Journal Pre-Proofs: Radiotherapy and Oncology

Journal Pre-proofs
Review Article
A review of long-term deficits in memory systems following radiotherapy for

pediatric posterior fossa tumor
Eloïse Baudou, Lisa Pollidoro, Stéphanie Iannuzzi, Anne-Isabelle Bertozzi,

Fatima Tensaouti, Yves Chaix, Anne Laprie
PII: S0167-8140(22)02278-2
DOI: https://doi.org/10.1016/j.radonc.2022.05.022
Reference: RADION 9254
To appear in: Radiotherapy and Oncology
Received Date: 21 December 2021

Revised Date: 10 May 2022
Accepted Date: 26 May 2022
Please cite this article as: Baudou, E., Pollidoro, L., Iannuzzi, S., Bertozzi, A-I., Tensaouti, F., Chaix, Y., Laprie,
A., A review of long-term deficits in memory systems following radiotherapy for pediatric posterior fossa tumor,
Radiotherapy and Oncology (2022), doi: https://doi.org/10.1016/j.radonc.2022.05.022
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2022 Published by Elsevier B.V.
Descargado para Anonymous User (n/a) en La Paz University Hospital de ClinicalKey.es por Elsevier en mayo 31, 2022. Para uso
personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2022. Elsevier Inc. Todos los derechos reservados.
A review of long-term deficits in memory systems following radiotherapy for pediatric posterior fossa tumor
Short title: Pediatric posterior fossa and memory
Eloïse Baudou1,2, Lisa Pollidoro1,2, Stéphanie Iannuzzi1, Anne-Isabelle Bertozzi3, Fatima Tensaouti4,2, Yves
Chaix1,2, Anne Laprie4,2.
1 Pediatric Neurology Department, Children's Hospital, Toulouse University Hospital, Toulouse, France
2 Toulouse NeuroImaging Center (ToNIC), INSERM University of Toulouse Paul Sabatier, Toulouse, France
3 Pediatric Oncology Department, Children's Hospital, Toulouse University Hospital, Toulouse, France
4 Radiation Oncology Department, Institut Claudius Regaud- Institut Universitaire du Cancer de Toulouse-
Oncopole, Toulouse, France.
Corresponding author: Eloïse BAUDOU, Pediatric Neurology Unit, Hôpital des Enfants, CHU Toulouse, 330 av
de Grande Bretagne-TSA, 31059 Toulouse Cedex, France [email protected]
Highlights
- All memory systems are impaired in irradiated PFT survivors.

