Modulation of cytokine-induced senescence in melanoma cells by JAK inhibitor: A possible way to improve therapeutic efficacy of immune checkpoint inhibitor

Apart from the direct blocking of immunosuppressive checkpoint signaling, a prominent feature of immune checkpoint inhibitors (ICIs) is the enhancement of cytokine production by immune cells to facilitate anti-cancer immunity in the tumor microenvironment. Prolonged cytokine stimulation may also lead to immune suppression. On the other hand, it is known these cytokines can act on cancer cells to induce cellular senescence (cytokine-induced senescence, CIS). It has also been recognized that cellular senescence is intertwined with stemness or cell plasticity, and these phenotypic changes can be adopted by cancer cells to evade the anticancer effect of therapeutics agents including ICIs, resulting in resistance to treatment and or relapse post treatment. In this study, we attempt to investigate whether senomorphics such as JAK inhibitors can ameliorate CIS in melanoma cells and whether this reduction is accompanied by changes in the stemness and senescence-associated secretory phenotype (SASP). Melanoma cell line SK-MEL-28 was treated with INF-gamma and TNF-alpha to establish CIS as indicated by SA-beta-gal staining and detected with flow cytometry. Ruxolitinib (Ruxo) as a JAK1/2 inhibitor was then used to treat the cells, and at the end of the treatment the expression of stemness markers, SASP components, and immune checkpoint ligands was measured with qPCR. We observed that Ruxo was able to partially reduce the SA-beta-gal fluorescence in SK-MEL-28 (express PD-L1) at 10 nM and to the levels comparable of non-senescent control with non-senescent cell morphology at 10 μM. Furthermore, the expression of SASP component IL-1beta and stemness markers CD271 was upregulated (about 4-fold) in the CIS state. Subsequent Ruxo treatment (10 μM) could partially attenuate the upregulation of IL-1beta (by about 23%) and CD271 (by about 40%). However, Ruxo was unable to downregulate the increased expression of the stemness marker CD133 (1.77-fold from control); on the contrary, further upregulation (by about 31%) was noted. In addition, the expression of immune checkpoint ligands PD-L1 and PD-L2 showed moderate increase (by about 37% and 61%, respectively) in CIS and was not largely altered by Ruxo (increased by about 3% and 13%, respectively) while the increase of TIGIT ligand CD155 in CIS (by about 43%) could virtually be reverted by Ruxo. In summary, Ruxo could revert the senescent morphology and elevated SA-beta-gal activity in the CIS of SK-MEL-28. Moreover, it could attenuate the expression of some stemness markers and SASP component elevated during CIS in melanoma cells. Further expansion of this study such as inclusion of more molecular markers and mechanism(s) is underway. Clinical study to include JAK inhibitor with ICI will also be initiated. This work is supported by the fund from Woman Cancer Association. Sylvester Comprehensive Cancer Ctr, U. of Miami and Miami VA Med. Ctr.