Venetoclax Has Activity in Patients with AML
See article, p. 1106.
Safety and activity of the BCL2 inhibitor venetoclax in AML was tested in a single-arm phase II trial.
Venetoclax has a 19% ORR and an acceptable tolerability in patients with heavily pretreated AML.
AML blasts dependent on BCL2 and independent of BCL-XL and MCL1 achieved the best response.
Venetoclax is a highly selective inhibitor of the antiapoptotic protein BCL2, which is often expressed in hematologic malignancies and promotes cancer cell survival. Preclinical data show that BCL2 inhibition induces cell death in acute myelogenous leukemia (AML) cells in vitro and in vivo, and also induces cell death in leukemic blasts, progenitor cells, and stem cells, providing a rationale for targeting BCL2 in patients with AML. Konopleva and colleagues evaluated the safety and efficacy of venetoclax monotherapy in a single-arm phase II trial in 32 patients with relapsed or refractory AML, or untreated AML unfit for intensive therapy. The primary objective was efficacy; secondary objectives included safety, and exploratory objectives evaluated biomarker utility. A total of 6 patients (19%) achieved an objective response, with 2 complete responses and 4 complete responses with incomplete blood count recovery (CRi), and 6 additional patients exhibited antileukemic activity that did not meet the response criteria. Venetoclax was generally well tolerated. BH3 profiling predicted clinical response and indicated that venetoclax resulted in on-target BCL2 inhibition. Of the 12 patients with IDH1/2 mutations, 4 achieved objective responses. Six patients showed a BCL2 family sensitive protein index, with >35% of tumor cells expressing BCL2 and/or <40% expressing BCL-XL, with 1 patient achieving a CRi and 3 exhibiting antileukemic activity. Altogether, these findings indicate that BH3 profiling may predict venetoclax sensitivity, and venetoclax warrants further investigation in combination with other agents in patients with AML.
Resistance Mechanisms Are Inhibitor-Specific in ALK-Driven Lung Cancer
See article, p. 1118.
First- and second-generation ALK inhibitors exhibit distinct ALK resistance mutational spectra.
Lorlatinib is efficacious against NSCLCs that harbor ALK resistance mutations.
Postrelapse genotyping of repeat biopsies will improve the design of targeted therapies.
Anaplastic lymphoma kinase (ALK) is activated by rearrangements in a subset of non–small cell lung cancers (NSCLC). However, patients with ALK-rearranged NSCLC eventually relapse after treatment with crizotinib, a first-generation ALK inhibitor, and second-generation ALK inhibitors, such as ceritinib, alectinib, and brigatinib. To further characterize the mechanisms underlying acquired resistance to ALK inhibitors, Gainor, Dardaei, and colleagues performed a comprehensive molecular and functional analysis of repeat biopsies from patients with ALK-rearranged NSCLC who had developed resistance to different ALK inhibitors. ALK resistance mutations were more common after treatment with the second-generation ALK inhibitors ceritinib, alectinib, and brigatinib compared to treatment with crizotinib. Further, each ALK inhibitor exhibited a specific pattern of frequency and distribution of ALK resistance mutations. Sequential treatment with crizotinib and then a second-generation ALK inhibitor resulted in the development of compound ALK resistance mutations in 6 of 48 patients. Although bypass signaling pathways were not the main mechanism of resistance to second-generation ALK inhibitors, the presence of epithelial–mesenchymal transition (EMT) in 5 of 12 ceritinib-resistant biopsies suggested that EMT may contribute to second-generation ALK inhibitor resistance. Lorlatinib, a third-generation ALK inhibitor, exhibited potency against single and compound ALK resistance mutations that arose after treatment with second-generation ALK inhibitors. Together, these results characterize the molecular mechanisms of resistance to first- and second-generation ALK inhibitors and emphasize the importance of longitudinal sampling.
PIM Kinases Promote Breast Cancer Resistance to PI3K Inhibition
See article, p. 1134.
PIM confers resistance to PI3K inhibitors by activating downstream PI3K effectors independent of AKT.