- Memory systems are more mildly affected in non-irradiated PFT survivors.
- Progress in radiotherapy has led to a reduction in long-term memory sequelae.
- Further investigation of procedural and episodic memory is warranted.
Abstract:
Introduction: In recent years, progress in pediatric posterior fossa tumor (PFT) treatments has improved survival
rates. However, the majority of survivors present neurocognitive sequelae that impact academic achievement.
Methods: This review examines the literature from 2000 to 2020 on long-term outcomes in different memory
systems for survivors of pediatric PFT, considering the impact of radiotherapy which is a well-known prognostic
factor for global neurocognitive function.
Results: Of the 43 articles selected, 31 explored working memory, 19 episodic memory, 9 semantic memory and
2 procedural memory. Irradiated survivors had scores of < -2 standard deviation (SD) (n = 4 studies/25) or between
-2SD and -1SD (n =7 studies/25) for working memory; < -1SD for anterograde memory (n = 11/13), with a
progressive decline in these two memory systems; < -1SD (n = 4/7) in semantic memory, and a deficit in
perceptual-motor procedural learning (n = 1/1). Reducing craniospinal irradiation dose, limiting tumor bed boosts,
and using proton therapy seem to have had a beneficial effect with better preservation of the memory score and a
reduction in the decline over time. Non-irradiated survivors had memory systems that were less affected, with
preservation of anterograde memory and maintenance of long-term stability.
Conclusion: Memory deficits are a core feature in survivors of pediatric PFT, especially when treatment requires
radiotherapy. To limit these effects, dose constraints for specific brain areas involved in memory should be defined.
During long-term follow-up, specific attention is essential to identify these deficits in order to limit their impact
on the quality of life.
Keywords: Posterior fossa tumors, neuropsychological outcomes, memory disorders, children, radiotherapy
1- Introduction:
Two-thirds of the central nervous system tumors, the most common solid neoplasm in children, occur in the
posterior fossa. With improvements in surgical technologies, chemotherapy and radiotherapy protocols over the
past decades, the 5-year survival rate for children with a posterior fossa tumor (PFT) have increased for the main
histological types of tumors diagnosed during childhood. The survival rate is 90% for astrocytoma (30% of PFT),
60% for ependymoma (10% of PFT), and 80% for medulloblastoma (40% of PFT). Treatments differ from one
histological form to another, currently including only surgery for astrocytomas, surgery and focal radiotherapy for
ependymomas and a combination of surgery, craniospinal radiotherapy and chemotherapy depending on the risk
factors and age for medulloblastomas.
Medical follow-up of patients with pediatric PFT showed a frequent occurrence of global and specific
neurocognitive disorders that impact academic achievement and professional integration. Many studies have
focused on changes in intellectual quotient (IQ), an age-adjusted composite index of several neuropsychological
processes, showing lower mean scores and a progressive decline in IQ over time for patients with medulloblastoma
[1]. Among the factors that cause late effects (tumor itself, hydrocephalus, surgery, chemotherapy), the most
significant for these neuropsychological sequalae is probably the whole brain irradiation dose. Recently, authors
have identified more specific neurocognitive disorders in PFT patients that affect learning, memory, processing
speed, attention, and executive function [2]. Palmer et al. showed that this impairment is not the consequence of a
loss of anterior learning but rather slowness in learning new skills [3].
Memory is a highly complex cognitive process that develops through childhood. Multiple systems are involved,
which concern short-term memory (working memory) or long-term memory (episodic, semantic and procedural
memory). These systems, which are often studied in isolation, constantly interact, based on the neuropsychology
of memory in the Memory Neo-Structural Inter-Systemic model (MNESIS) [4], and involve distinct and common
brain areas and neural circuits that can be affected in PFT (Table 1). While the involvement of the hippocampus
and parahippocampal regions in long-term episodic memory is well known [5], brain structures and networks
involved in memory systems are larger. The cerebellum and cortico-cerebellar circuits in particular, are both
involved in working memory (cerebellar posterior lobe (lobules VII and VIII), prefrontal and parietal cortices)
[6,7] and procedural memory (left lobules V & VI, dentate nuclei) [8]. Moreover, the cerebellum may also
participate in episodic memory, both in encoding [9] and retrieval [10]. Knowledge of the role of the cerebellum
in memory is currently limited [11], but some authors suggest that the internal predictive model initially developed
for motor skills could be applied to cognitive and memory skills [12,13]. Based on this idea, it can be hypothesized
that the localization of a tumor in the posterior fossa and the impact of surgery on the cerebellum and on afferent
and efferent tracts alters memory, especially working memory and procedural memory. Moreover, the
consequences of a tumor (such as hydrocephaly) and its treatment (chemotherapy, radiotherapy) on the
supratentorial brain parenchyma of PFT survivors could alter the different memory systems.
---- Insert Table 1-----
References in table 1: [5–8]
Except for infants, patients with malignant tumors receive multimodal treatment including radiotherapy. At
present, ependymoma radiotherapy is limited to the tumor bed with high doses ranging between 54 to 59.4 Gy.
Due to the high risk of craniospinal dissemination, medulloblastoma is treated by craniospinal irradiation (18 to
36 Gy depending on the tumor risk group) followed by a localized boost of up to a total of 54 Gy. The craniospinal
irradiation dose is an important factor in the pathogenesis of global neurocognitive sequelae, including specific
cognitive functions such as memory. One current challenge is to find the optimal balance between sufficient
irradiation to cure the patient while limiting the impact of irradiation by reducing the cranio-spinal dose of
radiotherapy and/or avoiding or limiting the dose to the normal brain.
In the past two decades, radiotherapy techniques used to manage the two main, malignant PFTs (medulloblastoma
and ependymoma) have improved. The changes in these techniques and their impact on therapeutic protocols are
indicated in figure 1. The common aim of these improvements is to limit the irradiation dose to the normal brain.
Knowledge of threshold doses with deleterious effects on the different brain structures helps to guide these
advances. However, these threshold doses are not known for the structures involved in memory development in
children and their determination is vital.
---- Insert Figure 1-----
The physiopathology of radiotherapy-induced brain lesions is better understood as a result of MRI studies. Brain
damage occurs early after radiotherapy (< 1 month), especially in white matter, with an alteration in diffusion
tensor imaging MRI that has been linked to cognitive deficit [11]. Damage to neuronal dendritic spines, neuronal
metabolic changes, oligodendrocyte injury and loss, neuroinflammation and vascular endothelial damage are early
damages that persist and change the signaling microenvironment in neuronal tissue, leading to a loss of cognitive
function and memory. The hippocampus, which is involved in episodic memory, seems to be particularly
vulnerable to irradiation with a specific impact on the proliferation of hippocampal subgranular zone progenitor
cells and their differentiation into neurons [11]. These damages largely depend on the irradiation dose.
Radiotherapy has been implicated in learning and memory impairments observed in brain tumor patients with an
impact of radiotherapy dose on the hippocampus [12,13]. The irradiation dose to the left hippocampus is linked
with difficulties in verbal learning and memory in adults [14] and children with infra- and supratentorial brain
tumors [15,16]. Some authors have built hippocampal dose volume histograms to predict verbal learning scores
after brain irradiation in adults [17]. These data are sparse and not widely replicated.
The aim of this work is to review the literature that reports on actual knowledge of long-term performances in the
different memory systems after PFT treated during childhood, detailing the impact of radiotherapy on the posterior
fossa and the entire developing brain.
2- Methods:
2.1- Eligibility criteria
To be eligible for this literature review, studies had to meet the following criteria: (1) patients had to be treated
for a PFT, or if supra- and infratentorial tumors were included, specific data had to be available for the
infratentorial group, (2) treatment had to be administered before the age of 18 years, (3) there had to be at least
one neuropsychological task used to assess memory, (4) the mean period between tumor treatment and
neuropsychological assessment had to be at least 3 years, (5) patients treated with radiotherapy were analyzed
separately from patients treated without, and (6) the study had to be published in English between 2000 and
2020.
2.2- Search strategy
A literature search was performed using the PubMed search engine. Combinations of the terms “Infratentorial
Neoplasms”, “Cerebellar Brain Tumors” and “Episodic Memory”, “Working Memory”, “Procedural Memory”,
and “Semantic Memory” were used. We then completed with a wider search including the terms “neuropsych*”
and “cognition” and we selected articles in which neuropsychological tests that assess memory were used.
References cited in relevant articles were also searched as an additional resource for articles. Case reports and case
series were excluded from the search.
2.3- Data extraction
From each study, the following data were extracted for the groups: number of participants, tumor type, age (mean,
standard deviation, range) and the period between tumor treatment and neuropsychological assessment. The
number of participants included in the control group, if any, was also reported.
The neuropsychological tests were numerous. Therefore, in order to clarify the results, we classified them: (1)
according to the different memory systems (working, episodic, semantic and procedural memory) and (2) for each
memory system, according to the memory processes involved. Table 2 summarizes memory tasks used in this
literature review according to the memory system and memory function assessed.
Depending on the article, the results of the neuropsychological tests is presented heterogeneously as raw,
standardized or z-score data. Moreover, the term "impaired" is sometimes used for scores < -1SD, < -1.65SD or <
-2SD. For simplicity, we associated each score with: a normal range (between -1SD and +1SD) which does not
reflect a deficit in memory performance; a low range (between -1 and -2SD) which represents poor memory
performance that can lead to repercussions on academic learning and everyday life; or an impaired range (< -2SD)
which reflects a deficit in memory performance with a memory disorder in everyday life.
Considering that the techniques and recommendations concerning radiotherapy have developed over the last
20 years, the data concerning these techniques, craniospinal irradiation dose and boost dose were specified when
they were available in the original articles.
Others prognostic factors for memory were also extracted based on neurocognitive prognostic factors reported in
the literature: age at treatment, length of time since treatment, hydrocephalus or the need for a shunt, tumor
characteristics (volume and localization), surgical complications and medical events, posterior fossa syndromes
and socioeconomic status.
Finally, we extracted neuroimaging data when available, for a better understanding of neural substrates of memory
functions in PFT survivors.
3- Results:
Forty-three studies met the eligibility criteria: 31 assessed working memory, 19 episodic memory, 9 semantic
memory and 2 procedural memory. See Figure 2 for a flow diagram of the study selection process. Details of
studies included are summarized in Table 3.
Nineteen studies reported memory data on PFT survivors treated without radiotherapy (pilocytic astrocytomas
or low-grade tumors). Most of the studies presented a relatively preserved verbal [18,19,28,20–27] and visual
[18,19,22,29,30] working memory load, regardless of the period between diagnosis and assessment in non-
irradiated patients. Concerning updating, studies showed a non-significant difference compared to the norm [28]
and/or the control group [21,28,31]. Studies reported results for preserved memory on the verbal and visual or
global score for anterograde memory in learning, long-delay free recall and recognition [20–22,25,28,32–35].
Aarsen et al. showed that in comparison to patients with supratentorial astrocytomas, patients with infratentorial
astrocytomas had lower scores for verbal memory and visuo-spatial memory. Studies that explored semantic
memory reported scores within the normal [21,22,25] or low range [26] in non-irradiated patients. Finally, only
two studies focused on procedural memory using a serial reaction time (SRT) task [33,36]. The authors showed
preserved learning scores with results similar to the control in Berger’s study. However, Quintero et al. found an
alteration in motor sequence learning with only a trend for statistical significance in a small sample of eleven non-
irradiated participants.
In summary, non-irradiated patients treated at least 3 years before for PFT, mainly showed results in the different
memory tasks that were between -1SD and the norm, stable over time and lower than the results of the control
subjects but not significantly. Although the results of the evaluation of semantic and procedural memory systems
should be viewed with caution due to the small number of studies and patients per study, it would appear that the
performance of non-irradiated patients is lower in cerebellum-dependent working and procedural memory and in
semantic memory, than in hippocampal-dependent anterograde memory.
Thirty-three studies included patients treated for high-grade PFT with radiotherapy. Radiotherapy
characteristics are indicated in supplementary data 1.
Studies exploring verbal working memory load in patients treated for medulloblastoma with PFB and CSI
(standard and reduced dose), reported scores in the normal [18,23,24,26,37–41], low [42–45] or deficit range
[25,28,37,46]. Studies showing a deficit score concerned children who were younger at the time of treatment [25]
or who had a higher mean time since treatment [28,46]. Differences were also linked to tests used. Impairment
was more significant with the backward than the forward digit span [37,46]. The two main findings of these studies
were that (1) irradiated patients performed significantly worse compared to non-irradiated patients [24,25,28] and
(2) in irradiated patients, the change in the working memory index showed a decline of approximately 2 points per
year [38,39,41,44]. Among patients treated for medulloblastoma with posterior fossa boost and CSI, patients
treated with a reduced dose of CSI had better results (lower range) compared to those treated with a standard dose
[42]. Studies that explored verbal working memory load in patients treated for medulloblastoma with a tumor bed
boost and adapted-risk CSI, reported scores in the normal [47–50] or low range for the posterior fossa syndrome
group [49] and in the standard dose CSI group [50]. Changes in the working memory index showed a decline of
approximately 2.2 points per year for standard CSI, while the reduced CSI dose improved relative stability with a
decrease of -0.2 points per year [50]. Knight et al. found a decline in the working memory index of -0.93 points
per year for standard and reduced dose CSI. In medulloblastoma treated with hyperfractionated radiation therapy,
Camara et al. found no significant difference in the working memory index (WISC/WAIS) compared to standard
radiation therapy, with results within the normal range for both groups. In medulloblastoma treated with risk-
adapted CSI with photon therapy and tumor bed boost (TBB) with proton therapy, Kahalley et al. found a working
memory index within the normal range with score stability over time (+0.1 point per year), while a decline of 2.2
points was reported in the photon therapy group. In ependymoma treated with TBB or posterior fossa boost (PFB),
von Hoff et al. and Zapotocky et al. found a working memory index within the normal range with relative stability
over time (-0.56 points per year in Zapotocky’s study). Only one study on ependymoma treated with TBB reported
mean scores within the normal range but the population was heterogeneous as 3 participants out of 19 had deficit
scores < -2SD. Studies that explored visual working memory in irradiated patients reported scores in the low
[29,30] or deficit range [18] using either forward or backward Wechsler block tasks in patients treated with risk-
adapted CSI. Irradiated patients were both slower and more inaccurate than non-irradiated patients and differences
between groups increased when working memory load was higher [30]. In medulloblastoma, Rønning et al. found
updating scores in the deficit range using PASAT, with a significant difference in non-irradiated patients. Hoang
et al. showed a greater difference between the medulloblastoma and the control group in the 2-back than the 1-
back tasks, both in accuracy and speed, and found an association between left posterior cerebellar lobe lesions and
working memory impairment.
Only one neuroimaging and behavioral study was published on episodic memory and its neural substrates in
posterior fossa medulloblastomas. Sekeres et al. evaluated autobiographical memory using the Children’s
Autobiographical Interview and assessed episodic and non-episodic details for events that either preceded (i.e.,
remote) or followed (i.e., recent) treatment [51]. The authors highlighted episodic memory preservation before
treatment, with equivalent episodic details in the PFT group compared to control subjects. However, they showed
an alteration in episodic memory after treatment, with fewer episodic details of post-treatment events in the PFT
group compared to the control subjects, which is consistent with a lower score on the Children’s Memory scale
(CMS). This suggests an alteration in anterograde memory and preservation of retrograde memory. Neuroimaging
results are discussed in the neuroimaging section of this review.
Studies that explored verbal anterograde memory in patients treated for medulloblastoma with posterior fossa
boost and CSI (standard and reduced dose), reported scores in the normal [35,38,52] or low range [28,37,46,53–
55]. Lower scores were found in studies that included children who were younger at the time of treatment, with a
higher mean time since treatment (respectively 14, 17 and 18 years old [28,46,54]) and a lower proportion of
reduced CSI doses or boosts limited to the tumor bed. Two findings were of interest in these studies: (1) irradiated
patients performed significantly worse than non-irradiated patients [28,35] and (2) in the irradiated patients,
changes in the scaled score showed relative stability in learning, long-delay free recall and the global verbal score
[38]. Among patients treated for medulloblastoma with a posterior fossa boost and CSI, those treated with a
reduced dose of CSI had lower learning scores and deficit scores in long-delay free recall, while patients treated
with a standard dose had scores in the deficit range [42]. Only one study was conducted on ependymoma treated
with a TBB and reported mean scores within the normal range but the population was heterogeneous as 3
participants out of 19 had deficit scores < -2SD.
Few studies explored visual anterograde memory in patients treated for medulloblastoma with PFB and CSI
(standard and reduced dose), and they reported scores within the normal [38], low [25,28,37,52,55] or deficit range
[54]. Two findings were of interest in these studies: (1) irradiated patients performed significantly worse than non-
irradiated patients [25,28] and (2) in irradiated patients, changes in the visual composite score on the Children’s
Auditory Verbal Learning Test showed a progressive decline in visual scores of -1.54 points per year [38]. This is
consistent with the fact that studies that assessed memory at a longer time after radiotherapy found more impaired
scores [11], [35]. In several visual memory assessment tests, Rønning et al. found lower results in the delayed
copy of the Rey figure than in the total recognition and learning of the Continuous Visual Memory Test. Among
patients treated for medulloblastoma with PFB and CSI, those treated with a reduced dose of CSI had non-
significantly higher scores than patients treated with a standard dose [42]. Global anterograde memory (including
both verbal and visual subtests) was in the lower range for patients irradiated by CSI with PFB or TBB
[46,52,55,56].
Results of studies were contradictory concerning memory process impairment in PFT survivors. Maddrey et al.
(2005) showed better performance in short and long delay recall, compared to immediate recall, suggesting an
encoding defect. Therefore, memory problems may be the result of information learning or encoding difficulties,
