High-grade gliomas in children and adolescents: is there a role for reoperation?

Marcos Devanir Silva da CostaDepartment of Neurology and Neurosurgery, Universidade Federal de São Paulo; and
Divisions of Neurosurgery,

Search for other papers by Marcos Devanir Silva da Costa in
jns
Google Scholar
PubMed
Close
 MD, PhD
,
Nicole Cavalari CamargoDepartment of Neurology and Neurosurgery, Universidade Federal de São Paulo; and

Search for other papers by Nicole Cavalari Camargo in
jns
Google Scholar
PubMed
Close
 BSc
,
Patricia Alessandra DastoliDepartment of Neurology and Neurosurgery, Universidade Federal de São Paulo; and
Divisions of Neurosurgery,

Search for other papers by Patricia Alessandra Dastoli in
jns
Google Scholar
PubMed
Close
 MD, PhD
,
Jardel Mendonça NicácioDepartment of Neurology and Neurosurgery, Universidade Federal de São Paulo; and
Divisions of Neurosurgery,

Search for other papers by Jardel Mendonça Nicácio in
jns
Google Scholar
PubMed
Close
 MD
,
Frederico Adolfo Benevides SilvaRadiology,

Search for other papers by Frederico Adolfo Benevides Silva in
jns
Google Scholar
PubMed
Close
 MD
,
Maria Luisa Sucharski FigueiredoRadiotherapy, and

Search for other papers by Maria Luisa Sucharski Figueiredo in
jns
Google Scholar
PubMed
Close
 MD
,
Michael Jenwei ChenRadiotherapy, and

Search for other papers by Michael Jenwei Chen in
jns
Google Scholar
PubMed
Close
 MD
,
Andrea Maria CappellanoNeuro-Oncology, Instituto de Oncologia Pediátrica, São Paulo—SP, Brazil

Search for other papers by Andrea Maria Cappellano in
jns
Google Scholar
PubMed
Close
 MD, PhD
,
Nasjla Saba da SilvaNeuro-Oncology, Instituto de Oncologia Pediátrica, São Paulo—SP, Brazil

Search for other papers by Nasjla Saba da Silva in
jns
Google Scholar
PubMed
Close
 MD
, and
Sergio CavalheiroDepartment of Neurology and Neurosurgery, Universidade Federal de São Paulo; and
Divisions of Neurosurgery,

Search for other papers by Sergio Cavalheiro in
jns
Google Scholar
PubMed
Close
 MD, PhD
Free access

OBJECTIVE

Tumors of the CNS are the main causes of childhood cancer and have an incidence that exceeds that of leukemia. In addition, they are the leading causes of cancer-related death in childhood. High-grade gliomas account for 11% of such neoplasms and are characterized by aggressive clinical behavior and high morbidity and mortality. There is a lack of studies focusing on the factors that can prolong survival in these patients or guide therapeutic interventions. The authors aimed to investigate the factors related to longer survival durations, with a focus on reoperation for gross-total resection (GTR).

METHODS

In this retrospective cohort study, the authors analyzed 78 patients diagnosed with high-grade gliomas occurring across all CNS locations except diffuse intrinsic pontine gliomas. Patients 0 to < 19 years of age were followed up at the Pediatric Oncology Institute. Overall survival (OS) and progression-free survival (PFS) were analyzed in the context of various prognostic factors, such as age, sex, histology, extent of tumor resection, reoperation for GTR, adjuvant treatment, and treatment initiation from 2010 onward.

RESULTS

With a mean age at diagnosis of 8.7 years, 50% of the patients were female and approximately 39% underwent GTR at some point, which was already achieved in approximately 46% of them in the first surgery. The median OS was 17 months, and PFS was 10 months. In terms of median OS, the authors found no significant difference between those with reoperation for GTR and patients without GTR during treatment. Significant differences were observed in the OS in terms of the extent of resection in the first surgery, age, sex, Ki-67 expression, adjuvant treatment, and treatment initiation from 2010 onward. Furthermore, the PFS values significantly differed between those with GTR in the first surgery and Ki-67 expression ≥ 50%.

CONCLUSIONS

This study demonstrates the importance of GTR for these neoplasms, highlights the role of surgeons in its achievement in the first attempt, and questions the role of reoperation for this purpose. Finally, this study further supports the use of combined adjuvant treatment for the improvement of OS and PFS.

ABBREVIATIONS

GTR = gross-total resection; OS = overall survival; PFS = progression-free survival; STR = subtotal resection.

OBJECTIVE

Tumors of the CNS are the main causes of childhood cancer and have an incidence that exceeds that of leukemia. In addition, they are the leading causes of cancer-related death in childhood. High-grade gliomas account for 11% of such neoplasms and are characterized by aggressive clinical behavior and high morbidity and mortality. There is a lack of studies focusing on the factors that can prolong survival in these patients or guide therapeutic interventions. The authors aimed to investigate the factors related to longer survival durations, with a focus on reoperation for gross-total resection (GTR).

METHODS

In this retrospective cohort study, the authors analyzed 78 patients diagnosed with high-grade gliomas occurring across all CNS locations except diffuse intrinsic pontine gliomas. Patients 0 to < 19 years of age were followed up at the Pediatric Oncology Institute. Overall survival (OS) and progression-free survival (PFS) were analyzed in the context of various prognostic factors, such as age, sex, histology, extent of tumor resection, reoperation for GTR, adjuvant treatment, and treatment initiation from 2010 onward.

RESULTS

With a mean age at diagnosis of 8.7 years, 50% of the patients were female and approximately 39% underwent GTR at some point, which was already achieved in approximately 46% of them in the first surgery. The median OS was 17 months, and PFS was 10 months. In terms of median OS, the authors found no significant difference between those with reoperation for GTR and patients without GTR during treatment. Significant differences were observed in the OS in terms of the extent of resection in the first surgery, age, sex, Ki-67 expression, adjuvant treatment, and treatment initiation from 2010 onward. Furthermore, the PFS values significantly differed between those with GTR in the first surgery and Ki-67 expression ≥ 50%.

