Abstract LB-146: T cell receptor immunosequencing improves prediction of melanoma recurrence

Abstract Background: The prognosis of patients with early-stage melanoma remains poor with overall survival at 10 years below 40% in patients with stage IIC disease. The use of anti-PD-1 adjuvant therapy has improved outcomes for some, but the role of the immune system in early disease and selection of ideal candidates for checkpoint blockade require further elucidation. Clonality of the tumor T-cell repertoire, as measured by next-generation immunosequencing of T-cell receptors (TCRs), recently demonstrated its utility as a biomarker of response to anti-PD-1 therapy in metastatic melanoma. Here, we aim to evaluate its ability to predict recurrence in patients who underwent resection of primary melanomas. Methods: Biomarkers of progression free survival (PFS) were evaluated in a cohort of 72 archival FFPE tumors from subjects who were followed for a minimum of five years. Tumor TCR-beta clonality and infiltrating T-cell fraction (fraction of nucleated cells that are T-cells) were quantified by the immunoSEQ® Assay (Adaptive Biotechnologies) while Breslow thickness, ulceration, mitotic rate, and briskness of tumor infiltrating lymphocytes were assessed by histopathology. Results: T-cell fraction, but not clonality, was a robust independent variable in predicting PFS (Cox Proportional Hazard Ratio = 0.71, p-value = 0.007) after resection of primary melanomas in patients who were not treated with immunotherapy regimens. In a gradient boosted model, T-cell fraction was second only to Breslow thickness in weighted importance for predicting PFS, which together accounted for 60% of the model's classification power. Variables comprising the remaining 40% included ulceration, patient age, tumor stage, and mitotic rate, and, except for stage, generally correlated with Breslow thickness. T-cell fraction, however, was not correlated with any clinical or histopathological variable, including briskness. The bivariate model Cox Regression of Breslow thickness and T-cell fraction performed better than Breslow thickness alone in predicting PFS (Likelihood Ratio Test p-value = 0.03). Furthermore, patients with high T-cell fractions had delayed recurrence (median of 19 vs 9 months, high vs low T-cell fraction in melanomas > 2.5mm; median PFS not reached for melanomas < 2.5mm). Conclusion: Patients with high T-cell fractions in intermediate and high-thickness melanomas have a more favorable prognosis and may be ideal candidates for adjuvant immunotherapy to prolong PFS. These findings are currently being validated in an expanded cohort of archival samples. Disclaimers/disclosures: For Research Use Only. Not for use in diagnostic procedures. Partial financial support provided by Adaptive Biotechnologies. Citation Format: Julie A. Rytlewski, Wiebke Pruessmann, James Wilmott, Martin C. Mihm, Beatrice Dyring-Andersen, Rachael A. Clark, Erik Yusko, Alexandra Snyder, Harlan Robins, Richard Scolyer, Thomas S. Kupper. T cell receptor immunosequencing improves prediction of melanoma recurrence [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr LB-146.