rather than information storage or retrieval difficulties. Quintero-Gallego et al. (2006) qualitatively highlighted
that in a medulloblastoma group, learning remained within the limits for a normal population. Khajuria et al. (2015)
found a tendency for medulloblastoma patients to learn less in the same period than astrocytoma patients. However,
Kieffer-Renaux et al. (2000) found that delayed recall was slightly more deficient than immediate recall in
medulloblastomas suggesting a storage or retrieval deficit.
Studies that explored semantic memory in patients treated for medulloblastoma with PFB and CSI (standard and
reduced dose), reported scores in the normal [37,53] or low range [3,25]. Differences in scores between these
studies may be explained by an older mean age at the time of treatment (8 and 8.5 yo vs. 4.6 and 4 yo) in both
studies with scores in the normal range. Two findings were of interest in these studies: (1) irradiated patients
performed significantly worse than non-irradiated patients [25] and (2) in irradiated patients, the change in
information scaled score significantly declined by 0.41 points per year, and was more pronounced in patient who
were younger at the time of treatment (<8.02 yo) than in those who were older (-0.53 vs. -0.17, respectively)[3].
Among patients treated for medulloblastoma with PFB and CSI, patients treated with a reduced dose of CSI
showed better results (normal or lower range scores depending on the tests used) vs. the standard dose group (lower
or deficit range) [42]. Moreover, in a study of only patients who received a standard CSI dose, Hazin et al. found
that results were more altered than in other studies with scores in the deficit range. Only one study focused on
ependymoma treated with a TBB and reported mean scores in the normal range but the population was
heterogeneous as 2 participants out of 23 had deficit scores < -2SD.
Only one study explored procedural memory using SRTT in a medulloblastoma group of 7 patients treated with
radiotherapy for which we do not have details of the type and doses of radiotherapy. They had a significantly lower
number of correct responses than the control group and motor sequence learning was absent.
In summary, irradiated patients showed worse results, with an impact on all the memory systems. However, these
results differ according to the type of irradiation received. Patients treated with standard-risk CSI and PFB had
scores in the lower range in the different memory systems, and a decline over time especially in working memory
and in visual anterograde memory. Reducing CSI dose, limiting TBB, and using proton therapy seem to have had
a beneficial effect with better preservation of the memory score and a reduction in decline over time.
Neuropsychological scores in different memory systems are summarized in figure 3 and are detailed in
Supplementary data 2.
Studies reported few prognostic factors of memory impairment. Radiotherapy is undoubtedly the main prognostic
factor for secondary memory impairment as shown in the previous section [18,23–26,28,30,33,35].
A younger age at the time of treatment was linked with a poor memory prognosis in the medulloblastoma group,
as shown by a comparison between children treated before 8 years old and those treated after in most of the studies
[37,53]. However, in non-irradiated patients, the impact of age at the time of treatment is controversial. Roncadin
et al. found poorer memory in younger patients [56] while Rønning et al. showed better results, which they linked
to more plasticity in the early stage of development [28]. In an astrocytoma population, Steilin et al. compared
preschool children (3.5-6.5 yo), elementary school children (7-9.5 yo) and middle school children (10-15.5 yo)
and found that elementary school children were most affected in terms of verbal anterograde memory and semantic
memory [22] than the other categories.
Concerning hydrocephalus, Rønning et al. negatively correlated neuropsychological effects with shunts in an
astrocytoma group but not in a medulloblastoma group [28]. In a medulloblastoma population, Hardy et al. also
found that the presence of hydrocephalus requiring the placement of a ventriculoperitoneal shunt was associated
with more severe intellectual and academic deficits but not with lower scores in working or anterograde memory
[55].
Whether tumor size is a risk factor is controversial. Khajuria et al. found that astrocytoma patients with a larger
cerebellar lesion had significantly decreased verbal memory performances in learning but not medulloblastoma
patients [35]. However, Steinlin et al. showed that tumor size had no influence on outcome in astrocytoma [22].
In an astrocytoma population, Aarsen et al. found a correlation between maximum tumor diameter and long-term
verbal memory, but this is probably related to hydrocephalus since a correlation was also found between maximum
tumor diameter and ventricular dilatation [32].
Undoubtedly, surgical complications impact memory outcomes. In an astrocytoma population, Roncadin et al.
found that poorer memory was predicted by a higher number of medical events in the first 5 years after surgery
[56]. Posterior fossa syndrome (PFS) was a factor in poorer neuropsychological outcomes with working memory
performances approximately 1SD below the performances of children without PFS [47,49]. Studies differed in
terms of the long-term changes, with some showing an increase in the decline in working memory compared to
non PFS medulloblastomas [49], or parallel changes [47].
Lastly, Khalil et al. linked socioeconomic status in a low-income Moroccan medulloblastoma population, with
global neurocognitive performance below -2SD and a low range for working memory [45].
In summary, radiotherapy is the most important prognostic factor for memory. In non-irradiated patients, more
refined prognostic factors could modulate the results, but reproducibility between studies is not always achieved.
Finally, concerning memory neuroimaging in PFT survivors, only six of the studies included used MRI to explore
the neural substrates of memory after radiotherapy, but none included an analysis of the radiotherapy doses
delivered to brain structures.
Sekeres et al. explored hippocampal volumes, white matter tracts involved in episodic memory (fornix) and five
cortical regions of recollection networks [51]. A significant difference was found, with smaller hippocampal,
fornix, precuneus and lateral temporal cortex volumes in the radiotherapy group compared to the control group.
Probably because of the small sample, no correlation was found between hippocampal volume and episodic
memory scores. Based on rodent models, the authors hypothesized that radiotherapy acted by suppressing
hippocampal neurogenesis and thereby produced dissociable anterograde versus retrograde effects on memory. No
studies were found on cerebellar macrostructural correlates of memory.
Diffusion Tensor Imaging (DTI) is a magnetic resonance neuroimaging technique based on the detection of the
diffusion of water molecules along the main direction of axons and myelin sheaths. This enables the estimation of
microstructural connectivity based on location, orientation, and anisotropy of the white matter tracts of the brain.
Two parameters are commonly used: mean diffusivity (MD) which reflects microstructural integrity and fractional
anisotropy (FA) which reflects microstructural orientation. Using DTI in adult survivors of pediatric
medulloblastoma, Brinkman et al. showed that FA in the parietal lobe was positively correlated with working
memory scores and in the right hemisphere and bilateral temporal lobes with visual memory scores [46]. In a
medulloblastoma group, Law et al. showed higher indices of diffusivity (FA, MD, and axial and radial diffusivity),
especially in the left cerebello-thalamo-cortical pathway related to variances in working memory outcome [29].
Riggs et al. explored the uncinate fasciculus using DTI and the size of the hippocampus in medulloblastomas [52].
They showed that the PFT group had a significantly lower FA on the bilateral uncinate fasciculus and significantly
smaller right hippocampal volumes compared to healthy controls. Medulloblastoma survivors had significantly
lower performances on the general Children’s Memory Scale index that are correlated with the FA of the left
uncinate fasciculus and the right hippocampal volume.
fMRI is a magnetic resonance neuroimaging technique based on the detection of fluctuations in blood oxygenation
level-dependent (BOLD) signals during a task that indirectly reflects neuronal activation. Using a N-back task in
eleven surgically treated low-grade PFTs compared to control, Reitchert et al. (2017) found that most group
differences in functional connectivity were observed in the least cognitively demanding tasks. Hoang et al. (2019)
used an experimental fMRI N-back task in children treated for PFT medulloblastoma and failed to find any
cerebellar activation in this group. The explanatory hypothesis put forward by the authors for the difficulties in
demonstrating cerebellar activation in PFT survivors, was a small sample size (low statistical power), fMRI
limitations to investigating the postoperative brain harboring magnetic susceptibility artifacts, anatomical
deformities of the posterior fossa uncompensated by the anatomical normalization and inter-individual spatially
significant activation variability.
4- Discussion:
The majority of studies show an alteration of all the memory systems in children who have received CSI, whereas
non-irradiated children have a lesser impairment, placing them below the performance of typically developing
children, but remaining mostly within the norm, except for working and procedural memory, whose neural
substrates depend in part on the cerebellum.
In non-irradiated children, despite a relative preservation of memory systems at the group level, individual
impairment could be seen especially when risk factors were present. This impairment could be underestimated for
procedural memory due to the fact that a repeated sequence learning task, which mainly involves the cortico-
striatal circuit, was used in the studies examined. We hypothesize that tests to explore motor adaptation such as
mirror writing would indicate more impaired results since they explore cerebello-cortical circuits. Interestingly,
scores for semantic and episodic memory were also lower, although treatment did not include chemotherapy or
radiotherapy which affect infratentorial areas. Few hypotheses could be formulated to explain these results. Firstly,
PFT and PFT surgery could alter cerebello-cortical networks involving brain areas concerned with episodic
memory. Secondly, preoperative hydrocephalus or postoperative complications such as meningitis could have an
impact on infratentorial area white matter microstructure [20]. Lastly, the different memory systems are
interconnected (MNESIS model), which could explain why some memory systems that do not directly depend on
cerebellar areas could be impaired by a PFT tumor. Neuroimaging studies suggest that the tumor has an impact on
left posterior localization and damages dentate nuclei in visual working memory [57]. But most studies failed to
highlight the relationship between specific memory impairment and tumor localization.
Radiotherapy has a major impact on all memory system outcomes in PFT survivors. In the past two decades,
progress in the global management of FPT tumor and especially the improvement in radiotherapy techniques has
allowed partial preservation of the memory process, limiting posterior fossa irradiation to the tumor bed and using
intensity-modulated radiation therapy or proton therapy. The addition of chemotherapy for chemosensitive tumors
(medulloblastoma) enabled the reduction of the CSI dose in patients with an average risk (non-metastatic disease
and no histological high-risk factors) at diagnosis with a clear beneficial impact on all neuropsychological
outcomes. At present, treatment protocols for children with medulloblastoma tend to become more and more
complex according to age, risk and the molecular biology of the tumor, and the dual aim is to improve survival
and limit the long-term sequelae of these treatments. Another way to limit these effects is to set dose constraints
for the specific brain areas involved in memory, such as the medial temporal lobe for episodic memory, and to
adapt conformational radiotherapy to avoid these areas as is already done for the pituitary to avoid endocrinal
sequelae or the optic chiasma to limit visual function impairment. Less is known about the impact of concomitant
treatment for medulloblastoma on the entire brain, with a potentially negative effect of chemotherapy, or a synergic
neurotoxic effect of concomitant administration of radiotherapy and chemotherapy [58,59], and medications such
as steroids [60].
In contrast to non-irradiated patients, longitudinal studies of patients treated with CSI show a progressive decline
in intellectual and memory functions over time [38,39,41,44] with a rapid decline in performance in the first 5 years
and slower decline thereafter [38]. Depending on the patient’s initial level of performance, this may place them in
a deficit or low performance zone and have an impact on their general life and school functioning.
In terms of neuroimaging, MRI helps to investigate the damage caused by radiotherapy on normal-appearing brain
in vivo, especially in infratentorial structures. CSI patients have both lower cerebral volumes and an alteration in
the microstructure of the brain area involved in memory.
It is actually well-established that a young age at the time of radiotherapy is a poor prognosis for IQ scores and
memory, with evolutive damage to brain structures causing progressive neuropsychological decline. Impairments
in memory acquired after radiotherapy treatment but not in memories acquired before [51] confirm that PFT
survivors have difficulties learning new skills, but no alteration in abilities acquired before tumor treatment [3].
Concerning non-irradiated patients, brain lesions caused by a tumor and tumor surgery can be considered as fixed,
with the possibility to improve over time thanks to neuroplasticity stimulated by rehabilitation. The impact of age
at the time of treatment and surgery is not so clear, with studies showing a worse prognosis for a young age at the
time of treatment [56] and other studies highlighting a worse prognosis when tumor surgery occurred at school
age (probably because it is a critical time to acquire general knowledge). This suggests that there should be specific
care for younger and school-age patients.
Results concerning the long term-memory process that is impacted in the pediatric population varies, and includes
deficits in learning [33,35,54] or storage and retrieval [42]. In adult patients, retrieval is the process that is most
affected [58]. Indeed, Durand et al. (2018) showed predominant impairment in retrieval (92%) compared to storage
(41%) or encoding (23%). These studies indicate the need to use tests to assess the different forms of antegrade
memory in order to gain a broader perspective of the processes that are affected.
Because of the high frequency of memory impairment, particular attention and systematic assessment of children
with risk factors are needed during long-term follow-up of PFT survivors. The main risks factors are radiation
therapy, neurologic complications, hydrocephalus, PFS and an age below 8 years at the time of treatment.
However, disease and treatment are not the only neurocognitive risk factors and patient characteristics and the
environmental context are also important to consider [61]. Assessment of prior neurocognitive and academic skills
and of the socio-familial context is important for neurocognitive management of these patients [62]. In fact,
cognitive decline in a child with good initial skills, good school integration and a supportive environment will
have less impact than in a child who was previously limited, poorly integrated or poorly supported [63]. The results
of neuropsychological tests carried out early during the first months of medical care are often difficult to interpret
because of the child's fatigue and the psychological context of a serious illness that may be life-threatening. An
alternative could be to identify elements of psychomotor, school and socio-familial development when
interviewing the child and their family and to systematically include a psychologist and a social worker in the care
in order to detect children at risk and to support their families.
Lastly, Various tests have been used in the literature which may partially explain variations in memory results.
Moreover, in some memory systems such as procedural memory, few tests are available. An effort should be made
to establish a panel of tests that are not too time-consuming, in several languages, and that can be easily carried
out in current practice and in world-wide research protocols to make results comparable.
This review has some limits. Firstly, memory is a broad and complex field and, to our knowledge, no study has
reported on assessments of all memory systems in the same population and which provides an overview of memory
deficit. Moreover, the multiplicity of tests used sometimes makes the interpretation and comparison of studies
difficult. As raw data, normalized data or z-scores used to present the results in articles are not directly comparable,
we transformed data into z-scores. However, this transformation does not make the data totally equivalent. In fact,
the normalized data include age correction which is not present in the raw data, and the z-transformation of raw-
data could have changed the magnitude of impairment. In addition to this heterogeneity, there is also the
multiplicity of protocols used over the past two decades, with different surgical and radiotherapy techniques and
various chemotherapies, as well as the heterogeneity of the population (ie. Including PFS who have a more
impaired neuropsychological profile. Finally, to date most studies have a small sample size due to the low number
of cases and difficulties in the long-term follow-up of patients, or do not include a control group.
5- Conclusion
All the studies are consistently show that PFT survivors have lower performances in all memory systems. Although
these scores are slightly lower than in the control and stay within the normal or low ranges in non-irradiated
patients, they are often in the low or deficit range for irradiated patients. Irradiation techniques that reduce doses
and avoid normal brain show better preservation of memory performances. There is a need to set dose constraints
on the specific brain areas involved in memory to further reduce the impact of radiotherapy on learning and
memory. We are currently conducting a prospective study to develop methods to define theses dose constraints
[66]. However, radiotherapy is not the only prognostic factor. Age at the time of treatment, hydrocephalus, surgical
complications, posterior fossa syndrome or socio-economic status can also affect the cognitive prognosis. During
the long-term follow-up of children treated for a PFT, specific attention is essential to identify learning and
memory deficits, and to address them at an early stage and adapt schooling in order to assist children individually
to achieve autonomy in their future adult life.
Declarations
Ethical approval and consent to participate
Not applicable
Consent for publication
Not applicable.
Availability of data and materials
Data sharing is not applicable to this article.
Competing interests
No competing interests
Funding and acknowledgements
This work was carried out within the framework of a neuroscience thesis financially supported by the following
associations that we would like to thank: “111 des Arts”, “TREC”, “Fondation de l’Avenir” and “Timeo mon
heros”.
Authors' contributions
EB conceived the methodology, selected articles, wrote the first draft of this manuscript and prepared the
figures and tables. LP selected and classified neuropsychological tests. LP, SI, AIB and YC reviewed the draft
with specific attention to the neuropsychological part. LP and SI are neuropsychologists and EB and YC are
pediatric neurologists who specialize in learning disorders in children. FT reviewed the draft with specific
attention to the neuro-imagery part. AL reviewed the draft with specific attention to the radiotherapy part. All
the authors read and approved the final manuscript.
Bibliography:
[1] Hanzlik E, Woodrome SE, Abdel-Baki M, Geller TJ, Elbabaa SK. A systematic review of
neuropsychological outcomes following posterior fossa tumor surgery in children. Child’s Nerv Syst
2015;31:1869–75. https://doi.org/10.1007/s00381-015-2867-3.
[2] Palmer SL, Reddick WE, Gajjar A. Understanding the cognitive impact on children who are treated for
medulloblastoma. J Pediatr Psychol 2007;32:1040–9. https://doi.org/10.1093/jpepsy/jsl056.
[3] Palmer SL, Goloubeva O, Reddick WE, Glass JO, Gajjar A, Kun L, et al. Patterns of intellectual
development among survivors of pediatric medulloblastoma: A longitudinal analysis. J Clin Oncol
2001;19:2302–8. https://doi.org/10.1200/JCO.2001.19.8.2302.
[4] Eustache F, Desgranges B. MNESIS: Towards the integration of current multisystem models of memory.
Neuropsychol Rev 2008;18:53–69. https://doi.org/10.1007/s11065-008-9052-3.
[5] Barrouillet P. Le développement de la mémoire de travail 2018.
[6] Guillery-Girard B, Martins S, Deshayes S, Hertz-Pannier L, Chiron C, Jambaqué I, et al. Developmental

