Neuroblastoma (NB) is a pediatric malignancy affects young children and responsible for approximately 10–12% of childhood cancer mortality [1]. It originals from the neural crest precursor cells of the developing sympathetic nervous system and is remarkably heterogenous in its malignant potential with diverse clinical presentations, courses, and overall prognosis. Although notably improvements have been achieved in the children with lower-risk NB in the past decades, children suffered with higher stage disease remains to be incurable with survival rates of less than 50%. Although 1p deletion, N-myc amplification, or gain of 17q may identify subtypes of NB and impact survival, yet there is no common NB-specific genomic alteration were uncovered [2].

Here we used whole exome sequencing (WES) to continuous screening the genetic somatic alteration in the NB individuals under the chemotherapy. We ought to explore the acquired somatic alteration, shedding light on genetic basis of somatic alteration in NB patient with chemotherapy.

By WES sequencing “diagnosis-2nd chemotherapy-4th chemotherapy” trios, recurrent new mutations in cell growth regulator CGREF1, transcriptional factor CUX2, diphosphohydrolase ENTPD2, solute carrier family 45 SLC45A1, and the component of RNA spliceosome TRA2A were identified as the novel somatic mutation during the chemotherapy. SLC45A1 was isolated initially from a region on chromosome 1p that is frequently deleted in human neuroblastoma. CGREF1, CUX2 and ENTPD2 are involved in mediating neural cell growth, as well.

NB originals from the neural crest precursor cells of the developing sympathetic nervous system and is remarkably heterogenous in its malignant potential with diverse clinical presentations, courses, and overall prognosis. NB with higher stage disease remains to be incurable with survival rates less than 50%. There is no common neuroblastoma-specific genomic alteration were uncovered [1, 2].

Screening gene variation in the process of therapy could provide us better and earlier information than the traditional imaging examination [6–8]. To determine tumor specific biomarkers for valuing the effect of chemotherapy in NB individuals, we focus on gene which its mutation rate changes consistent with tumor size variation.

Several novel somatic mutations were identified in all three trios comply or against the trendy of tumor size variation. Six recurrent mutations in BPTF were identified in NB patients under the continuous monitoring. BPTF gene codes nucleosome-remodeling factor subunit containing a DNA-binding domain and a zinc finger motif, indicating that BPTF might play a role in the regulation of transcription. High levels of BPTF were detected in fetal brain and in patients with neurodegenerative diseases. The gain of function mutation of BPTF could confer the abnormal expression itself and activate a set of target genes to contribute to the tumorigenesis and maintenance of NB [9–12]. Notably, BPFT gene mutation, were uncovered in the same trendy. Variation of the mutation rate in BPTF was coinciding with tumor size changes, indicating that there is strong positive relation between the mutations and prognosis. Our finding could uncover potential biomarkers of chemotherapy evaluation.

BPTF and TMCO3 mutations were detected at the time of diagnosis, after chemotherapy, BPTF mutation rate decreased, together with C20orf96 and LOC100996470, indicating tumor burden reduced, and BPTF may contribute to bone marrow metastasis.

Therapy-resistant relapse is the main cause of NB mortality. In this investigation, we identified several novel somatic mutations by WES sequencing “diagnosis-2nd chemotherapy-4th chemotherapy” trios. Recurrent new mutations in cell growth regulator CGREF1, transcriptional factor CUX2, diphosphohydrolase ENTPD2, solute carrier family 45 SLC45A1, and the component of RNA spliceosome TRA2A were identified as the novel somatic mutation during the chemotherapy. SLC45A1 was isolated initially from a region on chromosome 1p that is frequently deleted in human neuroblastoma. CGREF1, CUX2 and ENTPD2 are involved in mediating neural cell growth, as well [13–16]. The emergence of these mutations may indicate chemo-therapy resistant in NB patients.

When we look at the mutations detected in plasma cfDNA, we found that the mutations were highly consistent identified in bone marrow blood white cells and plasma cfDNA. Our results demonstrated that plasma cfDNA maybe a new option for MRD monitoring and the sensitivity of next generation sequencing is much higher than that of traditional bone marrow smear.

Consider that the limitation of sample size, we elevated the criteria of filtering, only the mutations shared by all patients remained maybe that is the reason we didn’t figure out the well-known mutations such as KRAS and NRAS. Only one recurrent mutation on MLL4 was identified in all the three samples when we do the baseline research, MLL4, lysine (K)-specific methyltransferase was the shared somatic mutation. The missense mutation was predicted to destroy protein structure by PROVEN software. MLL4 is a histone methyltransferase which methylates Lys-4 of histone H3 (H3K4me). H3K4me1/3 represents a specific tag for epigenetic transcriptional activation. Acts as a coactivator for estrogen receptor by being recruited by ESR1, thereby activating transcription. Abolish MLL4 function could impair the expression of a subset target genes. Sharing this mutation in the individuals with NB may indicate the dysregulation in the transcriptome in this malignancy tumor.

In the continuous screening genetic alterations, there are several mutation appear or disappear following the tumor size changes. Although due to the limitation of sample size, we missed some of molecular marker indicating the effects of therapy, the stricter filtering criteria allow us to identify the meaningful mutilations. This strategy could make sense to identify therapy specific genetic alteration in NB patients. Further, we ought to scale up the sample and preform functional validation of the mutations.

WES allows for novel mutation detection at low cost, could therefore be routinely applied clinically and reveal reliable bio-markers, has the potential for accelerating the personalized detection, therapy and monitoring of tumor. We anticipate that a panel of markers, such as BPTF, TMCO3, CUX2, combined with classic markers: PHOX2B, TH, DCX, will prove valuable in a variety of clinical settings, including the assessment of tumor, drug resistance, residual disease monitoring and prediction of recurrence.