See article, p. 630.

  • Never-smokers with NSCLCs lacking common mutations were screened for RET fusions.

  • Patients with RET fusion-positive tumors were enrolled in a phase II trial of cabozantinib.

  • The first 3 patients enrolled remain progression-free on treatment, with 2 partial responses.

Activating RET gene fusions have recently been found in 1% to 2% of non-small cell lung cancers (NSCLC), most frequently in never-smokers and “pan-negative” tumors that lack mutations in genes commonly mutated in lung cancer. Preclinical data showing that inhibition of RET with small-molecule kinase inhibitors reduces RET fusion-positive cell survival suggests that RET fusions are therapeutically actionable, but clinical data on the use of RET inhibitors in NSCLC is needed. Drilon and colleagues screened a cohort of never-smokers with advanced pan-negative NSCLC for RET gene fusions to identify candidates eligible to participate in a prospective, open-label phase II trial of the multi-tyrosine kinase inhibitor cabozantinib in RET fusion-positive NSCLC. Five of 31 (16%) NSCLCs in this enriched cohort were RET fusion-positive, including one with a previously uncharacterized TRIM33–RET fusion. Of the first 3 patients enrolled in the trial, 2 had partial responses, 1 experienced disease stabilization, and all 3 remained progression free after 4 to 8 months of cabozantinib therapy. Toxicities were manageable with dose modification and medication. Although completion of the trial awaits recruitment of additional patients and analysis of long-term follow-up data, including potential resistance mechanisms, these encouraging preliminary data suggest that cabozantinib use is feasible and initially effective in RET fusion-positive NSCLC and illustrate how use of molecularly enriched cohorts can facilitate the identification of patients eligible for clinical trials of targeted therapies.

See article, p. 636.

  • Clinical sequencing identified numerous distinct FGFR fusions in diverse solid tumor lineages.

  • FGFR fusion proteins are activated via protein oligomerization and induce cell proliferation.

  • FGFR fusion-positive tumors exhibit increased sensitivity to targeted FGFR kinase inhibitors.

Recurrent gene rearrangements have been implicated as driving events and actionable therapeutic targets in multiple cancer types, including prostate cancer and lung adenocarcinoma. For example, activating fusions involving the fibroblast growth factor receptor (FGFR) genes are present in hematologic tumors, glioblastoma multiforme, and bladder carcinomas. Wu and colleagues performed integrative clinical sequencing, which included whole-exome and transcriptome sequencing, of over 100 primary tumor samples and matched normal tissues as part of the Michigan Oncology Sequencing Program. Intriguingly, in-frame FGFR2 gene fusions were identified in 4 patients with diverse tumor types, including cholangiocarcinoma, breast cancer, and prostate cancer. Analysis of larger tumor cohorts identified additional 5′ and 3′ FGFR family member fusions to distinct partners in a wide range of solid tumor lineages such as thyroid cancer, lung squamous cell carcinoma, and head and neck squamous cell carcinoma. These fusion proteins contained intact kinase domains, which were activated by oligomerization mediated by FGFR fusion partners, and promoted cell proliferation. Furthermore, FGFR3 fusion proteins enhanced the sensitivity of bladder cancer cells to pharmacologic FGFR blockade; treatment of fusion-positive cells with FGFR inhibitors induced cell-cycle arrest and reduced xenograft tumor growth. These results suggest that patients with FGFR fusions may benefit from treatment with targeted FGFR kinase inhibitors and advocate the use of clinical sequencing to facilitate personalized therapeutic decisions for patients with cancer.

See article, p. 648.

  • SDH-deficient GIST tumors exhibit a divergent genomic DNA methylation profile.

  • SDH mutation is correlated with global DNA hypermethylation in multiple tumor lineages.

  • IDH-mutant glioma has a comparable methylome, linking the Krebs cycle to DNA demethylation.

The succinate dehydrogenase (SDH) complex is a key mitochondrial enzyme of the Krebs/tricarboxylic acid cycle that suppresses tumorigenesis via regulation of succinate metabolite levels. Intracellular succinate accumulation inhibits α-ketoglutarate-dependent enzymes, including histone demethylases and the TET methylcytosine dioxygenases, which catalyze the first step in the DNA demethylation pathway. To investigate a potential correlation between SDH deficiency and epigenomic reprogramming, Killian and colleagues examined the DNA methylation profiles of a panel of gastrointestinal stromal tumors (GIST), which can be clinically characterized by mutations in SDH subunits or the KIT tyrosine kinase. Strikingly, SDH-mutant GIST tumors exhibited a distinct methylation signature relative to the profile of both KIT-mutant tumors and normal reference tissues. This methyl-divergent profile was distinguished by a significant increase in global DNA hypermethylation, particularly at DNase hypersensitive sites, and recurrent differential methylation of genomic targets in the absence of significant genomic copy number alterations. In addition, SDH mutation in GIST tumors was associated with loss of 5-hydroxymethylcytosine, consistent with inhibition of TET-mediated maintenance demethylation in the presence of succinate accumulation. Elevated DNA hypermethylation was also present in other SDH-mutant tumor lineages, including paraganglioma and pheochromocytoma, supporting the oncogenotype dependence of this methylome signature. Furthermore, a similarly perturbed methylation profile was detected in gliomas harboring mutations in another Krebs cycle enzyme, isocitrate dehydrogenase (IDH). These findings support a strong association between the mitochondrial Krebs cycle and cancer epigenomic reprogramming.

See article, p. 658.

