Abstract
Background: In the molecular era, the relevance of tumor grade for prognostication of IDH1/2‐wildtype (WT) gliomas has been debated. It has been suggested that histologic grade II and III astrocytomas with molecular features of glioblastoma, IDH1/2‐WT have a similar prognosis to glioblastoma and should be considered for the same clinical trials. Methods: We integrated prospective clinical sequencing from 564 patients with IDH1/2‐WT gliomas (26 grade II, 71 grade III and 467 grade IV) with clinical and radiographic data to assess associations between molecular features, grade and outcome. Results: Compared to histologic grade IV IDH1/2‐WT astrocytomas, histologic grade II astrocytomas harbor fewer chromosome 7/10 alterations (P = 0.04), EGFR amplifications (P = 0.022) and alterations in cell‐cycle effectors (P = 1.9e‐11), but a similar frequency of TERT promoter mutations. In contrast, there is no difference in the frequency of these canonical molecular features in histologic grade III vs. IV IDH1/2‐WT disease. Progression‐free (PFS) and overall survival (OS) for histologic grade II tumors were significantly longer than grade III tumors (P = 0.02 and P = 0.008, respectively), whereas there was no difference in PFS and OS for histologic grade III compared to grade IV tumors. Median PFS for histologic grade II, III and IV tumors was 19, 11 and 9 months, respectively. Median OS for the same tumors was 44, 23 and 23 months, respectively. In histologic grade II and III IDH1/2 WT tumors, gliomatosis is associated with the absence of cell‐cycle alterations (P = 0.008) and enriched in grade II features (P = 0.1) and alterations in the PI3K‐AKT pathway (P = 0.09). Conclusions: Grade II histology has genotypic and phenotypic associations with prognostic implications in IDH1/2‐WT astrocytomas.
Keywords: astrocytoma, IDH1/2‐wildtype, genomics, histologic grade, prognosis
Introduction
Gliomas are one of the best characterized malignancies from a molecular standpoint. This molecular data has altered the way these tumors are classified and managed in clinic and is continuing to shape the design of our clinical trials.
Recent cIMPACT‐NOW guidelines have recommended that histologic grades II and III IDH1/2‐wildtype (WT) astrocytomas with EGFR amplification, chromosome 7 gain/10 loss or TERT promoter mutation should be re‐designated as “Diffuse astrocytic glioma, IDH1/2‐WT, with molecular features of glioblastoma, WHO grade IV” 3. This recommendation is based on numerous studies, including those examining histologic grade II and III gliomas in The Cancer Genome Atlas (TCGA) which revealed that diffuse gliomas should be segregated into three biologically distinct categories of disease, namely IDH1/2‐mutant astrocytomas, IDH1/2‐mutant oligodendrogliomas harboring 1p/19q codeletion and IDH1/2‐WT astrocytomas 5, 10, 12, 21. Further findings have indicated that a majority of histologic grade II and III IDH1/2‐WT diffuse gliomas have the molecular features and prognosis of IDH1/2‐WT glioblastomas, a prognosis that is significantly worse than that of IDH1/2‐mutant disease 1, 2, 13, 18, 20. The cIMPACT‐NOW guidelines recommend criteria for further stratification of IDH1/2‐WT gliomas to separate the aggressive subset from other biologically favorable glial and glioneuronal tumors 3, 11. At the same time, the clinical utility of histologic grading has been increasingly called into question. As the prognosis of IDH1/2‐WT histologic grade II and III gliomas with EGFR amplification, chromosome 7 gain/10 loss or TERT promoter mutation is similar to that of glioblastoma, IDH1/2‐WT, grade IV, it has been suggested that adult patients with these tumors should be eligible for the same clinical trials 3.
Molecular characterization of gliomas has resulted in another separate paradigm shift—one that has been fully adopted by the WHO 2016 classification. Gliomatosis cerebri was redesignated from a distinct pathologic entity to a growth pattern 17. Gliomatosis cerebri not only occur in IDH1/2‐mutant tumors, both 1p/19q codeleted and intact, where diffusely infiltrative and slow growth is expected, but also occurs in IDH1/2‐WT diffuse gliomas. The underlying biology and molecular basis of this growth pattern remains poorly understood and it is unclear if integration of this information is relevant for accurate prognostication.
