Radiotherapy for diffuse brainstem glioma in children and young adults

  • Review
  • Intervention

Authors


Abstract

Background

Diffuse brainstem glioma is a devastating disease with very poor prognosis. The most commonly used radiological treatment is conventional fractionated radiation. So far, there is no meta-analysis or systematic review available that assesses the benefits or harms of radiation in people with diffuse brainstem glioma.

Objectives

To assess the effects of conventional fractionated radiotherapy (with or without chemotherapy) versus other therapies (including different radiotherapy techniques) for newly diagnosed diffuse brainstem gliomas in children and young adults aged 0 to 21 years.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE/PubMed, and EMBASE to 19 August 2015. We scanned conference proceedings from the International Society for Paediatric Oncology (SIOP), International Symposium on Paediatric Neuro-Oncology (ISPNO), Society of Neuro-Oncology (SNO), and European Association of Neuro-Oncology (EANO) from 1 January 2010 to 19 August 2015. We searched trial registers including the International Standard Randomised Controlled Trial Number (ISRCTN) Register, the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), and the register of the National Institutes of Health to 19 August 2015. We imposed no language restrictions.

Selection criteria

All randomised controlled trials (RCTs), quasi-randomised trials (QRCTs), or controlled clinical trials (CCTs) that compared conventional fractionated radiotherapy (with or without chemotherapy) versus other therapies (including different radiotherapy techniques) for newly diagnosed diffuse brainstem glioma in children and young adults aged 0 to 21 years.

Data collection and analysis

Two review authors independently screened studies for inclusion, extracted data, assessed the risk of bias in each eligible trial, and conducted GRADE assessment of included studies. We resolved disagreements through discussion. We performed analyses according to the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions.

Main results

We identified two RCTs that fulfilled our inclusion criteria. The two trials tested different comparisons.

One multi-institutional RCT included 130 participants and compared hyperfractionated radiotherapy (six-week course with twice a day treatment of 117 cGy per fraction to a total dose of 7020 cGy) with conventional radiotherapy (six-week course with once a day treatment of 180 cGy per fraction to a total dose of 5400 cGy). The median time overall survival (OS) was 8.5 months in the conventional group and 8.0 months in the hyperfractionated group. We detected no clear evidence of effect on OS or event-free survival (EFS) in participants receiving hyperfractionated radiotherapy compared with conventional radiotherapy (OS: hazard ratio (HR) 1.07, 95% confidence interval (CI) 0.75 to 1.53; EFS: HR 1.26, 95% CI 0.83 to 1.90). Radiological response (risk ratio (RR) 0.94, 95% CI 0.54 to 1.63) and various types of toxicities were similar in the two groups. There was no information on other outcomes. According to the GRADE approach, we judged the quality of evidence to be low (i.e. further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate) for OS and EFS, and very low (i.e. we are very uncertain about the estimate) for radiological response and toxicities.

The second RCT included 71 participants and compared hypofractionated radiotherapy (39 Gy in 13 fractions over 2.6 weeks, 3 Gy per fraction) with conventional radiotherapy (54 Gy in 30 fractions over six weeks, 1.8 Gy per fraction). This trial reported a median OS of 7.8 months for the hypofractionated group and 9.5 months for the conventional group. It reported a progression-free survival (PFS) of 6.3 months for the hypofractionated group and 7.3 months for the conventional group. We found no clear evidence of effect on OS (HR 1.03, 95% CI 0.53 to 2.01) or PFS (HR 1.19, 95% CI 0.63 to 2.22) in participants receiving hypofractionated radiotherapy when compared with participants receiving conventional radiotherapy. The mainly observed adverse effect was local erythema and dry desquamation especially behind the auricles. There were some other toxicities, but there was no statistically significant difference between treatment groups. There was no information on other outcomes. We judged the quality of evidence to be moderate (i.e. further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate) for OS, and low for PFS and toxicities. It should be mentioned that the sample size in this RCT was small, which could lead to insufficient statistical power for a clinically relevant outcome.

Authors' conclusions

We could make no definitive conclusions from this review based on the currently available evidence. Further research is needed to establish the role of radiotherapy in the management of newly diagnosed diffuse brainstem glioma in children and young adults. Future RCTs should be conducted with adequate power and all relevant outcomes should be taken into consideration. Moreover, international multicentre collaboration is encouraged. Considering the potential advantage of hypofractionated radiotherapy to decrease the treatment burden and increase the quality of remaining life, we suggest that more attention should be paid to hypofractionated radiotherapy.

Resumen

Radioterapia para el glioma difuso del tronco encefálico en niños y adultos jóvenes

Antecedentes

El glioma difuso del tronco encefálico es una enfermedad devastadora y de muy mal pronóstico. El tratamiento radiológico utilizado con mayor frecuencia es la radioterapia fraccionada convencional. Hasta el presente no hay metanálisis o revisión sistemática disponibles que evalúen los efectos beneficiosos o perjudiciales de la radiación en los pacientes con glioma difuso del tronco encefálico.

Objetivos

Evaluar los efectos de la radioterapia fraccionada convencional (con o sin quimioterapia) versus otras terapias (que incluyen diferentes técnicas de radioterapia) para los gliomas difusos del tronco encefálico recién diagnosticados en niños y adultos jóvenes hasta los 21 años de edad.

Métodos de búsqueda

Se hicieron búsquedas en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL), MEDLINE/PubMed, y en EMBASE hasta el 19 agosto 2015. Se revisaron las actas de congresos de la International Society for Paediatric Oncology (SIOP), International Symposium on Paediatric Neuro-Oncology (ISPNO), Society of Neuro-Oncology (SNO) y de la European Association of Neuro-Oncology (EANO) desde el 1 enero 2010 hasta el 19 agosto 2015. Se buscaron registros de ensayos incluyendo el International Standard Randomised Controlled Trial Number (ISRCTN) Register, la World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), y el register of the National Institutes of Health hasta el 19 agosto 2015. No se impuso ninguna restricción de idioma.

Criterios de selección

Todos los ensayos controlados aleatorios (ECA), los ensayos cuasialeatorios o ensayos clínicos controlados (ECC) que compararon la radioterapia fraccionada convencional (con o sin quimioterapia) versus otras terapias (que incluyen diferentes técnicas de radioterapia) para el glioma difuso del tronco encefálico recién diagnosticados en niños y adultos jóvenes hasta los 21 años de edad.

Obtención y análisis de los datos

Dos autores de la revisión de forma independiente revisaron los estudios para inclusión, extrajeron los datos, evaluaron el riesgo de sesgo de cada ensayo elegible y realizaron la evaluación GRADE de los estudios incluidos. Los desacuerdos se resolvieron mediante discusión. Los análisis se realizaron según las guías del Manual Cochrane para Revisiones Sistemáticas de Intervenciones (Cochrane Handbook for Systematic Reviews of Interventions).

Resultados principales

Se identificaron dos ECA que cumplieron los criterios de inclusión. Los dos ensayos evaluaron diferentes comparaciones.

Un ECA multiinstitucional incluyó a 130 participantes y comparó radioterapia hiperfraccionada (ciclo de seis semanas con un tratamiento dos veces al día de 117 cGy por fracción a una dosis total de 7020 cGy) con radioterapia convencional (ciclo de seis semanas con un tratamiento una vez al día de 180 cGy por fracción a una dosis total de 5400 cGy). La mediana del tiempo de supervivencia (SG) general fue 8,5 meses en el grupo convencional y 8,0 meses en el grupo de radioterapia hiperfraccionada. No se detectaron pruebas claras de efectos sobre la SG o la supervivencia sin eventos (SSE) en los participantes que recibieron radioterapia hiperfraccionada en comparación con la radioterapia convencional (SG: cociente de riesgos instantáneos [CRI] 1,07; intervalo de confianza [IC] del 95%: 0,75 a 1,53; SSE: CRI 1,26; IC del 95%: 0,83 a 1,90). La respuesta radiológica (cociente de riesgos [CR] 0,94; IC del 95%: 0,54 a 1,63) y los diversos tipos de efectos tóxicos fueron similares en los dos grupos. No hubo información sobre otros resultados. Según el enfoque GRADE, la calidad de las pruebas se consideró baja (es decir, es muy probable que los estudios de investigación adicionales tengan una marcada repercusión sobre la confianza en la estimación del efecto y es probable que cambien la estimación) para la OS y la SSE, y muy baja (es decir, existe muy poca seguridad acerca de la estimación) para la respuesta radiológica y los efectos tóxicos.

El segundo ECA incluyó a 71 participantes y comparó radioterapia hipofraccionada (39 Gy en 13 fracciones durante 2,6 semanas, 3 Gy por fracción) con radioterapia convencional (54 Gy en 30 fracciones durante seis semanas, 1,8 Gy por fracción). Este ensayo informó una mediana de la SG de 7,8 meses para el grupo de radioterapia hipofraccionada y 9,5 meses para el grupo convencional. También informó una supervivencia sin progresión (SSP) de 6,3 meses para el grupo de radioterapia hipofraccionada y 7,3 meses para el grupo convencional. No se encontraron pruebas claras de efectos en la SG (CRI 1,03; IC del 95%: 0,53 a 2,01) o la SSP (CRI 1,19; IC del 95%: 0,63 a 2,22) en los participantes que recibieron radioterapia hipofraccionada en comparación con los participantes que recibieron radioterapia convencional. El efecto adverso que se observó principalmente fue el eritema local y la descamación seca especialmente detrás de los pabellones auriculares. Hubo algunos otros efectos tóxicos pero no hubo diferencias estadísticamente significativas entre los grupos de tratamiento. No hubo información sobre otros resultados. La calidad de las pruebas se consideró moderada (es decir, es probable que los estudios de investigación adicionales tengan una marcada repercusión sobre la confianza la estimación del efecto y puede cambiar la estimación) para la SG, y baja para la SSP y los efectos tóxicos. Se debe mencionar que el tamaño de la muestra de este ECA fue pequeño, lo que podría dar lugar a un poder estadístico insuficiente para un resultado clínicamente relevante.

Conclusiones de los autores

No es posible establecer conclusiones definitivas a partir de esta revisión según las pruebas actualmente disponibles. Se necesitan estudios de investigación adicionales para establecer la función de la radioterapia en el tratamiento del glioma difuso del tronco encefálico recién diagnosticado en niños y adultos jóvenes. Los ECA futuros que se realicen deben tener un poder estadístico suficiente y se deben considerar todos los resultados relevantes. Además, se recomienda la colaboración multicéntrica internacional. Al considerar la posible ventaja de la radioterapia hipofraccionada para reducir la carga de tratamiento y aumentar la calidad del tiempo de vida restante, se indica prestar más atención a la radioterapia hipofraccionada.

Plain language summary

Radiotherapy for diffuse brainstem glioma in children and young adults

Review question

To assess the effects of conventional radiotherapy (with or without chemotherapy) versus other therapies (including different radiotherapy techniques) for newly diagnosed diffuse brainstem gliomas in children and young adults aged 0 to 21 years.

Background

Diffuse brainstem glioma typically occurs in the pons (part of the brainstem) and expands and infiltrates at least 50% of the pons, with a characteristic appearance on magnetic resonance imaging (MRI). The prognosis is very poor, with a median overall survival (OS; time from cancer diagnosis, or treatment, to death from any cause) ranges from 8 to 11 months. So far, there is no analysis or review available that assessed the benefits or harms of radiation for newly diagnosed diffuse brainstem glioma in children and young adults aged 0 to 21 years.

Study characteristics

Through comprehensive search and screening of medical databases, we found two clinical studies that tested different treatments. One study, with 130 participants included, compared hyperfractionated radiotherapy (six-week course with treatment twice a day ) with conventional radiotherapy (six-week course with treatment once a day ). The second study, with 71 participants included, compared hypofractionated radiotherapy (three-week course with treatment once a day ) with conventional radiotherapy.

