The Janus kinases (JAKs), nonreceptor protein tyrosine kinases that mediate cytokine receptor signaling, are important drivers of multiple hematologic malignancies. Canonical JAK–signal transducer and activator of transcription (STAT) signaling results in phosphorylation of STAT transcription factors, which translocate to the nucleus, bind DNA, and transactivate expression of target genes. Noncanonical JAK signaling was first described in Drosophila through the identification of a global decrease in heterochromatin formation after JAK activation, resulting in derepression of gene expression. JAK2 was later demonstrated to phosphorylate histone 3 directly at tyrosine 41 in hematologic cells, thereby inhibiting heterochromatin protein 1 (HP1 ) binding and providing insight into the mechanism by which JAK2 can alter the expression of genes that are not direct STAT targets. JAK2 signaling is aberrantly activated in hematologic malignancies. Figure 1 illustrates the pathologic activation of JAK2 signaling in hematologic neoplasms. Multiple JAK2 chromosomal translocations have been described in various hematologic neoplasms, including a fusion of the oligomerization domain of TEL and the JH1 kinase domain of JAK2 (TEL-JAK2), which results in constitutive JAK2 tyrosine kinase activity in T-cell acute lymphoblastic leukemia (ALL). An activating mutation in JAK2, JAK2V617F, is found in the majority of patients with myeloproliferative neoplasms, and JAK2R683 activating mutations are found in 7% of high-risk B-cell ALLs and in approximately 25% of ALLs associated with Down syndrome. In lymphoma, JAK2 mutations have not been identified, but JAK2 signaling is pathologically activated. The SEC31A-JAK2 translocation, although rare, causes constitutively active JAK2 kinase activity in classical Hodgkin’s lymphoma (HL). JAK2 amplification, through copy number gain at 9p24, is present in 30% to 50% of HLs and primary mediastinal B-cell lymphomas (PMBLs). The gene expression profile of PMBL is different from that of other diffuse large B-cell lymphomas (DLBCL) and shares features with that of classical HL, a finding that partly relates to a gene expression signature indicative of increased JAK2 signaling common to both PMBL and HL. JAK2 has been postulated to play a role in lymphoma pathogenesis through other mechanisms as well, including enhanced autocrine interleukin-13 (IL-13) signaling, via amplified JAK2, leading to activation of STAT6. Amplification of 9p24 causes increased copy number of both JAK2 and the programmed death ligand-1 (PD-L1) and PD-L2 genes, which are also positively regulated by JAK2, causing evasion of immune surveillance through inhibition of T-cell activation. Similarly, the histone demethylase, JMJD2C, is also found in the recurrent 9p24 amplicon in HL and PMBL. JMJD2C demethylates histone 3 trimethylated lysine 9 (H3K9me3), thereby blocking HP1 recruitment, potentially acting synergistically with JAK2, reducing heterochromatin formation and increasing expression of key targets, including MYC. In a subset of lymphoma cases, the JAK-STAT signaling pathway is aberrantly regulated as a result of mutations in different pathway components. Suppressors of cytokine signaling (SOCS) proteins are negative regulators of JAK-STAT signaling. Loss-of-function mutations in SOCS1 have been described in PMBL and HL, and SOCS1 mutations were recently identified in DLBCL. Abnormal activation of JAK-STAT3 signaling also plays a key role in lymphoma, potentially driven in at least some cases by mutations in MYD88. MYD88 is an adaptor protein utilized by Toll-like receptors to activate nuclear factor B (NFB) and mitogen-activated protein kinase signaling. MYD88 mutations are found in approximately 40% of activated Bcell–like (ABC) DLBCL tumors and activate JAK-STAT3 signaling by means of an autocrine feed-forward loop, involving IL-6, IL-10, and type I interferon. A STAT3 gene expression signature is a wellrecognized feature of the ABC subtype of DLBCL, and a recent whole-exome sequencing study of 55 primary DLBCL tumors described mutations in STAT3 and STAT6. Pathogenic activation of JAK2 signaling in lymphomas led Younas et al to perform a phase I trial of SB1518, an oral JAK2/FLT3 inhibitor, in relapsed or refractory HL or NHL, described in the report accompanying this article. A total of 34 patients were treated in this dose-escalation study, 14 with classical HL and 20 with NHL (10 patients with follicular, five with mantle-cell lymphoma [MCL], four with DLBCL, two with small lymphocytic disease). A clinical benefit was observed in a subset of patients, with three partial responses, two in those with MCL and one with follicular disease; 15 patients demonstrated stable disease. Responses were enriched in patients who received more than 300 mg/d SB1518 and in patients with MCL, indolent lymphoma, or HL. A reduction in phosphorylated STAT3 and STAT5 levels was observed at all dose levels, suggesting that JAK2 inhibition was achieved. The median duration of treatment with SB1518 was just 88 days (range, 1 to 574 days), and 21 of 34 patients discontinued therapy for disease progression or lack of response. Despite the essential role of JAK2 in signaling downstream of the erythropoietin receptor, only 12% of patients developed anemia. Overall, SB1518 was reasonably well tolerated and seems to demonstrate some JOURNAL OF CLINICAL ONCOLOGY U N D E R S T A N D I N G T H E P A T H W A Y VOLUME 30 NUMBER 33 NOVEMBER 2