Melanoma, which arises from melanocytes, is one of the most aggressive cancers owing to its high metastatic potential and the associated mortality rates [1]. Although surgical resection remains the primary treatment for localized melanoma, effective therapies for patients with unresectable or advanced stages of the disease are urgently needed [[2], [3], [4]]. Immune checkpoint blockade (ICB) antibodies targeting PD-1/L1 and CTLA-4 have demonstrated success in clinical cancer treatment, with over 50 % of patients responding to combination therapy using nivolumab (anti-PD-1) and ipilimumab (anti-CTLA-4) [5,6]. However, approximately 50 % of patients are nonresponders, and many experience disease progression after therapy. Immunotherapy outcomes depend on the proper activation of multiple steps in the cancer-immunity cycle, including the release of tumor-associated antigens (TAAs), the capture and cross-presentation of TAAs by antigen-presenting cells (APCs), T cell priming and infiltration into tumors, and the recognition and killing of cancer cells [7]. Nonetheless, this cancer-immunity cycle often does not function well in cancer patients with an immunosuppressive tumor microenvironment (TME) [8]. Consequently, patients with immune-desert TME, characterized by inadequate tumor-infiltrating T lymphocytes (TILs), abundant immune suppressors, and low mutational burden, respond poorly to T cell-based therapy. Addressing the complex mechanisms of ICB resistance and immune evasion in patients with melanoma requires the development of novel therapeutic strategies that are both safe and effective, thereby enhancing tumor immunogenicity by potentiating the cancer-immunity cycle [9,10].

Mitochondria provide chemical energy for various cellular reactions [11]. Beyond ATP production, they are also crucial for tumorigenesis and immune responses [[12], [13], [14]]. Previous studies have shown that mitochondrial DNA and dynamics play a significant role in activating tumor immunogenicity [[15], [16], [17]]. Targeting cellular mitochondria can induce dysfunction and release damage-associated molecular patterns (DAMPs), thereby enhancing immunogenic cell death (ICD) over conventional photodynamic therapy (PDT) [[18], [19], [20]]. Hence, as a potential targetable vulnerability, targeting mitochondrial compartments with rationally designed therapies has emerged as a viable approach to rewire the cancer-immunity cycle and prime antitumor immunity. Most methods for the selective mitochondrial localization of therapeutic agents rely on the covalent conjugation of mitochondria-specific motifs such as the cationic triphenylphosphonium (TPP) moiety [21]. However, such highly cationic modification may result in the rapid clearance of the ligates from the body following their intravenous administration, leading to suboptimal treatment outcomes [22]. Hence, new selective drug strategies targeting mitochondria without compromising their properties of prolonged circulation and efficient delivery could elicit tumor cell immunogenicity and are in urgent need.

PDT has emerged as a minimally invasive modality for treating dermatological malignancies [23,24]. Upon near-infrared (NIR) laser irradiation, photosensitizers convert molecular oxygen into reactive oxygen species (ROS), causing potent cancer cell damage [25]. PDT has several advantages, such as reduced systemic toxicity, improved spatiotemporal control and enhanced therapeutic efficacy. Moreover, PDT holds the potential to induce ICD, which augments the antigenicity of tumor cells by locally releasing TAAs, DAMPs, and proinflammatory cytokines [26,27]. These events could promote the maturation of dendritic cells (DCs) and antigen presentation, evoking a specific antitumor immunity [28]. Nevertheless, conventional photosensitizers often lack specificity for cancer cells and cellular organelles such as mitochondria, which dampens their potential as an ICD inducer [[29], [30], [31]].

In our earlier study, a new concept of pseudo-stealthy mitochondria-targeting was proposed, and this approach was validated for the specific delivery of a taxane agent (e.g., cabazitaxel) to the mitochondria [32]. The low-density grafting of TPP cations was efficiently internalized by tumor cells for mitochondrial trafficking. Moreover, prolonged circulation, similar to that of conventional stealthy polyethylene glycol (PEG) cloaking, was maintained. Motivated by this design rationale, this research aimed to develop a mitochondria-targetable photosensitizer-formulated polymeric micelle (Mito-PM) with pseudo-stealthy characteristics for mitochondrial rewiring, ICD induction, and melanoma immunotherapy (Fig. 1). The shortened in vivo duration of the photosensitizer pyropheophorbide-a (PPa) was overcome by the rational engineering of a polymer-PPa conjugate (polyPPa), which was developed into a micelle-based vehicle with low-density surface conjugation of cationic TPP moieties. The findings indicated that the stable nanosystem was specifically localized in the mitochondria after cellular uptake and induced a potent ICD following NIR laser irradiation. Furthermore, intravenous administration of the stealthy Mito-PM in an animal model triggered the cancer-immunity cycle and suppressed tumor growth. The antitumor efficacy was further potentiated by combining it with ICB therapy. While positively charged surfaces upon TPP decoration render nanotherapeutics with efficient cellular uptake and subsequent mitochondrial trafficking, long-term blood circulation following intravenous administration has been impeded, which leads to insufficient drug delivery to diseased sites (e.g., tumors). This contradiction is expected to be addressed by our rationally engineered pseudo-stealthy nanosystem.