We also found that q does not

We also found that 1q+ does not show preference in evolutionary timing—it may occur as an early or late event—which suggests that its oncogenic effect is independent of other genetic changes. Therefore, we suspect that 1q+ may have a similar effect on WT outcome regardless of whether it is homogeneous or heterogeneous in the primary tumor, and that current studies based on single tumor sampling may have underestimated its prognostic significance. Our findings clearly imply that future clinical trials in WT must take this heterogeneity into account and multi-sample each tumor or attempt to detect this change in circulating tumor DNA at a level that can interrogate subclonality. In contrast, we find that brompheniramine maleate 11p15 CNNLOH, another common change in WT, is consistently an early event in WT tumorigenesis. Our finding therefore builds on previous observations of somatic 11p CNNLOH in WT precursor lesions (nephrogenic rests) (Charles et al., 1998). 11p15 CNNLOH is associated with several other pediatric small round cell tumors, and recently it was found as a recurrent lesion in the vast majority of pediatric adrenocortical tumors, also occurring as an early event preceding most point mutations (Pinto et al., 2015). These findings suggest that 11p15 CNNLOH may represent a common mechanism of tumorigenesis in a significant proportion of pediatric solid tumors, and its occurrence as an early event makes it a promising candidate for early detection of pediatric cancer by non-invasive screening for ctDNA in blood. In those tumors without 11p15 CNNLOH we identified a subset of five cases with isolated hypermethylation of the H19 DMR, and this abnormality was also present in adjacent histologically normal kidney. In one of these five cases, there was hemihypertrophy, whereas in the other four cases there were no features to suggest Beckwith-Wiedemann syndrome. This finding is in keeping with previous reports of mosaic hypermethylation of the H19 DMR in a significant proportion of normal cells in cases of WT with this abnormality, even in the absence of other features of Beckwith-Wiedemann syndrome (Moulton et al., 1994; Okamoto et al., 1997; Scott et al., 2008b). Indeed, the proportion and distribution of non-tumor cells with this WT-predisposing epimutation may at least in part underlie the expression and variability of the features of Beckwith-Wiedemann syndrome. Our findings on rarer biomarkers are more difficult to interpret in the absence of a larger multi-sampled tumor cohort. Nevertheless, our findings on 16q− highlight the importance of interpreting ITGH in the context of tumor evolution: in our series, 16q− is apparently heterogeneous only in Case 13, but it is erroneous to interpret this as evidence of 16q− ITGH, since it is present in the one sample from the smaller nodule that we showed arose independently of the remaining tumor mass. In the case of another biomarker, MYCN gain, we were able to relate subclonal acquisition of this change to a significantly better response to chemotherapy (as assessed on diffusion-weighted MR imaging before and after preoperative chemotherapy). MYCN gain may be expected to be associated with more rapid cell proliferation and therefore greater sensitivity to cytotoxic chemotherapy, and this may explain our finding. More generally, we have shown that it is feasible to integrate phylogenetic tumor analysis with monitoring of treatment response by imaging, provided that the imaging analysis is used as a guide to tumor sampling, in addition to current standard histological sampling.
Author Contributions
Declaration of Interests
Acknowledgements
This work was supported by the National Institute of Health Research (NIHR) Biomedical Research Centre at Great Ormond Street Hospital, Cancer Research UK (grant no. C1188/A4614) and the Great Ormond Street Hospital Children\'s Charity (W1090).