Incidence and survival of pediatric low-grade gliomas in Switzerland Authors: Ladina Greuter, Eleftheria Michalopoulou, Shelagh Redmond, Nicolas V. der Weid, Maria Licci, Johanna Wyss, Raphael Guzman, Jehuda Soleman Published: 2026 ABSTRACT BackgroundBrain tumors are the most common solid cancers occurring in children, with pediatric low-grade gliomas (pLGG) being the most frequent central nervous system (CNS) tumors in children. The overall survival of pLGG is excellent, with a ten-year survival of 85%–96%. Prognostic factors include age, gender, anatomical location, and histological and molecular subtype. Despite the excellent survival rates, children suffering from pLGG often have a diminished quality of life due to functional or neurologic deficits either caused by the primary tumor or as a side effect of their treatment. In this study, we describe the epidemiology of pLGG in Switzerland.MethodsThe target population comprised children aged 0–17 years, diagnosed with histologically verified pLGG during 1990–2019 while residing in Switzerland. We retrieved the data from the Swiss Childhood Cancer Registry (ChCR). We calculated crude incidence rates per 100,000 person-years and observed survival (OS) using the Kaplan-Meier estimator (non-parametric univariable analysis). We estimated OS separately according to age histological subgroup and treatment group.ResultsWe included 591 pLGG during 1990–2019 in Switzerland. Most pLGG were diagnosed at the age of 2–11 years (n = 352, 59.6%). Around half of pLGG were supratentorial located (n = 286, 48.4%), while the other half (n = 266, 45.0%) occurred infratentorial. Most pLGG (82.0%) were World Health Organization (WHO) grade 1. The vast majority of pLGG (n = 501, 84.8%) were treated with surgery, either alone (n = 441, 74.6%) or in combination with other treatments. The lowest 5- and 10-year survival was observed in children who received surgery followed by radiotherapy and/or chemotherapy (40% and 20%). Astrocytoma excluding pilocytic astrocytoma had a significantly lower 5-year, 10-year and total survival (98.8%, 85.8%, and 82.1%) compared to the other histopathological subtypes (P = 0.024).ConclusionsThis is the first report on the incidence and survival of children with pLGG in Switzerland. Children with pLGG have an excellent long-term survival rate of over 90%. Introduction Brain tumors are the most common solid cancers occurring in children, with low-grade pediatric gliomas (pLGG) being the most frequent central nervous system (CNS) tumors [1–4]. The pLGG comprises a heterogeneous group of tumors classified as World Health Organization (WHO) grade 1 or 2. It is reported that approximately (65%) of pLGGs are pilocytic astrocytoma (WHO 1) [4, 5]. While surgery is the main therapy for pLGG, further treatment is required in case of an incomplete resection due to their eloquent anatomical location or in case of a recurrence [6]. Adjuvant treatment conventionally included chemotherapy, sometimes radiotherapy. While chemotherapy remains the standard frontline therapy for most pLGG that recur after surgery, targeted inhibitors such as mitogen-activated protein kinase (MEK) and/or BRAF inhibitors have shown promising outcomes for relapsed tumors [4, 7–9]. Radiotherapy has generally been reserved for recurrent cases in the treatment of pLGG due to long-term toxicity, while it used to be a mainstay of therapy [10, 11]. The overall survival of pLGG is excellent, with a 10-year survival rate of 85%–96% [4, 5, 12, 13]. Prognostic factors influencing survival include age, gender, anatomical location, and histopathological and molecular subtype [14]. Despite the excellent survival rates, children suffering from pLGG often have a diminished quality of life due to functional or neurologic deficits either caused by the primary tumor or as a side effect of their treatment [4, 15], resulting in a high burden of long-term morbidity [4, 16]. Understanding the epidemiology of pLGG is important to estimate the long-term outcome for these patients and the public health impact [17]. In Germany, an increase in the incidence of pLGG has been observed over the last two decades from 0.94 to 2.12 per 100,000 children per year [18]. A similar development was observed in North America with geographic differences between the Northeast and Midwest [14]. No current study is available on the epidemiology of pLGG in Switzerland. National databases can play an essential role in conducting such epidemiological studies by collecting cancer data. In the US, the Surveillance, Epidemiology, and End Results Program (SEER) covers around half of the population’s information about cancer incidence and outcomes [17]. In Switzerland, the Childhood Cancer Registry (ChCR) is a comparable national database for cancer patients diagnosed at the age of 0–19 years. It was founded in the 1970s and all pediatric oncology centers report directly to this database [16, 19], making it a real population-based registry. In this study, we describe the epidemiology of pLGG in Switzerland, using national cancer registry data. Specifically, we report on the incidence and observed survival (OS) of pLGG and look for potential determinants of event-free survival (EFS). Methods Study population The target population comprised children (< 18 years) diagnosed with pLGG between 1990 and 2019 in Switzerland. Based on our predefined study protocol, we selected 2019 as the follow-up cutoff to ensure data completeness and accuracy, given potential delays in fully entering all information into the database. This study only included data from the ChCR with histopathologically verified pLGG. The ChCR is estimated to include over 90% of all cancer cases diagnosed in Switzerland for children younger than 16 years, whereas registration of cancer in adolescents older than 16 years at diagnosis has been less complete, due to the possibility of these children being treated in adult cancer centers [19, 20]. No data were acquired if the parents vetoed their child’s registration in the ChCR, despite anonymous collection. However, the veto rate was low at approximately 0.5%. The ChCR validated data