In this report, we leverage our prospectively collected, genomically annotated cohort of IDH1/2‐WT tumors to identify the prognostic and genomic determinants of histologic grade II and III disease with special attention to EGFR amplification, chromosome 7 gain/10 loss and TERT promoter mutation. Additionally, we analyze the genetic basis of gliomatosis, a growth pattern that is enriched in these histologic grade II and III tumors.
Materials and Methods
Patient cohort and clinical details
All patients with a histologic grade II or III IDH1/2‐WT astrocytoma that underwent prospective genomic profiling as part of their routine care during the time period July 2013 to November 2018 were included for analysis. A retrospective waiver was obtained from the Memorial Sloan Kettering Cancer Center Institutional Review Board (IRB) to integrate tumor sequencing and clinical data, including detailed demographic, pathologic, radiographic and treatment information. This collection of histologic grade II and III IDH1/2‐WT astrocytomas expands upon a previously reported cohort, which includes the collection of glioblastoma, IDH1/2‐WT used in this data analysis, in that it includes additional cases, more mature survival data and additional radiographic and histologic annotation 14.
An integrated diagnosis was established by incorporating histologic features of the tumor and class‐defining molecular features from the sequenced resection by a neuropathologist (T.A.B.) using the WHO 2016 Classification of Tumors of the Central Nervous System 17. When tumors were identified to harbor a H3 K27M mutation by immunohistochemistry or next generation sequencing, these tumors were excluded as these mutations define a distinct entity per the WHO 2016 classification. Tumors were also excluded if they harbored atypical histological features that were inconsistent with classification as an astrocytoma. The histologic grade used for this analysis was the grade conferred at the time of initial pathologic review; it reflects the grade conferred on the clinical report based on the pathologic material supplied by the clinician. Pathologic review was performed by at least two neuropathologists, including a senior neuropathologist (M.R.), for patients with histologic grade II IDH1/2‐WT gliomas without an additional mutation commonly seen in glioblastoma besides the TERT promoter mutation, such as a mutation in EGFR, NF1 or PTEN.
Clinical annotation was obtained for all patients and included: 1) demographics; 2) date of initial diagnosis; 3) extent of initial resection; 4) MGMT promoter methylation status; 5) date of first progression as determined by RANO criteria 19; and 6) date of patient death and/or date of last contact with hospital. For tumors that were histologic grades II or III at diagnosis, the presurgical MRI for each sequenced resection was available and reviewed for 95 of 97 (98%) cases and it was determined whether the resection included a nodular area of contrast enhancement and whether there was gliomatosis at the time of diagnosis, defined as confluent non‐enhancing tumor in three or more lobes of the brain.
MGMT promoter methylation status was assessed on the primary or recurrent surgery as per standard of care; only CLIA‐certified assays were included in this report. The majority of samples underwent assessment as per standard of care at Memorial Sloan Kettering Cancer Center either by pyrosequencing or methylation‐specific real‐time PCR after sodium bisulfite conversion. For patient‐level methylation status, borderline methylation of the promoter was deemed negative, and patients with discordant results from two time points were deemed methylation positive.
Genomic sequencing and analysis
Tumor DNA was extracted from formalin‐fixed paraffin embedded (FFPE) specimens while normal DNA was extracted from mononuclear cells from patient‐matched peripheral blood. Specimens underwent sequencing in CLIA/New York State Department of Health‐certified environments using either MSK‐IMPACT (n = 480, paired tumor‐normal sequencing) or FoundationOne (n = 84, unmatched tumor sequencing) sequencing platforms. Both platforms cover protein‐coding exons of cancer associated genes, and detect somatic single‐nucleotide variants, insertions, deletions, copy‐number alterations, and certain gene fusions and structural variants 9, 22. All variants identified by MSK‐IMPACT were manually reviewed. For the MSK‐IMPACT cohort, the median target coverage was 764X (the median coverage by grade is 626X, 774X and 771X).
Detection of chromosome 7 gain and 10 loss was performed on patients that were sequenced using MSK‐IMPACT as previously described 14. Briefly, for samples sequenced on MSK‐IMPACT, allele‐specific DNA copy number analysis was performed using FACETS (www.github.com/mskcc/facets, version 0.3.9). For detecting concurrent chromosome 7 gain and chromosome 10 loss, any sample with at least 90% of each chromosome affected was considered altered and all determinations were manually reviewed for accuracy.
Tumors were not classified as negative for all three cIMPACT alterations if any of the hallmark alterations was indeterminable or not detectable in the molecular subgroup survival analyses in Figure 2B.