trajectories of associative memory from childhood to adulthood: A behavioral and neuroimaging study.
Front Behav Neurosci 2013;7:1–12. https://doi.org/10.3389/fnbeh.2013.00126.
[7] Ofen N, Shing YL. From perception to memory: Changes in memory systems across the lifespan.
Neurosci Biobehav Rev 2013;37:2258–67. https://doi.org/10.1016/j.neubiorev.2013.04.006.
[8] Zwart FS, Vissers CTWM, Kessels RPC, Maes JHR. Procedural learning across the lifespan: A
systematic review with implications for atypical development. J Neuropsychol 2019;13:149–82.
https://doi.org/10.1111/jnp.12139.
[9] Schmahmann J. The cerebellar cognitive affective syndrome. Brain 1998;121:561–79.

https://doi.org/10.1093/brain/121.4.561.
[10] O’Halloran CJ, Kinsella GJ, Storey E. The cerebellum and neuropsychological functioning: A critical
review. J Clin Exp Neuropsychol 2012;34:35–56. https://doi.org/10.1080/13803395.2011.614599.
[11] Makale MT, McDonald CR, Hattangadi-Gluth JA, Kesari S. Mechanisms of radiotherapy-associated
cognitive disability in patients with brain tumours. Nat Rev Neurol 2016;13:52–64.
https://doi.org/10.1038/nrneurol.2016.185.
[12] Gondi V, Tomé WA MM. Why avoid the hippocampus? Radiother Oncol 2011;97:370–6.
https://doi.org/10.1016/j.radonc.2010.09.013.Why.
[13] Tsai PF, Yang CC, Chuang CC, Huang TY, Wu YM, Pai PC, et al. Hippocampal dosimetry correlates
with the change in neurocognitive function after hippocampal sparing during whole brain radiotherapy:
A prospective study. Radiat Oncol 2015;10:1–15. https://doi.org/10.1186/s13014-015-0562-x.
[14] Haldbo-Classen L, Amidi A, Lukacova S, Wu LM, Oettingen G von, Lassen-Ramshad Y, et al.
Cognitive impairment following radiation to hippocampus and other brain structures in adults with
primary brain tumours. Radiother Oncol 2020;148:1–7. https://doi.org/10.1016/j.radonc.2020.03.023.
[15] Acharya S, Wu S, Ashford JM, Tinkle CL, Lucas JT, Qaddoumi I, et al. Association between
hippocampal dose and memory in survivors of childhood or adolescent low-grade glioma: A 10-year
neurocognitive longitudinal study. Neuro Oncol 2019;21:1175–83.
https://doi.org/10.1093/neuonc/noz068.
[16] Zureick AH, Evans CL, Niemierko A, Grieco JA, Nichols AJ, Fullerton BC, et al. Left hippocampal
dosimetry correlates with visual and verbal memory outcomes in survivors of pediatric brain tumors.
Cancer 2018;124:2238–45. https://doi.org/10.1002/cncr.31143.
[17] Okoukoni C, McTyre ER, Ayala Peacock DN, Peiffer AM, Strowd R, Cramer C, et al. Hippocampal
dose volume histogram predicts Hopkins Verbal Learning Test scores after brain irradiation. Adv Radiat
Oncol 2017;2:624–9. https://doi.org/10.1016/j.adro.2017.08.013.
[18] Konczak J, Schoch B, Dimitrova A, Gizewski E, Timmann D. Functional recovery of children and
adolescents after cerebellar tumour resection. Brain 2005;128:1428–41.
https://doi.org/10.1093/brain/awh385.
[19] Benavides-Varela S, Lorusso R, Baro V, Denaro L, Estévez-Pérez N, Lucangeli D, et al. Mathematical

