The Machine-Learning-Mediated Interface of Microbiome and Genetic Risk Stratification in Neuroblastoma Reveals Molecular Pathways Related to Patient Survival

Abstract

Currently, most neuroblastoma patients are treated according to the Children’s Oncology Group (COG) risk group assignment; however, neuroblastoma’s heterogeneity renders only a few predictors for treatment response, resulting in excessive treatment. Here, we sought to couple COG risk classification with tumor intracellular microbiome, which is part of the molecular signature of a tumor. We determine that an intra-tumor microbial gene abundance score, namely M-score, separates the high COG-risk patients into two subpopulations (Mhigh and Mlow) with higher accuracy in risk stratification than the current COG risk assessment, thus sparing a subset of high COG-risk patients from being subjected to traditional high-risk therapies. Mechanistically, the classification power of M-scores implies the effect of CREB over-activation, which may influence the critical genes involved in cellular proliferation, anti-apoptosis, and angiogenesis, affecting tumor cell proliferation survival and metastasis. Thus, intracellular microbiota abundance in neuroblastoma regulates intracellular signals to affect patients’ survival.

Intratumor microbiome: selective colonization in the tumor microenvironment and a vital regulator of tumor biology

“…As a newly found player in the tumor microenvironment, numerous studies have proven that the composition of tumor microbiota is significantly different from that of adjacent normal tissues 12 and is widely associated with clinical features of patients (Table 1). 13–33 These results highlight the potential role of intratumor microbiome in regulating cancer development. Thus, the roles and mechanisms of mutual regulation among intratumor microbiome, tumor cells, and other components of the tumor environment urgently need to be uncovered.…”

Predicting Neuroblastoma Patient Risk Groups, Outcomes, and Treatment Response Using Machine Learning Methods: A Review

“…Interestingly, NB risk groups may also be linked to the tumour intracellular microbiome, and a study aimed to link these two aspects [ 43 ]. This study used 120 NB patient RNA-sequencing datasets from the National Cancer Institute for human and microbial genetic sequences.…”

“…This study used 120 NB patient RNA-sequencing datasets from the National Cancer Institute for human and microbial genetic sequences. In this cohort, the mean age at diagnosis was four years and three months, and the majority were male and were classified as high-risk based on COG criteria (80.8%) [ 43 ]. Microbiota was found in the NB patient RNA-sequencing data by Skmer, which extracted k-mer (K = 32) patterns of microbiome sequences, and this was referred to as the MKP profile; as a result, the group was split into MKP1 and MKP2.…”

“…The Pearson Chi-square test also showed that risk groups were linked to MKP groups ( p = 0.0195). Accordingly, the high-risk patients were represented in both MKP clusters, but the low- and intermediate-risk cases were represented in the MKP2 cluster only [ 43 ]. This suggested that the risk group was linked to the MPK profile.…”

“…As expected, the high-risk cases allocated to the MKP1 group showed lower survival than their counterparts in the MKP2 group ( p value of 6.422 × 10 −6 , HR = 3.78). High-risk NB patients allocated to MKP2 also had lower survival than the low-intermediate risk NB patients in MKP2 ( p value of 0.0004 and HR = 5.56) [ 43 ]. In addition, it was shown that microbial gene abundance was highly linked to prediction accuracy, and as such, the microbiome prediction score (M-score) was introduced.…”