In This Issue

See article, p. 711

Strategies to expand and activate tumor-infiltrating lymphocytes include high-dose IL2 (aldesleukin), which can have durable clinical benefit but has a significant risk of serious adverse events. NKTR-214 (bempegaldesleukin) is human recombinant IL2 conjugated to polyethylene glycol (PEG) chains that slowly release upon intravenous administration, which biases NKTR-214 binding away from the low-affinity IL2Rα chain toward the intermediate-affinity IL2Rβγ complex. IL2Rα-mediated signaling is thought to underlie many of the adverse effects of aldesleukin, suggesting that NKTR-214 may provide immunostimulatory activity with fewer adverse effects. Bentebibel, Hurwitz, Bernatchez, and colleagues report a first-in-human phase I study of NKTR-214 in 28 heavily pretreated patients with advanced solid tumors, more than half of whom had progressed on immune checkpoint therapy. The primary objectives were to determine the safety, maximum tolerated dose, recommended phase II dose, and antitumor activity; the secondary objectives were to characterize the pharmacokinetics and pharmacodynamics to evaluate immunologic changes. No grade 4 or 5 adverse events occurred, and other adverse events were manageable and reversible. The best overall response was stable disease in 14 of 26 evaluable patients (54%), with 9 of 26 patients (35%) experiencing maximum tumor reductions ranging from 2% to 30%. NKTR-214 significantly increased proliferation and activation of CD4⁺ T cells, CD8⁺ T cells, and natural killer cells in peripheral blood, and analyses of matched baseline and on-treatment biopsies showed that NKTR-214 increased T-cell activation signatures, CD8⁺ T-cell infiltration, and T-cell clonality within tumors. Based on its demonstrated tolerability and biological activity, NKTR-214 is being combined with immune checkpoint inhibitors in ongoing clinical trials.

See article, p. 722

Despite the promise of PARP inhibitors (PARPi) in the treatment of a subset of patients with triple-negative breast cancer (TNBC), resistance to these inhibitors has stimulated the investigation of combining PARPi with other therapies such as immune checkpoint blockade. Pantelidou, Sonzogni, and colleagues show that PARP inhibition in BRCA-deficient TNBC tumors in immunocompetent mice resulted in prolonged survival compared with immunodeficient mice, suggesting that an intact immune system is required for PARP efficacy, and resulted in increased total and tumor-localized CD8⁺ T-cell counts; conversely, CD8⁺ T-cell depletion significantly reduced survival of PARPi-treated immunocompetent mice. Further, PARP inhibition activated the cGAS/STING pathway in BRCA-deficient TNBC cells by increasing the formation of cGAS-associated cytoplasmic micronuclei and enhancing phosphorylation of the STING effectors TANK-binding kinase 1 (TBK1) and IFN regulatory factor 3 (IRF3). PARPi-induced cGAS/STING activation also increased expression of proinflammatory cytokines and chemokines, which are critical for T-cell recruitment, in tumor cells and subsequently activated dendritic cells to drive their maturation. Treatment of either BRCA-proficient or PARPi-resistant cell lines failed to activate cGAS/STING signaling, and knockout of either STING or cGAS abolished TBK1 phosphorylation and proinflammatory chemokine production in response to PARP inhibition. In vivo, intratumoral STING depletion abolished PARPi-induced recruitment of CD8⁺ T cells and reduced IRF3 activation, resulting in increased tumor burden. Taken together, these findings link PARP inhibitors with an antitumor immune response, providing further evidence for combining PARP inhibition and immunotherapy to treat patients with BRCA-deficient TNBC.

See article, p. 738

The most common oncogenic KRAS mutation is KRAS^G12D, yet the oncogenic potential of alternative KRAS mutants such as KRAS^A146T, prevalent in hematopoietic and intestinal cancers, has never been fully characterized. Poulin and colleagues assessed the oncogenic properties of the A146T mutant in comparison to the more common G12D mutant. Compared to wild-type KRAS (KRAS^WT), KRAS^A146T exhibited increased rates of intrinsic and SOS1-stimulated nucleotide exchange and decreased rates of intrinsic and GAP-stimulated GTP hydrolysis. The crystal structure of KRAS^A146T revealed an extended conformation of switch 1 and β2 from the nucleotide binding pocket, similar to HRAS in complex with SOS1, suggesting that enhanced SOS1-mediated GDP dissociation constitutes a structural mechanism for the increased rate of intrinsic and GEF-induced nucleotide exchange in KRAS^A146T. In murine models of colon and pancreatic cancers, Kras^A146T mice exhibited moderate hyperproliferative phenotypes and reduced survival compared with Kras^WT mice, albeit to lesser degrees than Kras^G12D mice. Mass spectrometry in all three genotypes identified allele-specific, opposing changes to the colon and pancreas proteomes, with co-enrichment of only three pathways between Kras^A146T and Kras^G12D. Among kinases, only ERK2 was predicted to be activated by both Kras mutants. Colons in both mutants expressed higher levels of phosphorylated ERK1/2 compared with Kras^WT, and pharmacologic inhibition of MEK reduced crypt proliferation; similar increases in phospho-ERK1/2 were observed in the pancreas. Collectively, these results demonstrate the distinct biochemical mechanisms of activation regulated by Kras mutants that translate into global signaling properties, many of which depend upon the ability of each mutant to activate MAPK signaling.