by receiving original pathology, surgery, and treatment or follow-up reports, as well as by submitting information on standardized forms that included individual treatment modalities. The retrieved data on histopathologically verified pLGG included information on the date of diagnosis (set as timepoint when histopathology was achieved), age at diagnosis (grouped as 0–1, 2–11, 12–15, and 16–17 years), sex (male, female), and geographic region of diagnosis according to language (German-, French-, or Italian-speaking areas). We additionally received data on the WHO grade (1, 2, or unknown), anatomical localization (supratentorial, infratentorial, spinal cord, multifocal, or unknown), and histopathological subgroup, which was coded according to the third version of the International Classification of Childhood Cancers (ICCC3) [21]. The histopathological subgroups comprised astrocytoma (not pilocytic), pilocytic astrocytoma, other specified and unspecified intracranial and intraspinal neoplasms (neuronal and mixed neuronal-glial tumors), and other gliomas (oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain origin). We also collected data on treatments during the first year after pLGG diagnosis and on treating hospitals, including Swiss Pediatric Oncology Group (SPOG) clinics and non-SPOG clinics. Treatments during the first year after pLGG diagnosis were grouped into surgery, chemotherapy, radiotherapy, and all the potential combinations thereof. In the case of surgery, we collected information on the extent of resection, including gross total resection (GTR), subtotal resection (STR), partial resection, and biopsy. GTR was defined as a resection with greater than 90% of the tumor mass. STR was defined as a resection of 50%–90%. Partial resection was defined as the removement of less than 50%. If available, data on underlying syndromes, such as neurofibromatosis type 1 (NF-1), tuberous sclerosis, and epilepsy, were acquired. Information on other underlying diseases was incomplete and was registered only for 13% of all the included pLGGs, which is why we did not analyze them. We acquired information on the vital status of patients at the end of the study follow-up (31st of December 2019) or the maximum available follow-up to estimate OS. We acquired the dates of progression, recurrence, and metastasis after a pLGG diagnosis to estimate EFS if those occurred. EFS was defined as the time to progression, recurrence, metastasis, or death, whichever came first. Progression was reported to the ChCR by the clinicians on a standard relapse or progression form. Statistical analysis To account for incomplete registration coverage of pLGG cases diagnosed at ages 16–17, we estimated the total number of pLGG that would have been observed for this age group under nationwide coverage by reweighting the observed number of cases [22]. The weights were calculated as the ratio of the Swiss resident population to the population covered by the existing Swiss cancer registries, stratified by age and calendar years, language region, and sex. The total number of pLGG for a specific calendar period was then estimated by taking the weighted sum of the observed cases across strata over the study period. To calculate incidence, we obtained data from the Federal Statistical Office on the Swiss resident population at year-end by sex, age (0–17 years), calendar year (1990–2019), and district (called canton) of residence. We calculated crude incidence rates per 100,000 person-years by histopathological subgroups and age groups across the entire study period (1990–2019). Crude incidence rates were calculated by dividing the aggregated estimated number of pLGG (observed for the age groups 0–1, 2–11, 12–15) by the corresponding sum of the mid-year population [22]. The mid-year population for each year was obtained by averaging the end-year populations of that year and the preceding year. The unit of incidence rate is per 100,000 children per year. Kaplan-Meier analysis was used to calculated OS, and log-rank test was used to assess differences. Follow-up time lasted until the date of death, the latest date of available information on vital status, or the end of follow-up, whichever came first. We estimated OS separately according to age at diagnosis, 10-year calendar period of pLGG diagnosis (1990–1999, 2000–2009, 2010–2019), histopathological subgroup, treatment group, and anatomical localization. To investigate potential determinants of EFS, we fitted a Cox model, assuming proportional hazards. Age at diagnosis, sex, calendar period of diagnosis, geographic region, anatomical localization, histopathological subgroup, WHO grade, the presence of NF-1 (yes or unknown), treating hospital and treatment group were included as covariates in the model, and estimated adjusted hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the combined event of progression, recurrence, metastases, and death (whichever came first). Overall P values were estimated by likelihood ratio test comparing the full model, which included all the covariates mentioned above, to the model excluding each time one covariate. We assessed 1-, 5- and 10-year survival (observed and event-free) trends for the periods 1995–2004, 2005–2014 and 2015–2019. For the most recent period (2015–2019), we estimated survival using the period monitoring scheme to get up-to-date long-term survival estimates [23]. We excluded one pLGG patient from all survival analyses, as he was diagnosed by death certificate only (DCO), meaning that the date of death was assigned to the date of diagnosis (zero follow-up time). All analyses were performed in R statistical software [24], and the survival analyses were conducted using “survival” R-package [25]. Results Baseline demographics A total of 591 pLGG patients were obtained, which account for 37.8% of all registered CNS tumors during 1990–2019 (Table 1). Out of those, 315 (53.3%) were diagnosed in males and 276 (46.7%) in females. Most pLGG were diagnosed between the ages of 2 and 11 years (n = 352, 59.6%). Around half of all pLGG were located supratentorial (n = 286, 48.4%), while the other half (n = 