CONCLUSIONS

This study demonstrates the importance of GTR for these neoplasms, highlights the role of surgeons in its achievement in the first attempt, and questions the role of reoperation for this purpose. Finally, this study further supports the use of combined adjuvant treatment for the improvement of OS and PFS.

In Brief

In this study the authors compared the survival between pediatric patients with high-grade gliomas who underwent gross-total resection after one or more reoperations versus those with subtotal resection. The authors found that median overall survival was 17 months, and progression-free survival was 10 months for all patients. The authors also found no differences in survival between patients with gross-total resection after reoperation when compared with those who achieved subtotal resection. These findings highlight the importance of the first attempt to remove the tumor, because the second surgery aiming to achieve gross-total resection may not provide the patient with longer survival.

Tumors of the CNS are the main cause of childhood cancer, the incidence of which exceeds that of leukemia.1,2 Approximately 5.74 individuals younger than 14 years of age per 100,000 inhabitants are diagnosed with CNS tumors, accounting for approximately 1 in every 2000 individuals in this age group.2 In addition to their high incidence in the pediatric population, these tumors are also the main cause of cancer-related death in childhood.1,2 High-grade gliomas account for approximately 11% of all such neoplasms.1

According to the WHO, high-grade gliomas are malignant tumors, categorized into grades III and IV, that include glioblastoma, anaplastic astrocytoma, anaplastic oligodendroglioma, anaplastic oligoastrocytoma, and anaplastic xanthoastrocytoma.3 By definition they are aggressive tumors that are associated with high morbidity and mortality values and a 5-year survival rate of 28.1%.2 Therefore, treatment is challenging and involves a combination of surgery, radiation, and chemotherapy.1

Several studies have investigated the demographic, clinical, and histological factors that affect patient survival in such settings. These prognostic factors, whether clinical, radiological, histological, or genetic-molecular, can be divided into two groups: a group of immutable factors (including age, sex, and histological type), and another group of factors subject to medical intervention (including the time between diagnosis and start of treatment, extent of tumor resection, number of surgeries and, more recently, with new discoveries, biological markers).

Published studies on high-grade gliomas predominantly include adult patients.4–7 In pediatric populations, the few studies conducted so far have had small sample sizes, with some of them indicating that sex, age, tumor location, and extent of resection affect survival.8–15

However, although previous cohort studies have described the prognostic factors in this population, their findings have shown low applicability in the achievement of therapeutic success. Likewise, the formulation of possible interventions in the natural history of the disease for the enhancement of patients’ survival or quality of life has been lacking. Therefore, we aimed to investigate the prognostic factors, especially those subject to interventions such as reoperation, that affect survival in a large cohort of pediatric patients.

Methods

This retrospective study analyzed patients admitted to the Instituto de Oncologia Pediátrica/Grupo de Apoio ao Adolescente e a Criança com Câncer (IOP/GRAACC)/Universidade Federal de São Paulo—(UNIFESP) between 1989 and 2019. The study was approved by the research ethics committee of the institution. No consent was required considering the retrospective nature of the study. The patients’ identities were anonymized.

Of 1508 patients with tumors of the CNS, those younger than 19 years with a histological diagnosis of glioblastoma and anaplastic astrocytoma, or high-grade glioma (grades III and IV), were included in the study after meeting the inclusion criteria. The histological diagnosis was defined by the pathologist at the institution even when the patient was initially treated in another institution. All CNS locations were included, except diffuse intrinsic pontine gliomas. Only primary cases of high-grade glioma were included; that is, patients whose results from the first anatomopathological examination did not reveal high-grade glioma were excluded from this series.

The medical records of patients selected based on the inclusion criteria were retrospectively analyzed, and we collected data on the date of diagnosis, age at diagnosis, sex, histological type, location, signs and symptoms, date of recurrence or progression, date of first surgery, extent of resection, occurrence of reoperation, Ki-67 index expression, adjuvant treatment (chemotherapy and radiotherapy), date of treatment initiation, and date of death or the last clinical follow-up.

The date of diagnosis was defined as the date of the first imaging tests (CT or MRI) that identified the tumor. Overall survival (OS) was calculated considering the period from the date of diagnosis to the date of all-cause mortality or last follow-up. Progression-free survival (PFS) was considered to be the time interval between the date of diagnosis and lesion recurrence or progression or, in cases with no recurrence or progression, death due to the disease or last follow-up date. Recurrence was defined as the reappearance of the lesion after total resection, and progression was characterized as an increase in the tumor volume > 25% on radiological follow-up.

For the extent of tumor resection three categories were considered: gross-total resection (GTR), subtotal resection (STR), and biopsy. GTR was confirmed on achievement of 100% resection of the preoperative volume of the Gd-enhanced lesion on MRI by using the (A × B × C)/2 method,16 which may have been reported by the radiologist, as directly measured by the authors of this article (M.D.S.C., N.C.C.), or described in the medical records by neuro-oncological report. STR was considered when the resection rate, as measured on postoperative MRI, was < 100% of the preoperative volume. Biopsy was considered when only small fragments were removed from the lesion for histological characterization through stereotaxic needle localization or neuronavigation. Postoperative MRI examinations were performed as part of the institution’s routine practice within 24–72 hours after surgery, to reduce potential artifacts such as enhancement of blood products. Radiological follow-up was performed according to the institution’s routine practice every 2–4 months, using MRI studies, whereas clinical follow-up was dependent on the chemotherapy and radiotherapy routine. Lesion type and histological grade were characterized according to the WHO criteria at the time of diagnosis.

