Altered Expression of Ribosome Biogenesis Regulators (TP53, C-MYC, FBL, and NCL) in Precursor B-cell Acute Lymphoblastic Leukemia and Neuroblastoma

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Synopsis: 

In their manuscript entitled “Altered expression of ribosome biogenesis regulators (TP53, C-MYC, FBL, and NCL) in precursor B-cell acute lymphoblastic leukemia and neuroblastoma,” Horochowska and colleagues conduct a single-institution retrospective study seeking to evaluate patterns of transcriptional expression of TP53, C-MYC, FBL, and NCL in pediatric pre-B ALL and neuroblastoma specimens relative to bone marrow from healthy donors to explore their roles in regulating ribosome biogenesis. They do so in an effort to address the relative paucity of data describing the regulation of ribosomal biogenesis in pediatric relative to adult malignancies. 

They pursued this goal through measurement of TP53, C-MYC, FBL, and NCL mRNA in un-annotated healthy donor bone marrow specimens and clinically annotated materials from patients with either of the two most common pediatric malignancies -- pre-B ALL and neuroblastoma. These measurements were compared in search of significant differences between pre-B ALL, neuroblastoma, and healthy donors. Additionally, correlations between expression of these genes within the pre-B ALL, neuroblastoma, and healthy donor specimens were also evaluated.

The authors’ primary findings were an increase C-MYC, NCL, and FBL expression in healthy donor bone marrow versus pre-B ALL bone marrow, and an increase in FBL expression in neuroblastoma tissue specimens versus pre-B ALL bone marrow. Within pre-B ALLs, the expression of TP53 correlated with that of C-MYC, FBL, and NCL. Within neuroblastomas, the expression of C-MYC correlated with that of FBL. In normal bone marrow, expression of NCL correlated with that of TP53 and FBL. Finally, no significant correlations were identified between mRNA levels of TP53, C-MYC, FBL, and NCL and annotated clinical or laboratory features of the pre-B ALL or the neuroblastoma specimens.

After speculating about the mechanisms underlying these associations, the authors conclude that altered expression of these genes disrupts ribosomal biogenesis, nucleolar function, and translational regulation, which may contribute to oncogenesis.

Overall impression: 

A positive aspect of this manuscript is the well-written and scholarly review of known biology in the introduction and discussion sections, with references that are germane and qualitatively sufficient. 

Inasmuch as this study set out to define aspects of basic biology in prevalent pediatric cancer subtypes, it addresses an import area of investigation. However, after reviewing the literature and delineating known mechanisms of ribosome biogenesis regulation, the authors state that most prior work was conducted in adult malignancies and suggest that there is thus a gap in our understanding of this process in pediatric malignancies. It would have helped solidify the premise of this work to provide any other data corroborating the likelihood that the regulation of ribosome biogenesis meaningfully differs between adults and children. In lieu of such data, the premise of this work remains speculative.

In terms of study design, the relatively small size of the cohrort evaluated, the narrow panel of genes assessed, and the single modality used to measure their expression (RT-qPCR) renders this work exploratory and descriptive. 

Constructive comments: 

Two specific aspects of the experimental design create the possibility of unaccounted confounding and are essential to consider when attempting to interpret these data. First, there is heterogeneity within the groups analyzed that is insufficiently documented and controlled for. For the malignant specimens like the pre-B ALL bone marrow samples, there are likely multiple cytomolecular subtypes of ALL represented, which biologically differ from each other in significant ways. These subtypes were neither reported in Table 1 nor accounted for in the gene expression analyses. In parallel, the bone marrow specimens were obtained from healthy donors who lack any clinical annotation. No demographic or clinical characteristics of these donors are provided to establish how well these specimens function as controls. In this reviewer's personal experience, most healthy bone marrow donors are young to middle age adults. For a study that states that its goal is to define pediatric tumor biology, the use of bone marrow specimens from donors with unclear demographic characteristics but who are likely to be adults introduces the possibility of critical age-related confounding. 

