Year: 2019
Type: article
Abstract: The phase II portion of Radiation Therapy Oncology Group 9802 for “low-risk” patients, defined as those aged <40 years who achieved a surgeon-defined gross total resection (GTR) in whom observation did not adversely affect overall survival (OS), identified 3 risk factors for worse progression-free survival (PFS): preoperative tumor size ≥4 cm, astrocytoma or oligoastrocytoma histology, and ≥1 cm of residual tumor on postoperative MRI.1 It is still unclear whether these factors continue to predict for progression when accounting for molecular markers. To identify risk factors for PFS in the molecular era among patients with low-grade glioma (LGG) who undergo GTR followed by initial observation, Tom et al reviewed patients with World Health Organization grade 2 LGG treated at a single institution and included only those who underwent MRI-confirmed GTR followed by initial observation.2 Molecular classification was obtained at either the time of diagnosis or pathology review. The study comprised 144 patients treated between 1994 and 2014. Median age was 29 years, median tumor size 2.7 cm, and median follow-up 81 months. Molecular classification was 13% IDH-mutant 1p19q-codeleted, 21% IDH-mutant 1p19q-intact, 39% IDH1-R132H-wildtype, and 28% undetermined. For the entire cohort, 5- and 10-year PFS and OS were 71% and 53%, and 98% and 90%, respectively. On multivariate analysis, factors associated with worse PFS included increasing age at diagnosis (P = 0.03), increasing preoperative tumor size (P < 0.0001), and IDH-mutant 1p19q-intact classification (P = 0.025). Median PFS for patients with IDH-mutant 1p19q-codeleted, IDH-mutant 1p19q-intact, and IDH1-R132H-wildtype tumors were 113 months, 56 months, and not reached, respectively. In multivariable analysis molecular classification was significantly associated with PFS (P < 0.0001) but not overall survival (P = 0.20). This study has several limitations. First of all, there are the inherent limitations of a retrospective study. Additionally, patients in the IDH1-R132H-wildtype group did not undergo sequencing for noncanonical IDH1/2 mutations. Thus, it is likely that the favorable outcomes among the IDH1-R132H-wildtype patients are overestimated and should be interpreted with caution. Other unfavorable prognostic molecular markers, such as CDKN2A loss, TERT promoter mutations, chromosome 7 gain, chromosome 10 loss, and EGFR amplification were not evaluated. In conclusion, for patients with LGG who undergo MRI-confirmed GTR and initial observation in the molecular era, IDH-mutant 1p19q-intact classification is associated with worse PFS in addition to the well-known clinical prognostic factors. Because tumor progression is associated with adverse health-related quality of life, these factors may aid clinicians and patients in the shared decision-making process regarding goals of surgery and timing of postoperative therapy. Further study is required to elucidate why IDH-mutant 1p19q-intact LGGs are at higher risk for early progression. Approximately 10–20% of patients with cancer will be diagnosed with and eventually treated for brain metastases over their disease course. Whereas classical treatments, surgery, radiation, and chemotherapy offer only limited prospects, the recent developments of specific immunotherapies and more targeted therapies have changed the game for relevant subgroups of patients. Amongst the benefitting subgroups are non-small cell lung cancer (NSCLC) patients with chromosomal rearrangements of anaplastic lymphoma kinase (ALK), ROS proto-oncogene 1 (ROS1), rearranged during transfection (RET), and neurotrophic receptor tyrosine kinases (NTRKs), which are each observed in 0.1–5% of NSCLC. ROS1 gene rearrangement is also observed in cholangiocarcinoma, glioblastoma, or colorectal cancer. Similarly, NTRK fusions, albeit rare, are also seen outside NSCLC and lung adenocarcinoma, in pediatric and adult cancers as well as sarcoma. ROS1 inhibitors are in frequent use and are effective, but several resistance mechanisms are described. Use in patients with brain metastases should focus on the presence of the target and use of inhibitors with established activity in brain metastases. In a recent article, Katayama et al. describe the activity and mechanism of action of a novel, brain-penetrant ROS1/NTRK inhibitor that is active after a secondary kinase-domain mutation (eg, ROS1-G2032R) has occurred with crizotinib therapy in preclinical models.1 Tumors with these mutations are also resistant to lorlatinib and entrectinib, next-generation ROS1 inhibitors. The selective