Statistical analysis was performed using Kaplan-Meier survival curves and log-rank (Mantel-Cox) tests to compare the groups. A multivariate analysis was conducted with multiple logistic regression models, in which the variables were selected according to the following criteria: effect and sample size, significance of univariate analysis (p < 0.05), and correlation of variables. The qualitative variables were analyzed using a contingency test as well as chi-square and/or Fisher’s exact tests, depending on the sample size and the testing for normal distribution for each set of samples. Differences with an α value ≤ 0.05 were considered to indicate statistical significance. Data were analyzed and graphs plotted using GraphPad Prism 8 for macOS version 8.0.1.

Results

Of a total of 1508 patients with tumors of the CNS, 78 (5.2%) presented with high-grade gliomas across all locations, excluding diffuse intrinsic pontine gliomas. The case distributions were equal between sexes. The ages at diagnosis were divided into three main groups: 0 to < 3 years, 15 patients (19.2%); 3–11 years, 36 patients (46.2%); and > 11 to < 19 years, 27 patients (34.6%). The mean age at diagnosis was 8.7 years.

The most frequently noted histological type was glioblastoma, which accounted for 54 (69.2%) cases, followed by anaplastic astrocytoma with 24 (30.8%) cases. The most commonly observed tumor location was the supratentorial region, as observed in 62 (79.5%) cases, followed by infratentorial and spinal cord locations with 8 (10.25%) cases each.

Thirty-four (43.6%) patients clinically presented with intracranial hypertension syndrome; 27 (79.4% of this group) experienced headache and 30 (88.2%) had vomiting. The incidence of papilledema in this symptomatic group was not reported. Thirty-three (42.3%) patients presented with focal deficits or focal neurological signs, of whom 26 (78.8%) had paresis, 23 (69.7%) showed cranial nerve alterations, and 10 (30.3%) presented with paresthesia. In 9 (11.5%) patients, the first symptoms were seizures. Two (2.6%) patients were diagnosed incidentally, one of them specifically prenatally (50%).

Sixty-seven (85.9%) patients underwent chemotherapy, 61 (78.2%) radiotherapy, and 53 (68%) both adjuvant treatments. One (1.3%) patient with congenital glioblastoma received an autologous stem cell transplant as part of the treatment. The patients’ demographic data are summarized in Table 1.

TABLE 1.

Summary of demographic data in 78 patients with high-grade glioma

VariableNo.%
Age
 0 to <3 yrs1519.2
 3 to 11 yrs3646.2
 >11 to <19 yrs2734.6
Sex
 Female3950
 Male3950
Clinical presentation
 Intracranial hypertension syndrome3443.6
 Focal deficits or signs3342.3
 Seizure911.5
 Incidental22.6
Histological type
 GBM5469.2
 AA2430.8
Tumor location
 Supratentorial6279.5
 Infratentorial810.25
 Spinal cord810.25
Resection at first surgery*
 GTR1117.7
 STR3353.2
 Biopsy1829
GTR and tumor location*
 Superficial1770.8
 Deep729.2
Partial resection or biopsy and tumor location*
 Superficial1436.8
 Deep2463.2
No. of patients with GTR at any point*
 GTR2438.7
 No GTR3861.3
No. of surgeries to achieve GTR*
 11145.8
 >11354.2
Adjuvant treatments
 Chemo6785.9
 RT6178.2
 Both chemo & RT5368
 Autologous stem cell transplantation11.3

AA = anaplastic astrocytoma; chemo = chemotherapy; GBM = glioblastoma; RT = radiotherapy.

Data regarding patients presenting with supratentorial tumors. The mean time until first resection or biopsy in patients with supratentorial tumors was 32.42 days.

The median OS of patients was 17 months, and the PFS was 6 months, with 1-, 2-, and 5-year OS rates of 62.7%, 36.35%, and 24.5%, respectively.

In the different age groups, the median overall survival durations in the age groups 0 to < 3 years, 3–11 years, and > 11 to < 19 years were 125, 16, and 14 months, respectively. The female group showed a median survival of 18 months, versus 15 months for males. Considering the different histological types, the median OS durations were 16 months for glioblastoma and 19 months for anaplastic astrocytoma, with a median OS for tumor with Ki-67 expression ≥ 50% of 14 months, in contrast to the expression < 50%, which was 24 months.

Patients with supratentorial tumors had a median overall survival duration of 17 months, whereas those with infratentorial or spinal cord tumors had a median overall survival duration of 12 months. Considering adjuvant treatment, patients who had combined treatment with chemo- and radiotherapy had a median OS of 19 months, whereas those who only underwent radiotherapy had a median survival of 11 months. Patients who underwent surgery without adjuvant treatment had a median OS of 5 months. Patients in whom treatment was initiated before 2010 had an OS duration of 11 months, whereas those treated from 2010 onward had a corresponding value of 23 months. The median OS curves are shown in Fig. 1.

FIG. 1.
FIG. 1.

Survival curves demonstrating OS in all patients (A), by age (B), sex (C), tumor location (D), histology (E), Ki-67 (F), first resection status (G), reoperation (H), adjuvant treatment (I), and treatment period (J). CT = chemotherapy; RT = radiotherapy. Figure is available in color online only.

Groups of patients treated before 2010 and from 2010 onward were compared according to the proportions of patients undergoing chemotherapy and radiotherapy, using the Fisher exact test. There was a significant difference between the number of patients who underwent regimens with temozolomide and irinotecan and radiotherapy during these intervals (p < 0.0001). However, there were no significant differences in age (p = 0.77), sex (0.16), or achievement of GTR during the first surgery (p = 0.43) between these groups.

Univariate Analysis

The median PFS duration was 17 months in the 0 to < 3 years age group, 5 months in the 3–11 years group, and 6 months in the > 11 to < 19 years group. In terms of the different histological types, the median progression durations were 5 and 7 months, respectively, for glioblastomas and anaplastic astrocytomas. The median PFS durations associated with the different locations were 6 months for infratentorial tumors and 7 months for supratentorial and spinal cord tumors.

