“Losing the Brakes”—Suppressed Inhibitors Triggering Uncontrolled Wnt/ß-Catenin Signaling May Provide a Potential Therapeutic Target in Elderly Acute Myeloid Leukemia

Round 1

Reviewer 1 Report

In their manuscript, Elyamany and colleagues described a differential expression of genes implicated in Wnt/b-catenin pathway regulation between samples from pediatric AML patients vs the elderly ones, with a downregulation of Wnt/b-catenin inhibitors in the latter.

Although the work would be potentially interesting, there are several points that need to be improved, that I detailed below.

Line 60: is the first time that the acronym "GEPs" appears in the manuscript, but its definition is reported only in the results paragraph (line 125). Move it at line 60.

Line 74: the Authors state "Our results authenticate Wnt/β-catenin overexpression in older AML patients as previously reported", but they did not report any reference where these results are previously reported. Add appropriate references.

In the "introduction" a more detailed description of the different Wnt/b-catenin pathway inhibitors and their functions seems to be missed.

Line 81: why the Authors choose to use formalin fixed paraffin embedded (FFPE) diagnostic bone marrow (BM) biopsy samples instead freshly isolated leukemic cells? Justify this choice.

Line 92: RNA analyses were done on samples taken from areas with the highest tumor concentration. Couldn't this selection lead to bias? Could the observed changes not be representative of the whole tumor?

Lines 116-125: the results regarding the mutational status of FLT3 and NPM1 in the analysed samples are deeply described, but the importance/role of these mutations in the disease (and specifically in this context) is missed. Add a paragraph about them.

Lines 155-156: delete the sentence "Utilizing them...for old age AML" which seems unnecessary and decontextualized.

Line 159: Is there any evidence? Without supporting data (even preliminary data on cell lines) it seems like excessive speculation.

Line 165: if the downregulation of Wnt/b-catenin inhibitors is due to methylation of their promoters, the effect of hypomethylating drugs (e.g. azacitidine), already used in the clinic for the standard treatment of old/unfit AML patients, should be sufficient for their reactivation, so it does not seem that the specific use of inhibitors of this signaling pathway can lead to a greater therapeutic response.

Line 226: the Authors state that "Restoring expression of suppressed Wnt/β-catenin inhibitors offer a novel strategy of managing AML". How? Explain.

Author Response

RESPONSE by the authors to the comments by reviewers

REVIEWER # 1

In their manuscript, Elyamany and colleagues described a differential expression of genes implicated in Wnt/ b-catenin pathway regulation between samples from pediatric AML patients vs the elderly ones, with a down regulation of Wnt/b-catenin inhibitors in the latter.

Although the work would be potentially interesting, there are several points that need to be improved, that I detailed below.

Reviewer’s comment #1

Line 60: is the first time that the acronym "GEPs" appears in the manuscript, but its definition is reported only in the results paragraph (line 125). Move it at line 60.

Author’s response: Thank you for identifying the deficiency in this manuscript. This deficiency is corrected as advised in the revised manuscript.

Reviewer’s comment #2

Line 74: the Authors state "Our results authenticate Wnt/β-catenin over-expression in older AML patients as previously reported", but they did not report any reference where these results are previously reported. Add appropriate references.

Author’s response: Thank you for identifying the weakness in this manuscript. This deficiency is corrected as advised in the revised manuscript.

Reviewer’s comment #3

In the "introduction" a more detailed description of the different Wnt/ b-catenin pathway inhibitors and their functions seems to be missed.

Author’s response: Thank you for identifying the weakness in this manuscript. This deficiency is corrected as advised in the revised manuscript.

Reviewer’s comment #4

Line 81: why the Authors choose to use formalin fixed paraffin embedded (FFPE) diagnostic bone marrow (BM) biopsy samples instead freshly isolated leukemic cells? Justify this choice.

Author’s response: Thank you for identifying the limitation of this study. This is a retrospective study where only formalin fixed paraffin embedded samples were available. This is addressed in the modified materials and methods section of the revised manuscript.

Reviewer’s comment #5

Line 92: RNA analyses were done on samples taken from areas with the highest tumor concentration. Couldn't this selection lead to bias? Could the observed changes not be representative of the whole tumor?

Author’s response: Thank you for identifying the limitation of this study. We used multiple cores of representative tissue cores containing more than 90% tumour. These cores contain more than 75% of the tumour and avoided the contamination of normal cellular elements. This is addressed in improved materials and methods section of the revised manuscript.

Reviewer’s comment #6

Lines 116-125: the results regarding the mutational status of FLT3 and NPM1 in the analyzed samples are deeply described, but the importance/role of these mutations in the disease (and specifically in this context) is missed. Add a paragraph about them.

Author’s response: Thank you for identifying the limitation of this study. The result section is improved with results and a paragraph is included in discussion section with adequate references.

