Isocitrate Dehydrogenase Mutations in Low-Grade Gliomas Correlate With Prolonged Overall Survival in Older Patients

Isocitrate Dehydrogenase Mutations in Low-Grade Gliomas Correlate With Prolonged Overall Survival... Abstract BACKGROUND Older age has been associated with worse outcomes in low-grade gliomas (LGGs). Given their rarity in the older population, determining optimal treatment plans and patient outcomes remains difficult. OBJECTIVE To retrospectively study LGG survival outcomes in an older population stratified by molecular genetic profiles. METHODS We included patients age ≥40 yr with pathologically confirmed World Health Organization grade II gliomas treated at a single institution between 1995 and 2015. We collected tumor genomic information when available. RESULTS Median overall survival for the entire group (n = 111, median age 51 yr, range 40-77 yr) was 15.75 yr with 5- and 10-yr survival rates of 84.3% and 67.7%, respectively. On univariate analysis, patients with isocitrate dehydrogenase (IDH) mutation had significantly increased survival compared to IDH wildtype (hazard ratio [HR] 0.17 [0.07-0.45], P < .001). Older age, seizure at presentation, larger tumor size, IDH wildtype, biopsy only, chemotherapy, and radiation were significantly associated with shorter survival based on univariate analyses. In patients with known IDH status (n = 73), bivariate analysis of IDH mutation status and age showed only IDH status significantly influenced overall survival (HR 0.22 [0.07-0.68], P = .008). Greater surgical resection was predictive of survival, although extent of resection significantly correlated with IDH mutation status (odds ratio 7.5; P < .001). CONCLUSION We show that genomic alterations in LGG patients ≥40 occur at high rates like the younger population and predict a similar survival advantage. Maximizing surgical resection may have survival benefit, although feasibility of resection is often linked to IDH status. Given the importance of molecular genetics, a redefinition of prognostic factors associated with these tumors is likely to emerge. Glioma outcomes, IDH, Low-grade gliomas ABBREVIATIONS ABBREVIATIONS EORTC European Organisation for Research and Treatment of Cancer FLAIR fluid-attenuated inversion recovery GTR gross total resection HR hazard ratio IDH isocitrate dehydrogenase KPS Karnofsky Performance score LGG low-grade glioma MRI magnetic resonance imaging OS overall survival PCV procarbazine, CCNU (lomustine), and vincristine chemotherapy RCT randomized controlled trial STR subtotal resection WHO World Health Organization Diffuse low-grade gliomas (LGGs) are typically encountered in younger patient populations and hold a favorable, yet highly variable, prognosis. Knowledge of the evolution and growth of these tumors has expanded greatly in recent decades, largely due to the discovery and characterization of genetic events such as IDH mutations and 1p19q codeletion.1-4 These molecular diagnostic insights have led to improved prognostication, a change in the pathological classification of all gliomas, and are now incorporated into clinical trial design and analysis.5-7 Prior to discovery and incorporation of these molecular and cytogenetic features, retrospective database studies and randomized controlled trials (RCTs) indicated that LGGs present more aggressively and have worse outcomes when they occur in patients older than 40.8-11 By analyzing some of these RCTs, a 2002 paper by Pignatti et al12 established 5 criteria that indicate poor prognosis for patients with LGGs: age >40 yr, astrocytoma histology subtype, largest diameter of the tumor >6 cm, tumor crossing the midline, and presence of neurological deficit before surgery. While older age remains a risk factor, it has been difficult to study the true impact of age in LGGs because of the relative rarity of these tumors in older patients. As such, predicting outcome and determining optimal treatment plans of surgery, radiation, and chemotherapy remains a challenge.13 Particularly, since prognostically favorable IDH mutations are more common in patients younger than 40 yr old, absence of molecular and genetic data in these past trials limits interpretation of whether patient outcomes are truly dependent on patient age or simply a worse genomic profile.14,15 In this study, we present outcomes from the largest reported single-institution population of patients with grade II glioma and high-risk criteria of age 40 yr or greater. Including IDH mutation and 1p19q codeletion status, we examined whether overall survival (OS) within an older population is influenced by molecular markers.1,8-12,16-19 METHODS Patient Data Acquisition We performed a retrospective review for all patients who received surgery for diagnosis and/or resection of a grade II glioma by World Health Organization (WHO) 2007 criteria who were ≥40 yr old at the time of diagnosis from January 1, 1995 to December 31, 2015 at a single institution. Neuropathologists determined tumor grade from operative specimens. Patients with histologic grade III gliomas, incomplete medical records, or who did not undergo surgery for brain tumor at the institution were excluded from this study. Institutional review board approval was obtained for the retrospective chart review. Of note, we did not obtain patient consent for this study since it was retrospective collection of data and many of the patients were deceased at time of data collection. The following information was obtained: date of birth, gender, date of glioma diagnosis, date of first intracranial disease progression, date(s) of craniotomy, treatment with chemotherapy (yes/no), treatment with radiation therapy (yes/no), and neurosurgical and oncologic visit notes to determine seizure as presenting symptom and preoperative Karnofsky Performance (KPS) score. We also retrospectively reviewed tumor pathology, including both histologic and genomic analysis, where possible, to avoid wrong diagnoses. Fluorescence in situ hybridization was used to detect deletion of chromosomes 1p and 19q. Tumors with deletions of only 1p or 19q were grouped with 1p19q non-codeleted tumors. Thirteen patients with tumors who were 1p19q codeleted but did not have IDH status information were grouped into the IDH-mutant category since virtually all 1p/19q codeleted oligodendrogliomas have a mutation in isocitrate dehydrogenase 1 (IDH1) at arginine 132 (R132) or the analogous residue arginine 172 in IDH2 (R172).3,20 In some cases, IDH1 mutation was detected using immunohistochemistry for cytoplasmic staining for R132H-mIDH1 and scored as positive by our institution's neuropathologists. For other cases, further genotyping was performed with SNaPshot genotyping of codon 132 of the IDH1 gene and codon 172 of the IDH2 gene.18 Preoperative imaging was reviewed to determine presence of enhancement, tumor location, and maximum tumor size. Following surgery for diagnosis, some patients received radiotherapy and/or adjuvant chemotherapy. This treatment typically involved fractionated 3-D-conformal radiotherapy and temozolomide, or procarbazine, CCNU (lomustine), and vincristine chemotherapy (PCV). The extent of surgical resection was determined by postoperative imaging. Operative reports were used in the case of planned biopsy-only cases. The definition of gross total resection (GTR) was no evidence of remaining T2/fluid-attenuated inversion recovery (FLAIR) hyperintensty not deemed to be edema in the region of previous abnormality after excision as judged by a neuroradiologist. There were no GTRs in cases where tumor demonstrated enhancement. Subtotal resection (STR) was defined as craniotomy procedure where tumor was debulked but there was residual disease noted by radiologists on postoperative imaging. If tissue was solely obtained for diagnosis without debulking, the procedure was classified as biopsy only. Progression of disease was determined by neuroradiologist interpretation of return or increase in T2/FLAIR hyperintensity not felt to be consistent with edema or treatment effect and/or development or increase of tumor enhancement pre- and postgadolinium contrast brain magnetic resonance imaging (MRI). Data Analysis and Statistical Methods The primary end point in this study was OS, calculated as the time from diagnosis to death. Additionally, the time to imaging progression, calculated as the time between the first surgery and worsened tumor burden on imaging, was evaluated as a secondary measure. Official death records, if not obtained in chart review, were obtained from the Social Security Death Index or by online obituary. At the time of the data analysis, 72 patients (64.9%) were known by recent clinical follow-up to be alive and were appropriately censored in the survival analysis. There were no patients lost to follow-up. If there was missing data, we did not include patients with missing data for that particular analysis. All genomic analyses were only performed for patients with data. Survival analyses were performed using Kaplan–Meier product limit estimators and univariate and bivariate Cox proportional hazards models. For the bivariate analysis, we included age as a continuous variable and IDH mutation status as a binary variable, as these were central to our research question. Given the small number of deaths in the dataset, we chose not to include other variables in order to maximize power in the model. Treatment strategies were compared using chi-square analysis and by logistic regression. Differences in age between IDH populations were compared using t-test. Calculations were performed using SAS (version 9.3; SAS Institute, Cary, North Carolina) and RStudio (Version 1.0.136, ©2009-2016 RStudio Inc, Boston, Massachusetts). RESULTS Clinical Data Among the total 111 cases, the mean age at time of diagnosis was 51.4 yr (Table 1). Median follow-up time was 94 mo. Preoperative tumor size was available for 99 cases with an average maximum diameter of 4.3 cm. Of 108 patients with preoperative exam details, 94% had a KPS >90 preoperatively. Resection data were available for all cases, with 36.9% undergoing biopsy alone, 27.9% undergoing STR, and 35.1% undergoing GTR. TABLE 1. Characteristics of 111 Patients with Supratentorial Low-Grade Gliomas, 73 Patients with Known Isocitrate Dehydrogenase 1 Mutation Status   Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached    Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached  IDH = isocitrate dehydrogenase; KPS = Karnofsky Performance score. Values given as number of patients (%) unless otherwise indicated. View Large TABLE 1. Characteristics of 111 Patients with Supratentorial Low-Grade Gliomas, 73 Patients with Known Isocitrate Dehydrogenase 1 Mutation Status   Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached    Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached  IDH = isocitrate dehydrogenase; KPS = Karnofsky Performance score. Values given as number of patients (%) unless otherwise indicated. View Large Molecular genomic data of tumor specimen were complete in 73 cases, allowing multivariate analysis in that cohort. The demographics for these 73 patients were similar to the total population (Table 1). Sixty-five percent of patients were alive at the time the data for this study were compiled. Outcome Analysis Median OS for all 111 patients was 189 mo (15.75 yr) with 5- and 10-yr survival rates of 84.3% and 67.7%, respectively. By log-rank comparison, there was no significant difference in survival based on histologic pathology, although oligodendroglioma and mixed glioma trended towards higher survival rates compared to patients with astrocytoma (P = .19). Median OS for patients with astrocytomas was 11.5 yr (138 mo), median OS for patients with oligodendrogliomas was 21 yr (252 mo), whereas median OS for patients with oligoastrocytomas had not yet been reached at time of data collection (Table 1). There were 39 cases of progression that led to a second surgery. Eleven of these cases progressed to grade III (28.2%) and 15 (38.5%) progressed to grade IV. Thirteen tumors remained grade II (33.3%). Median progression-free survival for the entire group was 8.2 yr. Molecular Data For patients with IDH mutations (n = 54), OS was significantly increased compared to patients without the mutation (n = 19). Five-year and 10-yr OSs were 94.2% and 85.8%, respectively, for IDH-mutant patients, and 67.4% and 36.8%, respectively, for IDH-wildtype patients (hazard ratio [HR] 0.175 [95% confidence interval 0.07-0.45], P < .001, Figure 1). Five-year and 10-yr progression-free survivals were 77.3% and 56.3%, respectively, for IDH-mutant patients and were 57.4% and 39.4%, respectively, for IDH-wildtype patients (HR 0.56 [0.28-1.15], P = .11). Of IDH-mutant cases, 44.4% progressed, whereas 57.9% progressed of IDH-wildtype cases (Table 2). There was a trend towards older median age in the IDH-wildtype population (53.8 yr) compared to the IDH-mutant population (50.1 yr), although this was not significant (P = .07). FIGURE 1. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant population vs IDH-wildtype population. FIGURE 1. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant population vs IDH-wildtype population. TABLE 2. Overall Survival and Progression-Free Survival of 73 Patients with Supratentorial Low-Grade Gliomas With Known Isocitrate Dehydrogenase 1 Mutation Status All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  IDH = isocitrate dehydrogenase. View Large TABLE 2. Overall Survival and Progression-Free Survival of 73 Patients with Supratentorial Low-Grade Gliomas With Known Isocitrate Dehydrogenase 1 Mutation Status All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  IDH = isocitrate dehydrogenase. View Large By univariate analysis, factors predicting worse OS included patients of older age (HR 1.07 [1.03-1.11], P < .001), patients who presented with seizure (HR 2.46 [1.07-5.67], P = .035), patients with larger tumor size (HR 1.25 [1.03-1.52], P = .027), patients who underwent biopsy rather than an STR or GTR (HR 0.36 [0.16-0.77; STR], HR 0.14 [0.05-0.37; GTR], P < .001), and patients who underwent chemotherapy (HR 3.48 [1.34-9.05], P = .012), and radiation (HR 4.36 [1.63-11.11], P = .003; Table 3). TABLE 3. Univariate Analysis of 111 Patients Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  CI = Confidence interval; HR = hazard ratio; KPS = Karnofsky Performance score; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large TABLE 3. Univariate Analysis of 111 Patients Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  CI = Confidence interval; HR = hazard ratio; KPS = Karnofsky Performance score; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large In a bivariate model with age and IDH status, IDH mutation was the only predictor of survival (HR 0.22 [0.07-0.68], P = .008; Table 4). Bivariate analysis of age and extent of resection showed that age and any surgical resection beyond biopsy predicted significantly improved survival (age HR 1.02 [1.02-1.16], P = .01, STR HR 0.13 [0.04-0.47], P = .002, GTR HR 0.026 [0.003-0.21], P = .0006). Bivariate analysis of IDH mutation status and extent of resection showed that IDH-mutant status and any surgical resection beyond biopsy predicted improved survival (IDH mutant 0.32 [0.11-0.9], P = .03, STR HR 0.19 [0.05-0.69], P = .01, GTR HR 0.04 [0.004-0.29], P = .002). Of these bivariate analyses, age became nonsignificant only in analysis with IDH status. TABLE 4. Bivariate Analyses Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  CI = Confidence interval; HR = hazard ratio; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large TABLE 4. Bivariate Analyses Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  CI = Confidence interval; HR = hazard ratio; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large Logistic regression analysis of extent of resection based on IDH mutation status showed there was significantly higher odds of getting a GTR in tumors that harbored IDH mutations. Conversely, there were significantly higher odds of biopsy in cases of IDH wildtype (OR 7.57, [95% confidence interval 2.59-24.85], P < .001; Table 5, Figure 2). FIGURE 2. View largeDownload slide Odds of extent of surgical resection based on IDH status. FIGURE 2. View largeDownload slide Odds of extent of surgical resection based on IDH status. TABLE 5. Surgical resection rates by IDH status   Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)    Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)  STR = subtotal resection; GTR = gross total resection; IDH = isocitrate dehydrogenase. View Large TABLE 5. Surgical resection rates by IDH status   Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)    Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)  STR = subtotal resection; GTR = gross total resection; IDH = isocitrate dehydrogenase. View Large Within the IDH-mutant population and those with known 1p19q status, 1p19q codeletion occurred in 75.5% of patients. Analysis of patients with 1p19q codeletion (37 patients) compared to 1p19q non-codeleted (12 patients) showed a significantly longer survival in the group with 1p19q codeletion (HR 0.172 [0.04-0.72]; P = .007; Figure 3). FIGURE 3. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant 1p19q codeleted population vs IDH-mutant 1p19q non-codeleted population. FIGURE 3. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant 1p19q codeleted population vs IDH-mutant 1p19q non-codeleted population. DISCUSSION This study represents the largest reported single-institution clinical outcomes of patients with grade II glioma and high-risk criteria of age 40 yr or greater in the postmolecular genomic era. In the management of LGGs, the “Pignatti criteria” have held that patients with age ≥40 yr have worse prognosis, along with patients with tumor size >6 cm, tumor crossing midline, neurological deficit at initial presentation, and astrocytoma histopathology.12 Recently, however, molecular alterations of these tumors, such as IDH mutations and 1p and 19q codeletion, have been found in high numbers of diffuse gliomas, ∼80% to 90%, and significantly influence outcome (Table 6).1 Although tumors harboring these mutations or genomic alterations typically present in the third or fourth decade of life, our study reports on the rarer population of older patients with LGG, and still finds a high IDH mutation rate of 74% and 1p19q codeletion rate of 75.5% in those with known genotype. Although univariate analysis showed older age was a significantly negative prognostic factor for survival, in the subgroup of patients with known IDH status, bivariate survival analysis examining IDH mutation and age determined that IDH-mutant status was prognostic of survival, while age was not. TABLE 6. Outcomes of IDH-Mutant Populations in the Literature Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  View Large TABLE 6. Outcomes of IDH-Mutant Populations in the Literature Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  View Large This finding suggests that molecular pathology of diffuse glioma may be more important than the age of the patient in determining clinical outcome. In the pre-molecular genomic era of the 1990s and 2000s, prognostic factors in LGGs were looked at extensively by both single-institution and multicenter RCTs. In their retrospective examination of prognostic factors of 379 patients with LGGs in 1997, Lote et al concluded that “prognosis in LGG following postoperative radiotherapy seemed largely determined by the inherent biology of the glioma and patient age at diagnosis.”16 In support of this, 20 yr later in 2017, Youland et al21 reported worse outcomes in their elderly group of 55 yr and older, although they did not use any genomic criteria in their analysis of patient outcomes.17 Under the scope of molecular genetics, however, recent studies suggest that the inherent tumor biology seems to trump more traditional prognostic factors, including age. In a 2014 study of patients of all ages with grade II and grade III gliomas, Sabha et al22 demonstrated that molecular data from the tumors was the only significant predictor of patient progression-free and OS. IDH mutation status, 1p/19q codeletion, and PTEN deletion were predictive of over overall survival in a multivariate model, while none of the clinical or demographic factors such as age, tumor size, performance status, or tumor histology were found to be predictive. More recently, Hayashi et al23 reported on 57 patients with 1p/19q codeleted gliomas and found that the only significant predictors of poor outcome were a gain of chromosome 19p and grade III histology. They found no significant difference in the OS of the patients with respect to age (≥40 vs <40 yr), degree of resection, maximum tumor diameter (≥5 vs <5 cm), histological subtype, and MGMT promoter methylation status. These results support our findings that traditional prognostic factors, such as age, may be less important in the setting of molecular markers. Older patients may not inherently have worse prognosis based on age, but rather, as our data suggest, due to a slightly lower chance of having an IDH mutation. Just as TERT promoter mutation frequency increases with patient age at diagnosis, it is possible that IDH mutation frequency decreases with patient age.24 While our study did not show a significant difference in IDH mutation rate by age, it did occur at a lower rate than in studies of LGGs including younger ages (68% vs up to 82%-91%, Table 6). In 2013, Gorlia et al validated new prognostic models from the European Organisation for Research and Treatment of Cancer (EORTC) by studying patient outcomes from tumors that had undergone central pathologic review rather than by local pathologist.25 Even more strikingly, in 2015 The Cancer Genome Atlas Research Network (TCGA) redefined glioma survival rates by molecular genomic classification rather than histologic classification.3 Along these lines, future outcomes analyses from large RCTs that include the molecular genomic information of tumors may allow re-examination and evolution of LGG prognostic factors. Extent of tumor resection is another common prognostic factor, with some large RCTs of gliomas, such as the EORTC 22844, finding significantly improved OS with increasing tumor resection rates.26 Although extent of surgical resection was highly significant on univariate analysis in our study, our study was limited to only bivariate models due to the small sample size of patients with known IDH status and limited number of events (deaths) per variable (n = 73, 18 deaths).27 Of these multiple bivariate analyses, the analysis of extent of resection and IDH status showed that both were significantly associated with survival. In logistic regression analysis, we found that IDH-mutant status and extent of surgical resection were highly correlated, with IDH-mutant status predicting a 7.5 times increased odds of a GTR than STR. This may be due to IDH-mutant tumors occurring in greater frequency frontal and temporal lobes, which, especially when right sided, lend themselves to safer maximum extent of resection. These findings have been similarly highlighted by other recent series. Youland et al21 found in that older patients with gliomas benefited from greater extent of resection, yet they did not have correlative mutation status of these patients.17 In studies of higher grade gliomas, Beiko et al28 were the first group to report on IDH1 mutation as an independent predictor of complete resection of enhancing disease, with 93% complete resections among mutants vs 67% among wildtype (P < .001). They further suggested that, given the decrease in IDH mutation after age 40, factoring in the clinical feature of age into risk stratification was an “incomplete surrogacy” for IDH status.28 Among grade III gliomas, Kawaguchi et al29 found that, while all patients undergoing GTR had longer median survival and progression-free survival times compared to those of non-GTR patients, the benefits of GTR was highest for the group with the IDH mutation without a 1p19q codeletion. The IDH mutant, 1p19q codeleted tumors, and the IDH-wildtype tumors did not show survival difference based on GTR vs non-GTR. In a study of 151 grade II gliomas, Leeper et al30 found no statistically significant difference in extent of resection achieved between any of the TCGA-defined molecular subgroups; however, only 9% of their population was IDH wildtype. While our study was not powered for this subgroup analysis, the role of resection and change in tumor volume with surgery based on molecular diagnosis is an important area of future research, with efforts already underway to allow for intraoperative determination of IDH status.31 Our paper did not find that adjuvant chemotherapy or radiation offered significant survival benefit, which differs from several recent phase II and phase III studies.32-34 This may be because our study looks at patients who already have at least 1 high-risk criteria and, therefore, our population, as a whole, may have tended to receive more aggressive treatment at baseline. Only 18.9% of our population was observed rather than receiving some form of postoperative chemotherapy or radiation. Similarly, in a study of 94 patients with LGG who were 55 yr and older treated at Mayo Clinic, Youland et al19,21 also found extent of surgical resection to be the only treatment intervention to significantly affect outcome. Neither radiation alone, chemotherapy alone, or combined chemoradiation had a survival impact, and their rate of postoperative observation was slightly higher than ours at 21%.19,21 To date, there have already been 2 phase III chemoradiation trials looking at treatment outcomes of LGG according to molecular diagnosis, and these include age ≥40 as a high-risk factor. In the EORTC 22033-26033, study of temozolomide chemotherapy vs radiotherapy in high-risk patients with LGG, subgroup analysis determined that IDH mutant, non-codeleted tumors had longer progression-free survival with radiation alone compared to temozolomide alone.32 However, patients with IDH mutant, 1p19q