266, 45.0%) occurred infratentorial, and only a few were located either in the spinal cord (n = 27, 4.6%) or in multiple or unknown locations (n = 12, 2.0%). Table 1Baseline population characteristicsCharacteristicObserved number of pLGG (N = 591) aDiagnosis age (y) 0–144 (7.4) 2–11352 (59.6) 12–15141 (23.9) 16–1754 (9.1)Sex Male315 (53.3) Female276 (46.7)Calendar period of diagnosis 1990–1999146 (24.7) 2000–2009219 (37.1) 2010–2019226 (38.2)Language region of diagnosis German/Romansh411 (69.5) French/Italian180 (30.5)Anatomical localization Supratentorial286 (48.4) Infratentorial266 (45.0) Spinal cord27 (4.6) Multiple or unknown12 (2.0)Treating hospital SPOG457 (77.3) Non-SPOG134 (22.7)Underlying diseasesb Yes79 (13.4) Not reported512 (86.6)Data is presented as n (%)a Number of histologically verified low-grade gliomas registered by the Childhood Cancer Registry (ChCR)b Underlying diseases consist of epilepsy or seizures, neurofibromatosis 1, and tuberous sclerosispLGG pediatric low-grade gliomas, SPOG Swiss Peadiatric Oncology Group Table 1Baseline population characteristicsCharacteristicObserved number of pLGG (N = 591) aDiagnosis age (y) 0–144 (7.4) 2–11352 (59.6) 12–15141 (23.9) 16–1754 (9.1)Sex Male315 (53.3) Female276 (46.7)Calendar period of diagnosis 1990–1999146 (24.7) 2000–2009219 (37.1) 2010–2019226 (38.2)Language region of diagnosis German/Romansh411 (69.5) French/Italian180 (30.5)Anatomical localization Supratentorial286 (48.4) Infratentorial266 (45.0) Spinal cord27 (4.6) Multiple or unknown12 (2.0)Treating hospital SPOG457 (77.3) Non-SPOG134 (22.7)Underlying diseasesb Yes79 (13.4) Not reported512 (86.6)Data is presented as n (%)a Number of histologically verified low-grade gliomas registered by the Childhood Cancer Registry (ChCR)b Underlying diseases consist of epilepsy or seizures, neurofibromatosis 1, and tuberous sclerosispLGG pediatric low-grade gliomas, SPOG Swiss Peadiatric Oncology Group Baseline population characteristics Data is presented as n (%) a Number of histologically verified low-grade gliomas registered by the Childhood Cancer Registry (ChCR) b Underlying diseases consist of epilepsy or seizures, neurofibromatosis 1, and tuberous sclerosis pLGG pediatric low-grade gliomas, SPOG Swiss Peadiatric Oncology Group Histopathology and treatment In Table 2, over half of the patients were pilocytic astrocytomas (n = 344, 58.2%), followed by other specified and unspecified intracranial and intraspinal neoplasms (n = 116, 19.6%), and other astrocytomas, excluding pilocytic (n = 103, 17.4%), hence, most pLGG were of WHO grade 1 (n = 482, 81.5%). During the first year after diagnosis, the vast majority of pLGG (n = 501, 84.8%) were treated with surgery, either alone (n = 441, 74.6%) or in combination with other treatments (42 with chemotherapy, 11 with radiation, and seven with chemotherapy and radiation). Out of all the surgeries, 52.1% were gross total resections (n = 261). Only very few pLGG were treated with either chemotherapy (n = 19, 3.2%) or radiotherapy (n = 5, 0.8%) or with a combination of both (n = 10, 1.7%) without any surgery beforehand. The NF-1 mutations were reported in 3.7% (n = 22) of our study population. Out of the 22 pLGG with confirmed NF-1, 18 (81.8%) had surgery with or without any additional treatment, which is comparable to the proportion of pLGG without confirmed NF-1 mutations having surgery with or without any other treatment (n = 483, 84.9%). Table 2Observed number of pLGGs between 1990 and 2019 in Switzerland by tumor characteristics and treatments related factorsCharacteristicObserved number of pLGG (N = 591)Histological subtype Astrocytomas (excluded pilocytic)103 (17.4) Pilocytic astrocytomas344 (58.2) Other specified and unspecified intracranial and intraspinal neoplasms a116 (19.6) Other gliomas b28 (4.7)WHO grade Grade 1482 (81.5) Grade 2105 (17.8) Unknown4 (0.7)NF-1 Yes22 (3.7) Not reported569 (96.3)Treatment within the 1 y of diagnosis Surgery441 (74.6) Chemotherapy19 (3.2) Radiotherapy5 (0.8) Surgery and chemotherapy42 (7.1) Surgery and radiotherapy11 (1.9) Radiotherapy and chemotherapy10 (1.7) Surgery, chemotherapy and radiotherapy7 (1.2) Incomplete information48 (8.1) No treatment8 (1.4)Extent of resection Gross total308 (52.1) Subtotal177 (29.9) Biopsy only c90 (15.2) Unclear16 (2.7)Data is presented as n (%)a Consist of neuronal and mixed neuronal-glial tumorsb Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain originc Biopsy was not counted to the treatment group “surgery” given it only served to obtain a histopathologypLGGs pediatric low-grade gliomas, WHO World Health Organization, NF-1neurofibromatosis 1 Table 2Observed number of pLGGs between 1990 and 2019 in Switzerland by tumor characteristics and treatments related factorsCharacteristicObserved number of pLGG (N = 591)Histological subtype Astrocytomas (excluded pilocytic)103 (17.4) Pilocytic astrocytomas344 (58.2) Other specified and unspecified intracranial and intraspinal neoplasms a116 (19.6) Other gliomas b28 (4.7)WHO grade Grade 1482 (81.5) Grade 2105 (17.8) Unknown4 (0.7)NF-1 Yes22 (3.7) Not reported569 (96.3)Treatment within the 1 y of diagnosis Surgery441 (74.6) Chemotherapy19 (3.2) Radiotherapy5 (0.8) Surgery and chemotherapy42 (7.1) Surgery and radiotherapy11 (1.9) Radiotherapy and chemotherapy10 (1.7) Surgery, chemotherapy and radiotherapy7 (1.2) Incomplete information48 (8.1) No treatment8 (1.4)Extent of resection Gross total308 (52.1) Subtotal177 (29.9) Biopsy only c90 (15.2) Unclear16 (2.7)Data is presented as n (%)a Consist of neuronal and mixed neuronal-glial tumorsb Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain originc Biopsy was not counted to the treatment group “surgery” given it only served to obtain a histopathologypLGGs pediatric low-grade gliomas, WHO World Health Organization, NF-1neurofibromatosis 1 Observed number of pLGGs between 1990 and 2019 in Switzerland by tumor characteristics and treatments related factors Data is presented as n (%) a Consist of neuronal and mixed neuronal-glial tumors b Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain origin c Biopsy was not counted to the