Targeting GSK3: beyond the mitotic spindle?

In this issue of Blood, Wu et al¹  present preclinical data on the novel therapeutic target GSK3, a pleotropic regulator of signaling pathways involved in cancer. The authors show effective kinase inhibition by the small molecule inhibitor 9-ING-41, leading to mitotic spindle interference, consequent mitotic arrest, and lymphoma-specific cytotoxicity in vitro and in vivo. These data points offer the first mechanistic insight into GSK3-related biology and onco-addiction in non-Hodgkin lymphomas (NHLs) and provide a strong preclinical rationale for a phase 1/2 clinical trial of GSK3 inhibition in relapsed NHL to evaluate safety and efficacy.

The GSK3 kinase is expressed as 2 closely homologous isoforms (GSK3α and GSK3β) that have been shown to phosphorylate >100 different protein substrates.²  The growing and complex body of literature describes pleiotropic effects of GSK3 activity in normal and malignant biology, linking GSK3 activity to lymphoid signaling pathways, such as Wnt/β-catenin, Hedgehog, Notch, NF-κB, and PI3K/AKT, as well as glucose hypermetabolism.³  The authors employed messenger RNA (mRNA) and protein expression analysis and in vitro and in vivo studies to establish GSK3 as a promising drug target for inhibitor therapy in NHL. Using a combination of CRISPR-Cas9–mediated knockouts in lymphoma-derived cell lines, as well as xenotransplantation experiments in vivo, the presented body of work demonstrates convincingly that inhibition of GSK3 by the selective GSK3β small molecule inhibitor, 9-ING-41, or genetic knockout of the GSK3 homologs decreases the proliferation of lymphoma cells and leads to cell-cycle arrest in the prophase. As complex and not always “protumor” effects of GSK3 activity have been reported in other hematologic and solid tumors,⁴  the preclinical data of this study address an important knowledge gap in NHL.

The published and here presented preclinical data support further evaluation in clinical trials for advanced cancers, including relapsed NHL, and a phase 1/2 clinical trial (#NCT03678883) has been recently activated. With the decision to advance the selective GSK3β kinase inhibitor approach into human trials comes the need to gain more detailed mechanistic insight and to anticipate drug synergies, toxicities, and potential resistance mechanisms. The presented cell-cycle and imaging studies demonstrate that GSK3 colocalizes with centrosomes and microtubules, suggesting that 9-ING-41 interferes with mitotic spindle formation and arrests cells in the prophase. Overall, the data are consistent with a classification of 9-ING-41 as a microtubule disrupting agent, but it remains an open question as to which degree other oncogenic pathways, such as the NF-κB, MYC, and apoptotic pathways, that have been shown to be regulated by GSK3,³  are affected. Beyond the focus on mitotic spindle formation, deeper mechanistic insights in lymphoma are not available at this point, and the current knockout and inhibitor studies do not yet comprehensively answer important questions pertaining to other GSK3-related pathway dependencies and additional on- and off-target effects. Future studies, ideally employing genome-wide transcriptome analysis and phosphoproteomic studies, will be able to dissect the full complement of GSK3-related functions fully leveraging the generated model systems across a spectrum of B- and T-cell lymphomas.

A critical question going forward will be the potential indication range for selective GSK3 inhibition in cancer therapy and lymphoma, in particular, given the significant heterogeneity in disease biology, clinical courses, and the importance of predictive biomarker considerations in the era of precision oncology. To provide first insight into this question, a previously published study⁵  started to explore 9-ING-41 activity in a number of diffuse large B-cell lymphoma (DLBCL)-derived cell lines, and the authors of the current study significantly expanded the spectrum by inclusion of more DLBCL, mantle cell lymphoma (MCL), and T-cell lymphoma (anaplastic large cell lymphoma)-derived cell lines. The current study also included a limited number of primary lymphoma specimens and analysis of publicly available expression data,⁶  confirming global increases of GSK3 expression on both the mRNA and the protein levels and revealing a correlation of high expression levels with poor outcome in DLBCL patients following standard-of-care treatment. Further studies will also need to explore GSK3-related biology and expression signatures in the context of molecular hallmark features in specific lymphoma subtypes. Important considerations should include cell-of-origin phenotypes and activation status of other pathways (eg, in DLBCL and MCL) that could be targeted in combination therapy, such as other small molecule inhibitors (Bruton tyrosine kinase, ibrutinib; anaplastic lymphoma kinase, crizotinib; B-cell lymphoma 2, venetoclax; phoshoinositide 3-kinase δ, idelalisib).^7-9  It might be a reasonable strategy to let first trial results inform on focus areas, at least to some degree, but this approach harbors the risk of missing activity in smaller, well-defined subtypes. Therefore, simultaneous deeper interrogation of GSK3 inhibition-related phenotypes across a wider spectrum of lymphomas seems warranted to prioritize clinical testing in molecularly defined disease entities, to refine combination strategies, and to anticipate side effects and treatment resistance.