Deciphering Mechanisms of UVR-Induced Tumoral Immune Checkpoint Regulation against Melanoma

The incidence and mortality rates due to malignant melanoma have risen globally over the last 50 years, and among various etiologic factors, exposure to solar ultraviolet radiation (UVR) is considered the paramount risk factor (1). Although the beneficial effects of UVR in modulating host immune responses to vitamin D synthesis have long been studied (2), the detailed mechanisms of UVR-induced effects such as immune tolerance to tumor antigens or inhibition of antitumor immunity have yet to be fully elucidated. The earlier discoveries including the demonstration that consecutive UVR modulates the immunologic and antigenic effects against skin carcinogenesis (3) has led to the exploration of immune-mediated mechanisms in UVR-induced melanoma initiation and progression. To that end, multiple studies have shown the roles of immunophenotypes including effector Th1/Th2/Th9 and CD8⁺ T cells as well as suppressive regulatory T cells (Tregs) in either mediating or inhibiting the antitumor immune responses in B16F10 or K1735 murine melanoma models with or without UVR, particularly UVB (4, 5). However, whether or not UVB-induced inhibition of melanoma antitumor immune responses is mediated via the modulation of immune checkpoint proteins remained a clinically relevant concern to be addressed.

Of significance, immune checkpoint inhibition has become a major focus in immune-based therapies for the treatment of human malignancies including melanoma. Various immunophenotypes including Tregs express negative regulatory immune checkpoint receptors such as cytotoxic T-lymphocyte associated protein 4 (CTLA-4) and programmed cell death protein 1 (PD-1). These checkpoints act to dampen or modulate T-cell activation/function, leading to diminished antitumor immune responses via mechanisms including inhibition of cytokine production such as IFNγ, TNFα, and IL2, and CD8⁺ T-cell–mediated cytotoxicity. Notably, several cancer types, including melanoma, express increased levels of PD-1 ligands, namely PD-L1 and PD-L2, respectively. The inhibition of CTLA-4 and PD-1 immune checkpoints in clinical settings have been shown to be associated with overall durable antitumor immunity and/or improved survival responses, yet in a subset of patients with melanoma (6). In addition, as a result of primary or acquired resistance following anti–CTLA-4 and anti–PD-1 therapy as well as identification of genomic landscapes and mutational signatures driving aberrant signal transduction, several therapeutic combinations were explored. Importantly, the combination of anti–PD-1 with BRAF and/or MEK inhibitors or anti–CTLA-4 in patients with melanoma have resulted in improved response rates, yet treatment-associated systemic toxicities including immune-related adverse events were documented (6). Nevertheless, given the advantages of superior efficacies of such combinational approaches, further exploratory studies including evaluation of dose-dependent effects with longer-term follow-up, complementary mechanisms of immune activation, and potential predictive indicators of therapeutic responses would help in maximizing the benefits and minimizing the toxicities. Moreover, such combinational approaches could also benefit from complementing other modifying factors, which play crucial roles in melanoma immune evasion and progression such as UVR.

As exposure to UVR causes DNA damage in the majority of skin-resident cells including melanocytes, The Cancer Genome Atlas melanoma study has identified the incidence of UVR-associated mutational signature in 76% of primary tumors and 84% of metastatic samples in patients with melanoma. In addition to the common driver mutations, the somatic mutation burden has been documented to be comparatively high in melanomas than other human malignancies (7). UVR significantly contributes to the accumulation of somatic mutations, which act as neoantigens to elicit antitumor immune responses. Understanding mechanistic insights into immune checkpoint regulation in response to UVR-induced mutational signature could offer new strategies to overcome immune tolerance or to circumvent melanoma immune evasion. Importantly, such strategies can be useful in designing immune checkpoint-based novel combination approaches for melanoma treatment.

The high somatic mutational burden in melanoma has been shown to be associated with long-term clinical benefits including overall improved survival responses to anti–CTLA-4 therapy (8). Wang and colleagues evaluated the significance of UVR-induced somatic mutation burden on antitumor immune responses against melanoma (9). Subsequently, authors generated a mutagenized mouse melanoma model, YUMMER 1.7 [i.e., Yale University Mouse Melanoma (YUMM) Exposed to Radiation], from the parental YUMM 1.7 melanoma cells that harbor three driver mutations (Braf^V600E, Pten^−/−, and Cdkn2a^−/−) by UVB irradiation and expansion of a single cell–derived clone. Implantation of YUMMER 1.7 cells harboring UVB-introduced high somatic mutation burden at low cell numbers exhibited tumor regression compared with YUMM 1.7 cells in immunocompetent C57BL/6J hosts. These data indicated that YUMMER 1.7 cells enhance immunogenicity via eliciting T-cell–mediated functional antitumor adaptive immune responses. This tumor regression was abrogated in immunodeficient Rag1^−/− mice and by the depletion of CD4⁺ and CD8⁺ T cells or inoculation of higher numbers of YUMMER 1.7 cells in C57BL/6J mice. The increased growth of YUMMER 1.7 tumors was attenuated by anti–CTLA-4 and anti–PD-1 immunotherapy, and their combination resulted in enhanced antitumoral and survival effects (9). Given these findings, further delineation of signaling pathways involved in regulating tumoral immune checkpoints in response to UVB may offer alternative therapeutic strategies for melanoma intervention. Such investigations are also important, given the delineation of transcription factors and upstream mediators such as p53, PTEN, and NF-κB, which regulate PD-L1 expression, and thus, add other pieces to the puzzle of this very ongoing challenge.