Second, Tables 1 and 2 indicate that the pre-B ALL bone marrow specimens had a range of involvement (5-99%). Tumor purity is not stated for the neuroblastoma specimens. The methods suggest that no tumor cell enrichment or isolation was performed, and total RNA was extracted from the tissues for bulk RT-qPCR of the selected genes. This suggests that a variable and uncontrolled proportion of the tissue sequenced from the pre-B ALL and the neuroblastoma specimens reflected non-tumor cells, thereby introducing an additional confounding variable into the comparison of gene expression in pre-B ALL or neuroblastoma and normal bone marrow. This factor is magnified by the fact that the non-malignant tissue present in the neuroblastoma cases was qualitatively distinct from the bone marrow control tissue. It is essential to have a better sense of these potential confounders in order to interpret the data in this study. 

Another important point acknowledged by the authors in the discussion is that the data provided in this manuscript are limited to a single modality (RT-qPCR). Given that RNA abundance can differ from protein expression and functional consequences based on numerous factors including translational efficiency, protein trafficking, and protein stability, the lack of protein-level corroboration -- to say nothing of functional validation -- tempers any biological interpretation of the RNA levels. This issue is particularly salient when data are counterintuitive, as they were in the case of C-MYC expression being higher in healthy donors compared to pre-B ALL and the positive correlation between TP53 and C-MYC expression in pre-B ALL. The authors speculate that this could reflet inherent biological differences between pediatric and adult leukemia, but it is equally possible that these findings reflect some degree of technical and biological confounding. 

The issues delineated above are what I would consider fundamental. 

Other minor comments: 

It would be helpful if the authors provided more justification of why they evaluated TP53, C-MYC, FBL, and NCL but not other genes implicated in regulating ribosome biogenesis. 

Table 1 and 2 lack some elements that would typically be included. For table 1, cytomolecular ALL subtype. For Table 2, patient age at diagnosis. For both tables, tumor involvement would help. 

Table 3 and Figure 2 would be easier to interpret if they were combined. It is much more conventional, and I would strongly suggest adding the statistical significance indicators to Figure 2 as brackets and asterisks above the box and whisker plots. 

Tables 4, 5, and 6 are displayed in a manner that is needlessly inefficient wherein every value is duplicated and there are blank cells for self-comparisons. I would strongly advocate for just listing the pairwise comparisons in separate rows with columns for the correlation coefficient and the p-value. 

The text does not match Table 5. On further examination, it appears that Table 5 is a duplicated version of Table 4. This must be corrected. 

The conclusions are vague. It is theoretically possible that dysregulated expression of TP53, C-MYC, FBL, and NCL contribute to oncogenesis, but that is several steps beyond the data provided here. I submit that this work would be stronger if more of the emphaisis was on providing additional complementary data rather than postulating about potential biological implications far downstream of the data shown. 

Author Response

Dear Editor,
We thank the Reviewers for their careful reading of our manuscript entitled “Altered expression of ribosome biogenesis regulators (TP53, C-MYC, FBL, and NCL) in precursor B-cell acute lymphoblastic leukemia and neuroblastoma.” We appreciate their constructive comments. We have revised the manuscript accordingly, and all changes are highlighted in the revised version. Below we provide a point-by-point response.

Major points:

Comment: The premise that ribosome biogenesis regulation differs in pediatric vs adult malignancies is speculative; more support would strengthen rationale.

Response: We agree and have strengthened the rationale. In the revised Introduction, we now discuss developmental context and pediatric-specific oncogenic programs that may alter nucleolar/ribosome biogenesis control compared with adult cancers. We also added supporting pediatric-focused citations in this section.

Group heterogeneity / lack of molecular subtypes and control annotation:

Comment: ALL cytomolecular subtypes not reported/controlled, healthy donors not annotated, likely adult donors could confound pediatric biology.

Response: 

We thank the Reviewer for highlighting this issue. We have added available cytogenetic/cytomolecular subtype information to Table 1 and stated explicitly that subgroup analyses were not feasible due to sample size. We also added a limitation noting that subtype-specific expression differences may exist.

Comment: The bone marrow specimens were obtained from healthy donors who lack any clinical annotation. No demographic or clinical characteristics of these donors are provided to establish how well these specimens function as controls.

Response: Healthy characteristics: We retrieved available demographic characteristics of healthy donors and added them to the control description (sex and age range/median where accessible). We further clarified in the Methods and Discussion that donor age mismatch could bias expression comparisons and thus results should be interpreted cautiously.

Tumor purity / bulk tissue confounding

Comment: Variable marrow blasts (5–99%) and unknown neuroblastoma purity, bulk RT-qPCR may reflect non-tumor cells, non-malignant tissue differs between NBL and BM controls.