inhibitor DS-6051b has demonstrated activity both in wildtype and G2032R-mutant ROS1-rearranged cancers and also in NTRK-rearranged cancers in vitro and in vivo. More importantly, DS-6051b is effective in treating crizotinib-resistant ROS1-positive cancer with secondary kinase domain mutations, especially the G2032R mutation. Once understood and targeted, the molecular lesions offer a logical, testable sequence of treatments. A good example is osimertinib,2 which is a very effective epidermal growth factor receptor inhibitor that is also active after the resistance mutation, T790M, has occurred, allowing for a switch from earlier-generation EGFR inhibitor, if it is not (already) considered treatment of choice in patients with EGFR-activated disease and brain metastases. We will have to develop rational sequences and filtering systems of testing and answer the question (for each compound and in general) of whether or not retesting of the brain lesions is warranted at all3 or whether the system’s mutation or a surrogate tissue (liquid) biopsy is sufficient. Pediatric glioma is the most common brain tumor of childhood and includes subsets of tumors driven by genetic lesions in the Ras-Raf pathway. Among these tumors are pilocytic astrocytoma and ganglioglioma, which commonly harbor BRAF fusions or the BRAF V600E mutation. It also includes brain tumors associated with neurofibromatosis 1 (NF1), which are thought to occur through Ras activation induced by loss of NF1. Targeted inhibitors of Ras are not available, and Raf inhibitors are only active against tumors with BRAF V600E mutations. However, inhibiting the pathway downstream, with MEK inhibitors already reaching the clinic, offers an alternative. Phase I trials of MEK inhibition in pediatric LGG and in NF1-associated plexiform neurofibromas have already shown promise,1 prompting a phase II trial with results from some groups presented in a new report by by Fangusaro et al.2,3 This report describes results from two of six strata in a phase II study, with these two strata having enrolled 50 patients at multiple centers with recurrent, refractory, or progressive low-grade gliomas: 25 patients with pilocytic astrocytoma with KIAA1549-BRAF fusions or BRAFV600E mutation and 25 patients with NF1-associated LGG. Treatment was with the MEK1/2 inhibitor selumetinib at 25 mg/m2 orally twice per day, and the primary outcome was objective response over at least an 8-week period. Nine out of 25 patients with pilocytic astrocytoma showed a sustained response, many lasting years and continuing through the end of the observation period, with 10/25 patients in the NF1 LGG group demonstrating similar prolonged responses. Notably, progression-free survival with selumetinib treatment was better for pediatric LGG with KIAA1549-BRAF fusions than for those with BRAFV600E mutation. Most common high-grade side effects were expected elevations in creatine phosphokinase (10%) and rash (10%). These finding more firmly establish the utility of MEK inhibition as an alternative to chemotherapy for pediatric LGG with RAF lesions or in the setting of NF1; in fact, the results have already spawned phase III trials comparing chemotherapy to MEK inhibition for these patients. However, the fact that a minority of patients derived long-term benefit raises other questions. Biomarkers of the sensitive subset remain to be identified and would be extremely useful. In addition, it may also be possible to identify synergistic combinations of MEK inhibitors with other agents to boost efficacy versus pediatric LGG driven by lesions in Ras and Raf. The nucleolus is the site of ribosomal RNA (rRNA) and transfer RNA (tRNA) biosynthesis. A common feature of many cancers, including glioblastoma (GBM), is an enlargement of the nucleolus (nucleolar hypertrophy). The mechanistic underpinnings and functional relevance of nucleolar hypertrophy are not well understood. Another common feature of cancers and GBM is the elevation of GTP (but not ATP) concentrations in the tumors. A recent study, published in the journal Nature Cell Biology, established a new link between nucleolar hypertrophy and de novo GTP biosynthesis that involves the enzyme IMP dehydrogenase-2 (IMPDH2) with implications for GBM malignancy.1 The study first showed that GTP de novo biosynthesis is significantly increased in GBM tissues, cells, and cancer stem cells as compared with normal brain and primary glia. A screen used to determine the mechanism of increased GTP biosynthesis uncovered elevated expression and activity of IMPDH2 and its association with poor survival in GBM. IMPDH2 is the