codeleted tumors or IDH-wildtype tumors showed no treatment-related difference in progression-free survival. Additionally, Buckner et al34 reported on high-risk patient outcomes after randomization to radiation alone vs PCV and radiation and found that patients with IDH-mutant tumors had significantly improved progression-free and overall survival compared to the IDH-wildtype cohort. More impressively, the IDH-mutant cohort treated with PCV and radiation had significantly improved survival over the IDH-mutant cohort treated with radiation alone. On their multivariate analysis for OS, age <40 was one of the significant factors indicating improved OS, while IDH-mutant status was not. Other positive prognostic factors included combined radiation and chemotherapy treatment and oligodenroglioma histology. These same variables were also found to significant influence improved progression-free survival, with the inclusion of IDH mutation status as an additional significant factor. Thus, while age still bore out to be influential in survival, similar to our findings, IDH mutation status predicted better survival than IDH-wildtype patients and had a significant influence on outcomes. Looking forward, these randomized trials will be very powerful for defining the role of molecular diagnosis. Study Limitations There are many inherent limitations of our study given its retrospective, chart-review design. Treatments were determined based on a variety of factors, including patient risk factors, patient or physician preference, and time period in which the patient was treated. Given the body of literature of the 1990s and 2000s that supported older age as a risk factor for worse outcomes, it is likely that our patients were treated more aggressively. In this population, adjuvant treatment of radiation and chemotherapy were not assigned randomly but chosen by clinicians based on perceived risk. Therefore, our results of worse outcome associated with adjuvant treatment more likely indicates that patients deemed more high risk for worse outcome were more likely to receive this therapy. The findings of our study were also diminished by the fact that we used the 2007 WHO classification of gliomas rather than the 2017 WHO grading scheme, mainly because most of our patients were diagnosed and treated based on 2007 classification. This renders survival based on histopathology less meaningful, as oligoastrocytoma no longer exists. Future analysis of molecular diagnosis and treatment outcomes across all ages is warranted, especially given recent findings from larger phase III trials with molecular genomic data showing that age <40 still has positive prognostic implications. Additionally, surgical treatment perhaps varied by surgeon and this could influence rates of resection. Furthermore, the extent of resection data based on MRI analysis was simplified into three tiers, and the size of tumors was recorded based on largest in a single dimension. We did not account for volume of T2/FLAIR hyperintense tumor or enhancing tumor, both pre- and postoperatively, which would be more sensitive and accurate to account for residual tumor burden and extent of resection. While we did look at Pignatti criteria data for this study, we limited ourselves to KPS as proxy for neurological deficit and only had two patients with tumor crossing midline so excluded these data from analysis. While size of tumor and seizure at diagnosis were significant on univariate analysis, we did not include this in our final models due to previously discussed limitations of our multivariate survival analyses. Astrocytic pathology trended towards predicting a worse survival, but not significantly so. As for molecular data, we were limited to only 73 of our 111 patients, limiting the robustness of our survival analyses since there were only 18 deaths in this population. We also did not capture large enough numbers of other molecular and genomic alterations to account for their influence in this study, such as p53 mutation, PTEN mutation, and TERT mutation. We did not have the age group of <40 for outcomes comparison, which is a significant limitation to this paper and can be reported on in upcoming analyses from our group. CONCLUSION The present study shows that IDH mutation and 1p19q chromosomal codeletion of LGG in patients ≥40 occurs at high rates like the younger population and predicts a similar survival advantage, and this should be considered when determining optimal management strategy in this population. Tumors harboring IDH mutations are known to present in more surgically accessible areas, and, as found in this study and many previous, maximizing extent of resection seems to hold a withstanding influence on improved survival. Prior analyses indicating older age as a poor prognostic indicator could have associated with IDH mutations occurring at slightly lower frequency in this population. Given that current and future studies include the molecular genetics of gliomas, it is likely time that the 15-yr-old 5 clinical “Pignatti criteria” give way to redefined prognostic factors that reflect the molecular era and, as such, more accurately predict outcome and treatment response. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. Notes Some of this work was presented previously at the Congress of Neurological Surgeons annual meeting in Boston, Massachusetts, as an Oral Poster Abstract on October 21, 2014. REFERENCES 1. Hartmann C, Meyer J, Balss J et al.   Type and frequency of IDH1 and IDH2 mutations are related to astrocytic and oligodendroglial differentiation and age: a study of 1,010 diffuse gliomas. Acta Neuropathol . 2009; 118( 4): 469- 474. Google Scholar CrossRef Search ADS PubMed  2. Louis DN, Perry A, Reifenberger G et al.   The 2016 world health organization classification of tumors of the central nervous system: a summary. Acta Neuropathol . 2016; 131( 6): 803- 820. Google Scholar CrossRef Search ADS PubMed  3. Cancer Genome Atlas Research Network. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. N Engl J Med . 2015; 372( 26): 2481- 2498. CrossRef Search ADS PubMed  4. Jones PS, Dunn GP, Barker FG, Curry WT, Hochberg FH, Cahill DP. Molecular genetics of low-grade gliomas: genomic alterations guiding diagnosis and therapeutic intervention. 11th Annual Frye-Halloran Brain Tumor Symposium. Neurosurg Focus . 2013; 34( 2): E9. Google Scholar CrossRef Search ADS PubMed  5. Chen R, Ravindra VM, Cohen AL et al.   Molecular features assisting in diagnosis, surgery, and treatment decision making in low-grade gliomas. Neurosurg Focus . 2015; 38( 3): E2. Google Scholar CrossRef Search ADS PubMed  6. Buckner J, Giannini C, Eckel-Passow J et al.   Management of diffuse low-grade gliomas in adults - use of molecular diagnostics. Nat Rev Neurol . 2017; 13( 6): 340- 351. Google Scholar CrossRef Search ADS PubMed  7. Delgado-López PD, Corrales-García EM, Martino J, Lastra-Aras E, Dueñas-Polo MT. Diffuse low-grade glioma: a review on the new molecular classification, natural history and current management strategies. Clin Transl Oncol . 2017; 19( 8): 931- 944. Google Scholar CrossRef Search ADS PubMed  8. Shaw E, Arusell R, Scheithauer B et al.   Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a north central cancer treatment Group/Radiation therapy oncology Group/Eastern cooperative oncology group study. J Clin Oncol . 2002; 20( 9): 2267- 2276. Google Scholar CrossRef Search ADS PubMed  9. Bauman G, Lote K, Larson D et al.   Pretreatment factors predict overall survival for patients with low-grade glioma: a recursive partitioning analysis. Int J Radiat Oncol Biol Phys . 1999; 45( 4): 923- 929. Google Scholar CrossRef Search ADS PubMed  10. Nicolato A, Gerosa MA, Fina P, Iuzzolino P, Giorgiutti F, Bricolo A. Prognostic factors in low-grade supratentorial astrocytomas: a uni-multivariate statistical analysis in 76 surgically treated adult patients. Surg Neurol . 1995; 44( 3): 208- 223. Google Scholar CrossRef Search ADS PubMed  11. Leighton C, Fisher B, Bauman G et al.   Supratentorial low-grade glioma in adults: an analysis of prognostic factors and timing of radiation. J Clin Oncol . 1997; 15( 4): 1294- 1301. Google Scholar CrossRef Search ADS PubMed  12. Pignatti F, Van den Bent M, Curran D et al.   Prognostic factors for survival in adult patients with cerebral low-grade glioma. J Clin Oncol . 2002; 20( 8): 2076- 2084. Google Scholar CrossRef Search ADS PubMed  13. Gallego Perez-Larraya J, Delattre J-Y. Management of elderly patients with gliomas. Oncologist . 2014; 19( 12): 1258- 1267. Google Scholar CrossRef Search ADS PubMed  14. Lai A, Kharbanda S, Pope WB et al.   Evidence for sequenced molecular evolution of IDH1 mutant glioblastoma from a distinct cell of origin. J Clin Oncol . 2011; 29( 34): 4482- 4490. Google Scholar CrossRef Search ADS PubMed  15. Hartmann C, Hentschel B, Wick W et al.   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 . 2010; 120( 6): 707- 718. Google Scholar CrossRef Search ADS PubMed  16. Lote K, Egeland T, Hager B et al.   Survival, prognostic factors, and therapeutic efficacy in low-grade glioma: a retrospective study in 379 patients. J Clin Oncol . 1997; 15( 9): 3129- 3140. Google Scholar CrossRef Search ADS PubMed  17. Schomas DA, Laack NN, Brown PD. Low-grade gliomas in older patients. Cancer . 2009; 115( 17): 3969- 3978. Google Scholar CrossRef Search ADS PubMed  18. Chi AS, Batchelor TT, Dias-Santagata D et al.   Prospective, high-throughput molecular profiling of human gliomas. J Neurooncol . 2012; 110( 1): 89- 98. Google Scholar CrossRef Search ADS PubMed  19. Youland RS, Schomas DA, Brown PD et al.   Changes in presentation, treatment, and outcomes of adult low-grade gliomas over the past fifty years. Neuro Oncol . 2013; 15( 8): 1102- 1110. Google Scholar CrossRef Search ADS PubMed  20. Sanson M, Marie Y, Paris S et al.   Isocitrate dehydrogenase 1 codon 132 mutation is an important prognostic biomarker in gliomas. J Clin Oncol . 2009; 27( 25): 4150- 4154. Google Scholar CrossRef Search ADS PubMed  21. Youland RS, Schomas DA, Brown PD, Parney IF, Laack NNI. Patterns of care and treatment outcomes in older adults with low grade glioma: a 50-year experience. J Neurooncol . 2017; 133( 2): 339- 346. Google Scholar CrossRef Search ADS PubMed  22. Sabha N, Knobbe CB, Maganti M et al.   Analysis of IDH mutation, 1p/19q deletion, and PTEN loss delineates prognosis in clinical low-grade diffuse gliomas. Neuro Oncol . 2014; 16( 7): 914- 923. Google Scholar CrossRef Search ADS PubMed  23. Hayashi S, Kitamura Y, Hirose Y, Yoshida K, Sasaki H. Molecular–Genetic and clinicopathological prognostic factors in patients with gliomas showing total 1p19q loss: gain of chromosome 19p and histological grade III negatively correlate with patient's prognosis. J Neurooncol . 2017; 132( 1): 119- 126. Google Scholar CrossRef Search ADS PubMed  24. Spiegl-Kreinecker S, Lötsch D, Ghanim B et al.   Prognostic quality of activating TERT promoter mutations in glioblastoma: interaction with the rs2853669 polymorphism and patient age at diagnosis. Neuro Oncol . 2015; 17( 9): 1231- 1240. Google Scholar CrossRef Search ADS PubMed  25. Gorlia T, Wu W, Wang M et al.   New validated prognostic models and prognostic calculators in patients with low-grade gliomas diagnosed by central pathology review: a pooled analysis of EORTC/RTOG/NCCTG phase III clinical trials. Neuro Oncol . 2013; 15( 11): 1568- 1579. Google Scholar CrossRef Search ADS PubMed  26. Karim AB, Maat B, Hatlevoll R et al.   A randomized trial on dose-response in radiation therapy of low-grade cerebral glioma: European Organization for Research and Treatment of Cancer (EORTC) Study 22844. Int J Radiat Oncol Biol Phys . 1996; 36( 3): 549- 556. Google Scholar CrossRef Search ADS PubMed  27. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol . 1996; 49( 12): 1373- 1379. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8970487. Accessed February 7, 2018. Google Scholar CrossRef Search ADS PubMed  28. Beiko J, Suki D, Hess KR et al.   IDH1 mutant malignant astrocytomas are more amenable to surgical resection and have a survival benefit associated with maximal surgical resection. Neuro Oncol . 2014; 16( 1): 81- 91. Google Scholar CrossRef Search ADS PubMed  29. Kawaguchi T, Sonoda Y, Shibahara I et al.   Impact of gross total resection in patients with WHO grade III glioma harboring the IDH 1/2 mutation without the 1p/19q co-deletion. J Neurooncol . 2016; 129( 3): 505- 514. Google Scholar CrossRef Search ADS PubMed  30. Leeper HE, Caron AA, Decker PA, Jenkins RB, Lachance DH, Giannini C. IDH mutation, 1p19q codeletion and ATRX loss in WHO grade II gliomas. Oncotarget . 2015; 6( 30): 30295- 30305. Google Scholar CrossRef Search ADS PubMed  31. Shankar GM, Francis JM, Rinne ML et al.   Rapid intraoperative molecular characterization of glioma. JAMA Oncol . 2015; 1( 5): 662. Google Scholar CrossRef Search ADS PubMed  32. Baumert BG, Hegi ME, van den Bent MJ et al.   Temozolomide chemotherapy versus radiotherapy in high-risk low-grade glioma (EORTC 22033-26033): a randomised, open-label, phase 3 intergroup study. Lancet Oncol . 