Response:

Response: We agree this is a relevant limitation.We now address it in two ways:

1. Reported blast burden: We corrected marrow blast percentage in Table 1 and referenced it in Results. 50, not 5%.

2. We added a clearer limitation emphasizing that bulk RNA includes stromal/immune components and that tissue-type differences (BM vs solid tumor) may influence baseline gene expression. “This concern is amplified by the use of bone marrow controls for comparison with solid-tumor neuroblastoma, where the non-malignant microenvironment differs qualitatively from marrow.”

Lack of protein-level/functional validation

Comment: Results are RNA-only, protein/functional corroboration lacking, especially when findings are counterintuitive.

Response: We agree fully. We have strengthened the Limitations and Conclusions to avoid over-interpretation and to propose protein-level validation as a necessary next step. We also moderated language in the Discussion and Conclusions to make clear that our findings are exploratory and hypothesis-generating.
Changes in manuscript:

1. Discussion: softened mechanistic claims; added explicit caution that mRNA doesn’t equal protein activity.

2. Conclusions: revised to emphasize exploratory nature and need for validation.

Minor points:

Justification for selected genes

Comment: Provide more justification for TP53, C-MYC, FBL, NCL selection.

Response: We added a short justification in the Introduction noting their central and complementary roles in Pol I/III regulation, rRNA processing, and nucleolar stress, our prior pilot data in pediatric ALL, and (iii) feasibility constraints in retrospective material.

Missing table elements

Comment: Table 1 lacks ALL subtype; Table 2 lacks age; tumor involvement could be added.
Response: Addressed as follows - Table 1: added cytomolecular subtype (as above).

Comment: Merge Table 3 and Figure 2
Response: We agree. We revised Figure 2 to include statistical brackets and p-value
indicators and re-organized Table 3 accordingly, so that the figure can be interpreted
stand-alone.

Comment: Reformat correlation Tables 4–6
Response: Tables 4–6 inefficient and duplicative; list pairwise correlations only.
Not changed.

Comment: Table 5 duplicated from Table 4
Response: This is not an oversight. Table 4 is pre-B-ALL- and 5 for neuroblastoma-specific correlations.

Comment: Conclusions too speculative
Response: We agree and revised Conclusions to avoid causal wording. We now
state that altered expression “suggests association with disrupted ribosome
biogenesis” rather than implying oncogenic causality, and highlight exploratory
nature.

We believe these revisions address all concerns, correct the table error, improve data presentation, and appropriately temper interpretation. We thank the Reviewers again for significantly improving the manuscript.

Sincerely,

Michalina Horochowska MD, prof. Marek Ussowicz

Reviewer 2 Report

Comments and Suggestions for Authors

 

1. Please describe interaction of the ribosomal protein-MDM-P53 in the introduction.
2. Do you have any data for influence of the c-Myc in Ribosome biogenesis?
3. Describe a little bit how c-Myc interact with ribosomal protein.
4. you wrote: For NCL, expression was highest in healthy donors, intermediate in pre-B ALL, and 182 lowest in neuroblastoma. The differences in NCL expression were statistically significant 183 between neuroblastoma and pre-B ALL. How fold is higher? If you have any data please give it!
5. Which is interaction of the c-Myc and P53 with ribosomal protein, rRNA, snoRNA, and so on?

Author Response

Dear Editor,
We thank the Reviewers for their careful reading of our manuscript entitled “Altered expression of ribosome biogenesis regulators (TP53, C-MYC, FBL, and NCL) in precursor B-cell acute lymphoblastic leukemia and neuroblastoma.” We appreciate their constructive comments. We have revised the manuscript accordingly, and all changes are highlighted in the revised version. Below we provide a point-by-point response.

1. Comment: Add RP–MDM2–p53 nucleolar stress interaction in Introduction
   Response: We added a dedicated paragraph in the Introduction describing the nucleolar stress pathway, including how free ribosomal proteins (RPL5/RPL11/RPL23) bind and inhibit MDM2, stabilizing p53 and linking ribosome biogenesis to cell-cycle arrest/apoptosis.
   
