI cessation aims to aid patients and physicians in informed decision-making.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Deutsche Krebshilfe (DKH).

Disclosure

All authors have declared no conflicts of interest.