skills in children with pilocytic astrocytoma. Acta Neurochir (Wien) 2019;161:161–9.
https://doi.org/10.1007/s00701-018-3744-0.
[20] Moberget T, Andersson S, Lundar T, Due-Tønnessen BJ, Heldal A, Endestad T, et al. Long-term
supratentorial brain structure and cognitive function following cerebellar tumour resections in childhood.
Neuropsychologia 2015;69:218–31. https://doi.org/10.1016/j.neuropsychologia.2015.02.007.
[21] Pletschko T, Felnhofer A, Lamplmair D, Dorfer C, Czech T, Chocholous M, et al. Cerebellar pilocytic
astrocytoma in childhood: Investigating the long-term impact of surgery on cognitive performance and
functional outcome. Dev Neurorehabil 2017:1–8. https://doi.org/10.1080/17518423.2017.1370502.
[22] Steinlin M, Imfeld S, Zulauf P, Boltshauser E, Lövblad KO, Lüthy AR, et al. Neuropsychological long-
term sequelae after posterior fossa tumour resection during childhood. Brain 2003;126:1998–2008.
https://doi.org/10.1093/brain/awg195.
[23] Vaquero E, Gómez CM, Quintero EA, González-Rosa JJ, Márquez J. Differential prefrontal-like deficit
in children after cerebellar astrocytoma and medulloblastoma tumor. Behav Brain Funct 2008;4:1–16.
https://doi.org/10.1186/1744-9081-4-18.
[24] Mabbott DJ, Penkman L, Witol A, Strother D, Bouffet E. Core Neurocognitive Functions in Children
Treated for Posterior Fossa Tumors. Neuropsychology 2008;22:159–68. https://doi.org/10.1037/0894-
4105.22.2.159.
[25] Callu D, Viguier D, Laroussinie F, Puget S, Boddaert N, Kieffer V, et al. Cognitive and academic
outcome after benign or malignant cerebellar tumor in children. Cogn Behav Neurol 2009;22:270–8.
https://doi.org/10.1097/WNN.0b013e3181bf2d4c.
[26] Hazin I, Dellatolas G, Garcia D, Pedrosa F, Pedrosa A. Intellectual impairment after treatment for
medulloblastoma and astrocytoma in childhood: The Brazilian experience. J Pediatr Hematol Oncol
2011;33:506–15. https://doi.org/10.1097/MPH.0b013e3182260e7c.
[27] Ait Khelifa-Gallois N, Puget S, Longaud A, Laroussinie F, Soria C, Sainte-Rose C, et al. Clinical
Evidence of the Role of the Cerebellum in the Suppression of Overt Articulatory Movements During
Reading. A Study of Reading in Children and Adolescents Treated for Cerebellar Pilocytic Astrocytoma.
Cerebellum 2015;14:97–105. https://doi.org/10.1007/s12311-014-0612-1.
[28] Rønning C, Sundet K, Due-Tønnessen B, Lundar T, Helseth E. Persistent cognitive dysfunction