PIM1 is overexpressed in a subset of biopsies from PI3K inhibitor–resistant breast tumors.
Dual PIM/PI3K targeting may overcome resistance to PI3K inhibitors in PIK3CA-mutant breast cancer.
Breast cancers frequently harbor mutations in PIK3CA, which encodes the alpha isoform of the class I PI3K catalytic subunit. Targeting PI3K with small-molecule inhibitors such as BYL719 has had some success in clinical trials, but acquired resistance limits the long-term efficacy of PI3K inhibition. Le, Antony, Razavi, and colleagues performed a gain-of-function screen in BYL719-treated PIK3CA-mutant breast cancer cells and identified 63 putative PI3K inhibitor resistance genes, including the serine/threonine kinase genes PIM1 and PIM3. PIM1 activated downstream PI3K effectors that are normally activated by AKT in an AKT-independent manner, and small-molecule inhibition of PIM1 prevented activation of AKT effectors and sensitized PIK3CA-mutant breast cancer cells to BYL719. RNA sequencing of six paired breast tumor biopsies from patients before and after BYL719 treatment indicated that a PIM kinase activation signature was upregulated in two patients, providing further support for the role of PIM kinases in promoting resistance to BYL719. In addition, data from The Cancer Genome Atlas revealed that PIM family gene overexpression or copy-number alterations occurred in 14% of patients with treatment-naïve invasive breast cancer, and PIM1 amplification or overexpression were largely mutually exclusive with PIK3CA or PTEN alterations, suggesting a functional redundancy between PIM kinases and PI3K pathway components that may be exploited to promote resistance to PI3K inhibition. These findings elucidate a mechanism of PIM-mediated resistance to PI3K inhibition and support further study of combined inhibition of PIM kinases and PI3K in PIK3CA-mutant cancer.
Integrin-α10 Signals through TRIO and RICTOR to Drive Myxofibrosarcoma
See article, p. 1148.
Integrin-α10 expression is significantly associated with poor outcome in high-grade myxofibrosarcoma.
Integrin-α10 interacts with TRIO and RICTOR to activate RAC and AKT and promote growth and survival.
High-grade myxofibrosarcoma may be sensitive to combined mTOR and RAC inhibition.
Our understanding of myxofibrosarcoma, one of the most common adult soft-tissue sarcomas, has been hindered by its genomic complexity. Identification of potential therapeutic targets is urgently needed, as 30% to 40% of patients with high-grade myxofibrosarcoma die of metastatic disease. Okada, Lee, and colleagues analyzed gene expression profiles of 64 primary high-grade myxofibrosarcomas and found that expression of ITGA10, which encodes integrin-α10, was most significantly associated with decreased survival. Knockdown of integrin-α10 specifically inhibited growth and survival and decreased RAC and AKT activation in myxofibrosarcoma cells but not normal mesenchymal cells, suggesting that integrin-α10 signals through RAC and AKT in a tumor-specific manner. Given that chromosome 5p, the most common myxofibrosarcoma amplicon, contains TRIO, encoding a guanine nucleotide exchange factor that activates RAC, and RICTOR, encoding an essential subunit of mTOR complex 2, which activates AKT, the authors hypothesized that integrin-α10 signals through TRIO and RICTOR to drive myxofibrosarcoma. Consistent with this model, integrin-α10 was found to physically interact with TRIO and RICTOR, and myxofibrosarcoma cells required TRIO-dependent activation of RAC and RICTOR-dependent AKT activation. Accordingly, pharmacologic inhibition of either RAC or mTOR reduced myxofibrosarcoma growth in vivo, and simultaneous targeting of both RAC and mTOR led to further tumor growth control. The finding that high-grade myxofibrosarcoma is dependent on integrin-α10 signaling through TRIO and RICTOR provides insight into the etiology of this genomically complex tumor type and provides a rationale for clinical testing of inhibitors of these pathways in this disease.
MLL1 and DOT1L May Be Therapeutic Targets in NPM1-Mutant Leukemia
See article, p. 1166.
