Abstract LB296: Clinical validation of a liquid biopsy microRNA assay for diagnosis and risk stratification of invasive cutaneous melanoma

Abstract Background: Rates of melanoma diagnoses have increased dramatically over the past 40 years, without the proportionate decrease in mortality observed in other common cancer types. Factors that make the early and accurate diagnosis of invasive cutaneous melanoma (ICM) challenging include (i) the large number of potential locations and small size of pigmented lesions, (ii) inherent subjectivity in visual diagnostic examinations, and (iii) a lack of objective molecular biomarkers. A 38-microRNA expression signature of melanoma (MEL38) was previously identified in a plasma-based pilot study and independent validated in multiple series of solid tissue biopsies. Patients & Methods: Using digital microRNA expression profiling, the diagnostic and prognostic performance of MEL38 was assessed using prospectively-collected or archival plasma from 582 patients. Criteria for inclusion was histologically confirmed ICM (n=372), or other conditions including nevi, melanoma in-situ (MIS) and non-melanoma skin cancer (n=210). A second microRNA profile was also developed, optimised to predict a patient's probability of 10-year melanoma-specific survival (MSS). Results: The MEL38 score correctly identified the ICM status of 551/582 (95%) plasma microRNA profiles, with an area under the curve of 0.98 (P<0.001). Using a MEL38 score of >5.5 (range 1 to 10) resulted an ICM detection sensitivity of 93% and specificity of 98%. Multivariate analysis indicated the MEL38 score is significantly different between ICM vs other disease states, independent of patient age, gender, biopsy site or plasma type (fresh vs archival) at P<0.001. In the ICM subset of patients (AJCC Stage I to 4), the MEL38 score was significantly associated with MSS (log rank P=0.0011). An optimised 12-microRNA signature (MEL12) specifically for melanoma prognosis was developed to compliment the diagnostic ability of MEL38. Rates of MSS between low, standard and high-risk MEL12 groups were 94%, 78% and 58% respectively (Log rank P<0.001). Risk group assignment by MEL12 was significantly associated with clinical staging (Chi square P<0.001) and SLNB status (P=0.027). Melanoma was detected in the SLNB of 87% of patients with the high risk MEL12 profile. The observed differences in MSS between MEL12 risk groups remained statistically significant when adjusted for variation in patient age, gender, systemic treatment and local vs metastatic melanoma. Conclusion: In a large series of clinically annotated plasma samples, the MEL38 signature was able to accurately stratify patients with ICM from those with non-ICM conditions associated with negligible or dramatically lower risk of mortality. In patients clinically or genomically-diagnosed with ICM, the MEL12 signature can provide a prognostic assessment, significantly related to SLNB status, clinical stage, and probability of 10-year MSS. Adoption of plasma microRNA profiling in the clinic may facilitate personalised, risk-informed treatment decisions, and potentially reducing the rates of melanoma under and over diagnosis. Citation Format: Ryan Van Laar, Babak Latif, Sam King, Christopher Love, Nadine Taubenheim, Esha Kalanooriya, Wandi Wang, Mirette Saad, Ingrid Winship, Tony Landgren. Clinical validation of a liquid biopsy microRNA assay for diagnosis and risk stratification of invasive cutaneous melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr LB296.