Clinico-radiological outcomes in WNT-subgroup medulloblastoma

1. Introduction

Medulloblastoma (MB), the most common malignant tumor involving the brain and central nervous system (CNS) in children is now considered to be a heterogeneous disease comprising of four broad molecular subgroups - wingless (WNT), sonic hedgehog (SHH), Group 3, and Group 4 respectively with subgroup-specific developmental origins, unique genetic profiles, distinct clinico-demographic characteristics, and diverse clinical outcomes “[1,2,3,4]”. This led to the incorporation of molecular/genetic information in the World Health Organization (WHO) CNS tumor classification that recommended an integrated layered diagnosis including available molecular/genetic information in the WHO 2016 update “[5]”. The 5^th edition of WHO classification of tumors involving the CNS (WHO CNS 5) “[6]” now classifies MB as molecularly-defined and histologically-defined groups to reflect this knowledge regarding the existence of clinical and biological heterogeneity. Molecularly-defined MB comprises of WNT-activated subgroup; SHH-activated subgroup which could be TP53-wildtype or TP53-mutant; and non-WNT/non-SHH subgroup. The WHO 2016 update “[5]” had classified MB into four histo-morphologic types labelled as classic; desmoplastic/nodular (D/N); medulloblastoma with extensive nodularity (MBEN); and large cell/anaplastic (LC/A). These histologic subtypes have now been combined into one section that describes them as morphologic patterns of an inclusive tumor type, MB-histologically defined in WHO CNS 5 “[6]”.

MB has a high propensity to spread throughout the craniospinal axis via cerebrospinal fluid (CSF) pathways with metastatic disease being identified on neuraxial staging in nearly one third of patients at initial diagnosis “[2,3]” necessitating treatment of the entire brain and spinal cord including its covering meninges for disease control. Contemporary management for non-infantile MB “[4,7,8]” comprises of maximal safe resection followed by post-operative risk-stratified adjuvant craniospinal irradiation (CSI) to a dose of 23.4-36Gy/13-20 fractions plus boost irradiation of the tumor-bed to a dose of 18-30.6Gy/10-17 fractions resulting in total primary site radiotherapy (RT) dose of 54Gy/30 fractions followed by 6-9 cycles of multi-agent adjuvant systemic chemotherapy “[4,7,9]”. Traditionally, children over the age of 3-years at diagnosis, with no/small residual tumor (<1.5cm²) and absence of metastatic disease (M0) were classified as average-risk disease “[10]” with >80% long-term survival “[3,4,5,6,7,8,9,10,11]”; while younger age (<3 years), large residual tumor (≥1.5cm²), and presence of leptomeningeal metastases (M+ disease) either alone or in combination were considered high-risk features “[10]” yielding much worse 5-year survival (30-60%) despite aggressive multi-modality therapy “[13,14]”. This traditional risk-stratification has been further refined by incorporating molecular/genetic information in the contemporary molecular era into low-risk, standard-risk, high-risk, and very high-risk categories with expected 5-year overall survival of >90%, 75-90%, 50-75%, and <50% respectively “[15]”.

Aggressive multi-modality treatment achieves good survival outcomes in MB but is associated with significant acute and late treatment-related toxicities including but not limited to neuro-cognitive impairment, neuro-psychological dysfunction, endocrinopathy particularly growth retardation, sensori-neural hearing loss (SNHL), vasculopathy specially cerebro-vascular accident (CVA), and second malignant neoplasm (SMN) “[16,17]”. Of all MBs, WNT-subgroup has the best outcomes (5-year survival >90%) particularly in children “[3,4,13,18]” making these long-term survivors more susceptible to dose-dependent treatment-related late morbidity prompting systematic attempts at de-intensification of therapy “[19]”. An appropriate risk-classification schema and optimal treatment regimen for WNT-MB is yet to be defined. For this, it is important to identify prognostic factors and assess patterns of failure to guide therapeutic decision-making for tailoring adjuvant therapy in WNT-subgroup MB.