   
2. Comment: Evidence for c-Myc influence on ribosome biogenesis
   Response: Our study did not include functional assays of C-MYC activity. We therefore clarified this and expanded the background summarizing established literature that MYC activates Pol I/II/III transcription, increases rRNA precursor synthesis, and upregulates ribosomal proteins and processing factors.
   
   
3. Comment: Describe how c-Myc interacts with ribosomal proteins
   Response: We added text explaining the reciprocal regulation: MYC promotes RP transcription, while specific RPs (e.g., L11) can inhibit C-MYC as part of a negative feedback loop during ribosomal stress.
4. Comment: Provide fold-change for NCL differences
   Response: We calculated fold differences from group medians of 2^-ΔCt values and added exact fold-change statements to Results, and we also included these values in the updated Figure 2 annotations.
   
   
5. Comment: Broaden description of c-Myc and p53 interactions with rRNA/snoRNA/RPs
   Response: We expanded the Introduction to summarize:

• p53 repression of Pol I/III transcription and downstream rRNA/snoRNA production,
• MYC activation of Pol I/II/III and induction of rRNA, 5S rRNA, and RP/biogenesis factor genes,
• the nucleolar stress checkpoint integrating RP abundance, MDM2 inhibition, and p53 stabilization.
  
  

We believe these revisions address all concerns, correct the table error, improve data presentation, and appropriately temper interpretation. We thank the Reviewers again for significantly improving the manuscript.

Sincerely,

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dr. Horochowska and colleagues here provide a revised manuscript entitled "Altered expression of ribosome biogenesis regulators (TP53, C-2 MYC, FBL, and NCL) in precursor B-cell acute lymphoblastic 3 leukemia and neuroblastoma," in which they have incorporated responses to a number of specific constructive critiques in the original review. 

The revised manuscript has addressed several previous gaps in rationale in the introduction, missing data in the results, and opportunities for added nuance in the discussion to temper conclusions in light of the limitations in the data. These improvements are duly noted, but critical issues and a non-trivial number of formatting and aesthetic issues remain, which should be addressed to improve clarity and readability. 

Major: 
1. The primary results of this study relate to differences in transcriptional expression of TP53, C-MYC, FBL, and NCL in pediatric B-ALLs, neuroblastomas, and healthy donors. The original review suggested potential confounding biological and clinical factors that might account for some of the counterintuitive results like C-MYC expression being lower in B-ALL than in healthy donor bone marrow. The revised manuscript now describes many of these limitations. It remains difficult to interpret these data, however, based on their measurement in relation to a single housekeeping gene, GAPDH, which is known to be variable in a number of cellular contexts. While the authors state that the biomaterial available from this retrospective cohort is limited and precludes protein analyses to corroborate these unexpected findings, might they have any RNA remaining with which they could repeat the qPCR studies using other housekeeping gene(s) commonly used in studies of B-ALL like EEF2, PGK1, TBP, or HPRT. This is a critical issue since GAPDH expression can vary in different cellular contexts. If it was higher in the leukemia cases than the normal controls, this would have blunted the relative expression values for the experimental genes and could be a reason for these counterintuitive findings where all of the genes tested had higher expression relative to GAPDH in the normal bone marrow specimens than the B-ALLs or the neuroblastomas. 
2. On page 8, lines 249-251, it would be important to note explicitly whether gene expression levels were evaluated for a correlation with B-ALL cytomolecular features. These were a new element of the Table 1 in the revised manuscript, but this sentence reflects no modification to make clear whether or not biological subtype of B-ALL influenced gene expression levels.  
3. Also on the subject of cytomolecular characteristics of the B-ALLs, the authors state on page 10, line 345 that "mutation status of TP53 and potential variants in the other analyzed genes were not assessed in this cohort." It would be necessary to explain why this was not done given that such alterations could confound the transcriptional results. Even if mutational profiling data were unavailable, could the authors provide or comment on the availability of fluorescence in situ hybridization data or relevant karyotype data (e.g., for evidence of 17p loss)?  