Figure 2.
Clinical outcome in MSK‐IMPACT cohort. Overall survival and progression‐free survival from pathological diagnosis (A) for all patients stratified by WHO grade of their tumor and (B) patients harboring grade II and III tumors, stratified by molecular subgroup, respectively. Patients for whom their tumor's molecular subgroup was indeterminable were excluded from (B). GBM mutated indicates the presence and triple negative indicates the absence of EGFR amplification, chromosome 7 gain/10 loss or TERT promoter mutation. Log‐rank test P‐values ≤0.05 are indicated in figure panels.
Somatic variant classification
Genomic variants were classified as oncogenic or likely oncogenic by the following criteria: 1) annotation with OncoKB 6; and 2) analysis of recurrent mutations across a collection of a total of 3,130 glioma samples 7, 8. All analyses described herein utilize only variants considered oncogenic or likely oncogenic according to these criteria. Alterations were mapped onto oncogenic pathways according to Supporting Table S1.
Statistical analyses
Statistical analyses were carried out using the R statistical programing environment, using base functions for statistical testing and the survival package for outcome analyses. All used tests are indicated throughout text and figures. Any comparisons of gene‐ or variant‐level counts were carried out on a per‐patient basis to account for patients for whom multiple sequenced specimens exist.
External data sources
Publicly available glioma mutational and clinical data were acquired from TCGA 4, 5.
Results
The study cohort consists of specimens from 564 patients with IDH1/2‐WT tumors. The majority of these tumors (n = 467) were IDH1/2‐WT glioblastomas (WHO grade IV). The histologic grade II diffuse astrocytomas were a rare species of diffuse glioma, accounting for only 26 tumors, almost universally non‐enhancing on MRI (20 of 25 evaluable cases). Among the 5 patients with enhancing components, the enhancement was minimal in 3/5 cases. Seven of 26 (27%) evaluable cases with histologic grade II IDH1/2‐WT disease met the criteria for gliomatosis, exhibiting confluent non‐enhancing disease involving at least three lobes of the brain. Comparatively, the histologic grade III anaplastic IDH1/2‐WT tumors were more common, accounting for 71 tumors. These histologic grade III tumors had higher rates of enhancement, occurring in 41 of the 70 patients with available imaging. Gliomatosis occurred less commonly in histologic grade III compared to grade II tumors (P = 0.1, Fisher's exact test), in that it was only present in 8 out of 70 evaluable patients (11%).
Compared to histologic grade IV IDH1/2‐WT tumors, histologic grade II astrocytomas harbor fewer chromosome 7/10 alterations (P = 0.04, Fisher's exact test), EGFR amplifications (P = 0.022, Fisher's exact test), but a similar frequency of TERT promoter mutations (Figure 1, proportion of tumors with 0 to 3 cIMPACT alterations is in Supporting Figure S1). Additionally, the entire cohort of histologic grade II tumors was less likely to harbor cell‐cycle alterations as compared to the histologic grade IV IDH1/2 WT gliomas (P = 1.9e‐11, Fisher's exact test). Likewise, comparing histologic grade II and III tumors, the grade II tumors were less likely to harbor cell‐cycle alterations (P = 6.9e‐07, Fisher's exact test), EGFR amplifications (P = 0.028, Fisher's exact test), with a similar trend for chromosome 7/10 alterations (P = 0.07, Fisher's exact test). No differences in these alterations were observed when comparing the entire cohorts of histologic grade III and IV tumors. With the exception of one tumor harboring a BRAF V600E mutation, sequencing of the histologic grade II tumors did not identify any tumors with characteristic alterations more commonly encountered in pediatric populations (FGFR, MYB, MYBL1, H3 K27M/G34R, see Supporting Table S2).
Figure 1.
Molecular features of IDH1/2‐WT gliomas. Bar chart showing the percentage of tumors per WHO grade harboring specific mutations or predicted oncogenic mutations in indicated pathways. Fisher's exact test P‐values ≤0.05 are indicated in figure.
