Metabolite profiles of medulloblastoma for rapid and non-invasive detection of molecular disease groups

BACKGROUND

The malignant childhood brain tumour, medulloblastoma, is classified clinically into molecular groups which guide therapy. DNA-methylation profiling is the current classification 'gold-standard', typically delivered 3-4 weeks post-surgery. Pre-surgery non-invasive diagnostics thus offer significant potential to improve early diagnosis and clinical management. Here, we determine tumour metabolite profiles of the four medulloblastoma groups, assess their diagnostic utility using tumour tissue and potential for non-invasive diagnosis using in vivo magnetic resonance spectroscopy (MRS).

METHODS

Metabolite profiles were acquired by high-resolution magic-angle spinning NMR spectroscopy (MAS) from 86 medulloblastomas (from 59 male and 27 female patients), previously classified by DNA-methylation array (WNT (n = 9), SHH (n = 22), Group3 (n = 21), Group4 (n = 34)); RNA-seq data was available for sixty. Unsupervised class-discovery was performed and a support vector machine (SVM) constructed to assess diagnostic performance. The SVM classifier was adapted to use only metabolites (n = 10) routinely quantified from in vivo MRS data, and re-tested. Glutamate was assessed as a predictor of overall survival.

FINDINGS

Group-specific metabolite profiles were identified; tumours clustered with good concordance to their reference molecular group (93%). GABA was only detected in WNT, taurine was low in SHH and lipids were high in Group3. The tissue-based metabolite SVM classifier had a cross-validated accuracy of 89% (100% for WNT) and, adapted to use metabolites routinely quantified in vivo, gave a combined classification accuracy of 90% for SHH, Group3 and Group4. Glutamate predicted survival after incorporating known risk-factors (HR = 3.39, 95% CI 1.4-8.1, p = 0.025).

INTERPRETATION

Tissue metabolite profiles characterise medulloblastoma molecular groups. Their combination with machine learning can aid rapid diagnosis from tissue and potentially in vivo. Specific metabolites provide important information; GABA identifying WNT and glutamate conferring poor prognosis.

FUNDING

Children with Cancer UK, Cancer Research UK, Children's Cancer North and a Newcastle University PhD studentship.

The proteomic landscape of soft tissue sarcomas

Soft tissue sarcomas (STS) are rare and diverse mesenchymal cancers with limited treatment options. Here we undertake comprehensive proteomic profiling of tumour specimens from 321 STS patients representing 11 histological subtypes. Within leiomyosarcomas, we identify three proteomic subtypes with distinct myogenesis and immune features, anatomical site distribution and survival outcomes. Characterisation of undifferentiated pleomorphic sarcomas and dedifferentiated liposarcomas with low infiltrating CD3 + T-lymphocyte levels nominates the complement cascade as a candidate immunotherapeutic target. Comparative analysis of proteomic and transcriptomic profiles highlights the proteomic-specific features for optimal risk stratification in angiosarcomas. Finally, we define functional signatures termed Sarcoma Proteomic Modules which transcend histological subtype classification and show that a vesicle transport protein signature is an independent prognostic factor for distant metastasis. Our study highlights the utility of proteomics for identifying molecular subgroups with implications for risk stratification and therapy selection and provides a rich resource for future sarcoma research.

Molecular characterisation defines clinically-actionable heterogeneity within Group 4 medulloblastoma and improves disease risk-stratification

Group 4 tumours (MBGrp4) represent the majority of non-WNT/non-SHH medulloblastomas. Their clinical course is poorly predicted by current risk-factors. MBGrp4 molecular substructures have been identified (e.g. subgroups/cytogenetics/mutations), however their inter-relationships and potential to improve clinical sub-classification and risk-stratification remain undefined. We comprehensively characterised the paediatric MBGrp4 molecular landscape and determined its utility to improve clinical management. A clinically-annotated discovery cohort (n = 362 MBGrp4) was assembled from UK-CCLG institutions and SIOP-UKCCSG-PNET3, HIT-SIOP-PNET4 and PNET HR + 5 clinical trials. Molecular profiling was undertaken, integrating driver mutations, second-generation non-WNT/non-SHH subgroups (1-8) and whole-chromosome aberrations (WCAs). Survival models were derived for patients ≥ 3 years of age who received contemporary multi-modal therapies (n = 323). We first independently derived and validated a favourable-risk WCA group (WCA-FR) characterised by ≥ 2 features from chromosome 7 gain, 8 loss, and 11 loss. Remaining patients were high-risk (WCA-HR). Subgroups 6 and 7 were enriched for WCA-FR (p < 0·0001) and aneuploidy. Subgroup 8 was defined by predominantly balanced genomes with isolated isochromosome 17q (p < 0·0001). While no mutations were associated with outcome and overall mutational burden was low, WCA-HR harboured recurrent chromatin remodelling mutations (p = 0·007). Integration of methylation and WCA groups improved risk-stratification models and outperformed established prognostication schemes. Our MBGrp4 risk-stratification scheme defines: favourable-risk (non-metastatic disease and (i) subgroup 7 or (ii) WCA-FR (21% of patients, 5-year PFS 97%)), very-high-risk (metastatic disease with WCA-HR (36%, 5-year PFS 49%)) and high-risk (remaining patients; 43%, 5-year PFS 67%). These findings validated in an independent MBGrp4 cohort (n = 668). Importantly, our findings demonstrate that previously established disease-wide risk-features (i.e. LCA histology and MYC(N) amplification) have little prognostic relevance in MBGrp4 disease. Novel validated survival models, integrating clinical features, methylation and WCA groups, improve outcome prediction and re-define risk-status for ~ 80% of MBGrp4. Our MBGrp4 favourable-risk group has MBWNT-like excellent outcomes, thereby doubling the proportion of medulloblastoma patients who could benefit from therapy de-escalation approaches, aimed at reducing treatment induced late-effects while sustaining survival outcomes. Novel approaches are urgently required for the very-high-risk patients.

Medulloblastoma group 3 and 4 tumors comprise a clinically and biologically significant expression continuum reflecting human cerebellar development

Medulloblastoma is currently subclassified into distinct DNA methylation subgroups/subtypes with particular clinico-molecular features. Using RNA sequencing (RNA-seq) in large, well-annotated cohorts of medulloblastoma, we show that transcriptionally group 3 and group 4 medulloblastomas exist as intermediates on a bipolar continuum between archetypal group 3 and group 4 entities. Continuum position is prognostic, reflecting a propensity for specific DNA copy-number changes, and specific switches in isoform/enhancer usage and RNA editing. Examining single-cell RNA-seq (scRNA-seq) profiles, we show that intratumoral transcriptional heterogeneity along the continuum is limited in a subtype-dependent manner. By integrating with a human scRNA-seq reference atlas, we show that this continuum is mirrored by an equivalent continuum of transcriptional cell types in early fetal cerebellar development. We identify distinct developmental niches for all four major subgroups and link each to a common developmental antecedent. Our findings show a transcriptional continuum arising from oncogenic disruption of highly specific fetal cerebellar cell types, linked to almost every aspect of group 3/group 4 molecular biology and clinico-pathology.