2016; 17( 11): 1521- 1532. Google Scholar CrossRef Search ADS PubMed  33. Fisher BJ, Hu C, Macdonald DR et al.   Phase 2 study of temozolomide-based chemoradiation therapy for high-risk low-grade gliomas: preliminary results of radiation therapy oncology Group 0424. Int J Radiat Oncol Biol Phys . 2015; 91( 3): 497- 504. Google Scholar CrossRef Search ADS PubMed  34. Buckner JC, Shaw EG, Pugh SL et al.   Radiation plus procarbazine, CCNU, and vincristine in low-grade glioma. N Engl J Med . 2016; 374( 14): 1344- 1355. Google Scholar CrossRef Search ADS PubMed  Neurosurgery Speaks! Audio abstracts available for this article at www.neurosurgery-online.com. COMMENTS The authors examine the question of whether “mutations” drive outcomes of older patients with low-grade gliomas. This is an interesting study but the results are limited by the retrospective nature of the study and patient selecton bias demonstrated by facts such as those of chemotherapy and radiation being associated with worse outcome. Some tumors were somehow deemed more aggressive and were treated with these modalities while others were not. It seems that this is a chicken and egg question. Furthermore, the findings of this study are diminished by the fact that with the new 2017 WHO grading scheme all grade II tumors require IDH-1 mutation and the oligodendrogliomas additionally require 1p19q deletion. Additionally, since oligoastrocytoma no longer exists in this classification scheme, so any discussion of this entity is unfruitful. Randy L. Jensen Salt Lake City, UT Maximal safe surgical resection of a presumed supratentorial diffuse low-grade glioma (LGG) is considered standard of care.1 This provides tumor tissue for histological and genetic analysis, reduces tumor burden and subsequent risk of tumor progression, and may reduce seizures associated with the tumor – it may also be associated with improved survival.1 Recent reports of the beneficial effect of postoperative radiation and chemotherapy in select patients were largely based on criteria proposed by Pignatti et al2 more than a decade ago.3,4 They proposed 5 criteria that portended a worse prognosis for adult patients with a supratentorial LGG; age type = "Other" >40 years, astrocytoma histology, tumor diameter >6 cm, tumors crossing the midline, and the presence of a neurological deficit before surgery.2 A critical factor was age; 40 years was considered the bellwether point for prognostication and treatment decision-making.2-4 The dawn of the new era of molecular medicine changed our philosophy on the management of LGG. In 2016, the World Health Organization (WHO) reclassified neuroepithelial tumors based on molecular markers such as 1p/19q and isocitrate dehydrogenase (IDH) mutations.5 Currently, these markers supercede histological criteria for tumors such as an oligodendroglioma and some categories such as oligoastrocytomas have been altogether culled.5 Overall, it has had a constructive effect on our understanding of gliomas. Management of these histologically and genetically diverse tumors is now predicated by the tumor molecular profile and henceforth, it is inconceivable that any clinical trial will be undertaken without consideration of these biomarkers. The p53 mutation (for astrocytomas) and 1p/19q deletion (for oligodendrogliomas) are pathognomic and mutually exclusive.5 A recent addition to this select list was isocitrate dehydrogenase (IDH) mutations.6 IDH mutations are reported in hematological malignancies such as acute myeloid leukemia but among solid tumors, they are particularly unique to gliomas. IDH mutations may occur as both IDH 1 and IDH 2 mutations. IDH 1 mutations appear to be more relevant and may be detected by standard immunohistochemical studies that are ubiquitously available – on the other hand, detection of IDH 2 mutations requires genomic sequencing studies. Both IDH mutations tend to occur early in glioma development and may be seen with either p53 mutations or 1p/19q deletions. They contribute to glioma progression through induction of the HIF-1 pathway and genome-wide histone and DNA methylation alterations through the oncometabolite 2-hydroxyglutarate.1,6,7 IDH mutations also alter the glioma microenvironment by affecting collagen maturation and impairing basement membrane function.7 They may predispose a patient to seizures or portend worse postoperative seizure occurrence. IDH 1 mutations are commonly seen in LGG and secondary glioblastoma (GBM) but rarely ever in primary GBM or malignant gliomas with EGFR amplification.7 They frequently accompany MGMT promoter hypermethylation and correlate with the Ki-67 labeling index.8,9 They are, hence, a major marker for OS in LGG and malignant gliomas – and are possibly more powerful than the Pignatti score in their predictive capacity.8,9 This study goes back to the tipping point effect of age viewed in light of this new molecular information. In LGG patients >40 years of age with IDH 1 mutations, the authors report improved survival; the presence of a 1p/19q deletion had a similar beneficial effect. Also, greater surgical resection was predictive of survival – and extent of resection significantly correlated with IDH mutation status. Their findings confirm previous reports of the salutary effect of IDH1 mutations in gliomas and the value of maximal safe surgical resection. It proves the need to use molecular markers to determine the optimal management of a LGG, regardless of age. It may be that age is a secondary factor, simply reflecting the frequency of occurrence of these changes. In other words, younger patients tend to have better outcomes because they have IDH 1 mutated tumors more frequently. Older patients who have the same molecular changes may fare the same. The characterization of genetic findings such as IDH mutations, p53 mutations, and 1p/19q deletions has revolutionized our understanding of diffuse LGG. It has led to a more precise pathological classification, the potential for better therapies, and improved prognostication. But this may simply be the proverbial “tip of the iceberg”– in addition to prognostication, there may be therapeutic benefits. In acute myeloid leukemia, IDH inhibition appears to the clinically effective – possibly by allowing malignant blasts cells to differentiate and become post mitotic. There may be similar value in the future with IDH inhibitors for gliomas. IDH1-R132H mutations are reportedly associated with a more severe phenotype of postoperative epilepsy – this suggests potential for an antiepileptic therapeutic effect of an IDH 1 inhibitor.10 This is a nice study that improves our understanding of the impact of genomic alterations on the clinical trajectory of a glioma in older patients. It provides clinicians useful information when counseling these patients. Such findings may alter our management of these patients significantly, causing us to look anew at older criteria used to determine patients who are at higher risk for tumor progression. The use of radiation and chemotherapy is not without side effects – molecular genomic information may lead to a more abstemious use of these modalities or at least a more targeted individualized approach. Vikram C. Prabhu Kevin Barton Ewa Borys Edward Melian Maywood, IL 1. Buckner J Giannini C Eckel-Passow J Lachance D Parney I Laack N Jenkins R. Management of diffuse low-grade gliomas in adults - use of molecular diagnostics. Nat Rev Neurol . 2017; 13( 6): 340- 351. Google Scholar CrossRef Search ADS PubMed  2. Pignatti F van den Bent M Curran D Debruyne C Sylvester R Therasse P Afra D Cornu P Bolla M Vecht C Karim AB; European Organization for Research and Treatment of Cancer Brain Tumor Cooperative Group; European Organization for Research and Treatment of Cancer Radiotherapy Cooperative Group. Prognostic factors for survival in adult patients with cerebral low-grade glioma. J Clin Oncol . 2002; 20( 8): 2076- 84. Google Scholar CrossRef Search ADS PubMed  3. Shaw EG Wang M Coons SW Brachman DG Buckner JC Stelzer KJ Barger GR Brown PD Gilbert MR Mehta MP. Randomized trial of radiation therapy plus procarbazine, lomustine, and vincristine chemotherapy for supratentorial adult low-grade glioma: initial results of RTOG 9802. J Clin Oncol . 2012; 30( 25): 3065- 70. Google Scholar CrossRef Search ADS PubMed  4. Youland RS Kreofsky CR Schomas DA Brown PD Buckner JC Laack NN. The impact of adjuvant therapy for patients with high-risk diffuse WHO grade II glioma. J Neurooncol . 2017; 135( 3): 535- 543. Google Scholar CrossRef Search ADS PubMed  5. Komori T. The 2016 WHO Classification of Tumours of the Central Nervous System: The Major Points of Revision. Neurol Med Chir (Tokyo) . 2017; 57( 7): 301- 311. Google Scholar CrossRef Search ADS PubMed  6. Parsons DW Jones S Zhang X Lin JC Leary RJ Angenendt P Mankoo P Carter H Siu IM Gallia GL Olivi A McLendon R Rasheed BA Keir S Nikolskaya T Nikolsky Y Busam DA Tekleab H Diaz LA Jr Hartigan J Smith DR Strausberg RL Marie SK Shinjo SM Yan H Riggins GJ Bigner DD Karchin R Papadopoulos N Parmigiani G Vogelstein B Velculescu VE Kinzler KW. An integrated genomic analysis of human glioblastoma multiforme. Science  2008; 321( 5897): 1807- 1812. Google Scholar CrossRef Search ADS PubMed  7. Zhang C Moore LM Li X Yung WK Zhang W. IDH1/2 mutations target a key hallmark of cancer by deregulating cellular metabolism in glioma. Neuro Oncol . 2013; 15( 9): 1114- 26. Google Scholar CrossRef Search ADS PubMed  8. Etxaniz O Carrato C de Aguirre I Queralt C Muñoz A Ramirez JL Rosell R Villà S Diaz R Estival A Teixidor P Indacochea A Ahjal S Vilà L Balañá C. IDH mutation status trumps the Pignatti risk score as a prognostic marker in low-grade gliomas. J Neurooncol . 2017 Nov; 135( 2): 273- 284. Google Scholar CrossRef Search ADS PubMed  9. Zeng A Hu Q Liu Y Wang Z Cui X Li R Yan W You Y. IDH1/2 mutation status combined with Ki67 labeling index defines distinct prong-ostic groups in glioma. Oncotarget  2015; 6( 30):30  232- 8. Google Scholar CrossRef Search ADS   10. Neal A Kwan P O’Brien TJ Buckland ME Gonzales M Morokoff A. IDH1 and IDH2 mutations in postoperative diffuse glioma-associated epilepsy. Epilepsy Behav . 2018; 78: 30- 36. Google Scholar CrossRef Search ADS PubMed  Neurosurgery Speaks (Audio Abstracts) Listen to audio translations of this paper's abstract into select languages by choosing from one of the selections below. English: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France English: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France Close French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France Close Chinese: Lin Song, MD. Department of Neurosurgery Beijing Tiantan Hospital Capital Medical University Beijing, China Chinese: Lin Song, MD. Department of Neurosurgery Beijing Tiantan Hospital Capital Medical University Beijing, China Close Italian: Alfredo Conti, MD, PhD. Department of Neurosurgery Department of Neurosurgery Charité Universitätsmedizin Berlin, Germany Italian: Alfredo Conti, MD, PhD. Department of Neurosurgery Department of Neurosurgery Charité Universitätsmedizin Berlin, Germany Close Russian: Sergei Kim. Department of Pediatric Neurosurgery Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Russian: Sergei Kim. Department of Pediatric Neurosurgery Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Close Japanese: Ryu Kurokawa, MD. Department of Neurosurgery Dokkyo University Hospital Tochigi, Japan Japanese: Ryu Kurokawa, MD. Department of Neurosurgery Dokkyo University Hospital Tochigi, Japan Close Korean: Sun Ha Paek, MD, PhD. Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Korean: Sun Ha Paek, MD, PhD. Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Close Portuguese: Eduardo Carvalhal Ribas, MD. Neurosurgery Department Hospital das Clínicas University of São Paulo Medicine School (HC-FMUSP) Hospital Israelita Albert Einstein São Paulo, Brazil Portuguese: Eduardo Carvalhal Ribas, MD. Neurosurgery Department Hospital das Clínicas University of São Paulo Medicine School (HC-FMUSP) Hospital Israelita Albert Einstein São Paulo, Brazil Close Greek: Alexiou A Georgios, MD. Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Greek: Alexiou A Georgios, MD. Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Close Spanish: Francisco Alberto Mannará, MD. Department of Neurosurgery Fernández Hospital Buenos Aires City, Argentina Spanish: Francisco Alberto Mannará, MD. Department of Neurosurgery Fernández Hospital Buenos Aires City, Argentina Close Copyright © 2018 by the Congress of Neurological Surgeons This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Neurosurgery Oxford University Press

Isocitrate Dehydrogenase Mutations in Low-Grade Gliomas Correlate With Prolonged Overall Survival in Older Patients

Loading next page...
 
/lp/ou_press/isocitrate-dehydrogenase-mutations-in-low-grade-gliomas-correlate-with-Dekqf62CYC
Publisher
Congress of Neurological Surgeons
Copyright
Copyright © 2018 by the Congress of Neurological Surgeons
ISSN
0148-396X
eISSN
1524-4040
D.O.I.