secondary to cerebellar injury in patients treated for posterior fossa tumors in childhood. Pediatr
Neurosurg 2005;41:15–21. https://doi.org/10.1159/000084860.
[29] Law N, Smith M Lou, Greenberg M, Bouffet E, Taylor MD, Laughlin S, et al. Executive function in
paediatric medulloblastoma: The role of cerebrocerebellar connections. J Neuropsychol 2017;11:174–
200. https://doi.org/10.1111/jnp.12082.
[30] Koustenis E, Hernáiz Driever P, De Sonneville L, Rueckriegel SM. Executive function deficits in
pediatric cerebellar tumor survivors. Eur J Paediatr Neurol 2016;20:25–37.
https://doi.org/10.1016/j.ejpn.2015.11.001.
[31] Reichert JL, Chocholous M, Leiss U, Pletschko T, Kasprian G, Furtner J, et al. Neuronal correlates of
cognitive function in patients with childhood cerebellar tumor lesions. PLoS One 2017;12:1–20.
https://doi.org/10.1371/journal.pone.0180200.
[32] Aarsen FK, Van Dongen HR, Paquier PF, Van Mourik M, Catsman-Berrevoets CE. Long-term sequelae
in children after cerebellar astrocytoma surgery. Neurology 2004;62:1311–6.
https://doi.org/10.1212/01.WNL.0000120549.77188.36.
[33] Quintero-Gallego EA, Gómez CM, Casares EV, Márquez J, Pérez-Santamaría FJ. Declarative and
procedural learning in children and adolescents with posterior fossa tumours. Behav Brain Funct
2006;2:1–9. https://doi.org/10.1186/1744-9081-2-9.
[34] Aarsen FK, Paquier PF, Arts WF, Van Veelen ML, Michiels E, Lequin M, et al. Cognitive deficits and
predictors 3 years after diagnosis of a pilocytic astrocytoma in childhood. J Clin Oncol 2009;27:3526–
32. https://doi.org/10.1200/JCO.2008.19.6303.
[35] Khajuria RK, Blankenburg F, Wuithschick I, Rueckriegel S, Thomale UW, Mansour M, et al.
Morphological brain lesions of pediatric cerebellar tumor survivors correlate with inferior
neurocognitive function but do not affect health-related quality of life. Child’s Nerv Syst 2015;31:569–
80. https://doi.org/10.1007/s00381-015-2635-4.
[36] Berger A, Sadeh M, Tzur G, Shuper A, Kornreich L, Inbar D, et al. Motor and non-motor sequence
learning in children and adolescents with cerebellar damage. J Int Neuropsychol Soc 2005;11:482–7.
https://doi.org/10.1017/S1355617705050587.
[37] Yoo HJ, Kim H, Park HJ, Kim D. Neurocognitive Function and Health-Related Quality of Life in
Pediatric Korean Survivors of Medulloblastoma 2016:1726–34.
[38] Spiegler BJ, Bouffet E, Greenberg ML, Rutka JT, Mabbott DJ. Change in neurocognitive functioning
after treatment with cranial radiation in childhood. J Clin Oncol 2004;22:706–13.
https://doi.org/10.1200/JCO.2004.05.186.
[39] Moxon-Emre I, Bouffet E, Taylor MD, Laperriere N, Scantlebury N, Law N, et al. Impact of
craniospinal dose, boost volume, and neurologic complications on intellectual outcome in patients with
medulloblastoma. J Clin Oncol 2014;32:1760–8. https://doi.org/10.1200/JCO.2013.52.3290.
[40] Cámara S, Fournier MC, Cordero P, Melero J, Robles F, Esteso B, et al. Neuropsychological Profile in
Children with Posterior Fossa Tumors with or Without Postoperative Cerebellar Mutism Syndrome
(CMS). Cerebellum 2020;19:78–88. https://doi.org/10.1007/s12311-019-01088-4.
[41] Kahalley LS, Peterson R, Douglas Ris M, Janzen L, Fatih Okcu M, Grosshans DR, et al. Superior
intellectual outcomes after proton radiotherapy compared with photon radiotherapy for pediatric
medulloblastoma. J Clin Oncol 2020;38:454–61. https://doi.org/10.1200/JCO.19.01706.
[42] Kieffer-Renaux V, Bulteau C, Grill J, Kalifa C, Viguier D, Jambaque I. Patterns of neuropsychological

deficits in children with medulloblastoma according to craniospatial irradiation doses. Dev Med Child
Neurol 2000;42:741–5. https://doi.org/10.1017/S0012162200001377.
[43] Edelstein K, Spiegler BJ, Fung S, Panzarella T, Mabbott DJ, Jewitt N, et al. Early aging in adult
survivors of childhood medulloblastoma: Long-term neurocognitive, functional, and physical outcomes.
Neuro Oncol 2011;13:536–45. https://doi.org/10.1093/neuonc/nor015.
[44] Szentes A, Erős N, Kekecs Z, Jakab Z, Török S, Schuler D, et al. Cognitive deficits and
psychopathological symptoms among children with medulloblastoma. Eur J Cancer Care (Engl)
2018;27:e12912. https://doi.org/10.1111/ecc.12912.
[45] Khalil J, Chaabi S, Oberlin O, Sialiti S, Hessissen L, Benjaafar N. Medulloblastoma in childhood: What
effects on neurocognitive functions? Cancer/Radiotherapie 2019;23:370–7.
https://doi.org/10.1016/j.canrad.2018.11.004.
[46] Brinkman TM, Reddick WE, Luxton J, Glass JO, Sabin ND, Srivastava DK, et al. Cerebral white matter
integrity and executive function in adult survivors of childhood medulloblastoma. Neuro Oncol
2012;14:25–36. https://doi.org/10.1093/neuonc/nos214.
[47] Knight SJ, Conklin HM, Palmer SL, Schreiber JE, Armstrong CL, Wallace D, et al. Working memory
abilities among children treated for medulloblastoma: Parent report and child performance. J Pediatr
Psychol 2014;39:501–11. https://doi.org/10.1093/jpepsy/jsu009.
[48] Glass JO, Ogg RJ, Hyun JW, Harreld JH, Schreiber JE, Palmer SL, et al. Disrupted development and
integrity of frontal white matter in patients treated for pediatric medulloblastoma. Neuro Oncol
2017;19:1408–18. https://doi.org/10.1093/neuonc/nox062.
[49] Schreiber JE, Palmer SL, Conklin HM, Mabbott DJ, Swain MA, Bonner MJ, et al. Posterior fossa
syndrome and long-term neuropsychological outcomes among children treated for medulloblastoma on a
multi-institutional, prospective study. Neuro Oncol 2017;19:1673–82.
https://doi.org/10.1093/neuonc/nox135.
[50] Palmer SL, Armstrong C, Onar-Thomas A, Wu S, Wallace D, Bonner MJ, et al. Processing speed,
attention, and working memory after treatment for medulloblastoma: An international, prospective, and
longitudinal study. J Clin Oncol 2013;31:3494–500. https://doi.org/10.1200/JCO.2012.47.4775.
[51] Sekeres MJ, Riggs L, Decker A, de Medeiros CB, Bacopulos A, Skocic J, et al. Impaired recent, but
preserved remote, autobiographical memory in pediatric brain tumor patients. J Neurosci 2018;38:8251–
61. https://doi.org/10.1523/JNEUROSCI.1056-18.2018.
[52] Riggs L, Bouffet E, Laughlin S, Laperriere N, Liu F, Skocic J, et al. Changes to memory structures in
children treated for posterior fossa tumors. J Int Neuropsychol Soc 2014;20:168–80.
https://doi.org/10.1017/S135561771300129X.
[53] Mulhern RK, Palmer SL, Reddick WE, Glass JO, Kun LE, Taylor J, et al. Risks of young age for
selected neurocognitive deficits in medulloblastoma are associated with white matter loss. J Clin Oncol
2001;19:472–9. https://doi.org/10.1200/JCO.2001.19.2.472.
[54] Maddrey AM, Bergeron JA, Lombardo ER, McDonald NK, Mulne AF, Barenberg PD, et al.
Neuropsychological performance and quality of life of 10 year survivors of childhood medulloblastoma.
J Neurooncol 2005;72:245–53. https://doi.org/10.1007/s11060-004-3009-z.
[55] Hardy KK, Bonner MJ, Willard VW, Watral MA, Gururangan S. Hydrocephalus as a possible additional
contributor to cognitive outcome in survivors of pediatric medulloblastoma. Psychooncology
2008;17:1157–61. https://doi.org/10.1002/pon.1349.
[56] Roncadin C, Dennis M, Greenberg ML, Spiegler BJ. Adverse medical events associated with childhood
cerebellar astrocytomas and medulloblastomas: Natural history and relation to very long-term
neurobehavioral outcome. Child’s Nerv Syst 2008;24:995–1002. https://doi.org/10.1007/s00381-008-
0658-9.
[57] Hoang DH, Pagnier A, Cousin E, Guichardet K, Schiff I, Icher C, et al. Anatomo-functional study of the
cerebellum in working memory in children treated for medulloblastoma. J Neuroradiol 2019;46:207–13.
https://doi.org/10.1016/j.neurad.2019.01.093.
[58] Durand T, Berzero G, Bompaire F, Hoffmann S, Léger I, Jego V, et al. Episodic Memory Impairments in
Primary Brain Tumor Patients. Arch Clin Neuropsychol 2018;33:949–55.
https://doi.org/10.1093/arclin/acx138.
[59] Phillips NS, Duke ES, Schofield HLT, Ullrich NJ. Neurotoxic Effects of Childhood Cancer Therapy and
Its Potential Neurocognitive Impact. J Clin Oncol 2021;39:1752–65.
https://doi.org/10.1200/JCO.20.02533.
[60] Klein M. Treatment options and neurocognitive outcome in patients with diffuse low-grade glioma. J
Neurosurg Sci 2015;59:383–92.
[61] Oyefiade A, Paltin I, De Luca CR, Hardy KK, Grosshans DR, Chintagumpala M, et al. Cognitive Risk in
Survivors of Pediatric Brain Tumors. J Clin Oncol 2021;39:1718–26.
https://doi.org/10.1200/JCO.20.02338.
[62] Laliberté Durish C, Moxon-Emre I, Bouffet E, Bartels U, Mabbott DJ. Family environment as a
predictor and moderator of cognitive and psychosocial outcomes in children treated for posterior fossa
tumors. Child Neuropsychol 2021:1–20. https://doi.org/10.1080/09297049.2021.1885639.
[63] Torres VA, Ashford JM, Wright E, Xu J, Zhang H, Merchant TE, et al. The impact of socioeconomic
status (SES) on cognitive outcomes following radiotherapy for pediatric brain tumors: a prospective,
longitudinal trial. Neuro Oncol 2021;23:1173–82. https://doi.org/10.1093/neuonc/noab018.
[64] Kesler SR, Lacayo NJ, Jo B. A pilot study of an online cognitive rehabilitation program for executive
function skills in children with cancer-related brain injury 2012;25:101–12.
https://doi.org/10.3109/02699052.2010.536194.A.
[65] Zucchella C, Capone A, Codella V, De Nunzio AM, Vecchione C, Sandrini G, et al. Cognitive
rehabilitation for early post-surgery inpatients affected by primary brain tumor: a randomized, controlled
trial. J Neurooncol 2013;114:93–100. https://doi.org/10.1007/s11060-013-1153-z.
[66] Baudou E, Patrice P, Arribarat G, Desirat J, Bertozzi A, Gambart M, et al. Clinical and Translational
Radiation Oncology A prospective behavioral and imaging study exploring the impact on long-term
memory of radiotherapy delivered for a brain tumor in childhood and adolescence 2022;33:7–14.
https://doi.org/10.1016/j.ctro.2021.10.006.
FIGURES
Fig 1: Changes in radiotherapy techniques in the management of malignant posterior fossa brain tumors
since 2000: Increased survival and decreased long-term side effects.
CSI: craniospinal irradiation; Gy: gray; PF: posterior fossa; RT: radiotherapy; yo: years old
* Inequality of access between countries and centers
Fig 2: PRISMA Flow diagram
Fig 3: Summary of the neuropsychological scores in different memory systems according to the memory
process assessed and the radiotherapy treatment.
Each colored squared corresponds to the results of one article. The color green represents scores within the normal
range (-1SD to 0SD), orange the low range (-2SD to -1SD) and red the deficit range (< -2SD). In this table, standard
and reduced CSI doses were not differentiated.
CSI: craniospinal irradiation; PF: posterior fossa; RT: radiotherapy; TB: tumor bed
Table 1: Memory systems
A short definition is given for the four main memory systems and memory processes of interest and specific tests
commonly used to asses them, brain areas involved and changes during the 2 first decades of life are summarized.
Yo: years old
Definition Process Neuropsychological tests Brain areas Development
Load: quantity
of information Verbal and visual span
Prefrontal Forward span:
that can be kept tasks
Allows cortex, anterior progressively
in mind
maintenance, cingulate gyrus, improves from
Short- Updating:
Working control and parietal lobe, 2 to 9 yo.
term ability to
memory processing of Broca's, Backward span:
memory replace the
information for occipital lobe, improve from
immediate use information N-back tasks cerebellar 6 yo to
stored a posterior lobe adolescence [5]
moment ago to
update it
Retrograde:
Information ability to
Questionnaires on
about remember
episodes specific to an Medial
personally information
individual’s life since temporal lobe, Moderately
experienced encoded before
birth especially improves from
events, the diagnosis of
brain tumor hippocampus, 3 to 9 yo, with a
Episodic associated with
fornix, cingulum quick
memory their Anterograde:
Word lists or picture bundle, increasement
spatiotemporal ability to
lists: learning, free recall, prefrontal and from 9 to 10 yo
context of encode new
an indexed recall and/or parietal [6]
acquisition and information
a task of recognition cortices.
their emotional after the
Declarative after a predefined time
content. diagnosis of
memory period
Long brain tumor
term General
memory Prefrontal
knowledge
cortex,
information
especially
(facts, ideas,
Language-based frontal inferior Progressively
meaning and
Semantic semantic Word generation, image gyrus (semantic improves from
concepts)
memory information naming representation birth to
regardless of
retrieval tasks access) and adolescence [7]
the
temporal
spatiotemporal
cortex (storing
context of
information).
acquisition.
Knowledge that Cognitive: Cerebellum and First memory
Non- Probabilistic
Procedural is acquired ability to learn a cerebello- system to
declarative classification task, Tour
memory during cognitive cortical circuit mature.
memory of Hanoï
perceptual- procedure (motor- Progressively
motor and Perceptual- adaptation), improves during
cognitive verbal: linked to Mirror writing test striatum and the two first
activities whose reading cortico-striatal decades or
learning circuit (motor stable over time
requires sequence [8]
repetition, and Perceptual- learning),
of which motor: ability to frontal
expression is learn a Sequence learning or associative
automatic. perceptual- motor adaptation tasks regions, medial
motor temporal lobe
procedure (hippocampus)
and temporal
cortex
Table 2: Memory tasks used in the literature review
MEMORY
PROCESS MEMORY TASK
SYSTEM
WORKING LOAD Verbal Digit span forward (WISC/WAIS)