The entire cohort of histologic grade II tumors demonstrated a longer median progression‐free survival (PFS) of 19 months vs. 11 months compared to grade III tumors (P = 0.02, log‐rank test, Figure 2A). Moreover, the entire cohort of histologic grade II tumors had a significantly longer median overall survival (OS) of 44 vs. 23 months for histologic grade III tumors (P = 0.008, log‐rank test). In contrast, the entire cohort of patients with tumors of grade III and IV histology fare similarly, the latter having a median PFS and OS of 9 and 23 months, respectively, with no statistically significant difference to histologic grade III tumors. This difference exists in spite of less aggressive treatment among the histologic grade II tumors, where radiotherapy was delayed in 7 patients, compared to just one patient with a histologic grade III tumor, in whom treatment was delayed because of treatment refusal. The pathologic diagnosis rendered in samples without a second molecular alteration that is common to glioblastoma (samples N7, 403, 404, 406, 925) was reviewed by a senior neuropathologist (M.R.), who confirmed that for samples N7, 404, 406 and 925 the diagnostic or the recurrent material available was consistent with the diagnosis of diffuse astrocytoma, IDH1/2‐WT. Sample 403 exhibited oligodendroglial features but is IDH1/2‐WT and 1p/19q intact, hence, the default position of many neuropathologist would be to call this an astrocytoma. With this additional review, there was consensus on the diagnosis among at least two neuropathologists for all five cases.
For grade II/III tumors, tumor grade appeared to be a superior prognostic indicator, as compared to the presence of an EGFR amplification, chromosome 7 gain/10 loss or TERT promoter mutation as the presence of any of these molecular alteration(s) did not identify a cohort of patients that were prognostically worse (Figure 2B). Of these 3 alteration, the only individual alteration that is prognostically significant for PFS and OS for histologic grade II and III tumors is chromosome 7/10 alteration (P = 0.01 and 0.045, respectively, log‐rank test). The presence of a cell‐cycle alteration in these histologic grade II and III tumors was also prognostically significant for PFS but not OS (P = 0.024 and 0.14, respectively, log‐rank test). When we investigated all possible combinations of TERT mutation, EGFR amplification and/or chromosome 7/10 alteration, no combination identified a cohort with a prognostically different outcome.
This difference in prognosis based on histologic grade cannot be accounted for by differences in MGMT promoter methylation status as only 2 histologic grade II tumors (of 21 evaluable) harbored promoter methylation, whereas 18 histologic grade III tumors (of 57 evaluable) were positive for MGMT promoter methylation (P = 0.08, Fisher's exact test, Table 1). Moreover, the difference is unlikely to be accounted for by extent of resection or age: Among the grade II tumors, which were predominantly diffusely infiltrative, non‐enhancing tumors, 38% (10/26) were biopsied, 54% (14/26) were subtotally resected and only 8% (2/26) were gross‐totally resected; among the grade III tumors, 55% (39/71) were biopsied, 32% (23/71) were subtotally resected and 10% (7/71) were gross‐totally resected (grade II vs. III, P = 0.25, chi‐square test). There was no statistically significant difference in age of diagnosis (median 58 and 57 years for grade II and III tumors, respectively; P = 0.4, two‐sided Student's t‐test).
Table 1.
Demographics.
| Characteristic | Grade 2 | Grade 3 |
|---|---|---|
| Age | ||
| Median | 58 | 57 |
| Range | 20‐79 | 19‐86 |
| Gender | ||
| Male | 14 (53%) | 44 (62%) |
| Female | 12(46%) | 27 (38%) |
| MGMT | ||
| Methylated | 2 (8%) | 18 (25%) |
| Unmethylated | 19 (73%) | 39 (55%) |
| Unknown | 5 (19%) | 14 (20%) |
| Extent of resection | ||
| Biopsy | 10 (38%) | 39 (55%) |
| STR | 14 (54%) | 23 (32%) |
| GTR | 2 (8%) | 7 (10%) |
| Unknown | ‐ | 2 (3%) |
When we queried the TCGA, we identified 18 grade II IDH1/2‐WT diffuse gliomas (histologically diagnosed as astrocytomas, oligoastrocytomas or oligodendrogliomas) with outcome data 16. Differences in OS were seen between the entire cohort of histologic grade II and histologic grade III/IV tumors in the TCGA data set (grade II vs. IV, P = 0.0009; grade II vs. III, P = 0.046, log‐rank test, Figure 3A), which was not seen when comparing histologic grade II and III tumors with and without EGFR amplification, chromosome 7/10 loss or TERT promoter mutations (P = 0.09, log‐rank test, Figure 3B). On closer examination, our prospectively collected study cohort differs from the TCGA cohort 5 in that a higher proportion of patients with histologic grade II and III tumors in our study cohort (89% vs. 77%) met the genomic criteria for the designation, “Diffuse astrocytic glioma, IDH1/2‐WT, with molecular features of glioblastoma, WHO grade IV.” Additionally, the cohort of patients with tumors that lack chromosome 7/10 alterations, TERT promoter mutation and EGFR amplification appears enriched in long‐term survivors in the TCGA data set relative to our cohort (56% vs. 27% alive at two years, respectively). This may be because of more stringent histologic and molecular curation that resulted in the exclusion of histologic grade II tumors that are better categorized as an entity other than diffuse astrocytoma, IDH1/2‐WT.