  • MET is amplified in 3 of 4 tumors without KRAS mutation that relapsed on anti-EGFR therapy.

  • Amplification of MET can be detected in plasma DNA before relapse is clinically evident.

  • MET inhibition leads to regression of patient-derived cetuximab-resistant xenografts.

Monoclonal antibodies targeting EGF receptor (EGFR) such as cetuximab and panitumumab are effective in a subset of patients with metastatic colorectal cancer, but resistance invariably develops, most commonly due to secondary KRAS mutations. However, the underlying mechanisms of acquired resistance to anti-EGFR therapies in patients who do not develop KRAS mutations are unknown. Bardelli and colleagues analyzed plasma DNA of 7 patients with metastatic colorectal cancer who relapsed after an initial response to cetuximab or panitumumab and found MET gene amplification and overexpression in the post-treatment tumor tissue of 3 of the 4 patients lacking KRAS mutations. The authors detected very low levels of MET amplification in plasma DNA prior to anti-EGFR therapy but increasing levels of MET amplification after the start of therapy prior to clinically evident relapse, suggesting that MET amplification could serve as a biomarker of acquired resistance and disease recurrence. MET amplification was also associated with intrinsic EGFR antibody resistance in patient-derived colorectal cancer xenografts. MET hyperactivity induced cetuximab and panitumumab resistance in vitro by counteracting MAPK and AKT inhibition, but these effects could be reversed by a MET inhibitor. Moreover, MET inhibition sensitized patient-derived xenografts with de novo or acquired MET amplification to cetuximab and halted tumor growth. Together, these findings implicate MET amplification as a potentially actionable mechanism of acquired and intrinsic resistance to anti-EGFR therapy in colorectal cancer

See article, p. 674.

  • WNT signaling is upregulated in mouse and human MPNSTs and increases with tumor grade.

  • WNT activation promotes Schwann cell transformation and MPNST cell survival.

  • WNT blockade suppresses MPNST cell growth alone or in combination with mTOR inhibition.

Malignant peripheral nerve sheath tumor (MPNST) is a type of soft-tissue sarcoma with a poor prognosis that originates from Schwann cells or their precursors in the peripheral nervous system connective tissue. MPNSTs can arise spontaneously or can develop from benign neurofibromas in patients with neurofibromatosis type I syndrome (NF1), a genetic disorder caused by biallelic loss of the neurofibromin 1 tumor suppressor gene, but the underlying mechanisms are unclear. Watson and colleagues used the Sleeping Beauty transposon system to identify drivers of Schwann cell tumor formation in mice and observed enrichment of WNT pathway genes among common transposon insertion sites in benign neurofibromas and MPNSTs. Consistent with a role of WNT dysregulation in these peripheral nervous system tumors, expression of nuclear β-catenin and WNT target genes significantly increased with tumor progression in mice and in patient samples, β-catenin activation increased Schwann cell proliferation and anchorage-independent growth, and β-catenin inhibition reduced sporadic and NF1-associated MPNST cell tumorigenicity in vitro and in vivo. MPNST cells activated WNT signaling through multiple mechanisms, including β-catenin destruction complex downregulation, WNT ligand overexpression, and R-spondin 2 fusion, and were highly sensitive to small-molecule inhibitors of WNT signaling. Moreover, a screen of targeted therapies revealed that mTOR and WNT pathway inhibition synergized to induce MPNST cell death. Together, these findings implicate WNT signaling in Schwann cell transformation and tumor formation and suggest potential therapeutic options for MPNST.

See article, p. 690.

  • GSK3α stabilizes the TAK1–TAB1 complex to promote canonical IKK/NF-κB activity.

  • GSK3α stimulates noncanonical NF-κB signaling by enhancing p100/p52 processing.

  • Inhibition of GSK3 suppresses pancreatic cancer cell proliferation and tumor growth in mice.

NF-κB activation in mutant KRAS-driven tumors, including pancreatic cancer, promotes oncogenic transformation and cancer cell growth and survival. Glycogen synthase kinase 3 (GSK3), a downstream KRAS effector protein, has been implicated in pancreatic cancer cell viability, in part via regulation of IκB kinase (IKK) and canonical NF-κB signaling, but the mechanisms by which GSK3 promotes constitutive NF-κB activity in pancreatic cancer are unclear. Bang and colleagues found that depletion of GSK3α but not GSK3β reduced pancreatic cancer cell viability and decreased the protein levels of TGFβ-activated kinase 1 (TAK1, also known as MAP3K7), a critical regulator of IKK. Oncogenic KRAS expression enhanced TAK1 interaction with a required cofactor, TAK1 binding protein 1 (TAB1); inhibition of TAK1 impaired canonical NF-κB activity and specifically diminished the proliferation of KRAS-mutant cells. This regulation of TAK1 was dependent on KRAS-mediated upregulation of GSK3α, which interacted with TAK1 and maintained the stability of the TAK1–TAB1 complex. In addition, GSK3α knockdown inhibited NF-κB2/p100 processing and nuclear accumulation of the p52 subunit, indicating that GSK3α also regulates noncanonical NF-κB signaling. Consistent with a role for GSK3α as a potential therapeutic target, pharmacologic GSK3 inhibition suppressed the growth of human pancreatic tumor explants in mice and was associated with downregulation of proproliferative NF-κB target genes. These results identify GSK3α as a key activator of NF-κB downstream of mutant KRAS and support further testing of GSK3α-targeted therapeutics for pancreatic cancer.

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