Key results

For the comparison of hyperfractionated radiotherapy and conventional radiotherapy, there was no clear evidence of effect on OS, event-free survival (EFS; time from diagnosis, study entry, or treatment to disease progression, disease relapse, a second tumour, or death), radiological response (a reduction in tumour size of more than 50%), and toxicities (damage to the body due to radiotherapy).

For the comparison of hypofractionated radiotherapy and conventional radiotherapy, there was no clear evidence of effect on OS, progression-free survival (PFS; time from diagnosis, study entry, or treatment to disease progression), and side effects.

Quality of the evidence

For the hyperfractionated radiotherapy, when compared with conventional therapy, the quality of evidence was low for OS and EFS, and very low for radiological response and toxicities.

For the hypofractionated radiotherapy, when compared with conventional therapy, the quality of evidence was moderate for OS, and low for PFS and toxicities.

Laički sažetak

Radioterapija za liječenje difuznog glioma moždanog debla u djece i adolescenata

Istraživačko pitanje

Cilj ovog Cochrane sustavnog pregleda literature bio je procijeniti učinak standardne radioterapije (s ili bez kemoterapije) naspram drugih terapija (uključujući i druge tehnike radioterapije) za novo-dijagnosticirane difuzne gliome moždanog debla u djece i adolescenata od rođenja do 21. godine života.

Dosadašnje spoznaje

Difuzni gliomi moždanog debla u pravilu se pojavljuju u ponsu (dio moždanog debla) te se šire i infiltriraju najmanje 50% ponsa s karakterističnim prikazom na magnetskoj rezonanciji (MRI). Prognoza je vrlo loša. Srednje vrijeme ukupnog preživljavanja (eng. overall survival, OS; vrijeme od postavljanja dijagnoze pa do vremena smrti iz bilo kojeg razloga) je 8 do 11 mjeseci. Do sada nije bilo analiza ili sustavnih preglednih radova u kojima se procijenio učinak ili šteta zračenja novo-dijagnosticiranih difuznih glioma moždanog debla u djece i adolescenata između rođenja i 21. godine života.

Značajke istraživanja

Temeljitim pretraživanjem medicinskih baza podataka su pronađena dva istraživanja koja su ispitala različite terapije. Jedna studija, sa 130 sudionika, je usporedila hiperfrakcioniranu radioterapiju (6 tjedna terapije 2 puta na dan) s konvencionalnom radioterapijom (6 tjedna terapije jednom na dan). Druga studija, sa 71 sudionikom, je usporedila hipofrakcioniranu radioterapiju (3 tjedna terapije jednom na dan) s konvencionalnom radioterapijom.

Rezultati

Usporedbom hiperfrakcionirane i konvencionalne radioterapije nije pronađen utjecaj na ukupnu stopu preživljavanja (eng. overall survival rate, OS), preživljavanje bez znakova bolesti (eng. event-free survival, EFS; vrijeme od dijagnoze, ulaska u studiji ili početka liječenja do progresije bolesti, relapsa bolesti, sekundarnog tumora ili smrti), radiološkog odgovora (smanjenje tumora više od 50%) te toksičnosti (oštećenje organizma zbog radioterapije).

Usporedbom hipofrakcionirane i konvencionalne radioterapije nije pronađen utjecaj na: OS, preživljavanje bez napredovanja bolesti (eng. progression-free survival, PFS; vrijeme od dijagnoze, ulaska u studiji ili početka liječenja do napredovanja bolesti) te nuspojave.

Kvaliteta dokaza

Za hiperfrakcioniranu radioterapiju uspoređenu s konvencionalnom radioterapijom, kvaliteta dokaza je bila niska za OS i EFS te vrlo niska za radiološki odgovor i toksičnost.

Za hipofrakcioniranu radioterapiju uspoređenu s konvencionalnom radioterapijom, kvaliteta dokaza je bila umjerena za OS te niska za PFS i nuspojave.

Bilješke prijevoda

Hrvatski Cochrane
Preveo: Igor Vlatković
Ovaj sažetak preveden je u okviru volonterskog projekta prevođenja Cochrane sažetaka. Uključite se u projekt i pomozite nam u prevođenju brojnih preostalih Cochrane sažetaka koji su još uvijek dostupni samo na engleskom jeziku. Kontakt: cochrane_croatia@mefst.hr

Resumen en términos sencillos

Radioterapia para el glioma difuso del tronco encefálico en niños y adultos jóvenes

Pregunta de la revisión

Evaluar los efectos de la radioterapia convencional (con o sin quimioterapia) versus otras terapias (que incluyen diferentes técnicas de radioterapia) para los gliomas difusos del tronco encefálico recién diagnosticados en niños y adultos jóvenes hasta los 21 años de edad.

Antecedentes

El glioma difuso del tronco encefálico aparece habitualmente en el puente (parte del tronco encefálico) y se expande e infiltra al menos el 50% del puente, con una apariencia característica en la imagenología de resonancia magnética (IRM). Tiene muy mal pronóstico, con una mediana de la supervivencia general (SG; tiempo desde el diagnóstico del cáncer o el tratamiento, hasta la muerte por cualquier causa) que varía de ocho a 11 meses. Hasta el presente no hay análisis o revisión disponibles que se evalúen los efectos beneficiosos o perjudiciales de la radiación para el glioma difuso del tronco encefálico recién diagnosticado en niños y adultos jóvenes hasta los 21 años de edad.

Características de los estudios

Mediante la búsqueda exhaustiva y la revisión de las bases de datos médicas se encontraron dos estudios clínicos que evaluaron diferentes tratamientos. Un estudio, con 130 participantes incluidos, comparó la radioterapia hiperfraccionada (ciclo de seis semanas con tratamiento dos veces al día) con la radioterapia convencional (ciclo de seis semanas con tratamiento una vez al día). El segundo estudio, con 71 participantes incluidos, comparó la radioterapia hipofraccionada (ciclo de tres semanas con tratamiento una vez al día) con la radioterapia convencional.

Resultados clave

Para la comparación de radioterapia hiperfraccionada y radioterapia convencional, no hubo pruebas claras de efectos sobre la SG, la supervivencia sin eventos (SSE; tiempo desde el diagnóstico, el ingreso al estudio o el tratamiento hasta la progresión de la enfermedad, la recidiva de la enfermedad, un segundo tumor, o la muerte), la respuesta radiológica (una reducción del tamaño del tumor de más del 50%) y los efectos tóxicos (daño al cuerpo debido a la radioterapia).

Para la comparación de la radioterapia hipofraccionada y la radioterapia convencional, no hubo pruebas claras de efectos sobre la SG, la supervivencia sin progresión (SSP; tiempo desde el diagnóstico, el ingreso al estudios o el tratamiento hasta la progresión de la enfermedad) y los efectos secundarios.

Calidad de la evidencia

Para la radioterapia hiperfraccionada comparada con el tratamiento convencional la calidad de las pruebas fue baja para la SG y la SSE y muy baja para la respuesta radiológica y los efectos tóxicos.

Para la radioterapia hipofraccionada comparada con el tratamiento convencional la calidad de las pruebas fue moderada para la SG y baja para la SSP y los efectos tóxicos.

Notas de traducción

La traducción y edición de las revisiones Cochrane han sido realizadas bajo la responsabilidad del Centro Cochrane Iberoamericano, gracias a la suscripción efectuada por el Ministerio de Sanidad, Servicios Sociales e Igualdad del Gobierno español. Si detecta algún problema con la traducción, por favor, contacte con Infoglobal Suport, cochrane@infoglobal-suport.com.

Summary of findings(Explanation)

Summary of findings for the main comparison. Hyperfractionated radiotherapy compared with conventional radiotherapy for diffuse brainstem glioma in children and young adults
  1. 1 The assumed risk is based on the prevalence in the control group of the included study.

    2 Small sample size with a total number of events fewer than 300 (the threshold rule-of-thumb value stated in the GRADEpro software).

    3 Presence of selection bias and other bias was unclear.

    4 Presence of selection bias, detection bias, and other bias was unclear.

    5 Presence of selection bias, detection bias, and other bias was unclear; high risk of attrition bias.

Hyperfractionated radiotherapy compared with conventional radiotherapy for diffuse brainstem glioma in children and young adults

Patient or population: children and young adults with diffuse brainstem glioma

Settings: hospital setting in USA and Canada

Intervention: hyperfractionated radiotherapy

Comparison: conventional radiotherapy

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Conventional radiotherapyHyperfractionated radiotherapy
Overall survivalSee commentsSee commentsHR 1.07 (0.75 to 1.53)130 (1 study)⊕⊕⊝⊝
low2,3

Time-to-event data; no assumed risk could be calculated

Participants were enrolled between June 1992 and March 1996. All participants were followed until death or October 1997 except 5 participants who were lost during follow-up

Progression-free survivalSee commentSee commentNot estimable-See commentNo adequate information on this outcome was provided
Event-free survivalSee commentSee commentHR 1.26 (0.83 to 1.90)130 (1 study)⊕⊕⊝⊝
low2,4
Time-to-event data; no assumed risk could be calculated
Quality of lifeSee commentSee commentNot estimable-See commentNo information on this outcome was provided
Neurological responseSee commentSee commentNot estimable-See commentNo adequate information on this outcome was provided

Radiological response

defined as the number of participants with a reduction in tumour size > 50%

Follow-up: 4 or 8 weeks
(median 4 weeks) post treatment

333 per 10001313 per 1000
(180 to 543)
RR 0.94 (0.54 to 1.63)108
(1 study)
⊕⊝⊝⊝
very low2,5
-
ToxicitiesSee commentSee commentSee comments

112 or 113

(1 study)

⊕⊝⊝⊝
very low2,5

Various types of toxicities were evaluated (see text)

No significant differences in examined toxicities between the 2 groups

For fungal stomatitis, 112 participants were available while for the other toxicities, 113 participants were available

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; HR: hazard ratio; RR: risk ratio.
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Summary of findings 2 Hypofractionated radiotherapy compared with conventional radiotherapy for diffuse brainstem glioma in children and young adults

Summary of findings 2. Hypofractionated radiotherapy compared with conventional radiotherapy for diffuse brainstem glioma in children and young adults
  1. 1 Small sample size with a total number of events fewer than 300 (the threshold rule-of-thumb value stated in the GRADEpro software).