Figure 3.
Clinical outcome in TCGA cohort. Overall survival and progression‐free survival from pathological diagnosis (A) for all patients stratified by WHO grade of their tumor and (B) patients harboring grade II and III tumors, stratified by molecular subgroup, respectively. Patients for whom their tumor's molecular subgroup was indeterminable were excluded from (B). GBM mutated indicates the presence and triple negative indicates the absence of EGFR amplification, chromosome 7 gain/10 loss or TERT promoter mutation. Log‐rank test P‐values ≤0.05 are indicated in figure panels.
When we investigated the genomic determinants of gliomatosis in histologic grade II and III tumors, we found that cell‐cycle alterations were less common in tumors with this diffusely infiltrative growth pattern regardless of grade (4/15 vs. 54/81 in those with no gliomatosis; P = 0.008, Fisher's exact test) and there was a higher, statistically insignificant, rate of PI3K‐AKT pathway alterations (10/15 vs. 33/81; P = 0.09, two‐sided Fisher's exact test). In patients with gliomatosis, the cell‐cycle alterations were predominantly found in the patients with a large contrast enhancing component of the tumor. There was no difference in PFS or OS in patients with histologic grade II and III tumors with or without gliomatosis.
Discussion
A major function of the WHO classification is to enable neuropathologists, neuro‐oncologists, radiation oncologist, neuro‐radiologists and neurosurgeons around the world to use a common language when making treatment decisions and designing clinical trials. It has been suggested that clinical trials for glioblastoma should allow inclusion of tumors meeting criteria for “molecular glioblastoma,” regardless of histologic grade, based on data from the TCGA and other retrospectively collected cohorts 5, 18. Our data indicate that histologic grade II IDH1/2 gliomas carry a poor prognosis compared to IDH1/2‐mutant gliomas and yet have a differential prognosis to glioblastoma, IDH1/2‐WT in spite of meeting criteria for the proposed entity, “Diffuse astrocytic glioma, IDH1/2‐WT, with molecular features of glioblastoma, WHO grade IV.” While overall relatively small in number, our data suggest that inclusion of histologic grade II, IDH1/2 WT into clinical trials for glioblastoma has the potential to distort outcomes results. In contrast, our data do suggest that it would be reasonable to enroll patients with histologic grade III IDH1/2‐WT astrocytomas onto clinical trials for glioblastoma, IDH1/2‐WT because of similar PFS and OS.
Eighty‐eight percent of histologic grade II IDH1/2‐WT gliomas in our study cohort met criteria for “Diffuse astrocytic glioma, IDH1/2‐WT, with molecular features of glioblastoma, WHO grade IV” because of the high rate of TERT promoter mutations alone, in the setting of relatively infrequent EGFR amplification, chromosome 7/10 alterations and cell‐cycle alterations. While this may seem contrary to prior work suggesting that TERT promoter mutations occur later in the evolutionary trajectory of glioblastoma than chromosome 7 gain and 10 loss 15, we cannot exclude the possibility that this is caused by the technical limitations for detection of copy number alterations in low cellularity/low tumor purity specimens, frequently encountered among histologic grade II infiltrative gliomas (though of note, only six samples in our MSK‐IMPACT cohort was without a mutation with an observed allele frequency above 0.10, implying low purity and there was no enrichment for any histologic grade among these six samples). An alternative explanation is that a subset of patients with low‐grade, infiltrative growth patterns may in fact have a different molecular evolution than IDH1/2‐WT glioblastoma and the development of canonical molecular alterations, including cell‐cycle alterations, might play a role in malignant transformation in these tumors, as is the case in histologic grade II and III IDH1/2‐mutant tumors 14. In this scenario, histologic grade is reflective of an earlier timepoint in tumor evolution. Notably, we have observed examples of patients with histologic grade II lesions being reprofiled following malignant transformation with enhancing disease with the subsequent development of cell‐cycle alterations.
