e21542 Background: Many melanoma patients do not respond to anti-PD1 therapy due to lack of antigen specific responses. IFx-Hu2.0 (plasmid DNA encoding the streptococcal membrane protein, Emm55, contained within a cationic polymer) primes innate and antigen dependent responses in murine/equine melanoma models to produce an environment needed for checkpoint inhibitor efficacy. We describe the first in human study utilizing IFx-Hu2.0 in unresectable melanoma – NCT03655756. Methods: Melanoma patients (unresectable stage III/IV) had cutaneous lesions injected with IFx-Hu2.0 to test safety and feasibility. Patients were refractory to standard of care (anti-PD1, BRAF/MEK) or did not wish these treatments. 1-3 lesions (> 3 mm – 0.1 mg/0.2 mL) were injected, pre/post treatment biopsies obtained, and the primary endpoint of 5/6 patients without dose limiting toxicity (DLT) was assessed at 28 days. Retreatment was permitted. ≥2 lesions were needed: one for injection and uninjected lesion for biopsy. Tissue samples were analyzed for mRNA profiles, antigen responses (PEPperPRINT assay), and multiplex immunofluorescence (markers: CD3, CD8, FOXP3, PD1, PDL1, SOX10, DAPI). Results: The primary endpoint was met in 6 evaluable patients out of 7 enrolled. Observed toxicities included: G1-2 Injection site reactions – 5/7; G1 Bleeding – 1/7; G1-2 Pain – 2/7, G1 Lymphopenia – 1/7, G1 Pruritis – 1/7; with no ≥ G3 toxicities related to study drug observed. One G5 toxicity (Clostridium septicum infection 20 days post injection) was deemed unlikely related to study drug. 5/6 patients received 1 cycle prior to post-protocol immune-based therapy. One treatment naïve patient retreated once with IFx-Hu2.0 required no additional therapy > 9 months. Available paired tissue and plasma sampling revealed increased T cell infiltration into treated lesions, increase in IgM and IgG epitope recognition to melanoma associated antigens in the plasma (detected by PEPperPRINT assay), an increase in mRNA associated with innate immune responses in the injected lesion (CXCL13, LAG3, CXCL11, CXCL10, ICOS) and an adaptive immune response (IL-12, HLA-DRB5, WNT4, CD3D, Arg I) in uninjected lesions associated with downregulation of known melanoma antigens. Of 4 anti-PD1 refractory patients, three patients had clinical benefit to post-protocol retreatment with anti-PD1 based therapy (Stable Disease (SD) lasting > 2 years followed by surgical resection, Partial Response (PR) lasting > 9 months, PR subsequently surgical resected and rendered no evidence of disease). Conclusions: In this pilot study, intralesional IFx-Hu2.0 demonstrated a favorable safety profile. These data support encouraging immunological correlative responses and further study of IFx-Hu2.0 as a priming agent to enhance or restore sensitivity to immune checkpoint inhibitor therapy in melanoma. Clinical trial information: NCT03655756.