The median PFS duration of those who underwent combined treatment was 7 months, whereas that of those who underwent surgery and chemotherapy was 10 months. The corresponding values were 4.5 months in those who underwent combined surgery and radiotherapy and 1 month in patients who underwent only surgery. Patients with initiation of treatment before 2010 had a median PFS duration of 5 months, whereas those treated from 2010 onward had a median survival duration of 6.5 months. The median PFS curves are shown in Fig. 2.

FIG. 2.
FIG. 2.

Survival curves demonstrating PFS in all patients (A), by age (B), sex (C), tumor location (D), histology (E), Ki-67 (F), first resection status (G), reoperation (H), adjuvant treatment (I), and treatment period (J). CT = chemotherapy; RT = radiotherapy. Figure is available in color online only.

Statistical analysis of the factors affecting OS showed significant differences in the following prognostic factors: age (p = 0.0038), sex (p = 0.0245), and those who received adjuvant treatment (p = 0.0001). Significant differences were also found between Ki-67 expression groups (p = 0.0429) and between patients treated before 2010 and from 2010 onward (p = 0.0092). No significant differences were observed in terms of histology and tumor location. These results are outlined in Table 2.

TABLE 2.

Summary of statistical data for OS in 78 patients with high-grade glioma

VariableNo.Median Survival (mos)p Value
Age0.0038
 0 to <3 yrs15125
 3 to 11 yrs3616
 >11 to <19 yrs2714
Sex0.0245
 Female3918
 Male3915
Histological type0.9904
 GBM5416
 AA2419
Ki-670.0429
 Ki-67 expression <50%3324
 Ki-67 expression ≥50%1214
Tumor location0.7579
 Supratentorial6217
 Infratentorial812
 Spinal cord812
Resection at first surgery*0.0175
 GTR1129
 STR3319
 Biopsy1813
Reop for GTR*0.1530
 GTR after reop1324
 STR only2616
 Biopsy only1212
Adjuvant treatments0.0001
 Both chemo & RT5319
 RT611
 Chemo12ND
 None75
Treatment period0.0092
 Before 20103211
 Between 2010 & 20194623

ND = not done.

Data regarding patients presenting with supratentorial tumors.

In terms of median PFS, the analysis revealed significant differences in the Ki-67 expression groups (p = 0.0135), whereas no significant differences were found regarding the presence of adjuvant treatments (p = 0.3423), or in patients’ age, sex, tumor histology, tumor location, and year of diagnosis. These results are summarized in Table 3.

TABLE 3.

Summary of statistical data for PFS in 78 patients with high-grade glioma

VariableNo.Median Survival (mos)p Value
Age
 0 to <3 yrs1517
 3 to 11 yrs3650.2891
 >11 to <19 yrs276
Sex
 Female3950.1889
 Male397
Histological type
 GBM5450.6345
 AA247
Ki-67
 Ki-67 expression <50%33130.0135
 Ki-67 expression ≥50%123.5
Tumor location
 Supratentorial627
 Infratentorial860.8293
 Spinal cord87
Resection at first surgery*
 GTR1112
 STR3350.0114
 Biopsy184
Reop for GTR*
 GTR after reop133
 STR only2650.6892
 Biopsy only123.5
Adjuvant treatments
 Both chemo & RT5370.3423
 RT64.5
 Chemo124
 None71
Treatment period
 Before 20103250.1076
 Between 2010 & 2019466.5

Data regarding patients presenting with supratentorial tumors.

Multivariate Analysis

In the multivariate analysis for OS, the following variables were included: age < 3 years, female sex, extent of resection in the first surgery of patients with supratentorial tumors, and presence of adjuvant treatment. Although Ki-67 index expression and treatment initiation after 2010 were significant in the univariate analysis, they were not considered in the model because of missing data in the Ki-67 expression group and due to collinearity concerns in the treatment initiation variable. Our results indicated that age < 3 years, female sex, and GTR in the first surgery are predictors of a better chance of survival. Also, the generated model showed high accuracy, with a calculated area under the curve of 0.84 (95% CI 0.73–0.96) with p < 0.0001. These results are summarized in Table 4.

TABLE 4.

Statistical data from multivariate analysis, considering death as the outcome in 78 patients with high-grade glioma

VariableMedian OS (mos)SR at 1 YrSR at 2 YrsSR at 5 YrsOR95% CIp Value
Age <3 yrs12580%73.3%40%0.070.01–0.41<0.0001
Female sex1864.1%38.5%23.1%0.120.02–0.5
GTR at first surgery*2982%54.5%45.4%0.120.02–0.7
Adjuvant treatment1960.6%33.8%18.3%3.060.25–32.8

SR = survival rate.

Data regarding patients presenting with supratentorial tumors.

Regarding PFS, multivariate analysis could not be generated because there were only two significant predictors, one of which had to be excluded due to missing data.

Supratentorial Tumors Analysis

Considering only the 62 (79.5%) supratentorial tumors, 24 (38.7%) patients underwent GTR, including 13 (54.2%) who underwent two or more surgeries. Considering the first surgery, 33 (53.2%) patients underwent STR and 18 (29%) biopsy. Thirteen (20.9%) patients underwent reoperation to achieve GTR, with 12 of them (92.3%) undergoing consecutive resections, without adjuvant treatment between surgeries. The mean time from diagnosis to the first resection or biopsy was 32.42 days, in which we did not consider other surgeries such as shunts to treat hydrocephalus.

Among the 24 (38.7%) patients who underwent GTR, 17 (70.8%) had superficially localized tumors, whereas 7 (29.2%) had deeply localized tumors. Of the 38 patients who underwent STR or only biopsy, 24 (63.2%) presented with deeply localized tumors, in locations such as thalamus, whereas 14 (36.8%) were superficial. Testing the difference in proportions of deep and superficial localized tumors between GTR and STR/biopsy groups, the Fisher exact test revealed a significant difference (p = 0.02). Demographic data from patients can be seen in Table 1.