treatment group “surgery” given it only served to obtain a histopathology pLGGs pediatric low-grade gliomas, WHO World Health Organization, NF-1neurofibromatosis 1 Incidence rates We estimated 641 pLGG diagnosed at age 0–17 years during 1990–2019 and a total incidence of 1.4 pLGG per 100,000 children per year in Switzerland, after accounting for a possibly incomplete registration coverage for pLGG diagnosed at the age of 16–17 (Table 3). Incidence was similar across the older age groups (2–11 years, 12–15 years, and 16–17 years), ranging from 1.4 to 1.5except during infancy (0–1 year), with the lowest estimated incidence of 0.9. Among the histopathological subgroups, pilocytic astrocytoma had the highest incidence of 0.8, whereas other gliomas had the lowest incidence of 0.1. Over the observed interval (1990–1999 versus 2010–2019), incidence rates increased in the infant and adolescent (12–15 years) age groups from 0.85 to 1.01, and from 1.09 to 1.54, respectively. Table 3Incidence of pLGGs per 100,000 children per year between 1990 and 2019 in Switzerland by age groups and histological subtypesCharacteristicNumber of pLGGsIRObserved (N = 591) aEstimated (N = 614) bAge groups (y) 0–144 (7.4)44 (7.2)0.92 2–11352 (59.6)352 (57.3)1.45 12–15141 (23.9)141 (23.0)1.42 16–1754 (9.1%)77 (12.5)1.52Histological subtypes Astrocytomas (excluded pilocytic)103 (17.4)112 (18.2)0.25 Pilocytic astrocytomas344 (58.2)352 (57.3)0.80 Other specified and unspecified intracranial and intraspinal neoplasms c116 (19.6)122 (19.9)0.28 Other gliomas d28 (4.7)29 (4.7)0.07a Number of LGGs registered by the Childhood Cancer Registry (ChCR)b Number of LGGs that would have been observed if registration of cases in 16- to 17-year olds had been nationwidec Consist of neuronal and mixed neuronal-glial tumorsd Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain originpLGGs pediatric low-grade gliomas; IR incidence rate Table 3Incidence of pLGGs per 100,000 children per year between 1990 and 2019 in Switzerland by age groups and histological subtypesCharacteristicNumber of pLGGsIRObserved (N = 591) aEstimated (N = 614) bAge groups (y) 0–144 (7.4)44 (7.2)0.92 2–11352 (59.6)352 (57.3)1.45 12–15141 (23.9)141 (23.0)1.42 16–1754 (9.1%)77 (12.5)1.52Histological subtypes Astrocytomas (excluded pilocytic)103 (17.4)112 (18.2)0.25 Pilocytic astrocytomas344 (58.2)352 (57.3)0.80 Other specified and unspecified intracranial and intraspinal neoplasms c116 (19.6)122 (19.9)0.28 Other gliomas d28 (4.7)29 (4.7)0.07a Number of LGGs registered by the Childhood Cancer Registry (ChCR)b Number of LGGs that would have been observed if registration of cases in 16- to 17-year olds had been nationwidec Consist of neuronal and mixed neuronal-glial tumorsd Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain originpLGGs pediatric low-grade gliomas; IR incidence rate Incidence of pLGGs per 100,000 children per year between 1990 and 2019 in Switzerland by age groups and histological subtypes a Number of LGGs registered by the Childhood Cancer Registry (ChCR) b Number of LGGs that would have been observed if registration of cases in 16- to 17-year olds had been nationwide c Consist of neuronal and mixed neuronal-glial tumors d Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain origin pLGGs pediatric low-grade gliomas; IR incidence rate Observed and event-free survival At the end of follow-up, 91% of children were still alive, and 26% experienced progression, recurrence, metastasis, or death. The vast majority of pLGG patients showed excellent short and long-term survival after their diagnosis. Survival increased during the observed period; for example, during 2015–2019, 5-year OS was 97.1% and 10-year survival was 95.2%, compared with 95.2% and 91.6% during 1995–2004, respectively. On the contrary, we observed a decrease in EFS over time, especially at 5- and 10-year after the pLGG diagnosis (Fig. 1).Fig. 11-, 5-,10-year, and global overall survival curves by calendar period at diagnosis (y) 1-, 5-,10-year, and global overall survival curves by calendar period at diagnosis (y) The OS appeared to be lower for the older age group compared to the rest of the age groups at the end of follow-up, but with no evidence for a significant difference in survival among the different age groups (Table 4, Fig. 2a). However, in terms of EFS after adjusting for other potential determinants, we found that the age at pLGG diagnosis was associated with the risk for recurrence, metastasis, progression, or death. The HR was lower for the age of 2–11 years (adjusted HR = 0.51, 95% CI 0.31–0.84) and 12–15 years (adjusted HR = 0.27, 95% CI 0.14–0.52) compared to the 0–1 year (Table 5). The univariable analysis showed evidence for a difference in OS by histopathological subtype. Astrocytoma, excluding pilocytic astrocytoma, showed a systematically lower OS after the first year of diagnosis compared to the other histopathological subtypes (P = 0.024), with the steepest decrease in the survival of this subtype being observed between two and five years of follow-up (Fig. 2b, Table 4). The highest OS at the end of follow-up was estimated for pilocytic astrocytoma (92.3%), while other intracranial neoplasms showed the highest 5-year OS (98.8%, Table 4). No associations were found between histopathological subtypes and EFS (P = 0.920, Table 5). Table 4Observed survival at 1, 5, and 10 y after the pLGG diagnosis and at the end of the follow-up by clinical and tumor-related characteristicsCharacteristic1 y5 y10 yEnd of follow-upP value aAge group at diagnosis (y) 0–1100.0 (93.0, 100.0)97.1 (81.4, 99.6)97.1 (81.4, 99.6)87.6 (65.2, 96.0)0.833 2–1198.0 (95.8, 99.0)94.8 (91.8, 96.8)92.0 (88.2, 94.6)91.3 (87.3, 94.1) 12–1598.6 (94.3, 99.6)95.4 (90.1, 97.9)92.5 (86.0, 96.1)91.3 (84.2, 95.3) 16–1798.1 (87.1, 99.7)96.2 (85.5, 99.0)89.6 (76.6, 95.5)81.4 (56.4, 92.9)Calendar period of diagnosis 1990–199996.6 (92.0, 98.6)92.5 (86.8, 95.8)86.9 (80.3, 91.5)86.2 (79.4, 90.8)0.107 2000–200998.2 (95.2, 99.3)95.9 (92.2, 97.8)94.0 (89.9, 96.5)90.5 (84.4, 94.3) 2010–201999.6 (96.9, 99.9)96.5 (92.2, 98.4)95.4 (90.4, 97.9)95.4 (90.4, 97.9)Histological subtype Astrocytomas (excluded pilocytic)97.0 (91.1, 99.0)89.8 (81.8, 94.4)85.8 (76.7, 91.6)82.1 (71.5, 89.1)0.024 Pilocytic astrocytomas98.2 (96.1, 99.2)96.0 (93.1, 97.6)94.3 (91.0, 96.5)92.3 (87.3, 95.4) Other specified and unspecified intracranial and intraspinal neoplasms b100.0 (97.4, 100.0)98.9 (92.7, 99.8)92.3 (83.5, 96.5)90.4 (80.6, 95.4) Other gliomas c96.4 (77.2, 99.5)92.9 (74.3, 98.2)88.4 (68.0, 96.2)88.4 (68.0, 96.2)Treatment within 1 y of diagnosis Surgery99.5 (98.2, 99.9)99.0 (97.5, 99.6)97.2 (94.8, 98.5)95.7 (91.8, 97.8)< 0.0001 Chemo94.4 (66.6, 99.2)88.9 (62.4, 97.1)88.9 (62.4, 97.1)77.8 (42.4, 92.9) Radio100.0 (40.0, 100.0)80.0 (20.4, 96.9)60.0 (12.6, 88.2)60.0 (12.6, 88.2) Surgery-chemo100.0 (92.9, 100.0)94.7 (80.3, 98.6)87.1 (68.7, 95.0)70.8 (40.7, 87.5) Surgery-radio100.0 (72.7, 100.0)80.8 (42.3, 94.9)57.7 (22.1, 81.9)57.7 (22.1, 81.9) Radio-chemo80.0 (40.9, 94.6)48.0 (16.1, 74.5)48.0 (16.1, 74.5)48.0 (16.1, 74.5) Surgery-radio-chemo100.0 (57.1, 100.0)40.0 (5.2, 75.3)20.0 (0.8, 58.2)20.0 (0.8, 58.2) Incomplete89.3 (76.3, 95.4)84.5 (70.2, 92.3)82.0 (67.0, 90.6)82.0 (67.0, 90.6) No treatment100.0 (62.5, 100.0)100.0 (62.5, 100.0)100.0 (62.5, 100.0)100.0 (62.5, 100.0)Anatomical localization Supratentorial98.5 (96.2, 99.5)95.5 (92.2, 97.4)93.1 (89.1, 95.7)88.4 (81.9, 92.7)0.0068 Infratentorial98.1 (95.5, 99.2)96.0 (92.7, 97.8)93.8 (89.6, 96.3)93.8 (89.6, 96.3) Spinal100.0 (88.9, 100.0)87.4 (65.6, 95.8)72.8 (49.0, 86.9)72.8 (49.0, 86.9) Multiple or unknown90.9 (50.8, 98.7)90.9 (50.8, 98.7)80.8 (42.3, 94.9)80.8 (42.3, 94.9)Data is presented as observed survival (95% confidence interval, 95%CI)a The P values are calculated by log-rank testsb Consist of neuronal and mixed neuronal-glial tumorsc Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain originpLGG pediatric low-grade gliomas, Chemo chemotherapy, Radio radiotherapy, Surgery-Chemo surgery with chemotherapy, Surgery-Radio surgery with radiotherapy, Surgery-Radio-Chemo surgery with radiotherapy and chemotherapyFig. 25-, 10-year, and total observed survival curves by subgroups. a age at diagnosis; b histopathological subgroup; c treatment type received the first year of diagnosis; d anatomical locationTable 5Adjusted hazard ratios for patient, tumor, and treatment characteristics associated with the composite outcome of death, progression, recurrence, or metastasisCharacteristicHR95% CIP valueAge at diagnosis (y) 0–11.00< 0.0001 2–110.500.31, 0.81 12–150.250.13, 0.47 16–170.650.27, 1.60Sex Female1.000.702 Male0.940.67, 1.30Calendar period of diagnosis 1990–19991.000.071 2000–20090.720.47, 1.09 2010–20191.120.73, 1.71Language region of diagnosis German1.000.344 French/Italian1.200.85, 1.70Anatomical localization Infra1.000.104 Supra1.200.82, 1.75 Spinal2.061.10, 3.88Histological subtype Pilocytic astrocytomas1.000.857 Astrocytomas (excluded pilocytic)0.850.29, 2.49 Other specified and unspecified intracranial and intraspinal neoplasms0.940.57, 1.55 Other gliomas0.630.19, 2.08WHO grade Grade 11.000.615 Grade 21.310.45, 3.84Type of treating hospital Non SPOG1.000.003 SPOG2.621.33, 5.18Treatment within 1 y of diagnosis Surgery only1.00< 0.0001 Surgery with other5.293.45, 8.13 Other a3.331.93, 5.74 Incomplete or no treatment2.641.48, 4.71a Includes radiotherapy, and chemotherapy alone or as a combination treatmentHR hazard ratio, 95% CI 95% confidence interval, WHO World Health Organization, SPOG Swiss Peadiatric Oncology Group Table 4Observed survival at 1, 5, and 10 y after the pLGG diagnosis and at the end of the follow-up by clinical and tumor-related characteristicsCharacteristic1 y5 y10 yEnd of follow-upP value aAge group at diagnosis (y) 0–1100.0 (93.0, 100.0)97.1 (81.4, 99.6)97.1 (81.4, 99.6)87.6 (65.2, 96.0)0.833 2–1198.0 (95.8, 99.0)94.8 (91.8, 96.8)92.0 (88.2, 94.6)91.3 (87.3, 94.1) 12–1598.6 (94.3, 99.6)95.4 (90.1, 97.9)92.5 (86.0, 96.1)91.3 (84.2, 95.3) 16–1798.1 (87.1, 99.7)96.2 (85.5, 99.0)89.6 (76.6, 95.5)81.4 (56.4, 92.9)Calendar period of diagnosis 1990–199996.6 (92.0, 98.6)92.5 (86.8, 95.8)86.9 (80.3, 91.5)86.2 (79.4, 90.8)0.107 2000–200998.2 (95.2, 99.3)95.9 (92.2, 97.8)94.0 (89.9, 96.5)90.5 (84.4, 94.3) 2010–201999.6 (96.9, 99.9)96.5 (92.2, 98.4)95.4 (90.4, 97.9)95.4 (90.4, 97.9)Histological subtype Astrocytomas (excluded pilocytic)97.0 (91.1, 99.0)89.8 (81.8, 94.4)85.8 (76.7, 91.6)82.1 (71.5, 89.1)0.024 Pilocytic astrocytomas98.2 (96.1, 99.2)96.0 (93.1, 97.6)94.3 (91.0, 96.5)92.3 (87.3, 95.4) Other specified and unspecified intracranial and intraspinal neoplasms b100.0 (97.4, 100.0)98.9 (92.7, 99.8)92.3 (83.5, 96.5)90.4 (80.6, 95.4) Other gliomas c96.4 (77.2, 99.5)92.9 (74.3, 98.2)88.4 (68.0, 96.2)88.4 (68.0, 96.2)Treatment within 1 y of diagnosis Surgery99.5 (98.2, 99.9)99.0 (97.5, 99.6)97.2 (94.8, 98.5)95.7 (91.8, 97.8)< 0.0001 Chemo94.4 (66.6, 99.2)88.9 (62.4, 97.1)88.9 (62.4, 97.1)77.8 (42.4, 92.9) Radio100.0 (40.0, 100.0)80.0 (20.4, 96.9)60.0 (12.6, 88.2)60.0 (12.6, 88.2) Surgery-chemo100.0 (92.9, 100.0)94.7 (80.3, 98.6)87.1 (68.7, 95.0)70.8 (40.7, 87.5) Surgery-radio100.0 (72.7, 100.0)80.8 (42.3, 94.9)57.7 (22.1, 81.9)57.7 (22.1, 81.9) Radio-chemo80.0 (40.9, 94.6)48.0 (16.1, 74.5)48.0 (16.1, 74.5)48.0 (16.1, 74.5) Surgery-radio-chemo100.0 (57.1, 100.0)40.0 (5.2, 75.3)20.0 (0.8, 58.2)20.0 (0.8, 58.2) Incomplete89.3 (76.3, 95.4)84.5 (70.2, 92.3)82.0 (67.0, 90.6)82.0 (67.0, 90.6) No treatment100.0 (62.5, 100.0)100.0 (62.5, 100.0)100.0 (62.5, 100.0)100.0 (62.5, 100.0)Anatomical localization Supratentorial98.5 (96.2, 99.5)95.5 (92.2, 97.4)93.1 (89.1, 95.7)88.4 (81.9, 92.7)0.0068 Infratentorial98.1 (95.5, 99.2)96.0 (92.7, 97.8)93.8 (89.6, 96.3)93.8 (89.6, 96.3) Spinal100.0 (88.9, 100.0)87.4 (65.6, 95.8)72.8 (49.0, 86.9)72.8 (49.0, 86.9) Multiple or unknown90.9 (50.8, 98.7)90.9 (50.8, 98.7)80.8 (42.3, 94.9)80.8 (42.3, 94.9)Data is presented as observed survival (95% confidence interval, 95%CI)a The P values are calculated by log-rank testsb Consist of neuronal and mixed neuronal-glial tumorsc Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain originpLGG pediatric low-grade gliomas, Chemo chemotherapy, Radio radiotherapy, Surgery-Chemo surgery with chemotherapy, Surgery-Radio surgery with