MEMORY
Digit span backward (WISC/WAIS)
Letter-Number sequencing (WIE/HAWIK)
Digit span total (WISC/WAIS)
Working Memory Index (WISC/WAIS)
Working memory (WJ-IV)
Visual and Corsi or Wechsler blocs forward

visuo-spatial
Corsi or Wechsler blocs backward
Block board test
Feature identification (Amsterdam Neuropsychological task)
Working Memory Test Battery for Children
UPDATING Paced Auditory Serial Addition Test
N-back task
Working memory reaction times (Test-battery for Attentional Performance)
Working memory omissions (Test-battery for Attentional Performance)
EPISODIC & EPISODIC Recall of a personal life event, one that occurred before and one after radiotherapy
ANTEROGRADE
MEMORY VERBAL Learning California Verbal Learning Test
ANTEROGRADE
Rey Auditory Verbal Learning Test
Verbal Learning and Memory Test
Signoret BEM 144 module
Children’s Auditory Verbal Learning Test
Children's memory scale
Korean version of the auditory verbal learning test
Long delay Free California Verbal Learning Test

recall
Rey Auditory Verbal Learning Test
Signoret BEM 144 module
Children’s Auditory Verbal Learning Test
Verbal delayed index (Children's memory scale)
Recognition Verbal Learning and Memory Test
Global California Verbal Learning Test
Children's memory scale
Wide Range Assessment of Memory and Learning
VISUAL Learning Rey visual design learning test

ANTEROGRADE
Brief Visuospatial Memory Test - revised learning
Continuous Visual Memory Test total learning
Korean complex figure test immediate recall

Children's memory scale visual immediate index
Long delay Free Rey Complex Figure Test and Recognition Trial
recall
Brief Visuospatial Memory Test - revised delayed recall
Continuous Visual Memory Test delayed recognition
Korean complex figure test delayed recall

Recognition Rey Complex Figure Test and Recognition Trial
Visual recognition test
Global Children's memory scale visual composite score
GLOBAL ANTEROGRADE Wechsler Memory Scale - Revised
Children's memory scale general index
Recognition Memory Test
Wide Range Assessment of Memory and Learning.
SEMANTIC Subtest Information (WISC/WAIS)

MEMORY
Subtest Information (HIE/HAWIK)
Known people and places
PROCEDURAL PERCEPTUAL-MOTOR Serial reaction time task

MEMORY
BEM: Batterie d'Efficience Mnésique; HAWIK: Hamburg-Wechsler-Intelligenztest für Kinder und Jugendliche;
HIE: Hamburg-Wechsler-Intelligenztest für Erwachsene; WAIS: Wechsler Adult Intelligence Scale; WISC:
Wechsler Intelligence Scale for Children; WJ: Woodcock-Johnson Tests of Cognitive Abilities.
Table 3: Characteristics of the study population and the memory system assessed
MEMORY SYSTEM
POPULATION EXAMINED
STUDIES
Control
Non-irradiated Group Irradiated Group Group
Tumor Age at diagnosis Follow-up time Tumor Age at diagnosis Follow-up time
N type (year) (year) N type (year) (year) WM EM SM PM
Aarsen et al. 2004 23 AS 9.3 (SD = 3.7; R: 3,4 (SD = 2.2; No +
3.9-16.6) R: 1.0-8.1)
Aarsen et al. 2009 35 AS 7.7 (R: 3.2-11.4) 3,5 (R: 2.0-5.0) No +
Ait Khelifa et al. 17 AS 5 (SD = 2.1; R: 2- 6 (SD= 4; R: 1- 61 +

2015 10) 15)
Benavides et al. 11 AS 11.2 (SD = 1.8; R: 2.45 (R: 0.6- 11 +
2019 6.2-12.5) 5.5)
Berger et al. 2004 8 AS/NM 7.2 (R: 1-11) 5.9 (SD = 2.76) 8 +
C
Brinkman et al. 20 MB 29 (R: 2-17) 18 No + +
2012
Callu et al. 2009 19 AS 6.1 (SD = 1.8; R: 2.9 (SD = 2.0) 20 HGT 4.7 (SD = 1.9; R: 2.9 (SD = 2.0) No + + +
1.8-8.5) 0.5-8.9)
Camara et al. 137 MB 7.1 (SD= 4.1) 12.9 (SD= 4.3; No +
2015 R: 5–22)
Edelstein et al. 20 MB 7.2 (SD = 3.8; R: No +
2011 1.1-13.8)
Glass et al. 2017 92 MB 8.7 (R: 3.2-21.6) 3 72 +
Hardy et al. 2008 25 MB 8.2 (SD = 2.8; R: 2.2 (SD = 1.83; No +