MEDB-36. Clinical and molecular heterogeneity withinMYC andMYCN amplified medulloblastoma

MYC and MYCN are the most commonly amplified oncogenes in medulloblastoma. Their overall association with a poor prognosis has supported their adoption as high-risk disease biomarkers in trials. However, emerging evidence suggests that certain patients with MYN/MYCN focally-amplified tumours can achieve long-term survival and therefore may suffer unnecessary late-effects associated with intensified therapies. To investigate this heterogeneity, we characterised the molecular and clinico-pathological features of curated cohorts of MYC (n=64) and MYCN (n=95) amplified tumours, drawn from >1000 diagnostic cases, and assessed their associations with disease outcome. Within the MYCN-amplified cohort, survival was related to molecular group; patients with MYCN_Grp3 or MYCN_Grp4 tumours with no other clinico-pathological risk factors (subtotal resection (STR), metastatic disease, LCA pathology) were intermediate-risk (n=25;70% 5-year PFS). In contrast, a very-high-risk group was defined by positivity for MYCN_SHH, STR and/or LCA (n=64;32% 5-year PFS). 22/35 assessable MYCN_SHH harboured TP53 mutations; 9/12 with data were germline. MYC_Grp3 represented the majority (46/58; 79%) of molecularly-grouped MYC-amplified tumours. Importantly, while radiotherapy receipt conferred a modest survival advantage, for MYC-amplified tumours with additional clinico-molecular risk factors (LCA, metastasis, STR, Grp3), survival was dismal, irrespective of radiotherapy receipt. A very-high-risk group of MYC-amplified tumours was identified (n=51;10% 5-year PFS), defined by positivity for ≥1 additional risk factors (STR, LCA and/or metastasis). Alternatively, membership of subgroups II/V defined a smaller, very-high-risk patient group (n=28;7% 5-year PFS). Long-term survival was seen in the majority of remaining MYC-amplified tumours negative for these specified features (61% 5-year PFS; high-risk). MYC and MYCN-amplified medulloblastomas are biologically heterogeneous with diverse clinical outcomes. Molecular subgroup assignment and established clinical features are critical for their improved stratification. Patient subgroups identified may be eligible for therapy de-escalation; in contrast, the very-high-risk patient groups are incurable using current therapies and urgently require novel experimental treatment strategies upfront.

ATRT-20. Novel prognostic molecular signatures for improved risk-classification of Atypical Teratoid Rhabdoid Tumours

Malignant Rhabdoid Tumours (MRT) are aggressive paediatric malignancies seen in the central nervous system (Atypical Teratoid Rhabdoid Tumours (ATRT)), and kidney and other soft tissues (Extra-cranial Rhabdoid Tumours (ECRT)). With current therapies often proving ineffective and a lack of clear prognostic associations with consensus subgroups, we explored the possibility of using prognostic molecular signatures to further identify the biological characteristics of high risk ATRT patients. By employing a cross-validated feature selection method the methylation profiles of 121 MRT patients were analysed with clinical data to obtain meta-CpG signatures associated with prognosis for ATRT, ECRT and MRT. The relationship between these meta-CpG signatures and the consensus subgroups were further explored, along with the correlation of meta-CpGs with gene expression to establish biological significance. By selecting CpGs for their ability to predict survival this method obtained three novel prognostic methylation signatures which predict MRT outcome (ATRT-5, ECRT-14 and MRT-42). These signatures are independent of molecular subgroup and each signature was significantly associated with overall survival (OS) and event free survival (EFS) in their respective cohorts (p&lt;0.001). Both ATRT-5 and MRT-42 maintained their significant association with OS in an independent ATRT cohort (n=64) and each meta-CPG signature is prognostically independent of other major clinical risk factors (e.g. receipt of radiotherapy and presence of metastases). Biologically, individuals with high-risk methylation signatures showed a gene expression profile suggestive of higher proliferative rates and tumours with low-risk scores in ATRT-5 and MRT-42 had an upregulated inflammatory response and increased immune infiltration. Combining these meta-CpGs with other significant clinical risk-factors produced high performing multivariate Cox-models enabling us to propose new stratification models for ATRT and MRT patients. These subgroup-independent prognostic signatures represent a distinct biology in ATRT and, if validated in prospective studies, could progress the use and efficacy of precision-based medicine in this therapeutically challenging disease.

MEDB-71. Molecular characterisation of group 4 medulloblastoma improves risk-stratification and its biological understanding

Group 4 (MB_Grp4) accounts for ~40% of medulloblastoma and the majority of non-WNT/non-SHH cases, yet its underpinning biology is poorly understood, and survival outcomes are not sufficiently explained by established clinico-pathological risk factors. We investigated the clinical and molecular correlates of MB_Grp4, including second-generation methylation non-WNT/non-SHH subtypes (I-VIII) and whole chromosome aberration (WCA) subtypes (defined by chromosome 7 gain, 8 loss, and 11 loss; WCA-favourable risk [WCA-FR] ≥2 features, WCA-high risk [WCA-HR] ≤1 feature). A clinically-annotated MB_Grp4 discovery cohort (n=378) was assembled from UK CCLG institutions, collaborating centres and SIOP-UKCCSG-PNET3/HIT-SIOP-PNET4 clinical trials. Contemporary molecular profiling integrating methylation/WCA subtypes and next-generation sequencing was performed. Survival modelling was carried out with patients >3 years old who received craniospinal irradiation (n=336). Association analysis confirmed relationships between methylation and WCA subtypes. Subtypes VI and VII were enriched for WCA-FR (p<0.0001) and aneuploidy, whereas subtype VIII was defined solely by i17q (p<0.0001). Whilst we observed an overall low mutational burden, WCA-HR harboured recurrent mutations in genes involved in chromatin remodelling (p=0.007). No gene-specific events were associated with disease risk, however integration of both methylation subtype and WCA groups enabled improved risk-stratification survival models that outperformed current schemes. The optimal MB_Grp4-specific model stratified patients into: favourable-risk (local disease, subtype VII or subtype VI with WCA-FR; 39/194 [20%], 97% 5-year PFS), very-high-risk (metastatic disease with WCA-HR; 71/194 [37%], 50% 5-year PFS) and high-risk (remaining patients; 84/194 [43%], 67% 5-year PFS). Findings were validated in independent cohorts. Comprehensive clinico-molecular assessment of MB_Grp4 provides important understanding of its clinical and biological heterogeneity. Our novel MB_Grp4 stratification scheme removes standard risk disease and identifies a favourable risk group (20% of MB_Grp4) with potential for therapy de-escalation. Current therapeutic strategies are insufficient for the very-high risk group (encompassing 37% of MB_Grp4), for whom novel therapies are urgently required.