2. Materials and Methods

Patients with molecularly-confirmed WNT-MB treated with maximal safe resection followed by post-operative standard-of-care risk-stratified adjuvant radio(chemo)therapy were identified retrospectively via electronic search of the neuro-oncology database.

Molecular subgrouping: Molecular subgroup assignment of MB was based on an inhouse developed assay combining differential expression of 12 protein-coding genes and 9 miRNAs using real-time reverse transcriptase polymerase chain reaction (RT-PCR) as described previously “[20]”. Briefly, RNA (1–2 mg) was reverse transcribed using random hexameric primers and M-MLV reverse transcriptase (Invitrogen). The primers for real-time PCR analysis were designed such that they corresponded to 2 adjacent exons and, wherever possible, were located at exon boundaries to avoid amplification of genomic DNA. The amplicon size was maintained below 75 –80 bp, so as to enable amplification of the fragmented RNA from formalin-fixed paraffin-embedded (FFPE) tissues. The expression was analyzed by SYBR Green PCR amplification assay on an Applied Biosystems 7900HT real-time PCR system using 10 ng cDNA per reaction for frozen tissues and 10-100 ng cDNA per reaction for FFPE tissues. For miRNA expression analysis, 50 ng RNA from fresh tissues and 50–200 ng RNA from FFPE tissues were reverse transcribed using multiplex RT primer pools and the Taqman MicroRNA Reverse Transcription Kit (Applied Biosystems) according to manufacturer’s instructions. The expression of each miRNA was analyzed by TaqMan real-time miRNA assay (Applied Biosystems) on the ABI 7900HT real-time PCR system using 10 ng cDNA from frozen tissues and 10–40 ng cDNA from FFPE tissues. The relative quantity (RQ) of each protein-coding gene/miRNA compared with GAPDH/RNU48 was determined by the comparative cycle threshold (Ct) method. Genes that were significantly differentially expressed in the 4 molecular subgroups were identified by Significance Analysis of Microarray (MeV, http://www.TM4.com) of expression profiling data previously obtained using Affymetrix Gene 1.0 ST array. The selection of 12 protein-coding marker genes for classification from the significantly differentially expressed genes was based on the standardized fold-change in expression of the gene in a particular subgroup. Concomitant over-expression of WIF1, DKK2, and MYC identified WNT-MB. Over-expression of HHIP, EYA1, and MYCN with under-expression of OTX2 served as markers for the SHH-subgroup. The over-expression of EOMES helped to identify Group 3 and Group 4 tumors, while higher expression of NPR3, MYC, and IMPG2 with lower expression of GRM8 and UNC5D helped to distinguish Group 3 from Group 4 tumors. Similar to gene-expression profiling, the differential expression of 9 selected miRNAs was used for subgroup assignment. WNT-activated tumors showed significant over-expression of miR-193a-3p, miR-224, miR-148a, miR-23b, and miR-365 compared with other subgroups. MiR-182 was found to be over-expressed in all WNT-MBs and in many Group 3 and some Group 4 MBs, while miR-204 was over-expressed in all WNT-MBs and in most Group 4 MBs. MiR-182, miR-135b, and miR-204 were found to be under-expressed in SHH-activated MBs. MiR-135b was found to be over-expressed in Group 3 and Group 4 tumors. MiR-592, a miRNA that is located within the GRM8 gene was over-expressed in Group 4 MB. This aforesaid assay had previously been successfully validated “[20]” against a set of 34 well-annotated FFPE MB samples with subgroup assignment based on the 22-gene set NanoString assay from the German Cancer Research Centre (DFKZ). In recent times (after 2017), confirmation of WNT-activation was further supplemented by testing for monosomy 6 (fluorescence in-situ hybridization), CTNNB1 mutation analysis (Sanger sequencing), and/or nuclear beta-catenin positivity (immunohistochemical staining) as orthogonal techniques.