enzyme that commits IMP to GTP biosynthesis. The study showed that IMPDH2 is elevated in several GBM databases, patient tumor specimens, and GBM tumor cells and models. Genetic and pharmacological inhibitions of IMPDH2 led to a strong decrease in GTP biosynthesis in GBM cells but not in primary glia, demonstrating that upregulation of IMPDH2 leads to increased GTP biosynthesis in GBM. Pharmacological inhibition of IMPDH1/2 (to avoid compensation of IMPDH2 loss by IMPDH1) led to decreased GBM cell and stem cell proliferation and suppression of in vivo GBM xenograft growth, demonstrating a role for IMPDH in maintaining GBM malignancy. The study then linked IMPDH2 upregulation and associated increase in GTP biosynthesis to rRNA and tRNA biosynthesis and nucleolar hypertrophy using several approaches including tracking of ribonucleotide fate and analysis of Pol I (transcribes rRNA) and Pol III (transcribes tRNA) activities and promoter binding. Also, transcriptome screening of IMPDH2-dependent gene expression showed that it regulated genes that are implicated in nucleolar function and morphology. Finally, inhibition of IMPDH2 in GBM cells led to decreased nucleolar density and size. Altogether, this study established a novel link between IMPDH2 upregulation, the increase in GTP biosynthesis, nucleolar hypertrophy, and GBM malignancy. Cellular defense against oxidative and electrophilic stress include the Kelch-like ECH-associated protein 1 (Keap1) – nuclear factor E2 related factor 2 (Nrf2). Under stressful conditions, critical thiol residues in Keap1 are modified, leading to Nrf2 stabilization and subsequently target genes that provide resistance against endogenous and exogenous stresses. Aberrant Keap1-Nrf2 signaling can contribute to radio- and chemoresistance by expression of cytoprotective genes, as described in lung, pancreatic, ovarian, and biliary malignancies. In a recent article, Pölönen et al analyzed the Broad-Novartis Cancer Cell Line Encyclopedia (CCLE) and the Cancer Genome Atlas (TCGA) and other multi-omics datasets to identify and characterize cancer types where Nrf2 is active.1 The authors benchmarked Nrf2 activation signature based on gene expression profiles. They identified cancer types with an overactive Nrf2 pathway across more than 900 samples. In addition to cancers known to have a Nrf2 overactivity profile, they reported that Nrf2 is hyperactive in gliomas. They confirmed this in other sample sets including the TCGA, the Rembrandt data set, and the Human Glioma Cell Cultures (HGCC) collection. Analyzing 550 patient microarray gene expression profiles from the Rembrandt data set, Nrf2 activity was high in 13% of the samples. None of the grade I glioma patient samples had high Nrf2 activity, while it was higher in grades II-IV. In the TCGA, 17% of the samples had high Nrf2 activity. Interestingly, the percentage of Nrf2 overactive samples was the highest in GBM. The authors used Gene Set Enrichment Analysis (GSEA) to identify processes linked to Nrf2 activity. They linked significant pathways from TCGA, CCLE, and HGCC datasets to other biological processes and ranked them based on mean enrichment scores. The highest-ranked cancer-related process was mesenchymal transition. They concluded that increased Nrf2 activity may be linked to the mesenchymal subtype and potentially be involved in mesenchymal transition. Mesenchymal and stem cell-like properties render the mesenchymal glioma subtype highly resistant to treatment. Nrf2 and mesenchymal transition in another 33 cancer types were then compared using correlation analysis between the Nrf2 Gene Set Variation Analysis (GSVA) score and epithelial-to-mesenchymal transition (EMT)-associated gene sets in the other cancer types. Correlation between Nrf2 activity and EMT was particularly strong in GBM. The authors also reported a link between Nrf2 activity and cell cycle and energy metabolism. Invasion was associated with Nrf2 activity. This could have implications for therapy resistance in the clinic. They authors observed additionally that p62, a classical receptor of autophagy, creates a positive feedback loop with Nrf2 in glioma. This in turn, may promote tumorigenesis and stemness properties and contribute to autophagy dysregulation. Increased understanding of Nrf2 and p62 and their role in the mesenchymal subtype molecular network and the function of Nrf2 in metabolic reprogramming may help efforts to overcome therapy resistance in glioma and specifically in GBM.
Source: Neuro-Oncology
Language: English
Subfield: Genetics
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4252922268