10.1093/neuros/nyy149
Publisher site
See Article on Publisher Site

Abstract

Abstract BACKGROUND Older age has been associated with worse outcomes in low-grade gliomas (LGGs). Given their rarity in the older population, determining optimal treatment plans and patient outcomes remains difficult. OBJECTIVE To retrospectively study LGG survival outcomes in an older population stratified by molecular genetic profiles. METHODS We included patients age ≥40 yr with pathologically confirmed World Health Organization grade II gliomas treated at a single institution between 1995 and 2015. We collected tumor genomic information when available. RESULTS Median overall survival for the entire group (n = 111, median age 51 yr, range 40-77 yr) was 15.75 yr with 5- and 10-yr survival rates of 84.3% and 67.7%, respectively. On univariate analysis, patients with isocitrate dehydrogenase (IDH) mutation had significantly increased survival compared to IDH wildtype (hazard ratio [HR] 0.17 [0.07-0.45], P < .001). Older age, seizure at presentation, larger tumor size, IDH wildtype, biopsy only, chemotherapy, and radiation were significantly associated with shorter survival based on univariate analyses. In patients with known IDH status (n = 73), bivariate analysis of IDH mutation status and age showed only IDH status significantly influenced overall survival (HR 0.22 [0.07-0.68], P = .008). Greater surgical resection was predictive of survival, although extent of resection significantly correlated with IDH mutation status (odds ratio 7.5; P < .001). CONCLUSION We show that genomic alterations in LGG patients ≥40 occur at high rates like the younger population and predict a similar survival advantage. Maximizing surgical resection may have survival benefit, although feasibility of resection is often linked to IDH status. Given the importance of molecular genetics, a redefinition of prognostic factors associated with these tumors is likely to emerge. Glioma outcomes, IDH, Low-grade gliomas ABBREVIATIONS ABBREVIATIONS EORTC European Organisation for Research and Treatment of Cancer FLAIR fluid-attenuated inversion recovery GTR gross total resection HR hazard ratio IDH isocitrate dehydrogenase KPS Karnofsky Performance score LGG low-grade glioma MRI magnetic resonance imaging OS overall survival PCV procarbazine, CCNU (lomustine), and vincristine chemotherapy RCT randomized controlled trial STR subtotal resection WHO World Health Organization Diffuse low-grade gliomas (LGGs) are typically encountered in younger patient populations and hold a favorable, yet highly variable, prognosis. Knowledge of the evolution and growth of these tumors has expanded greatly in recent decades, largely due to the discovery and characterization of genetic events such as IDH mutations and 1p19q codeletion.1-4 These molecular diagnostic insights have led to improved prognostication, a change in the pathological classification of all gliomas, and are now incorporated into clinical trial design and analysis.5-7 Prior to discovery and incorporation of these molecular and cytogenetic features, retrospective database studies and randomized controlled trials (RCTs) indicated that LGGs present more aggressively and have worse outcomes when they occur in patients older than 40.8-11 By analyzing some of these RCTs, a 2002 paper by Pignatti et al12 established 5 criteria that indicate poor prognosis for patients with LGGs: age >40 yr, astrocytoma histology subtype, largest diameter of the tumor >6 cm, tumor crossing the midline, and presence of neurological deficit before surgery. While older age remains a risk factor, it has been difficult to study the true impact of age in LGGs because of the relative rarity of these tumors in older patients. As such, predicting outcome and determining optimal treatment plans of surgery, radiation, and chemotherapy remains a challenge.13 Particularly, since prognostically favorable IDH mutations are more common in patients younger than 40 yr old, absence of molecular and genetic data in these past trials limits interpretation of whether patient outcomes are truly dependent on patient age or simply a worse genomic profile.14,15 In this study, we present outcomes from the largest reported single-institution population of patients with grade II glioma and high-risk criteria of age 40 yr or greater. Including IDH mutation and 1p19q codeletion status, we examined whether overall survival (OS) within an older population is influenced by molecular markers.1,8-12,16-19 METHODS Patient Data Acquisition We performed a retrospective review for all patients who received surgery for diagnosis and/or resection of a grade II glioma by World Health Organization (WHO) 2007 criteria who were ≥40 yr old at the time of diagnosis from January 1, 1995 to December 31, 2015 at a single institution. Neuropathologists determined tumor grade from operative specimens. Patients with histologic grade III gliomas, incomplete medical records, or who did not undergo surgery for brain tumor at the institution were excluded from this study. Institutional review board approval was obtained for the retrospective chart review. Of note, we did not obtain patient consent for this study since it was retrospective collection of data and many of the patients were deceased at time of data collection. The following information was obtained: date of birth, gender, date of glioma diagnosis, date of first intracranial disease progression, date(s) of craniotomy, treatment with chemotherapy (yes/no), treatment with radiation therapy (yes/no), and neurosurgical and oncologic visit notes to determine seizure as presenting symptom and preoperative Karnofsky Performance (KPS) score. We also retrospectively reviewed tumor pathology, including both histologic and genomic analysis, where possible, to avoid wrong diagnoses. Fluorescence in situ hybridization was used to detect deletion of chromosomes 1p and 19q. Tumors with deletions of only 1p or 19q were grouped with 1p19q non-codeleted tumors. Thirteen patients with tumors who were 1p19q codeleted but did not have IDH status information were grouped into the IDH-mutant category since virtually all 1p/19q codeleted oligodendrogliomas have a mutation in isocitrate dehydrogenase 1 (IDH1) at arginine 132 (R132) or the analogous residue arginine 172 in IDH2 (R172).3,20 In some cases, IDH1 mutation was detected using immunohistochemistry for cytoplasmic staining for R132H-mIDH1 and scored as positive by our institution's neuropathologists. For other cases, further genotyping was performed with SNaPshot genotyping of codon 132 of the IDH1 gene and codon 172 of the IDH2 gene.18 Preoperative imaging was reviewed to determine presence of enhancement, tumor location, and maximum tumor size. Following surgery for diagnosis, some patients received radiotherapy and/or adjuvant chemotherapy. This treatment typically involved fractionated 3-D-conformal radiotherapy and temozolomide, or procarbazine, CCNU (lomustine), and vincristine chemotherapy (PCV). The extent of surgical resection was determined by postoperative imaging. Operative reports were used in the case of planned biopsy-only cases. The definition of gross total resection (GTR) was no evidence of remaining T2/fluid-attenuated inversion recovery (FLAIR) hyperintensty not deemed to be edema in the region of previous abnormality after excision as judged by a neuroradiologist. There were no GTRs in cases where tumor demonstrated enhancement. Subtotal resection (STR) was defined as craniotomy procedure where tumor was debulked but there was residual disease noted by radiologists on postoperative imaging. If tissue was solely obtained for diagnosis without debulking, the procedure was classified as biopsy only. Progression of disease was determined by neuroradiologist interpretation of return or increase in T2/FLAIR hyperintensity not felt to be consistent with edema or treatment effect and/or development or increase of tumor enhancement pre- and postgadolinium contrast brain magnetic resonance imaging (MRI). Data Analysis and Statistical Methods The primary end point in this study was OS, calculated as the time from diagnosis to death. Additionally, the time to imaging progression, calculated as the time between the first surgery and worsened tumor burden on imaging, was evaluated as a secondary measure. Official death records, if not obtained in chart review, were obtained from the Social Security Death Index or by online obituary. At the time of the data analysis, 72 patients (64.9%) were known by recent clinical follow-up to be alive and were appropriately censored in the survival analysis. There were no patients lost to follow-up. If there was missing data, we did not include patients with missing data for that particular analysis. All genomic analyses were only performed for patients with data. Survival analyses were performed using Kaplan–Meier product limit estimators and univariate and bivariate Cox proportional hazards models. For the bivariate analysis, we included age as a continuous variable and IDH mutation status as a binary variable, as these were central to our research question. Given the small number of deaths in the dataset, we chose not to include other variables in order to maximize power in the model. Treatment strategies were compared using chi-square analysis and by logistic regression. Differences in age between IDH populations were compared using t-test. Calculations were performed using SAS (version 9.3; SAS Institute, Cary, North Carolina) and RStudio (Version 1.0.136, ©2009-2016 RStudio Inc, Boston, Massachusetts). RESULTS Clinical Data Among the total 111 cases, the mean age at time of diagnosis was 51.4 yr (Table 1). Median follow-up time was 94 mo. Preoperative tumor size was available for 99 cases with an average maximum diameter of 4.3 cm. Of 108 patients with preoperative exam details, 94% had a KPS >90 preoperatively. Resection data were available for all cases, with 36.9% undergoing biopsy alone, 27.9% undergoing STR, and 35.1% undergoing GTR. TABLE 1. Characteristics of 111 Patients with Supratentorial Low-Grade Gliomas, 73 Patients with Known Isocitrate Dehydrogenase 1 Mutation Status   Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached    Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached  IDH = isocitrate dehydrogenase; KPS = Karnofsky Performance score. Values given as number of patients (%) unless otherwise indicated. View Large TABLE 1. Characteristics of 111 Patients with Supratentorial Low-Grade Gliomas, 73 Patients with Known Isocitrate Dehydrogenase 1 Mutation Status   Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached    Total  IDH status  Patient Characteristics      No of patients  111  73  Mean age in yr (+SD)  51.4 (±8.04)  51.1 (±7.7)  Median age (yr)  51  51  Females (%)  61 (54.95)  41 (56)  KPS (n = 108)       ≥90  102 (94.4)  69 (95.8)   <90  6 (5.6)  3 (4.2)  Seizure at presentation (%)  65 (62.5)  46 (63.9)  Mean size of tumor (cm)  4.3  4.3  Presence of tumor enhancement  14 (14.4)  7 (10.9)  Surgery (%)       Biopsy  41 (36.9)  24 (32.9)   Subtotal resection  31 (27.9)  18 (24.7)   Gross total resection  39 (35.1)  31 (42.5)  Pathology (%)       Astrocytoma  42 (37.8)  28 (38.4)   Oligodendroglioma  40 (36.04)  23 (31.5)   Oligoastrocytoma  29 (26.13)  22 (30.1)  IDH mutation (n = 73)       IDH wildtype (%)  19 (26.0)  19 (26.0)   IDH mutant (%)  54 (74.0)  54 (74.0)  Postoperative chemotherapy (%)  61 (59.2)  38 (54.3)  Postoperative radiation therapy (%)  65 (63.1)  40 (58.0)  No adjuvant treatment (%)  21 (18.9)  14 (19)  Median overall survival (mo)  189  N/A   Percent 5-yr survival  84.3  87.0   Percent 10-yr survival  67.7  74.1  Median progression-free survival (mo)  121  121   Percent 5-yr survival  71.3  72.3   Percent 10-yr survival  50.1  52.0  Median survival by histopathology (mo)       Astrocytoma  138  182   Oligodendroglioma  252  Not reached   Oligoastrocytoma  Not reached  Not reached  IDH = isocitrate dehydrogenase; KPS = Karnofsky Performance score. Values given as number of patients (%) unless otherwise indicated. View Large Molecular genomic data of tumor specimen were complete in 73 cases, allowing multivariate analysis in that cohort. The demographics for these 73 patients were similar to the total population (Table 1). Sixty-five percent of patients were alive at the time the data for this study were compiled. Outcome Analysis Median OS for all 111 patients was 189 mo (15.75 yr) with 5- and 10-yr survival rates of 84.3% and 67.7%, respectively. By log-rank comparison, there was no significant difference in survival based on histologic pathology, although oligodendroglioma and mixed glioma trended towards higher survival rates compared to patients with astrocytoma (P = .19). Median OS for patients with astrocytomas was 11.5 yr (138 mo), median OS for patients with oligodendrogliomas was 21 yr (252 mo), whereas median OS for patients with oligoastrocytomas had not yet been reached at time of data collection (Table 1). There were 39 cases of progression that led to a second surgery. Eleven of these cases progressed to grade III (28.2%) and 15 (38.5%) progressed to grade IV. Thirteen tumors remained grade II (33.3%). Median progression-free survival for the entire group was 8.2 yr. Molecular Data For patients with IDH mutations (n = 54), OS was significantly increased compared to patients without the mutation (n = 19). Five-year and 10-yr OSs were 94.2% and 85.8%, respectively, for IDH-mutant patients, and 67.4% and 36.8%, respectively, for IDH-wildtype patients (hazard ratio [HR] 0.175 [95% confidence interval 0.07-0.45], P < .001, Figure 1). Five-year and 10-yr progression-free survivals were 77.3% and 56.3%, respectively, for IDH-mutant patients and were 57.4% and 39.4%, respectively, for IDH-wildtype patients (HR 0.56 [0.28-1.15], P = .11). Of IDH-mutant cases, 44.4% progressed, whereas 57.9% progressed of IDH-wildtype cases (Table 2). There was a trend towards older median age in the IDH-wildtype population (53.8 yr) compared to the IDH-mutant population (50.1 yr), although this was not significant (P = .07). FIGURE 1. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant population vs IDH-wildtype population. FIGURE 1. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant population vs IDH-wildtype population. TABLE 2. Overall Survival and Progression-Free Survival of 73 Patients with Supratentorial Low-Grade Gliomas With Known Isocitrate Dehydrogenase 1 Mutation Status All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  IDH = isocitrate dehydrogenase. View Large TABLE 2. Overall Survival and Progression-Free Survival of 73 Patients with Supratentorial Low-Grade Gliomas With Known Isocitrate Dehydrogenase 1 Mutation Status All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  All patients with IDH status (73)  IDH wildtype (19)  IDH mutant (54)  Median age  53.8  50.1  Median overall survival (mo)  91  Not reached   Percent 5-yr survival  67.4  94.2   Percent 10-yr survival  36.8  85.8  Median progression-free survival (mo)  77  131   Percent 5-yr survival  57.4  77.3   Percent 10-yr survival  39.4  56.3  Patients with 1p19q status (49)  1p19q maintained (12)  1p19q codeleted (37)        Median overall survival (mo)  77  Not reached   Percent 5-yr survival  83.1  94.2   Percent 10-yr survival  49.6  78.01  IDH = isocitrate dehydrogenase. View Large By univariate analysis, factors predicting worse OS included patients of older age (HR 1.07 [1.03-1.11], P < .001), patients who presented with seizure (HR 2.46 [1.07-5.67], P = .035), patients with larger tumor size (HR 1.25 [1.03-1.52], P = .027), patients who underwent biopsy rather than an STR or GTR (HR 0.36 [0.16-0.77; STR], HR 0.14 [0.05-0.37; GTR], P < .001), and patients who underwent chemotherapy (HR 3.48 [1.34-9.05], P = .012), and radiation (HR 4.36 [1.63-11.11], P = .003; Table 3). TABLE 3. Univariate Analysis of 111 Patients Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  CI = Confidence interval; HR = hazard ratio; KPS = Karnofsky Performance score; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large TABLE 3. Univariate Analysis of 111 Patients Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  Variable  HR (95% CI)  P value (Wald)  Age  1.07 (1.03–1.11)  <.001  Female  0.59 (0.31–1.13)  .11  KPS  0.90 (0.79–1.01)  .070  Seizure  2.46 (1.07–5.67)  .035  Enhancement  1.37 (0.57–3.31)  .485  Size  1.25 (1.03–1.52)  .027  Pathology       Astrocytoma  1 (referent)  .22   Oligodendroglioma  0.61 (0.30–1.25)     Oligoastrocytoma  0.47 (0.17–1.29)    IDH mutation (73 patients)  0.17 (0.07–0.45)  <.001  1p19q codeletion (49 patients)  0.25 (0.07–0.98)  .03  Surgery       Biopsy  1 (referent)  <.001   Partial resection  0.36 (0.16–0.77)     Gross total resection  0.14 (0.05–0.37)    Chemotherapy  3.48 (1.34–9.05)  .012  Radiation  4.36 (1.63–11.11)  .003  CI = Confidence interval; HR = hazard ratio; KPS = Karnofsky Performance score; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large In a bivariate model with age and IDH status, IDH mutation was the only predictor of survival (HR 0.22 [0.07-0.68], P = .008; Table 4). Bivariate analysis of age and extent of resection showed that age and any surgical resection beyond biopsy predicted significantly improved survival (age HR 1.02 [1.02-1.16], P = .01, STR HR 0.13 [0.04-0.47], P = .002, GTR HR 0.026 [0.003-0.21], P = .0006). Bivariate analysis of IDH mutation status and extent of resection showed that IDH-mutant status and any surgical resection beyond biopsy predicted improved survival (IDH mutant 0.32 [0.11-0.9], P = .03, STR HR 0.19 [0.05-0.69], P = .01, GTR HR 0.04 [0.004-0.29], P = .002). Of these bivariate analyses, age became nonsignificant only in analysis with IDH status. TABLE 4. Bivariate Analyses Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  CI = Confidence interval; HR = hazard ratio; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large TABLE 4. Bivariate Analyses Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  Variable  HR (95% CI)  P value  Analysis 1       Age  1.03 (0.96–1.11)  .410   IDH mutation  0.22 (0.07–0.68)  .008  Analysis 2       Age  1.02 (1.02–1.16)  .01   Surgery        Biopsy  1 (Reference)      Partial resection  0.13 (0.04–0.47)  .002    Gross total resection  0.026 (0.003–0.21)  .0006  Analysis 3       IDH mutation  0.32 (0.11–0.9)  .03   Surgery        Biopsy  1 (Reference)      Partial resection  0.19 (0.05–0.69)  .01    Gross total resection  0.04 (0.004–0.29)  .002  CI = Confidence interval; HR = hazard ratio; IDH = isocitrate dehydrogenase. Bold values are statistically significant (P < .05). View Large Logistic regression analysis of extent of resection based on IDH mutation status showed there was significantly higher odds of getting a GTR in tumors that harbored IDH mutations. Conversely, there were significantly higher odds of biopsy in cases of IDH wildtype (OR 7.57, [95% confidence interval 2.59-24.85], P < .001; Table 5, Figure 2). FIGURE 2. View largeDownload slide Odds of extent of surgical resection based on IDH status. FIGURE 2. View largeDownload slide Odds of extent of surgical resection based on IDH status. TABLE 5. Surgical resection rates by IDH status   Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)    Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)  STR = subtotal resection; GTR = gross total resection; IDH = isocitrate dehydrogenase. View Large TABLE 5. Surgical resection rates by IDH status   Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)    Biopsy  STR  GTR  IDH wildtype  13 (68)  3 (16)  3 (16)  IDH mutant  11 (20)  15 (28)  28 (52)  STR = subtotal resection; GTR = gross total resection; IDH = isocitrate dehydrogenase. View Large Within the IDH-mutant population and those with known 1p19q status, 1p19q codeletion occurred in 75.5% of patients. Analysis of patients with 1p19q codeletion (37 patients) compared to 1p19q non-codeleted (12 patients) showed a significantly longer survival in the group with 1p19q codeletion (HR 0.172 [0.04-0.72]; P = .007; Figure 3). FIGURE 3. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant 1p19q codeleted population vs IDH-mutant 1p19q non-codeleted population. FIGURE 3. View largeDownload slide Kaplan–Meier OS curve of IDH-mutant 1p19q codeleted population vs IDH-mutant 1p19q non-codeleted population. DISCUSSION This study represents the largest reported single-institution clinical outcomes of patients with grade II glioma and high-risk criteria of age 40 yr or greater in the postmolecular genomic era. In the management of LGGs, the “Pignatti criteria” have held that patients with age ≥40 yr have worse prognosis, along with patients with tumor size >6 cm, tumor crossing midline, neurological deficit at initial presentation, and astrocytoma histopathology.12 Recently, however, molecular alterations of these tumors, such as IDH mutations and 1p and 19q codeletion, have been found in high numbers of diffuse gliomas, ∼80% to 90%, and significantly influence outcome (Table 6).1 Although tumors harboring these mutations or genomic alterations typically present in the third or fourth decade of life, our study reports on the rarer population of older patients with LGG, and still finds a high IDH mutation rate of 74% and 1p19q codeletion rate of 75.5% in those with known genotype. Although univariate analysis showed older age was a significantly negative prognostic factor for survival, in the subgroup of patients with known IDH status, bivariate survival analysis examining IDH mutation and age determined that IDH-mutant status was prognostic of survival, while age was not. TABLE 6. Outcomes of IDH-Mutant Populations in the Literature Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  View Large TABLE 6. Outcomes of IDH-Mutant Populations in the Literature Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  Study  Findings  Cancer Genome Atlas Research Network3  Performed genome-wide analyses of 293 lower grade gliomas from adults. Patients who had lower grade gliomas with wildtype IDH were older than those who had lower grade gliomas with mutated IDH. Persons who had lower grade gliomas with an IDH mutation and 1p/19q codeletion had a median survival of 8.0 yr, and those with an IDH mutation and no codeletion had a median survival of 6.3 yr.    Subanalysis of the TCGA dataset shows 80 patients with supratentorial grade II glioma age <40, 78 with known IDH status. The IDH mutation rate is 92.3% (72/78). For patients ≥40, the mutation rate was 87.7% (50/57).    Median survival of patients ≥40 was 7.9 yr with IDH mutation and 5.5 yr with IDH wildtype. Median survival of patients <40 was 7.9 yr with IDH mutation and not reached with IDH wildtype.  Beiko et al27  Investigated the impact of surgical resection on survival after controlling for IDH1 status in malignant astrocytomas—128 patients with grade III anaplastic astrocytomas and 207 patients with grade IV glioblastoma. Median age of the IDH-mutant population was significantly younger (median age 37 yr) than the IDH-wildtype population (57 yr, P < 0.001). IDH1 mutations were identified in 86 of 128 AAs (67%) and 27 of 207 GBM (13%).  Sabha et al22  Analyzed outcomes of 108 patients with grade II and grade III nonenhancing gliomas. A total of 93 cases with IDH mutations were detected (86% of total cohort). Patient age was examined as >50 and ≤50, which, along with clinical variables (tumor diameter, extent of resection, performance status), and pathology (tumor type and grade), were not predictive of OS or PFR. IDH mutation status alone was predictive of longer OS and PFR for the entire group of tumors; 1p/19q deletion alone was predictive of OS but not PFR.  Hartmann et al1  Determined mutation types and frequencies in 10 010 diffuse gliomas. The average age of patients with gliomas of WHO grade II carrying IDH1 mutations was 41.3 yr while glioma patients without mutations averaged 42.8 yr (not significant). The mutation rate was 72.7% for grade II astrocytoma, 82.0% for grade II oligodendroglioma, and 81.6% for grade II oligoastrocytoma.  View Large This finding suggests that molecular pathology of diffuse glioma may be more important than the age of the patient in determining clinical outcome. In the pre-molecular genomic era of the 1990s and 2000s, prognostic factors in LGGs were looked at extensively by both single-institution and multicenter RCTs. In their retrospective examination of prognostic factors of 379 patients with LGGs in 1997, Lote et al concluded that “prognosis in LGG following postoperative radiotherapy seemed largely determined by the inherent biology of the glioma and patient age at diagnosis.”16 In support of this, 20 yr later in 2017, Youland et al21 reported worse outcomes in their elderly group of 55 yr and older, although they did not use any genomic criteria in their analysis of patient outcomes.17 Under the scope of molecular genetics, however, recent studies suggest that the inherent tumor biology seems to trump more traditional prognostic factors, including age. In a 2014 study of patients of all ages with grade II and grade III gliomas, Sabha et al22 demonstrated that molecular data from the tumors was the only significant predictor of patient progression-free and OS. IDH mutation status, 1p/19q codeletion, and PTEN deletion were predictive of over overall survival in a multivariate model, while none of the clinical or demographic factors such as age, tumor size, performance status, or tumor histology were found to be predictive. More recently, Hayashi et al23 reported on 57 patients with 1p/19q codeleted gliomas and found that the only significant predictors of poor outcome were a gain of chromosome 19p and grade III histology. They found no significant difference in the OS of the patients with respect to age (≥40 vs <40 yr), degree of resection, maximum tumor diameter (≥5 vs <5 cm), histological subtype, and MGMT promoter methylation status. These results support our findings that traditional prognostic factors, such as age, may be less important in the setting of molecular markers. Older patients may not inherently have worse prognosis based on age, but rather, as our data suggest, due to a slightly lower chance of having an IDH mutation. Just as TERT promoter mutation frequency increases with patient age at diagnosis, it is possible that IDH mutation frequency decreases with patient age.24 While our study did not show a significant difference in IDH mutation rate by age, it did occur at a lower rate than in studies of LGGs including younger ages (68% vs up to 82%-91%, Table 6). In 2013, Gorlia et al validated new prognostic models from the European Organisation for Research and Treatment of Cancer (EORTC) by studying patient outcomes from tumors that had undergone central pathologic review rather than by local pathologist.25 Even more strikingly, in 2015 The Cancer Genome Atlas Research Network (TCGA) redefined glioma survival rates by molecular genomic classification rather than histologic classification.3 Along these lines, future outcomes analyses from large RCTs that include the molecular genomic information of tumors may allow re-examination and evolution of LGG prognostic factors. Extent of tumor resection is another common prognostic factor, with some large RCTs of gliomas, such as the EORTC 22844, finding significantly improved OS with increasing tumor resection rates.26 Although extent of surgical resection was highly significant on univariate analysis in our study, our study was limited to only bivariate models due to the small sample size of patients with known IDH status and limited number of events (deaths) per variable (n = 73, 18 deaths).27 Of these multiple bivariate analyses, the analysis of extent of resection and IDH status showed that both were significantly associated with survival. In logistic regression analysis, we found that IDH-mutant status and extent of surgical resection were highly correlated, with IDH-mutant status predicting a 7.5 times increased odds of a GTR than STR. This may be due to IDH-mutant tumors occurring in greater frequency frontal and temporal lobes, which, especially when right sided, lend themselves to safer maximum extent of resection. These findings have been similarly highlighted by other recent series. Youland et al21 found in that older patients with gliomas benefited from greater extent of resection, yet they did not have correlative mutation status of these patients.17 In studies of higher grade gliomas, Beiko et al28 were the first group to report on IDH1 mutation as an independent predictor of complete resection of enhancing