The determination of “molecular glioblastoma” criteria to some extent circumvents the nearly universal problem of grading based on limited material, wherein a lower grade is often rendered because of undersampling. A molecular criteria offers a more objective and arguably reproducible approach, as compared to the relatively subjective process of ascribing histologic grade.
Recognizing that this study, and even the TCGA data set, is limited by relatively small numbers, these data suggest that further studies are necessary to better understand the clinical and molecular characteristics of histologic grade II, IDH1/2‐WT gliomas with molecular features of glioblastoma, ahead of incorporating these tumors into glioblastoma clinical trials. At the very least, histologic grade should continue to be considered when evaluating outcomes of these trials. Finally, our findings underscore the utility of more comprehensive molecular characterization of gliomas for accurate tumor classification.
Conclusion
Histologic grade remains a prognostically relevant variable in a clinical cohort of IDH1/2‐WT gliomas with molecular features of glioblastoma (TERT promoter mutation, chromosome 7/10 alterations or EGFR amplification). Molecular differences between low and high grade disease can be identified among IDH1/2‐WT tumors; namely, histologic grade II tumors are less likely to harbor cell‐cycle alterations, EGFR amplifications and chromosome 7/10 alterations compared to their histologic grade III and IV counterparts. Gliomatosis appears to be mediated not only by the absence of cell‐cycle alterations, but also activation of the PI3K‐AKT pathway.
Conflict of Interest
A.L.L. reports research funding from Bristol‐Myers Squibb. I.K.M. reports research funding from General Electric, Amgen and Lilly; advisory roles with Agios, Puma Biotechnology, Debiopharm Group and Voyager Therapeutics; and honoraria from Roche. V.S.T. is a co‐founding investigator and consultant for BlueRock Therapeutics. T.J.Y. reports research funding from AstraZeneca and Kazia Therapeutics; consulting for Debiopharm. R.J.Y. reports research funding from Agios; consulting roles with Agios, Puma, NordicNeuroLabs and ICON plc. No other disclosures are reported.
Supporting information
Table S1 . Members of oncogenic pathways.
Table S2 . Molecular profiles of the grade II astrocytomas in the MSK‐IMPACT cohort.
Figure S1 . The number of cIMPACT alterations. Bar chart showing the number of cIMPACT alterations in the cohort of tumors where the status of all 3 alterations is known, expressed as a percentage of tumors per grade.
Acknowledgments
We thank Dr. Lisa M. DeAngelis for inspiring this study. This work was supported by National Institutes of Health (P30 CA008748).
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Aibaidula A, Chan AK, Shi Z, Li Y, Zhang R, Yang R et al (2017) Adult IDH wild‐type lower‐grade gliomas should be further stratified. Neuro Oncol 19:1327–1337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Aoki K, Nakamura H, Suzuki H, Matsuo K, Kataoka K, Shimamura T et al (2018) Prognostic relevance of genetic alterations in diffuse lower‐grade gliomas. Neuro Oncol 20:66–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Brat DJ, Aldape K, Colman H, Holland EC, Louis DN, Jenkins RB et al (2018) cIMPACT‐NOW update 3: recommended diagnostic criteria for “Diffuse astrocytic glioma, IDH‐wildtype, with molecular features of glioblastoma, WHO grade IV”. Acta Neuropathol 136:805–810. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Brennan CW, Verhaak RG, McKenna A, Campos B, Noushmehr H, Salama SR et al (2013) The somatic genomic landscape of glioblastoma. Cell 155:462–477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Cancer Genome Atlas Research Network , Brat DJ, Verhaak RG, Aldape KD, Yung WK et al (2015) Comprehensive, integrative genomic analysis of diffuse lower‐grade gliomas. N Engl J Med 372:2481–2498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Chakravarty D, Gao J, Phillips SM, Kundra R, Zhang H, Wang J et al (2017) OncoKB: a precision oncology knowledge base. JCO Precis Oncol 2017. [Epub ahead of print; 10.1200/PO.17.00011]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Chang MT, Asthana S, Gao SP, Lee BH, Chapman JS, Kandoth C et al (2016) Identifying recurrent mutations in cancer reveals widespread lineage diversity and