    2 High risk of detection bias.

Hypofractionated radiotherapy compared with conventional radiotherapy for diffuse brainstem glioma in children and young adults

Patient or population: children and young adults with diffuse brainstem glioma

Settings: Children's Cancer Hospital, Egypt

Intervention: hypofractionated radiotherapy

Comparison: conventional radiotherapy

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Conventional radiotherapyHypofractionated radiotherapy

Overall survival

Follow up: mean 9 months

See commentSee commentHR 1.03 (0.53 to 2.01)71
(1 study)
⊕⊕⊕⊝
moderate1
Time-to-event data; no assumed risk could be calculated
Progression-free survivalSee commentSee commentHR 1.19 (0.63 to 2.22)

71

(1 study)

⊕⊕⊝⊝
low1,2
Time-to-event data; no assumed risk could be calculated
Event-free survivalSee commentSee commentNot estimable-See commentNo information on this outcome was provided
Quality of lifeSee commentSee commentNot estimable-See commentNo information on this outcome was provided
Neurological responseSee commentSee commentNot estimable-See commentNo information on this outcome was provided
Radiological responseSee commentSee commentNot estimable-See commentNo information on this outcome was provided
ToxicitiesSee commentSee commentSee comments

71

(1 study)

⊕⊕⊝⊝
low1,2

Various types of toxicities were evaluated (see text)

No significant differences in examined toxicities between the 2 groups

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; HR: hazard ratio.
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Gliomas located in the brainstem account for 10% to 20% of all brain tumours in children (Stiller 1994; Jallo 2004). Many classification schemes have been devised for brainstem tumours (Jallo 2003). Currently, radiographic findings and clinical symptoms are used to categorise brainstem gliomas into four groups, namely: diffuse (widely distributed), focal, exophytic (growing outwards from the epithelium), and cervicomedullary (within the organ) (Roonprapunt 2002). Focal tumours are defined as demarcated lesions less than 2 cm in diameter without associated oedema; dorsally exophytic brainstem gliomas are a group of tumours arising from the subependymal glial tissue; and cervicomedullary brainstem gliomas are similar to intramedullary spinal cord gliomas. Focal and dorsally exophytic brainstem gliomas comprise about 15% to 20% of brainstem tumours; they are low-grade astrocytomas (i.e. arise from embryonic tissue that forms the nervous system) that have characteristic clinical features, growth patterns of low-grade glial tumours, and generally follow an indolent course (i.e. cause no trouble or pain). Most of the remaining 80% of tumours are diffuse brainstem gliomas (DBG) (Maria 1993); usually a fibrillary (World Health Organization (WHO) grade 2) or malignant (WHO grade 3 or 4) astrocytoma (Freeman 1998). DBGs are the most aggressive subgroup of brainstem tumours, and typically occur in the pons (part of the brainstem) with a characteristic appearance on magnetic resonance imaging (MRI) of a T1 hypointensity and T2 hyperintensity diffuse tumour, which expands and infiltrates at least 50% of the pons. The encasement of the basilar artery is one of the most common findings in typical diffuse intrinsic pontine gliomas (DIPG) (Hargrave 2008). The typical clinical presentation includes a neurological signs triad: multiple cranial nerve deficits (cranial neuropathy), long tract signs (hyper-reflexia, a Babinski sign, and weakness), and ataxia (problems with muscular co-ordination) (Donaldson 2006; Khatua 2011).

Diffuse pontine (i.e. relating to the pons) gliomas rarely affect children younger than three years old. Broniscer et al. suggested that DBG in children below the age of three years may be distinct biologically from similar tumours in older age groups (Broniscer 2008). The difference between children aged below three years and older children with diffuse pontine glioma is unclear. In addition, most contemporary clinical trials for children with diffuse pontine glioma are restricted to children aged three years and older.

Despite collaborative efforts to improve treatments, DBGs carry the worst prognosis of all brainstem gliomas. Survival has remained dismal since the mid-1990s with the median overall survival (OS) ranging from 8 to 14 months (Kaplan 1996; Mandell 1999; Vanan 2015). Various interventions have been investigated including radiation, surgical resection, chemotherapy, radiation sensitisers, and biological agents (Epstein 1988; Pierre-Kahn 1993; Hargrave 2006; Jansen 2012). However, most of them showed disappointing efficacy. Radiotherapy is the only treatment that has consistently shown clinical and radiographic improvement in people with DBG (Khatua 2011).

Without radiation, median survival is approximately 20 weeks (Langmoen 1991). The most commonly used radiological treatment is conventional fractionated radiation, with local field radiation for a total dose of 54 to 60 Gy delivered in 30 fractions (1.8 to 2 Gy per fraction per day) over a period of six weeks (Albright 2004; Donaldson 2006). Preclinical studies have shown that initial radiation effects are mediated through direct vascular effects on the endothelium (i.e. thin layer of cells that lines the interior surface of blood vessels) and through cytokine (i.e. small cell-signalling protein molecules that are secreted by cells) activation, with attendant oedema and disruption of the blood-brain barrier. Doses delivered within the therapeutic range result in progressive, diffuse changes in white matter as part of a continuous, dynamic series of events, including direct and indirect effects on glial (brain tissue that does not conduct electrical impulses) and neuronal elements, as well as on small vessels (Tofilon 2000; Coderre 2006). Studies reported no toxicity in children with brainstem glioma (treated with opposed lateral fields that encompassed the majority of the brainstem) to doses of 54 to 60 Gy at 2 Gy per fraction, 75.6 Gy at 1.26 Gy twice daily, or 78 Gy at 1 Gy twice daily (Freeman 1993; Packer 1994; Mayo 2010). Radiotherapy induces neurological improvement, allows reduction or discontinuation of steroids, and is associated with the radiological response.

Researchers attempted to seek more effective radiotherapeutic approaches given the evidence of transient response to radiation, the tumours' tendencies to progress locally (Donahue 1998), and the radiation dose-response relationship observed for DBGs (Hibi 1992). These regimens were designed to increase the dose-intensity by delivering a higher total dose in the same time (hyperfractionated regimen) (Freeman 1991; Freeman 1993; Packer 1994; Prados 1995), or a higher daily dose given over a shorter period of time (hypofractionated regimen) (Janssens 2009; Negretti 2011). In hyperfractionated regimens, two fractions are delivered each day, with a reduced dose per fraction equal to 1.1 to 1.2 Gy. The reduction of the dose per fraction might reduce the risk of late toxicity, despite the increased total dose. In 1984, the Pediatric Oncology Group began a phase I/II study to test the efficacy and toxicity of sequentially escalated doses of hyperfractionated (twice daily) radiation in children with brainstem gliomas (Freeman 1988). The final results of this study revealed that there appeared to be a trend towards increased survival and time to progression with an increase in the dose level from 6600 to 7020 cGy (Freeman 1991). The next dose escalation level of 7560 cGy did not result in an improvement in survival over that observed with the second dose level (Freeman 1993); this observation was corroborated by other investigators who used hyperfractionated radiotherapy to 78 Gy (Packer 1994; Prados 1995). In hypofractionated regimens a total dose of 45 Gy is delivered in daily fractions of 3 to 5.5 Gy over three weeks (Janssens 2009; Negretti 2011). The advantages of hypofractionated radiotherapy include the shorter overall treatment time with a reduction in the associated economic burden, reduced supportive care, limited toxicity, and the same efficacy as conventionally fractionated radiotherapy.

So far, there is no meta-analysis or systematic review available that assesses the benefits or harms of radiation in people with DBG. This review aimed to evaluate the existing evidence on radiotherapy in treating children and young adults aged 0 to 21 years with newly diagnosed DBG.

Objectives

To assess the effects of conventional fractionated radiotherapy (with or without chemotherapy) versus other therapies (including different radiotherapy techniques) for newly diagnosed diffuse brainstem gliomas in children and young adults aged 0 to 21 years.

Methods

Criteria for considering studies for this review

Types of studies

We included all randomised controlled trials (RCTs) and quasi-randomised trials (QRCTs) or controlled clinical trials (CCTs), as defined by the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), that compared conventional fractionated radiotherapy (with or without chemotherapy) versus other therapies (including different radiotherapy techniques). We did not include uncontrolled observational trials.

Types of participants

Participants met the following criteria:

  • aged of 0 to 21 years at treatment;

  • diagnosis of DBG made as described in the original studies;

  • DBG was newly diagnosed and previously untreated with radiotherapy.

Types of interventions

Conventional fractionated radiotherapy (with or without chemotherapy) versus other therapies (including different radiotherapy techniques). Trials differed only in respect of radiotherapy modalities. The definition of conventional radiotherapy was external-beam radiotherapy and local field radiotherapy.

Types of outcome measures

Primary outcomes
  • Overall survival (OS), defined as time from tumour diagnosis, or treatment, to death from any cause.

Secondary outcomes
  • Progression-free survival (PFS): progression defined clinically or radiologically (or both) as an increase of 25% or more in the size of the tumour on imaging, or as the appearance of a new lesion (or both). PFS defined as time from diagnosis, study entry, or treatment to progression.

  • Event-free survival (EFS): defined as time from diagnosis, study entry, or treatment to disease progression, disease relapse, a second malignant neoplasm, or death from any cause.

  • Quality of life (QoL) (as defined by the trial authors).

  • Neurological response: defined as the number of participants with improved neurological function.

  • Radiological response: defined as the number of participants with a reduction in tumour size of more than 50% (Gnekow 1995).

  • Toxicity (as defined by the trial authors).

Search methods for identification of studies

Electronic searches

We searched the following electronic databases:

  • the Cochrane Central Register of Controlled Trials (CENTRAL) (2015, Issue 7);

  • MEDLINE/PubMed (1945 to 19 August 2015);

  • EMBASE/Ovid (1980 to 19 August 2015).

The appendices show the search strategies for the different electronic databases (using a combination of controlled vocabulary and text words) (Appendix 1; Appendix 2; Appendix 3).

Searching other resources

We searched the reference lists of relevant articles and review articles to identify potentially eligible citations.

We scanned the following conference proceedings electronically, from 1 January 2010 to 19 August 2015 (Appendix 4):

  • International Society for Paediatric Oncology (SIOP);

  • International Symposium on Paediatric Neuro-Oncology (ISPNO);

  • Society of Neuro-Oncology (SNO);

  • European Association of Neuro-Oncology (EANO).

We searched the following trials registers to 19 August 2015 (Appendix 5):

  • International Standard Randomised Controlled Trial Number (ISRCTN) Register (www.isrctn.com/);

  • the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP, apps.who.int/trialsearch);

  • the register of the National Institutes of Health (NIH) (clinicaltrials.gov/) for ongoing trials.

We did not impose any language restrictions and we will update the searches every two years.

Data collection and analysis

Selection of studies

Two review authors independently screened the titles and abstracts of studies identified through the searches, and selected trials that possibly met the inclusion criteria. We retrieved full-text articles for further assessment. We used discussion and consultation with a third review author to resolve any disagreements.

Data extraction and management

For included studies, two review authors extracted the following information independently, using a standard form.

General information
  • Title.

  • Authors.

  • Whether trial was published or unpublished.

  • Year of publication.

  • Language of publication.

  • Duplicate publications.

  • Study design.

  • Country.

  • Reference/source.

  • Contact address.

  • Sponsor(s).

  • Setting.

  • Funding.

  • Declaration of interest.

Intervention
  • Radiation techniques.

  • Dose.

  • Duration.

  • Control intervention.

  • Concomitant therapy and timing.

Participants
  • Sampling.

  • Total number and number in comparison groups.

  • Sex.

  • Age.

  • Trial inclusion and exclusion criteria.

  • Withdrawals and losses to follow-up (with reasons and descriptions).

  • Subgroups.

Outcomes
  • Outcomes specified above that are included in the trial report(s).

  • Length of follow-up.

  • Quality of reporting of outcomes.

If there were differences in data extraction, we resolved them by discussion, referring back to the original paper, or by consultation with a third review author. In addition, we contacted authors about information missing from their trial reports, if necessary.

Assessment of risk of bias in included studies

Two review authors assessed the methodological quality of included RCTs, QRCTs, and CCTs using the Cochrane 'Risk of bias' tool (Higgins 2011). The tool considered six domains of bias: sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other issues. For each domain, the study method was described using verbatim quotes (i.e. quotations from the original paper(s)) and judged for adequacy ('Yes' (adequate), 'No' (inadequate), 'Unclear' (insufficient information to decide 'Yes' or 'No')). A judgement of 'Yes' indicated low risk of bias, while 'No' indicated high risk of bias. We added items for the assessment of risk of bias as described in the module of the Cochrane Childhood Cancer Group (Kremer 2014). We resolved any disagreements by discussion, or by using a third-party arbitrator, and presented the results in the 'Risk of bias' table, and in both graphical and written summary form.

Measures of treatment effect

For time-to-event data (e.g. OS, EFS and PFS), we used hazard ratios (HR) with 95% confidence intervals (CI). If HRs were not explicitly presented in the study, we used Parmar's method (Parmar 1998).

For dichotomous outcomes (e.g. radiological and neurological response to treatment, toxicities), we calculated risk ratios (RR) with 95% CI for each trial.

For continuous outcomes (e.g. QoL), we evaluated mean difference (MD) or standardised mean difference (SMD) with 95% CI. We used the MD between the treatment arms at the end of follow- up if all trials measured the outcome on the same scale, otherwise, we used the SMD.

