Transformed cancer cells can be detected and eliminated by the immune system. Some tumours can evade immune-surveillance by a number of mechanisms resulting in “immune escape”. Loss of antigenicity, loss of immunogenicity and development of a tumour suppressive microenvironment have been investigated as possible mechanisms of immune escape.
In this study, Prestipino et al., (Prestipino et al., Sci. Transl. Med. 10, eaam7729 (2018)) postulate that immune escape and oncogenic transformation occur early in the development of tumours and suggest that cooperation between the two can occur resulting in immune escape. They investigated a potential role for programmed death ligand‑1 in myeloproliferative neoplasms (MPNs) harbouring the JAK2V617F oncogenic mutation.
In summary:
Increased expression of PD‑L1 in multiple cell types in a knock-in JAK2V617F mouse model. Increased PD‑L1 expression was also observed in bone marrow cells transfected with a JAK2V617F expressing plasmid.
Murine myeloid 3D cells transfected with a JAK2V617F vector also showed increased levels of PD‑L1 and treatment of these cells with Ruxolitinib and SD‑1029, both JAK2 inhibitors, reversed this expression. These effects were also observed in Ba/F3 cells.
There was a concomitant increase in phosphorylated STAT1, STAT3 and STAT5 in 3D and Ba/F3 cells transfected with the JAK2V617F plasmid compared to WT. Similar experiments utilising gain of function (GoF) and loss of function (LoF) mutants for the STATs demonstrated that STATS 3 and 5 but not STAT1 modulated the expression of PD‑L1 in JAK2V617F transfected cells.
Similar experiments in K562 cells confirmed the link between JAK2V617F and PS‑L1 in human cells.
In the peripheral blood of MPN patients with the JAK2V617F mutation, T‑cells, monocytes myeloid derived suppressor cells (MDSCs) and platelets all expressed higher levels of PD‑L1 compared to healthy controls.
More advanced grades of MPN had higher levels of PD-L1 expression and treatment with Ruxlitinib decreased PD‑L1 levels in platelets and monocytes of MPN patients.
Higher levels of PD‑L1 were observed in the spleen and bone marrow of mice transfected JAK2V617F compared to WT‑JAK2 when monitored by PET‑CT using a 64Cu‑labelled antibody.
Higher Allele burden for JAK2V617F was observed in monocytes and platelets of MPN patients compared with T- and B‑cells consistent with higher levels of PD‑L1 in these cells. Higher levels of Stat3 were also observed in monocytes and platelets from peripheral blood.
Improved survival after treatment with anti‑PD‑1 or anti‑PD‑L1 antibodies was demonstrated in mice engrafted with human PBMCs from MPN patients. Consistent with this allele burden for JAK2V617F, human platelets and CD45+ cells were lower in mice (Peripheral blood and bone marrow) treated with anti‑PD‑1 antibodies.
Gene expression analysis revealed changes in amino acid metabolism and cell cycle genes in T‑cells exposed to JAK2V617F myeloid 32D cells. A lower oxygen consumption rate and reduced levels of cdks 6 and 8 confirmed these effects. Cyclin G2 and cyclin‑dependent kinase inhibitor 2D, which reduce cell cycle progression, were upregulated.
In this study, the authors suggest a connection between JAK2 and PD‑L1 expression via phosphorylation of Stat3 or Stat5. Oncogenic JAK2 mutants, such as V617F, may increase PD-L1 expression resulting in decreases amino acid metabolism and cell cycle progression in T‑cells. The consequent reduction in T‑cell activation may be sufficient to allow immune escape in MPN and that immunotherapies targeting PD‑1, PD‑L1 may be useful in this disease setting.