Regarding the resection status of the first surgery, those who underwent GTR had a median OS duration of 29 months, whereas those who underwent STR or biopsy showed median survival durations of 19 and 13 months, respectively. Patients who underwent reoperation for GTR had a median OS of 24 months, whereas patients who underwent STR had 16 months and those who underwent biopsy had 12 months of median survival. The curves showing OS by extent of resection and reoperation can be seen in Fig. 1.

In terms of PFS regarding the extent of resection in the first surgery, patients who underwent GTR had a median PFS duration of 12 months, whereas those who underwent STR and biopsy had median PFS durations of 5 and 4 months, respectively. Patients who underwent reoperation for GTR had a median PFS duration of 3 months, whereas patients who underwent STR or biopsy had PFS durations of 5 and 3.5 months, respectively. The curves showing PFS regarding the extent of resection and reoperation can be seen in Fig. 2.

The statistical analysis for OS showed significant differences regarding the extent of resection in the first surgery, with the highest median OS duration observed in those who underwent GTR (p = 0.0175). In addition, no significant differences were found between the groups of patients who underwent GTR after reoperation and those who underwent biopsy or STR during treatment (p = 0.1530). The 5-year survival rates were 15.4% for the patients who underwent reoperation to achieve GTR, 19.2% for patients who underwent STR, and 8.3% for the biopsy group. The results regarding univariate analysis for OS are summarized in Table 2.

Regarding PFS, the univariate analysis revealed significant differences in first resection status, with GTR presenting the longer PFS (p = 0.0114). No significant differences were found between those who underwent reoperation to achieve GTR. These results are outlined in Table 3.

Considering only patients who achieved GTR in the first surgery or after reoperation, there were no significant differences in age, sex, tumor histology, depth of tumor location, adjuvant treatment, or Ki-67 expression. Additional data on patients who underwent reoperation are shown in Table 5.

TABLE 5.

Data from 13 patients with high-grade glioma who underwent reoperation

Case No.Age, Sex1st OpHistologyImmunohistochemistryTumor LocationNSChemoRTOS/PFS (mos)Death
111 yrs 6 mos, FSTRGBMH3K27: +; GFAP: +; ALK: −; synaptophysin: +; Ki-67: 20–30%; p53: 50–60%; IDH-1: +; ATRX: +; cyclin D1: +; INI1: +; CD34: + in vesselsParieto-occipitotemporal2Irinotecan & temozolomide60 Gy7/3No
26 yrs 7 mos, MSTRGBMGFAP: +; Ki-67: 90%; neurofilament: −; INI1: +Thalamus2Irinotecan60 Gy & reirradiation w/ 40 Gy9/6Yes
38 yrs 9 mos, MSTRGBMGFAP: +; neurofilament: −; CD34: + in vessels; Ki-67: 60%; INI1: +; enolasis: −; chromogranin: −; synaptophysin: +; CD56: +Frontoparietal2Irinotecan & temozolomide | vincristine, lomustine, & procarbazine59.4 Gy & reirradiation w/ 59.4 Gy37/20Yes
415 yrs 7 mos, MSTRGBMp53: 10%; IDH-1: −; Ki-67: 10–30%; BRAF: −; GFAP: +Parietal2Irinotecan & temozolomide60 Gy24/24Yes
52 yrs 11 mos, FSTRGBMParietal2ND59.4 Gy11/3Yes
64 yrs 3 mos, FBiopsyAAParietal2ND50.4 Gy15/2Yes
74 mos, FSTRGBMGFAP: +; synaptophysin: −; INI1: +; Ki-67: 25%; p53: 25%Hemispheric3Vincristine, etoposide, & cisplatinND17/17No
816 yrs, MSTRGBMParieto-occipital3ND60 Gy7/5Yes
95 mos, FSTRGBMGFAP: +; IDH-1: +; INI1: +; Ki-67: 25–50%; p53: 90%Hemispheric3Temozolomide & irinotecan | cisplatin, cyclophosphamide, etoposide, & vincristine | methotrexate, cisplatin, vincristine, & etoposideND32/3No
101 yr 4 mos, MSTRGBMGFAP: +; Ki-67: 40%; INI1: +; EMA: +; AE1–AE3: −; enolasis: +; chromogranin: −; myeloperoxidase: + in neutrophils permeating the tumor; LCA: + in typical lymphocytesTemporal3Cisplatin, cyclophosphamide, etoposide, & vincristine59.4 Gy49/19Yes
110, F*STRGBMGFAP: +; EMA: −; S-100: +; enolasis: +; synaptophysin: −; chromogranin: −; neurofilament: −; desmin: −; 1A4: −; CD56: +; Ki-67: 15–20%Lateral ventricle3Cisplatin, cyclophosphamide, etoposide, & vincristineND83/3No
126 yrs 10 mos, FSTRGBMKi-67: 60%; neurofilament: −; CD56: −; GFAP: +; synaptophysin: −; monoclonal enolasis: +; chromogranin: −; vimentin: +Parietal3Cisplatin, cyclophosphamide, etoposide, & vincristine | temozolomide594 Gy9/2Yes
130, F*STRGBMCD99: −; INI1: +; EMA: −; S100: −; GFAP: +; synaptophysin: −; chromogranin: −; Ki-67: 90%; enolasis: −Lateral ventricle5Carboplatin & etoposide | cisplatin, cyclophosphamide, & etoposide | irinotecan & temozolomide | ifosfamide, carboplatin, & etoposide | vinblastineND50/2No

NS = number of surgeries to achieve GTR; + = positive; − = negative; | = separates subsequent chemotherapy protocols.

Diagnosis made at birth or prenatally.

