Patterns and quality of therapeutic care for glioma: a population-based study in Belgium between 2016 and 2019 - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice ESMO Open . 2026 Apr 9;11(4):106303. doi: 10.1016/j.esmoop.2026.106303 Search in PMC Search in PubMed View in NLM Catalog Add to search Patterns and quality of therapeutic care for glioma: a population-based study in Belgium between 2016 and 2019 D Vanhauwaert D Vanhauwaert 1 Department of Neurosurgery, AZ Delta, Roeselare, Belgium Find articles by D Vanhauwaert 1, ∗ , K Vanschoenbeek K Vanschoenbeek 2 Belgian Cancer Registry, Brussels, Belgium Find articles by K Vanschoenbeek 2 , G Silvermit G Silvermit 2 Belgian Cancer Registry, Brussels, Belgium Find articles by G Silvermit 2 , A Lombard A Lombard 3 Department of Neurosurgery, CHR Citadelle, Liège, Belgium 4 Department of Neurosurgery, CHU Liège, Liège, Belgium Find articles by A Lombard 3, 4 , N Liefhooghe N Liefhooghe 5 Department of Radiation Oncology, AZ Groeninge, Kortrijk, Belgium Find articles by N Liefhooghe 5 , M Lammens M Lammens 6 Department of Pathology, Antwerp University Hospital and CORE University of Antwerp, Antwerp, Belgium 7 Department of Pathology, CHU St. Luc, Brussels, Belgium 8 Department of Pathology, AZ St. Jan, Brugge, Belgium Find articles by M Lammens 6, 7, 8 , F Dedeurwaerdere F Dedeurwaerdere 9 Department of Pathology, AZ Delta, Roeselare, Belgium Find articles by F Dedeurwaerdere 9 , N Whenham N Whenham 10 Department of Oncology, Clinique St. Pierre, Ottignies, Belgium 11 Department of Medical Oncology, CHU St. Luc, Brussels, Belgium Find articles by N Whenham 10, 11 , C De Gendt C De Gendt 2 Belgian Cancer Registry, Brussels, Belgium Find articles by C De Gendt 2 , T Boterberg T Boterberg 12 Department of Radiation Oncology, Ghent University Hospital, Ghent, Belgium Find articles by T Boterberg 12 , S De Vleeschouwer S De Vleeschouwer 13 Department of Neurosurgery, UZ Leuven, Leuven, Belgium 14 Department of Neurosciences and Leuven Brain Institute (LBI), KU Leuven, Leuven, Belgium Find articles by S De Vleeschouwer 13, 14 Author information Article notes Copyright and License information 1 Department of Neurosurgery, AZ Delta, Roeselare, Belgium 2 Belgian Cancer Registry, Brussels, Belgium 3 Department of Neurosurgery, CHR Citadelle, Liège, Belgium 4 Department of Neurosurgery, CHU Liège, Liège, Belgium 5 Department of Radiation Oncology, AZ Groeninge, Kortrijk, Belgium 6 Department of Pathology, Antwerp University Hospital and CORE University of Antwerp, Antwerp, Belgium 7 Department of Pathology, CHU St. Luc, Brussels, Belgium 8 Department of Pathology, AZ St. Jan, Brugge, Belgium 9 Department of Pathology, AZ Delta, Roeselare, Belgium 10 Department of Oncology, Clinique St. Pierre, Ottignies, Belgium 11 Department of Medical Oncology, CHU St. Luc, Brussels, Belgium 12 Department of Radiation Oncology, Ghent University Hospital, Ghent, Belgium 13 Department of Neurosurgery, UZ Leuven, Leuven, Belgium 14 Department of Neurosciences and Leuven Brain Institute (LBI), KU Leuven, Leuven, Belgium ∗ Correspondence to: Dr Dimitri Vanhauwaert, Department of Neurosurgery, AZ Delta Hospital Roeselare, Deltalaan 1, 8800 Roeselare, Belgium. Tel: +32-51237446 [email protected] Collection date 2026 Apr. © 2026 The Author(s) This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). PMC Copyright notice PMCID: PMC13091517 PMID: 41962311 Abstract Background The study aimed to evaluate the patterns and quality of therapeutic care and pathological diagnosis for glioma patients by calculating process indicators across Belgian hospitals. Materials and methods All Belgian patients newly diagnosed with glioma between 2016 and 2019 were identified in the Belgian Cancer Registry (BCR). The National Social Security Number was used to link pathology reports and databases providing information on diagnostic and therapeutic procedures and vital status. Eighteen process indicators for therapeutic care and seven for pathological diagnosis were assessed for measurability based on administrative data availability. Measurable indicators were reviewed by a multidisciplinary expert board, and a validation study was conducted in six pilot hospitals. Associations between process indicators and outcomes were explored where possible. Results Nine indicators for therapeutic care and five for pathological diagnosis were measurable, though five required reformulations due to shortcomings in administrative data, mainly regarding radiotherapy dose and number of fractions. Most indicators met the predefined targets, indicating the provision of qualitative care. No low-grade gliomas received chemo- or radiotherapy without prior pathological confirmation. Concomitant chemoradiotherapy was initiated in 93.1% of surgically treated and 88.9% of biopsied glioblastoma patients (target 90%). For grade 3 astrocytoma and oligodendroglioma, the use of adjuvant chemo- and radiotherapy increased over time, meeting the 90% target by 2019. However, timely initiation of chemoradiotherapy (within 6 weeks of resection) in high-grade glioma could be improved (79.8% versus target 95%). The proportion of patients receiving chemotherapy (3.5%) or radiotherapy (2.5%) in the last 14 days of life remained below the 5% target. Conclusions This research provides valuable feedback to Belgian hospitals and health care providers in neuro-oncology care programmes, opening opportunities for quality enhancement. On an international level, it underscores the significance of standardized quality assessment methods in neuro-oncology, which could serve as a foundation for future benchmark initiatives. Key words: brain tumours, quality of care, chemotherapy, radiotherapy, surgery, glioma Highlights • Process indicators for therapeutic care for glioma were calculated. • Population- and hospital-level results indicate high-quality care in Belgium. • Timely initiation of chemoradiotherapy can be improved. • Unprecise assessment of radiotherapy dose and number of fractions was identified. • Internationally, awareness of quality assessment in neuro-oncology should increase. Introduction Quality assessments through process and outcome indicators in the field of neuro-oncology are scarce, in contrast to other oncological domains. Our group recently developed a set of both process and outcome indicators that could provide insights into care paths and through which an estimation of the quality of care for glioma patients can be made. 1 The objective of this study is to calculate process indicators assessing (therapeutic) care and pathological diagnosis in glioma patients across Belgian hospitals. Whenever possible, comparisons with earlier time frames (2008-2011 and 2012-2015) were conducted to identify evolving trends. In addition, associations between the evaluated process indicators and outcome were explored. Outcome indicators such as 30-day mortality and survival for glioblastoma, and indicators related to the diagnostic process and imaging have been published previously by our group. 2 , 3 Materials and methods Data collection and study cohort In Belgium, cancer registration is compulsory and all new diagnoses ought to be reported to the Belgian Cancer Registry (BCR) by the so-called ‘oncological care programmes’ of Belgian hospitals and also independently by the laboratories of pathological anatomy. Incidence date, basis of diagnosis (e.g. autopsy, histology of primary tumour, technical exam, etc.), International Classification of Diseases for Oncology, third edition (ICD-O-3) topography and morphology code, World Health Organization (WHO) performance status and performed and/or planned treatments are among the required data. 4 Since 2004, the BCR has been legally charged with the collection of data on all new oncological diseases, and the completeness of incidence in the database is estimated to be 98%. 5 For this study, we identified all newly diagnosed gliomas between 2016 and 2019 in the BCR database. Inclusion criteria were patients aged ≥18 years, official residence in Belgium at diagnosis and known National Social Security Number (NSSN). The following exclusion criteria applied: patients without reimbursement data for diagnostic or therapeutic procedures within the time frame of 1 month before until 3 months after incidence date, patients whose incidence date corresponded with the date of death and patients lost to follow-up from the day of incidence. The NSSN served as a unique identifier to link the BCR database with other data sources. Firstly, the database of the Intermutualistic Agency (IMA) provides details on all diagnostic and therapeutic procedures reimbursed by the national health insurance. Data are available from 1 January of the year preceding the incidence year up to 31 December of the fifth year after the incidence year. However, a delay of 2 years should be taken into account before IMA data are considered 100% complete. Secondly, the vital status of the patients is retrieved from the Crossroad Bank of Social Security. Thirdly, molecular marker information is manually extracted from pathology reports derived from the laboratories of pathological anatomy. An approval for linkage of the databases is provided by the Data Protection Authority (DPA), previously known as the Belgian Privacy Commission. 