Overcoming Steroid Resistance in T Cell Acute Lymphoblastic Leukemia

In a Perspective, Pieter Van Vlierberghe and Steven Goossens discuss Meijerink and colleagues' findings on steroid resistance in pediatric T cell acute lymphoblastic leukemia.


Genetics Meet Drug Sensitivity in Human T-ALL
In this issue of PLOS Medicine, Jules Meijerink and colleagues [4] integrated next-generation sequencing data and genome-wide copy number profiles with ex vivo drug sensitivity and clinical outcome in a large panel of primary T-ALL specimens obtained at diagnosis. Interestingly, the authors found that mutations in the interleukin 7 receptor (IL7R) signaling pathway, including JAK1 and KRAS mutations, were significantly associated with prednisolone resistance and reduced survival in an initial discovery cohort of 69 primary T-ALL samples. Of note, these results are in line with the poor prognostic implications of JAK1 [5] and N/K-RAS [6] mutations that have previously been reported in adult T-ALL. In addition, previous studies have also shown an association between RAS mutations and steroid resistance in the context of precursor B-lineage ALL (B cell acute lymphoblastic leukemia [B-ALL]) [7,8]. Given that various mutations affecting the IL7R signaling cascade have also been reported in precursor B-ALL, it is tempting to speculate that the relation between IL7R signaling and steroid resistance could be a general feature of both T-and B-lineage ALL.
Next, Meijerink and colleagues extended their analyses to 146 leukemia samples, for which they determined the mutational status of a selected panel of genes required for IL7R-RAS-MAPK-AKT signaling (IL7Ra, JAK1, JAK3, NF1, NRAS, KRAS, and AKT) by PCR-based Sanger sequencing. This analysis revealed that T-ALL specimens mutated for any of these genes were significantly associated with steroid resistance and poor clinical outcome. Interestingly, T-ALLs that harbor mutations in IL7R-RAS-MAPK-AKT signaling were associated with specific molecular genetic subtypes, including immature, TLX1, TLX3-rearranged, or HOXA-activated leukemias. However, it should be noted that Meijerink and colleagues are the first to establish that these specific genetic alterations are truly associated with steroid resistance in primary T-ALL specimens. In steroid-sensitive T-ALL cells, steroid-mediated activation, dimerization, and subsequent nuclear translocation of the steroid receptor NR3C1 results in transactivation of the pro-apoptotic BIM and PUMA, which will activate the apoptosome complex after cytochrome C release from the mitochondria. The anti-apoptotic protein MCL1 in steroid-sensitive T-ALL cells is actively targeted for proteasomal degradation via GSK3β. βP, phosphorylation; Ub, ubiquination; GRE, glucocorticoid response element. doi:10.1371/journal.pmed.1002208.g001

