You are here

Article Text

Article menu

Download PDFPDF

Original research
Symptom interval and treatment burden for patients with malignant central nervous system germ cell tumours
  1. James Hayden1,2,
  2. Matthew J Murray3,
  3. Ute Bartels4,
  4. Thankamma Ajithkumar5,
  5. Brinda Muthusamy6,
  6. Anthony Penn7,8,
  7. Gabriele Calaminus9,
  8. James Nicholson10
  1. 1 Department of Paediatric Haematology and Oncology, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK
  2. 2 Department of Paediatric Haematology and Oncology, Great North Children’s Hospital, Newcastle upon Tyne, UK
  3. 3 Department of Paediatric Haematology and Oncology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK
  4. 4 Department of Paediatric Haematology and Oncology, Paediatric Brain Tumour Program, SickKids, Toronto, Ontario, Canada
  5. 5 Department of Radiation Oncology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, Cambridgeshire, UK
  6. 6 Department of Paediatric Neuro-Ophthalmology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK
  7. 7 Children’s Brain Tumour Research Network (CBTRN), Royal Manchester Children's Hospital, Manchester, UK
  8. 8 Faculty of Biology, Medicine and Health, University of Manchester, Manchester, UK
  9. 9 Department of Paediatric Haematology and Oncology, University Hospital Bonn, Bonn, Germany
  10. 10 Paediatric Oncology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK
  1. Correspondence to Dr Matthew J Murray, Department of Paediatric Haematology and Oncology, Cambridge University Hospitals NHS Foundation Trust, Cambridge CB2 0QQ, UK; mjm16{at}cam.ac.uk

Abstract

Objective Patients with central nervous system germ cell tumours (CNS-GCTs) commonly initially present to primary care or general paediatricians. Prolonged symptom intervals (SI) are frequently seen in CNS-GCTs and have been associated with inferior outcomes in other brain tumours. This study reviewed the clinical presentation of CNS-GCTs and examined the effect of prolonged SI.

Design/Setting/Patients/Outcomes International multicentre 10-year retrospective study (2002–2011 inclusive), across six international paediatric oncology treatment centres. All newly diagnosed patients with CNS-GCT were included. Main outcome measure was time interval from first symptom to diagnosis.

Results The study cohort included 86 (58 males:28 female) patients (59 ‘germinoma’ and 27 ‘non-germinomatous’ GCTs), with tumours being pineal (n=33), suprasellar (n=25), bifocal (pineal+suprasellar; n=24) and ‘other’ site (n=4), of which 16 (19%) were metastatic. Median age at diagnosis was 14 years (0–23 years). The time to diagnosis from first symptom (SI) was 0–69 months (median 3 months, mean 9 months). A prolonged SI (>6 months) was observed in 28/86 patients (33%) and significantly associated with metastatic disease (11/28 (39%) vs 5/58 (9%); p=0.002)) at diagnosis, but not overall survival. With prolonged SI, endocrine symptoms, particularly diabetes insipidus, were more common (21/28 (75%) vs 14/58 (24%) patients; p<0.002), but raised intracranial pressure (RICP) was less frequent (4/28 (14%) vs 43/58 (74%) patients; p<0.001)) at first symptom.

Conclusions One-third of patients with CNS-GCT have >6 months of symptoms prior to diagnosis. Delayed diagnosis is associated with metastatic disease. Early symptom recognition, particularly related to visual and hormonal disturbances in the absence of RICP, may improve timely diagnosis, reduce metastatic disease frequency and consequently reduce treatment burden and late effects.

  • brain tumour
  • symptom interval
  • prolonged symptom interval
  • Outcomes research
  • intervention

https://doi.org/10.1136/archdischild-2019-317245

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    What is already known on this topic?

    • There is wide variation in the time taken from initial symptom development to diagnosis of a brain tumour.

    • Prolonged symptom intervals (>6 months) in patients with intracranial malignant germ cell tumours have previously been shown to be associated with increased risk of metastasis in single-institution series.

    • Educational interventions have been shown to reduce the interval to diagnosis from first symptom in patients with brain tumours.

    What this study adds?

    • The increased risk of metastasis is confirmed for patients with prolonged symptom intervals (>6 months) diagnosed with malignant central nervous system germ cell tumours in this international multi-institution series.

    • Such patients receive a greater burden of subsequent treatment (craniospinal irradiation) than patients with localised disease, with associated increased risk of late effects.