Conceptually minor but striking: 
1. On page 7, lines 209-211, there is a sentence that has placeholder variables for data. "Based on group median 2^-ΔCt values, NCL expression in pre-B ALL was approximately x-fold lower than in controls, while neuroblastoma showed an approximately y-fold decrease versus pre-B ALL and z-fold decrease versus controls." These values must be included. It is disconcerting that this new sentence, which is highlighted, has placeholder values.  
2. The original review noted that Tables 5 and 6 are identical. The response to the review was that this was not an oversight, as one table encompasses gene expression correlations for pre-B-ALL patients and the other is for neuroblastoma patients. However, the numbers in these two tables are identical, and the values in Table 6 deviate from numbers provided in the text. For example, on page 8, line 232-234, the sentence reads, "a statistically significant positive correlation was iden-232 tified between C-MYC and FBL expression (p = 0.004), while no other correlations were 233 found to be significant (Table 6)." However, in Table 6, the p-value for the correlation between C-MYC and FBL is 0.258. If I am misinterpreting the fact that the values in these tables is identical, I suspect that many readers will have the same experience. 

Minor/formatting/aesthetic: 
1. It would help if the manuscript reflected the convention that gene names including when referring to RNA are italicized. 
2. It would help if the authors used p53 when referring to the protein encoded by the TP53 gene. 
3. Gene names and other abbreviations should be spelled out the first time they are mentioned. Currently, as two examples, fibrillarin and nucleolin are not spelled out until page 3, after they have been referenced several times. 
4. The tables in general remain formatted in a manner that is difficult to read. Suggestions for improving the tables are: (a) Left justify the first column (e.g., in Table 1); (b) only show one value for binary variables (e.g., just show the yes values and percentages); (c) be consistent in whether to separate factor names and numeric values with a dash (e.g., in Table 1 cytomolecular subtype, some of the values have a dash whereas most other entries do not); (d) somehow separate clusetered rows by incorporating vertical padding, row borders, or background shading (e.g., in Table 4 visually distinguish the four row clusters correlating with each gene more clearly and also vertically center the gene names in those clustered rows, as the gene names appear low compared to the patient groups). 
5. In Table 1, for cytomolecular subtype-CDKN2A heterozygous deletion, the decimal point is missplaced. This value should be 7% not 0.07% (since it reflects 3/45). 
6. In Table 2, the values for MYCN amplification do not add up to 19 (positive n=5, negative n=9). The percentages reflect a denominator of 14, which is misleading. Were the remaining 5 patients in this cohort not tested for MYCN? If so, they should be shown as such (e.g., not tested - 5 patients (26%)). 
6. The legends for Tables 5, 6, and 7 state that "statistically significant results are shown in bold and marked with an asterisk." However, only asterisks are shown. The significant values in the table are not bold. 

Overall, the gene expression results (particularly for C-MYC) deviate from conventional wisdom, so the fact that the qPCR data are the only data provided in this paper, apparently due to unavoidable limitations of this small retrospective series of sparse specimens, underscores the need to demonstrate technical rigor. This is why it is so important to evaluate whether these gene expression patterns are maintained when a different, commonly used housekeeping gene is measured. Otherwise, the divergence between the rigor of the data and the degree of very thoughtful mechanistic speculation in the discussion is very large. 

Author Response

We thank the Reviewer for the careful, detailed, and constructive re-evaluation of our revised manuscript entitled “Altered expression of ribosome biogenesis regulators (TP53, C-MYC, FBL, and NCL) in precursor B-cell acute lymphoblastic leukemia and neuroblastoma.” We appreciate the Reviewer’s recognition of the improvements made in the Introduction, Results, and Discussion. In the current revision, we have addressed the remaining critical scientific concerns and the formatting/aesthetic issues to improve interpretability, technical rigor, and readability.

We reproduce comments followed by our response and a concise summary of changes implemented in the manuscript.

Major comments

1) Reference gene (GAPDH) and risk of confounding normalization

Reviewer comment: Interpretation remains difficult because expression was measured relative to a single housekeeping gene (GAPDH), which can be variable. Could the authors repeat qPCR using other housekeeping gene(s) (EEF2, PGK1, TBP, HPRT), or otherwise address the stability of GAPDH?

Response:

We reviewed Ct(GAPDH) distributions across groups and did not observe a systematic shift; this information is now reported as a technical QC item. No remaining RNA/cDNA was available from this retrospective cohort to repeat qPCR with alternative reference genes, we therefore explicitly frame the results as exploratory and emphasize reference-gene variability as a central limitation.

2) Explicit statement about correlations with B-ALL cytomolecular features

Reviewer comment: It should be stated explicitly whether gene expression was evaluated for correlation with B-ALL cytomolecular features (newly shown in Table 1).