MEDB-79. MYC-driven upregulation of thede novo serine and glycine pathway is a novel therapeutic target for Group 3 MYC-amplified Medulloblastoma

Despite advances in the molecular sub-classification and risk-stratification of medulloblastoma (MB), a subset of tumours remain refractory to current multimodal therapies. Group 3 (MB_Group3) patients represent around 25% of MBs, and amplification and elevated expression of MYC in this group correlates with dismal clinical outcomes. Since direct targeting of MYC remains elusive, understanding and exploiting metabolic dependencies in MYC-amplified MB_Group3 may reveal novel therapeutic opportunities. We engineered three independent regulable MYC-amplified MB_Group3 cell-based models, each harbouring doxycycline-inducible anti-MYC shRNAs (two independent species) or a non-silencing shRNA control. In all three models, MYC knockdown (KD) revealed persistent MYC-dependent cancer phenotypes, reduction in proliferation and cell cycle progression. We utilised ¹H high-resolution magic angle spectroscopy (HRMAS) and stable isotope-resolved metabolomics to assess changes in intracellular metabolites and pathway dynamics when MYC expression was modulated. Profiling revealed consistent MYC-dependent changes in metabolite concentrations across models. Notably, glycine was consistently accumulated following MYC KD suggesting altered pathway dynamics. ¹³C-glucose tracing further revealed a reduction in glucose-derived serine and glycine (de novo synthesis) following MYC KD which was attributable to lower expression of PHGDH, the rate-limiting enzyme of this pathway. Furthermore, in human primary tumours, elevated expression of PHGDH was associated with MYC amplification and poorer survival outcomes. MYC expressing cells showed greater sensitivity to pharmacological inhibition of PHGDH compared to MYC KD (MB_Group3) and MB_SHH subgroup cell lines in vitro. Critically, targeting PHGDH in vivo, using MYC-dependent xenografts and genetically engineered mouse models, consistently slowed tumour progression and increased survival. In summary, metabolic profiling has uncovered MYC-dependent metabolic alterations and revealed the de novo serine/glycine synthesis pathway as a novel and clinically relevant therapeutic target in MYC-amplified MB_Group3. Together, these findings reveal metabolic vulnerabilities of MYC-amplified MB_Group3 which represent novel therapeutic opportunities for this poor-prognosis disease group.

Single-cell DNA sequencing identifies risk-associated clonal complexity and evolutionary trajectories in childhood medulloblastoma development

We reconstructed the natural history and temporal evolution of the most common childhood brain malignancy, medulloblastoma, by single-cell whole-genome sequencing (sc-WGS) of tumours representing its major molecular sub-classes and clinical risk groups. Favourable-risk disease sub-types assessed (MB_WNT and infant desmoplastic/nodular MB_SHH) typically comprised a single clone with no evidence of further evolution. In contrast, highest risk sub-classes (MYC-amplified MB_Group3 and TP53-mutated MB_SHH) were most clonally diverse and displayed gradual evolutionary trajectories. Clinically adopted biomarkers (e.g. chromosome 6/17 aberrations; CTNNB1/TP53 mutations) were typically early-clonal/initiating events, exploitable as targets for early-disease detection; in analyses of spatially distinct tumour regions, a single biopsy was sufficient to assess their status. Importantly, sc-WGS revealed novel events which arise later and/or sub-clonally and more commonly display spatial diversity; their clinical significance and role in disease evolution post-diagnosis now require establishment. These findings reveal diverse modes of tumour initiation and evolution in the major medulloblastoma sub-classes, with pathogenic relevance and clinical potential.

Emergence and maintenance of actionable genetic drivers at medulloblastoma relapse

Background

Less than 5% of medulloblastoma (MB) patients survive following failure of contemporary radiation-based therapies. Understanding the molecular drivers of medulloblastoma relapse (rMB) will be essential to improve outcomes. Initial genome-wide investigations have suggested significant genetic divergence of the relapsed disease.

Methods

We undertook large-scale integrated characterization of the molecular features of rMB—molecular subgroup, novel subtypes, copy number variation (CNV), and driver gene mutation. 119 rMBs were assessed in comparison with their paired diagnostic samples (n = 107), alongside an independent reference cohort sampled at diagnosis (n = 282). rMB events were investigated for association with outcome post-relapse in clinically annotated patients (n = 54).

Results

Significant genetic evolution occurred over disease-course; 40% of putative rMB drivers emerged at relapse and differed significantly between molecular subgroups. Non-infant MBSHH displayed significantly more chromosomal CNVs at relapse (TP53 mutation-associated). Relapsed MB_Group4 demonstrated the greatest genetic divergence, enriched for targetable (eg, CDK amplifications) and novel (eg, USH2A mutations) events. Importantly, many hallmark features of MB were stable over time; novel subtypes (>90% of tumors) and established genetic drivers (eg, SHH/WNT/P53 mutations; 60% of rMB events) were maintained from diagnosis. Critically, acquired and maintained rMB events converged on targetable pathways which were significantly enriched at relapse (eg, DNA damage signaling) and specific events (eg, 3p loss) predicted survival post-relapse.

Conclusions

rMB is characterised by the emergence of novel events and pathways, in concert with selective maintenance of established genetic drivers. Together, these define the actionable genetic landscape of rMB and provide a basis for improved clinical management and development of stratified therapeutics, across disease-course.

Abstract 2120: Interrogation of regulatory and expression changes in Malignant Rhabdoid Tumours to identify new therapeutic approaches

Malignant Rhabdoid Tumors (MRTs) are a textbook model of epigenetic dysregulation within cancer. The biallelic inactivation of a single gene SMARCB1 in &gt;95% of MRTs is sufficient to provoke a lethal pediatric cancer. SMARCB1 encodes a core subunit of the SWI/SNF complex and inactivation results in genome-wide deregulation of gene expression/pathways. SMARCB1 loss was previously shown to impair SWI/SNF enhancer targeting at TSS-distal enhancers and a model proposed whereby mutant residual SWI/SNF complexes preferentially binds to super-enhancers over typical-enhancers; thereby inappropriately maintaining progenitor lineage processes and blocking differentiation. This model deemphasizes the action of mutant SWI/SNF complexes upon cell cycle/oncogenic processes. We performed a consensus analysis on a cohort of novel and published ChIP-seq profiles (n=6 +/- SMARCB1 models) to define consensus SMARCB1-dependent epigenetic changes. We further used parallel RNA-seq, methylation and whole-genome CRISPR screen data to emphasize the significance of specific regulatory processes from the perspective of likely biological effects and functional essentiality. We bioinformatically identified consensus super and typical-enhancers from H3K27ac peaks, multiple datasets enabled high confidence regions to be called. SMARCB1 loss did not significantly impact SWI/SNF occupancy of super-enhancers, permutation test revealed significant (p&lt;0.001) overlap of binding across both conditions; in agreement with residual SWI/SNF complex binding theories. However, we noted the same at typical-enhancers implying all enhancers are similarly affected. We hypothesized that the resulting SMARCB1-dependent transcriptional impact, not location of SWI/SNF binding would be paramount, InTAD and RNAseq data were used to assign target genes to enhancers. Of the SMARCB1-dependent genes linked to typical-enhancers 13% showed significantly altered expression compared to 9% for super-enhancers. Ontology analysis identified classic oncogenic pathways such as cell cycle and metabolic genes as enriched among SMARCB1-dependent enhancer targets. Gene Set Enrichment Analysis (GSEA) confirmed significant enrichment of mucopolysaccharide metabolic genes (p&lt;0.002, NES 1.42). Correlating dysregulated gene targets to epigenetically altered enhancers/promoters we ranked loci which drive MRT tumorigenesis. We identified a 40kb upstream region of SMARCB1-dependent methylation and ChIPseq binding strongly correlated to expression of CDK6. We further employed CRISPR activation/repression tools to mimic SMARCB1 re-expression and model their effects. Successful modulation of candidate loci will help define potential future novel targeted therapeutics - e.g use of HDACi - to recapitulate the effects of SMARCB1 re-expression.

Citation Format: Emma L. Lishman-Walker, Martina A. Finetti, James P. Hacking, Matthew Selby, Yura Grabovska, Stephen Crosier, Simon Bailey, Steven C. Clifford, Daniel Williamson. Interrogation of regulatory and expression changes in Malignant Rhabdoid Tumours to identify new therapeutic approaches [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2120.

