In This Issue

See article, p. 368.

Myeloproliferative neoplasms (MPN) are a class of clonal hematopoietic diseases characterized by an excess production of mature red blood cells harboring frequent activating mutations in the non–receptor tyrosine kinase JAK2, or in the thrombopoietin receptor (MPL), leading to aberrant cytokine signaling. Recently, whole exome sequencing efforts have identified recurrent frameshift mutations within the C-terminal domain of the endoplasmic reticulum chaperone protein calreticulin (CALR) in the majority of patients without JAK pathway mutations. However, the mechanisms by which CALR mutations transform hematopoietic cells and cause MPN have yet to be elucidated. Elf, Abdelfattah, Chen, and colleagues showed that bone marrow transplantation with cells expressing CALR^MUT, but not CALR^WT, led to MPN-associated hematopoietic features, including megakaryocyte lineage-specific hyperplasia, which led to the investigation of the role of MPL in CALR^MUT-driven MPN. In line with this, expression of CALR^MUT, but not CALR^WT, led to IL3-independent growth of IL3-dependent Ba/F3 murine hematopoietic cells when co-expressed with MPL, but not other type 1 hematopoietic cytokine receptors. Importantly, CRISPR/Cas9-mediated introduction of a frameshift mutation into the endogenous Calr locus also drove IL3-independent growth in cooperation with MPL. Mechanistically, cells transformed with CALR^MUT displayed activated STAT signaling and were sensitive to JAK2 inhibition. Mutagenesis experiments revealed that the positively charged amino acids within the mutant-specific C-terminus of mutant CALR were required for CALR^MUT-driven transformation by facilitating binding to MPL. Together, these data provide a mechanistic explanation for the interaction of mutant CALR with MPL to drive the development of MPN.

See article, p. 382.

Kinases are frequently mutated in cancers and easily druggable, and kinase inhibitors have shown significant clinical efficacy. However, resistance to kinase inhibitors inevitably arises due to multiple mechanisms of resistance, one of which is the bypass mechanism of the compensatory activation of other kinases to maintain the activation of key downstream signaling pathways. Miller, Oudin, and colleagues evaluated MEK inhibition in cancers which exhibit bypass signaling in response to kinase inhibition. Treatment of cancer cell lines and xenografts with MEK inhibitors (MEKi) resulted in reduced levels of circulating receptor tyrosine kinases (RTK), particularly MET and AXL. In patients with melanoma, the level of circulating RTKs, but not tumor-specific expression of RTKs, was predictive of MAPKi resistance, and concurrent measurement of circulating AXL and tumor expression of AXL revealed that MAPKi-induced decreases in circulating AXL were due to reduced shedding of AXL from tumors. Consistent with these findings, MEKi treatment resulted in the accumulation of phosphorylated AXL on the cell surface and decreased metalloproteinase (MP) activity. Co-treatment of xenografts with MEKi and a neutralizing antibody targeting tissue inhibitor of MP 1 (TIMP1) resulted in decreased tumor growth and delayed tumor recurrence. Mechanistically, MEKi or MPi promoted the association of cell surface TIMP1 with MPs, which resulted in the reduction of MP-mediated shedding of AXL and the induction of AXL-driven JNK bypass signaling. Together, these findings describe a post-translational bypass mechanism of resistance to kinase inhibitors which may potentially impact kinase inhibitor–based therapy.

See article, p. 400.

The tumor microenvironment in pancreatic ductal adenocarcinoma (PDAC) is characterized by extensive fibrosis and infiltration of immunosuppressive cell types that reduce the efficacy of chemotherapy and immunotherapy. Thus, targeting the tumor microenvironment represents a potentially promising therapeutic strategy. One PDAC therapy under investigation is the use of a CD40 agonist to stimulate monocyte-derived macrophages to promote the depletion of extracellular proteins, leading to tumor regression. However, the mechanisms by which CD40 redirects antifibrotic monocytes are unknown. Long, Gladney, and colleagues used a mutant Kras, mutant Trp53 mouse model of PDAC to determine the role of monocytes in reducing fibrosis. Treatment with a CD40 agonist stimulated a Ly6C⁺ inflammatory subset of monocytes to infiltrate tumors in a process coordinated by resident monocytes/macrophages and requiring the chemokine CCL2. Mechanistically, CD40 activation triggered the release of cytokines including IFNγ, which activated peripheral monocytes through downstream phosphorylation of STAT1 in both the mouse model and human patients. IFNγ was required for CD40 agonist-mediated degradation of the extracellular matrix. Further, CD40 activation induced Ly6C-dependent expression of multiple matrix metalloproteinases (MMP), which were required for fibrosis degradation. The chemotherapeutic efficacy of gemcitabine was increased when it was administered after treatment with a CD40 agonist, and the improved efficacy required IFNγ-, MMP-, and Ly6C-dependent antifibrotic activity. Altogether, these results elucidate a mechanism by which Ly6C⁺ monocytes/macrophages, commonly reported to be tumor-promoting, can be redirected to facilitate degradation of PDAC fibrosis and provide a framework for strategies to enhance chemosensitivity by reducing fibrosis.