Treatment & follow-up: Information regarding patient demographics, clinical features, histopathological features, molecular profiling, risk-stratification, treatment details, and outcomes were retrieved from hospital case files and/or electronic medical records as was appropriate. All patients underwent maximal safe resection followed by post-operative risk-stratified adjuvant radio(chemo)therapy. Risk-stratification after surgery was based on conventional criteria without upfront knowledge of the molecular subgroup. Children (≤16-years) with average-risk MB defined as residual tumor <1.5cm² with no evidence of metastases (M0) were treated with CSI (23.4Gy) plus boost irradiation (30.6Gy) for total primary-site dose of 54Gy followed by 6-cycles of adjuvant systemic chemotherapy. For adolescents and young adults (AYA) over 16-years of age at initial diagnosis with average-risk MB, RT alone was considered and comprised of full-dose CSI (35-36Gy) plus boost (18-19.8Gy) for total primary-site dose of 54-54.8Gy without adjuvant chemotherapy. The presence of any high-risk features such as large residue (≥1.5cm²), metastatic disease (M+) or adverse histology (LC/A) mandated full-dose/extended-dose CSI (35-40Gy) plus boost irradiation of primary-site (14.4-19.8Gy) with or without boost (5.4-9Gy) to the metastatic deposits followed by 6-cycles of adjuvant systemic chemotherapy. Following completion of adjuvant radio(chemo)therapy, patients were followed up clinically at 3-4 monthly intervals for the first two years, 6-monthly intervals till 5-years, and annually thereafter with periodic surveillance magnetic resonance imaging (MRI) scans as per institutional policy.

Statistical analysis: Clinical and demographic variables were analyzed and summarized using descriptive statistics with measures of central tendency and dispersion being reported. Patterns of relapse were defined as local recurrence (in and around the surgical cavity/resected tumor-bed); metastatic disease either involving the leptomeningeal space outside the initial tumor-bed in the cranial and/or spinal leptomeninges or extra-neural metastases (ENM) involving the bones, lymph nodes, or bone marrow; or a combination of the above. Progression free survival (PFS) was defined as the time interval from the date of surgery till documented clinico-radiological progression, or death due to any cause, or last follow-up. Overall survival (OS) was defined from the date of surgery till death due to any cause or last documented follow-up. Median follow up of surviving patients was calculated by the reverse Kaplan-Meier method. Time-to-event outcomes were analyzed using the product-limit method of Kaplan Meier and presented as 5-year estimates with 95% confidence interval (CI). Univariate analysis of variables of known and/or presumed prognostic significance was done using the log-rank test after dichotomization at median values or cut-offs established from earlier literature as appropriate. Statistical analysis was performed using SPSS version 25.0 (IBM Corporation, Armonk, USA) and R Studio version 3.2.7 (R Corporation, Vienna). The study was duly reviewed and approved by the Institutional Ethics Committee (IEC) that functions in accordance with the Declaration of Helsinki. IEC also granted waiver of consent due to retrospective nature of the study with no/minimal risk to participants.

3. Results

Electronic search of the neuro-oncology database identified a total of 504 MB patients registered in the neuro-oncology unit of the institute between 2004 till 2020 of which 74 (14.6%) were diagnosed as having WNT-subgroup MB “[20]”. Seven patients who were treated on a prospective protocol of therapy de-intensification in WNT-MB “[21]” were excluded from the dataset leaving 67 patients which constitute the present study cohort.

