*2.5. Talimogene Laherparepvec*

Talimogene laherparepvec (IMLYGICTM, T-Vec, OncoVEXGM-CSF by Amgen Inc. Cambridge, MA, USA) is a genetically engineered OV based on the HSV-1 strain JS1, which has been modified by deletion of γ*34.5* and *ICP47* as well as an insertion of the gene for GM-CSF [19] to render the virus more immunogenic. The first phase I clinical trial was performed in patients with (sub)cutaneous metastases of breast, gastrointestinal adenocarcinoma, malignant melanoma, and epithelial cancer of the head and neck to determine safety, biological activity and adequate dosing [41]. For the first part of the study, patients were categorized in three cohorts with the HSV-seropositive patients receiving the highest dose. The second part of the study focused on evaluating three dose regimens with the HSV-seropositive patients receiving the highest doses. The authors recorded pyrexia, low-grade anorexia, nausea and vomiting, fatigue, and reaction at the injection site as the main adverse events. <sup>1</sup> <sup>×</sup> <sup>10</sup><sup>7</sup> pfu/mL was declared as the maximum-tolerated dose (MTD) for seronegative patients, while no MTD for seropositive patients could be stated. All HSV-seronegative patients seroconverted, whereas in the seropositive cohort, an increase in HSV antibody titer was noted. No treatment-associated effects on cytokines were recorded. Histological analyses of tumor tissue frequently showed necrotic areas and positive HSV1 staining primarily in necrotic tumor tissues suggesting a correlation. In three patients, stable disease was achieved and in some cases size reductions of the injected tumor was seen.

Further studies on the effects of T-Vec on clinical response and survival were conducted by Senzer et al. in a phase II study for patients with unresectable stage IIIc and stage IV melanoma [42]. The patients each received initial intratumoral injections, followed three weeks later by injections every two weeks for a possible total of 24 treatments. All seronegative patients seroconverted. Eighty-five percent of the patients experienced grade I/II adverse effects with the most common being fever, chills, fatigue, nausea, and vomiting, as well as headache. Treatment was associated with local as well as

distant responses in lung, liver, pancreas, lymph nodes, and soft tissue. Clinical response assessment resulted in 20% complete responses; 13% of the patients were classified as having "no evidence of disease" with some cases involving additional surgery. Overall median survival was 16 months, and the one-year survival rate of patients with complete or partial response totaled 93%. Kaufman et al. [43] further analyzed local and distant immune responses of this patient cohort. The authors used peripheral blood mononuclear cells (PBMCs) from study patients, non-study patients, and healthy donors as well as tumor tissue from study patients and non-study melanoma patients to compare the immune cell status. Higher amounts of activated CD8+-T-cells expressing Perforin and Granzyme B as well as PD-1 expressing T-cells and Tregs in the local tumor tissue compared to the periphery in non-study melanoma patients were observed. Functional analysis of tumor infiltrating lymphocytes (TILs) and PBMCs from a study patient showed an enrichment of MART-1-specific T-cells, indicating the development of a T-Vec-mediated systemic anti-tumor immunity. Moreover, a decrease of CD4+-T-cells, Tregs, T-suppressor cells, and myeloid-derived suppressor cells (MDSCs) within TILs of study patients compared to non-study patients was noted. A comparison of immune cell populations between treated tumor sites and peripheral tumor sites showed more distinct local responses but still provided evidence for the induction of a systemic anti-tumor immunity.

A randomized phase III trial of T-Vec compared to GM-CSF in patients with unresected stage IIIB-IV melanoma [13] showed that T-Vec treatment resulted in a prolonged median overall survival (23.3. vs. 18.9 months T-Vec vs. GM-CSF) and an improved durable response rate in T-Vec patients (16.3%) in contrast to GM-CSF-treated patients (2.1%). The T-Vec dosing scheme followed previous strategies [42], while GM-CSF was administered daily for two weeks in 28-day cycles. The most common adverse events in the T-Vec cohort included chills, pyrexia, pain at the injection site, nausea, influenza-like symptoms, and fatigue, therefore matching the profile of adverse events seen in preceding studies. In the T-Vec group, the authors further observed decreased size of more than 50% in injected as well as in uninjected lesions [44], which points to the development of a systemic anti-tumor response as previously reported [43].

Additional clinical data show that oHSV therapy appears to work well with immune checkpoint blockade. Combined treatment of T-Vec (1 <sup>×</sup> <sup>10</sup>6–1 <sup>×</sup> <sup>10</sup><sup>8</sup> pfu/mL) with the CTLA-4 blocking monoclonal antibody ipilimumab in 19 patients with stage III and IV melanoma did not lead to dose-limiting toxicities [45]. Moreover, Puzanov et al. [45] reported 22% complete responses, 28% partial responses, and 22% stable disease and an objective response rate of 50% referring to immune-related response criteria. As already noted in previous studies with T-Vec monotherapy, both injected and uninjected tumor lesions showed a size reduction after treatment with T-Vec and ipilimumab. Significant enrichment of total CD8<sup>+</sup> and activated CD8+-T-cells during T-Vec monotherapy as well as a gain of ICOS-expressing CD4+-T-cells during combination therapy was observed.