radiotherapy, Surgery-Radio-Chemo surgery with radiotherapy and chemotherapy Observed survival at 1, 5, and 10 y after the pLGG diagnosis and at the end of the follow-up by clinical and tumor-related characteristics Data is presented as observed survival (95% confidence interval, 95%CI) a The P values are calculated by log-rank tests b Consist of neuronal and mixed neuronal-glial tumors c Consist of oligodendrogliomas, mixed and unspecified gliomas, neuroepithelial glial tumors of uncertain origin pLGG pediatric low-grade gliomas, Chemo chemotherapy, Radio radiotherapy, Surgery-Chemo surgery with chemotherapy, Surgery-Radio surgery with radiotherapy, Surgery-Radio-Chemo surgery with radiotherapy and chemotherapy 5-, 10-year, and total observed survival curves by subgroups. a age at diagnosis; b histopathological subgroup; c treatment type received the first year of diagnosis; d anatomical location Table 5Adjusted hazard ratios for patient, tumor, and treatment characteristics associated with the composite outcome of death, progression, recurrence, or metastasisCharacteristicHR95% CIP valueAge at diagnosis (y) 0–11.00< 0.0001 2–110.500.31, 0.81 12–150.250.13, 0.47 16–170.650.27, 1.60Sex Female1.000.702 Male0.940.67, 1.30Calendar period of diagnosis 1990–19991.000.071 2000–20090.720.47, 1.09 2010–20191.120.73, 1.71Language region of diagnosis German1.000.344 French/Italian1.200.85, 1.70Anatomical localization Infra1.000.104 Supra1.200.82, 1.75 Spinal2.061.10, 3.88Histological subtype Pilocytic astrocytomas1.000.857 Astrocytomas (excluded pilocytic)0.850.29, 2.49 Other specified and unspecified intracranial and intraspinal neoplasms0.940.57, 1.55 Other gliomas0.630.19, 2.08WHO grade Grade 11.000.615 Grade 21.310.45, 3.84Type of treating hospital Non SPOG1.000.003 SPOG2.621.33, 5.18Treatment within 1 y of diagnosis Surgery only1.00< 0.0001 Surgery with other5.293.45, 8.13 Other a3.331.93, 5.74 Incomplete or no treatment2.641.48, 4.71a Includes radiotherapy, and chemotherapy alone or as a combination treatmentHR hazard ratio, 95% CI 95% confidence interval, WHO World Health Organization, SPOG Swiss Peadiatric Oncology Group Adjusted hazard ratios for patient, tumor, and treatment characteristics associated with the composite outcome of death, progression, recurrence, or metastasis a Includes radiotherapy, and chemotherapy alone or as a combination treatment HR hazard ratio, 95% CI 95% confidence interval, WHO World Health Organization, SPOG Swiss Peadiatric Oncology Group The different treatment methods during the first year after diagnosis were associated with different OS (Table 4, Fig. 2c). By the end of follow-up, no deaths were observed among the eight children (1.4%) who didn’t receive any treatment within the first year after diagnosis. Children solely treated with surgery had the highest 5- and 10-year OS (99.0% and 97.2%, respectively), whereas children treated with surgery followed by radio- and/or chemotherapy had the lowest 5- and 10-year OS (40% and 20%, respectively). Treatment during the first year after diagnosis appeared to be associated the strongest with EFS, with children being treated with surgery only showing the lowest risk for recurrence, metastasis, progression or death compared to the children that received other treatment methods or combination treatment with surgery (Table 5). The OS appeared to differ depending on the anatomical localization of the pLGG (P = 0.007). Tumors located in the spine showed significantly lower 5-,10-year and OS compared to pLGG located elsewhere (Table 4). The 10-year survival rate for supratentorial pLGG was 93.1% (95% CI 89.1%–95.7%), for infratentorial pLGG was 93.8% (95% CI 89.6%–96.3%), for spinal pLGG was 72.8% (95% CI 49.0%–86.9%), and for multifocal or unknown location pLGG was 80.8 (95% CI 42.3%–94.9%). Spinal pLGG also showed a higher risk for recurrence, metastasis, progression, or death compared to infratentorial tumors (adjusted HR = 2.09, 95% CI 1.10–3.97). Nevertheless, there was no evidence for association between anatomical localization and EFS (Table 5, Fig. 2d). Discussion This is the first study on the incidence and survival of children with pLGG in Switzerland. We estimated a total incidence of 1.4 pLGGs per 100,000 children per year during 1990–2019. The histopathological subtype of pLGG appears to influence OS, with pilocytic astrocytoma showing the highest OS rates. Few children had spinal pLGG, which was associated with the lowest OS). Treatment during the first year after diagnosis appeared to influence OS and be the strongest determinant of EFS, a finding most likely biased by tumor characteristics, meaning that more severe tumors might demand exposure to more treatments and have worse outcomes. Three-quarters of the children underwent surgery alone as treatment and showed the highest OS compared to those who required radio- and/or chemotherapy after surgery. The increased incidence rate over the observed time periods could be due to improved detection rates with increased patients receiving imaging. Further, other factors could be due to the overall higher surgical rates, and more patients being deemed eligible for surgery. The reported demographics in our epidemiological cohort, including age, sex, anatomical location, and histopathology, are comparable to previously reported studies [5, 26–28]. Since the proportion of children with registered underlying diseases was small in our study (13%), we decided to discuss only the presence of NF-1. The NF-1 mutations were reported for 4% in our study, which is lower compared to other cohorts where it has been reported in up to 10.5% [6]. This difference could be due to underdiagnosis of disease, especially when genetic testing was not as widespread as nowadays, delayed diagnosis in adulthood, or underreporting (underlying diseases have not been systematically collected by the ChCR and were only registered when indicated on the pathology reports of patients). Further, patients with known NF-1 might not undergo biopsy due to very pathognomonic tumors and, as a consequence, might not be registered in the ChCR, which could be contributing