Mutations in the interleukin-7 receptor (IL7R) or the Janus kinase 3 (JAK3) kinase occur frequently in T-cell acute lymphoblastic leukemia (T-ALL) and both are able to drive cellular transformation and the development of T-ALL in mouse models. However, the signal transduction pathways downstream of JAK3 mutations remain poorly characterized. Here we describe the phosphoproteome downstream of the JAK3(L857Q)/(M511I) activating mutations in transformed Ba/F3 lymphocyte cells. Signaling pathways regulated by JAK3 mutants were assessed following acute inhibition of JAK1/JAK3 using the JAK kinase inhibitors ruxolitinib or tofacitinib. Comprehensive network interrogation using the phosphoproteomic signatures identified significant changes in pathways regulating cell cycle, translation initiation, mitogen-activated protein kinase and phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K)/AKT signaling, RNA metabolism, as well as epigenetic and apoptotic processes. Key regulatory proteins within pathways that showed altered phosphorylation following JAK inhibition were targeted using selumetinib and trametinib (MEK), buparlisib (PI3K) and ABT-199 (BCL2), and found to be synergistic in combination with JAK kinase inhibitors in primary T-ALL samples harboring JAK3 mutations. These data provide the first detailed molecular characterization of the downstream signaling pathways regulated by JAK3 mutations and provide further understanding into the oncogenic processes regulated by constitutive kinase activation aiding in the development of improved combinatorial treatment regimens.

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive leukemia that is most frequent in children and is characterized by the presence of few chromosomal rearrangements and 10 to 20 somatic mutations in protein-coding regions at diagnosis. The majority of T-ALL cases harbor activating mutations in NOTCH1 together with mutations in genes implicated in kinase signaling, transcriptional regulation or protein translation. To obtain more insight in the level of clonal heterogeneity at diagnosis and during treatment, we used single-cell targeted DNA sequencing with the Tapestri platform. We designed a custom ALL panel and obtained accurate single-nucleotide variant and small insertion-deletion mutation calling for 305 amplicons covering 110 genes in about 4400 cells per sample and time point. A total of 108,188 cells were analyzed for 25 samples of 8 T-ALL patients. We typically observed a major clone at diagnosis (>35% of the cells) accompanied by several minor clones of which some were less than 1% of the total number of cells. Four patients had >2 NOTCH1 mutations some of which present in minor clones, indicating a strong pressure to acquire NOTCH1 mutations in developing T-ALL cells. By analyzing longitudinal samples, we detected the presence and clonal nature of residual leukemic cells as well as clones with a minor presence at diagnosis that evolved to clinically relevant major clones at later disease stages. Therefore, single-cell DNA amplicon sequencing is a sensitive assay to detect clonal architecture and evolution in T-ALL.

No abstract

Background T cell acute lymphoblastic leukemia (T-ALL) is a high-risk subtype that comprises 10–15% of childhood and 20–25% of adult ALL cases. Over 70% of T-ALL patients harbor activating mutations in the NOTCH1 signaling pathway and are predicted to be sensitive to gamma-secretase inhibitors. We have recently demonstrated that selective inhibition of PSEN1-containing gamma-secretase complexes can overcome the dose-limiting toxicity associated with broad gamma-secretase inhibitors. In this study, we developed combination treatment strategies with the PSEN1-selective gamma-secretase inhibitor MRK-560 and other targeted agents (kinase inhibitors ruxolitinib and imatinib; XPO-1 inhibitor KPT-8602/eltanexor) for the treatment of T-ALL. Methods We treated T-ALL cell lines in vitro and T-ALL patient-derived xenograft (PDX) models in vivo with MRK-560 alone or in combination with other targeted inhibitors (ruxolitinib, imatinib or KPT-8602/eltanexor). We determined effects on proliferation of the cell lines and leukemia development and survival in the PDX models. Results All NOTCH1-signaling-dependent T-ALL cell lines were sensitive to MRK-560 and its combination with ruxolitinib or imatinib in JAK1- or ABL1-dependent cell lines synergistically inhibited leukemia proliferation. We also observed strong synergy between MRK-560 and KPT-8602 (eltanexor) in all NOTCH1-dependent T-ALL cell lines. Such synergy was also observed in vivo in a variety of T-ALL PDX models with NOTCH1 or FBXW7 mutations. Combination treatment significantly reduced leukemic infiltration in vivo and resulted in a survival benefit when compared to single treatment groups. We did not observe weight loss or goblet cell hyperplasia in single drug or combination treated mice when compared to control. Conclusions These data demonstrate that the antileukemic effect of PSEN1-selective gamma-secretase inhibition can be synergistically enhanced by the addition of other targeted inhibitors. The combination of MRK-560 with KPT-8602 is a highly effective treatment combination, which circumvents the need for the identification of additional mutations and provides a clear survival benefit in vivo. These promising preclinical data warrant further development of combination treatment strategies for T-ALL based on PSEN1-selective gamma-secretase inhibition.