Discussion

In this study we found that the median OS duration of patients with supratentorial tumors who underwent GTR in the first surgery was longer than that of patients who received STR and biopsy (29, 19, and 13 months, respectively). In addition, we found no significant differences between the survival curves of patients who underwent reoperation for GTR and those with STR and biopsy. This is revealed by the 5-year survival rates of these groups, with 15.4%, 19.2%, and 8.3% of patients alive in the GTR, STR, and biopsy groups, respectively, showing in a long-term comparison of these groups that patients have similar outcomes as shown in the Kaplan-Meier curve. Therefore, our data indicate the impact of the first surgery on the survival curve of patients, because patients with supratentorial tumors who underwent reoperation do not have longer survival durations than those without GTR.

Perkins et al. found significant differences in the 2-year survival values between those who underwent GTR (60%) and STR or biopsy (25%),17 and McCrea et al. observed differences in the median OS value of patients treated with GTR (40.8 months) and biopsy (15.6 months).9 In another cohort of patients, Yang et al. showed differences in the survival values between patients who underwent GTR (45.1 months) and those with STR or biopsy (8.7 and 11.5 months, respectively).11 None of those studies, however, reported on the impact of reoperation on survival. However, similar to our findings, the results published by those authors highlighted the importance of the first surgery for patient survival.

Another key issue is the extent of resection regarding the tumor location. We found significant differences in the analysis of the primary tumor as deep or superficial and its corresponding extent of resection, without a significant difference on survival durations of those groups when considering only the depth (p = 0.66). Patients with lobar tumors more frequently underwent GTR (70.8% vs 36.8%), indicating a greater degree of difficulty in the surgical access to deep tumors. In their cohort, McCrea et al. demonstrated that patients with supratentorial tumors had longer survival durations than those with thalamic, brainstem, or cerebellum tumors (57.6, 31.2, 24, and 45.6 months, respectively).9

We also found significant differences in terms of age, with the group that was 0 to < 3 years of age showing greater survival values. This may be related to the biology of these tumors; these individuals had a median OS duration of 125 months, whereas those in the group that was > 11 to < 19 years old showed corresponding values of only 14 months, similar to the findings of Lam et al., indicating that younger age is associated with longer survival durations.10

Significant differences in the median OS were also noted between adjuvant treatments. The use of chemotherapy and radiotherapy after surgery was associated with a survival duration of 19 months, whereas the OS value related to surgery alone was 5 months, highlighting the effect of adjuvant treatment on the survival curve of these patients. The patients who underwent surgery alone were those who experienced early progression or dissemination a few weeks after surgery, not allowing the use of adjuvant treatment. Boudaouara et al. demonstrated differences in the survival values associated with radiotherapy, with patients who received this treatment showing a median OS value of 30 months, and those who did not receive it showing a median OS duration of 5 months.18

The survival of patients treated before 2010 (median OS of 11 months) and from 2010 onward (23 months) may also reflect the effect of the adjuvant treatment. During the first decade of the 2000s, our institution underwent a series of changes reflected also in the treatment of high-grade gliomas. During this period a new radiotherapy device was acquired by our hospital, which reduced the waiting time for radiotherapy. We incorporated protocols using temozolomide combined with irinotecan into our chemotherapy regimens, and our analysis also revealed different proportions of adjuvant treatment regimens from 2010 onward. This hypothesis is in line with the results of the study conducted by Jakacki et al., who favored the use of temozolomide as an adjuvant treatment to resection and radiotherapy.19 In addition, because younger age is a significant prognostic factor for longer survival durations, another possibility is the presence of a larger number of individuals younger than 3 years in this group, proportionally, compared to those treated before 2010, although no significant difference was found in the chi-square tests.

In this study we found significant differences in the median OS durations between females (18 months) and males (15 months), in line with the findings reported by McCrea et al., who showed higher survival values in girls (57.6 months) than in boys (33.6 months).9

Regarding PFS, we found significant differences in the following factors: extent of resection of the first surgery and adjuvant treatment. Patients who underwent GTR (12 months) had greater median PFS than those with STR (5 months) or biopsy (4 months). Patients who underwent adjuvant treatment had greater median PFS (7 months) than those with only surgery (1 month). Jung et al. reported a significant difference in the status of the first resection, with longer PFS noted in patients with GTR than in those with STR and biopsy (15, 7.8, and 8.5 months, respectively).14 Similarly, Muhammed et al. reported that 30% and 52% of those with total resection and STR, respectively, in the first surgery showed progression in 1 year.20 Jung et al. demonstrated longer PFS durations in patients who underwent chemotherapy and radiotherapy (10.5 months) than in those who underwent only radiotherapy (4.7 months).

The impact on the PFS of the reoperation to achieve GTR was not noted either. Although the group who underwent GTR after initial surgery had the shorter survival (3 months vs 3.5 and 5 months for the biopsy and STR groups, respectively), this difference is not statistically demonstrated considering the whole observation period. This is better illustrated by comparison of 1-, 2-, and 5-year PFS rates: the group who underwent reoperation for GTR had rates of 38.5%, 7.7%, and 7.7%, whereas the STR group had rates of 34.6%, 19.2%, and 11.5%. Finally, the biopsy group had rates of 25%, 16.7%, and 8.3% in the 1st, 2nd, and 5th years, respectively. This shows that despite the specific difference at the 2-year follow-up, for example, the final outcome is similar in these groups, which explains the absence of a significant difference (p = 0.69).

In contrast to our study, McCrea et al. demonstrated a longer median PFS in females (14.4 months) than in males (6 months).9

Regarding Ki-67 expression, we have found significant differences between patients with Ki-67 expression of < 50%, with longer OS and PFS. These results may be due to tumor aggressiveness because Ki-67 is a proliferation marker. In contrast to our study, Alkhaibary et al. found no significant difference between tumors with Ki-67 expression ≤ 27% when analyzing a cohort of 44 patients.21

The absence of a unique method to determine lesion size due to the lack of digital images from past patients limited the evaluation of the extent of resection. Alternatively, we defined GTR strictly as a 100% resection of the tumor, which is more rigorous than that reported in other cohorts.22,23 However, this may also have underestimated the proportion of patients who underwent GTR at our institution.