Dealing with missing data

During our review, we tried to contact authors of both included studies to clarify or gather missing data with regard to our study selection, risk of bias assessment, and data extraction. We successfully contacted Dr. Peter C. Adamson, chair of the Children's Oncology Group, and the corresponding author of Zaghloul 2014 for additional information. If appropriate, we performed intention-to-treat analyses.

Assessment of heterogeneity

We planned to assess heterogeneity both by visual inspection of the forest plots and by a formal statistical test for heterogeneity, the I2 statistic. In the absence of significant heterogeneity (I2 less than 50%) (Higgins 2011), we planned to use a fixed-effect model for the estimation of treatment effects. Otherwise, we intended to explore possible reasons for the occurrence of heterogeneity and take appropriate measures, such as using a random-effects model. However, assessment of heterogeneity was not applicable since, for each comparison, only one study was available.

Assessment of reporting biases

We evaluated reporting bias as described in the 'Assessment of risk of bias' section. We planned to construct a funnel plot for reporting bias evaluation when there were a sufficient number of included studies (i.e. at least 10 studies included in a meta-analysis) because otherwise the power of the test is too low to distinguish chance from real asymmetry (Higgins 2011). Since pooling of results was not possible, this was not applicable.

Data synthesis

We carried out data synthesis and analyses using Review Manager 5 (RevMan 2011). For time-to-event data, we used the generic inverse variance function of Review Manager 5 to combine logs of the HRs. We used Pamar's method to obtain the necessary data (Parmar 1998). Dichotomous outcomes were related to risk using the RR. We used the fixed-effect model throughout the review. Since the two included studies tested different comparisons, pooling of results was not possible.

For each comparison, we prepared a 'Summary of findings' table using the GRADE profiler software, in which we presented the following outcomes: OS, EFS, PFS, QoL, neurological response, radiological response, and toxicities. Two review authors independently assessed the quality of the evidence using the five GRADE considerations, that is, study limitations, inconsistency, indirectness, imprecision, and publication bias.

Subgroup analysis and investigation of heterogeneity

If possible, we planned to carry out the following subgroup analyses:

  • age at onset of radiotherapy (less than three years old versus three years old or older);

  • different radiation techniques;

  • different concomitant therapies.

Sensitivity analysis

Since for each comparison we found only one study, sensitivity analyses were not applicable.

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of ongoing studies tables.

Results of the search

The initial search identified 3414 records, 112 from CENTRAL, 771 from MEDLINE/PubMed, 2180 from EMBASE, 71 from trials registers (45 from the NIH register; 10 from the WHO register; 16 from the ISRCTN register), 279 from conference abstracts (69 from SIOP, 128 from ISPNO, 62 from SNO, and 20 from EANO), and 1 identified by screening reference lists (Janssens 2013). Figure 1 shows the process of study selection. After excluding 379 duplicating records, we screened the titles and abstracts of the remaining 3035 records. We subsequently excluded another 3022 records as they did not meet the inclusion criteria. We obtained the full-text publications of the remaining 13 studies (including one conference abstract and one ongoing trial) and read them thoroughly to assess eligibility. Finally, we considered two studies (Mandell 1999; Zaghloul 2014) and one ongoing trial (NCT01878266) eligible in this review.

Figure 1.

Study flow diagram.

Included studies

See: Characteristics of included studies table.

Only two trials were eligible for inclusion, both of which were RCT design (Mandell 1999; Zaghloul 2014).

The Mandell 1999 trial was a multi-institutional phase III RCT. This study focused on the comparison between conventional radiotherapy versus hyperfractionated radiotherapy. A total of 132 children and young adults were enrolled between June 1992 and March 1996. Two participants were considered ineligible due to an error in diagnosis. Diagnosis of diffuse brainstem tumour was based on clinical and radiological findings. Histological diagnosis was established in 22 participants, 10 in the conventional radiotherapy group and 12 in the hyperfractionated radiotherapy group. Median age was 78 months in the conventional radiotherapy group, and 74 months in the hyperfractionated radiotherapy group. The treatment was initiated within 28 days of diagnosis, consisting of a six-week course of local field radiotherapy with either conventional regimen or hyperfractionated regimen (details in Characteristics of included studies table). Cisplatin was added as a potential radiosensitiser in all participants. All participants received steroid medication during radiotherapy. Participants were followed until death or October 1997, and five were lost to follow-up.

The second RCT included 71 eligible participants from July 2007 to July 2011 in Children's Cancer Hospital, Egypt (CCHE) (Zaghloul 2014). Diagnosis of diffuse brainstem tumour was based on clinical and radiological findings. Participants were randomised to receive hypofractionated radiotherapy or conventional radiotherapy. Mean (± standard deviation (SD)) age was 8.3 ± 3.8 years in the hypofractionated radiotherapy group and 7.5 ± 4.1 years in the conventional radiotherapy group. Treatment was initiated within two weeks of diagnosis. Hypofractionated radiotherapy was delivered over 2.6 weeks to a total dose of 39 Gy. Conventional radiotherapy was delivered over six weeks to a total dose of 54 Gy (details in Characteristics of included studies table). No chemotherapy was administered during or after radiation. The dose of steroids was reduced in a tapering way identically in the two groups. The median follow-up period was 9.0 months (range 1.3 to 25 months). Only one participant in the conventional radiotherapy group was lost to follow-up. We contacted the corresponding author of this trial and were informed about exclusion criteria, funding, and conflicts of interest.

Excluded studies

See: Characteristics of excluded studies table.

We excluded 10 studies: two studies were retrospective design (Negretti 2011; Sun 2013), five were cohort studies (Packer 1990; Freeman 1993; Kretschmar 1993; Packer 1993; Packer 1994), one was matched-cohort analysis (Janssens 2013), one was a comment letter (Roos 2014), and one was a conference abstract (Ahmed 2012) that was reported in full-text (Zaghloul 2014).

Ongoing studies

See: Characteristics of ongoing studies table.

One RCT is ongoing comparing two hypofractionated radiotherapy regimens versus conventional radiotherapy in children with newly diagnosed DIPG.

Risk of bias in included studies

Figure 2 shows the risk of bias on seven items for each included trial. The 'Risk of bias' table details judgement of bias (see Characteristics of included studies table). We contacted Dr. Peter C. Adamson, chair of the Children's Oncology Group, who provided the study protocol of Mandell 1999. We contacted the corresponding author of Zaghloul 2014, who provided information about random sequence generation, allocation concealment, blinding of outcome assessment, incomplete outcome data, follow-up time, and median age.

Figure 2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

For evaluating selection bias, we assessed random sequence generation and allocation concealment. The risk was unclear in Mandell 1999 and low in Zaghloul 2014.

Blinding

For evaluating performance bias, we assessed the blinding of participants and personnel. Although neither of the studies clearly stated whether blinding was done, it was impossible to blind the participants or personnel, considering the different nature of the intervention (conventional: daily for 30 fractions; hyperfractionated: twice a day for 60 fractions; hypofractionated: daily for 13 fractions), Thus, we considered both trials to have a high risk of performance bias.

For evaluating detection bias, we assessed the blinding of outcome assessors for each separate outcome. For OS, we considered both Mandell 1999 and Zaghloul 2014 to have a low risk of detection bias since this outcome was unlikely to be affected by assessors' knowledge of the assignment status. Mandell 1999 also reported radiological response and toxicities: risk of detection bias for these outcomes was unclear. Zaghloul 2014 also reported toxicities: risk of detection bias for these outcomes was high.

Incomplete outcome data

For evaluating attrition bias, we assessed incomplete outcome data for each separate outcome. Both studies reported OS: the risk of attrition bias was low. Both studies reported toxicities: the risk of attrition bias was high in Mandell 1999 and low in Zaghloul 2014. Mandell 1999 reported radiological response: the risk of attrition bias was high.

Selective reporting

For evaluating reporting bias, we assessed selective reporting. The risk of bias was unclear for Mandell 1999 and low for Zaghloul 2014.

Other potential sources of bias

For evaluating other potential sources of bias, we assessed the following items: block randomisation in unblinded trials, baseline imbalance between treatment groups related to outcome (prior chemotherapy and radiotherapy, age, sex), and difference in length of follow-up between treatment arms. The risk was unclear for Mandell 1999. Zaghloul 2014 did not conduct block randomisation; baseline characteristics were balanced; and there was no significant difference in length of follow-up between treatment groups. Taken together, we considered there was low risk of other bias for Zaghloul 2014,

Effects of interventions

See: Summary of findings for the main comparison Hyperfractionated radiotherapy compared with conventional radiotherapy for diffuse brainstem glioma in children and young adults; Summary of findings 2 Hypofractionated radiotherapy compared with conventional radiotherapy for diffuse brainstem glioma in children and young adults

See: Summary of findings for the main comparison; Summary of findings 2.

We did not combine the data of these two RCTs using meta-analysis since they reported different comparisons. Instead, we presented results for each trial separately.

Comparison 1: hyperfractionated radiotherapy versus conventional radiotherapy

Only one RCT compared hyperfractionated radiotherapy versus conventional radiotherapy (Mandell 1999).

Overall survival

The analysis included all 130 participants (66 for conventional radiotherapy and 64 for hyperfractionated radiotherapy). Mandell 1999 reported that the median time to death was 8.5 months (range 3 to 24 months) for participants receiving conventional radiotherapy, and 8 months (range 1 to 23 months) for participants receiving hyperfractionated radiotherapy. We detected no significant difference in OS between the groups (one study, 130 participants, HR 1.07, 95% CI 0.75 to 1.53, P value = 0.70; Figure 3; Analysis 1.1).

Figure 3.

Forest plot of comparison: 1 Hyperfractionated radiotherapy versus conventional radiotherapy, outcome: 1.1 Overall survival.

Progression-free survival

The Mandell 1999 trial did not assess PFS.

Event-free survival

The analysis included all 130 participants (66 for conventional radiotherapy and 64 for hyperfractionated radiotherapy). Mandell 1999 reported that the median EFS was six months (range 2 to 15 months) for participants receiving conventional radiotherapy, and five months (range 1 to 12 months) for participants receiving hyperfractionated radiotherapy. We detected no significant difference in EFS between the participants who received hyperfractionated radiotherapy and participants who received conventional radiotherapy (one study, 130 participants, HR 1.26, 95% CI 0.83 to 1.90, P value = 0.27; Figure 4; Analysis 1.2).

Figure 4.

Forest plot of comparison: 1 Hyperfractionated radiotherapy versus conventional radiotherapy, outcome: 1.2 Event-free survival.

Quality of life

The Mandell 1999 trial did not assess QoL.

Neurological response

Mandell 1999 reported the neurological response for the whole group of participants rather than for treatment groups. Thus, we could not calculate RR for corresponding neurological response.

Radiological response

Only 108/130 participants (57 in the conventional radiotherapy group and 51 in the hyperfractionated radiotherapy group) in Mandell 1999 were available for review of post-treatment MRI. For the conventional radiotherapy group, the authors reported that there was complete response in one participant, partial response (greater than 50% decrease in size) in 18 participants, stable response in 25 participants, and progressive response (increase in size of disease on MRI) in 13 participants. For the hyperfractionated radiotherapy group, the authors observed complete response in one participant, partial response in 15 participants, stable response in 23 participants, and progressive response in 12 participants. There is no evidence of difference in radiological response between the two groups (RR 0.94, 95% CI 0.54 to 1.63, P value = 0.83; Figure 5; Analysis 1.3).

Figure 5.

Forest plot of comparison: 1 Hyperfractionated radiotherapy versus conventional radiotherapy, outcome: 1.3 Radiological response.