Clinico-demographic features: Patient, disease, and treatment characteristics of the study cohort are summarized in Table 1. Median age of the study cohort was 12 years with an inter-quartile range (IQR) of 9-18 years and preponderance of male gender (2:1 ratio). Pediatric WNT-MB (defined as age ≤ 16 years) comprised 73.1% (n=49) of patients compared to 26.9% (n=18) of AYA WNT-MB (defined as age >16 years). All patients underwent maximal safe resection with gross total resection (GTR) achieved in 49% of patients. Classic histology was the most common histological subtype seen in 61.2% (n=41) patients. Metastatic disease status by CSF cytology and/or neuro-imaging was available in 62 patients with majority (91.9%, n=57) being non-metastatic at initial diagnosis. Presence of any one or more of the following adverse features such as large residual tumor (≥1.5cm²), metastatic disease (M+) and LC/A histology classified 16 (29.1%) patients as having high-risk disease and 39 (70.9%) patients as average-risk disease. All included patients were treated post-operatively with contemporary risk-stratified RT comprising of CSI plus boost irradiation with or without adjuvant systemic chemotherapy. The median dose of CSI was 35Gy (IQR: 23.4-35Gy) with median tumor-bed boost dose of 19.8Gy (IQR: 19.8-30.6Gy). Extended dose CSI (40Gy) and boost irradiation of metastatic deposits was also done at the discretion of the treating radiation oncologist. Most of the patients were treated with conformal techniques either three-dimensional conformal radiotherapy (3D-CRT) or intensity modulated radiation therapy (IMRT) using 6MV photons on modern linear accelerators including tomotherapy. Adjuvant systemic chemotherapy was delivered in 72.2% (n=39) patients whereas 27.8% (n=15) patients did not receive any chemotherapy after completion of RT. Chemotherapy was initiated 4-6 weeks after completion of RT after sufficient myelo-recovery defined as absolute neutrophil count (ANC) >1500/dl and platelet count >100000/dl. Adjuvant chemotherapy generally comprised of 6 cycles of cisplatin (75mg/m² intravenously on d1 in alternate cycles 2,4,6), cyclophosphamide (1000mg/m² intravenously on d1-d2 in cycles 1,3,5 and d2-d3 in cycles 2,4,6) and vincristine (1.5mg/m² intravenously d1 & d8 in all 6 cycles) given at 4-weekly intervals with adequate hydration, forced saline diuresis, mesna prophylaxis and requisite dose modifications as appropriate (7). Two of 5 children with metastatic disease at initial diagnosis also received 1-year of maintenance chemotherapy post-completion of standard therapy using the modified combined oral metronomic bio-differentiating anti-angiogenic therapy (COMBAT) regimen comprising of temozolomide, etoposide, celecoxib, fenofibrate, and retinoic acid.

Patterns of failure, causes of death, and survival outcomes: Six patients (1 post-operative mortality and 5 without adequate details of treatment or outcomes) were excluded from the survival analysis which was restricted to 61 patients. Nine of the 61 included patients experienced an event of interest (relapse and/or death). Seven patients were detected with relapse on follow-up with leptomeningeal dissemination seen in 4 patients (including one with synchronous local recurrence), local tumor-bed recurrence in 3 patients (including one with synchronous neuraxial relapse), and isolated extra-neural metastases (lymph nodes, bones) in a single patient. Images from one such case scenario each of tumor-bed recurrence only, synchronous local recurrence with neuraxial failure, and isolated extra-neural (ENM) metastases from the study cohort are illustrated in Figure 1. Seven of 61 patients have died by the time of this analysis, six of recurrent/progressive disease and one due to chemotherapy-induced febrile neutropenia leading to septic shock and death. Clinico-demographic details, pattern of relapse, and outcomes of all these 9 patients experiencing an event are summarized in Table 2. Of the five WNT-MB patients who were treated with salvage therapy at relapse, two patients (W1: tumor-bed recurrence and W2: diffuse leptomeningeal metastases) achieved post-relapse survival of 55 and 29 months respectively, while the lone patient with ENM (W8) was alive with disease on salvage systemic chemotherapy at the time of this analysis. Two patients treated with re-excision alone without further re-irradiation or salvage chemotherapy succumbed to further progressive disease within 6-9 months of relapse. All the three WNT-MB patients offered best supportive care at relapse died of progressive disease within 3-months of first relapse. At a median follow-up of 72 months (IQR: 51-101 months) for the entire study cohort (N=61), the 5-year Kaplan-Meier estimates of PFS and OS were 87.7% (95%CI: 75.1-96.1%) and 91.2% (95%CI: 83.0-100%) respectively (Figure 2). Univariate analysis of various patient, disease, and treatment related factors did not identify any putative prognostic factor impacting upon PFS or OS (Table 3). Multi-variate analysis was considered inappropriate due to small number of events in the study cohort.

