Strong therapeutic efficacy of oncolytic virus KLS-3021 in orthotopic head and neck cancer models regardless of PD-L1 expression via tumor microenvironment remodeling

Treatment of confined or locally advanced head and neck squamous cell carcinoma (HNSCC) remains challenging, even though surgery, radiotherapy, and chemotherapy are standard therapeutic options. High recurrence rates after treatment and limited systemic therapeutic efficacy always meet persistent unmet needs. Recent advances have highlighted the value of neoadjuvant immuno-oncology strategies, which can reduce tumor burden and prime systemic immunity to improve postsurgical outcomes. In recurrent or metastatic HNSCC, first-line therapy is guided by PD-L1 expression, using immune checkpoint inhibitors, chemotherapy, or their combinations; however, clinical benefit remains limited regardless of PD-L1 status. These limitations emphasize the need for new treatment modalities capable of reducing tumor burden and eliciting robust antitumor immunity regardless of PD-L1 levels. KLS‑3021 is a genetically engineered oncolytic vaccinia virus encoding PH-20 (hyaluronidase), interleukin‑12, and a soluble PD1-Fc. This design enables extracellular matrix (ECM) degradation, intratumoral immune activation, and localized blockade of PD‑1/PD‑L1 signaling. In this study, the therapeutic activity of KLS‑3021 was evaluated in orthotopic HNSCC models. The CAL27 xenograft model was used to investigate antitumor efficacy, viral spread, and apoptotic tumor cell death after a single intratumoral dose of KLS-3021. Syngeneic NOOC1 (PD‑L1-high) and MOC2 (PD‑L1-low) models were used to study PD-L1-independent tumor control by KLS-3021 and to compare efficacy against standard therapies-anti-PD-1 antibody and cisplatin, respectively-while characterizing TME remodeling and immune activation. In CAL27 model, KLS‑3021 produced robust antitumor efficacy, including complete regression with prolonged survival. In NOOC1 model, KLS‑3021 outperformed anti‑PD‑1 therapy in both tumor regression and TME reprogramming. Notably, KLS-3021 markedly degraded the intratumoral ECM, enhancing viral spread and immune cell infiltration, and shifted the TME toward an M1-dominant state with increased CD86 and TNF-alpha and reduced CD206 levels. This M1-skewed milieu was accompanied by CXCL10 upregulation and increased infiltration of activated T cells. KLS-3021 also induced immunogenic cell death establishing conditions favorable for durable antitumor immunity. In MOC2 model, KLS‑3021 provided superior tumor control and survival benefit compared to cisplatin. Collectively, these findings indicate that KLS-3021 elicits potent antitumor immunity and remodels the TME regardless of PD-L1 levels, exceeding current conventional standards. Given its capacity for local tumor debulking and immune activation, KLS-3021 implicates strong translational potential as a next-generation therapeutic, including neoadjuvant applications for HNSCC.