LAUSR

e14039 Background: Combining stereotactic ablative radiotherapy (SABR) and anti-PD1 therapy may result in synergism but the response and toxicity rates has not been established. Biomarkers to predict clinical outcomes in this setting could improve patient selection, informed consent and cost-effectiveness. Here, we report exploratory analyses from Phase I clinical trial. Methods: Metastatic melanoma patients with at least two metastases received SABR to a single metastatic site and standard dose immunotherapy with anti-PD1 and/or anti-CTLA4 therapy. Peripheral blood mononuclear cells (PBMC) were collected prior to each immunotherapy cycle and analysed with 17-color flow cytometry to quantify CD4+ and CD8+ T cell subsets as potential biomarkers for response and toxicity. RECIST v1.1 at six months dichotomized responders (stable disease or any response) from non-responders (progressive disease). The presence of CTCAE v4 defined grade 3 or higher toxicity within six months dichotomized those with and without significant toxicity. Mann-Whitney tests were used to assess differences in biomarkers between response and toxicity groups. The Phi Coefficient (pr) was used to correlate these groups. Results: Sixteen of 25 patients had bloods available for testing. There were 11 responders (5 non-responders) and three with significant toxicity (13 without). At 12 months all responders were still alive, while all non-responders had died. Response significantly correlated with developing toxicity (pr = 0.869, p = 0.027). Prior to treatment, responders had a significantly higher percentage of total regulatory T cells (Tregs; CD4+CD25+CD127-) compared to non-responders (p = 0.011). After SABR and four cycles of immunotherapy, Treg frequencies had significantly changed (p = 0.046), with decreases in most responders and increase in all non-responders. Other CD4+ and CD8+ did not correlate with response or toxicity. Conclusions: Patients developing significant toxicities appear more likely to respond, reinforcing the importance of aggressively managing toxicities rather than ceasing treatment. Patients with higher pre-treatment Treg population and decreasing levels with treatment were significantly more likely to respond to SABR and immunotherapy. Clinical trial information: ACTRN12616001064493.