disease, with 93% complete resections among mutants vs 67% among wildtype (P < .001). They further suggested that, given the decrease in IDH mutation after age 40, factoring in the clinical feature of age into risk stratification was an “incomplete surrogacy” for IDH status.28 Among grade III gliomas, Kawaguchi et al29 found that, while all patients undergoing GTR had longer median survival and progression-free survival times compared to those of non-GTR patients, the benefits of GTR was highest for the group with the IDH mutation without a 1p19q codeletion. The IDH mutant, 1p19q codeleted tumors, and the IDH-wildtype tumors did not show survival difference based on GTR vs non-GTR. In a study of 151 grade II gliomas, Leeper et al30 found no statistically significant difference in extent of resection achieved between any of the TCGA-defined molecular subgroups; however, only 9% of their population was IDH wildtype. While our study was not powered for this subgroup analysis, the role of resection and change in tumor volume with surgery based on molecular diagnosis is an important area of future research, with efforts already underway to allow for intraoperative determination of IDH status.31 Our paper did not find that adjuvant chemotherapy or radiation offered significant survival benefit, which differs from several recent phase II and phase III studies.32-34 This may be because our study looks at patients who already have at least 1 high-risk criteria and, therefore, our population, as a whole, may have tended to receive more aggressive treatment at baseline. Only 18.9% of our population was observed rather than receiving some form of postoperative chemotherapy or radiation. Similarly, in a study of 94 patients with LGG who were 55 yr and older treated at Mayo Clinic, Youland et al19,21 also found extent of surgical resection to be the only treatment intervention to significantly affect outcome. Neither radiation alone, chemotherapy alone, or combined chemoradiation had a survival impact, and their rate of postoperative observation was slightly higher than ours at 21%.19,21 To date, there have already been 2 phase III chemoradiation trials looking at treatment outcomes of LGG according to molecular diagnosis, and these include age ≥40 as a high-risk factor. In the EORTC 22033-26033, study of temozolomide chemotherapy vs radiotherapy in high-risk patients with LGG, subgroup analysis determined that IDH mutant, non-codeleted tumors had longer progression-free survival with radiation alone compared to temozolomide alone.32 However, patients with IDH mutant, 1p19q codeleted tumors or IDH-wildtype tumors showed no treatment-related difference in progression-free survival. Additionally, Buckner et al34 reported on high-risk patient outcomes after randomization to radiation alone vs PCV and radiation and found that patients with IDH-mutant tumors had significantly improved progression-free and overall survival compared to the IDH-wildtype cohort. More impressively, the IDH-mutant cohort treated with PCV and radiation had significantly improved survival over the IDH-mutant cohort treated with radiation alone. On their multivariate analysis for OS, age <40 was one of the significant factors indicating improved OS, while IDH-mutant status was not. Other positive prognostic factors included combined radiation and chemotherapy treatment and oligodenroglioma histology. These same variables were also found to significant influence improved progression-free survival, with the inclusion of IDH mutation status as an additional significant factor. Thus, while age still bore out to be influential in survival, similar to our findings, IDH mutation status predicted better survival than IDH-wildtype patients and had a significant influence on outcomes. Looking forward, these randomized trials will be very powerful for defining the role of molecular diagnosis. Study Limitations There are many inherent limitations of our study given its retrospective, chart-review design. Treatments were determined based on a variety of factors, including patient risk factors, patient or physician preference, and time period in which the patient was treated. Given the body of literature of the 1990s and 2000s that supported older age as a risk factor for worse outcomes, it is likely that our patients were treated more aggressively. In this population, adjuvant treatment of radiation and chemotherapy were not assigned randomly but chosen by clinicians based on perceived risk. Therefore, our results of worse outcome associated with adjuvant treatment more likely indicates that patients deemed more high risk for worse outcome were more likely to receive this therapy. The findings of our study were also diminished by the fact that we used the 2007 WHO classification of gliomas rather than the 2017 WHO grading scheme, mainly because most of our patients were diagnosed and treated based on 2007 classification. This renders survival based on histopathology less meaningful, as oligoastrocytoma no longer exists. Future analysis of molecular diagnosis and treatment outcomes across all ages is warranted, especially given recent findings from larger phase III trials with molecular genomic data showing that age <40 still has positive prognostic implications. Additionally, surgical treatment perhaps varied by surgeon and this could influence rates of resection. Furthermore, the extent of resection data based on MRI analysis was simplified into three tiers, and the size of tumors was recorded based on largest in a single dimension. We did not account for volume of T2/FLAIR hyperintense tumor or enhancing tumor, both pre- and postoperatively, which would be more sensitive and accurate to account for residual tumor burden and extent of resection. While we did look at Pignatti criteria data for this study, we limited ourselves to KPS as proxy for neurological deficit and only had two patients with tumor crossing midline so excluded these data from analysis. While size of tumor and seizure at diagnosis were significant on univariate analysis, we did not include this in our final models due to previously discussed limitations of our multivariate survival analyses. Astrocytic pathology trended towards predicting a worse survival, but not significantly so. As for molecular data, we were limited to only 73 of our 111 patients, limiting the robustness of our survival analyses since there were only 18 deaths in this population. We also did not capture large enough numbers of other molecular and genomic alterations to account for their influence in this study, such as p53 mutation, PTEN mutation, and TERT mutation. We did not have the age group of <40 for outcomes comparison, which is a significant limitation to this paper and can be reported on in upcoming analyses from our group. CONCLUSION The present study shows that IDH mutation and 1p19q chromosomal codeletion of LGG in patients ≥40 occurs at high rates like the younger population and predicts a similar survival advantage, and this should be considered when determining optimal management strategy in this population. Tumors harboring IDH mutations are known to present in more surgically accessible areas, and, as found in this study and many previous, maximizing extent of resection seems to hold a withstanding influence on improved survival. Prior analyses indicating older age as a poor prognostic indicator could have associated with IDH mutations occurring at slightly lower frequency in this population. Given that current and future studies include the molecular genetics of gliomas, it is likely time that the 15-yr-old 5 clinical “Pignatti criteria” give way to redefined prognostic factors that reflect the molecular era and, as such, more accurately predict outcome and treatment response. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. Notes Some of this work was presented previously at the Congress of Neurological Surgeons annual meeting in Boston, Massachusetts, as an Oral Poster Abstract on October 21, 2014. REFERENCES 1. Hartmann C, Meyer J, Balss J et al.   Type and frequency of IDH1 and IDH2 mutations are related to astrocytic and oligodendroglial differentiation and age: a study of 1,010 diffuse gliomas. Acta Neuropathol . 2009; 118( 4): 469- 474. Google Scholar CrossRef Search ADS PubMed  2. Louis DN, Perry A, Reifenberger G et al.   The 2016 world health organization classification of tumors of the central nervous system: a summary. Acta Neuropathol . 2016; 131( 6): 803- 820. Google Scholar CrossRef Search ADS PubMed  3. Cancer Genome Atlas Research Network. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. N Engl J Med . 2015; 372( 26): 2481- 2498. CrossRef Search ADS PubMed  4. Jones PS, Dunn GP, Barker FG, Curry WT, Hochberg FH, Cahill DP. Molecular genetics of low-grade gliomas: genomic alterations guiding diagnosis and therapeutic intervention. 11th Annual Frye-Halloran Brain Tumor Symposium. Neurosurg Focus . 2013; 34( 2): E9. Google Scholar CrossRef Search ADS PubMed  5. Chen R, Ravindra VM, Cohen AL et al.   Molecular features assisting in diagnosis, surgery, and treatment decision making in low-grade gliomas. Neurosurg Focus . 2015; 38( 3): E2. Google Scholar CrossRef Search ADS PubMed  6. Buckner J, Giannini C, Eckel-Passow J et al.   Management of diffuse low-grade gliomas in adults - use of molecular diagnostics. Nat Rev Neurol . 2017; 13( 6): 340- 351. Google Scholar CrossRef Search ADS PubMed  7. Delgado-López PD, Corrales-García EM, Martino J, Lastra-Aras E, Dueñas-Polo MT. Diffuse low-grade glioma: a review on the new molecular classification, natural history and current management strategies. Clin Transl Oncol . 2017; 19( 8): 931- 944. Google Scholar CrossRef Search ADS PubMed  8. Shaw E, Arusell R, Scheithauer B et al.   Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a north central cancer treatment Group/Radiation therapy oncology Group/Eastern cooperative oncology group study. J Clin Oncol . 2002; 20( 9): 2267- 2276. Google Scholar CrossRef Search ADS PubMed  9. Bauman G, Lote K, Larson D et al.   Pretreatment factors predict overall survival for patients with low-grade glioma: a recursive partitioning analysis. Int J Radiat Oncol Biol Phys . 1999; 45( 4): 923- 929. Google Scholar CrossRef Search ADS PubMed  10. Nicolato A, Gerosa MA, Fina P, Iuzzolino P, Giorgiutti F, Bricolo A. Prognostic factors in low-grade supratentorial astrocytomas: a uni-multivariate statistical analysis in 76 surgically treated adult patients. Surg Neurol . 1995; 44( 3): 208- 223. Google Scholar CrossRef Search ADS PubMed  11. Leighton C, Fisher B, Bauman G et al.   Supratentorial low-grade glioma in adults: an analysis of prognostic factors and timing of radiation. J Clin Oncol . 1997; 15( 4): 1294- 1301. Google Scholar CrossRef Search ADS PubMed  12. Pignatti F, Van den Bent M, Curran D et al.   Prognostic factors for survival in adult patients with cerebral low-grade glioma. J Clin Oncol . 2002; 20( 8): 2076- 2084. Google Scholar CrossRef Search ADS PubMed  13. Gallego Perez-Larraya J, Delattre J-Y. Management of elderly patients with gliomas. Oncologist . 2014; 19( 12): 1258- 1267. Google Scholar CrossRef Search ADS PubMed  14. Lai A, Kharbanda S, Pope WB et al.   Evidence for sequenced molecular evolution of IDH1 mutant glioblastoma from a distinct cell of origin. J Clin Oncol . 2011; 29( 34): 4482- 4490. Google Scholar CrossRef Search ADS PubMed  15. Hartmann C, Hentschel B, Wick W et al.   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 . 2010; 120( 6): 707- 718. Google Scholar CrossRef Search ADS PubMed  16. Lote K, Egeland T, Hager B et al.   Survival, prognostic factors, and therapeutic efficacy in low-grade glioma: a retrospective study in 379 patients. J Clin Oncol . 1997; 15( 9): 3129- 3140. Google Scholar CrossRef Search ADS PubMed  17. Schomas DA, Laack NN, Brown PD. Low-grade gliomas in older patients. Cancer . 2009; 115( 17): 3969- 3978. Google Scholar CrossRef Search ADS PubMed  18. Chi AS, Batchelor TT, Dias-Santagata D et al.   Prospective, high-throughput molecular profiling of human gliomas. J Neurooncol . 2012; 110( 1): 89- 98. Google Scholar CrossRef Search ADS PubMed  19. Youland RS, Schomas DA, Brown PD et al.   Changes in presentation, treatment, and outcomes of adult low-grade gliomas over the past fifty years. Neuro Oncol . 2013; 15( 8): 1102- 1110. Google Scholar CrossRef Search ADS PubMed  20. Sanson M, Marie Y, Paris S et al.   Isocitrate dehydrogenase 1 codon 132 mutation is an important prognostic biomarker in gliomas. J Clin Oncol . 2009; 27( 25): 4150- 4154. Google Scholar CrossRef Search ADS PubMed  21. Youland RS, Schomas DA, Brown PD, Parney IF, Laack NNI. Patterns of care and treatment outcomes in older adults with low grade glioma: a 50-year experience. J Neurooncol . 2017; 133( 2): 339- 346. Google Scholar CrossRef Search ADS PubMed  22. Sabha N, Knobbe CB, Maganti M et al.   Analysis of IDH mutation, 1p/19q deletion, and PTEN loss delineates prognosis in clinical low-grade diffuse gliomas. Neuro Oncol . 2014; 16( 7): 914- 923. Google Scholar CrossRef Search ADS PubMed  23. Hayashi S, Kitamura Y, Hirose Y, Yoshida K, Sasaki H. Molecular–Genetic and clinicopathological prognostic factors in patients with gliomas showing total 1p19q loss: gain of chromosome 19p and histological grade III negatively correlate with patient's prognosis. J Neurooncol . 2017; 132( 1): 119- 126. Google Scholar CrossRef Search ADS PubMed  24. Spiegl-Kreinecker S, Lötsch D, Ghanim B et al.   Prognostic quality of activating TERT promoter mutations in glioblastoma: interaction with the rs2853669 polymorphism and patient age at diagnosis. Neuro Oncol . 2015; 17( 9): 1231- 1240. Google Scholar CrossRef Search ADS PubMed  25. Gorlia T, Wu W, Wang M et al.   New validated prognostic models and prognostic calculators in patients with low-grade gliomas diagnosed by central pathology review: a pooled analysis of EORTC/RTOG/NCCTG phase III clinical trials. Neuro Oncol . 2013; 15( 11): 1568- 1579. Google Scholar CrossRef Search ADS PubMed  26. Karim AB, Maat B, Hatlevoll R et al.   A randomized trial on dose-response in radiation therapy of low-grade cerebral glioma: European Organization for Research and Treatment of Cancer (EORTC) Study 22844. Int J Radiat Oncol Biol Phys . 1996; 36( 3): 549- 556. Google Scholar CrossRef Search ADS PubMed  27. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol . 1996; 49( 12): 1373- 1379. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8970487. Accessed February 7, 2018. Google Scholar CrossRef Search ADS PubMed  28. Beiko J, Suki D, Hess KR et al.   IDH1 mutant malignant astrocytomas are more amenable to surgical resection and have a survival benefit associated with maximal surgical resection. Neuro Oncol . 2014; 16( 1): 81- 91. Google Scholar CrossRef Search ADS PubMed  29. Kawaguchi T, Sonoda Y, Shibahara I et al.   Impact of gross total resection in patients with WHO grade III glioma harboring the IDH 1/2 mutation without the 1p/19q co-deletion. J Neurooncol . 2016; 129( 3): 505- 514. Google Scholar CrossRef Search ADS PubMed  30. Leeper HE, Caron AA, Decker PA, Jenkins RB, Lachance DH, Giannini C. IDH mutation, 1p19q codeletion and ATRX loss in WHO grade II gliomas. Oncotarget . 2015; 6( 30): 30295- 30305. Google Scholar CrossRef Search ADS PubMed  31. Shankar GM, Francis JM, Rinne ML et al.   Rapid intraoperative molecular characterization of glioma. JAMA Oncol . 2015; 1( 5): 662. Google Scholar CrossRef Search ADS PubMed  32. Baumert BG, Hegi ME, van den Bent MJ et al.   Temozolomide chemotherapy versus radiotherapy in high-risk low-grade glioma (EORTC 22033-26033): a randomised, open-label, phase 3 intergroup study. Lancet Oncol . 2016; 17( 11): 1521- 1532. Google Scholar CrossRef Search ADS PubMed  33. Fisher BJ, Hu C, Macdonald DR et al.   Phase 2 study of temozolomide-based chemoradiation therapy for high-risk low-grade gliomas: preliminary results of radiation therapy oncology Group 0424. Int J Radiat Oncol Biol Phys . 2015; 91( 3): 497- 504. Google Scholar CrossRef Search ADS PubMed  34. Buckner JC, Shaw EG, Pugh SL et al.   Radiation plus procarbazine, CCNU, and vincristine in low-grade glioma. N Engl J Med . 2016; 374( 14): 1344- 1355. Google Scholar CrossRef Search ADS PubMed  Neurosurgery Speaks! Audio abstracts available for this article at www.neurosurgery-online.com. COMMENTS The authors examine the question of whether “mutations” drive outcomes of older patients with low-grade gliomas. This is an interesting study but the results are limited by the retrospective nature of the study and patient selecton bias demonstrated by facts such as those of chemotherapy and radiation being associated with worse outcome. Some tumors were somehow deemed more aggressive and were treated with these modalities while others were not. It seems that this is a chicken and egg question. Furthermore, the findings of this study are diminished by the fact that with the new 2017 WHO grading scheme all grade II tumors require IDH-1 mutation and the oligodendrogliomas additionally require 1p19q deletion. Additionally, since oligoastrocytoma no longer exists in this classification scheme, so any discussion of this entity is unfruitful. Randy L. Jensen Salt Lake City, UT Maximal safe surgical resection of a presumed supratentorial diffuse low-grade glioma (LGG) is considered standard of care.1 This provides tumor tissue for histological and genetic analysis, reduces tumor burden and subsequent risk of tumor progression, and may reduce seizures associated with the tumor – it may also be associated with improved survival.1 Recent reports of the beneficial effect of postoperative radiation and chemotherapy in select patients were largely based on criteria proposed by Pignatti et al2 more than a decade ago.3,4 They proposed 5 criteria that portended a worse prognosis for adult patients with a supratentorial LGG; age type = "Other" >40 years, astrocytoma histology, tumor diameter >6 cm, tumors crossing the midline, and the presence of a neurological deficit before surgery.2 A critical factor was age; 40 years was considered the bellwether point for prognostication and treatment decision-making.2-4 The dawn of the new era of molecular medicine changed our philosophy on the management of LGG. In 2016, the World Health Organization (WHO) reclassified neuroepithelial tumors based on molecular markers such as 1p/19q and isocitrate dehydrogenase (IDH) mutations.5 Currently, these markers supercede histological criteria for tumors such as an oligodendroglioma and some categories such as oligoastrocytomas have been altogether culled.5 Overall, it has had a constructive effect on our understanding of gliomas. Management of these histologically and genetically diverse tumors is now predicated by the tumor molecular profile and henceforth, it is inconceivable that any clinical trial will be undertaken without consideration of these biomarkers. The p53 mutation (for astrocytomas) and 1p/19q deletion (for oligodendrogliomas) are pathognomic and mutually exclusive.5 A recent addition to this select list was isocitrate dehydrogenase (IDH) mutations.6 IDH mutations are reported in hematological malignancies such as acute myeloid leukemia but among solid tumors, they are particularly unique to gliomas. IDH mutations may occur as both IDH 1 and IDH 2 mutations. IDH 1 mutations appear to be more relevant and may be detected by standard immunohistochemical studies that are ubiquitously available – on the other hand, detection of IDH 2 mutations requires genomic sequencing studies. Both IDH mutations tend to occur early in glioma development and may be seen with either p53 mutations or 1p/19q deletions. They contribute to glioma progression through induction of the HIF-1 pathway and genome-wide histone and DNA methylation alterations through the oncometabolite 2-hydroxyglutarate.1,6,7 IDH mutations also alter the glioma microenvironment by affecting collagen maturation and impairing basement membrane function.7 They may predispose a patient to seizures or portend worse postoperative seizure occurrence. IDH 1 mutations are commonly seen in LGG and secondary glioblastoma (GBM) but rarely ever in primary GBM or malignant gliomas with EGFR amplification.7 They frequently accompany MGMT promoter hypermethylation and correlate with the Ki-67 labeling index.8,9 They are, hence, a major marker for OS in LGG and malignant gliomas – and are possibly more powerful than the Pignatti score in their predictive capacity.8,9 This study goes back to the tipping point effect of age viewed in light of this new molecular information. In LGG patients >40 years of age with IDH 1 mutations, the authors report improved survival; the presence of a 1p/19q deletion had a similar beneficial effect. Also, greater surgical resection was predictive of survival – and extent of resection significantly correlated with IDH mutation status. Their findings confirm previous reports of the salutary effect of IDH1 mutations in gliomas and the value of maximal safe surgical resection. It proves the need to use molecular markers to determine the optimal management of a LGG, regardless of age. It may be that age is a secondary factor, simply reflecting the frequency of occurrence of these changes. In other words, younger patients tend to have better outcomes because they have IDH 1 mutated tumors more frequently. Older patients who have the same molecular changes may fare the same. The characterization of genetic findings such as IDH mutations, p53 mutations, and 1p/19q deletions has revolutionized our understanding of diffuse LGG. It has led to a more precise pathological classification, the potential for better therapies, and improved prognostication. But this may simply be the proverbial “tip of the iceberg”– in addition to prognostication, there may be therapeutic benefits. In acute myeloid leukemia, IDH inhibition appears to the clinically effective – possibly by allowing malignant blasts cells to differentiate and become post mitotic. There may be similar value in the future with IDH inhibitors for gliomas. IDH1-R132H mutations are reportedly associated with a more severe phenotype of postoperative epilepsy – this suggests potential for an antiepileptic therapeutic effect of an IDH 1 inhibitor.10 This is a nice study that improves our understanding of the impact of genomic alterations on the clinical trajectory of a glioma in older patients. It provides clinicians useful information when counseling these patients. Such findings may alter our management of these patients significantly, causing us to look anew at older criteria used to determine patients who are at higher risk for tumor progression. The use of radiation and chemotherapy is not without side effects – molecular genomic information may lead to a more abstemious use of these modalities or at least a more targeted individualized approach. Vikram C. Prabhu Kevin Barton Ewa Borys Edward Melian Maywood, IL 1. Buckner J Giannini C Eckel-Passow J Lachance D Parney I Laack N Jenkins R. Management of diffuse low-grade gliomas in adults - use of molecular diagnostics. Nat Rev Neurol . 2017; 13( 6): 340- 351. Google Scholar CrossRef Search ADS PubMed  2. Pignatti F van den Bent M Curran D Debruyne C Sylvester R Therasse P Afra D Cornu P Bolla M Vecht C Karim AB; European Organization for Research and Treatment of Cancer Brain Tumor Cooperative Group; European Organization for Research and Treatment of Cancer Radiotherapy Cooperative Group. Prognostic factors for survival in adult patients with cerebral low-grade glioma. J Clin Oncol . 2002; 20( 8): 2076- 84. Google Scholar CrossRef Search ADS PubMed  3. Shaw EG Wang M Coons SW Brachman DG Buckner JC Stelzer KJ Barger GR Brown PD Gilbert MR Mehta MP. Randomized trial of radiation therapy plus procarbazine, lomustine, and vincristine chemotherapy for supratentorial adult low-grade glioma: initial results of RTOG 9802. J Clin Oncol . 2012; 30( 25): 3065- 70. Google Scholar CrossRef Search ADS PubMed  4. Youland RS Kreofsky CR Schomas DA Brown PD Buckner JC Laack NN. The impact of adjuvant therapy for patients with high-risk diffuse WHO grade II glioma. J Neurooncol . 2017; 135( 3): 535- 543. Google Scholar CrossRef Search ADS PubMed  5. Komori T. The 2016 WHO Classification of Tumours of the Central Nervous System: The Major Points of Revision. Neurol Med Chir (Tokyo) . 2017; 57( 7): 301- 311. Google Scholar CrossRef Search ADS PubMed  6. Parsons DW Jones S Zhang X Lin JC Leary RJ Angenendt P Mankoo P Carter H Siu IM Gallia GL Olivi A McLendon R Rasheed BA Keir S Nikolskaya T Nikolsky Y Busam DA Tekleab H Diaz LA Jr Hartigan J Smith DR Strausberg RL Marie SK Shinjo SM Yan H Riggins GJ Bigner DD Karchin R Papadopoulos N Parmigiani G Vogelstein B Velculescu VE Kinzler KW. An integrated genomic analysis of human glioblastoma multiforme. Science  2008; 321( 5897): 1807- 1812. Google Scholar CrossRef Search ADS PubMed  7. Zhang C Moore LM Li X Yung WK Zhang W. IDH1/2 mutations target a key hallmark of cancer by deregulating cellular metabolism in glioma. Neuro Oncol . 2013; 15( 9): 1114- 26. Google Scholar CrossRef Search ADS PubMed  8. Etxaniz O Carrato C de Aguirre I Queralt C Muñoz A Ramirez JL Rosell R Villà S Diaz R Estival A Teixidor P Indacochea A Ahjal S Vilà L Balañá C. IDH mutation status trumps the Pignatti risk score as a prognostic marker in low-grade gliomas. J Neurooncol . 2017 Nov; 135( 2): 273- 284. Google Scholar CrossRef Search ADS PubMed  9. Zeng A Hu Q Liu Y Wang Z Cui X Li R Yan W You Y. IDH1/2 mutation status combined with Ki67 labeling index defines distinct prong-ostic groups in glioma. Oncotarget  2015; 6( 30):30  232- 8. Google Scholar CrossRef Search ADS   10. Neal A Kwan P O’Brien TJ Buckland ME Gonzales M Morokoff A. IDH1 and IDH2 mutations in postoperative diffuse glioma-associated epilepsy. Epilepsy Behav . 2018; 78: 30- 36. Google Scholar CrossRef Search ADS PubMed  Neurosurgery Speaks (Audio Abstracts) Listen to audio translations of this paper's abstract into select languages by choosing from one of the selections below. English: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France English: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France Close French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France French: Georges Abi Lahoud, MD, MSc, MS. Department of Neurosurgery Sainte-Anne University Hospital Paris Descartes University Paris, France Close Chinese: Lin Song, MD. Department of Neurosurgery Beijing Tiantan Hospital Capital Medical University Beijing, China Chinese: Lin Song, MD. Department of Neurosurgery Beijing Tiantan Hospital Capital Medical University Beijing, China Close Italian: Alfredo Conti, MD, PhD. Department of Neurosurgery Department of Neurosurgery Charité Universitätsmedizin Berlin, Germany Italian: Alfredo Conti, MD, PhD. Department of Neurosurgery Department of Neurosurgery Charité Universitätsmedizin Berlin, Germany Close Russian: Sergei Kim. Department of Pediatric Neurosurgery Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Russian: Sergei Kim. Department of Pediatric Neurosurgery Novosibirsk Federal Centre of Neurosurgery Novosibirsk, Russia Close Japanese: Ryu Kurokawa, MD. Department of Neurosurgery Dokkyo University Hospital Tochigi, Japan Japanese: Ryu Kurokawa, MD. Department of Neurosurgery Dokkyo University Hospital Tochigi, Japan Close Korean: Sun Ha Paek, MD, PhD. Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Korean: Sun Ha Paek, MD, PhD. Department of Neurosurgery Seoul National University College of Medicine Seoul, Republic of Korea Close Portuguese: Eduardo Carvalhal Ribas, MD. Neurosurgery Department Hospital das Clínicas University of São Paulo Medicine School (HC-FMUSP) Hospital Israelita Albert Einstein São Paulo, Brazil Portuguese: Eduardo Carvalhal Ribas, MD. Neurosurgery Department Hospital das Clínicas University of São Paulo Medicine School (HC-FMUSP) Hospital Israelita Albert Einstein São Paulo, Brazil Close Greek: Alexiou A Georgios, MD. Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Greek: Alexiou A Georgios, MD. Department of Neurosurgery University Hospital of Ioannina Ioannina, Greece Close Spanish: Francisco Alberto Mannará, MD. Department of Neurosurgery Fernández Hospital Buenos Aires City, Argentina Spanish: Francisco Alberto Mannará, MD. Department of Neurosurgery Fernández Hospital Buenos Aires City, Argentina Close Copyright © 2018 by the Congress of Neurological Surgeons This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

NeurosurgeryOxford University Press

Published: May 25, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off