ABSTRACT: This dataset spans two studies which investigate the clinico-pathology and molecular features of infant medulloblastoma, defined as children aged <5 years at diagnosis. Both abstracts follow. Background Infant medulloblastoma (iMB; <3-5-years old at diagnosis) is a major challenge in paediatric oncology. The favourable-risk SHH iMB molecular group with desmoplastic histology (40% of iMB) is commonly spared upfront craniospinal irradiation (CSI) to avoid neurocognitive side-effects. However, clinical studies of the remaining non-WNT/non-SHH iMB group (~60% of iMBs) have not been undertaken, and it is currently considered uniformly high-risk. Understanding the potential for its biological subclassification and clinical stratification is an essential goal towards improved outcomes. This study therefore aimed to directly compare different therapeutic approaches in non-WNT/non-SHH iMB and determine relationships between outcomes and clinico-molecular features. Methods We assembled a global cohort of non-WNT/non-SHH iMBs (n=375). Inclusion criteria was a principal non-WNT/non-SHH MB classification (Group 3 or Group 4); confidence score >0.8) using DNA methylation array-based classification. The survival cohort (n=313) was defined by availability of complete progression-free (PFS) and overall (OS) survival data. Using these cohorts, we characterised their molecular pathology, treatments received, and relationships to outcomes, principally PFS. Findings The total collected cohort (n=375) comprised 262 males and 110 females, with a median age of 3.0 (IQR 2.53-4.0) years. Molecular group 3 tumours (iMBGrp3) predominated (n=246, 66%), primarily molecular subgroups 2 (n=75; 32%), 3 (n=46; 19%) and 4 (n=98; 42%). Regimens incorporating upfront CSI were most common (54% of patients); such patients had significantly better PFS than those treated with focal or no radiotherapy (5-year PFS; CSI, 57.9%; Focal-radiotherapy, 35.3%; No radiotherapy, 26.8%; p<0.001), and equivalent PFS to CSI-treated non-infant patients. Upfront chemotherapy -only approaches were used in 33% (66/199) of iMBGrp3; high-dose (HIGH-DOSE) chemotherapy produced better survival rates (45.7%, 5-year PFS) than intraventricular methotrexate-based (21.1%, 5-year PFS) or standard-dose (11%, 2-year PFS) chemotherapy (p=0.003). Subgroup 4 represents a novel chemo-sensitive iMB group, with 64.3% 5-year PFS using high-dose -chemotherapy-only and with evidence of rescue post-relapse. MYC-amplified iMBGrp3 ( subgroup2/3) have dismal prognosis (0% 2-year overall survival (OS)), with rapid progression regardless of treatment. Remaining subgroup 2/3 patients treated with Chemotherapy-only had 30% 5-year PFS, showed some evidence of post-relapse rescue, and could be randomised to test alternative therapeutic strategies. Interpretation Non-WNT/non-SHH iMB outcomes are associated with specific biomarkers and therapy received. Capturing this ‘real-world’ experience identifies favourable ( subgroup 4) and very-high-risk ( subgroup 2/3-MYC) patient groups, providing a critical foundation for first biomarker-driven iMBGrp3 clinical trials. Background Medulloblastoma in infants (iMB; <5 years) presents the challenge of achieving cure while minimizing deleterious cranio-spinal irradiation (CSI)-associated late-effects. Non-randomised Phase-II studies have examined upfront CSI omission and chemotherapy intensification for favourable-risk desmoplastic/nodular (DN) tumours associated with the SHH molecular group (iMBSHH). Comparison of these therapies in large molecularly-defined iMBSHH cohorts, alongside investigations of prognostic biomarkers in therapy-specific context, is urgently required to define future therapeutic strategies. Methods A multi-national cohort of molecularly and clinically-annotated iMBSHH (n=267), not previously reported in clinical trials, was assembled. We investigated molecular pathology, upfront treatments, and relationships to outcome, in this ‘real-world’ setting. Findings iMBSHH patients treated upfront with intensified chemotherapy-only regimens (n=132) based on intraventricular methotrexate (IVT-MTX; 5-year progression-free survival (PFS), 72.6%) or high-dose (HD; 73.0%) therapy, achieved (PFS) outcomes comparable to upfront CSI-based regimens (n=49; 74.0%)(p=0.51). Lower-intensity standard-dose chemotherapy-only regimens were inferior (SD; 48.4% PFS)(p=0.006). Rescue was common post-relapse after IVT-MTX/HD protocols and translated into 5-year overall survival (OS) of 85.6% and 88.6%, respectively. iMBSHH encompassed SHH-1 (38%), SHH-2 (47%) and SHH-3 (14%) WHO subgroups. In CSI-naïve iMBSHH, SD associated with worse PFS in SHH-1 (p=0.001), but not SHH-2. Non-DN/MBEN histology (21.2% of iMBSHH) conferred worse PFS in upfront CSI-treated and SD (p<0.001 and p=0.003, respectively) groups. Metastatic disease only associated with prognosis with upfront IVT-MTX-only therapies (p=0.013), while established high-risk features of non-infant MBSHH (TP53-mutation, LCA-histology, MYCN-amplification) only associated with poor prognosis in older SHH-3 (7/7 relapsed). Finally, CSI-naïve PFS findings were validated in a re-evaluation of smaller historical trials cohorts. Interpretation iMBSHH outcomes and prognostic biomarkers are therapy-dependent. Non-metastatic iMBSHH treated with HD or IVT-MTX chemotherapy-only have equivalent favourable outcomes, independent of histology and subgroup. With outcomes established, clinical trials must now focus on quality-of-life following different intensified approaches, to identify kindest curative strategies.