Background:T‐cell acute lymphoblastic leukemia (T‐ALL) is an aggressive malignancy of the lymphocytes that mainly affects children. High cure rates can be achieved in pediatric T‐ALL at the cost of long‐term side‐effects of conventional chemotherapy. Despite progress in the past years, cure rates remain below 50% in adults and for refractory or relapsed patients the prognosis remains dismal. Given that over 60% of T‐ALL patients carry activating mutations in NOTCH1, and that mutant NOTCH1 requires cleavage by the g‐secretase complex for its oncogenic activity, gamma‐secretase inhibitors have been investigated for the treatment of T‐ALL. MRK‐560 is a selective inhibitor of Presenilin 1 containing gamma‐secretase complexes for which we have recently shown that it has potent anti‐leukemia efficacy with minimal effects on normal cells and hence little gastro‐intestinal toxicity compared to broad gamma‐secretase inhibitors.Aims:We investigated whether MRK‐560 could act synergistically with other targeted inhibitors for the treatment of T‐ALL patients with NOTCH1 mutations. We tested the combination of MRK‐560 with a general antineoplastic inhibitor such as the exportin1‐inhibitor KPT‐8602 or a kinase inhibitor specifically chosen based on the mutational landscape of the T‐ALL patient.Methods:We cultured T‐ALL cell lines in vitro in the presence of MRK‐560, ruxolitinib (JAK inhibitor), imatinib (ABL1 inhibitor) or KPT‐8602 (XPO‐1 inhibitor) alone or in combination to assess cell growth and viability. We selected 3 patient‐derived T‐ALL xenograft samples with NOTCH1 or FBXW7 mutation and a mutation in JAK1/3 or an ABL1 fusion. These samples were treated in vivo with MRK‐560, ruxolitinib, imatinib and/or KPT‐8602. We monitored disease progression by measuring the percentage of human CD45+ cells in the blood of the mice and bio luminescence imaging of Luciferase‐positive PDX samples. Endpoint analysis included survival as well as leukemic infiltration in bone marrow and spleen. To evaluate gastro‐intestinal toxicity we weighed the mice on a weekly basis and performed histological analysis of the small intestine with Periodic acid‐Schiff staining.Results:Treatment of T‐ALL cell lines with MRK‐560 in combination with a second inhibitor resulted in a strong inhibition of proliferation that was more profound than the effect of either drug alone in cell lines that depend on NOTCH1 signaling. In PDX samples carrying a mutation in NOTCH1 or FBXW7 we observed a significantly stronger reduction of disease burden when MRK‐560 was combined with ruxolitinib, imatinib or KPT‐8602 compared to vehicle or single drug treatment. Moreover, the mice that were treated with a combination regimen (MRK‐560 + KPT‐8602, ruxolitinib or imatinib) had a significantly longer survival than with single drug treatment. There was no difference in weight or abundance of goblet cells in the gut of the mice from the different treatment groups, indicating that MRK‐560 was not toxic as single drug and also not in the combinations.Summary/Conclusion:Combination of the selective gamma‐secretase inhibitor MRK‐560 with other targeted therapies effectively inhibits T‐ALL cell growth in vitro as well as in vivo. Selective targeting of the gamma‐secretase complex is well tolerated and toxicity is not increased when MRK‐560 is combined with ruxolitinib, imatinib or KPT‐8602. These findings create new targeted treatment options for a large cohort of T‐ALL patients.

Following the publication of this article the authors noted that data describing precisely where phosphorylation sites in proteins modulated following JAK1 or JAK3 inhibition in mutant T-ALL samples was not clearly annotated. Therefore an additional sheet has been added to Supplementary Table 2.

Acute lymphoblastic leukemia (ALL), which is the most common cancer in children, shows extensive genetic intra-tumoral heterogeneity. This heterogeneity might be the underlying reason for an incomplete response to treatment and for the development of relapse. In order to envision the clinical implementation of a refined risk-category strategy based on ALL subclonal composition, it is essential to first generate a reference single-cell map and accumulate evidence on how the subclonal composition affects the response to treatment. For that, we performed large-scale and integrative single-cell genome and transcriptome profiling of pediatric samples at diagnosis, during drug treatment and in case of relapse. We used the 10x Genomics platform for single-cell RNA-sequencing analysis (around 4000 cells per sample) and the Tapestri Platform (Mission Bio) for targeted single-cell DNA-sequencing (around 5000 cells per sample) of the most mutated genomic regions in ALL. For the later, we developed a custom panel that covers 305 ALL mutational hotspots across 110 genes. We have determined a reference single-cell map of the cellular (based on the gene expression profile) and the clonal composition (based on the co-occurrence of mutations at each individual cell) for pediatric ALL at diagnosis (8 T-ALL and 10 B-ALL patients). We have also reconstructed the tumor phylogeny highlighting the order of mutational acquisition and the most likely pattern of evolution. Moreover, we have studied how T-ALL evolves during drug treatment at single-cell resolution in 4 patients, unraveling which are the most sensitive clones to the therapy, which are the most resistant ones and when relapse clones originated. Single-cell RNA-sequencing also provided information on how normal hematopoiesis recovers during chemotherapy treatment. The results show that ALL is typically composed by a major clone and 5-10 smaller clones that have different sensitivities to the therapy. We have been able to detect minor clones (<1% of the cells) at diagnosis that are clinically relevant. These very rare clones are either not responding to the therapy (present at minimal residual disease) or form the relapsed leukemia. These findings could have clinical implications for improved risk-stratification methods based on individualized patient's molecular profiles. Disclosures Maertens: Gilead Sciences: Other: Grants, personal fees and non-financial support; Cidara: Other: Personal fees and non-financial support; Amplyx: Other: Personal fees and non-financial support; F2G: Other: Personal fees and non-financial support; Merck: Other: Personal fees and non-financial support; Pfizer: Other: Grant and personal fees; Astellas Pharma: Other: Personal fees and non-financial support.