5 This study was approved by the Ethics Committee of Ghent University Hospital (DA 2020-042). Written informed consent from participants was not required in accordance with the local/national guidelines. Process indicators From the previously published list of indicators for assessing glioma care, 17 process indicators were related to treatment (surgery, chemotherapy and radiotherapy) and 7 to pathology ( Table 1 ). 1 One additional indicator assessing the number of deceased glioma patients who received chemo- or radiotherapy within 14 days before death was added. Measurability was judged based on the availability of administrative data for every single element in the numerator and denominator of an indicator. To overcome shortcomings in the available data, some indicators were reformulated to become measurable. In line with the methodology used by the Belgian Healthcare Knowledge Centre (KCE) in other projects, for each indicator a technical documentation sheet (TDS) was developed, detailing the rationale, possibly required reformulation, numerator and denominator of the calculation, target value (as defined by a multidisciplinary team of experts based on relevant literature and international preceding experience, if available), data sources and technical definitions, risk adjustment if indicated, limitations, subgroup analyses, sensitivity analyses, benchmarking and international results if available. Also time frames were defined wherein biopsy, surgical resection, primary or adjuvant chemo- and radiotherapy were considered relevant for the calculation. 6 Before definite calculation, these TDSs were reviewed by a multidisciplinary expert board. By agreement, one outcome indicator was partially converted into a process indicator and added to the indicator set related to therapeutic care (see ‘Discussion’ section). In order to assess concordance between diagnostic and therapeutic procedures identified in the administrative population-based database (IMA) and the information directly available in the hospitals (e.g. medical files), a validation study and data checks were carried out in six pilot hospitals. The complete TDSs can be found in Supplementary Material S1 , available at https://doi.org/10.1016/j.esmoop.2026.106303 . Table 1. Initial QIs, reformulation based on availability of administrative data, applied allocation algorithm, target and limitations QI Original number Quality indicator Reformulation Allocation to centre of Target Limitations 12 Proportion of patients with a histological diagnosis of glioblastoma with no measurable residual tumour demonstrated by post-operative MRI AND no new neurological deficit after surgery a Residual tumour not identifiable New neurological deficit not identifiable 13 Proportion of patients who underwent surgery for low-grade glioma with no measurable residual tumour demonstrated by MRI and no new neurological deficit at 3 months after surgery a Residual tumour not identifiable New neurological deficit not identifiable T1 14 Proportion of patients with suspected low-grade glioma receiving chemotherapy or radiotherapy without pathological confirmation Proportion of patients with low-grade (grade 2) glioma without pathological confirmation before chemo- and/or radiotherapy b Chemo- or radiotherapy 0% 22 Proportion of patients where the radiotherapy planning process included MRI fusion a Fusion of radiotherapy planning with MRI not identifiable 23 Proportion of patients undergoing radiotherapy where dose-volume histograms of critical structures were calculated a Radiotherapy plans and dose-volume histograms not identifiable 24 Proportion of patients who were provided chemotherapy education AND for whom informed consent was obtained before prescription of chemotherapy a Provided information and informed consent not identifiable T2 26 Proportion of patients with glioblastoma <70 years of age with KPS >60 in whom concomitant chemoradiotherapy is initiated (Stupp protocol) Proportion of patients with a glioblastoma <70 years of age with a WHO performance score ≤2 (or missing) in whom concomitant chemoradiotherapy is initiated (Stupp protocol) b , c Surgery or biopsy 90% 27 Proportion of patients with newly diagnosed glioblastoma receiving bevacizumab outside a clinical trial a Participation to clinical trial cannot be identified 29 Proportion of patients with newly diagnosed glioblastoma >65 years of age treated with either TMZ or radiotherapy (not Stupp protocol) in whom the MGMT promotor methylation status is analysed before treatment a Data on MGMT promotor methylation test performance and test results not accurately available MGMT promotor methylation test not reimbursed by national health insurance T3 30 Proportion of patients with grade 3 oligodendroglioma treated with radiotherapy and adjuvant chemotherapy (at least four cycles of PCV) Proportion of anaplastic (grade 3) oligodendroglioma patients who receive both (adjuvant) radiotherapy and chemotherapy b , c Surgery or biopsy 90% T4 31 Proportion of patients with grade 3 astrocytoma treated with radiotherapy and adjuvant 12 cycles of TMZ Proportion of anaplastic (grade 3) astrocytoma patients who received both (adjuvant) radiotherapy and chemotherapy b , c Surgery or biopsy 90% T5 25 Proportion of patients with HGG who commence oncological treatment (chemotherapy, radiotherapy or chemoradiotherapy) within 6 weeks of surgery Proportion of patients with HGG who commence oncological treatment (chemotherapy, radiotherapy or chemoradiotherapy) within 6 weeks of surgical resection b Surgery 95% T6 45 Proportion of patients with glioma who die within 30 days of treatment (surgery, radiotherapy or chemotherapy) Proportion of deceased glioma patients who received chemotherapy or radiotherapy within 14 days before death b Last care (chemotherapy or radiotherapy) <5% Total dose, number of fractions and dose per fraction not identifiable T7 28 Proportion of patients with HGG <70 years of age with good performance status (KPS ≥60) who receive radiotherapy in doses of 60 Gy in 2 Gy daily fractions Proportion of patients with high-grade (grade 3/4) glioma <70 years of age with a WHO performance score ≤2 (or missing) who started a long course of radiotherapy b , c Radiotherapy 95% Total dose, number of fractions and dose per fraction not identifiable T8 32 Proportion of patients treated with radiotherapy for anaplastic astrocytoma receiving a maximum total dose of 59.4-60 Gy Proportion of patients with anaplastic (grade 3) astrocytoma who started a long course of radiotherapy b , c Radiotherapy 95% Total dose, number of fractions and dose per fraction not identifiable 33 Proportion of patients operated for grade II ependymoma with residual tumour who receive post-operative radiotherapy with a dose of 45-54 Gy a Residual tumour not identifiable Total dose, number of fractions and dose per fraction not identifiable T9 34 Proportion of patients operated for grade III ependymoma who receive post-operative radiotherapy with a dose of 54-60 Gy Proportion of patients with anaplastic (grade 3) ependymoma who started a long course of adjuvant radiotherapy b Radiotherapy 95% Total dose, number of fractions and dose per fraction not identifiable 35 Proportion of patients with metastatic disease of grade III ependymoma who receive craniospinal axis radiotherapy a Metastatic disease of ependymoma not identifiable Total dose, number of fractions and dose per fraction not identifiable P1a 15 (i) Proportion of glioma patients with a diffuse astrocytoma, oligodendroglioma or glioblastoma for whom the histopathological diagnosis in their pathology reports integrates the molecular status correctly in an entity as defined in the WHO 2016 classification Pathology P1b 15 (ii) Proportion of glioma patients excluding diffuse astrocytoma, oligodendroglioma and glioblastoma for whom the histopathological diagnosis in their pathology reports is an entity as defined in the WHO 2016 classification Pathology P2 16 Proportion of patients with a diagnosis of oligodendroglioma for whom a tissue sample was tested for 1p/19q codeletion Proportion of oligodendroglioma patients in whom 1p19q codeletion is identified b Pathology P3 17 Proportion of glioma patients for whom the tumour was tested for IDH-1/2 mutation (i) Proportion of glioma patients with a diffuse astrocytoma, oligodendroglioma or