    • Interventions to raise awareness and promote early diagnosis may minimise the need for craniospinal irradiation and the subsequent development of late effects in such patients.

    Background

    Central nervous system (CNS) germ cell tumours (GCT) are a group of rare tumours predominantly occurring in childhood and adolescence (incidence 0.1–0.2 per 100 000) and account for 3.4% of all tumours in childhood.1 2 There is variation in these figures, with an increased incidence in Asian countries where they account for up to 14% of all CNS tumours.3–6 These tumours are considered to originate from primordial germ cells in the developing embryo that then migrate along the midline of the body during embryogenesis.7–10 For classification and treatment purposes, malignant CNS-GCT can be further divided into ‘germinomas’ and ‘non-germinomatous’ GCT (NGGCT) based on circulating markers and histology. Germinoma may be treated effectively with chemotherapy and limited field radiotherapy for localised cases, usually whole ventricular fields with varying practices internationally in relation to focal tumour boosts, and with craniospinal irradiation (CSI) for metastatic cases, both with excellent outcomes.11 NGGCT still have good outcomes, with progression-free survival rates of 72% and 68%, and overall survival rates of 82% and 75%, for localised and metastatic patients, respectively, but require a more intensive combined chemoradiotherapy approach for cure than germinomas.12 While patients with metastatic NGGCT are treated with CSI, the radiotherapy management of localised NGGCT differ globally and varies from focal radiotherapy to CSI.13 14

    CNS-GCTs typically arise in midline locations, most commonly in the pineal and suprasellar regions.15 Tumours arising within the pineal region commonly obstruct cerebrospinal fluid (CSF) flow, giving rise to symptoms and signs of raised intracranial pressure (RICP) including nausea, vomiting and headaches.16 Pineal tumours may also present with ‘Parinaud’s syndrome’, caused by tectal plate compression. The syndrome comprises a limitation of upgaze, loss of the pupillary light reflex with preservation of the accommodation pupil reflex and convergence retraction nystagmus. The latter consists of both eyes converging and retracting within the globe on attempted quick upgaze. The upper lids can also retract bilaterally as part of the syndrome. Patients may report double vision and oscillopsia (where objects in the visual field appear to oscillate); and can develop a ‘chin-up’ head posture due to the limitations of upgaze. Suprasellar tumours may compress the optic chiasm giving rise to visual field defects and endocrinopathies, including most frequently diabetes insipidus (DI), and effects on puberty and growth from effects on the hypothalamus and pituitary gland.17 18 ‘Bifocal’ tumours are those that arise synchronously at both suprasellar and pineal sites; in the absence of disease elsewhere these are considered localised.11 12 Patients with CNS-GCTs may also present with behavioural changes, altered school performance and altered neurology, especially in those tumours arising within the basal ganglia.19

    A prolonged symptom interval (SI; defined as the time from first reported symptom to diagnosis) has been reported to occur in the diagnosis of several paediatric brain tumours.20–23 Delays in diagnosis have also been identified as a leading cause for malpractice claims as well as a source of remorse for both parents and clinicians.23 Reports from two institutional series have identified a wide variation in time from first symptom to diagnosis, with diagnosis occurring up to 3 years after symptoms first develop.20 21 In recent single-centre studies, 38/70 (54%)21 and 9/30 (30%)19 of patients with CNS-GCT had an SI>6 months. In another study of 181 patients with CNS-GCT, 17 patients (9%) had a delayed diagnosis of >3 months and survival in a small subset of germinoma patients (13 of 119 cases; 11%) was negatively affected by this.24 However, it should be noted that these data were not for time from first presentation to diagnosis, but instead for patients with >3 months delay from initial brain MRI to the definitive CNS-GCT diagnosis.24 The authors noted that clinical factors other than diagnostic delay, and which remain undetermined, may therefore have contributed to these adverse outcomes.24 A study of 53 patients presenting with suprasellar GCTs showed that a substantial proportion experienced a delay in the time to diagnosis.25 Although not affecting survival, the authors concluded that the delay may increase the risk of short stature and highlighted the importance of a detailed medical history and early neuroimaging.25 Evidence has also recently emerged for the role of education in improving the time taken to diagnosis. Following a national survey in the UK in 2006, the ‘HeadSmart: Be Brain Tumour Aware’ campaign was launched in the UK in 2011 to promote awareness of symptoms associated with brain tumours.26

    Here, we have undertaken the first international, multicentre review of the impact of prolonged SI on patients with CNS-GCTs. The findings have relevance for opticians, ophthalmologists, general practitioners and paediatricians, who are likely to be the first clinicians to interface with such patients. A high index of suspicion for CNS-GCT needs to be maintained for such patients, particularly in patients presenting without RICP but with subtle new visual field defects and endocrinopathies causing DI and/or affecting growth/puberty.

