(P 237) INTRALESIONAL ADMINISTRATION OF TALIMOGENE LAHERPAREPVEC (T-VEC) INDUCES TUMOR REGRESSION IN PATIENTS WITH CUTANEOUS ANGIOSARCOMA

Objective: Cutaneous angiosarcoma (CAS) is a rare and highly malignant soft tissue sarcoma with poor survival rate. Complete resection with wide margins is the only curative treatment but often aggressive resection is limited by anatomic considerations or tumor size given the diffuse pattern of disease. For this reason, there is a significant unmet need for effective local control therapy in such cases. Talimogene Laherparepvec (T-VEC, Imlygic®, Amgen) is an FDA-approved oncolytic virus for the treatment of cutaneous melanoma which effects an anti-tumor response through the recruitment of an immune response to the tumor microenvironment of injected lesions. The purpose of this clinical trial was to test the efficacy T-VEC in unresectable CAS and to assess the degree of immune response following treatment.

Methods: Treatment consisted of initial intralesional injection of up to 4ml 10^6 PFU/mL T-VEC (volume dependent on tumor size) followed by bi-weekly intralesional injections of up to 4ml 10^8 PFU/mL T-VEC for up to 12 months. Overall response to treatment was assessed by direct caliper tumor measurement prior to each injection with T-VEC and RECIST-based response categories were assigned based on the sum of the diameters. Skin punch biopsies were performed on treated lesions before first injection and at the conclusion of treatment. Multiplex Immunofluorescence Microscopy (mIF) was performed on these tissue samples to describe the immune infiltration in the tumor microenvironment of injected lesions. Data are presented as median (range).

Results: Five patients (69-86 years old; 4 female, 1 male) were enrolled and received treatment. The median aggregate baseline tumor size was 3.1cm (0.9-39.0) and 3/5 (60%) patients completed 52 weeks of treatment. The most common treatment-related adverse events were fever (60%), fatigue (60%) and injection-site pain (40%) (all G1-2). Overall response rate at 1-year follow up was 60% (3/5 PR, 2/5 PD). Two patients with clinical PR were found to have pathologic CR on post-treatment biopsy. The median overall reduction in tumor size for partial responders was 71.3% (30.7-88). Of those with PD, one had distant progression despite partial response of injected disease (75.7% reduction) and the other had local progression (25.8% increase). Treatment was well-tolerated with no G3 or higher AEs. Pre- and post-treatment biopsy samples were available for three patients. Post-treatment biopsies had increased immune cell infiltration after T-VEC treatment (CD3+ median 0.88-fold change, 0.47 – 11.25). CD8+ T-cell infiltration was increased in all patients (median 14.95-fold change, 1.80 – 51.22), while median fold change of CD4+ T cells was 0.80 (0.28 – 11.94) and 1.81 (0.48 – 337.39) for CD20+ B-cells (FIGURE).

Conclusion: This is the first report that describes efficacy of T-VEC as monotherapy for unresectable CAS. Furthermore, we have shown that treatment induces a marked increase in immune infiltration, predominantly of CD8+ T cells. Despite low sample size (though concordant with the incidence of disease), we show a 40% pathologic complete response rate which suggests further exploration in a larger study is warranted.