Toxicities

Mandell 1999 presented various types of toxicities, most of which were both recorded in both the conventional radiotherapy group (58 participants, unless otherwise stated (NOS)) and hyperfractionated radiotherapy group (55 participants). This trial graded the toxicities into six categories (i.e. 1 to 5 and unknown (UNK)). All the observed toxicities were minimal to mild. There was no grade 4 or 5 toxicity of hearing loss or renal dysfunction. Given the fact that the distribution of each reported toxicity in both groups was similar, we combined the count in each grade for specific toxicity item, with the aim of transferring it to dichotomous variable (i.e. toxicity and non-toxicity). There was no evidence of difference between the hyperfractionated radiotherapy group and the conventional radiotherapy group with regards to platelets (RR 1.58, 95% CI 0.27 to 9.11, P value = 0.61), haemoglobin (RR 3.16, 95% CI 0.34 to 29.51, P value = 0.31), infection NOS/UNK (RR 1.05, 95% CI 0.22 to 5.00, P value = 0.95), sepsis and bacteria (RR 2.11, 95% CI 0.20 to 22.61, P value = 0.54), upper respiratory infection/otitis (RR 1.05, 95% CI 0.43 to 2.61, P value = 0.91), abscess (RR 2.11, 95% CI 0.20 to 22.61, P value = 0.54), other bacterial (RR 1.05, 95% CI 0.15 to 7.23, P value = 0.96), fungal NOS/UNK (RR 0.35, 95% CI 0.04 to 3.28, P value = 0.36), fungal stomatitis (RR 1.24, 95% CI 0.40 to 3.84, P value = 0.70; 57 participants in the conventional radiotherapy group), alopecia (RR 1.76, 95% CI 0.44 to 7.01, P value = 0.42), decrease of magnesium (electrolyte imbalance) (RR 0.41, 95% CI 0.15 to 1.06, P value = 0.07), decrease of sodium (electrolyte imbalance) (RR 2.11, 95% CI 0.20 to 22.61, P value = 0.54), hearing (subjective) (RR 2.11, 95% CI 0.20 to 22.61, P value = 0.54), and hearing (objective) (RR 1.51, 95% CI 0.62 to 3.68, P value = 0.37) (see: Figure 6; Analysis 1.4).

Figure 6.

Forest plot of comparison: 1 Hyperfractionated radiotherapy versus conventional radiotherapy, outcome: 1.4 Toxicities.

Subgroup analysis

We did not perform a subgroup analysis due to lack of data from the included study.

Comparison 2: hypofractionated radiotherapy versus conventional radiotherapy

Only one RCT compared hypofractionated radiotherapy versus conventional radiotherapy (Zaghloul 2014). In this trial, one participant in the conventional radiotherapy was lost to follow-up. The corresponding author from Zaghloul 2014 stated that the participant lost for follow-up was enrolled in OS and adverse effects with the last observation carried forward (not as best or worst case).

Overall survival

The analysis included 71 participants (36 in the conventional radiotherapy group and 35 in the hypofractionated radiotherapy group). For the hypofractionated radiotherapy group, Zaghloul 2014 reported a median OS of 7.8 months. For the conventional radiotherapy group, Zaghloul 2014 reported a median OS of 9.5 months. There was no significant difference between the groups (one study, 71 participants, HR 1.03, 95% CI 0.53 to 2.01, P value = 0.93; Figure 7; Analysis 2.1).

Figure 7.

Forest plot of comparison: 2 Hypofractionated radiotherapy versus conventional radiotherapy, outcome: 2.1 Overall survival.

Progression-free survival

The analysis included 71 participants (36 in the conventional radiotherapy group and 35 in the hypofractionated radiotherapy group). For the hypofractionated radiotherapy group, Zaghloul 2014 reported a median PFS of 6.3 months (range 3.7 to 8.8 months). For the conventional radiotherapy group, Zaghloul 2014 reported a median PFS of 7.3 months (range 5.5 to 9.2 months). There was no significant difference between groups (one study, 71 participants, HR 1.19, 95% CI 0.63 to 2.22, P value = 0.60; Figure 8; Analysis 2.2).

Figure 8.

Forest plot of comparison: 2 Hypofractionated radiotherapy versus conventional radiotherapy, outcome: 2.2 Progression-free survival.

Event-free survival

The Zaghloul 2014 trial did not assess EFS.

Quality of life

The Zaghloul 2014 trial did not assess QoL

Neurological response

The Zaghloul 2014 trial did not assess neurological response.

Radiological response

The Zaghloul 2014 trial did not assess radiological response.

Toxicities

The trial reported information on adverse effects for all included participants. The most frequent adverse effect was local erythema and dry desquamation, especially behind the auricles, with 16 participants in the hyperfractionated radiotherapy group and 17 participants in conventional radiotherapy group. The difference was not significant (RR 0.97, 95% CI 0.59 to 1.60, P value = 0.90). For other less frequent adverse effects, there were no differences in hearing (RR 1.54, 95% CI 0.27 to 8.68, P value = 0.62), decreased appetite (RR 1.54, 95% CI 0.27 to 8.68, P value = 0.62), dysphagia (RR 1.37, 95% CI 0.33 to 5.69, P value = 0.66), fatigue (RR 1.10, 95% CI 0.63 to 1.93, P value = 0.73), insomnia (RR 0.69, 95% CI 0.12 to 3.86, P value = 0.67), night mares (RR 1.03, 95% CI 0.15 to 6.90, P value = 0.98), or seizures (RR 0.34, 95% CI 0.01 to 8.14, P value = 0.51) (see: Figure 9; Analysis 2.3).

Figure 9.

Forest plot of comparison: 2 Hypofractionated radiotherapy versus conventional radiotherapy, outcome: 2.3 Toxicities.

Subgroup analysis

We did not perform a subgroup analysis due to lack of data from the included study.

Discussion

Summary of main results

This systematic review included two RCTs that investigated the benefits and harms of two different radiotherapy techniques (compared with conventional radiotherapy) in children and young adults aged 0 to 21 years with newly diagnosed DBG. Mandell 1999, a multi-institutional RCT, enrolled 130 participants and compared hyperfractionated radiotherapy with conventional radiotherapy. We detected no clear evidence of effect on OS or EFS in participants received hyperfractionated radiotherapy when compared with participants received conventional radiotherapy. Radiological response and various types of toxicities were similar between the two groups. There were no severe toxicities. The second trial included 71 participants and compared hypofractionated radiotherapy with conventional radiotherapy (Zaghloul 2014). Based on this trial, we found no clear evidence of effect on OS or PFS in participants who received hypofractionated radiotherapy compared with participants who received conventional radiotherapy. The main adverse effect was local erythema and dry desquamation, especially behind the auricles, with a similar proportion in the two groups. There were other toxicities, but there was no statistically significant different between the treatment groups.

Overall completeness and applicability of evidence

We attempted to identify all relevant studies. We are confident that majority of published trials are included in this review. In both included RCTs, the diagnosis of DBG was established based on MRI findings (in Mandell 1999, at least two-thirds of the diffuse lesion was intrinsic to the pons; in Zaghloul 2014, diffuse infiltration of more than 50% of the pons with or without extension to the midbrain or medulla oblongata (or both)) and clinical findings (at least two of the three typical brainstem symptoms, cranial nerve deficit, long tract signs, and ataxia). However, in our protocol, we also specified the T1 and T2 findings from MRI as one of the diagnostic criteria, which were actually unavailable in both included studies. Therefore, we modified the diagnostic criteria as 'Diagnosis of DBG made as described in the original studies'. EFS, focusing on time to disease progression, disease relapse, a second malignant neoplasm, or death from any cause, could also provide valuable information regarding the treatment outcome. Therefore, it was included in this review although we did not include it as secondary outcome in prior protocol. Both included studies reported the primary outcome. For the secondary outcomes, only toxicity was reported in both included studies; only Mandell 1999 reported EFS and radiological response; only Zaghloul 2014 reported PFS. Neither study included QoL and neurological response. It should be noted that Mandell 1999 provided information about neurological response for all available participants rather than per treatment group; they reported that all participants received steroid medication during radiotherapy but presented no more details. Instead of taking the neurological symptoms as a whole, Zaghloul 2014 compared each symptom along the course of therapy between the hypofractionated radiotherapy group and the conventional radiotherapy group. They reported a non-significant trend towards more rapid improvement in cranial nerve palsy by the hyperfractionated radiotherapy group.

It is noteworthy that Mandell 1999 recommended conventional rather than hyperfractionated radiotherapy as the radiotherapeutic regimen of choice due to the similar outcome and there being less treatment burden for the participant and family as well as the radiation oncology department. Furthermore, one participant in the hyperfractionated radiotherapy group in this study was histologically diagnosed as hemangioblastoma, which is not a diagnosis of interest for this review and which might have a more favourable prognosis. Moreover, only 22 of the 130 participants had a histological diagnosis. Therefore, it is possible that there are more ineligible participants included. These confounding factors could influence the applicability of its results to current clinical practice. However, some studies shifted their aim to decrease the treatment burden and increase the quality of remaining life (Janssens 2009; Negretti 2011; Janssens 2013; Zaghloul 2014). Effect of hypofractionated radiotherapy for DIPG was thus examined.

Since only two RCTs were eligible and only one RCT was included for each specific comparison with limited sample size, combined with the limitation mentioned above, we could not draw conclusions or make any recommendations. It should also be noted that 'no evidence of effect', as identified in this review, is not the same as 'evidence of no effect'.

Quality of the evidence

We used the GRADE approach to assess the quality of evidence, and our judgements are presented in Summary of findings for the main comparison and Summary of findings 2.

For the hyperfractionated radiotherapy versus conventional radiotherapy comparison, we included only one multi-institutional RCT. This trial was well organised. Although the risk of bias was high for blinding of participants and personnel, it was acceptable since these two treatment modalities are very different from each other. The quality of evidence was low for OS and EFS, and very low for radiological response and toxicities. The main reason for downgrading was that only one trial with a small sample size was eligible. The sample size in Mandell 1999 (130 participants) was estimated based on the assumption that there could be 80% power (one sided, P value = 0.05) to detect a 15% improvement in two-year survival in hyperfractionated radiotherapy over the 15% expected on the conventional radiotherapy. In addition, the process of randomisation sequence generation and allocation concealment was not described. Furthermore, for the outcomes radiological response and toxicities, the incomplete outcome data also contributed to downgrading.

For the hypofractionated radiotherapy versus conventional radiotherapy comparison, we included only one RCT. This trial was well organised, with protocol available, sample size estimation performed, and allocation concealed. The risk of bias was high for blinding of participants and personnel, but was reasonable since these two treatment modalities are very different from each other. The quality of evidence was moderate for OS and low for PFS and toxicities. The quality was downgraded mainly because only one trial with a small sample size was found for this comparison. In Zaghloul 2014, the sample size estimation was based on non-inferiority test. However, it should be noted that the small sample size (only 71 children were included) might generate wider CI values. Furthermore, for PFS and toxicities, the presence of high risk of detection bias also contribute to the consideration of downgrading.

Potential biases in the review process

We followed a well-designed search strategy to identify potentially eligible studies, with no restriction on language, publication status, or sample size. Two review authors independently conducted the study selection and data extraction processes. Therefore, we made every attempt to minimise bias during the review process.