4. Discussion

The clinico-demographic characteristics of this large cohort of WNT-MB patients treated at an academic neuro-oncology unit of tertiary-care comprehensive cancer centre are largely in accordance with previously published literature with minor differences. The present study had more males with WNT-MB than females (2:1) possibly due to socio-cultural differences and patriarchal mindset prevalent in low-middle income country setting in South-East Asia compared to the fairly balanced gender ratio reported previously from high-income countries of the West “[18]”. Median age at diagnosis of the present study cohort was also slightly higher (12 years, IQR: 9-18 years) reflecting an increased representation of adult WNT-MB compared to an international reference cohort (median 10 years, IQR: 8-14.2 years) which was largely limited to the pediatric age group “[22]”.

Given the low prevalence of WNT-MB (constituting around 10% of all MBs) coupled with very low risk of failure in appropriately treated patients, prognostic factors impacting upon survival, patterns of relapse, and drivers of metastatic dissemination are relatively poorly understood. Nobre et al. “[22]”assembled a retrospective multi-institutional clinically annotated cohort of 93 WNT-pathway medulloblastoma patients using an integrated genomic approach. Fifteen patients with relapse were identified, 12 in metastatic compartment including one with ENM and 3 in the surgical cavity. Interestingly, 8 of 11 neuraxial relapses were in lateral ventricles (6 confined to frontal horns) leading to the hypothesis that ependymal lining of the lateral ventricles may be more conducive to homing of WNT-MB. Maintenance systemic chemotherapy (p=0.033), specifically lower cumulative dose of cyclophosphamide/ifosfamide (<12mg/m²) was reported to be associated with increased risk of relapse. It was proposed that the paracrine signals driven by mutant β-catenin protein induce a fenestrated tumor vasculature promoting accumulation of chemotherapeutic agents within the tumor-bed. The authors also reported that male gender (p=0.032) was associated with significantly increased risk of relapse in WNT-MB. Age at diagnosis, extent of resection, metastatic status at presentation, dose of CSI, and additional molecular/genetic alterations did not predict the risk of relapse in their study. In another cohort of 191 WNT-MB patients registered in the HIT database “[23]”, mutations in CTNNB1, APC, and TP53 were analyzed by DNA sequencing and chromosomal copy number aberrations by molecular inversion probe technology to identify the prognostic impact of TP53 mutations and other chromosomal aberrations in WNT-subgroup. Patients with tumors harboring TP53 mutation showed worse outcomes (5-year PFS: 68% vs 93%, p=0.001 and 5-year OS: 81% vs 95%, p=0.105) compared to TP53 wild-type tumors. Gain of OTX2 was associated with inferior survival outcomes (5-year PFS: 72% vs 93%, p=0.017 and 5-year OS:83% vs 97%, p=0.006). Multivariable Cox regression analysis identified both genetic alterations as independent prognostic markers for survival raising concerns regarding inclusion of such patients in ongoing prospective trials of therapy de-intensification.

The presence of intra-tumoral heterogeneity within the four broad molecular subgroups prompted several researchers to perform large-scale integrative clustering analysis combining DNA methylation and gene-expression profiling to identify further subtypes within each broad molecular subgroup “[24,25,26]” resulting in consensus definition of 12 subtypes of MB in second generation molecular subgrouping “[27]”. WNT-pathway MB typically demonstrate homogenous genome-wide expression patterns and methylation profiles; however, two molecular subtypes of WNT-activated MB have been identified “[27]” referred to as WNT-α and WNT-β that differ in age at diagnosis (median age of 10 vs 20 years), frequency of monosomy 6 (>85% vs <50%), histo-morphology (typically classic vs sometimes LC/A) and metastatic disease (absent vs occasionally present) respectively. Very rarely WNT-MB may harbor distinct genetic alterations typical of another molecular subgroup (such as SHH and non-WNT/non-SHH) in addition to WNT-activation referred to as hybrid molecular subtypes “[28]” indicating intra-tumoral heterogeneity with potential prognostic implications. MYC oncogenes are the most commonly amplified loci in MB “[29,30]” that are generally associated with non-WNT/non-SHH disease (particularly subgroup 3), LC/A histology, and metastatic dissemination making them known biomarkers of poor prognosis. Although overexpression of MYC can also be seen in WNT-subgroup MB with no detrimental impact on survival “[30]”, MYC-amplification has rarely been described “[31]” in WNT-activated tumors. It may be pertinent to note that increased MYC-signaling has been shown to accelerate tumor growth and promote metastases in a murine model of WNT-MB “[32]”.