NPM1mut leukemic transformation requires MLL1- and DOT1L-dependent expression of HOX and MEIS1.
Inhibition of DOT1L and the menin–MLL1 interaction promote NPM1mut leukemic cell differentiation.
MLL1 and DOT1L may be therapeutic targets in patients with NPM1mut AML.
Nucleophosmin (NPM1) is frequently mutated in acute myeloid leukemia (AML), but it is unclear how NPM1 mutations initiate and maintain AML, which has limited the development of targeted therapies. NPM1-mutant (NPM1mut) tumors exhibit aberrant HOX expression, and because MLL1 is known to regulate HOX expression, Kühn and colleagues hypothesized that MLL1 may be required for NPM1mut leukemia. CRISPR/Cas9-mediated disruption of the menin–LEDGF binding motif of MLL1 in NPM1mut AML cells indicated that the interaction between MLL1 and its cofactor menin was required in NPM1mut AML. Disrupting the menin–MLL1 interaction with the small-molecule inhibitor MI-503 reduced proliferation; suppressed expression of HOX genes, MEIS1, and FLT3, a downstream target of MEIS1; and induced differentiation in NPM1mut cells. In vivo, MI-503 reduced the leukemic burden and extended survival, further suggesting that disrupting the menin–MLL1 interaction reverses leukemic gene expression to promote differentiation in NPM1mut leukemia. MI-503 reduced menin occupancy of the HOX and MEIS1 loci and corresponded with a locus-specific decrease in H3K4me3 and H3K79me2, suggesting a role for the H3K79 methyltransferase DOT1L in HOX regulation. Consistent with this finding, DOT1L inhibition reduced proliferation and colony-forming ability, suppressed MEIS1 and HOX gene expression, and induced differentiation. Additionally, combined inhibition of DOT1L and the menin–MLL1 interaction synergistically reduced HOX, MEIS1, and FLT3 expression to promote differentiation of NPM1mut cells, delayed leukemia onset, and extended survival in vivo. These findings suggest that therapeutic targeting of MLL1 and/or DOT1L may be effective in patients with NPM1mut AML.
Protumorigenic IL21+ TFH-Like Cells Foster Immune Privilege in Hepatoma
See article, p. 1182.
Innate monocyte activation induces protumorigenic IL21+ T follicular helper (TFH)–like cells in HCC.
Tissue-resident TFH-like cells stimulate plasma-cell maturation via IFNγ and IL21 signaling.
TFH-like cell–driven B-cell differentiation promotes M2b macrophage polarization and HCC growth.
IL21 secreted by T follicular helper cells (TFH) is a key regulator of B-cell maturation and has been implicated in various inflammatory diseases. However, the role of IL21+ immune cells in human cancer remains unclear. Chen, Xiao, Lao, and colleagues identified a subset of IL21+ T helper cells in human hepatocellular carcinoma (HCC), which were abundant in the peritumoral stroma and associated with accumulation of plasma B cells and monocytes/macrophages. These tumor tissue–resident IL21+ TFH-like cells expressed the TFH-specific transcription factor BCL6 and promoted the maturation of plasma cells, but displayed a distinct activated phenotype from lymph node TFH cells and were frequently characterized by production of IFNγ. Differentiation of IL21+ TFH-like cells required toll like receptor 4–mediated innate activation of tumor monocytes/macrophages, which triggered STAT1 and STAT3 signaling via secretion of the inflammatory cytokines IL1β and IL23. IL21+ cell density in the peritumoral stroma was inversely correlated with overall and disease-free survival in patients with HCC, suggesting that TFH-like cells have a protumorigenic function. Consistent with this idea, induction of plasma cell differentiation by TFH-like cells stimulated the polarization of macrophages to a protumorigenic M2b phenotype and enhanced hepatoma growth in mouse models. Taken together, these findings define a protumorigenic IL21+ TFH-like cell subset that links activation of innate immune responses to the establishment of immune privilege in HCC.
Note: In This Issue is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details.