Agreements and disagreements with other studies or reviews

Both RCTs used well-specified eligibility criteria: at least two of the three typical brainstem symptoms with a clinical history of symptoms less than six months for Mandell 1999 and no longer than three months for Zaghloul 2014; extent of pontine involvement of tumour on MRI was required (greater than 66% for Mandell 1999 and 50% or greater for Zaghloul 2014). Median age and age range in both studies were in agreement with many other clinical trials according to previous critical reviews (Hargrave 2006; Jansen 2012). In addition, according to the reported baseline characteristics, neither study included children younger than three years. For Zaghloul 2014, a Karnofsky/Lansky play status of 50% or greater was required. For Mandell 1999, median OS in the hyperfractionated radiotherapy group (8.0 months) was within the range (8.0 to 10.5 months) in studies assessing hyperfractionated radiotherapy with specified clinical and radiological eligibility criteria applied (Hargrave 2006). For Zaghloul 2014, the median OS and PFS in the hypofractionated radiotherapy group (7.8 months for OS and 6.3 months for PFS) were within the range (7 to 14 months for median OS and 5 to 8 months for median PFS) reported in studies with specified MRI criteria for DIPG (greater than 50% pontine involvement) (Jansen 2012). Furthermore, the median OS and PFS in Zaghloul 2014 appeared similar, although slightly different, with that reported in studies focusing on hypofractionated radiotherapy (Janssens 2009; Negretti 2011; Janssens 2013). Based on only one available RCT (Zaghloul 2014), Zaghloul recommended hypofractionated radiotherapy as standard radiotherapy due to the findings that the child and his/her family could spend less than 10% of his/her remaining survival time on treatment, while conventional radiotherapy needs more than double this time, with minimal affection of survival rates (Zaghloul 2015). In addition, we are waiting for the result from an important ongoing RCT comparing two hypofractionated radiotherapy regimens versus conventional radiotherapy in children with newly diagnosed DIPG (NCT01878266).

There are no reviews available discussing the effects of radiotherapy (compared between different radiotherapy techniques or with other therapies) based on RCTs, QRCTs, or CCTs.

Authors' conclusions

Implications for practice

We identified only two randomised controlled trials (RCTs) evaluating radiotherapy in newly diagnosed diffuse brainstem glioma in children and young adults aged 0 to 21 years. One RCT compared hyperfractionated versus conventional radiotherapy and was conducted from 1992 to 1997 (Mandell 1999). There was no clear evidence of effect of hyperfractionated radiotherapy with regard to overall survival (OS), event-free survival (EFS), radiological response, and various types of toxicities. Considering the similar outcomes and treatment burden for participant and family as well as radiation oncology department, Mandell et al. recommended conventional radiotherapy as standard radiotherapy strategy. The second study compared hypofractionated versus conventional radiotherapy and was conducted from 2007 to 2013 (Zaghloul 2014). There was no clear evidence of effect of hypofractionated radiotherapy on OS, progression-free survival (PFS), or toxicities. Based on the results of this most recent trial, Zaghloul et al. recommended hypofractionated radiotherapy instead of conventional radiotherapy as standard radiotherapy due to the lesser time spent on treatment with minimal effect on survival rates. Neither study reported information on quality of life (QoL) and neurological response, and only one study reported radiological response. It should be noted that no evidence of effect, as identified in this review, is not the same as evidence of no effect.

Implications for research

Further research is needed to establish the role of radiotherapy in management of diffuse brainstem glioma in children and young adults. Future RCTs should be conducted with adequate power and all relevant outcomes should be taken into consideration, including OS, EFS, PFS, QoL, neurological response, radiological response, toxicities, and steroid consumption. Given the incidence of this rare entity, international multicentre collaboration is encouraged. Considering the potential advantage of hypofractionated radiotherapy to decrease the treatment burden and increase the quality of remaining life, we suggest that more attention should be given to hypofractionated radiotherapy.

Acknowledgements

We are grateful to Dr. Edith Leclercq for helping with the development of the search strategy; running the searches in CENTRAL, MEDLINE/PubMed, and EMBASE; and providing us with titles/abstracts. Dr. Edith Leclercq also provided us with valuable suggestions for searching conference abstracts. We are also grateful to other members from the editorial base of the Cochrane Childhood Cancer group for their kind review and valuable suggestions for our work. The Stichting Kinderen Kankervrij (KiKa) funds the editorial base of Cochrane Childhood Cancer Group. We are grateful to Dr. Zaghloul and Dr. Peter C. Adamson for providing additional information of the included studies. We are also grateful to the peer reviewers, Dr E. Bouffet and Prof. D.N. Sharma.

Data and analyses

Download statistical data

Comparison 1. Hyperfractionated radiotherapy versus conventional radiotherapy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Overall survival1130Hazard Ratio (Fixed, 95% CI)1.07 [0.75, 1.53]
2 Event-free survival1130Hazard Ratio (Fixed, 95% CI)1.26 [0.83, 1.90]
3 Radiological response1108Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.54, 1.63]
4 Toxicities1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
4.1 Platelets1113Risk Ratio (M-H, Fixed, 95% CI)1.58 [0.27, 9.11]
4.2 Haemoglobin1113Risk Ratio (M-H, Fixed, 95% CI)3.16 [0.34, 29.51]
4.3 Infection (not otherwise specified/unknown)1113Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.22, 5.00]
4.4 Sepsis, bacteria1113Risk Ratio (M-H, Fixed, 95% CI)2.11 [0.20, 22.61]
4.5 Upper respiratory infection/otitis1113Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.43, 2.61]
4.6 Abscess1113Risk Ratio (M-H, Fixed, 95% CI)2.11 [0.20, 22.61]
4.7 Other bacterial1113Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.15, 7.23]
4.8 Fungal (not otherwise specified/unknown)1113Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.04, 3.28]
4.9 Fungal stomatitis1112Risk Ratio (M-H, Fixed, 95% CI)1.24 [0.40, 3.84]
4.10 Alopecia1113Risk Ratio (M-H, Fixed, 95% CI)1.76 [0.44, 7.01]
4.11 Decrease of magnesium (electrolyte imbalance)1113Risk Ratio (M-H, Fixed, 95% CI)0.41 [0.15, 1.06]
4.12 Decrease of sodium (electrolyte imbalance)1113Risk Ratio (M-H, Fixed, 95% CI)2.11 [0.20, 22.61]
4.13 Hearing (subjective)1113Risk Ratio (M-H, Fixed, 95% CI)2.11 [0.20, 22.61]
4.14 Hearing (objective)1113Risk Ratio (M-H, Fixed, 95% CI)1.51 [0.62, 3.68]
Analysis 1.1.

Comparison 1 Hyperfractionated radiotherapy versus conventional radiotherapy, Outcome 1 Overall survival.

Analysis 1.2.

Comparison 1 Hyperfractionated radiotherapy versus conventional radiotherapy, Outcome 2 Event-free survival.

Analysis 1.3.

Comparison 1 Hyperfractionated radiotherapy versus conventional radiotherapy, Outcome 3 Radiological response.

Analysis 1.4.

Comparison 1 Hyperfractionated radiotherapy versus conventional radiotherapy, Outcome 4 Toxicities.

Comparison 2. Hypofractionated radiotherapy versus conventional radiotherapy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Overall survival171Hazard Ratio (Fixed, 95% CI)1.03 [0.53, 2.01]
2 Progression-free survival171Hazard Ratio (Fixed, 95% CI)1.19 [0.63, 2.22]
3 Toxicities1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
3.1 Skin171Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.59, 1.60]
3.2 Hearing171Risk Ratio (M-H, Fixed, 95% CI)1.54 [0.27, 8.68]
3.3 Decreased appetite171Risk Ratio (M-H, Fixed, 95% CI)1.54 [0.27, 8.68]
3.4 Dysphagia171Risk Ratio (M-H, Fixed, 95% CI)1.37 [0.33, 5.69]
3.5 Fatigue171Risk Ratio (M-H, Fixed, 95% CI)1.10 [0.63, 1.93]
3.6 Insomnia171Risk Ratio (M-H, Fixed, 95% CI)0.69 [0.12, 3.86]
3.7 Night mares171Risk Ratio (M-H, Fixed, 95% CI)1.03 [0.15, 6.90]
3.8 Seizures171Risk Ratio (M-H, Fixed, 95% CI)0.34 [0.01, 8.14]
Analysis 2.1.

Comparison 2 Hypofractionated radiotherapy versus conventional radiotherapy, Outcome 1 Overall survival.

Analysis 2.2.

Comparison 2 Hypofractionated radiotherapy versus conventional radiotherapy, Outcome 2 Progression-free survival.

Analysis 2.3.

Comparison 2 Hypofractionated radiotherapy versus conventional radiotherapy, Outcome 3 Toxicities.

Appendices

Appendix 1. Search strategy for Cochrane Central Register of Controlled Trials (CENTRAL)

1. ForGlioma, we used the following text words:

glioma OR gliomas OR glioma* OR Glial Cell Tumors OR Glial Cell Tumor OR Tumor, Glial Cell OR Tumors, Glial Cell OR Mixed Glioma OR Glioma, Mixed OR Gliomas, Mixed OR Mixed Gliomas OR Malignant Glioma OR Glioma, Malignant OR Gliomas, Malignant OR Malignant  Gliomas OR brain stem glioma OR brainstem glioma OR brain stem tumor OR brain stem tumors OR brainstem tumor OR brainstem tumors OR brain stem tumour OR brainstem tumour OR  brain stem tumours OR brainstem tumours OR brain stem neoplasm OR brain stem neoplasms OR pons neoplasm

2. For Radiotherapy, we used the following text words:

Radiotherapy OR radiotherap* OR radiotherapies OR Targeted Radiotherapies OR Targeted Radiotherapy OR adjuvant radiotherapy OR radiation oncology OR radiation OR radiation* OR radiation therapy OR radiation therap* OR "radiotherapy fractionation" OR irradiation or irradiation* OR chemoradiotherapy OR chemoradiotherap* OR chemoradiotherapies OR Combined Modality Therapy OR Multimodal Treatment OR Multimodal Treatments OR  Combined Modality Therapies OR hyperfractionated radiotherapy OR hyperfractionated radiotherap* OR hyperfractionated radiotherapies OR hyperfractionated radiation OR hyperfractionated radiation* OR hyperfractionated irradiation OR hyperfractionated irradiation* OR accelerate* hyperfractionate* OR hyperfractionation OR radiotherapy dosage OR Radiotherapy Dosages OR Radiation Dose-Response Relationship OR Radiation Dose-Response Relationships OR dose fractionation OR Dose Fractionations OR Radiotherapy Dose Fractionation OR Radiotherapy Dose Fractionations

3. For Children and young adults, we used the following text words:

infant OR infan* OR newborn OR newborn* OR new-born* OR baby OR baby* OR babies OR neonat* OR perinat* OR postnat* OR child OR child* OR schoolchild* OR schoolchild OR school child OR school child* OR kid OR kids OR toddler* OR adolescent OR adoles* OR teen* OR boy* OR girl* OR minors OR minors* OR underag* OR under ag* OR juvenil* OR youth* OR kindergar* OR puberty OR puber* OR pubescen* OR prepubescen* OR prepuberty* OR pediatrics OR pediatric* OR paediatric* OR peadiatric* OR schools OR nursery school* OR preschool* OR pre school* OR primary school* OR secondary school* OR elementary school* OR elementary school OR high school* OR highschool* OR school age OR schoolage OR school age* OR schoolage* OR infancy OR young adult OR young adults OR young adult*

Final search: 1 AND 2 AND 3

[*=zero to many characters] 

Appendix 2. Search strategy for PubMed

1. ForGlioma, we used the following MeSH headings and text words:

glioma OR gliomas OR glioma* OR Glial Cell Tumors OR Glial Cell Tumor OR Tumor, Glial Cell OR Tumors, Glial Cell OR Mixed Glioma OR Glioma, Mixed OR Gliomas, Mixed OR Mixed Gliomas OR Malignant Glioma OR Glioma, Malignant OR Gliomas, Malignant OR Malignant Gliomas OR brain stem glioma OR brainstem glioma OR brain stem tumor OR brain stem tumors OR brainstem tumor OR brainstem tumors OR brain stem tumour OR brainstem tumour OR brain stem tumours OR brainstem tumours OR brain stem neoplasm OR brain stem neoplasms OR pons neoplasm OR pons neoplasms

