Oral Presentation Australasian Melanoma Conference 2026

Circulating Antibody Profiling Identifies Predictive Biomarkers of Outcome and irAEs in Melanoma   (#13)

Joanna Chan 1 2 , Eliza Grills 3 , Ashkan Mehrnejad 4 , Mustafa Mohamed 4 , Samantha Hopkins 2 , Anupama Pasam 4 , Natalie Brajanovski 4 , Lavinia Spain 1 2 , Ramin Alipour 5 , Aparna Rao 1 2 , Damien Kee 1 2 , Belinda Lee 1 2 , Benjamin Brady 2 , Grant McArthur 1 2 , Shahneen Sandhu 1 2 , Jessica Da Gama Duarte 3 6
  1. Sir Peter MacCallum Department of Oncology,University of Melbourne, , Melbourne, VIC, 3010
  2. Department of Medical Oncology, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia
  3. Department of Cancer Medicine, Alfred Hospital, School of Translational Medicine, Monash University, Melbourne, VIC, Australia
  4. Peter Maccallum Cancer Centre, Parkville, VIC, Australia
  5. Prostate Cancer Theranostics and Imaging Centre of Excellence, Molecular Imaging and Therapeutic Nuclear Medicine, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia
  6. Olivia Newton-John Cancer Research Institute, and La Trobe School of Cancer Medicine, Heidelberg, VIC, Australia

Background: Immune checkpoint inhibitors (ICI) have transformed outcomes in stage III/IV melanoma. However, immune-related adverse events (irAEs) are common and associated with significant morbidity. Emerging evidence suggests that autoantibodies may serve as biomarkers of immune activation and predictors of toxicity with pre-existing and treatment-emergent autoantibodies associated with an increased risk of irAEs. MicroMAC is an observational longitudinal study investigating microbiome and immune dynamics in patients with stage III/IV melanoma receiving ICI. By integrating longitudinal biospecimen collection with toxicity and response data, this study aims to identify predictors of irAEs, recurrence and response to ICI.

Methods: Baseline serum samples were collected from 133 patients, 55 with resected stage III/IV melanoma receiving adjuvant single-agent anti-PD-1 (cohort 1), and 78 with unresectable stage III/IV melanoma treated with single-agent anti-PD-1 or combination anti-PD-1/anti-CTLA4 (cohort 2). Patients underwent serial imaging with FDG-PET/CT and MRI Brain 6 monthly in cohort 1 and 3-4 monthly in cohort 2. The i-Ome protein discovery array (>1800 proteins) assessed circulating antigen-specific antibodies and IgG profiles to identify biomarkers predictive of recurrence (cohort 1), response (cohort 2), and irAEs as per CTCAE (low-grade [≤G2], high-grade [≥G3]) across cohorts. Statistical analyses were conducted in R using univariate and multivariate receiver operating characteristic (ROC) approaches (pROC, combiroc).

Results: In Cohort 1, anti-ZKSCAN3 antibodies, reflecting dysregulated cancer hallmarks, predicted recurrence (AUC 0.796), improving to 0.926 with a multivariate signature (NFE2, PRKD2, E2F6, HUS1, SOCS6, TDRKH, MIF). Anti-INA antibodies, linked to paraneoplastic neurological syndromes, predicted high-grade (G3–4) irAEs (AUC 0.921), increasing to 0.933 with TWF2, PHLDA1, CBLC, RALBP1, and PRKAA1. In Cohort 2, anti-CBX5 antibodies, associated with immune infiltration, predicted response (AUC 0.771), improving to 0.874 with NDUFAB1, GNG11, NFIL3, and DGUOK. Anti-XIAP antibodies, linked to autoimmunity and inflammation, predicted high-grade irAEs (AUC 0.714), increasing to 0.779 with LTA4H, SPATS1, PADI4, PAK6, GNG3, and HBG1.

Conclusions: Serum antibody profiling using high-content protein arrays may identify predictive biomarkers of recurrence, response, and risk of irAEs in patients with stage III/IV melanoma receiving ICI. These findings highlight the potential use of circulating antibody signatures as clinically actionable tools for patient stratification and prospective toxicity risk prediction.