glioblastoma who were tested for IDH with IDH1-IHC or IDH1/ 2 -NGS b Pathology P4 21 Proportion of patients with diffuse glioma <55 years of age who had sequencing or NGS examination of IDH1 and IDH2 mutations, when IDH-R132H immunostaining was negative or in the absence of IDH-R132H immunostaining (ii) Proportion of glioma patients with a diffuse astrocytoma, oligodendroglioma or glioblastoma <55 years of age whose IDH1-IHC was negative or absent and who had IDH1/ 2 -NGS b Pathology 18 Proportion of IDH -mutated not TP53 -mutated diffuse gliomas (regardless of their morphological picture of astrocytoma, oligodendroglioma or oligoastrocytoma) that are tested for 1p/19q a Pathology TP53 not sufficiently annotated in pathology reports Morphological picture of tumour not identifiable 19 Proportion of patients with biopsied or resected gliomas who undergo relevant molecular analysis of tumour tissue within 21 days of surgery a Pathology 21 days is not identifiable 21 days is arbitrary time frame P5 20 Proportion of patients with HGG >65 years of age in whom a MGMT promoter methylation analysis was carried out Proportion of glioblastoma patients >65 years of age in whom MGMT methylation analysis is carried out b Pathology Open in a new tab HGG, high-grade glioma; KPS, Karnofsky performance status; MRI, magnetic resonance imaging; NGS, next-generation sequencing; QI, quality indicator; TMZ, temozolomide; WHO, World Health Organization. a QIs that remain unmeasurable. b QI formulations that were calculated. c Results for patients who underwent surgical resection and for those who underwent only a diagnostic (stereotactic) biopsy are presented separately. Assignment of each patient to one centre For the benchmarking of the indicators between hospitals, it is essential to identify in which hospital(s) patients received their diagnostic and therapeutic care. 5 Since patients often receive care in more than one hospital, allocation algorithms were designed to identify the hospital with the highest impact on every indicator. Details regarding the allocation algorithms and applied priority rules are provided in Supplementary Material S2 , available at https://doi.org/10.1016/j.esmoop.2026.106303 . Statistical analysis Population-level results for the main cohort (incidence years 2016-2019 for process indicators related to treatment and 2017-2019 for process indicators related to pathological diagnosis) are calculated and compared with earlier cohorts when possible (incidence years 2012-2015 and 2008-2011). Descriptive presentations without statistical tests were employed. For the indicators related to treatment with a sufficient number of patients in the denominator, the observed results per hospital are visualized in funnel plots. The estimate of an indicator is plotted on the y -axis versus the number of observations per hospital on the x -axis. The agreed target is used as the population or reference value and added to the plot, as well as assumed prediction limits (95% and 99%) and observed overall (national) indicator result. Each dot represents a different hospital, but in the exceptional case that two or more centres have the same number of patients and the same proportion, dots are superimposed. 6 The indicators related to pathological diagnosis are considered as purely descriptive and no benchmarking is done. Process indicator/outcome associations For the process indicators related to glioblastoma—the only glioma subgroup large enough for outcome analysis—associations with outcomes were initially explored using non-parametric testing ( P values were obtained from a Pearson chi-square test for independence for mortality, and from Kaplan–Meier analysis for observed survival). This included not only the process indicators related to treatment described in this paper, but also those related to diagnosis and imaging that were published earlier. 3 For those indicators where an association was found, further parametric analysis with case-mix adjustment was carried out. Specifically, logistic regression models were used for post-biopsy or post-surgery mortality, while Cox proportional hazards model was applied for observed survival. Follow-up time was split into two consecutive intervals (0-0.5 and 0.5-2 years). The case-mix correction included sex, age at diagnosis, WHO performance score, comorbidities (diabetes, respiratory disease, cardiovascular disease) and presence of multiple tumours diagnosed between 5 years before and 2 years after diagnosis. Results Cohort description and patient to centre assignment A total of 3137 newly diagnosed gliomas were identified in the BCR database between 2016 and 2019, belonging to 3136 patients. In total, 70 gliomas (2.2%) were excluded, because of the unavailability of health insurance (IMA) data ( n = 67; 2.1%) or an incidence date corresponding with the day of death ( n = 3; 0.1%), resulting in 3067 remaining gliomas in this study set. Detailed patient characteristics can be found in Supplementary Material S3 , available at https://doi.org/10.1016/j.esmoop.2026.106303 . This indicator set on therapeutic care only considered diffuse astrocytic, oligodendroglial and ependymal tumours, and glioblastoma ( n = 2959). Tumour characteristics are shown in Table 2 . The assignment of patients to hospitals according to allocation algorithms can be found in Table 3 . Table 2. Tumour characteristics Glioma (2016-2019) N = 3067 Morphology/behaviour Glioma subtype Patients, n % of patients (ICD-O-3) Glioma ( n = 67) 9380/1 Borderline glioma 1 0 9380/3 Malignant glioma 65 2.1 9381/3 Gliomatosis cerebri 1 0 Ependymoma ( n = 39) 9391/3 Ependymoma 32 1 9392/3 Anaplastic ependymoma 6 0.2 9393/3 Papillary ependymoma 1 0 Astrocytoma ( n = 377) 9400/3 Diffuse astrocytoma 185 6 9411/3 Gemistocytic astrocytoma 16 0.5 9420/3 Fibrillary astrocytoma 17 0.6 9401/3 Anaplastic astrocytoma 159 5.2 Oligodendroglioma ( n = 191) 9450/3 Oligodendroglioma 112 3.7 9451/3 Anaplastic oligodendroglioma 79 2.6 Glioblastoma ( n = 2270) 9440/3 Glioblastoma 2 181 71.1 9441/3 Giant-cell glioblastoma 37 1.2 9442/3 Gliosarcoma 36 1.2 9445/3 IDH -mutant glioblastoma 16 0.5 Diffuse midline glioma ( n = 9) 9385/3 Diffuse midline glioma H3 K27 -mutant 9 0.3 Mixed glioma ( n = 6) 9382/3 (Anaplastic) oligoastrocytoma, NOS 6 0.2 Open in a new tab ICD-O-3, International Classification of Diseases for Oncology, third edition. Table 3. Allocation of patients and patient distribution for ‘centre of biopsy’, ‘centre of surgical resection’, ‘centre of radiotherapy’, ‘centre of chemo- or radiotherapy’, ‘centre of last care radiotherapy’ or ‘centre of last care chemotherapy’ for all patients diagnosed with glioma, incidence years 2016-2019 Number of distinct known centres Number of patients with centre Number of patients by centre n Known Unknown Mean Min Q1 Median Q3 Max n (%) n (%) Centre of biopsy 50 906 (100.0) 0 (0.0) 18.1 1 6 15 25 77 Centre of surgical resection 56 2210 (99.9) 2 (0.1) 38.8 1 12 31 53 160 Centre of radiotherapy 25 2097 (100.0) 0 (0.0) 83.8 6 53 70 98 247 Centre of chemo- or radiotherapy 54 2233 (100.0) 0 (0.0) 41.4 1 2 6 63 247 Centre of last care chemotherapy 93 2261 (97.9) 49 (2.1) 24.6 1 3 12 38 156 Centre of last care radiotherapy 90 2269 (97.8) 51 (2.2) 25.4 1 2 11 36 156 Open in a new tab Indicator measurability Nine of the 18 process indicators related to treatment were eventually measurable although some reformulations were necessary ( Table 1 ). The major shortcoming in this set of reformulated process indicators is the lack of information on the total dose of radiotherapy, the radiation dose per fraction and the number of fractions. Eight indicators were considered not measurable based on available data sources. For one indicator about the relationship between molecular status and applied treatment, although measurable, the expert board in charge of reviewing the TDS agreed not to benchmark the results because, at the population level, accurate data on the performance of the O6-methylguanine-DNA methyltransferase (MGMT) promotor methylation test and its results are not available. Moreover, the MGMT promotor methylation test was not (and still is not) reimbursed by the national health insurance. Where applicable, results for patients who underwent surgical resection and for those who underwent only a diagnostic (stereotactic) biopsy (in the absence of a subsequent surgical resection) are presented separately. Subgroup analyses, sensitivity analyses and historical comparison can be found in Supplementary Material S4 , available at https://doi.org/10.1016/j.esmoop.2026.106303 . For the process indicators on pathology reporting, five of the seven indicators were measurable, but only by using manually extracted data from pathology reports submitted to the BCR by the laboratories of pathological anatomy. Because of the