2. ForRadiotherapy, we used the following MeSH headings and text words:

Radiotherapy OR radiotherap* OR radiotherapies OR Radiotherapy, Targeted OR Radiotherapies, Targeted OR Targeted Radiotherapies OR Targeted Radiotherapy OR radiotherapy[sh] OR adjuvant radiotherapy OR radiation oncology OR radiation[tiab] OR radiation* OR radiation therapy[tiab] OR radiation therap* OR "radiotherapy fractionation" OR (radiotherapy AND fractionation[tiab]) OR irradiation OR irradiation* OR chemoradiotherapy OR chemoradiotherap* OR chemoradiotherapies OR Combined Modality Therapy OR Multimodal Treatment OR Multimodal Treatments OR Treatment, Multimodal OR Treatments, Multimodal OR Therapy, Combined Modality OR Combined Modality Therapies OR Modality Therapies, Combined OR Modality Therapy, Combined OR Therapies, Combined Modality OR hyperfractionated radiotherapy OR hyperfractionated radiotherap* OR hyperfractionated radiotherapies OR hyperfractionated radiation OR hyperfractionated radiation* OR hyperfractionated irradiation OR hyperfractionated irradiation* OR accelerate* hyperfractionate* OR hyperfractionation OR radiotherapy dosage OR Dosage, Radiotherapy OR Dosages, Radiotherapy OR Radiotherapy Dosages OR dose-response relationship, radiation OR Dose Response Relationship, Radiation OR Dose-Response Relationships, Radiation OR Radiation Dose-Response Relationship OR Radiation Dose-Response Relationships OR Relationship, Radiation Dose-Response OR Relationships, Radiation Dose-Response OR dose fractionation OR Dose Fractionations OR Fractionation, Dose OR Fractionations, Dose OR Radiotherapy Dose Fractionation OR Dose Fractionation, Radiotherapy OR Dose Fractionations, Radiotherapy OR Fractionation, Radiotherapy Dose OR Fractionations, Radiotherapy Dose OR Radiotherapy Dose Fractionations

3. ForChildren and young adults, we used the following MeSH headings and text words:

infant OR infan* OR newborn OR newborn* OR new-born* OR baby OR baby* OR babies OR neonat* OR perinat* OR postnat* OR child OR child* OR schoolchild* OR schoolchild OR school child OR school child* OR kid OR kids OR toddler* OR adolescent OR adoles* OR teen* OR boy* OR girl* OR minors OR minors* OR underag* OR under ag* OR juvenil* OR youth* OR kindergar* OR puberty OR puber* OR pubescen* OR prepubescen* OR prepuberty* OR pediatrics OR pediatric* OR paediatric* OR peadiatric* OR schools OR nursery school* OR preschool* OR pre school* OR primary school* OR secondary school* OR elementary school* OR elementary school OR high school* OR highschool* OR school age OR schoolage OR school age* OR schoolage* OR infancy OR schools, nursery OR infant, newborn OR young adult OR young adults OR young adult*

4. Forrandomised controlled trials (RCTs) and controlled clinical trials (CCTs), we used the following MeSH headings and text words:

(randomized controlled trial[pt] OR controlled clinical trial[pt] OR randomized[tiab] OR placebo[tiab] OR drug therapy[sh] OR randomly[tiab] OR trial[tiab] OR groups[tiab]) AND humans[mh]

Final search: 1 AND 2 AND 3 AND 4

[pt = publication type; tiab = title, abstract; sh = subheading; mh = MeSH term; * = zero to many characters; RCT = randomized controlled trial; CCT = controlled clinical trial]

Appendix 3. Search strategy for EMBASE (Ovid)

1 For Glioma, we used the following Emtree terms and text words:

1. exp glioma/ or exp pontine glioma/
2. (glioma or gliomas or glioma$).mp.
3. (Glial Cell Tumor or Glial Cell Tumors).mp.
4. (Mixed Glioma or Mixed Gliomas).mp.
5. (Malignant Glioma or Malignant Gliomas).mp.
6. (brain stem glioma or brain stem gliomas or brainstem glioma or brainstem gliomas).mp.
7. exp brain stem tumor/
8. (brain stem tumor or brain stem tumors or brainstem tumor or brainstem tumors or brain stem tumour or brainstem tumour or brain stem tumours or brainstem tumours).mp.
9. brain neoplasm.mp.
10. (brain stem neoplasm or brain stem neoplasms).mp.
11. (pons neoplasm or pons neoplasms).mp.
12. or/1-11

2. For Radiotherapy, we used the following Emtree terms and text words:

1. exp Radiotherapy/ or exp Cancer Radiotherapy/
2. (radiotherapy or radiotherapies or radiotherap$).mp.
3. (targeted radiotherapy or targeted radiotherapies or targeted radiotherap$).mp.
4. radiotherapy.sh.
5. adjuvant radiotherapy.mp.
6. radiation oncology.mp.
7. (radiation or radiation therapy).ti,ab.
8. (radiation$ or radiation therap$).mp.
9. exp Radiation Dose Fractionation/ or radiotherapy fractionation.mp.
10. (radiotherapy and fractionation).ti,ab.
11. (irradiation or irradiation$).mp.
12. (chemoradiotherapy or chemoradiotherap$ or chemoradiotherapies).mp.
13. Combined Modality Therapy.mp. or exp Multimodality Cancer Therapy/
14. (multimodal treatment or multimodal treatments).mp.
15. combined modality therapies.mp.
16. (hyperfractionated radiotherapy or hyperfractionated radiotherap$ or hyperfractionated radiotherapies).mp.
17. (hyperfractionated radiation or hyperfractionated radiation$).mp.
18. (hyperfractionated irradiation or hyperfractionated irradiation$).mp.
19. accelerate$ hyperfractionate$.mp.
20. hyperfractionation.mp.
21. radiotherapy dosage.mp. or exp Radiation Dose/ or exp Radiation Dose Distribution/
22. radiotherapy dosages.mp.
23. (radiation dose-response relationship or radiation dose-response relationships).mp.
24. (dose fractionation or dose fractionations).mp.
25. (radiotherapy dose fractionation or radiotherapy dose fractionations).mp.
26. or/1-25

3. For Children and young adults, we used the following Emtree terms and text words:

1. infant/ or infancy/ or newborn/ or baby/ or child/ or preschool child/ or school child/
2. adolescent/ or juvenile/ or boy/ or girl/ or puberty/ or prepuberty/ or pediatrics/ or female/
3. primary school/ or high school/ or kindergarten/ or nursery school/ or school/
4. or/1-3
5. (infant$ or newborn$ or (new adj born$) or baby or baby$ or babies or neonate$ or perinat$ or postnat$).mp.
6. (child$ or (school adj child$) or schoolchild$ or (school adj age$) or schoolage$ or (pre adj school$) or preschool$).mp.
7. (kid or kids or toddler$ or adoles$ or teen$ or boy$ or girl$).mp.
8. (minors$ or (under adj ag$) or underage$ or juvenil$ or youth$ or young adult or young adults or young adult$).mp.
9. (puber$ or pubescen$ or prepubescen$ or prepubert$).mp.
10. (pediatric$ or paediatric$ or peadiatric$).mp.
11. (school or schools or (high adj school$) or highschool$ or (primary adj school$) or (nursery adj school$) or (elementary adj school) or (secondary adj school$) or kindergar$).mp.
12. or/5-11
13. 4 or 12

 4. For randomised controlled trials (RCTs) and controlled clinical trials (CCTs), we used the following Emtree terms and text words:

1. Randomized Controlled Trial/
2. Controlled Clinical Trial/
3. randomized.ti,ab.
4. placebo.ti,ab.
5. randomly.ti,ab.
6. trial.ti,ab.
7. groups.ti,ab.
8. drug therapy.sh.
9. or/1-8
10. Human/
11. 9 and 10

Final search: 1 and 2 and 3 and 4

[mp = title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer; $= zero or more characters; /= Emtree term; sh = subheading; ti,ab = title or abstract] 

Appendix 4. Search strategy for conference abstracts (Embase.com)

1.'international society of paediatric oncology':nc AND [embase]/lim

2.'international society of pediatric oncology':nc AND [embase]/lim

3.'international symposium on pediatric neuro-oncology':nc AND [embase]/lim

4.'society for neuro-oncology':nc AND [embase]/lim

5.'european association of neuro-oncology':nc AND [embase]/lim

6.'european association of neurooncology':nc AND [embase]/lim

7.#1 OR #2 OR #3 OR OR #4 OR #5 OR #6 AND [1-1-2010]/sd NOT [19-8-2015]/sd

8.pons OR pontine OR brainstem OR 'brain stem' AND [embase]/lim

9.glioma* OR tumor* OR neoplasm* AND [embase]/lim

10.radiotherap* OR radiation AND [embase]/lim

11.#7 AND #8 AND #9 AND #10

[nc = conference name; lim = limit; sd = since date; * = zero or more characters]

Appendix 5. Search strategy for ongoing trial registers

1. www.isrctn.com (ISRCTN register)

Search terms were: glioma

2. apps.who.int/trialsearch (WHO register)

In advanced search page, search terms were glioma ORtumor ORneoplasm for Conditions, radiotherapy OR radiation for Interventions, brainstem OR brain stem OR pontine OR pons for Titles.

3. https://clinicaltrials.gov (NIH registered)

In advanced search page, search terms were glioma ORtumor ORneoplasm for Conditions, radiotherapy OR radiation for Interventions, brainstem OR brain stem OR pontine OR pons for Titles.

Contributions of authors

X Hu: selection of trials, risk of assessment of trials, data extraction, data analyses, GRADE assessment, preparation of the Summary of Findings tables, contributed to the results and discussion, and developed the draft of the review.

Y Fang: protocol and review development; searching for trials, selection of trials, data extraction, data analyses, GRADE assessment, preparation of the Summary of Findings tables , and contributed to the results and discussion.

X Hui: protocol and review development, searching for trials, risk of bias assessment of trials, data extraction, data analyses, and review development.

Y Jv: protocol development, searching for trials, and contributed to discussion.

C You: organization of the revision process, and contributed to the discussion.

Declarations of interest

None known.

Differences between protocol and review

In the protocol, we specified that diagnostic criteria for diffuse brainstem glioma should be made by magnetic resonance imaging (MRI) or histological tests. Furthermore, the MRI criteria for the diagnosis of diffuse brainstem glioma was a T1 isointense or hypointense, T2 hyperintense tumour originating in the pons, with at least 50% involvement of the pontine region. In our review, both included studies performed MRI but neither mentioned T1 or T2. If the diagnostic criteria remained unchanged as the protocol, neither of the included study could be considered as eligible and important information might have been missed. Therefore, we modified the diagnostic criteria as 'Diagnosis of diffuse brainstem glioma made as described in the original studies'.

Our previous definition of the neurological response (i.e. defined as the number of participants with a complete or partial remission) was actually in duplicate with that of radiological response (i.e. number of participants with a reduction in tumour size of more than 50%). Therefore, we changed the definition of the neurological response into 'the number of participants with improved neurological function'.

In the protocol, PFS was defined as time from study entry to progression. However, the starting time point for PFS calculation varied among different studies. Therefore, in this review, definition of PFS has been changed and PFS was calculated from time of diagnosis, study entry, or treatment to progression.

We added 'event-free survival' as one of the secondary outcome since some studies measure event-free survival while some measured progression-free survival and some measured both parameters.

