CML stem cells: evasion for better invasion

In this issue of Blood, Tarafdar et al demonstrate the downregulation of major histocompatibility complex (MHC)-II molecules (and their main regulator class II transactivator [CIITA]) in chronic myeloid leukemia (CML) stem cells, allowing them to evade the immune surveillance exerted by the innate immune system that can be reverted or enhanced by JAK2 inhibitors and interferon γ (IFN-γ).¹ 

For more than 2 decades, the immunogenicity of CML cells has been demonstrated and therapeutically successfully exploited in the allogeneic setting,^2,3  leading to indefinite complete molecular remissions in many CML patients, especially those in chronic phase. In addition, the presence of host immune effectors directed against CML aberrantly expressed proteins⁴  or BCR-ABL peptides⁵  able to induce an effective lysis of malignant cells has been also demonstrated and exploited in the clinical arena in few patients with real, but less success.⁶  These issues are still currently under investigation.

In this original work, using a microarray approach, Tarafdar et al carefully examined the expression of MHC-II molecules in progenitor cells of CML patients (CD34⁺38⁺, CD34⁺38⁻, CD34⁺38⁻90⁺), comparing them with normal bone marrow counterparts. They explored different hypotheses that may explain why these cells evade the host immune surveillance and persist despite life-long tyrosine kinase inhibitor (TKI) treatment. The expression of MHC-II genes and molecules are specifically reduced in CML stem cells, as is the major regulator of these molecules, CIITA. The addition in culture media of IFN-γ increased the expression of MHC-II molecules, but to a lesser degree than it would do for normal counterparts. The exposure of CML stem cells to 3 different TKIs in vitro for 2 to 7 days did not result in substantial modification of the expression of MHC-II and CIITA, suggesting that this phenomenon is BCR-ABL–independent. The decreased expression does not seem to be mediated via epigenetic mechanisms. Because interleukin-4, a natural antagonist of MHC-II and CIITA expression, signals through the JAK/STAT pathway (as do many other cytokines), the authors demonstrate that the incubation of CML stem cells with the JAK1/2-specific inhibitor Ruxolitinib restored the expression levels of CIITA and MHC-II as a result of the inhibition of JAK-downstream mediators. Finally, Tarafdar et al show that IFN-γ and ruxolitinib increase the proliferation of responder CD4⁺CD69⁺ allogeneic T cells in mixed lymphocyte reactions, which could be abolished by the addition of antibodies against MHC-II molecules.

The authors thus provide novel information describing an additional BCR-ABL–independent mechanism that might contribute to the resistance of CML stem cells that persist despite TKI therapy by down-regulating MHC-II molecules, which can be reverted in vitro by ruxolitinib and IFN-γ. These data are in line with previous in vitro and in vivo observations on the efficacy of type I interferons (ie, IFN-α) in CML patients with residual disease on TKI, which may work by restoring MHC-I and MHC-II expression in CML stem cells and allow their recognition by the host immune effectors.⁷  This study also helps explain data from previous or ongoing trials combining TKIs with (pegylated forms of) IFN-α since diagnosis, showing unexpectedly high molecular response rates.^8-10  Responses of 4 ongoing clinical trials in different clinical situations in CML, combining TKI and ruxolitinib as discussed in Tarafdar et al, also support these in vitro observations. That the downregulation of MHC-II and CIITA is intrinsic to CML stem cells or related to some cytokine deregulations within the CML niche, or both, requires further exploration.