Reviewer’s comment #7

Lines 155-156: delete the sentence "Utilizing them...for old age AML" which seems unnecessary and decontextualized.

Author’s response: Thank you for identifying the limitation of this study. This sentence is removed from the revised and improved manuscript.

Reviewer’s comment #8

Line 159: Is there any evidence? Without supporting data (even preliminary data on cell lines) it seems like excessive speculation.

Author’s response: Thank you for identifying the weakness in this study. Additional references from clinical studies are provided to support the statement in the revised and improved manuscript.

Reviewer’s comment #9

Line 165: if the down regulation of Wnt/ b-catenin inhibitors is due to methylation of their promoters, the effect of hypo-methylating drugs (e.g. azacitidine), already used in the clinic for the standard treatment of old/unfit AML patients, should be sufficient for their reactivation, so it does not seem that the specific use of inhibitors of this signaling pathway can lead to a greater therapeutic response.

Author’s response: Thank you for your comment. We completely agree that hyper methylation of the Wnt/β-catenin inhibitors may be a common mechanism of low expression which can be addressed with hypo-methylating drugs (e.g., azacytidine and others) in elderly AML patients. However, there are additional mechanisms of low Wnt/β-catenin inhibitors expression, such as low expression of micro-RNA impacting DKK3 upregulation impacting the Wnt/β-catenin signalling as reported in adult B-cell lymphoblastic lymphoma (Oncotarget. 2017 Jul 18;8(38):64114-64128) and several other mechanisms as reviewed by David J. Stewart (J Natl Cancer Inst. 2014 Jan;106(1): djt356. doi: 10.1093/jnci/djt356). The efficiency of hypo-methylating drugs in elderly AML patients is moderate, such as 67% complete remission with 8 months study period in a recent report (Blood. 2019 Jan 3;133(1):7-17). This suggest that there are other mechanisms at play in addition to hypermethylation. Hence, additional approaches including combination therapies like Venetoclax and other investigational agents are being under extensive evaluation (Expert Rev Hematol. 2020 Jun;13(6):619-643). Since, supplemental pharmacotherapies with exogenous Wnt/b-catenin inhibitors are providing benefits in solid cancers (Tissue Cell. 2022 Aug; 77: 101853), such agents may have therapeutic potential in old age AML as well. To this effect, a dedicated paragraph is added into the discussion section of the revised manuscript.

Reviewer’s comment #10

Line 226: the Authors state that "Restoring expression of suppressed Wnt/β-catenin inhibitors offer a novel strategy of managing AML". How? Explain.

Author’s response: Thank you for your comment. Please see response above.

Reviewer 2 Report

In this study, the authors used available formalin-fixed paraffin-embedded (FFPE) diagnostic bone marrow biopsy samples to conduct a comparative gene expression analysis on the NanoString platform focused on the expression levels of Wnt/β-catenin pathway inhibitors in elderly versus pediatric AML patients. The authors hypothesized that the dysregulation of the Wnt/β-catenin signaling pathway plays key roles in the initiation, progression and prognosis of old-age AML, which distinguishes this pathology from pediatric AML. According to their gene expression analysis methodology, the authors found that Wnt/β-catenin target genes (i.e., MYC, MYB, RUNX1) were upregulated while Wnt/β-catenin inhibitors (e.g., CXXR, DKK1-4, SFRP1-4, SOST, and WIF1) were suppressed in old-age AML compared to pediatric AML. Based on these findings, the authors concluded that the Wnt pathway is overactive in old-age AML and  the therapeutic restoration of suppressed Wnt/β-catenin inhibitors might offer a novel strategy for managing elderly AML patients. 

 

I am not convinced that the evidence presented by the authors, strictly based on the methodology employed in the study, is enough to support the above conclusion. My main comments for the authors are as follows:

 

1.     The assertion that the Wnt/β-catenin pathway is dysregulated (overactive) in old-age AML versus pediatric AML also needs to be properly documented at protein level on the analyzed patient samples—i.e., IHC staining for nuclear β-catenin to demonstrate canonical Wnt pathway overactivation and/or IHC staining for ROR1 activity if the overactivation of non-canonical Wnt pathway is suspected instead. The simple measurement of various levels of Wnt-related transcripts conducted in isolation cannot clearly ascertain the overactivation of this pathway between patient samples. 

2.     Similarly, the findings at transcript level which show differential expression of Wnt-related transcripts between patient samples need to be properly validated by IHC staining at protein level. This needs to be done at least for some of the inhibitors that were quantified by the authors (e.g., CXXC4, DKK1, SFRP1-4, etc.). Once again, transcriptomic data in isolation cannot be used to draw any definitive conclusions on the functional activity of the Wnt pathway. 

