Off-target effects of tyrosine kinase inhibitors: Beauty or the Beast?

Targeted therapy celebrates the concept of ‘designer therapy,’ i.e. designing therapies based on pre-identified targets. Tyrosine kinase inhibitors (TKIs), being one of the most successful stories of targeted therapy, have revolutionized the prognosis and management of chronic myeloid leukemia (CML) [1]. Its target, the fusion protein BCR–ABL, beautifully illustrates the three criteria of identifying the target molecule: the molecule must be present in most of the cancer cells; the molecule should not present in most of the normal cells; the molecule should be essential for the persistence and proliferation of cancer cells [2]. Various TKIs have been developed to target the fusion BCR–ABL protein, such as those targeting the inactive conformation (e.g. imatinib, nilotinib) and both the active and inactive conformations (e.g. dasatinib) [3]. However, even with this classic target, modern TKIs remain entangled by off-target effects which result in normal tissue damage [3]. Large granular lymphocytes (LGLs) consist of natural killer (NK) cells and cytotoxic T lymphocytes (CTLs) [4,5]. They are essential for innate and acquired immunity against viral infection and neoplasm. However, they are also potential culprits of autoimmunity. The association between LGL expansion and autoimmune diseases is particularly prominent in LGL leukemia, which features a clonal expansion of LGLs [6]. Interestingly, as a manifestation of the off-target effect of TKIs, some TKI treatments are associated with a mono/oligoclonal LGL expansion [7,8].

In this issue of Leukemia and Lymphoma, Powers and his colleagues [9] characterize the molecular signatures of the expanded LGL population in patients with CML treated with TKIs. Moreover, they analyzed the potential mechanisms of this expansion and its association with side-effects of TKI treatment. Compared to the normal composition of memory repertoire, peripheral T lymphocytes in patients treated with different TKIs all showed a decrease in naive or central memory subsets and an increase in effector/terminal memory subsets, with the trend more prominent in the CD4+ than in the CD8+ population. This is in line with an abnormal T-cell response due to chronic and/or suboptimal stimulation [10]. Consistent with T-cell activation, a clonal T-cell expansion was identified in most of the TKI-treated patients regardless of the TKI used (imatinib, nilotinib, or dasatinib). Interestingly, this universal clonal expansion did not result in universal expansion of LGLs. A significant LGL population was only identified in dasatinib-treated patients. There were correlations between NK-cell or CD8+ T-cell expansion and the presence of adverse reactions, indicating the potential functional importance of LGLs in dasatinib-treated patients. Analyzing peripheral blood samples following different TKI treatments highlighted the molecular ‘royal flush’ that is essential for LGL expansion: interleukin-15 (IL-15) transpresentation, increased levels of plasma platelet derived growth factor BB (PDGF-BB), nuclear factor κB (NF-κB), and T-beta activation and lack of T-regulatory cell (T_reg) supervision.

The off-target effects of targeted therapy, although not desirable, can sometimes lead to discoveries of novel signaling details otherwise hidden. The work of Powers et al. [9] is important not only for understanding the causes of adverse effects related to TKI treatment but also potentially for improving current T-cell immunotherapy. On the one hand, the correlation between LGL expansion and adverse effects of TKI treatment raised the possibility of using the size of the LGL population or factors essential for LGL expansion as markers to guide therapy adjustment. It is even conceivable that immune modulatory strategies used to treat LGL leukemia may be useful for control of serious side-effects following TKI treatment. On the other hand, the effect of TKIs on T-cell response may aid the generation and expansion of long-term competent CTLs necessary for immunotherapy. Consistent with previous studies [11], Powers et al. showed that IL-15 transpresentation and plasma PDGF-BB elevation occur together with CTL expansion [9]. IL-15 has been shown to be important for T-cell activation and generation of CD8+ memory cells [12,13]. PDGF is also a master switch controlling leukemic LGL survival [11,14]. This study reinforces the concept that both pro- and anti-activation signals are needed for CTL expansion and persistence [11]. Different from previous studies, Powers et al. showed that CTL expansion can be achieved by signaling permutations resulting from a single agent, TKI [9]. This is not entirely surprising, considering the presence of neoplastic myeloid cells as a chronic immune stimulus (the pro-activation signal) and inhibition of tyrosine kinase which is essential for normal T-cell activation [15] by TKI (the anti-activation signal). The immune-boosting effects and promising therapeutic outcomes of TKIs have been reported when used together with allogeneic transplant or vaccination in patients with CML [16,17]. Understanding the molecular mechanisms of CTL survival will help us harness CTLs for cancer and virus immunotherapy not limited to CML treatment.

Further studies are required to fully define the relationships among treatment with TKIs, T-cell clonal expansion, LGL clonal expansion, and adverse effects. In this study, a significant LGL population was only identified in dasatinib-treated patients. However, autoimmune-like cytotoxicity was detected in samples from both dasatinib- and nilotinib-treated patients, with samples from imatinib-treated patients not examined [9]. This necessitates additional studies regarding how LGL might cause dasatinib-specific side-effects. Powers et al. explained the immune modulatory effect of TKIs from the perspective of altered T_reg level [9]. However, without knowing the basal T_reg level in untreated patients with CML, this causal relationship is difficult to establish. The relationship between NK-cell or T-cell expansion and adverse effects of dasatinib treatment will require verification with a larger patient pool and a detailed longitudinal look into the LGL population. It would also be worthwhile to conduct similar longitudinal studies in patients treated with other TKIs, which may shed light on the molecular mechanisms of adverse effects of TKI treatment in general.

Potential conflict of interest: Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.