EMBR-25. GENOME-WIDE GENETIC AND EPIGENETIC ASSESSMENT OF GROUP 4 MEDULLOBLASTOMA FOR IMPROVED, BIOMARKER DRIVEN, PROGNOSTICATION AND RISK-STRATIFICATION

Introduction

Medulloblastoma (MB) is the most common malignant brain tumour in children. The most frequent molecular subgroup, Group 4 (MBGrp4) accounts for ~35/40% of cases, however it has the least understood underlying biology. Clinical outcomes are heterogeneous in MBGrp4 and are not accounted for by established clinico-pathological risk factors. There is now a requirement for a comprehensive study of MBGrp4, considering established clinico-pathological features and novel molecular biomarkers to enhance risk-stratification and identify novel therapeutic targets.

Methods

A clinically-annotated, retrospective MBGrp4 discovery cohort (n = 420) was generated from UK CCLG institutions, collaborating European centres and SIOP-UKCCSG-PNET3 and HIT-SIOP-PNET4 clinical trials. Contemporary, multi-omics profiling was performed. Focal and arm level copy number aberrations (CNAs) were determined from molecular inversion probe (MIP) or DNA methylation array which additionally provided next generation non-WNT/non-SHH (Grp3/Grp4) subtype classifications. Targeted next-generation DNA sequencing was performed to overlay the mutational landscape. Survival modelling was carried out with patients &gt;3 years old who received craniospinal irradiation.

Results

MBGrp4 subtypes were assigned to 88% of tumours with available data. Subtype VIII was strongly associated with i17q (p&lt;0.0001). The favourable-risk cytogenetic signature (2 or 3 of; chromosome 7 gain, chromosome 8 loss and/or chromosome 11 loss) associated with both subtypes VI and VII (p&lt;0.0001). MYCN amplifications were strongly associated with subtype V (p&lt;0.0001) in addition to 16q loss (p&lt;0.0001). The high-risk CNA group was enriched for mutations in genes involved in chromatin remodelling (p&lt;0.0001). Risk factors were identified from multivariate survival modelling. Subtype and CNA groups contributed to improved risk-stratification models that outperformed current clinical schemes.

Conclusion

Comprehensive genetic and epigenetic profiling in this large retrospective cohort has improved our understanding of the molecular and clinical heterogeneity within MBGrp4. Incorporation of molecular biomarkers improved risk-stratification for MBGrp4.

Advanced molecular pathology for rare tumours: A national feasibility study and model for centralised medulloblastoma diagnostics

Aims

Application of advanced molecular pathology in rare tumours is hindered by low sample numbers, access to specialised expertise/technologies and tissue/assay QC and rapid reporting requirements. We assessed the feasibility of co‐ordinated real‐time centralised pathology review (CPR), encompassing molecular diagnostics and contemporary genomics (RNA‐seq/DNA methylation‐array).

Methods

This nationwide trial in medulloblastoma (<80 UK diagnoses/year) introduced a national reference centre (NRC) and assessed its performance and reporting to World Health Organisation standards. Paired frozen/formalin‐fixed, paraffin‐embedded tumour material were co‐submitted from 135 patients (16 referral centres).

Results

Complete CPR diagnostics were successful for 88% (120/135). Inadequate sampling was the most common cause of failure; biomaterials were typically suitable for methylation‐array (129/135, 94%), but frozen tissues commonly fell below RNA‐seq QC requirements (53/135, 39%). Late reporting was most often due to delayed submission. CPR assigned or altered histological variant (vs local diagnosis) for 40/135 tumours (30%). Benchmarking/QC of specific biomarker assays impacted test results; fluorescent in‐situ hybridisation most accurately identified high‐risk MYC/MYCN amplification (20/135, 15%), while combined methods (CTNNB1/chr6 status, methylation‐array subgrouping) best defined favourable‐risk WNT tumours (14/135; 10%). Engagement of a specialist pathologist panel was essential for consensus assessment of histological variants and immunohistochemistry. Overall, CPR altered clinical risk‐status for 29% of patients.

Conclusion

National real‐time CPR is feasible, delivering robust diagnostics to WHO criteria and assignment of clinical risk‐status, significantly altering clinical management. Recommendations and experience from our study are applicable to advanced molecular diagnostics systems, both local and centralised, across rare tumour types, enabling their application in biomarker‐driven routine diagnostics and clinical/research studies.

MBRS-44. TIME, PATTERN AND OUTCOME OF MEDULLOBLASTOMA RELAPSE ARE ASSOCIATED WITH TUMOUR BIOLOGY AT DIAGNOSIS AND UPFRONT THERAPY: A COHORT STUDY

Disease relapse occurs in ~30% of children with medulloblastoma, and is fatal in the majority. We sought to establish whether clinico-molecular characteristics at diagnosis are associated with the nature of relapse, subsequent disease-course, and whether these associations could inform clinical management. We surveyed the clinical features of medulloblastoma relapse (time-to-relapse, pattern-of-relapse, time-to-death and overall outcome) in 247 centrally-reviewed patients who relapsed following standard-upfront-therapies. We related these to clinico-molecular features at diagnosis, prognostic factors, and first-line/relapse treatment. Patients who received upfront craniospinal irradiation (CSI-treated) displayed prolonged time-to-relapse compared to CSI naïve patients (p<0.001). Similarly, in CSI naïve patients, CSI at relapse, alongside re-resection and desmoplastic/nodular histology, were associated with long-term survival. In CSI-treated patients, the nature of relapse was subgroup-dependent. Local-nodular relapse patterns were enriched in relapsed-MB_SHH patients (p<0.001), but a notable proportion (65%) also acquired distant-diffuse disease (p=0.010). MB_Group3 relapsed quickly (median 1.3 years), MB_Group4 slowly (median 2.1 years). Distant-disease was prevalent in MB_Group3 and MB_Group4 relapses (90%) but, in contrast to relapsed-MB_SHH, nodular and diffuse patterns of distant-disease were observed. Furthermore, nodular disease was associated with a prolonged time-to-death post-relapse (p=0.006). Investigation of second-generation MB_Group3/4 subtypes refined our understanding of heterogeneous relapse characteristics. Subtype VIII had prolonged time-to-relapse; subtype II a rapid time-to-death. Subtypes II/III/VIII developed a significantly higher incidence of distant-disease at relapse, whereas subtypes V/VII did not. The nature of medulloblastoma relapse are biology and therapy-dependent, providing immediate translational opportunities for improved disease management through biology-directed surveillance, post-relapse prognostication and risk-stratified selection of second-line treatment.