    Methods

    Study design and patient eligibility

    We undertook a retrospective clinical chart review of all patients diagnosed with a malignant CNS-GCT over a 10-year period, from January 2002 to December 2011 from six international paediatric haematology and oncology centres across three countries, four within the UK (Alder Hey Children’s NHS Trust, Liverpool, UK; Cambridge University Hospitals NHS Foundation Trust, UK; Royal Manchester Children’s Hospital, Manchester University NHS Foundation Trust, Manchester, UK and Great North Children’s Hospital, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK), one in Germany (University Hospital, Bonn) and one in Canada (The Hospital for Sick Children, Toronto, Canada). Resources were not available to undertake retrospective data collection at other paediatric haematology and oncology centres.

    Clinicopathological information was extracted from the medical record including tumour type (germinoma or NGGCT), tumour location, presence or absence of metastasis (other sites of radiological disease or positive CSF cytology), age at first reported symptom, age at diagnosis, presenting symptom and symptoms at diagnosis. Symptoms were classified into three groups, namely: (1) RICP (headaches and vomiting); (2) endocrinopathy (polydipsia and/or polyuria, suggestive of DI, and failure to thrive/slow growth) and (3) visual changes/disturbance. It should be noted that pubertal status was poorly and unreliably documented at the time of referral and therefore pubertal delay/arrest was not included in reported symptoms for the purposes of this study.

    Outcome measures and statistical analysis

    The SI was calculated as the time from first symptom to diagnosis. A prolonged SI was defined as an SI>6 months, in order to allow direct comparison of our results with previous studies.19 21 Associations between SI and clinical/tumour characteristics were evaluated with the Fisher’s exact test and performed using IBM SPSS Statistics Premium 24 (IBM SPSS Statistics for Windows, V.24.0, IBM, Armonk, New York, USA). Comparisons of clinical characteristics were only analysed when there were at least five patients within a group. Due to multiple testing, an increased stringency of p<0.01 was considered significant.

    Results

    Eighty-seven patients were initially identified. For one patient, information regarding the first symptom could not be identified, and therefore this patient was excluded from subsequent analyses, as SI information was missing. The remaining 86 patients in the study cohort are summarised in table 1. Forty-five patients were from the UK (Alder Hey Children’s NHS Trust, Liverpool, UK, n=11; Cambridge University Hospitals NHS Foundation Trust, UK, n=13; Royal Manchester Children’s Hospital, Manchester University NHS Foundation Trust, Manchester, UK, n=10 and Great North Children’s Hospital, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK, n=11), 30 from Canada (The Hospital for Sick Children, Toronto, Canada) and 11 from Germany (University Hospital, Bonn). Two-thirds (n=58/86; 67%) of patients were male. By local report, 59/86 (69%) were diagnosed with a germinoma and 27/86 (31%) with an NGGCT. Thirty-three tumours were located in the pineal region, 25 were suprasellar, 24 were bifocal (pineal and suprasellar site) and 4 were located at other sites within the brain (figure 1A). The tumours were metastatic at diagnosis in 16 of 86 cases (19%) (figure 1B), of which three-quarters (12/16) were germinomas and one-quarter (4/16) NGGCT. The median age at diagnosis was 14 years (0–23 years).

    Figure 1
    Figure 1

    Clinicopathological information for the 86 patients at the time of central nervous system germ cell tumour (CNS-GCT) diagnosis: (A) by tumour site and (B) by metastatic status. Bifocal disease involves both pineal and suprasellar sites.

    View this table:
    Table 1

    Clinical and diagnostic characteristics of the 86 patients in the study cohort

    The initial symptoms experienced by the patients in the cohort included those attributable to RICP (47/86; 55%), endocrine (35/86; 41%) and visual (21/86; 24%) disturbance (table 1). By the time of diagnosis however, the range of symptoms had increased, and >70% of patients had developed RICP symptoms (n=62/86; 72%). In addition, over half the patients had developed endocrine (n=46/86; 53%) or visual (n=45/86; 52%) symptoms by this time (figure 2A).

