Predicting clinical outcomes in neuroblastoma with genomic data integration

Abstract

BackgroundNeuroblastoma is a heterogeneous disease with diverse clinical outcomes. Current risk group models require improvement as patients within the same risk group can still show variable prognosis. Recently collected genome-wide datasets provide opportunities to infer neuroblastoma subtypes in a more unified way. Within this context, data integration is critical as different molecular characteristics can contain complementary signals. To this end, we utilized the genomic datasets available for the SEQC cohort patients to develop supervised and unsupervised models that can predict disease prognosis.ResultsOur supervised model trained on the SEQC cohort can accurately predict overall survival and event-free survival profiles of patients in two independent cohorts. We also performed extensive experiments to assess the prediction accuracy of high risk patients and patients without MYCN amplification. Our results from this part suggest that clinical endpoints can be predicted accurately across multiple cohorts. To explore the data in an unsupervised manner, we used an integrative clustering strategy named multi-view kernel k-means (MVKKM) that can effectively integrate multiple high-dimensional datasets with varying weights. We observed that integrating different gene expression datasets results in a better patient stratification compared to using these datasets individually. Also, our identified subgroups provide a better Cox regression model fit compared to the existing risk group definitions.ConclusionAltogether, our results indicate that integration of multiple genomic characterizations enables the discovery of subtypes that improve over existing definitions of risk groups. Effective prediction of survival times will have a direct impact on choosing the right therapies for patients.ReviewersThis article was reviewed by Susmita Datta, Wenzhong Xiao and Ziv Shkedy.Electronic supplementary materialThe online version of this article (10.1186/s13062-018-0223-8) contains supplementary material, which is available to authorized users.

The ZNF750–RAC1 axis as potential prognostic factor for breast cancer

“…Each of these subtypes is characterized by distinct disease progression, therapeutic response and clinical outcome. Therefore, a stratification of the patients is necessary to achieve better clinical outcome and for predicting the clinical course of the disease 9,10 . This is a general problem in clinical cancer research [11][12][13] , and current research is very active in this respect 14 , including for example on neuroblastoma 9,[15][16][17] .…”

“…Therefore, a stratification of the patients is necessary to achieve better clinical outcome and for predicting the clinical course of the disease 9,10 . This is a general problem in clinical cancer research [11][12][13] , and current research is very active in this respect 14 , including for example on neuroblastoma 9,[15][16][17] . To achieve this, novel molecular pathways that are implicated in the pathogenesis of cancer should be uncovered and identification of novel biomarkers is needed for predicting patient clinical outcome 15,16,18 .…”

Identification of novel prognostic markers of survival time in high-risk neuroblastoma using gene expression profiles

“…Outcomes ranging from spontaneous regression to relentless progression despite extensive therapies indicate the heterogeneity of neuroblastoma [6]. The Children's Oncology Group (COG) classifies neuroblastoma patients into low, intermediate and high-risk groups.…”

Computational Identification of Gene Networks as a Biomarker of Neuroblastoma Risk

“…In practice, several clinical parameters, such as age at diagnosis, tumor stage, genomic amplification of MYCN oncogene, and ploidy, are widely used as the markers of neuroblastoma risk [ 3 , 4 , 5 , 6 ]. Increasing studies have attempted to stratify neuroblastoma risk based on the pattern of differential gene expression [ 1 , 4 , 5 , 7 , 8 ]. Sets of genes of prognostic importance have been identified to link with neuroblastoma; for example, Oberthuer et al [ 4 ] established a 144-gene predictor for classifying the stratification of neuroblastoma patients, Formicola et al [ 5 ] used 18 genes to predict the outcome of stage 4 patients, and Utnes et al [ 6 ] identified 20 mRNAs and six lncRNAs of clinical relevance to the prediction of tumor recurrence and response to neuroblastoma therapy.…”