Response:

We explicitly state that these analyses were not performed due to small subgroup sizes and incomplete data, and we added this as a limitation/future direction. The information on this subject has been included in the article.

3) Lack of mutation profiling / structural alterations (e.g., 17p loss) and rationale

Reviewer comment: Explain why TP53 mutation status and variants in other genes were not assessed. Could the authors comment on availability of FISH/karyotype data (e.g., 17p loss)?

Response: We agree this point requires clearer justification. This was a retrospective cohort and the study used archival RNA derived from diagnostic material, mutational profiling of TP53 and the other genes was not part of routine diagnostic work-up for all cases during the study period at our center and therefore was not uniformly available in the medical record. In addition, remaining material was insufficient to perform consistent post hoc DNA-based profiling across all cases.

Regarding cytogenetics/FISH: we clarified exactly what was available in the record (and what was not). Specifically, Table 1 reports the available cytomolecular data that were routinely collected (ploidy, ETV6/RUNX1, CDKN2A deletion, and other alterations). We additionally reviewed clinical diagnostic documentation for evidence of 17p loss / TP53 locus assessment, “17p/TP53 locus evaluation (e.g., 17p deletion FISH) was not routinely performed and was not available for the majority of patients; thus we could not incorporate it.”

We also added text clarifying that unmeasured TP53 genetic alterations could confound transcript levels and interpretation, reinforcing that our findings are hypothesis-generating.

Conceptually minor but striking

1) Placeholder x/y/z fold-changes for NCL must be filled

Reviewer comment: Sentence contains placeholder variables x/y/z that must be included.

Response: We apologize for this oversight. We have replaced the placeholders with the actual fold-changes calculated from the group median 2^-ΔCt values reported in Table 4. The placeholder sentence is now populated with the correct fold-change values.

2) Tables 5 and 6 identical + mismatch with text (p-values)

Thank you—this was a critical error. Tables are replaced with the correct correlations, duplication removed.

1) Gene symbols italicized

Response: Implemented throughout: gene symbols (TP53, MYC/C-MYC, FBL, NCL, MYCN, ETV6, RUNX1, CDKN2A, GAPDH, etc.) are now italicized when referring to genes/transcripts, consistent with convention.

2) Use p53 for protein encoded by TP53

Response: Implemented throughout: we now use p53 for the protein and TP53 for the gene.

3) Spell out gene names/abbreviations at first mention

Response: Implemented: fibrillarin (FBL) and nucleolin (NCL) are now spelled out at first mention (and similarly for other abbreviations, including GAPDH, MYCN, etc.).

4) Table readability improvements (alignment, binary variables, consistency, visual grouping)

Response: Implemented: tables were reformatted to improve readability.

5) Table 1 CDKN2A heterozygous deletion percentage incorrect

Response: Corrected: 3/45 = 6.7% (~7%), not 0.07%. Table 1 now reports 7%.

6) Table 2 MYCN amplification denominator discrepancy

Response: Corrected and clarified: the table now explicitly reports positive – 5 pts (26,5%), negative – 9 pts (47%)  and no data – 5 pts (25,5%) and the correct denominator for percentages (e.g., “not tested – 5 patients (26%)” if applicable), so totals sum to n=19 and percentages are not misleading.

7) Bold + asterisk inconsistency in Tables 5–7 legends

Response: Corrected: significant results are now bolded and marked with an asterisk, matching the legend wording .

We appreciate the Reviewer’s point that the apparent divergence of C-MYC expression from conventional expectations makes technical rigor and normalization assumptions especially important. In the revision, we therefore strengthened the explicit limitations around RNA-only analysis and reference-gene dependence, corrected and quality-checked all correlation tables and narrative p-values, and  tightened the Discussion language to ensure mechanistic interpretations are framed as hypothesis-generating and proportionate to the dataset.

Kind regards

Reviewer 2 Report

Comments and Suggestions for Authors

N/A

Author Response

We thank the Reviewer for the careful, detailed, and constructive re-evaluation of our revised manuscript entitled “Altered expression of ribosome biogenesis regulators (TP53, C-MYC, FBL, and NCL) in precursor B-cell acute lymphoblastic leukemia and neuroblastoma.” 

Kind regards,

Authors