Our results highlight the role of reoperation, emphasizing the absence of difference on survival curves. No similar studies have been published yet. Further research with larger cohorts may be needed to establish that.

The limitations of our study include its retrospective design and the sample size, although our study included one of the largest cohorts ever reported in the literature. Moreover, the influence of the institutional philosophy and surgeon aggressiveness may limit the study generalizability.

Conclusions

Our cohort study reiterates how challenging the approach to these neoplasms is and highlights the role of the surgeon in the achievement of GTR right from the first surgery, because this factor is subject to intervention and alters the duration of survival. In addition, reoperation for GTR does not appear to have a significant impact on OS in patients with supratentorial high-grade gliomas, further reinforcing the role of the first surgery. Finally, the importance of combined chemotherapy and radiotherapy is reinforced through the increase observed in the survival values of these patients.

Acknowledgments

This study was funded by the National Council of Scientific and Technological Development (Conselho Nacional de Desenvolvimento Científico e Tecnológico—CNPq), which awarded a scientific initiation scholarship to Miss Camargo. We thank Marta Maria da Silva for helping with the patients’ charts and Daniela Barbosa de Almeida, clinical nurse, who helped in clinical assistance and protocols management.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Silva da Costa, Camargo, Saba da Silva, Cavalheiro. Acquisition of data: Silva da Costa, Camargo. Analysis and interpretation of data: Silva da Costa, Camargo. Drafting the article: Silva da Costa, Camargo. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Silva da Costa. Statistical analysis: Silva da Costa, Camargo. Study supervision: Saba da Silva, Cavalheiro.

References

  • 1

    Ostrom QT, Gittleman H, Truitt G, et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2011-2015. Neuro Oncol. 2018;20(suppl_4):iv1iv86.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Ostrom QT, de Blank PM, Kruchko C, et al. Alex’s Lemonade Stand Foundation infant and childhood primary brain and central nervous system tumors diagnosed in the united states in 2007–2011. Neuro Oncol. 2015;16(suppl 10):x1x36.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    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):803820.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Gerges N, Fontebasso AM, Albrecht S, et al. Pediatric high-grade astrocytomas: a distinct neuro-oncological paradigm. Genome Med. 2013;5(7):6677.

  • 5

    El-Ayadi M, Ansari M, Sturm D, et al. High-grade glioma in very young children: a rare and particular patient population. Oncotarget. 2017;8(38):6456464578.

  • 6

    Chamdine O, Gajjar A. Molecular characteristics of pediatric high-grade gliomas. CNS Oncol. 2014;3(6):433443.

  • 7

    Paugh BS, Qu C, Jones C, et al. Integrated molecular genetic profiling of pediatric high-grade gliomas reveals key differences with the adult disease. J Clin Oncol. 2010;28(18):30613068.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Tamber MS, Rutka JT. Pediatric supratentorial high-grade gliomas. Neurosurg Focus. 2003;14(2):e1.

  • 9

    McCrea HJ, Bander ED, Venn RA, et al. Sex, age, anatomic location, and extent of resection influence outcomes in children with high-grade glioma. Neurosurgery. 2015;77(3):443453.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Lam S, Lin Y, Zinn P, et al. Patient and treatment factors associated with survival among pediatric glioblastoma patients: a Surveillance, Epidemiology, and End Results study. J Clin Neurosci. 2018;47:285293.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Yang T, Temkin N, Barber J, et al. Gross total resection correlates with long-term survival in pediatric patients with glioblastoma. World Neurosurg. 2013;79(3-4):537544.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Walston S, Hamstra DA, Oh K, et al. A multi-institutional experience in pediatric high-grade glioma. Front Oncol. 2015;5:28.

  • 13

    Nikitović M, Stanić D, Pekmezović T, et al. Pediatric glioblastoma: a single institution experience. Childs Nerv Syst. 2016;32(1):97103.

  • 14

    Jung TY, Lee JY, Kim DS, et al. Pediatric supratentorial high-grade glioma: multicenter retrospective observational study of the Korean Society for Pediatric Neuro-Oncology. J Neurooncol. 2015;121(2):413419.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    Finlay JL, Boyett JM, Yates AJ, et al. Randomized phase III trial in childhood high-grade astrocytoma comparing vincristine, lomustine, and prednisone with the eight-drugs-in-1-day regimen. J Clin Oncol. 1995;13(1):112123.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Sreenivasan SA, Madhugiri VS, Sasidharan GM, Kumar RV. Measuring glioma volumes: A comparison of linear measurement based formulae with the manual image segmentation technique. J Cancer Res Ther. 2016;12(1):161168.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Perkins SM, Rubin JB, Leonard JR, et al. Glioblastoma in children: a single-institution experience. Int J Radiat Oncol Biol Phys. 2011;80(4):11171121.

  • 18

    Boudaouara O, Charfi S, Bahri M, et al. Pediatric high grade gliomas: Clinico-pathological profile, therapeutic approaches and factors affecting overall survival. Neurochirurgie. 2019;65(2-3):6368.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Jakacki RI, Cohen KJ, Buxton A, et al. Phase 2 study of concurrent radiotherapy and temozolomide followed by temozolomide and lomustine in the treatment of children with high-grade glioma: a report of the Children’s Oncology Group ACNS0423 study. Neuro Oncol. 2016;18(10):14421450.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Muhammed A, Gaber MS, Elbeltagy M, et al. Risk stratification of pediatric high-grade glioma: a newly proposed prognostic score. Childs Nerv Syst. 2019;35(12):23552362.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Alkhaibary A, Alassiri AH, AlSufiani F, Alharbi MA. Ki-67 labeling index in glioblastoma; does it really matter? Hematol Oncol Stem Cell Ther. 2019;12(2):8288.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Bloch O, Han SJ, Cha S, et al. Impact of extent of resection for recurrent glioblastoma on overall survival: clinical article. J Neurosurg. 2012;117(6):10321038.