mutational specificity. Nat Biotechnol 34:155–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Chang MT, Bhattarai TS, Schram AM, Bielski CM, Donoghue MTA, Jonsson P et al (2018) Accelerating discovery of functional mutant alleles in cancer. Cancer Discov 8:174–183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Cheng DT, Mitchell TN, Zehir A, Shah RH, Benayed R, Syed A et al (2015) Memorial sloan kettering‐integrated mutation profiling of actionable cancer targets (MSK‐IMPACT): a hybridization capture‐based next‐generation sequencing clinical assay for solid tumor molecular oncology. J Mol Diagn 17:251–264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Eckel‐Passow JE, Lachance DH, Molinaro AM, Walsh KM, Decker PA, Sicotte H et al (2015) Glioma groups based on 1p/19q, IDH, and TERT promoter mutations in tumors. N Engl J Med 372:2499–2508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Ellison DW, Hawkins C, Jones DTW, Onar‐Thomas A, Pfister SM, Reifenberger G, Louis DN (2019) cIMPACT‐NOW update 4: diffuse gliomas characterized by MYB, MYBL1, or FGFR1 alterations or BRAF(V600E) mutation. Acta Neuropathol 137:683–687. [DOI] [PubMed] [Google Scholar]
- 12. Hartmann C, Hentschel B, Wick W, Capper D, Felsberg J, Simon M et al (2010) Patients with IDH1 wild type anaplastic astrocytomas exhibit worse prognosis than IDH1‐mutated glioblastomas, and IDH1 mutation status accounts for the unfavorable prognostic effect of higher age: implications for classification of gliomas. Acta Neuropathol 120:707–718. [DOI] [PubMed] [Google Scholar]
- 13. Hirose Y, Sasaki H, Abe M, Hattori N, Adachi K, Nishiyama Y et al (2013) Subgrouping of gliomas on the basis of genetic profiles. Brain Tumor Pathol 30:203–208. [DOI] [PubMed] [Google Scholar]
- 14. Jonsson P, Lin AL, Young RJ, DiStefano NM, Hyman DM, Li BT et al (2019) Genomic correlates of disease progression and treatment response in prospectively characterized gliomas. Clin Cancer Res 25:5537–5547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Korber V, Yang J, Barah P, Wu Y, Stichel D, Gu Z et al (2019) Evolutionary trajectories of IDH(WT) glioblastomas reveal a common path of early tumorigenesis instigated years ahead of initial diagnosis. Cancer Cell 35:692–704.e12. [DOI] [PubMed] [Google Scholar]
- 16. Liu J, Lichtenberg T, Hoadley KA, Poisson LM, Lazar AJ, Cherniack AD et al (2018) An integrated TCGA pan‐cancer clinical data resource to drive high‐quality survival outcome analytics. Cell 173:400–416.e11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella‐Branger D, Cavenee WK et al (2016) The 2016 World Health Organization classification of tumors of the central nervous system: a summary. Acta Neuropathol 131:803–820. [DOI] [PubMed] [Google Scholar]
- 18. Reuss DE, Kratz A, Sahm F, Capper D, Schrimpf D, Koelsche C et al (2015) Adult IDH wild type astrocytomas biologically and clinically resolve into other tumor entities. Acta Neuropathol 130:407–417. [DOI] [PubMed] [Google Scholar]
- 19. Wen PY, Macdonald DR, Reardon DA, Cloughesy TF, Sorensen AG, Galanis E et al (2010) Updated response assessment criteria for high‐grade gliomas: response assessment in neuro‐oncology working group. J Clin Oncol 28:1963–1972. [DOI] [PubMed] [Google Scholar]
- 20. Wijnenga MMJ, Dubbink HJ, French PJ, Synhaeve NE, Dinjens WNM, Atmodimedjo PN et al (2017) Molecular and clinical heterogeneity of adult diffuse low‐grade IDH wild‐type gliomas: assessment of TERT promoter mutation and chromosome 7 and 10 copy number status allows superior prognostic stratification. Acta Neuropathol 134:957–959. [DOI] [PubMed] [Google Scholar]
- 21. Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W et al (2009) IDH1 and IDH2 mutations in gliomas. N Engl J Med 360:765–773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Zehir A, Benayed R, Shah RH, Syed A, Middha S, Kim HR et al (2017) Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med 23:703–713. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1 . Members of oncogenic pathways.
Table S2 . Molecular profiles of the grade II astrocytomas in the MSK‐IMPACT cohort.
Figure S1 . The number of cIMPACT alterations. Bar chart showing the number of cIMPACT alterations in the cohort of tumors where the status of all 3 alterations is known, expressed as a percentage of tumors per grade.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.