Multiple studies have implicated UVB in the induction of oxidative stress via the production of reactive oxygen species (ROS), and cancer cells require high ROS demand compared with normal cells to support their accelerated metabolism and high proliferation rate. A recent study examined the role of nuclear factor E2–related transcription factor (NRF2), which regulates the expression of antioxidant proteins, in melanoma intervention (10). These authors demonstrated that UVB upregulates PD-L1 and NRF2 expression in human primary keratinocytes (HPK) and human primary melanocytes (HPM), and the analysis of human melanoma tissue array also revealed a significant correlation between PD-L1 and NRF2 expression. The question of whether UVB-induced PD-L1 expression or transcription is NRF2-dependent was supported by the findings including that UVB upregulates PD-L1 expression on the dorsal skin including melanocytes of wild-type, but not from Nrf2−/− mice, and NRF2 overexpression induced and its silencing attenuated UVB-mediated PD-L1 expression in HPK and HPM. Importantly, NRF2 silencing combined with anti–PD-1 therapy resulted in greater suppression of melanoma tumor growth compared with single treatments, respectively (10). These combination treatment–induced effects were mediated via mechanisms including increased infiltration of CD8⁺ and CD4⁺ T cells as well as reduced Tregs and myeloid-derived suppressor cells compared with NRF2-silencing or anti–PD-1 therapy alone. As the efficacy of silencing NRF2 and anti–PD-1 therapy was relatively similar in inducing antitumor immune responses, these findings suggested the potential of targeting NRF2 as an alternative strategy of PD-1/PD-L1 inhibition for melanoma treatment (10). However, the question of whether UVB exposure can impede the antitumor immunity mediated via NRF2-silencing and/or anti–PD-1 therapy as well as the mechanisms involved in these events remained to be elucidated.

Notably, one of the major limitations of UVB-induced effects on tumoral PD-L1 expression and anti–PD-1–mediated antitumor immune responses was thoroughly addressed by Wang and colleagues (11). The authors demonstrated that UVR-induced secretion of damage-associated molecular patterns (DAMP) molecule, HMGB1, by melanocytes and keratinocytes activated the receptor for advanced glycation endproducts (RAGE). These changes in turn activated NF-κB and IFN regulatory factor 3 (IRF3) pathways in melanocytes to promote PD-L1 transcription (11). Although NRF2 depletion did not affect UVB-induced PD-L1 expression in melanoma cells, NF-κB and IRF3 were found to be indispensable for PD-L1 regulation by UVB in primary melanocytes. These data indicated a cell type–specific mechanism and suggested that within primary melanocytes, UVB may mediate PD-L1 transcription with NRF2 alternatively and/or collaboratively. The findings that deletion of HMGB1/RAGE abolished UVB-induced phosphorylation of TBK1, NF-κBp65, and IRF3 as well as PD-L1 upregulation, validated the role of HMGB1/RAGE in activating the downstream TBK1/NF-κB/IRF3 pathway and PD-L1 induction by UVB. Further studies confirmed that the upstream kinase, TBK1, activated both NF-κB and IRF3, and IRF3 formed a transcriptional complex with NF-κBp65 within PD-L1 promoter to coordinately promote PD-L1 transcription, and that inhibition of the NF-κB upstream kinases, IKKβ or TBK1, attenuated UVR-induced increased PD-L1 transcription in melanoma cells (11). Importantly, the inhibition of PD-1/PD-L1 signaling attenuated the UVB-induced increased growth of melanoma tumors via CD8⁺ T-cell–mediated enhanced antitumor immunity. These data indicated that blocking immune checkpoint may also mitigate UVR-induced immune suppression, which resulted in immune evasion of premalignant melanocytes and melanoma cells, leading to melanoma initiation and progression.

Several clinical studies have documented improved checkpoint immunotherapy responses with tumor profiles including PD-L1 positivity, high mutational burdens, tumor-infiltrating lymphocytes such as increased CD8⁺ T cells and immune gene signatures against malignancies including melanoma. However, additional evidence is needed to utilize the relevance of such tumor profiles as predictive indicators in selecting patients with cancer for immune checkpoint–based combination treatments and/or correlating their levels with treatment outcomes. Importantly, as novel therapeutic strategies are being actively explored, the validation of cellular pathways such as HMGB1/TBK1/IRF3/NF-κB in clinical settings will allow us to determine whether the benefits of alleviating UVR-induced melanoma immune evasion by immune checkpoint inhibition can be replicated pharmacologically as alternative combinational approaches for melanoma treatment.

No potential conflicts of interest were disclosed.

The grant support from NIH K22 (ES023850) is greatly acknowledged. The author sincerely thanks Dr. Anita Thyagarajan for providing helpful comments.