MBRS-69. METABOLITE PROFILING OF SHH MEDULLOBLASTOMA IDENTIFIES A SUBSET OF CHILDHOOD TUMOURS ENRICHED FOR HIGH-RISK MOLECULAR BIOMARKERS AND CLINICAL FEATURES

SHH medulloblastoma patients have a variable prognosis. Infants (&lt;3–5 years at diagnosis) are associated with a good prognosis, while disease-course in childhood is associated with specific prognostic biomarkers (MYCN amplification, TP53 mutation, LCA histology; all high-risk). There is an unmet need to identify prognostic subgroups of SHH tumours rapidly in the clinical setting, to aid in real-time risk stratification and disease management. Metabolite profiling is a powerful technique for characterising tumours. High resolution magic angle spinning NMR spectroscopy (HR-MAS) can be performed on frozen tissue samples and provides high quality metabolite information. We therefore assessed whether metabolite profiles could identify subsets of SHH tumours with prognostic potential. Metabolite concentrations of 22 SHH tumours were acquired by HR-MAS and analysed using unsupervised hierarchical clustering. Methylation profiling assigned the infant and childhood SHH subtypes, and clinical and molecular features were compared between clusters. Two clusters were observed. A significantly higher concentration of lipids was observed in Cluster 1 (t-test, p=0.012). Cluster 1 consisted entirely of childhood-SHH whilst Cluster 2 included both childhood-SHH and infant-SHH subtypes. Cluster 1 was enriched for high-risk markers - LCA histology (3/7 v. 0/5), MYCN amplification (2/7 v. 0/5), TP53 mutations (3/7 v. 1/5) and metastatic disease - whilst having a lower proportion of TERT mutations (0/7 v. 2/5) than Cluster 2. These pilot results suggest that (i) it is possible to identify childhood-SHH patients linked to high-risk clinical and molecular biomarkers using metabolite profiles and (ii) these may be detected non-invasively in vivo using magnetic-resonance spectroscopy.

ATRT-25. INTEGRATED QUANTITATIVE SWATH-MS PROTEOMICS ANALYSIS OF ATRTs UNCOVERS NEW THERAPEUTIC OPPORTUNITIES

The consequences of SMARCB1 loss in Atypical Teratoid Rhabdoid Tumors (ATRTs) have been extensively characterized at the epigenetic/transcriptomic level. In this study we detail the functional effect of SMARCB1 mutation on the MRT proteome, its relationship with RNA deregulation or lack thereof. We performed unlabeled discovery proteomics using MS-SWATH on MRT cells in which SMARCB1 was forcibly re-expressed (5 cell lines, +/-SMARCB1); analyzing changes in protein abundance within 3 fractions (total, membrane, nuclear). We generated a custom spectral library, covering >8,000 proteins, for analysis of the ATRT proteome using a pH fractionated pool of each cellular subfraction. This SMARCB1-dependent ATRT spectral library constitutes a powerful tool for profiling proteins of potentially therapeutic relevance in both model systems and primary ATRT samples. We show that whilst gene expression and protein abundance are significantly related there are many instances whereby expression changes do not reliably predict protein abundances. Several hundred proteins show significantly increased abundance (p<0.01) with no concomitant change by RNA-seq. SMARCB1 mutation is able to invoke critical changes in post-transcriptional/translational regulation as well as sub-cellular localization. By integration with whole-genome CRISPR/cas9 screening we describe functionally essential SMARCB1 dependent pathway/membrane biomarkers, evident at the protein but not the RNA level. We describe several which are druggable and suggest certain therapeutic modalities e.g. specific combinations of RTKs, RNA-binding proteins/splicing factors (SpliceosomeA, U4:U5:U6 tri-snRNP complexes). Our analysis links, for the first time in ATRT, genome-wide transcriptomic and proteome aberrations and reveals broad mechanisms underlying the effect of SMARCB1 mutation.

MBRS-59. SINGLE-CELL WHOLE-GENOME SEQUENCING DISSECTS INTRA-TUMOURAL GENOMIC HETEROGENEITY AND CLONAL EVOLUTION IN CHILDHOOD MEDULLOBLASTOMA

Medulloblastomas harbor clinically-significant intra-tumoral heterogeneity for key biomarkers (e.g. MYC/MYCN, β-catenin). Recent studies have characterized transcriptional heterogeneity at the single-cell level, however the underlying genomic copy number and mutational architecture remains to be resolved. We therefore sought to establish the intra-tumoural genomic heterogeneity of medulloblastoma at single-cell resolution. Copy number patterns were dissected by whole-genome sequencing in 1024 single cells isolated from multiple distinct tumour regions within 16 snap-frozen medulloblastomas, representing the major molecular subgroups (WNT, SHH, Group3, Group4) and genotypes (i.e. MYC amplification, TP53 mutation). Common copy number driver and subclonal events were identified, providing clear evidence of copy number evolution in medulloblastoma development. Moreover, subclonal whole-arm and focal copy number alterations covering important genomic loci (e.g. on chr10 of SHH patients) were detected in single tumour cells, yet undetectable at the bulk-tumor level. Spatial copy number heterogeneity was also common, with differences between clonal and subclonal events detected in distinct regions of individual tumours. Mutational analysis of the cells allowed dissection of spatial and clonal heterogeneity patterns for key medulloblastoma mutations (e.g. CTNNB1, TP53, SMARCA4, PTCH1) within our cohort. Integrated copy number and mutational analysis is underway to establish their inter-relationships and relative contributions to clonal evolution during tumourigenesis. In summary, single-cell analysis has enabled the resolution of common mutational and copy number drivers, alongside sub-clonal events and distinct patterns of clonal and spatial evolution, in medulloblastoma development. We anticipate these findings will provide a critical foundation for future improved biomarker selection, and the development of targeted therapies.

Time, pattern, and outcome of medulloblastoma relapse and their association with tumour biology at diagnosis and therapy: a multicentre cohort study

Background

Disease relapse occurs in around 30% of children with medulloblastoma, and is almost universally fatal. We aimed to establish whether the clinical and molecular characteristics of the disease at diagnosis are associated with the nature of relapse and subsequent disease course, and whether these associations could inform clinical management.

Methods

In this multicentre cohort study we comprehensively surveyed the clinical features of medulloblastoma relapse (time to relapse, pattern of relapse, time from relapse to death, and overall outcome) in centrally reviewed patients who relapsed following standard upfront therapies, from 16 UK Children's Cancer and Leukaemia Group institutions and four collaborating centres. We compared these relapse-associated features with clinical and molecular features at diagnosis, including established and recently described molecular features, prognostic factors, and treatment at diagnosis and relapse.

Findings

247 patients (175 [71%] boys and 72 [29%] girls) with medulloblastoma relapse (median year of diagnosis 2000 [IQR 1995–2006]) were included in this study. 17 patients were later excluded from further analyses because they did not meet the age and treatment criteria for inclusion. Patients who received upfront craniospinal irradiation (irradiated group; 178 [72%] patients) had a more prolonged time to relapse compared with patients who did not receive upfront craniospinal irradiation (non-irradiated group; 52 [21%] patients; p<0·0001). In the non-irradiated group, craniospinal irradiation at relapse (hazard ratio [HR] 0·27, 95% CI 0·11–0·68) and desmoplastic/nodular histology (0·23, 0·07–0·77) were associated with prolonged time to death after relapse, MYC amplification was associated with a reduced overall survival (23·52, 4·85–114·05), and re-resection at relapse was associated with longer overall survival (0·17, 0·05–0·57). In the irradiated group, patients with MB_Group3 tumours relapsed significantly more quickly than did patients with MB_Group4 tumours (median 1·34 [0·99–1·89] years vs 2·04 [1·39–3·42 years; p=0·0043). Distant disease was prevalent in patients with MB_Group3 (23 [92%] of 25 patients) and MB_Group4 (56 [90%] of 62 patients) tumour relapses. Patients with distantly-relapsed MB_Group3 and MB_Group4 displayed both nodular and diffuse patterns of disease whereas isolated nodular relapses were rare in distantly-relapsed MB_SHH (1 [8%] of 12 distantly-relapsed MB_SHH were nodular alone compared with 26 [34%] of 77 distantly-relapsed MB_Group3 and MB_Group4). In MB_Group3 and MB_Group4, nodular disease was associated with a prolonged survival after relapse (HR 0·42, 0·21–0·81). Investigation of second-generation MB_Group3 and MB_Group4 molecular subtypes refined our understanding of heterogeneous relapse characteristics. Subtype VIII had prolonged time to relapse and subtype II had a rapid time from relapse to death. Subtypes II, III, and VIII developed a significantly higher incidence of distant disease at relapse whereas subtypes V and VII did not (equivalent rates to diagnosis).