potential data gap between what is available in individual hospitals and at the BCR used for indicator calculations, the expert board agreed not to benchmark the results for pathology indicators (see also ‘Discussion’ section). T1. Proportion of low-grade glioma patients receiving chemo- and/or radiotherapy without pathological confirmation In this cohort, no low-grade glioma (LGG) patients were treated with chemo- and/or radiotherapy without prior pathological confirmation. Since the national result for this process indicator was 0%, which was the predefined target, no funnel plot is presented. T2. Proportion of glioblastoma patients <70 years of age with a WHO performance score ≤2 in whom concomitant chemoradiotherapy is initiated In this cohort, 93.1% of surgically treated and 88.9% of biopsied glioblastoma patients started with concomitant chemoradiotherapy within, respectively, 12 months after surgical resection or 9 months after incidence, which is in line with the predefined target of 90%. Also, in the previous incidence periods, the target was met for surgically treated patients (90.5% in 2012-2015 and 90.8% in 2008-2011), but not always for biopsied patients (81.4% in 2012-2015 and 91.8% in 2008-2011). On hospital level, all centres perform within the funnel, with the majority performing above the target and some even at 100% ( Figure 1 A and B). The presence of a main or satellite radiotherapy facility does not seem to influence the scores. Figure 1. Open in a new tab Funnel plots for process indicators of therapeutic care. (A) Funnel plot of proportion of surgically treated glioblastoma patients <70 years of age with a World Health Organization (WHO) performance score ≤2 (or missing) who started concomitant chemoradiotherapy, by centre of surgical resection (49 hospitals reported in this funnel plot, 16 had <10 patients in the denominator and 2 patients could not be allocated to a centre of surgical resection and are thus not represented in this graph. Twenty-two hospitals have a recognized main RT centre, 7 hospitals have a recognized satellite RT centre and 20 hospitals have no RT centre. Patient distribution among the centres of surgical resection: min = 1; Q1 = 8; median = 15; Q3 = 26; max = 71). (B) Funnel plot of proportion of non-surgically treated glioblastoma patients <70 years of age with a WHO performance score ≤2 (or missing) who received a biopsy and who started concomitant chemoradiotherapy, by centre of biopsy. (36 hospitals reported in this funnel plot, 32 had <10 patients in the denominator. Sixteen hospitals have a recognized main RT centre, 5 hospitals have a recognized satellite RT centre and 15 hospitals have no RT centre. Patient distribution among the centres of biopsy: min = 1; Q1 = 1; median = 3; Q3 = 7; max = 18). (C) Funnel plot of proportion of surgically treated high-grade (grade 3/4) glioma patients who started chemotherapy or (chemo)radiotherapy within 6 weeks after surgical resection, by centre of surgical resection [51 hospitals reported in this funnel plot, 11 had <10 patients in the denominator and 2 patients could not be allocated to a centre of surgical resection and are thus not represented in this graph. Twenty-two hospitals have a recognized main radiotherapy (RT) centre, 8 hospitals have a recognized satellite RT centre and 21 hospitals have no RT centre. Patient distribution among the centres of surgical resection: min = 1; Q1 = 11; median = 26; Q3 = 39; max = 108]. (D) Funnel plot of proportion of deceased glioma patients who received chemotherapy maximum 14 days before death, by centre of last care (chemotherapy) [93 hospitals reported in this funnel plot, 44 had <10 patients in the denominator and 46 patients could not be allocated to a centre of last care (chemotherapy) and are thus not represented in this graph. Patient distribution among the centres of last care (chemotherapy): min = 1; Q1 = 3; median = 12; Q3 = 38; max = 156]. (E) Funnel plot of proportion of deceased glioma patients who received radiotherapy (cat 1-4) maximum 14 days before death, by centre of last care (radiotherapy) [90 hospitals reported in this funnel plot, 42 had <10 patients in the denominator and 50 patients could not be allocated to a centre of last care (radiotherapy) and are thus not represented in this graph. Patient distribution among the centres of last care (radiotherapy): min = 1; Q1 = 2; median = 11; Q3 = 36; max = 156]. PI, prediction interval. T3. Proportion of anaplastic oligodendroglioma patients who received both adjuvant chemo- and radiotherapy In grade 3 oligodendroglioma, 81.0% of surgically treated patients received both chemo- and radiotherapy within 12 months after surgical resection, while the target was set at 90%. A marked increase is noted in 2019 (93.3%) as compared with 2016 (72.0%) and with the historical comparison periods: 43.8% (2008-2011) and 57.0% (2012-2015). Because of the low number of patients involved ( n = 63), a funnel plot is not presented. Forty-four percent of grade 3 oligodendroglioma patients received temozolomide (TMZ) as chemotherapy, while 20.6% received a procarbazine, CCNU (lomustine) and vincristine (PCV) regimen. The remainder received a ‘light’ PCV regimen (i.e. only one or two of the three PCV components) or both TMZ and (light) PCV chemotherapy regimens. In the biopsy group, 80.0% received both chemo- and radiotherapy, but the number of patients here is very low ( n = 9). T4. Proportion of anaplastic astrocytoma patients who received both adjuvant chemo- and radiotherapy In grade 3 astrocytoma, 74.1% of surgically treated and 85.4% of biopsied patients received both chemo- and radiotherapy within, respectively, 12 months after surgical resection or 9 months after incidence, while the target was set at 90%. A marked increase, however, is noted from 2016 (42.1%) to 2019 (94.1%). Hospital benchmarking is not carried out because of the low number of both surgically treated ( n = 85) and biopsied ( n = 41) grade 3 astrocytoma patients. T5. Proportion of high-grade glioma patients who started chemotherapy or (chemo)radiotherapy within 6 weeks of surgical resection For this indicator, assessing the timing of adjuvant treatment in high-grade glioma (HGG), an overall score of 79.8% is obtained, which is below the target of 95%. The 5% tolerance was accepted for patients in poor clinical condition due to (post-operative) complications or for patients refraining from treatment. The score of the indicator is lower in the current cohort as compared with earlier incidence periods: 83.6% (2012-2015) and 85% (2008-2011). Regarding the most recent incidence period (2016-2019), the score is worse in 2018 and 2019 (76.6% and 77.1%) as compared with 2016 (85.3%) and 2017 (80.3%). If the interval is enlarged to 8 weeks, adjuvant treatment was initiated in 94.0% of patients. As illustrated in the funnel plot ( Figure 1 C), only a few centres (most of them with <10 patients/year) have a score for timely referral to a medical and/or radiation oncology department <60%. Five centres have a score above the target of 95%. The presence of a main or satellite radiotherapy facility does not seem to influence the score. T6. Proportion of deceased glioma patients who received chemo- or radiotherapy within 14 days before death The proportion of deceased glioma patients who received chemotherapy in the last 14 days before death is 3.5%, which is below the predefined target of 5%. The result was comparable in the earlier time frames: 3.4% in 2012-2015 and 3.0% in 2008-2011. Also, the proportion of deceased glioma patients who received radiotherapy (category 1, 2, 3 or 4) in the last 14 days before death (2.3%) is below the target of 5%. Also, in the previous time frames, the target was not surpassed (3.3% in 2008-2011; 4.0% in 2012-2015). The results at hospital level are presented in Figure 1 D and E. The majority of centres score within the random variability, with only one outlier where ∼20% of patients received radiotherapy in the last 14 days before death. When expanding the time frame up to 30 days before death, 10.7% received chemotherapy and 5.5% received radiotherapy. T7. Proportion of HGG (grade 3/4) patients <70 years of age with a WHO performance score ≤2 who started a long course of radiotherapy Both for surgically treated (99.6%) and biopsied patients (98.5%), the target of 95% is reached at national level, as illustrated in Figure 2 A and B. On hospital level, all centres score within the random variability around the target. In the historical comparison periods, results were similar. Figure 2. Open in a new tab Funnel plots for process indicators of therapeutic care. (A) Funnel plot of proportion of surgically treated high-grade (grade 3/4) astrocytoma patients <70 years of age with a WHO performance score ≤2 (or missing) who started a long series of