    Figure 2
    Figure 2

    Symptoms and symptom interval (SI) of the patient cohort: (A) number of patients affected by raised intracranial pressure (RICP), endocrine and visual symptoms at the time of first symptom (pale grey bars) and at the time of diagnosis (darker grey bars); (B) SI at the time of diagnosis. Prolonged SI defined as >6 months.

    A prolonged SI (>6 months) was observed in 28/86 (33%) of patients (figure 2B). Overall, the time to diagnosis from first symptom (SI) ranged from 0 to 69 months (median 3 months and mean 9 months) (figure 3). There was no predilection for prolonged SI with age at diagnosis, sex or CNS-GCT subtype (table 2). Regarding location of disease, pineal site was observed substantially less frequently in those with prolonged SI (3/28 patients; 11%) than in those with an SI≤6 months (30/58 patients; 52%) (p=0.003) (figure 4A). A prolonged SI was also significantly associated with metastatic disease at diagnosis (11/28 vs 5/58; p=0.002 at diagnosis) (figure 4B). There were no other clinicopathological correlations identified with prolonged SI. Two patients in the whole study cohort died of disease; both had an NGGCT and the SI for both patients was <2 months; this was not significant (p=1.00; data not shown).

    Figure 3
    Figure 3

    Individual symptom intervals at the time of diagnosis. The vertical dotted line indicates the threshold for definition of prolonged symptom interval (SI), namely >6 months (28 of 86 patients).

    Figure 4
    Figure 4

    Clinical associations with prolonged symptom interval (SI): (A) negative association of pineal tumour site with prolonged SI (p=0.003); (B) association of prolonged SI with metastatic disease at diagnosis (p=0.002).

    View this table:
    Table 2

    Clinical characteristics of the cohort of 86 patients and association with symptom interval (SI)

    The pattern of symptoms differed significantly in those with a prolonged SI, both at the time of first symptom and at the time of diagnosis (table 2). At first symptom, endocrine symptoms, particularly DI, were more frequent (n=21/28; 75%) in those with prolonged SI compared with those who were diagnosed ≤6 months (n=14/58; 24%) (p<0.0001). In contrast, RICP was an infrequent initial symptom (n=4/28; 14%) in those with a prolonged SI, but was typically observed in those with an SI ≤6 months (n=43/58; 74%) (p<0.001). At the time of diagnosis, these differences in endocrine symptoms and RICP were still maintained (table 2).

    Discussion

    This is the first international multi-institutional study investigating SI in patients diagnosed with CNS-GCTs. We have shown that the interval from first symptom to diagnosis may vary significantly, and in some cases may extend to several years. A progression of symptoms is identified from first initial presentation report to eventual diagnosis, with the majority of patients (72%) developing symptoms of RICP by the time of actual diagnosis. Unsurprisingly, pineal site of disease, which frequently presents relatively early and acutely with RICP, was less frequently associated with prolonged SI. For most patients therefore, it is the development of RICP, rather than endocrine or visual symptoms, that appears to prompt referral and investigation leading to diagnosis. In suprasellar tumours, hydrocephalus is usually a late and more insidious event. The development of RICP and hydrocephalus in itself, however, can be life-threatening and may contribute to visual and neurocognitive sequelae.27 Consequently, opticians and ophthalmologists, and general practitioners and general paediatricians, need to maintain a low threshold for appropriate investigation of patients presenting with new endocrine complaints and/or visual symptoms, which may be suggestive of a CNS-GCT. From an ophthalmological perspective, patients presenting with transient visual obscurations, double vision, oscillopsia, loss of vision or visual field defects (particularly, but not exclusively, in the presence of RICP) should be seen urgently by an optometrist or ophthalmologist for a full assessment including oculomotility and retinal examination for papilloedema. However, it is important that at initial presentation, the receiving healthcare professional attempts a basic evaluation of pupils, visual fields, eye movements and retinal check and acts on positive findings to prevent delays that can occur within the referral process to ophthalmology.