Interpretation

This study suggests that the nature and outcome of medulloblastoma relapse are biology and therapy-dependent, providing translational opportunities for improved disease management through biology-directed disease surveillance, post-relapse prognostication, and risk-stratified selection of second-line treatment strategies.

Funding

Cancer Research UK, Action Medical Research, The Tom Grahame Trust, The JGW Patterson Foundation, Star for Harris, The Institute of Child Health - Newcastle University - Institute of Child Health High-Risk Childhood Brain Tumour Network (co-funded by The Brain Tumour Charity, Great Ormond Street Children's Charity, and Children with Cancer UK).

The molecular landscape and associated clinical experience in infant medulloblastoma: prognostic significance of second-generation subtypes

AIMS

Biomarker-driven therapies have not been developed for infant medulloblastoma (iMB). We sought to robustly sub-classify iMB, and proffer strategies for personalized, risk-adapted therapies.

METHODS

We characterized the iMB molecular landscape, including second-generation subtyping, and the associated retrospective clinical experience, using large independent discovery/validation cohorts (n = 387).

RESULTS

iMB_Grp3 (42%) and iMB_SHH (40%) subgroups predominated. iMB_Grp3 harboured second-generation subtypes II/III/IV. Subtype II strongly associated with large-cell/anaplastic pathology (LCA; 23%) and MYC amplification (19%), defining a very-high-risk group (0% 10yr overall survival (OS)), which progressed rapidly on all therapies; novel approaches are urgently required. Subtype VII (predominant within iMB_Grp4 ) and subtype IV tumours were standard risk (80% OS) using upfront CSI-based therapies; randomized-controlled trials of upfront radiation-sparing and/or second-line radiotherapy should be considered. Seventy-five per cent of iMB_SHH showed DN/MBEN histopathology in discovery and validation cohorts (P < 0.0001); central pathology review determined diagnosis of histological variants to WHO standards. In multivariable models, non-DN/MBEN pathology was associated significantly with worse outcomes within iMB_SHH . iMB_SHH harboured two distinct subtypes (iMB_SHH-I/II ). Within the discriminated favourable-risk iMB_SHH DN/MBEN patient group, iMB_SHH-II had significantly better progression-free survival than iMB_SHH-I , offering opportunities for risk-adapted stratification of upfront therapies. Both iMB_SHH-I and iMB_SHH-II showed notable rescue rates (56% combined post-relapse survival), further supporting delay of irradiation. Survival models and risk factors described were reproducible in independent cohorts, strongly supporting their further investigation and development.

CONCLUSIONS

Investigations of large, retrospective cohorts have enabled the comprehensive and robust characterization of molecular heterogeneity within iMB. Novel subtypes are clinically significant and subgroup-dependent survival models highlight opportunities for biomarker-directed therapies.

Infant High-Grade Gliomas Comprise Multiple Subgroups Characterized by Novel Targetable Gene Fusions and Favorable Outcomes

Infant high-grade gliomas appear clinically distinct from their counterparts in older children, indicating that histopathologic grading may not accurately reflect the biology of these tumors. We have collected 241 cases under 4 years of age, and carried out histologic review, methylation profiling, and custom panel, genome, or exome sequencing. After excluding tumors representing other established entities or subgroups, we identified 130 cases to be part of an "intrinsic" spectrum of disease specific to the infant population. These included those with targetable MAPK alterations, and a large proportion of remaining cases harboring gene fusions targeting ALK (n = 31), NTRK1/2/3 (n = 21), ROS1 (n = 9), and MET (n = 4) as their driving alterations, with evidence of efficacy of targeted agents in the clinic. These data strongly support the concept that infant gliomas require a change in diagnostic practice and management. SIGNIFICANCE: Infant high-grade gliomas in the cerebral hemispheres comprise novel subgroups, with a prevalence of ALK, NTRK1/2/3, ROS1, or MET gene fusions. Kinase fusion-positive tumors have better outcome and respond to targeted therapy clinically. Other subgroups have poor outcome, with fusion-negative cases possibly representing an epigenetically driven pluripotent stem cell phenotype.See related commentary by Szulzewsky and Cimino, p. 904.This article is highlighted in the In This Issue feature, p. 890.

Dual Targeting of PDGFRα and FGFR1 Displays Synergistic Efficacy in Malignant Rhabdoid Tumors

Subunits of the SWI/SNF chromatin remodeling complex are mutated in a significant proportion of human cancers. Malignant rhabdoid tumors (MRTs) are lethal pediatric cancers characterized by a deficiency in the SWI/SNF subunit SMARCB1. Here, we employ an integrated molecular profiling and chemical biology approach to demonstrate that the receptor tyrosine kinases (RTKs) PDGFRα and FGFR1 are coactivated in MRT cells and that dual blockade of these receptors has synergistic efficacy. Inhibitor combinations targeting both receptors and the dual inhibitor ponatinib suppress the AKT and ERK1/2 pathways leading to apoptosis. MRT cells that have acquired resistance to the PDGFRα inhibitor pazopanib are susceptible to FGFR inhibitors. We show that PDGFRα levels are regulated by SMARCB1 expression, and assessment of clinical specimens documents the expression of both PDGFRα and FGFR1 in rhabdoid tumor patients. Our findings support a therapeutic approach in cancers with SWI/SNF deficiencies by exploiting RTK coactivation dependencies.

•

Malignant rhabdoid tumors display coactivation of PDGFRα and FGFR1

•

Dual inhibition of PDGFRα and FGFR1 leads to synergistic apoptosis

•

FGFR1 inhibition overcomes acquired resistance to pazopanib treatment

•

PDGFRα and FGFR1 are expressed in rhabdoid tumor patient specimens

Minimal methylation classifier (MIMIC): A novel method for derivation and rapid diagnostic detection of disease-associated DNA methylation signatures

Rapid and reliable detection of disease-associated DNA methylation patterns has major potential to advance molecular diagnostics and underpin research investigations. We describe the development and validation of minimal methylation classifier (MIMIC), combining CpG signature design from genome-wide datasets, multiplex-PCR and detection by single-base extension and MALDI-TOF mass spectrometry, in a novel method to assess multi-locus DNA methylation profiles within routine clinically-applicable assays. We illustrate the application of MIMIC to successfully identify the methylation-dependent diagnostic molecular subgroups of medulloblastoma (the most common malignant childhood brain tumour), using scant/low-quality samples remaining from the most recently completed pan-European medulloblastoma clinical trial, refractory to analysis by conventional genome-wide DNA methylation analysis. Using this approach, we identify critical DNA methylation patterns from previously inaccessible cohorts, and reveal novel survival differences between the medulloblastoma disease subgroups with significant potential for clinical exploitation.