Although MB largely remains a disease of childhood with much lower incidence in the AYA population, it is common perception that excellent survival outcomes achieved in the pediatric population may not be exactly mirrored in the AYA cohort “[33,34]”. However, contrary to popular belief, analysis of the Surveillance, Epidemiology, and End Results (SEER) database “[35]” from 1992-2013 reported comparable 2-year, 5-year, and 10-year survival outcomes between childhood (n=616) and adult MB (n=349). The first comprehensive molecular analysis adult MB “[36]” defined three broad molecular subgroups viz. WNT, SHH, and Group D (later reclassified as Group 4) with absence of Group 3 tumors. The authors reported worse prognosis of adult WNT-MB and Group 4 disease compared to corresponding subgroups of childhood MB; however, survival in SHH-subgroup MB was similar across both age groups. In another large multi-institutional dataset of adult MB “[37]”, there was no prognostic impact of molecular subgrouping with 5-year survival of 45%, 67%, 62% and 67% for WNT, SHH, Group 3, and Group 4 respectively. The largest integrative analysis of adult MB “[38]” also reported no statistically significant survival differences between the four broad molecular subgroups with 5-year PFS (95%CI) of 64.4% (48.0-86.5%), 61.9% (51.6-74.2%), 80.0% (51.6-100%), and 44.9% (28.6-70.7%) for WNT (n=30), SHH (n=112), Group 3 (n=6), and Group 4 (n=41) respectively. However, what stands out clearly is the substantially lower survival in adults with WNT-activated MB (5-year survival 45-70%) compared to benchmark outcomes in childhood WNT-MB (5-year survival >90%). However, this notion has recently been challenged with molecular subgrouping emerging as a significant prognostic factor in AYA-MB. In a large single-institutional dataset “[39]” of molecularly-characterized AYA-MB (≥15-years at initial diagnosis), the reported 5-year survival was 87.5%, 62.2%, and 50.1% for WNT (n=14), SHH (n=71), and non-WNT/non-SHH (n=21) subgroups respectively. A comparative analysis of pediatric versus AYA WNT-MB also reported similar survival outcomes “[40]” suggesting that age alone should not be used to intensify treatment in WNT-MB.