    In our study, one-third of patients (33%) had symptoms for >6 months prior to diagnosis, that is, had a prolonged SI. This result is consistent with reports from previous single institutional studies. Of particular concern is the strong association of prolonged SI with increased risk of metastatic disease. For such patients, CSI will be required for cure, increasing the risk and burden of late effects that would be expected in such patients. These late effects include endocrine deficiency due to radiotherapy to the pituitary gland and hypothalamus, impairment of growth and bone development due to growth hormone deficiency and impaired spinal growth in prepubertal children, neurocognitive dysfunction and vasculopathies with increased risk of stroke. Less common late effects include dose-dependent ototoxicity, cardiac toxicity due to exit dose to the heart from the spinal field and secondary induced cancers. Most of these side effects are dependent on age at radiotherapy, and total dose to, and irradiation volume of, critical organs. Early diagnosis thus may avoid the need for CSI and thereby minimise the risk of significant late-radiation sequelae. Acute radiation toxicities are generally minimal and self-limiting; but acute haematological toxicities (particularly common with CSI) can be a significant problem, especially after chemotherapy.

    Previous reports have only described single-centre studies examining the incidence and/or impact of delayed diagnosis in patients with CNS-GCT.19 21 24 These studies are limited in that they do not reflect any wider national or international issues that may affect the patient diagnostic pathway and it is therefore difficult to extrapolate from them to a broader population. Indeed, while Phi et al 24 reported a significant association of delayed diagnosis with an inferior outcome in a subset of germinoma cases (n=13) among a cohort of 181 patients with localised CNS-GCT from their institution, this result was not seen in the other single-centre studies19 21 24 nor replicated in our multicentre study. This is likely due to a number of reasons. First, definition of delay for this outcome in the study by Phi et al was based on time to definitive CNS-GCT diagnosis from first MRI scan, and the authors acknowledge that undetermined clinical factors besides delayed diagnosis may therefore have contributed to this observation.24 Second, it is also not clear what treatment these patients received, particularly important as only 8 of the 13 patients with localised germinoma (62%) subsequently survived, a very low figure given current reported outcomes.11 Although limitations of our study include its retrospective nature and the relatively modest total number of patients included (n=86 cases), patient data were derived from six tertiary paediatric oncology centres across three countries (four centres within the UK, one in Germany and one in Canada). The strong association between prolonged SI and increased risk of metastatic disease found in this study therefore likely represents a true reflection of the global challenges which need to be addressed in order to advance definitive CNS-GCT diagnosis for such patients. As has been identified from the HeadSmart experience,26 strategies to improve the time to diagnosis may be undertaken at many levels across the patient diagnostic pathway, including public, primary care and secondary care awareness and education.28 29 The HeadSmart intervention has been successful in the UK in reducing the interval to diagnosis for brain tumours in general from a median of 14 weeks in 2006 to just 6.7 weeks in 2013.26

    We acknowledge that the first symptom as defined in this study was derived retrospectively based on referral information and clinical notes and, consequently, the symptoms that patient/families described may not necessarily have been reported to a healthcare professional at the time. In this study, we are able to confirm more advanced (metastatic) disease in those with a prolonged SI prior to diagnosis, but cannot determine whether such a prolonged SI is primarily due to delayed reporting by patients/families or whether reporting is occurring, but onward referral is then not being made. This is where maintaining the high profile of established campaigns such as HeadSmart, which increase both public and healthcare professional awareness, may potentially make a difference in the future.

    In summary, in order to improve the outcome for patients with CNS-GCTs, a focus on the prediagnostic pathways is necessary. Our findings have relevance for clinicians who are generally the first to see such patients, namely general practitioners and paediatricians, where a high index of suspicion needs to be maintained for CNS-GCT and other brain tumours. Reduction in the SI may enable a reduction in overall treatment burden and, in particular, the need for CSI, thus reducing the burden of late effects in these patients. Future clinical trials should record details pertaining to the prediagnostic pathway if feasible, to facilitate further analysis in larger multicentre trial cohorts of the effect of prolonged SI in patients with CNS-GCT.