Abstract 4875: Identification of the cellular origin and "stemness" phenotype of Malignant Rhabdoid Tumors (MRT) may represent a new therapeutic approach in paediatric oncology

Malignant Rhabdoid Tumors (MRT) are highly aggressive childhood malignancies characterized by a single mutation; biallelic inactivation of SMARCB1, a component of the SWI/SNF chromatin remodeling complex. These tumors may occur anywhere, most frequently in the brain (Atypical Teratoid/Rhabdoid Tumor, ATRT) and in the kidneys and soft tissues (Extra Cranial Rhabdoid tumor, ECRT). MRT presents immune markers from multiple lineages exhibiting a teratoid phenotype, characterized by cells with heterogeneous morphology within the same tumor leading to speculation that the MRT cell of origin is a type of multipotent stem cell. Despite recent advances in treating other solid tumors, treatment for MRT still remains ineffective and we hypothesized that MRT “stemness” characteristics are key to its aggressive clinical course and resistance to treatment.

Here we analyze the expression profiles of primary MRT (n= 134) and catalog the expression of a program of “stemness” genes capable of driving aberrant self-renewal. Further, we show by re-expression of SMARCB1 in MRT cells that several of these key “stemness” genes are aberrantly activated by SMARCB1 mutation. One such gene, BMI1 was further demonstrated to be critical to MRT tumorigenesis by shRNA knockout, and a novel anti-BMI1 drug, PTC209 was tested to show efficacy in MRT. Knockdown of BMI1 in MRT cells reduced the self-renewal capability of cells as seen from the number of tumorspheres formed in limiting dilution assay (LDA). Interestingly, BMI1 loss triggers upregulation of p16(INK4a) expression and this mimics the expression profile when SMARCB1 was re-expressed into MRT cells.

To identify a putative MRT cell of origin we performed a meta-analysis cross-referencing expression profiles from primary MRT (n= 20), and functional models in which SMARCB1 was re-expressed (n=5 lines) with expression profiles from multiple stem cell types including epithelial, embryonic, neural mesenchymal and neural crest (n= 446). Analysis by t-SNE and NMF consensus clustering suggested that MRT bore the closest resemblance to either early neural progenitors (ATRT) or early neural crest cells (ECRT). In MRT cells in which SMARCB1 was forcibly re-expressed, cells became gradually more differentiated and their expression profile altered from that of an early neural crest to a more differentiated mesenchymal pattern.

For the first time, we show evidence that ECRTs and ATRTs appear to have different cells of origin, despite having the same mutation and tumor appearance. We further demonstrate that SMARCB1 expression is necessary for MRT cells to maintain a de-differentiated “stem-like” state and finally that the SMARCB1-dependent stemness gene BMI1 shows potential as a novel therapeutic target in MRT.

Citation Format: Ras A. Ramli, Martina A. Finetti, Matthew P. Selby, Yura Grabovska, Stephen Crosier, Amanda J. Smith, Steven C. Clifford, Daniel Williamson. Identification of the cellular origin and "stemness" phenotype of Malignant Rhabdoid Tumors (MRT) may represent a new therapeutic approach in paediatric oncology [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4875. doi:10.1158/1538-7445.AM2017-4875

Combined MYC and P53 defects emerge at medulloblastoma relapse and define rapidly progressive, therapeutically targetable disease

We undertook a comprehensive clinical and biological investigation of serial medulloblastoma biopsies obtained at diagnosis and relapse. Combined MYC family amplifications and P53 pathway defects commonly emerged at relapse, and all patients in this group died of rapidly progressive disease postrelapse. To study this interaction, we investigated a transgenic model of MYCN-driven medulloblastoma and found spontaneous development of Trp53 inactivating mutations. Abrogation of p53 function in this model produced aggressive tumors that mimicked characteristics of relapsed human tumors with combined P53-MYC dysfunction. Restoration of p53 activity and genetic and therapeutic suppression of MYCN all reduced tumor growth and prolonged survival. Our findings identify P53-MYC interactions at medulloblastoma relapse as biomarkers of clinically aggressive disease that may be targeted therapeutically.

c-MYC is a radiosensitive locus in human breast cells

Ionising radiation is a potent human carcinogen. Epidemiological studies have shown that adolescent and young women are at increased risk of developing breast cancer following exposure to ionising radiation compared with older women, and that risk is dose-dependent. Although it is well understood which individuals are at risk of radiation-induced breast carcinogenesis, the molecular genetic mechanisms that underlie cell transformation are less clear. To identify genetic alterations potentially responsible for driving radiogenic breast transformation, we exposed the human breast epithelial cell line MCF-10A to fractionated doses of X-rays and examined the copy number and cytogenetic alterations. We identified numerous alterations of c-MYC that included high-level focal amplification associated with increased protein expression. c-MYC amplification was also observed in primary human mammary epithelial cells following exposure to radiation. We also demonstrate that the frequency and magnitude of c-MYC amplification and c-MYC protein expression is significantly higher in breast cancer with antecedent radiation exposure compared with breast cancer without a radiation aetiology. Our data also demonstrate extensive intratumor heterogeneity with respect to c-MYC copy number in radiogenic breast cancer, suggesting continuous evolution at this locus during disease development and progression. Taken together, these data identify c-MYC as a radiosensitive locus, implicating this oncogenic transcription factor in the aetiology of radiogenic breast cancer.

Abstract LB-201: MYC and TP53 defects interact at medulloblastoma relapse to define rapidly progressive disease and can be targeted therapeutically

Disease recurrence following multi-modal therapy is the single most adverse event in medulloblastoma (MB). Currently &gt;90% of relapsing patients die, accounting for ∼10% of childhood cancer deaths. MB is heterogeneous at diagnosis, comprising four molecular subgroups with distinct clinicopathological and molecular features and outcomes. The relevance of these features at relapse is unknown, making characterisation, modelling and targeted therapy of relapse biology essential to improve outcomes. However, relapsed MBs are not routinely biopsied in clinical practice.

We undertook a first comprehensive investigation of the molecular, clinical and pathological features of 29 relapsed MBs and paired tumour samples taken at diagnosis, including the assessment of features with established significance at diagnosis (e.g. chromosome 17 and TP53 pathway status, MYC family (MYC, MYCN) gene amplification, polyploidy, CTNNB1 mutation and molecular subgroup status). Molecular subgroup was concordant at diagnosis and relapse, however evidence of alteration of all other features examined was found in relapsed tumours, with the majority of changes (30/44) representing acquired high-risk events. Most notably, MYC family gene amplifications and TP53 pathway defects commonly emerged in combination at relapse following conventional multimodal treatment (P=0.02, 7/22, 32%) and predicted rapid progression to death (P=0.016).

These observations suggested aberrant activation of MYC family genes synergizes with TP53 inactivation in the genesis of biologically aggressive MB. To investigate any such relationship, we examined Trp53 status in our transgenic mouse model of spontaneously-arising MYCN-driven MB (GTML; Glt1-tTA/TRE-MYCN-Luc). Somatic Trp53 mutations were found in 83% of tumors (n=10/12). Direct modelling of this interaction in GTML/Trp53KI/KI mice dramatically enhanced MB formation with 100% penetrance (43/43, median survival 47 days) in GTML/Trp53KI/KI versus 6% (3/50) in GTML; P&lt;0.0001), faithfully mimicked clinicopathological characteristics of TP53-MYC family gene-associated relapsed human tumors, and validated the essential role of TP53 in potentiating the growth of MYCN-driven MB. Finally, therapeutic inhibition of Aurora-A kinase using MLN8237 in these tumours, and in derived neurospheres in vitro, promoted degradation of MYCN, reduced tumor growth and prolonged survival.