to the low NF-1 rate [12]. Our analysis showed that despite the gradual increase in survival over the observed calendar periods, indicating an improvement in the overall care of pLGG, the EFS subsequently decreased over the calendar periods, with the lowest 5-year EFS of 70.9% between 2015 and 2019 compared to 81.2% between 1995 and 2004. This could be due to improved follow-ups and diagnostics, allowing treating physicians to detect even small recurrences. Furthermore, this trend has been shifting towards an intended subtotal resection in eloquent areas to avoid neurological deficits, due to improved adjuvant treatments, which could lead to a higher rate of radiological progression, but can be successfully managed with medication such as chemotherapy [29, 30]. Further, due to less toxic therapies, the interval to treatment might be shorter, given different indications over time. The high survival rates of over 90% for pLGG have also been reported in other studies from first-world countries [5, 26, 27, 31]. Despite the excellent outcomes in high-income countries, there is a large disparity in middle- and low-income countries [31–33], where survival rates range from approximately 60% to over 90% [31]. Given this disparity around the globe, pLGG is one of the tumors included in the WHO initiative against cancer, aiming to increase survival by 2030 worldwide [34]. Further, it has to be expected that pLGG is underdiagnosed in developing countries, given the limited access to medical resources [31]. Prediction models estimated that 2.9 million childhood cancer cases will be missed between 2015 and 2030, of which a large proportion will include pLGG [35]. Ultimately, one of the key prognostic factors for survival and outcome remains the country of residence when being diagnosed with pLGG [33]. Further, in certain countries, access to more novel treatments such as MEK inhibitors might be limited. Apart from these epidemiological and social challenges, other reported prognostic factors on outcome and survival are histopathology, molecular pathology, and the need for radiation therapy [5]. In our study, children undergoing radiation, either alone or in combination with surgery and/or chemotherapy, comprised only 6% of our population. At the end of our study follow-up, OS after radiation only, after radiation and chemotherapy, and after surgery, radiation, and chemotherapy was 60%, 58%, and 20%, respectively. Comparable results were observed by an analysis of the SEER database including 4040 children diagnosed with pLGG from 1973 to 2008 [5]. Given that our study only focused on including treatment in the first year, we might have recorded lower adjuvant treatment rates, which might have been administered years after initial diagnosis. Further, children requiring radiation have been shown to experience a significantly lower quality of life during long-term follow-up [5]. However, this could also be biased by the fact that children requiring radiation have either a remaining tumor with a less favorable histology or are not amenable to surgery or recurrence, which could influence the OS more than the radiation itself [5]. The pLGG is a heterogeneous group of different tumors. Given our study period (1990–2019), the histopathological subtypes in our study do not follow the newest WHO 2021 guidelines. Further, our data does not include any molecular characteristics of the included pLGG. Molecular information is now included in the WHO classification and is a key determining factor of prognosis. A BRAF aberration is found in up to 70% of all sporadic pLGG and the somatic driver mutation is aberrant signaling of the mitogen-activated protein kinase (MAPK) pathway [32, 36–38]. These molecular aberrations, which have been discovered with increasing precision over the last 1–2 decades, are not only important for classification but also have an impact on prognosis and treatment [7, 37]. We observed in our study that children aged 16–17 years had a lower survival compared to the other age groups (Table 4). This could be because pLGG occurring in late teenage years might resemble adult LGGs, possibly harboring different mutations [5, 12, 39, 40]. A large analysis of mutations in adolescents and young adults showed that only 30% harbored typical rat sarcoma virus (RAS) / MAPK pathway mutations, which were associated with a better outcome than gliomas lacking these molecular mutations [41, 42]. Some patients harbor specific mutations which predispose them to a poorer outcome even without malignant transformation, which is exceedingly rare, and only occurs in approximately 3% of all pLGG [39, 43]. Despite advances in targeted therapies, the mainstay of treatment is still surgical resection, which was also reflected in our cohort and other studies, with 75% of all children undergoing only surgery as their primary treatment, which is comparable to other studies [5, 32, 44, 45]. However, tumors that cannot be safely removed due to their eloquent region, recurrent or multifocal tumors, often require additional treatment. Even in the molecular era, this usually includes chemotherapy as a first adjunct. The first chemotherapy regimens used were a combination of carboplatin or vincristine or multi-agent combinations of thioguanine, procarbazine, chloroethyl-cyclohexyl-nitrosourea (CCNU, lomustine), and vincristine (TPCV regimen). Single-agent vinblastine was then shown as equally effective and less burdensome [6, 32, 46–49] and is, in principle, still the first-line treatment approach for non-BRAF V600E-mutated pLGG today. However, given that pLGG with its high OS rates, should be regarded as a chronic disease, the toxicity-induced side effects of these chemotherapeutic agents are not to be underestimated [15, 27]. In our cohort, 7.1% of all children received surgery followed by chemotherapy in their first year of treatment, while 3.2% received upfront chemotherapy alone. The severe long-term side effects, including neurocognitive and endocrine deficiencies in these children, must be considered when recommending adjuvant treatments. Nowadays, it is standard practice that adjuvant drug