From Association to Functional Implications: Genetic Drivers of Steroid Resistance
To functionally validate the putative contribution of each of these mutations to steroid resistance in T-ALL, Meijerink and colleagues generated a panel of isogenic tumor lines with inducible expression of IL7R-RAS-MAPK-AKT pathway mutations. Using these elegant in vitro model systems, the authors confirmed that IL7R, JAK1, and NRAS mutations, as well as overexpression of wild-type NRAS or AKT, could drive steroid resistance in human T-ALL cell lines (Fig 2). In contrast, no effect was observed for mutant JAK3 or expression of wildtype JAK1, JAK3, or IL7R.
Notably, no differences in NR3C1 nuclear translocation or target gene transactivation upon glucocorticoid exposure were observed between steroid-resistant and steroid-sensitive isogenic tumor cell lines. These data suggest that, at least in these in vitro model systems, the molecular mechanisms controlling steroid resistance are downstream or independent of the NR3C1 transactivation response. Strikingly, these results are in sharp contrast with other work in which glucocorticoid resistance in ALL was associated with the inability to induce the proapoptotic NR3C1 target gene BCL2L11 upon dexamethasone treatment of patient-derived xenografts [9,10]. These contradictory results may be related to a differential epigenetic state of the BCL2L11 (BIM) locus in the T-ALL cell lines versus patient-derived xenografts. In addition, these results suggest that steroid resistance can be gained via various means at different essential nodes in these converging signaling pathways.
Instead of a dysfunctional glucocorticoid-induced transcriptional response, Meijerink and colleagues identified aberrant activation of specific signaling pathways downstream of mutant IL7R, JAK1 or NRAS, or wild-type AKT as major determinants of steroid resistance in the isogenic tumor lines (Fig 2). Indeed, expression of IL7R, JAK1, or NRAS mutations provoked a strong activation of MEK-ERK and AKT signaling with concomitant phosphorylation of p70-S6K and an increased expression of anti-apoptotic BCL-XL. In addition, activated MEK-ERK and AKT signaling also triggered a robust inactivation of GSK3β, which would lead to stabilization of anti-apoptotic MCL1 (Fig 2). Interestingly, the association between IL7R-JAK signaling and steroid resistance described in this study is in line with the interleukin 7-induced anti-apoptotic rescue of primary T-ALLs against dexamethasone, as reported more than a decade ago [11]. However, in these studies, the inhibition of apoptosis by interleukin 7 was shown to correlate with up-regulation of BCL2 [12]. In addition, recurrent mutations in the IL7R and RAS signaling pathways mainly occur in early immature T-ALLs, which are also more often characterized by increased concentrations of BCL2 [13,14].

Small Molecules with a Future in Steroid-Resistant T-ALL
Meijerink and colleagues' study has unraveled specific pathways that influence glucocorticoid sensitivity in T-ALL by affecting the balance between pro-and anti-apoptotic factors. Therefore, specific targeting of these signaling cascades could provide a unique opportunity to overcome glucocorticoid resistance in human T-ALL. Indeed, the JAK1 inhibitor ruxolitinib, the MEK inhibitor CI1040, and the AKT inhibitor MK2206 were able to reverse steroid resistance in specific isogenic tumor lines (Fig 3). Interestingly, the authors also showed that the GSK3β inhibitor IX provoked strong steroid resistance in sensitive T-ALL tumor lines, further confirming the pivotal role of this molecule in the regulation of glucocorticoid resistance. Moreover, given the central role for the balance between pro-and anti-apoptotic factors in this process, one could hypothesize that specific targeting of the apoptotic machinery using BCL2 [13] or MCL1 [15] inhibitors could also serve as a useful strategy to alter steroid resistance in the context of malignant T cell transformation.
In a more clinically relevant approach, the authors subsequently evaluated the putative drug synergism between prednisolone and seven small molecule inhibitors targeting JAK1, MEK1/ 2, or PI3K/AKT/mTOR in a panel of 11 genetically well-characterized primary T-ALL samples at diagnosis. Notably, small molecule inhibitors targeting MEK-ERK or PI3K/AKT/mTOR signaling enhanced the steroid response in most primary T-ALL samples (Fig 3). In line with this notion, recent work showed that the dual PI3K/AKT/mTOR inhibitor BEZ235 enhanced dexamethasone sensitivity in vivo using a patient-derived T-ALL xenograft [16]. Surprisingly, and in contrast with the cell line data, ruxolitinib had a synergistic effect on prednisolone treatment in only 1 out of 11 primary T-ALL samples. However, this might be related to the lack of proliferative capacity of T-ALL cells in this system, given that previous work convincingly showed preclinical in vivo efficacy of ruxolitinib in patient-derived T-ALL xenografts [14]. Nevertheless, these results could suggest that ruxolitinib might not be suitable for therapeutic targeting of dormant leukemic stem cells.
In conclusion, Meijerink and colleagues' study nicely illustrates how somatic alterations implicated in T-ALL disease biology influence the differential steroid response observed in the clinic. These novel insights strongly suggest that inhibition of MEK-ERK or PI3K/AKT/mTOR signaling could enhance steroid sensitivity in T-ALL and potentially improve patient treatment outcome, a notion that warrants further investigation in future prospective clinical trials.