  • 23

    Wisoff JH, Boyett JM, Berger MS, et al. Current neurosurgical management and the impact of the extent of resection in the treatment of malignant gliomas of childhood: a report of the Children’s Cancer Group trial no. CCG-945. J Neurosurg. 1998;89(1):5259.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • Collapse
  • Expand
  • View in gallery
    FIG. 1.

    Survival curves demonstrating OS in all patients (A), by age (B), sex (C), tumor location (D), histology (E), Ki-67 (F), first resection status (G), reoperation (H), adjuvant treatment (I), and treatment period (J). CT = chemotherapy; RT = radiotherapy. Figure is available in color online only.

  • View in gallery
    FIG. 2.

    Survival curves demonstrating PFS in all patients (A), by age (B), sex (C), tumor location (D), histology (E), Ki-67 (F), first resection status (G), reoperation (H), adjuvant treatment (I), and treatment period (J). CT = chemotherapy; RT = radiotherapy. Figure is available in color online only.

  • 1

    Ostrom QT, Gittleman H, Truitt G, et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2011-2015. Neuro Oncol. 2018;20(suppl_4):iv1iv86.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Ostrom QT, de Blank PM, Kruchko C, et al. Alex’s Lemonade Stand Foundation infant and childhood primary brain and central nervous system tumors diagnosed in the united states in 2007–2011. Neuro Oncol. 2015;16(suppl 10):x1x36.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    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):803820.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Gerges N, Fontebasso AM, Albrecht S, et al. Pediatric high-grade astrocytomas: a distinct neuro-oncological paradigm. Genome Med. 2013;5(7):6677.

  • 5

    El-Ayadi M, Ansari M, Sturm D, et al. High-grade glioma in very young children: a rare and particular patient population. Oncotarget. 2017;8(38):6456464578.

  • 6

    Chamdine O, Gajjar A. Molecular characteristics of pediatric high-grade gliomas. CNS Oncol. 2014;3(6):433443.

  • 7

    Paugh BS, Qu C, Jones C, et al. Integrated molecular genetic profiling of pediatric high-grade gliomas reveals key differences with the adult disease. J Clin Oncol. 2010;28(18):30613068.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Tamber MS, Rutka JT. Pediatric supratentorial high-grade gliomas. Neurosurg Focus. 2003;14(2):e1.

  • 9

    McCrea HJ, Bander ED, Venn RA, et al. Sex, age, anatomic location, and extent of resection influence outcomes in children with high-grade glioma. Neurosurgery. 2015;77(3):443453.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Lam S, Lin Y, Zinn P, et al. Patient and treatment factors associated with survival among pediatric glioblastoma patients: a Surveillance, Epidemiology, and End Results study. J Clin Neurosci. 2018;47:285293.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Yang T, Temkin N, Barber J, et al. Gross total resection correlates with long-term survival in pediatric patients with glioblastoma. World Neurosurg. 2013;79(3-4):537544.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Walston S, Hamstra DA, Oh K, et al. A multi-institutional experience in pediatric high-grade glioma. Front Oncol. 2015;5:28.

  • 13

    Nikitović M, Stanić D, Pekmezović T, et al. Pediatric glioblastoma: a single institution experience. Childs Nerv Syst. 2016;32(1):97103.

  • 14

    Jung TY, Lee JY, Kim DS, et al. Pediatric supratentorial high-grade glioma: multicenter retrospective observational study of the Korean Society for Pediatric Neuro-Oncology. J Neurooncol. 2015;121(2):413419.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    Finlay JL, Boyett JM, Yates AJ, et al. Randomized phase III trial in childhood high-grade astrocytoma comparing vincristine, lomustine, and prednisone with the eight-drugs-in-1-day regimen. J Clin Oncol. 1995;13(1):112123.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Sreenivasan SA, Madhugiri VS, Sasidharan GM, Kumar RV. Measuring glioma volumes: A comparison of linear measurement based formulae with the manual image segmentation technique. J Cancer Res Ther. 2016;12(1):161168.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Perkins SM, Rubin JB, Leonard JR, et al. Glioblastoma in children: a single-institution experience. Int J Radiat Oncol Biol Phys. 2011;80(4):11171121.

  • 18

    Boudaouara O, Charfi S, Bahri M, et al. Pediatric high grade gliomas: Clinico-pathological profile, therapeutic approaches and factors affecting overall survival. Neurochirurgie. 2019;65(2-3):6368.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Jakacki RI, Cohen KJ, Buxton A, et al. Phase 2 study of concurrent radiotherapy and temozolomide followed by temozolomide and lomustine in the treatment of children with high-grade glioma: a report of the Children’s Oncology Group ACNS0423 study. Neuro Oncol. 2016;18(10):14421450.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Muhammed A, Gaber MS, Elbeltagy M, et al. Risk stratification of pediatric high-grade glioma: a newly proposed prognostic score. Childs Nerv Syst. 2019;35(12):23552362.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Alkhaibary A, Alassiri AH, AlSufiani F, Alharbi MA. Ki-67 labeling index in glioblastoma; does it really matter? Hematol Oncol Stem Cell Ther. 2019;12(2):8288.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Bloch O, Han SJ, Cha S, et al. Impact of extent of resection for recurrent glioblastoma on overall survival: clinical article. J Neurosurg. 2012;117(6):10321038.

  • 23

    Wisoff JH, Boyett JM, Berger MS, et al. Current neurosurgical management and the impact of the extent of resection in the treatment of malignant gliomas of childhood: a report of the Children’s Cancer Group trial no. CCG-945. J Neurosurg. 1998;89(1):5259.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 2036 0 0
Full Text Views 1404 1133 35
PDF Downloads 810 584 37
EPUB Downloads 0 0 0