therapies are only administered if tumors show progression or symptoms, and the associated risk of a bad outcome outweighs the risk of therapy [32]. Additionally, patients with NF-1 have a higher risk of developing secondary radiation-induced neoplasms and should generally be avoided in this population. However, radiation therapy has also undergone modernization over the years, becoming more precise with lower rates of side effects [50]. However, nowadays it is only administered in very rare cases. In Switzerland, children standardly undergo proton beam radiation therapy due to its improved clinical outcomes [51]. Due to the additional molecular and genetic information gained over the last two decades, newer targeted therapies have been established for pLGG. One of the main paradigm changes was the introduction of the MEK inhibitors such as selumetinib or trametinib. With these targeted therapies up to 50% of all children reach a partial response [52]. A recent randomized controlled trial (RCT) comparing MEK/BRAF inhibitors (trametinib plus dabrafenib) versus chemotherapy (carboplatin plus vincristine) in patients with the BRAF V600E mutation showed that children in the MEK/BRAF inhibitor group had a significantly higher response rate of 47% compared to 11% in the chemotherapy group, improved quality of life, and a longer progression-free survival of 20.1 months versus 7.4 months [53]. After this trial, the combination of MEK/BRAF inhibitors became the new first-line treatment for BRAF V600E mutated pLGG. Given the inclusion period in our cohort, the data regarding the molecular findings was absent, and all patients were diagnosed and treated before the widespread use of MEK and or BRAF inhibitors in pLGG. The main challenges in the use of targeted therapies are the development of resistance and determining the optimal treatment duration, due to the well-known rebound phenomenon after discontinuation of targeted drugs (especially in BRAF V600E mutated tumors) [32]. In our cohort, only 1.4% children did not undergo any treatment. This could either be due to very small lesions which were just followed radiographically, or the refusal of therapy. Especially in children with NF-1, radiographic findings can be very pathognomonic, for which surgery did not seem warranted [54, 55]. Overall, 7.1% did not undergo any surgical treatment in their first year. This number is very low compared to a German and a British series, where up to 35.2% did not undergo any surgery [6, 56]. This difference could be partially due to their higher percentage of NF-1 children or different tumor sizes upon presentation given the different healthcare systems or heterogeneity in the different treatment paradigms. The main strength of our study was that our population was determined by using comprehensive cancer registry data in Switzerland. The ChCR has acquired high completeness since 1995, especially for the age group 0–15 years a 95% completeness can be achieved [19]. Its further benefits from a high-quality long-term follow-up for most registered patients with a lost-to-follow-up rate less than 2%. This study still has several limitations. We did not include any information about molecular diagnostics, and the classification of tumors likely differ in some cases compared to the current WHO classification. Given the lack of any molecular findings, no comparison to the up-to-date treatment guidelines and studies can be drawn. Further, for any midline pLGG, no comment regarding a possible H3K27 mutation is possible, which might lead to some high-grade gliomas being wrongfully diagnosed as pLGG based on histopathology alone. This might lead to a lower OS for pLGG, excluding pilocytic astrocytoma, than in other studies. We only registered treatment in the first year after diagnosis and hence cannot comment on any further treatment during long-term follow-up. Given the long-term survival, many children might undergo additional treatment years after their initial diagnosis, which might not be reflected in our results. Further, patients with high-risk tumor features may have been exposed to treatments because their risk of progression was higher in the first place. Similarly, patients who experience recurrence or progression within the first year following diagnosis typically require further therapeutic interventions beyond primary surgery. Therefore, poor EFS or increased risk factors for poor overall survival may correlate with greater cumulative treatment exposures. Due to the study interval from 1990–2019, targeted therapies are not listed as a treatment option, which, in the current era, are frequently used in relapsed or refractory pLGG after chemotherapy and in BRAF V600-altered pLGG as frontline therapy. Additionally, we do not have any information about the chemotherapy regimens or radiation protocols these children were treated with. The ChCR further does not contain any information about the cognitive or quality of life outcomes of these children, and we cannot make any statement about any possible side effects of their treatment or primary tumor. Moreover, our data did not distinguish between different causes of death in our study; hence do not know if the children succumbed to their primary diagnosis or treatment side effects. In conclusion, this the first nationwide study in Switzerland to report on the epidemiology of pLGG, and found a total incidence of 1.4 per 100,000 children per year. OS was excellent, with 91% of patients, while 25.8% experienced progression, recurrence, metastasis, or death. Survival was notably lower in 16–17 years group, and the event hazard was significantly higher among 0–1 year group compared to older children. Compared to other histopathological subtypes, children with non-pilocytic astrocytoma had lower survival after the first year of diagnosis. The poorest outcomes were observed in children with spinal tumors and in those who underwent radiation therapy, either alone or in combination with other treatments. Future studies incorporating molecular diagnostics and treatment regimens should be conducted to enhance the understanding of pLGG epidemiology.