We did not mention 'Summary of findings' tables in the protocol. We reconsidered this and decided that 'Summary of findings' tables were necessary and would help the reader to obtain the important information quickly. Therefore, we added the tables to the review.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Mandell 1999

MethodsPhase III prospective RCT
Participants

Sample: 130 participants (aged 37-266 months; 57 boys and 73 girls; conventional radiotherapy group: 30 boys and 36 girls; median age 78 months (range 40-266 months); hyperfractionated radiotherapy: 27 boys and 37 girls; median age 74 months (range 37-212 months))

Setting: multi-institutional; USA (Departments of Radiation Oncology and Pediatrics, Mount Sinai Medical Center; Children's Hospital of San Diego; Zeneca Pharmaceuticals; Department of Radiation Oncology, Harper Hospital; Department of Neurology, Buffalo Children's Hospital; MACC Fund Research Center; Department of Pathology, Johns Hopkins Hospital, Baltimore; Semmes-Murphy Clinic, Memphis, TN University of Florida; Department of Pediatrics, Duke University Medical School; Department of Radiation Oncology, St. Jude Children's Research Hospital) and Canada (Department of Radiation Oncology, McGill University)

Period of trial conducted: June 1992 to October 1997

Inclusion criteria: aged 3-21 years; newly diagnosed brainstem tumour based on radiological and clinical findings; at least two-thirds of the lesion was intrinsic to the pons on MRI; clinical history of < 6 months; at least 2 of the neurological triad (cranial nerve deficit, long tract signs, and ataxia); no prior chemotherapy and radiotherapy; life expectancy > 6 weeks; adequate haematological and renal function

Exclusion criteria: not mentioned

Interventions

Treatment was initiated within 28 days of diagnosis and consisted of a 6-week course of local field radiotherapy

Conventional radiotherapy (66 participants): once a day treatment of 180 cGy per fraction to a total dose of 5400 cGy for a 6-week course

Hyperfractionated radiotherapy (64 participants): twice a day regimen of 117 cGy per fraction to a total dose of 7020 cGy for a 6-week course

Treatment volume was to include a 2-cm margin around MRI-visualised abnormalities in all directions. All participants received 3 courses of cisplatin (delivered by continuous infusion over 120 hours, beginning on the first day of radiotherapy on weeks 1, 3, and 5). All participants received steroid medication during radiotherapy

Outcomes

Overall survival (defined as time from diagnosis to time of death)

Event-free survival (defined as time from diagnosis to off-study) (off-study criteria: progressive or recurrent disease; death) (based on the study protocol provided by Peter C. Adamson)

Radiological response (number of participants with > 50% decrease in size on MRI)

Toxicity (graded by standard National Cancer Institute common toxicity criteria; including platelets, haemoglobin, infection NOS/UNK, sepsis, bacteria, URI/otitis, abscess, other bacterial, fungal NOS/UNK, fungal stomatitis, alopecia, magnesium decrease, sodium decrease, hearing - subjective, hearing - objective considering grade ≥ 1 toxicities)

Notes

Sponsor: National Cancer Institute

Funding and declaration of interest: not mentioned

The study stated that 130 participants were finally eligible. However, in a figure caption it was presented that there were 66 participants in the conventional radiotherapy group and 65 participants in the hyperfractionated radiotherapy group (so total of 131 participants). We were unable to contact the author to clarify this. We considered that this was possibly a typographical error

Some toxicities were not reported for both groups and we were unsure if that meant that the participants did not have that toxicity. Therefore, we did not include these toxicities in the review

Histological diagnosis was confirmed in 22/130 participants. 1 participant in the hyperfractionated radiotherapy group was diagnosed as having a hemangioblastoma, which is not a diffuse brainstem glioma and thus ineligible for this review.

According to the protocol (provided by Peter C. Adamson) and full-text of this study, the eligible participants should be between 3 and 21 years of age. However, the age range of participants in conventional radiotherapy group was 40-266 months.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "prospective randomized trial"

Comment: no description of sequence generation process

Allocation concealment (selection bias)Unclear risk

Quote: "prospective randomized trial"

Comment: no information regarding allocation concealment provided

Blinding of participants and personnel (performance bias)
All outcomes
High riskComment: not mentioned. Considering different nature of the intervention, we assumed that participants and care providers were not blinded
Blinding of outcome assessment (detection bias)
Overall survival
Low riskComments: although no information was provided, we judged this as a low risk of bias considering that this outcome is unlikely to be affected by the participant's or personnel's knowledge of the assignment status
Blinding of outcome assessment (detection bias)
Event-free survival
Unclear riskComment: insufficient information to judge
Blinding of outcome assessment (detection bias)
Radiological response
Unclear riskComment: insufficient information to judge
Blinding of outcome assessment (detection bias)
Toxicities
Unclear riskComment: insufficient information to judge
Incomplete outcome data (attrition bias)
Overall survival
Low riskComment: all participants were analysed
Incomplete outcome data (attrition bias)
Event-free survival
Low riskComment: all participants were analysed
Incomplete outcome data (attrition bias)
Radiological response
High risk

Quote: "available for review in 108 of the 130 patients"

Comment: status of radiological response was not available in 22 participants

Incomplete outcome data (attrition bias)
Toxicities
High risk

Quote: "Toxicity reported among 58 patients evaluable for Treatment 01 (180 cGy/d to 5400)" (conventional radiotherapy) and "Toxicity reported among 55 patients evaluable for Treatment 02 (117 cGy/d to 7020)" (hyperfractionated radiotherapy); for fungal stomatitis toxicity, grade was unknown for 1 additional participant in the conventional radiotherapy group

Comment: toxicities was not reported in 17 or 18 participants

Selective reporting (reporting bias)Low riskComment: protocol provided by Dr. Peter C. Adamson; the outcomes were reported accordingly
Other biasUnclear risk

Block randomisation in unblinded trials: unclear (not mentioned)

Baseline imbalance between treatment arms related to outcome (prior chemotherapy and radiotherapy, age, sex): low (age and sex was similar between the 2 arms; all participants received no prior chemotherapy and radiotherapy)

Difference in length of follow-up between treatment arms: unclear (not mentioned)

Central review: copies of all clinical, pathological, and radiological data were to be gathered centrally and reviewed by the study co-ordinators

Zaghloul 2014

  1. a

    MRI: magnetic resonance imaging; NOS: not otherwise specified; RCT: randomised controlled trial; UNK: unknown; URI; upper respiratory infection.

MethodsPhase III RCT
Participants

Sample: 71 participants (median age 7.9 years; 37 boys and 34 girls; conventional radiotherapy group: 18 boys and 18 girls; median age 7.7 years (range 3.9-12.0 years); hypofractionated radiotherapy group: 19 boys and 16 girls; median age 8.1 years (range 4.0-14.2 years))

Setting: Radiation Oncology Department, Children's Cancer Hospital, Egypt

Period of trial conducted: July 2007 to Jan 2013;

Inclusion criteria: aged < 18 years; MRI showing diffuse infiltration of > 50% of the pons with or without extension to the midbrain or medulla oblongata (or both); at least 2 of the 3 typical brainstem symptoms; clinical history no longer than 3 months; Karnofsky/Lansky play status ≥ 5, unless the reason for decrease in status was a direct result of neurological involvement; no prior chemotherapy or radiotherapy

Exclusion criteria: opposite of the inclusion criteria (provided by the corresponding author)

Interventions

Treatment was initiated as soon as possible and within 2 weeks of diagnosis

Hypofractionated radiotherapy (35 participants): total dose of 39 Gy in 13 daily fractions of 3 Gy each, given over 2.6 weeks

Conventional radiotherapy (36 participants): 54 Gy in 30 fractions over 6 weeks, 1.8 Gy per fraction

No chemotherapy was administered during or after radiation

Steroids were allowed and usually initiated before radiation treatment was started

Outcomes

Overall survival (calculated from time of diagnosis to death or time of reporting)

Progression-free survival (calculated from time of diagnosis to time of documented failure) (documented failure: clinical or radiological progression on MRI)

Toxicity (recorded according to Radiation Therapy Oncology Group criteria; including skin, hearing, decreased appetite, dysphagia, fatigue, insomnia, night mares, seizures; considering grade ≥ 1 toxicities)

Notes

Sponsor: Children's Cancer Hospital Egypt 57357

Funding and declaration of interest: funded by Children's Cancer Hospital Egypt and no conflict of interest to declare (provided by the corresponding author).

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskQuote: "simple randomisation, sequence of numbers was computer generated" (provided by the corresponding author)
Allocation concealment (selection bias)Low riskQuote: "Concealment of randomization was performed using sequential, sealed opaque envelopes" (provided by the corresponding author)
Blinding of participants and personnel (performance bias)
All outcomes
High riskComment: this trial was open-label (NCT01635140)
Blinding of outcome assessment (detection bias)
Overall survival
Low riskComment: although no information was provided, we judged this as a low risk of bias considering that this outcome is unlikely to be affected by the participant's or personnel's knowledge of the assignment status
Blinding of outcome assessment (detection bias)
Progression-free survival
High riskQuote: "assessment of progression-free survival was not blinded on the clinical examination level" (provided by the corresponding author)
Blinding of outcome assessment (detection bias)
Toxicities
High riskQuote: "assessment of side-effects was not blinded on the clinical examination level" (provided by the corresponding author)
Incomplete outcome data (attrition bias)
Overall survival
Low riskComment: all participants were analysed (provided by the corresponding author)
Incomplete outcome data (attrition bias)
Progression-free survival
Low riskComment: progression-free survival was assessed in all participants (provided by the corresponding author)
Incomplete outcome data (attrition bias)
Toxicities
Low riskComment: toxicities were assessed in all participants (provided by the corresponding author)
Selective reporting (reporting bias)Low risk

Quote: "The study was registered in Clinical Trial. Gov (NCT01635140)"

Comment: prospectively registered protocol available and outcomes reported accordingly

Other biasLow risk

Block randomisation in unblinded trials: used simple randomisation (provided by the corresponding author)

Baseline imbalance between treatment groups related to outcome (prior chemotherapy and radiotherapy, age, sex): low (age and sex were similar between the 2 groups; all participants received no prior chemotherapy or radiotherapy). In addition to the items defined by us as being important for baseline imbalance, this study also reported that signs and symptoms, site, Karnofsky/Lanesky status, shunt application, and steroid needs were similar between the 2 groups

Difference in length of follow-up between treatment groups: no significant difference (provided by the corresponding author)

Central review: this was a single-centre RCT, so central review is not needed

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    CCT: controlled clinical trial; quasi-RCT; quasi-randomised controlled trial; RCT: randomised controlled trial.

Ahmed 2012Conference abstract was in a full-text report
Freeman 1993Not an RCT, quasi-RCT, or CCT; cohort study
Janssens 2013Not an RCT, quasi-RCT, or CCT; matched-cohort study
Kretschmar 1993Not an RCT, quasi-RCT, or CCT; cohort study
Negretti 2011Not an RCT, quasi-RCT, or CCT; retrospective studies
Packer 1990Not an RCT, quasi-RCT, or CCT; cohort study
Packer 1993Not an RCT, quasi-RCT, or CCT; cohort study
Packer 1994Not an RCT, quasi-RCT, nor CCT; cohort study
Roos 2014Not an RCT, quasi-RCT, nor CCT; comment letter
Sun 2013Not an RCT, quasi-RCT, nor CCT; retrospective study

Characteristics of ongoing studies [ordered by study ID]

NCT01878266

Trial name or titleProspective Randomized Trial of Two Hypofractionated Radiotherapy Regimens Versus Conventional Radiotherapy in Pediatric Diffuse Brainstem Glioma
MethodsRandomised
Participants

Children aged 2-18 years with newly diagnosed diffuse intrinsic brainstem glioma

Estimated enrolment: 119

Interventions

Hypofractionated radiotherapy: total dose of 39 Gy in daily fractions of 3 Gy, 5 fractions per week

Hypofractionated radiotherapy: total dose to 4500 cGy in 15 fractions in 3 weeks

Conventional radiotherapy: total dose of 54 Gy in 30 fractions giving 1.8 Gy per fraction

Outcomes

Primary outcome: median overall survival (time frame: 3 years)

Secondary outcome: progression-free survival (time frame: 3 years)

Starting dateFebruary 2013
Contact information

Children's Cancer Hospital Egypt 57357, Cairo, Egypt, 11441 (Mohamed S Zaghloul)

www.clinicaltrials.gov (NCT01878266)

NotesThe status of this study is 'Recruiting' (last checked 4 April 2016), although the estimated completion date was "December 2013"