Molecular markers for prediction of response and progression free survival to novel therapies in cutaneous malignant melanoma

Drugs inhibiting the MAPK-pathway (MAPKi) and immune checkpoint inhibitors (ICI) have changed the clinical outcome of metastatic cutaneous malignant melanoma (CMM) in a significant way. Nonetheless, many patients have primary resistance or develop acquired resistance to these therapies within a relatively short period of time. This thesis was performed to explore mechanisms of resistance and possible predictive biomarkers to further improve treatment outcome and to help individualize treatment in this patient population.

In Paper I, we compared mRNA and protein expression in MAPKi resistant and sensitive melanoma cell lines. By applying gene expression and proteome profiling we identified two previously described (MET and EPHA2) and two novel (FLI1 and CD13/ANPEP) candidate biomarkers that, when overexpressed, were associated with treatment resistance to MAPKi. The overexpression of MET and EPHA2 was confirmed in melanoma samples from patients with metastatic CMM when comparing samples taken before and after treatment with MAPKi. In cell lines, we demonstrated that an inhibitor of EPHA2 (the multikinase inhibitor dasatinib), re-sensitized cells to MAPKi treatment.

In Paper II, we analyzed plasma samples from 28 patients with metastatic CMM before and during treatment with MAPKi. Micro-RNA (miRNA) was extracted from plasmatic extra cellular microvesicles (EVs) and miRNA profiling was performed by microarray, using a panel with 372 human miRNAs. We assessed the association of the miRNA levels with response to treatment and progression free survival (PFS) and found that an increased level of let-7g-5p during treatment, compared to before treatment, was correlated with better response. A high level of miRNA 497-5p during treatment was associated with longer PFS.

In Paper III, we investigated if plasmatic proteins were related to response and PFS to MAPKi or ICI in 109 patients with metastatic CMM. Proteomic profiling of plasma samples collected before and during treatment was performed by mass spectroscopy and the abundance of proteins was then correlated with treatment response and PFS. We identified that the plasma levels of 43 proteins, either before or during treatment, were prognostic/predictive of treatment outcome. An inverse correlation between acute-phase inflammatory proteins and apolipoproteins was observed. Patients with high levels of acute-phase inflammatory proteins and low levels of apolipoproteins had worse outcome to therapy.

In Paper IV, we analyzed mRNA expression by targeted RNA sequencing of pre-treatment tumor samples from 28 patients with metastatic CMM who underwent treatment with MAPKi or ICI. Transcriptomic data was correlated with treatment response and PFS in gene set enrichment analysis (GSEA). Enrichment of genes in IFN-gamma and inflammatory response was associated with longer PFS to MAPKi therapy, and decreased expression of proliferative genes and increased expression of immune genes correlated with longer PFS to ICI. Finally, lower expression of proliferation genes and immune evasion genes was associated with increased response to ICI.

In summary, we have identified possible mechanisms of resistance and potential predictive biomarkers to novel therapies in patients with metastatic CMM. Our studies were performed in small cohorts of patients and further studies to validate our findings are warranted.

List of scientific papers

I. Silencing FLI or targeting CD13/ANPEP lead to dephosphorylation of EPHA2, a mediator of BRAF inhibitor resistance, and induce growth arrest or apoptosis in melanoma cells. Alireza Azimi, Rainer Tuominen, Fernanda Costa Svedman, Stefano Caramuta, Maria Pernemalm, Marianne Frostvik Stolt, Lena Kanter, Pedram Kharaziha, Janne Lehtiö, Carolina Hertzman Johansson, Veronica Höiom, Johan Hansson, Suzanne Egyhazi Brage. Cell Death Dis. 8, e3029 (2017).
https://doi.org/10.1038/cddis.2017.406

II. Extracellular microvesicle microRNAs as predictive biomarkers for targeted therapy in metastastic cutaneous malignant melanoma. Fernanda Costa Svedman, Warangkana Lohcharoenkal, Matteo Bottai, Suzanne Egyhazi Brage, Enikö Sonkoly, Johan Hansson, Andor Pivarcsi, Hanna Eriksson. PLoS One. 13, e0206942 (2018).
https://doi.org/10.1371/journal.pone.0206942

III. Inflammation and apolipoproteins as potential biomarkers for stratification of cutaneous melanoma patients for immunotherapy and targeted therapy. Max J. Karlsson, Fernanda Costa Svedman, Abdellah Tebani, David Kotol, Veronica Höiom, Linn Fagerberg, Fredrik Edfors, Mathias Uhlén, Suzanne, Egyházi Brage, Gianluca Maddalo. [Accepted]
https://doi.org/10.1158/0008-5472.CAN-20-2000

IV. Gene enrichment signatures in proliferation and immune response as potential biomarkers for clinical outcome to immune checkpoint inhibitors and to targeted therapy in metastatic cutaneous malignant melanoma. Fernanda Costa Svedman, Ishani Das, Rainer Tuominen, Eva Darai Ramqvist, Johan Hansson, Veronica Höiom, Suzanne Egyhazi Brage. [Manuscript]