The time to recurrence (early vs delayed), pattern of failure (local, metastatic, or combined) and post-relapse survival in MB is largely dictated by disease biology and varies across the four broad molecular subgroups “[41,42,43]”. Time to relapse even within the WNT-subgroup has been variable across studies with both early as well as delayed relapses being reported “[22,34] “sometimes even beyond 10-years from initial diagnosis. Management of relapsed MB after appropriate and adequate upfront radio(chemo)therapy is not clearly defined with no universally acceptable standard-of-care salvage treatment “[44]”. A substantial and large proportion of these patients particularly with disseminated disease are offered best supportive care alone with only a small minority being treated with aggressive multi-modality salvage therapies including a combination of re-excision (isolated local relapse), re-irradiation, and systemic therapies that might include high-dose chemotherapy with autologous stem cell rescue and targeted therapy as appropriate. The prognosis of relapsed MB in patients previously treated with CSI in the upfront setting is typically poor and considered non-curative with <5% long-term survival despite aggressive salvage therapies “[44]”. However, it is now being increasingly appreciated that post-relapse outcomes might be somewhat subgroup-dependent with WNT-MB and Group 4 tumors demonstrating a more indolent clinical course with favorable outcomes compared to SHH-MB and Group 3 disease. This study also reported favorable outcomes in a subset of relapsed WNT-MB patients further raising the question whether these patients can be treated with upfront de-intensified therapy at initial diagnosis reserving treatments associated with high morbidity at the time of relapse. The patterns of failure and 5-year survival outcomes of appropriately treated WNT-MB patients in various molecularly-informed prospective cohort studies including randomized controlled trials are summarized in Table 4 “[12,13,42,45,46,47,48]” that reaffirms excellent prognosis and provides justification for ongoing global efforts towards de-escalation of therapy “[19]”. However, such an approach warrants caution as two prospective de-intensification studies had to be terminated prematurely due to unacceptably high risk of failures. The first of these “[21]” treated rigorously-defined low-risk WNT-MB patients with focal-only conformal RT to the index tumor-bed (54Gy) plus adjuvant systemic chemotherapy with omission of upfront CSI. Three of the first 7 patients accrued on the study were detected with neuraxial dissemination within 2-years of index diagnosis and although they were subsequently successfully salvaged in the short-term with aggressive multi-modality therapy including full-dose CSI and systemic chemotherapy, mature outcomes of salvage therapy remain to be reported. The second study “[49]” utilized a post-surgery primary chemotherapy approach eliminating RT completely in low-risk WNT-MB. Once again, three of the first 6 children on the study developed local recurrence and neuraxial dissemination shortly after completing chemotherapy leading to early closure due to safety concerns. Two of them were successfully salvaged with RT including CSI plus additional chemotherapy, but one child succumbed to further progressive disease at 35-months from initial diagnosis. Of the remaining 3 patients, 2 children proceeded to immediate RT after completion of primary chemotherapy to protect against early relapse, while the remaining child was switched mid-treatment to high-dose chemotherapy with autologous stem-cell rescue. Both these studies reinforce the need of RT, particularly CSI for effective disease control even in low-risk, favorable-biology WNT-pathway MB “[50]”.

Strengths & limitations: This study represents the largest descriptive analysis of WNT-MB treated with contemporary risk-stratified radio(chemo)therapy at a single institution anywhere in the world. Access to advanced molecular diagnostics for subgroup assignment and therapeutic decision-making in multi-disciplinary neuro-oncology clinic add further strength to the study. However, despite the above-mentioned strengths, several caveats and limitations remain. Retrospective nature of the study makes it susceptible to inherent biases that could potentially confound interpretation of results. Various platforms exist for robust molecular subgrouping of MB including immunohistochemistry panel, gene-expression analysis, microRNA profiling, and DNA methylation array. The study used combined gene-expression analysis and microRNA profiling for molecular subgroup assignment; however, DNA methylation which is considered the current gold-standard and method of choice for molecular classification of MB was not performed due to issues with availability, accessibility, and affordability. Although rare, the co-occurrence of additional genetic alterations to identify any hybrid molecular subtypes was not assessed in the study. Analysis of survival outcomes was restricted to 61 patients (after excluding 6 patients) which could be a potential source of bias. Follow-up duration though long (median of 72 months) may be considered inadequate to capture very delayed relapses and SMNs. Although therapeutic decision-making was largely based on discussion in a multi-disciplinary tumor board, patients may not have been treated uniformly over the long period of the study potentially impacting upon outcomes. Finally, lack of documented data on neuro-cognitive impairment, neuro-psychological dysfunction, SNHL, endocrinopathies, CVA, and resultant quality-of-life precludes assessment of the impact of treatment-related late toxicity in these long-term survivors.

5. Conclusions

This retrospective clinical audit confirms excellent survival in WNT-MB patients treated with contemporary multi-modality therapy comprising of maximal safe resection followed by risk-stratified appropriate radio(chemo)therapy. Lack of prognostic impact of conventional high-risk features suggests the need for refined risk-stratification and potential for de-intensification of therapy in WNT-MB.