Gene Expression Profiles and ctDNA for Risk Stratification in Patients With Melanoma

Cutaneous melanoma is a malignancy arising from melanocytes of the skin. Incidence rates are rising, particularly in White populations.

For patients with clinically node-negative (cN0) disease, sentinel lymph node biopsy (SLNB) is indicated for all patients with a melanoma of thickness of 0.8 mm or more (\>pT1b) to stratify the prognosis and guarantee access to systemic adjuvant treatments.

Indeed, adjuvant immune checkpoint inhibitors or anti-proto-oncogene B-Raf (anti- BRAF) / Mitogen-activated protein kinase kinase (MEK) targeted agents were first demonstrated to improve clinical outcomes in patients with nodal involvement (stage III melanoma). Subsequently, immune-checkpoint inhibitors demonstrated a benefit in terms of disease-free survival also in patients with stage IIB and IIC melanoma, i.e., with melanoma of thickness of 2 mm or more (\>pT3b). As a consequence, in this subgroup of patients, SLNB has lost its therapeutical implications and maintains only a role for prognostication, i.e., in the distinction between stages IIB-IIC (SLNB-negative; 71% of cases) and stage III (SLNB-positive; 29% of cases).

However, SLNB can be complicated by seroma, bleeding, wound infection, nerve damage, and even low rates of lymph edema have been reported. Moreover, depending on the country and healthcare system, it is a costly procedure for patients and society. According to an idea of professor Umberto Veronesi, Gentilini et al. demonstrated that the omission of SLNB in patients with cT1N0M0 breast cancer has not an impact on distant disease-free survival at 5 years (SOUND trial). Novel prognostic biomarkers for melanoma can substitute the SLNB role in prognostication and eventually lead to SLNB de-escalation.

Gene expression profiles (GEPs) and circulating tumor DNA (ctDNA) represents promising methods to assess tumor burden and tumor invasiveness, thus stratifying the prognosis.

ctDNA demonstrated to predict recurrence in patients with resected stage II-III melanoma, with a sensitivity and a specificity of 11-80% and 55-100%, according to the method (tumor-informed vs. tumor-naive), the techniques (digital droplet PRC \[ddPCR\] or others), the approach (single time point vs. dynamic assessment). In patients with non-resected stage II-III melanoma, the ctDNA-positivity rate is 35-37%. However, no data about pre-operative ctDNA as a predictor of SLNB status is available.

GEPs, alone or combined with clinicopathological features (CP-GEP), demonstrated to predict outcomes in patients with non-metastatic melanoma. In particular, MerlinTM assay is a CP-GEP model, developed by logistic regression modeling, which combines clinicopathologic factors (age and Breslow thickness) with the gene expression profiling component of 8 specific genes involved in cancer metastasis and melanosome biogenesis. It identifies patients at high or low risk of nodal metastasis. However, most patients included in studies validating this assay had a pathologic tumore stage 1-2 (pT1-2) melanoma; in patients with pathologic tumore stage 3 (pT3) melanoma this assay is characterized by high sensitivity but low specificity, and a negative predictive value of 75%.

The combination of CP-GEP and ctDNA may improve the prognostication and the prediction of SLNB status in patients with a \>pT3b melanoma (candidate to adjuvant therapy regardless of nodal status), eventually leading to the de-escalation of SLNB. Indeed, in this population, in case of concordance between biomarker-positivity and SLNB status, SLNB can be definitively omitted, without missing any information.

• the estimation of the correlation between ctDNA-positivity and SLNB-positivity [Time frame: 36 months]
• The estimation of the correlation between MerlinTM/ctDNA-positivity and SLNB-positivity [Time frame: 36 months]