See article, p. 414.

STAT5, a transcription factor activated by JAK kinases, is constitutively active in a variety of lymphoid malignancies, which has led to the development of STAT5-directed therapies. STAT5 is also essential for natural killer (NK) cell survival, and therefore inhibitors of STAT5 will target immune cells in addition to tumor cells. To better predict potential side effects of STAT5-directed therapies caused by immune targeting, Gotthardt and colleagues investigated the effects of STAT5 on NK cell activity. STAT5 depletion prevented the maturation and survival of NK cells, which could be rescued by BCL2 expression. The absence of STAT5 accelerated tumor growth in 2 mouse models, an effect that was correlated with increased angiogenesis and increased levels of VEGFA, and genetic depletion of Vegfa decreased tumor burden. Furthermore, analysis of human NK-cell populations revealed that low levels of STAT5 were associated with high levels of VEGFA. Cytokines present in the tumor microenvironment, including IL10, IL12, IL18, IL21, and IFNβ, led to downregulation of phosphorylated STAT5 accompanied by an increase in Vegfa expression in NK cells. In addition, ruxolitinib, a JAK/STAT inhibitor, increased VEGFA expression in mouse and human NK cells and increased the tumor burden in mice. Taken together, these findings demonstrate that STAT5 inhibition may convert NK cells from cytotoxic killers to tumor promoters by inducing VEGFA expression and angiogenesis. This unintended tumor-promoting effect provides impetus to reconsider STAT5 therapies.

See article, p. 430.

The histone acetyltransferase (HAT) CREB binding protein (CREBBP, also known as CBP) is a frequent target of loss-of-function mutations in human cancer. To identify potential therapeutic targets in CBP-deficient tumors, Ogiwara and colleagues performed a functional synthetic-lethal siRNA screen and identified the CBP paralog EP300, which encodes the HAT p300. In lung cancer cells with CBP loss or deleterious mutations, p300 knockdown induced G1 arrest followed by apoptosis. The HAT activity of p300 was essential for this synthetic lethality, as a HAT-deficient p300 mutant could not rescue the survival defect. Genome-wide expression data identified MYC as the gene whose expression levels most frequently changed when both CBP and p300 were lost, and exogenous MYC expression reversed the growth suppression mediated by p300 loss. Chromatin immunoprecipitation showed that p300 and CBP bound to the MYC locus, and p300 depletion reduced acetylation of histone 3 lysine 18 (H3K18) and H3K27 and RNA polymerase II recruitment at the MYC promoter in CBP-deficient cells. In xenograft and orthotopic mouse models of lung cancer, genetic depletion of p300 or pharmacologic inhibition of p300 HAT activity suppressed the growth of CBP-deficient tumors. Furthermore, inhibition of p300 also suppressed the growth of CBP-mutant hematologic cancer cells, indicating that p300 blockade may be effective in multiple tumor types. These findings demonstrate that paralog targeting of p300 suppresses tumor growth in CBP-deficient cells via suppression of MYC and suggest that p300 inhibitors have promise in treating CBP-deficient tumors.

See article, p. 446.

Natural killer (NK) cells reduce early tumor growth and metastasis, and are controlled in part by the expression of activating and inhibitory receptors. Following their recent observation that the checkpoint protein CD96 negatively regulates NK cell–mediated activity, Blake and colleagues tested whether blockade of CD96 would inhibit metastasis either alone or in combination with other antitumor therapies in mouse models of lung metastasis. Genetic deletion of CD96 or treatment with an anti-CD96 mAb led to fewer lung metastases, which were dependent on NK cells and IFNγ, but independent of Fc receptors. Genetic deletion of TIGIT, another negative regulator of NK cells and a marker of T-cell exhaustion, led to a further reduction in metastasis when combined with an anti-CD96 mAb, suggesting that combined inhibition of CD96 and TIGIT may protect against metastasis. As a monotherapy, anti-CD96 mAb showed superior antimetastatic activity compared with antibodies targeting the immune checkpoint proteins CTLA4 or PD-1; however, combination therapy with anti-CD96 and either anti-CTLA4 or anti–PD-1 or chemotherapy more effectively controlled metastasis and extended overall survival. Mechanistically, treatment with anti-CD96 alone or in combination with anti–PD-1 led to increased immune cell infiltration and enhanced IFNγ production and IL2 secretion. In addition to metastatic control, treatment with anti-CD96 inhibited primary tumor formation in a fibrosarcoma mouse model. Together, these data indicate that CD96 blockade may be a potential strategy to inhibit tumor progression and metastasis either alone or in combination with other therapies.

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