See article, p. 756

Isocitrate dehydrogenase 1 (IDH1) mutations, particularly the R132H mutation, promote epigenetic dysregulation in cancers such as glioblastoma and acute myeloid leukemia (AML). To elucidate the mechanisms underlying the regulation of mutant and wild-type (WT) IDH1 in AML, Chen and colleagues utilized a comprehensive biochemical approach to systematically survey the role of tyrosine kinases (TK) in altering IDH1 function through post-translational phosphorylation. Treatment with a FLT3 inhibitor decreased the enzymatic activity of both mutant and WT IDH1 in patient AML regardless of FLT3 status, and in vitro kinase assays revealed that two distinct groups of TKs, Group 1 (including EGFR, JAK2, ABL1, PDGFR, and FGFR1) and Group 2 (including FLT3 and SRC), activated both WT and R132H mutant IDH. In many WT IDH1 cancers, Group 1 TKs phosphorylated monomeric WT IDH1 on the Y42 residue to promote IDH1 dimerization and substrate binding and subsequently activate the Group 2 TK SRC, which phosphorylates the Y391 residue of either monomeric or dimerized WT IDH1 to promote NADP+ cofactor binding. In AML, the Group 2 TK FLT3 phosphorylated Y391 and promoted the phosphorylation of Y42 via JAK2 to drive WT IDH1 homodimerization and IDH1 R132H homo- and heterodimerization in IDH1 R132H AML. These results characterize the intrinsic link between FLT3 and IDH1 in AML, describe the mechanisms underlying the regulation of wild-type and mutant IDH1 which are isoform-specific (both Y42 and Y391 are replaced with corresponding phenylalanine residues in IDH2), and suggest a therapeutic strategy for patients with mutant IDH–positive AML.

See article, p. 778

TET enzymes catalyze the formation of 5-hydroxymethylcytosine (5-hmC) modifications in DNA, which are critical for hematopoietic differentiation. However, the mechanism by which extracellular cytokine stimuli regulate TET activity to drive differentiation remains unclear. Jeong and colleagues show that phosphorylation of TET2 by the JAK2 kinase increases TET2 DNA hydroxymethylation activity. Cytokine-induced differentiation of hematopoietic stem cells or erythroid progenitors resulted in enhanced phosphorylation of TET2, which corresponded with an increase in 5-hmC levels. Pharmacologic inhibition of either FLT3 or JAK2 prevented TET2 phosphorylation and impaired its activation, and mass spectrometry revealed that JAK2 directly phosphorylates the TET2 catalytic domain at tyrosines 1939 and 1964. Phosphorylation of TET2 sustained the interaction between TET2 and JAK2 and increased the affinity of TET2 for KLF1, a transcription factor implicated in erythroid differentiation. Mutation of these tyrosine residues to alanine diminished hematopoietic colony formation and erythroid differentiation in vitro, suggesting an essential role for JAK2-mediated phosphorylation of TET2 in hematopoiesis. Furthermore, expression of constitutively active JAK2^V617F in murine erythroid cells or primary samples from patients with idiopathic myelofibrosis resulted in increased accumulation of 5-hmC and a genome-wide loss of cytosine methylation, in particular at a subset of overexpressed genes encoding proteins associated with erythropoiesis or involved in oncogenic pathways. Collectively, these results highlight JAK2 as a critical regulator of TET2-dependent chromatin remodeling in response to extracellular stimuli during hematopoiesis and suggest a potential role for JAK2-mediated modification of TET2 in the pathogenesis of myeloproliferative neoplasms.

See article, p. 796

The ETS family member SPI1 plays a critical role in B-cell maturation and activation, and inactivating mutations of SPI1 have previously been described in myeloid malignancies. Roos-Weil, Decaudin, Armand, and colleagues identify a recurrent missense somatic SPI1^Q226E mutation affecting the alpha 3 helix of the SPI1 DNA binding domain in 6% of patients with Waldenström macroglobulinemia (WM). Patients with the SPI1^Q226E mutation also had decreased overall survival compared with patients with wild-type SPI1. SPI1^Q226E maintained DNA binding and transactivation capacity but bound with reduced affinity to wild-type SPI1 motifs and with increased affinity toward other ETS transcription factor motifs. In multiple cell lines, expression of SPI1^Q226E increased proliferation compared with wild-type SPI1, and SPI1^Q226E induced a transcriptional profile similar to those of other ETS proteins, including several key transcription factors and signaling pathways central to B-cell physiology and pathology such as MYC and IRF4. Coexpression of SPI1^Q226E with mutant MYD88 in murine primary B cells in vivo conferred growth and survival advantages and reduced terminal B-cell differentiation, and RNA sequencing and bone marrow biopsy samples revealed decreased differentiation markers in patients with WM with mutant SPI1 compared with patients with WM with wild-type SPI1. Moreover, pharmacologic inhibition of MYC with JQ1 or IRF4 with lenalidomide blunted the proliferative stimulation induced by SPI1^Q226E. Taken together, these results identify SPI1^Q226E as a driver of WM and implicate altered DNA binding as a mechanism underlying oncogenicity of transcription factor missense mutations.

Note: In This Issue is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details.