adjuvant radiotherapy, by centre of radiotherapy (25 hospitals reported in this funnel plot, 1 had <10 patients in the denominator. Patient distribution among the centres of radiotherapy: min = 5; Q1 = 19; median = 37; Q3 = 46; max = 121). (B) Funnel plot of proportion of non-surgically treated high-grade (grade 3/4) astrocytoma patients <70 years of age with a WHO performance score ≤2 (or missing) who started a long series of radiotherapy, by centre of radiotherapy (23 hospitals reported in this funnel plot, 13 had <10 patients in the denominator. Patient distribution among the centres of radiotherapy: min = 1; Q1 = 3; median = 8; Q3 = 14; max = 26). HGG, high-grade glioma. T8. Proportion of anaplastic astrocytoma patients who started a long course of radiotherapy All surgically treated grade 3 astrocytoma patients treated with radiotherapy received a long series of radiotherapy (target 95%). For the group of patients who underwent only a biopsy, 95.1% received a long series of radiotherapy. Hospital benchmarking is not carried out because of the low number of both surgically treated ( n = 85) and biopsied ( n = 41) cases. T9. Proportion of surgically treated anaplastic ependymoma patients who started a long course of radiotherapy All surgically treated grade 3 ependymoma patients treated with radiotherapy received a long series of radiotherapy (target 95%). Hospital benchmarking is not carried out because of the low number of patients ( n = 4). Pathology indicators Pathology reports were available for 2117 (93.6%) of 2263 glioma patients diagnosed between 2017 and 2019 who underwent a biopsy and/or surgical resection. At the population level, the isocitrate dehydrogenase (IDH) status was documented in 88.6% of glioma, while next-generation sequencing for IDH1/2 in patients <45 years of age with negative or absent immunohistochemistry for IDH1 was carried out in 52.3%. For glioblastoma, the MGMT promotor methylation test result was available in 32.2%. Details and scatter plots are provided in Supplementary Material S4 , available at https://doi.org/10.1016/j.esmoop.2026.106303 . Process indicator/outcome associations Non-parametric testing did not reveal any significant associations between the indicators related to the initiation of the Stupp protocol (T2), timely initiation of adjuvant treatment (T5), or follow-up imaging (D8) and outcome. In contrast, the indicator on the performance of a diagnostic magnetic resonance imaging (MRI) (D3) was associated with 30-day post-biopsy mortality ( P = 0.0136), though no significant association was found with post-resection mortality ( P = 0.3458) or survival ( P = 0.0775). Additionally, the indicator on early post-operative MRI (D11) showed significant associations with both 30-day mortality ( P = 0.0003) and survival ( P < 0.0001). After adjustment for case mix, the association between diagnostic MRI and 30-day post-biopsy mortality was no longer statistically significant ( P = 0.1392). However, the association for early post-operative MRI remained significant for 30-day mortality ( P = 0.0055) and survival (0-0.5 years: P < 0.0001; 0.5-2 years: P = 0.0902) ( Table 4 ). Detailed calculations are provided in Supplementary Material S5 , available at https://doi.org/10.1016/j.esmoop.2026.106303 . Table 4. Unadjusted and adjusted hazard ratios for all-cause death associated with early post-operative MRI Label HR 95% CI P value D11, 0-0.5 years, unadjusted 0.5415 0.443-0.662 <0.0001 D11, 0.5-2 years, unadjusted 0.8392 0.737-0.956 0.0085 D11, 0-0.5 years, adjusted 0.6117 0.499-0.749 <0.0001 D11, 0.5-2 years, adjusted 0.8923 0.782-1.018 0.0902 Open in a new tab CI, confidence interval; HR, hazard ratio; MRI, magnetic resonance imaging. Discussion Although the predefined target was met for the majority of the nine process indicators calculated for treatment, indicating the provision of high-quality neuro-oncological care in Belgium, there is still some room for improvement. Histopathological diagnosis is key in order to be able to discuss prognosis and treatment options with patients and should be pursued in all suspected glioma patients, through resection or (stereotactic) biopsy in deep-seated lesions. However, in this cohort, no LGG patients were treated with chemo- and/or radiotherapy without prior pathological confirmation. For the indicator evaluating the use of the Stupp protocol in glioblastoma (in this study 93.1% of surgically treated and 88.9% of biopsied patients), international comparison is difficult, since most studies only report the total number of glioblastoma patients receiving chemoradiotherapy, without subgroup analysis for patients <70 years of age with good (≤2) WHO performance score, as used in our indicator’s definition. In a ‘real-world’ single-institution retrospective analysis in the Czech Republic, 66.9% of patients (all ages) with Karnofsky index ≥60 received the Stupp protocol. 7 However, according to the definition of long course of radiotherapy (see further), patients who received a Perry regimen (40 Gy in 15 fractions) were not excluded from the numerator in this indicator. 8 For the indicator assessing adjuvant treatment with both chemo- and radiotherapy in anaplastic oligodendroglioma, the remarkable increase noted in 2019 (93.3%) as compared with 2016 (72.0%) (and earlier periods) can possibly be explained by evidence provided by the RTOG 9402 and EORTC 26951 studies that adding PCV chemotherapy, either prior or after radiotherapy, in first line approximately doubles the overall survival. 9 , 10 International comparison is difficult since most studies are not assessing the same era as this cohort. In an analysis of the United States National Cancer Database, 62.2% of all 1p19q co-deleted tumours between 2010 and 2014 received surgery and chemoradiotherapy. 11 The same is observed for grade 3 astrocytoma, where again the result increased from 42.1% in 2016 to 94.1% in 2019 (target 90%). The published interim results of the CATNON trial, where a clear benefit of 12 cycles of adjuvant TMZ was observed, might be the explanation. 12 Hospital benchmarking is not carried out because of the low number of both surgically treated ( n = 85) and biopsied ( n = 41) anaplastic astrocytoma patients. International data for benchmarking were not available. To provide the best results for patients with HGG (grade 3/4), adjuvant treatment, whether chemoradiotherapy, radiotherapy or chemotherapy alone, should be initiated within 6 weeks after surgical resection. 1 , 13 , 14 , 15 , 16 , 17 For the indicator assessing the timeliness of adjuvant treatment in HGG, there is an opportunity for improvement. Since ∼15% of patients start treatment between 6 and 8 weeks after surgical resection, with probably little effort and alertness, this could be improved so that most centres would reach the predefined target of 95%. However, as compared with the Czech Republic, France and Scotland, treatments in Belgium are initiated earlier. In the Czech Republic, only 37% of patients with HGG start with the Stupp regimen within 6 weeks and only 34% with radiotherapy alone. 7 In France, in 2006, only 56% of patients received radiotherapy within 6 weeks. 18 In Scotland, where the target for this indicator was increased from 90% to 95% in 2021, the proportion starting adjuvant oncological treatment was 77.4% in 2018 and 74.7% in 2019. 18 , 19 In 2020, only 71.2% could start within 6 weeks, but this result could have been impacted by the coronavirus disease 2019 (COVID-19) pandemic. The median waiting time after surgery is 30-45 days in Sweden and 27 days in Germany, but the proportion of patients starting within 6 weeks is not available. 20 , 21 The proportion of glioma patients who die within 30 days of treatment (surgery, radiotherapy and chemotherapy) was originally proposed as an outcome indicator. The multidisciplinary expert board reviewing the TDS confirmed this for mortality after surgical resection or biopsy, although, besides mortality as a complication of the surgery, in HGG, death within the first 30 days after surgery might also be attributed to early disease progression. This will be further elaborated in a subsequent analysis on mortality and survival. Contrarily, glioma patients generally do not die as an undesired effect of radio- or chemotherapy, but rather from disease progression, while still under this (those) treatment(s). Therefore, this part of the outcome indicator was reformulated towards a process indicator aiming to identify potential futile therapy by assessing the proportion of patients who received radio- or chemotherapy in the 14 days before death. In a single-centre retrospective study, in the United States (2010-2015), 6% of glioblastoma patients received chemotherapy in the last 14 days of life. 22 In Finland, 21% of patients with malignant brain tumours treated in a single institution received systemic treatments and 10% received chemoradiotherapy in the last 30 days of life. In the last 14 days before death, 17% received their last anticancer treatment. 