    References

    1. ↵
      2. Kong Z ,
      3. Wang Y ,
      4. Dai C , et al
      . Central nervous system germ cell tumors: a review of the literature. J Child Neurol 2018;33:610–20.doi:10.1177/0883073818772470
      OpenUrl
    2. ↵
      2. Dolecek TA ,
      3. Propp JM ,
      4. Stroup NE , et al
      . CBTRUS statistical report: primary brain and central nervous system tumors diagnosed in the United States in 2005-2009. Neuro Oncol 2012;14 Suppl 5(Suppl 5):v1–49.doi:10.1093/neuonc/nos218
      OpenUrl
    3. ↵
      2. Poynter JN ,
      3. Fonstad R ,
      4. Tolar J , et al
      . Incidence of intracranial germ cell tumors by race in the United States, 1992-2010. J Neurooncol 2014;120:381–8.doi:10.1007/s11060-014-1562-7
      OpenUrl
    4. ↵
      2. Parkin D ,
      3. Kramarova E ,
      4. Draper G , et al
      . International incidence of childhood cancer(Volume II), 1998.
    5. ↵
      2. Echevarría ME ,
      3. Fangusaro J ,
      4. Goldman S
      . Pediatric central nervous system germ cell tumors: a review. Oncologist 2008;13:690–9.doi:10.1634/theoncologist.2008-0037
      OpenUrlAbstract/FREE Full Text
    6. ↵
      2. Lee SH ,
      3. Jung K-W ,
      4. Ha J , et al
      . Nationwide population-based incidence and survival rates of malignant central nervous system germ cell tumors in Korea, 2005-2012. Cancer Res Treat 2017;49:494–501.doi:10.4143/crt.2016.129
      OpenUrl
    7. ↵
      2. Teilum G
      . Classification of endodermal sinus tumour (mesoblatoma vitellinum) and so-called "embryonal carcinoma" of the ovary. Acta Pathol Microbiol Scand 1965;64:407–29.doi:10.1111/apm.1965.64.4.407
      OpenUrlPubMedWeb of Science
    8. ↵
      2. Rescorla FJ
      . Pediatric germ cell tumors. Semin Surg Oncol 1999;16:144–58.doi:10.1002/(SICI)1098-2388(199903)16:2<144::AID-SSU6>3.0.CO;2-M
      OpenUrlCrossRefPubMed
    9. ↵
      2. Schneider DT ,
      3. Schuster AE ,
      4. Fritsch MK , et al
      . Multipoint imprinting analysis indicates a common precursor cell for gonadal and nongonadal pediatric germ cell tumors. Cancer Res 2001;61:7268–76.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      2. Oosterhuis JW ,
      3. Stoop H ,
      4. Honecker F , et al
      . Why human extragonadal germ cell tumours occur in the midline of the body: old concepts, new perspectives. Int J Androl 2007;30:256–64. discussion 63-4.doi:10.1111/j.1365-2605.2007.00793.x
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      2. Calaminus G ,
      3. Kortmann R ,
      4. Worch J , et al
      . Siop CNS GCT 96: final report of outcome of a prospective, multinational nonrandomized trial for children and adults with intracranial germinoma, comparing craniospinal irradiation alone with chemotherapy followed by focal primary site irradiation for patients with localized disease. Neuro Oncol 2013;15:788–96.doi:10.1093/neuonc/not019
      OpenUrlCrossRefPubMed
    12. ↵
      2. Calaminus G ,
      3. Frappaz D ,
      4. Kortmann RD , et al
      . Outcome of patients with intracranial non-germinomatous germ cell tumors-lessons from the SIOP-CNS-GCT-96 trial. Neuro Oncol 2017;19:1661–72.doi:10.1093/neuonc/nox122
      OpenUrl
    13. ↵
      2. Bowzyk Al-Naeeb A ,
      3. Murray M ,
      4. Horan G , et al
      . Current management of intracranial germ cell tumours. Clin Oncol 2018;30:204–14.doi:10.1016/j.clon.2018.01.009
      OpenUrl
    14. ↵
      2. Goldman S ,
      3. Bouffet E ,
      4. Fisher PG , et al
      . Phase II trial assessing the ability of neoadjuvant chemotherapy with or without second-look surgery to eliminate measurable disease for Nongerminomatous germ cell tumors: a children's Oncology Group study. J Clin Oncol 2015;33:2464–71.doi:10.1200/JCO.2014.59.5132
      OpenUrlAbstract/FREE Full Text
    15. ↵
      2. Keene D ,
      3. Johnston D ,
      4. Strother D , et al
      . Epidemiological survey of central nervous system germ cell tumors in Canadian children. J Neurooncol 2007;82:289–95.doi:10.1007/s11060-006-9282-2
      OpenUrlPubMed
    16. ↵
      2. Matsutani M ,
      3. Sano K ,
      4. Takakura K , et al