4–14) R: 1-7)
Hazin et al. 2011 13 AS 9.2 (R: 4-13) 9 <= 3 yo 7 MB 6,.5 (R: 2-12) 4 <= 3 yo No + +
Hoang et al. 2019 11 MB NA 3.6 (SD = 1.2; 23 +

R: 2.25-6.1)
Kahalley 2020 79 MB 8.9 (SD = 2.9; R: 4 No +
3.5-14.4)*
Khajura et al. 17 AS 6.7 (R: 0.9-12.2) 6.3 (SD = 2.6) 17 MB 7.6 (R: 2.2-16.6) 5.6 (SD = 3.2) No +
2015
Khalil et al. 2019 16 MB 6.8 (SD= 2.3; R: 4 No +
4-11)
Kieffer-Renaux et 36 MB 13.1 (SD = 4.1) 4.3 (SD = 4.7) No + + +
al. 2000
Knight et al. 2014 167 MB 9.2 (SD= 3.9) 5 No +
Konczak et al. 14 LGT 8.79 (SD= 4.69; R: 8.36 (SD= 4.31; 6 HGT 9.66 (SD= 2.66; 7.5 (SD= 3.02; 14 +
2005 1-17) R: 3-17) R:7-13) R: 4-13)
Koustenis et al. 17 AS 9.21 (SD = 5.2) 2.52 (SD = 25 MB/EP 9.7 (SD = 4.5) 4.3 (SD = 3.01) No +
2015 2.10)
Law et al. 2017 25 MB 13.3 (SD = 3.5; R: 6.3 (SD = 4.1; 20 +
8.0–19.0) R: 1.2–13.6)
Mabbott et al. 32 LGT 5.2 (SD = 3.6) 6.3 (SD = 2.6) 32 HGT 6.85 (SD = 2.66) 4.6 (SD = 2.5) 10 +
2008
Maddrey et al. 16 MB 7.3 (SD = 4.5; R: 14.6 (SD = 3.5) No +
2005 1-15)
Moberget et al. 20 AS 7.1 (SD= 4.1) 12.9 (SD = 4.3; 26 + +
2015 R: 5-22)
Moxon et al. 2014 115 MB 7.5 (SD = 3.4, R: 6,1 (SD = 3.4, No +
1.1-15.0) R: 1.5-14.2)
Mulhern et al. 42 MB 8.2 (SD = 3.8) 4.9 (SD = 2.5) No + +
2001
Palmer et al. 2001 44 MB 4.6 (SD = 3.3; R: 5.2 (SD = 2.4; No +
1,1-12,5) R: 1.9-12.6)
Palmer et al. 2013 126 MB 9.8 (SD = 4.4) 5 No +
Pletschko et al. 14 AS 13.3 (R: 3–21) 8.1 (SD = 2.8; 14 + + +

2018 R:3.7–13.7)
Quintero-Gallego 11 AS 8.0 (SD = 3.2; R: 4.8 (SD = 3.6; 7 MB 8.0 (SD = 3.2; R: 4.8 (SD = 3.6; 12 +
et al. 2006 1.9-10.8)** R: 0.4-12.6)** 1.9-10.8)** R: 0.4-12.6)** +
Reitchert et al. 11 LGT 6.6 (SD = 3.3) 16 (SD = 3.9; 17 +
2017 R:7.8-20.5)
Riggs et al. 2014 20 HGT 12.4 (R: 7.2-17.2) 5.1 (R: 1.1- 13 +
11.6)
Roncadin et al. 29 AS 6.4 (SD = 3.8; R: 11.1 (SD = 6.1; No +
2008 1.2-15.9) R: 4.8-22.2)
Ronning et al. 12 AS 8.6 (SD = 3.9; R: 14.9 (SD = 3.1; 11 MB 6.1 (SD = 3.4; R: 17.0 (SD = 4.9; No + +
2004 3-14.9) R: 10.0-21.1) 1.8-12.1) R: 10.7-27.0)
Schreiber et al. 36 MB 8.4 (SD = 2.7) 5 36 +
2018
Sekeres et al. 2018 13 MB/EP 6.59 (SD = 2.7; R: 7.42 (SD = 4.1; 28 +
2.8–11.8) R: 1.6–13.8)
Spiegler et al. 34 MB/EP 6.1 (SD = 2.7) 4.7 (R: 1.3- No + +
2004 15.3)
Steinlin et al. 2003 23 LGT 8.3 (R: 3.6-15.5) 7,5 (R: 2.1- No + + +
18.3)
Szentes et al. 2019 34 MB 7.53 (SD = 3.3) 2.71 (SD = 1.8) 46 +
Vaquero et al. 13 AS 8.2 (SD = 4.0) 3.3 (SD = 2.7) 7 MB 7.1 (SD = 2.1) 6.5 (SD = 2.8) 12 +
2008
von Hoff et al. 23 EP 7.2 (R: 0.3-14.2) 4.5 (R: 1-15.5) No + +
2008
Yoo et al. 2016 58 MB 8 (R: 1-22) 5.7 No + + +
Zapotocky et al. 24 EP 4.94 (R: 0.43- 5.54 No +

2019 17.68)
AS: astrocytoma; EM: episodic memory; EP: ependymoma; HGT: high-grade tumor; LGT: low-grade tumor;
MB: medulloblastoma, NMC: non-malignant cyst, PM; procedural memory; SM: semantic memory; WM:
working memory
* mean only for 37 patients treated with proton therapy. For 42 patients treated with photon therapy: mean age at
diagnosis 8.4 yo (SD: 3.1; R: 3.6-15.3).
** mean for both groups AS and MB.

Journal Pre-Proofs: Radiotherapy and Oncology

Uploaded by

Copyright:

Available Formats

Journal Pre-Proofs: Radiotherapy and Oncology

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal Pre-Proofs: Radiotherapy and Oncology

Uploaded by

Copyright:

Available Formats

Journal Pre-proofs

A review of long-term deficits in memory systems following radiotherapy for

Eloïse Baudou, Lisa Pollidoro, Stéphanie Iannuzzi, Anne-Isabelle Bertozzi,

To appear in: Radiotherapy and Oncology

Received Date: 21 December 2021

© 2022 Published by Elsevier B.V.

Short title: Pediatric posterior fossa and memory

- All memory systems are impaired in irradiated PFT survivors.

---- Insert Table 1-----

References in table 1: [5–8]

---- Insert Figure 1-----

2.1- Eligibility criteria

2.2- Search strategy

2.3- Data extraction

---- Insert Table 2-----

---- Insert Figure 2-----

---- Insert Table 3-----

---- Insert Figure 3-----

Ethical approval and consent to participate

Consent for publication

Availability of data and materials

Data sharing is not applicable to this article.

Funding and acknowledgements

the authors read and approved the final manuscript.

[5] Barrouillet P. Le développement de la mémoire de travail 2018.

[6] Guillery-Girard B, Martins S, Deshayes S, Hertz-Pannier L, Chiron C, Jambaqué I, et al. Developmental

[9] Schmahmann J. The cerebellar cognitive affective syndrome. Brain 1998;121:561–79.

[19] Benavides-Varela S, Lorusso R, Baro V, Denaro L, Estévez-Pérez N, Lucangeli D, et al. Mathematical

[28] Rønning C, Sundet K, Due-Tønnessen B, Lundar T, Helseth E. Persistent cognitive dysfunction

[42] Kieffer-Renaux V, Bulteau C, Grill J, Kalifa C, Viguier D, Jambaque I. Patterns of neuropsychological

Fig 2: PRISMA Flow diagram

Table 1: Memory systems

Yo: years old

Definition Process Neuropsychological tests Brain areas Development

Table 2: Memory tasks used in the literature review

WORKING LOAD Verbal Digit span forward (WISC/WAIS)

Letter-Number sequencing (WIE/HAWIK)

Digit span total (WISC/WAIS)

Working Memory Index (WISC/WAIS)

Working memory (WJ-IV)

Visual and Corsi or Wechsler blocs forward

Block board test

Feature identification (Amsterdam Neuropsychological task)

Working Memory Test Battery for Children

UPDATING Paced Auditory Serial Addition Test

Working memory reaction times (Test-battery for Attentional Performance)

Working memory omissions (Test-battery for Attentional Performance)

Verbal Learning and Memory Test

Signoret BEM 144 module

Children’s Auditory Verbal Learning Test

Children's memory scale

Korean version of the auditory verbal learning test

Long delay Free California Verbal Learning Test

Verbal Learning and Memory Test

Signoret BEM 144 module

Verbal delayed index (Children's memory scale)

Korean version of the auditory verbal learning test

Recognition Verbal Learning and Memory Test

Korean version of the auditory verbal learning test