23 This seems higher than in our cohort where <5% received chemo- or radiotherapy in the last 14 days of life. In Scotland, the Scottish Adult Neuro-Oncology Network audit evaluated post-treatment mortality at 30 days (as an outcome indicator) and reported that in 2021 4.5% of patients with central nervous system tumours died within 30 days after chemoradiotherapy, while 9.8% died within 30 days after radiotherapy (target <5%). 24 For all radiotherapy indicators, it has to be noted that the administered total dose and the dose per fraction are not identifiable in administrative databases in Belgium. Only the differentiation between a long and a short course of radiotherapy was used as a surrogate in reformulated indicators on radiotherapy, which unfortunately resulted in a less detailed evaluation (see Appendix 1 ). However, since all radiotherapy departments have to register all fractions and dose per fraction for each treatment to comply with radiation and patient protection requirements, it could be an opportunity to make these data available to BCR. Another option is to reform or adapt nomenclature rules for radiotherapy enabling accurate assessment of dose and number of fractions. By using these ‘reimbursement categories’ for radiotherapy-related indicators, both for surgically treated (99.6%) and for biopsied (98.5%) HGG patients <70 years of age and with a WHO score ≤2 undergoing radiotherapy, the predefined target of 95% is reached. Internationally, according to the Danish Neuro-Oncology Registry (2010-2014), 92%-96% of patients finish high-dose radiotherapy as planned out of all patients starting this treatment. 25 Data from single-institution retrospective analyses can give an indication but of course cannot be extrapolated to the population of that country. In a single institution in India (2005-2009), 92% of glioblastoma patients received a total radiotherapy dose of 60 Gy, 26 while in the Ontario region, Canada, only 47.3% completed 60 Gy radiotherapy. 27 For the process indicators on pathological diagnosis, it has to be underlined that information on molecular markers is not available as structured data within the BCR. An incomplete overlap between the ICD-O-3 classification used in cancer registration and the clinically used WHO classification has been previously highlighted by our group. 4 As a result, molecular marker data had to be manually extracted. This approach introduces several issues leading to a potential data gap between what is available in individual hospitals and the data used for process indicator calculation at the population level. Firstly, glioma patients with at least one glioma-related pathology report available at the BCR were included for the calculation of indicators related to molecular marker status. This may have excluded crucial information from subsequent reports not submitted to the BCR. Secondly, manual data extraction carries the risk of incomplete or inaccurate data. As a result, indicators on pathology were not benchmarked. However, IDH status availability in this series (88.6%) was superior to data from the CBTRUS database (75.1% for the incidence year 2018), while the Dutch Brain Tumour Registry reported performance of molecular testing for IDH in 81.1% of gliomas. 28 , 29 In this study, associations between process indicators related to treatment and previously reported indicators on diagnosis and imaging and outcome (30-day mortality and survival) were explored. This was only conducted for the glioblastoma subgroup, as it was the only subgroup large enough to evaluate outcome associations with case-mix adjustments. No significant associations were observed between any treatment-related process indicators and outcome, and only one imaging-related indicator (early post-operative MRI) showed a statistically significant association with both 30-day mortality and survival. Despite these findings, the evaluation of process indicators remains relevant in the context of glioblastoma care. Given the inherently incurable nature of glioblastomas and their generally poor prognosis, detecting associations with survival endpoints might be more challenging than in other malignancies. The indicators used in this study were developed and refined through a Delphi survey and reflect a peer-reviewed expert consensus on good clinical practice. 1 As such, they still serve as standardized tools to assess and improve the quality of care. The absence of associations in this cohort does not exclude their potential significance in other populations or settings. The observed association with early post-operative MRI may in part reflect selection bias: patients in poor clinical condition are less likely to undergo early imaging, either due to clinical decisions favouring comfort care or because they are physically less stable or difficult to transport for MRI acquisition. Moreover, while no associations were identified between treatment-related indicators and hard outcomes, process indicators may still have an association with soft outcome indicators which are more related to the quality of life of patients. However, patient-reported outcomes and patient experience are unexplored in this context and should be a focus of future studies. A major strength of this study is that this cohort, selected from the BCR database, is population based and therefore includes all registered adult glioma cases fulfilling the inclusion criteria, reducing the risk of selection bias to a minimum. In this way the care patterns across all Belgian hospitals are evaluated. As previously described, this kind of study also incorporates limitations. 30 A first limitation is the retrospective nature of indicator calculation, which in 2025 revealed population-level and hospital-specific results of glioma patients treated from 2016 to 2019. Of course, subsequent future calculations to identify possible improvement will be conducted more efficiently because the methodology is developed and described in detail in the TDS. Still, the inherent delay in population databases reaching completeness will remain. Secondly, and most importantly, the use of administrative data for the calculation of process indicators implies that adaptations to the original formulation of the indicators are inevitable. The most important limitation for the calculation of process indicators regarding radiotherapy is the fact that, as already described, the total dose, number of fractions and dose per fraction are not available on a population level, and that a surrogate administrative billing code had to be used. Moreover, also the use of MRI fusion-based radiotherapy planning and the methods by which target and organ-at-risk doses were evaluated are not identifiable in the current population-based databases. In recent years, radiotherapy-specific quality indicators have been developed in Belgium on behalf of the Belgian College for Physicians in Radiation Oncology. Data from a randomly selected group of patients from all radiation oncology departments were collected and analysed. 31 Up to now, besides structure indicators, unfortunately, only process or outcome indicators related to radiotherapy for breast, prostate and head and neck cancers have been established. Probably, in the future, this could be extended to the field of neuro-oncology and collaboration with the BCR could be considered. With regard to process indicators related to surgery, two essential elements for the concept of maximal safe resection are actually very difficult to identify on a population level: the presence of post-operative residual tumour and the occurrence of new neurological deficits after surgery. In this way, qualitative evaluation of the surgical process remains impossible. Outcome indicators such as post-operative mortality and survival can only partially overcome this issue. Registration of, for instance, the Neurological Assessment in Neuro-Oncology (NANO) score or National Institutes of Health Stroke Scale (NIHSS) before and at certain time points after surgical resection might contribute to differentiating iatrogenic neurological worsening from disease progression. 32 , 33 Recently, the three-tiered Onco-Functional Outcome (OFO) score, integrating the extent of resection and neurological status, proved to be correlated with overall survival and might be useful in future assessments to cover the concept of maximal safe resection on a population level. 