      . Primary intracranial germ cell tumors: a clinical analysis of 153 histologically verified cases. J Neurosurg 1997;86:446–55.doi:10.3171/jns.1997.86.3.0446
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      2. Jorsal T ,
      3. Rørth M
      . Intracranial germ cell tumours. A review with special reference to endocrine manifestations. Acta Oncol 2012;51:3–9.doi:10.3109/0284186X.2011.586000
      OpenUrlPubMed
    18. ↵
      2. Janmohamed S ,
      3. Grossman AB ,
      4. Metcalfe K , et al
      . Suprasellar germ cell tumours: specific problems and the evolution of optimal management with a combined chemoradiotherapy regimen. Clin Endocrinol 2002;57:487–500.doi:10.1046/j.1365-2265.2002.01620.x
      OpenUrlCrossRefPubMed
    19. ↵
      2. Crawford JR ,
      3. Santi MR ,
      4. Vezina G , et al
      . Cns germ cell tumor (CNSGCT) of childhood: presentation and delayed diagnosis. Neurology 2007;68:1668–73.doi:10.1212/01.wnl.0000261908.36803.ac
      OpenUrlCrossRef
    20. ↵
      2. Klitbo DM ,
      3. Nielsen R ,
      4. Illum NO , et al
      . Symptoms and time to diagnosis in children with brain tumours. Dan Med Bull 2011;58:A4285.
    21. ↵
      2. Sethi RV ,
      3. Marino R ,
      4. Niemierko A , et al
      . Delayed diagnosis in children with intracranial germ cell tumors. J Pediatr 2013;163:1448–53.doi:10.1016/j.jpeds.2013.06.024
      OpenUrl
    22. ↵
      2. Brasme J-F ,
      3. Chalumeau M ,
      4. Doz F , et al
      . Interval between onset of symptoms and diagnosis of medulloblastoma in children: distribution and determinants in a population-based study. Eur J Pediatr 2012;171:25–32.doi:10.1007/s00431-011-1480-z
      OpenUrlCrossRefPubMed
    23. ↵
      2. Shay V ,
      3. Fattal-Valevski A ,
      4. Beni-Adani L , et al
      . Diagnostic delay of pediatric brain tumors in Israel: a retrospective risk factor analysis. Childs Nerv Syst 2012;28:93–100.doi:10.1007/s00381-011-1564-0
      OpenUrlPubMed
    24. ↵
      2. Phi JH ,
      3. Kim S-K ,
      4. Lee YA , et al
      . Latency of intracranial germ cell tumors and diagnosis delay. Childs Nerv Syst 2013;29:1871–81.doi:10.1007/s00381-013-2164-y
      OpenUrl
    25. ↵
      2. Zhang Y ,
      3. Deng K ,
      4. Zhu H , et al
      . Delays in diagnosis of pediatric histologically confirmed sellar germ cell tumors in China: a retrospective risk factor analysis. World Neurosurg 2019;122:e472–9.doi:10.1016/j.wneu.2018.10.082
      OpenUrl
    26. ↵
      2. Walker D ,
      3. Wilne S ,
      4. Grundy R , et al
      . A new clinical guideline from the Royal College of Paediatrics and Child Health with a national awareness campaign accelerates brain tumor diagnosis in UK children--"HeadSmart: Be Brain Tumour Aware". Neuro Oncol 2016;18:445–54.doi:10.1093/neuonc/nov187
      OpenUrlCrossRefPubMed
    27. ↵
      2. Chapman CA ,
      3. Waber DP ,
      4. Bernstein JH , et al
      . Neurobehavioral and neurologic outcome in long-term survivors of posterior fossa brain tumors: role of age and perioperative factors. J Child Neurol 1995;10:209–12.doi:10.1177/088307389501000308
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      2. Wilne S ,
      3. Collier J ,
      4. Kennedy C , et al
      . Progression from first symptom to diagnosis in childhood brain tumours. Eur J Pediatr 2012;171:87–93.doi:10.1007/s00431-011-1485-7
      OpenUrlCrossRefPubMed
    29. ↵
      2. Walker D ,
      3. Hamilton W ,
      4. Walter FM , et al
      . Strategies to accelerate diagnosis of primary brain tumors at the primary-secondary care interface in children and adults. CNS Oncol 2013;2:447–62.doi:10.2217/cns.13.36
      OpenUrl

    Footnotes

    • JH and MJM are joint first authors.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Patient consent for publication Not required.

    • Ethics approval Ethical approvals were obtained according to institutional and national standards for retrospective service evaluations, requirements which varied between participating countries.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data availability statement Data are available on reasonable request.

    Linked Articles