In summary, while subgroup status remains stable, MBs display altered molecular, pathological and clinical features at relapse, and the emergence of combined TP53-MYC family gene defects is common following conventional therapy. Their association with rapid demise, coupled with their biological validation as driving and therapeutically exploitable events in a novel mouse MB model, strongly support further investigation and routine biopsy of relapse disease to drive future individualised therapeutic strategies.

Citation Format: Rebecca M. Hill, Sanne Kuijper, Janet Lindsey, Ed C. Schwalbe, Karen Barker, Jessica Boult, Daniel Williamson, Zai Ahmad, Albert Hallsworth, Sarra Ryan, Evon Poon, Simon Robinson, Ruth Ruddle, Florence Raynaud, Louise Howell, Colin Kwok, Abhijit Joshi, Sarah Nicholson, Stephen Crosier, Stephen Wharton, Tom Jacques, Keith Robson, Antony Michalski, Darren Hargrave, Barry Pizer, Simon Bailey, Fredrik J. Swartling, Kevin Petrie, William A. Weiss, Louis Chesler, Steve Clifford. MYC and TP53 defects interact at medulloblastoma relapse to define rapidly progressive disease and can be targeted therapeutically. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-201. doi:10.1158/1538-7445.AM2014-LB-201

A 9 series microRNA signature differentiates between germinal centre and activated B-cell-like diffuse large B-cell lymphoma cell lines

The microRNAs are endogenous, non-coding RNAs that play key roles in a range of pathophysiological processes by up- or down-regulating gene expression. Recent studies have shown that some microRNAs have oncogenic or tumour suppressor activity. Diffuse large B-cell lymphoma (DLBCL) is an aggressive non-Hodgkin's lymphoma with a heterogeneous biology, which has impeded the clinical assessment of patients. The currently-used clinically-based IPI provides useful information for treatment decision making, but has limited predictive power. Recent immunohistochemical approaches have identified two different prognostic groups: the more indolent germinal centre (GC)- and the higher risk activated B-cell (ABC)-like phenotypes. Although useful, prediction based on immunophenotype has limitations. The present study uses microRNA profiling and a number of well-characterised B-cell lymphoma cell lines to identify microRNA signatures that are correctly assigned to the DLBCL prognostic subgroups and distinguish DLBCL from other more indolent lymphoma, including follicular lymphoma (FL). MicroRNA microarray analysis was based on miRBase version 12.0 and analysis was performed using an unsupervised hierarchical clustering model. Discriminatory microRNAs were validated by qRT-PCR. We identified a 9 microRNA signature that discriminated between ABC- and GC-like DLBCL. This included 3 newly identified microRNAs, not previously associated with DLBCL and predicted to target genes that are de-regulated in lymphoma. DLBCL was distinguished from FL by 4 microRNAs and a total of 18 microRNAs were identified that differentiated between all lymphoma and control populations. Most of the discriminatory microRNAs have been reported previously to be known oncomiRs or act as tumour suppressors. In conclusion, the present study identified a microRNA signature that correctly classified GC and ABC phenotypes in DLBCL cell lines. This signature has yet to be assessed for prediction in clinical samples.

Immunophenotyping of diffuse large B-cell lymphoma (DLBCL) defines multiple sub-groups of germinal centre-like tumours displaying different survival characteristics

Diffuse large B-cell lymphoma (DLBCL) forms a heterogeneous collection of aggressive non-Hodgkin's Lymphoma in which three principle classes of neoplasia have been defined according to gene expression and immunophenotyping studies. The present investigation sought to examine the immunophenotype of proposed subgroups and relate these to patient survival. A series of 155 DLBCL treated uniformly with anthracycline therapy in clinical trials, were stratified upon the basis of common biomarker expression with combination immunophenotype being related to patient overall survival. Stratification of tumours with respect to combined expression profiles of the three biological markers (CD10, Bcl-6 and MUM-1) revealed six groups showing significant differences in survival (p=0.014). The greatest difference resided between distinct populations of germinal centre (GC) cell tumours; the first being CD10-, Bcl-6+, MUM-1- and the second CD10+ Bcl-6+ MUM-1+ (p=0.002). The former group displayed median survival time of 143 months, the latter only 11 months. A third population of GC tumours (CD10+ Bcl-6+ and MUM-1-) also displayed a relative short median survival (32 months). Of the three groups presenting a non-GC or activated B cell (NGC/ABC) phenotype, only one (CD10-, Bcl-6+ and MUM-1+) presented short-term median survival (27 months) comparable with poor prognosis GC sub-populations. Within the remaining ABC tumour groups (CD10- Bcl-6- MUM-1- and CD10- Bcl-6- MUM-1+) patients presented intermediate median survival times of 54 and 58 months, respectively. Thus, the GC phenotype did not act as a universal indicator of good clinical prognosis, but rather multiple groups of GC tumours were associated with distinct overall survival profiles. Ultimately, the data allowed definition of a predictive algorithm defining three groups predicting poor, intermediate and good clinical prognosis. The first of these comprised two patient sub-populations with GC-like tumours together with one sub-population of NGC/ABC, the second two sub-populations of ABC-like tumours, and the final a single group of GC-like tumours associated with optimal long-term survival.

A new subtype-specific monoclonal antibody for IAP-survivin identifies high-risk patients with diffuse large B-cell lymphoma and improves the prognostic value of bcl-2

Anti-apoptotic factors including IAP-survivin and bcl-2 are involved in carcinogenesis and predict for disease outcome for patients with cancer. We used RT-PCR and specific primers to generate two recombinant IAP-survivin proteins; one encoding for the full-length protein and the second comprising the survivin sequence incorporating amino acids 98 to 142. Both proteins were used to immunize mice and as capture antigens to screen NS1/immune splenocyte hybridoma supernatants for anti-survivin antibody in ELISA assays. The antibody designated F2-9C3 was most effective and reacted with both recombinant proteins and with the native protein present in lysates of A549 (lung carcinoma) and Jurkat cells in Western blots, immunoprecipitation and formalin-fixed tissue sections. Immunohistochemical staining of normal and neoplastic tissues showed association of the F2-9C3 antibody with the mitotic spindles. Expression of survivin was not detected elsewhere in sections of normal tissue while all neoplastic tissues examined, including those from patients with diffuse large B-cell lymphoma (DLBCL), showed significant expression of survivin. The intensity and localization of staining in these tumours varied and was observed in cytoplasm and/or nuclei. High nuclear expression of survivin predicted the disease outcome in patients with DLBCL. This association was evident when relating intensity to patient survival (p=0.0321) and strengthened when a score was calculated based on both staining intensity and the proportion of the reactive tumour cells (p=0.0128; reduction in the mean survival times: 35% and 46%, respectively). Elevated expression of bcl-2 protein also identified the high-risk patients (p=0.0095; reduction in mean survival time: 37%). Over-expression of both factors was a more powerful indicator of poor prognosis than either marker alone (p=0.0054, 70% reduction in mean survival time). In conclusion, our novel F2-9C3 monoclonal antibody is effective in determination of expression of IAP-survivin in neoplastic tissue. Nuclear overexpression of IAP-survivin using this antibody predicts the disease outcome in patients with DLBCL and significantly improves the predictive power of bcl-2 in these patients.