34 Central registration of this OFO score 6 weeks post-operatively, when adjuvant treatment is (about to be) initiated, should be feasible. Other elements that are currently not available in population-based databases in Belgium are the presence of craniospinal metastatic disease in grade 3 ependymoma, participation in clinical trials and the presence of patient education and informed consent before the prescription of chemotherapy. A compulsory prospective registry could collect more (clinical) data points that are currently unavailable in population-based administrative databases, but it would probably be time-consuming for physicians already burdened with administrative demands. Alternatively, voluntary participation in prospective registries yields incomplete and biased results, with variable inclusion of patients among centres and therefore should be avoided. 35 Probably, in the future, artificial intelligence will enable linkage between population-based cancer registries and patient data directly from hospital-based medical records. For the moment, this idea remains attractive but is not yet feasible. 36 The large amount of unstructured (plain text) data in electronic patient files, the absence of uniformity in the (commercially) available electronic health record systems and the regulatory framework to comply with in terms of General Data Protection Regulation are the three main hurdles on the way. Thirdly, although exploring the potential influence of socioeconomic status on calculated process indicators would be worth exploring, such information is currently not accessible from cancer registry or reimbursement data. Next, comparing indicator results on an international level remains difficult because of the limited number of studies that can be retrieved from the literature and because of the lack of standardization of applied indicators across nations. Finally, as scientific evidence evolves, so too must the rationale for certain indicators. Consequently, future (re-)assessments will require adaptations or revisions to certain indicators to align with the prevailing WHO classification and guidelines at that time. We hope this study will help increase awareness of quality assessments in glioma care internationally and may contribute to harmonization of different process indicators, ultimately resulting in an improvement in the quality of care in neuro-oncology. Conclusion In general, the results on a population level for these process indicators related to therapeutic care are good with regard to the predefined targets, indicating the delivery of high-quality care for neuro-oncological patients in Belgium. Although there is a lack of association between process indicators and certain outcomes, evaluating process indicators remains relevant for assessing the quality of the care trajectory of glioma patients. For grade 3 astrocytoma and oligodendroglioma, administration of both adjuvant chemo- and radiotherapy increased throughout the timespan of the cohort and reached the predefined target (90%) for patients diagnosed in 2019. However, for the timing of adjuvant treatment in HGG and for many (radiotherapy- and/or surgical outcome-related) registration requirements, there is room for improvement. Individual feedback reports on the assessment of these process indicators were provided to Belgian hospitals, enabling reflection among the physicians involved in neuro-oncology care programmes as well as hospital administration. As mentioned earlier, a first calculation of indicators should not be used to criticize individual hospitals; rather, it should help identify gaps in clinical practice, create opportunities for quality improvement and motivate physicians and hospitals already performing well. On an international level, increasing awareness about quality assessment in neuro-oncology, and harmonization of applied process indicators, could enhance benchmarking possibilities in the future. Acknowledgements We thank all employees of the Belgian Cancer Registry (BCR) who participated in data collection and processing. We also thank the following experts who were involved in the validation of the methodology of the technical documentation sheets before definite calculation: Steven De Vleeschouwer, Frank Weyns, Arnaud Lombart, Ludo Vanopdenbosch, Alex Michotte, Tom Boterberg, Nick Liefhooghe, Martin Lammens, Raf Sciot, Paul Clement, Nicolas Whenham, Ann Tieleman, Siska Dedeurwaerdere, Karolien Goffin. We are very grateful to the following colleagues for their help and support: Jeroen Van Lerbeirghe, Stephanie Du Four, Wim Maenhoudt and Olivier Van Damme. Funding This work was supported by Stichting tegen Kanker, AZ Delta vzw and the Ghent University Arne Lannoy, A.K.A. Zorro Fund (no grant number). Disclosure SDV is a certified Gliolan (Medac GmbH) Trainer and has carried out consultancy for Lamepro NV (currently Pharmanovia) (Gliolan) for which a fee was received. All other authors have declared no conflicts of interest. Data sharing The cancer cohort data used and analysed during the study are available via the Belgian Cancer Registry (BCR) upon reasonable request. The pseudonymized data can be provided within the secured environment of the BCR after having been guaranteed that the applicable General Data Protection Regulation regulations are applied. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work, the author(s) used OpenAI/ChatGPT in order to improve language and readability. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication. Appendix 1 According to reimbursement rules in Belgium, the number of fractions should normally be available at the end of the radiotherapy treatment, since all fractions of radiotherapy ought to be registered separately. In practice, however, often only the last fraction is registered because reimbursement of the complete cycle is related to that last fraction. This means that it is not possible to reliably identify the number of fractions and duration of the treatment for all patients, except by exploring all individual medical patient files, which was beyond the scope of this project. Instead of evaluating total dose of radiotherapy, number of fractions and dose per fraction, the billing code of the ‘category of radiotherapy’ was used as a weak surrogate, enabling only to differentiate between: a palliative or short series of <10 fractions (category 1), a simple series of 11-35 fractions with curative intent (category 2), a complex or three-dimensional series regardless of the number of fractions (category 3) or intensity-modulated radiotherapy (IMRT) with at least 15 fractions (category 4). In this study category, 2, 3 and 4 are considered a long course of radiotherapy. Supplementary data Supplementary Materials S1 mmc1.pdf (561.8KB, pdf) Supplementary Materials S2 mmc2.pdf (107.2KB, pdf) Supplementary Materials S3 mmc3.pdf (67.9KB, pdf) Supplementary Materials S4 mmc4.pdf (1.6MB, pdf) Supplementary Materials S5 mmc5.pdf (321.2KB, pdf) Supplementary Appendix mmc6.docx (13.7KB, docx) References 1. Vanhauwaert D., Pinson H., Sweldens C., et al. Quality indicators in neuro-oncology: review of the literature and development of a new quality indicator set for glioma care through a two-round Delphi survey. J Neurooncol. 2022;157(2):365–376. doi: 10.1007/s11060-022-03971-3. [ DOI ] [ PubMed ] [ Google Scholar ] 2. Vanhauwaert D., Silversmit G., Vanschoenbeek K., et al. Association of hospital volume with survival but not with postoperative mortality in glioblastoma patients in Belgium. J Neurooncol. 2024;170(1):79–87. doi: 10.1007/s11060-024-04776-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Vanhauwaert D., Vanschoenbeek K., Weyns F., et al. Measuring the diagnostic management and follow-up imaging for patients diagnosed with glioma in Belgium, across hospitals. Cancer Med. 2024;13(21) doi: 10.1002/cam4.70045. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Vanhauwaert D., Pinson H., Vanschoenbeek K., et al. Cancer registration, molecular marker status and adherence to the WHO 2016 classification of pathology reports for glioma diagnosed during 2017-2019 in Belgium. Pathobiology. 2023;90(6):365–376. doi: 10.1159/000529320. [ DOI ] [ PubMed ] [ Google Scholar ] 5. Leroy R., De Gendt C., Stordeur S., et al. Head and neck cancer in Belgium: quality of diagnostic management and variability across Belgian hospitals between 2009 and 2014. Front Oncol. 2019;9:1006. doi: 10.3389/fonc.2019.01006. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Savoye I., De Gendt C., Bourgeois J., et al. Quality Indicators for the management of epithelial ovary cancer – KCE report 357. Brussels: Belgian Health Care Knowledge Center (KCE) 2022. https://kce.fgov.be/sites/default/files/2022-06/KCE_357_Quality_Indicators_Ovarian_Cancer_Report.pdf Available at. 7. Lakomy R., Kazda T., Selingerova I., et al. Real-world evidence in glioblastoma: Stupp’s regimen after a decade. Front Oncol. 2020;10:840. doi: 10.3389/fonc.2020.00840. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Perry J.R., Laperriere N., O’Callaghan C.J., et al. Short-course radiation plus temozolomide in elderly patients with glioblastoma. N Engl J Med. 2017;376(11):1027–1037. doi: 10.1056/NEJMoa1611977. [ DOI ] [ PubMed ] [ Google Scholar ] 9. Cairncross G., Wang M., Shaw E., et al. Phase III trial of chemoradiotherapy for anaplastic oligodendroglioma: long-term results of RTOG 9402. J Clin Oncol. 2013;31(3):337–343. doi: 10.1200/JCO.2012.43.2674. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. van den Bent M.J., Brandes A.A., Taphoorn M.J.B., et al. Adjuvant procarbazine, lomustine, and vincristine chemotherapy in newly diagnosed anaplastic oligodendroglioma: long-term follow-up of EORTC brain tumor group study 26951. J Clin Oncol. 2013;31(3):344–350. doi: 10.1200/JCO.2012.43.2229. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Yeboa D.N., Yu J.B., Liao E., et al. Differences in patterns of care and outcomes between grade II and grade III molecularly defined 1p19q co-deleted gliomas. Clin Transl Radiat Oncol. 2019;15:46–52. doi: 10.1016/j.ctro.2018.12.003. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. van den Bent M.J., Baumert B., Erridge S.C., et al. Interim results from the CATNON trial (EORTC study 26053-22054) of treatment with concurrent and adjuvant temozolomide for 1p/19q non-co-deleted anaplastic glioma: a phase 3, randomised, open-label intergroup study. Lancet. 2017;390(10103):1645–1653. doi: 10.1016/S0140-6736(17)31442-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Landelijke werkgroep Neuro-oncologie Kwaliteitscriteria neuro-oncologie. Diagnostiek, behandeling en begeleiding van patiënten met een glioom. 2014. https://richtlijnendatabase.nl/uploaded/docs/IKNL_in_ontw/Kwaliteitscriteria_Gliomen_def_mei2014.pdf Available at. 14. Sulman E.P., Ismaila N., Armstrong T.S., et al. Radiation therapy for glioblastoma: American Society of Clinical Oncology Clinical Practice Guideline Endorsement of the American Society for Radiation Oncology Guideline. J Clin Oncol. 2017;35(3):361–369. doi: 10.1200/JCO.2016.70.7562. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Sulman E.P., Ismaila N., Chang S.M. Radiation therapy for glioblastoma: American Society of Clinical Oncology Clinical Practice Guideline Endorsement of the American Society for Radiation Oncology Guideline. J Oncol Pract. 2017;13(2):123–127. doi: 10.1200/JOP.2016.018937. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Warren K.T., Liu L., Liu Y., Milano M.T., Walter K.A. The impact of timing of concurrent chemoradiation in patients with high-grade glioma in the era of the Stupp protocol. Front Oncol. 2019;9:186. doi: 10.3389/fonc.2019.00186. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Buszek S.M., Al Feghali K.A., Elhalawani H., Chevli N., Allen P.K., Chung C. Optimal timing of radiotherapy following gross total or subtotal resection of glioblastoma: a real-world assessment using the National Cancer Database. Sci Rep. 2020;10(1):4926. doi: 10.1038/s41598-020-61701-z. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Noel G., Huchet A., Feuvret L., et al. Waiting times before initiation of radiotherapy might not affect outcomes for patients with glioblastoma: a French retrospective analysis of patients treated in the era of concomitant temozolomide and radiotherapy. J Neurooncol. 2012;109(1):167–175. doi: 10.1007/s11060-012-0883-7. [ DOI ] [ PubMed ] [ Google Scholar ] 19. Liaquat I., Campbell L., McMahon J. Scottish Adult Neuro Oncology Network; 2022. Audit Report: Brain and Central Nervous System Cancers Quality Performance Indicators. Report of the 2020 Clinical Audit Data; p. 41. [ Google Scholar ] 20. Seidlitz A., Siepmann T., Löck S., Juratli T., Baumann M., Krause M. Impact of waiting time after surgery and overall time of postoperative radiochemotherapy on treatment outcome in glioblastoma multiforme. Radiat Oncol. 2015;10:172. doi: 10.1186/s13014-015-0478-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Asklund T., Malmström A., Bergqvist M., Björ O., Henriksson R. Brain tumors in Sweden: data from a population-based registry 1999-2012. Acta Oncol. 2015;54(3):377–384. doi: 10.3109/0284186X.2014.975369. [ DOI ] [ PubMed ] [ Google Scholar ] 22. Hemminger L.E., Pittman C.A., Korones D.N., et al. Palliative and end-of-life care in glioblastoma: defining and measuring opportunities to improve care. Neurooncol Pract. 2017;4(3):182–188. doi: 10.1093/nop/npw022. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Nåhls N.S., Leskelä R.-L., Saarto T., Hirvonen O., Anttonen A. Effect of palliative care decisions making on hospital service use at end-of-life in patients with malignant brain tumors: a retrospective study. BMC Palliat Care. 2023;22(1):39. doi: 10.1186/s12904-023-01154-z. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Liaquat I., Campbell L., McMahon J. Scottish Adult Neuro Oncology Network; 2021. Audit Report 2021 on Brain and Central Nervous System Cancers: Quality Performance Indicators. https://www.nn.nhs.scot/sanon/wp-content/uploads/sites/32/2022/10/Final-Published-2021_BrainCNS_QPI_Audit_Report_2021_v1.pdf Available at: [ Google Scholar ] 25. Hansen S. The Danish Neuro-Oncology Registry. Clin Epidemiol. 2016;8:629–632. doi: 10.2147/CLEP.S99459. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Julka P.K., Sharma D.N., Mallick S., Gandhi A.K., Joshi N., Rath G.K. Postoperative treatment of glioblastoma multiforme with radiation therapy plus concomitant and adjuvant temozolomide: a mono-institutional experience of 215 patients. J Cancer Res Ther. 2013;9(3):381–386. doi: 10.4103/0973-1482.119310. [ DOI ] [ PubMed ] [ Google Scholar ] 27. Lwin Z., MacFadden D., Al-Zahrani A., et al. Glioblastoma management in the temozolomide era: have we improved outcome? J Neurooncol. 2013;115(2):303–310. doi: 10.1007/s11060-013-1230-3. [ DOI ] [ PubMed ] [ Google Scholar ] 28. Dutch Brain Tumour Registry . Integraal Kankercentrum Nederland and Landelijke Werkgroep Neuro-Oncologie; 2019. Glioomzorg in Nederland. https://lwno.nl/media/1050/rapportage-glioomzorg-dbtr-2014-2017.pdf Available at: [ Google Scholar ] 29. Ostrom Q.T., Cioffi G., Waite K., Kruchko C., Barnholtz-Sloan J.S. CBTRUS Statistical Report: Primary brain and other central nervous system tumors diagnosed in the United States in 2014-2018. Neuro Oncol. 2021;23(suppl 3):iii1–iii105. doi: 10.1093/neuonc/noab200. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Vanhauwaert D, Vanschoenbeek K, Weyns F, et al. Measuring the diagnostic management and follow-up imaging for glioma patients across Belgian hospitals between 2016 and 2019. Cancer Med. 2024;13 doi: 10.1002/cam4.70045. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Vaandering A., Jansen N., Weltens C., et al. Radiotherapy-specific quality indicators at national level: how to make it happen. Radiother Oncol. 2023;178 doi: 10.1016/j.radonc.2022.11.022. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Nayak L., DeAngelis L.M., Brandes A.A., et al. The Neurologic Assessment in Neuro-Oncology (NANO) scale: a tool to assess neurologic function for integration into the Response Assessment in Neuro-Oncology (RANO) criteria. Neuro Oncol. 2017;19(5):625–635. doi: 10.1093/neuonc/nox029. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Zeitlberger A.M., Flynn M.-C., Hollenstein M., Hundsberger T. Assessment of neurological function using the National Institute of Health Stroke Scale in patients with gliomas. Neurooncol Pract. 2021;8(6):699–705. doi: 10.1093/nop/npab046. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Gerritsen J.K.W., Zwarthoed R.H., Kilgallon J.L., et al. Impact of maximal extent of resection on postoperative deficits, patient functioning, and survival within clinically important glioblastoma subgroups. Neuro Oncol. 2023;25(5):958–972. doi: 10.1093/neuonc/noac255. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Jegou D., Penninckx F., Vandendael T., Bertrand C., Van Eycken E. PROCARE. Completeness and registration bias in PROCARE, a Belgian multidisciplinary project on cancer of the rectum with participation on a voluntary basis. Eur J Cancer. 2015;51(9):1099–1108. doi: 10.1016/j.ejca.2014.02.025. [ DOI ] [ PubMed ] [ Google Scholar ] 36. Wang J.W., Williams M. Registries, databases and repositories for developing artificial intelligence in cancer care. Clin Oncol (R Coll Radiol) 2022;34(2):e97–e103. doi: 10.1016/j.clon.2021.11.040. [ DOI ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Materials S1 mmc1.pdf (561.8KB, pdf) Supplementary Materials S2 mmc2.pdf (107.2KB, pdf) Supplementary Materials S3 mmc3.pdf (67.9KB, pdf) Supplementary Materials S4 mmc4.pdf (1.6MB, pdf) Supplementary Materials S5 mmc5.pdf (321.2KB, pdf) Supplementary Appendix mmc6.docx (13.7KB, docx) ACTIONS View on publisher site PDF (1016.5 KB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top