3.     The activity of Wnt-pathway inhibitors in cancer is highly contextual and the finding that certain Wnt inhibitors are suppressed in cancer cells doesn’t always and/or necessarily correlate with a worse prognosis. For instance, it was shown that WNT/β-catenin signaling inhibition in cancer cells or in the tumor microenvironment (TME) owing to DKK1 upregulation could lead to immune evasion through the accumulation of myeloid-derived suppressor cells (MDSC) and exclusion of natural killer and cytotoxic T cells from the TME. In this context, a pro-WNT signaling therapy was proposed using an anti-DKK1 mAb which may be applicable for the treatment of immune evasion in cancer patients with DKK1 upregulation. Therefore, a more nuanced discussion should be added by the authors on the contextual significance of their findings pertaining Wnt inhibitors.  

 

 

 

 

 

 

Author Response

REVIEWER # 2

In this study, the authors used available formalin-fixed paraffin-embedded (FFPE) diagnostic bone marrow biopsy samples to conduct a comparative gene expression analysis on the NanoString platform focused on the expression levels of Wnt/β-catenin pathway inhibitors in elderly versus pediatric AML patients. The authors hypothesized that the dysregulation of the Wnt/β-catenin signaling pathway plays key roles in the initiation, progression and prognosis of old-age AML, which distinguishes this pathology from pediatric AML. According to their gene expression analysis methodology, the authors found that Wnt/β-catenin target genes (i.e., MYC, MYB, RUNX1) were upregulated while Wnt/β-catenin inhibitors (e.g., CXXR, DKK1-4, SFRP1-4, SOST, and WIF1) were suppressed in old-age AML compared to pediatric AML. Based on these findings, the authors concluded that the Wnt pathway is overactive in old-age AML and  the therapeutic restoration of suppressed Wnt/β-catenin inhibitors might offer a novel strategy for managing elderly AML patients. 

I am not convinced that the evidence presented by the authors, strictly based on the methodology employed in the study, is enough to support the above conclusion. My main comments for the authors are as follows:

Reviewer Comment #1 The assertion that the Wnt/β-catenin pathway is dysregulated (overactive) in old-age AML versus pediatric AML also needs to be properly documented at protein level on the analyzed patient samples—i.e., IHC staining for nuclear β-catenin to demonstrate canonical Wnt pathway over-activation and/or IHC staining for ROR1 activity if the over-activation of non-canonical Wnt pathway is suspected instead. The simple measurement of various levels of Wnt related transcripts conducted in isolation cannot clearly ascertain the over-activation of this pathway between patient samples. 

Authors response.  Thank you for your comments related to the deficiency in current study. We fully acknowledge this limitation in this retrospective study, as limited tissue was available in the diagnostic bone marrow biopsy samples, hence construction of tissue microarray (TMA) was not possible to confirm protein expression levels by immunohistochemistry. We are also in full agreement that this pilot project require additional validation proteomic studies. In view of this limitation, the manuscript is submitted as “Communication” and not as “Original research article”. In such situation, we can take guidance from the published studies which have reported this correlation between transcript and protein. We have added the following paragraph to reflect such limitation and have supported our observations through indirect evidence citing relevant references from other studies.

“The observations presented here are based on our preliminary data and should be inferred in the context of the limitations of current pilot project. These extrapolations require validation through additional comprehensive studies, such as linkage of RNA levels with protein expression. However, there is ample indirect evidence by other investigators linking Wnt/β-catenin transcripts with protein (either by immunohistochemistry or western blot) in solid cancers and other pathological processes. In this regard additional reference ( # 47-50 ) has been provided.

 

Reviewer Comment #2- Similarly, the findings at transcript level which show differential expression of Wnt related transcripts between patient samples need to be properly validated by IHC staining at protein level. This needs to be done at least for some of the inhibitors that were quantified by the authors (e.g., CXXC4, DKK1, SFRP1-4, etc.). Once again, transcriptomic data in isolation cannot be used to draw any definitive conclusions on the functional activity of the Wnt pathway. 

Authors response.  Thank you for your comments related to the deficiency in current study. We fully acknowledge these limitations. We are in full agreement that this pilot project require additional validation studies. In view of limited samples available, confirmation at protein level by immunohistochemistry was not possible. Please see above for more detail response.

Reviewer Comment # 3.   The activity of Wnt-pathway inhibitors in cancer is highly contextual and the finding that certain Wnt inhibitors are suppressed in cancer cells doesn’t always and/or necessarily correlate with a worse prognosis. For instance, it was shown that WNT/β-catenin signalling inhibition in cancer cells or in the tumor microenvironment (TME) owing to DKK1 up-regulation could lead to immune evasion through the accumulation of myeloid-derived suppressor cells (MDSC) and exclusion of natural killer and cytotoxic T cells from the TME. In this context, a pro-WNT signalling therapy was proposed using an anti-DKK1 mAb which may be applicable for the treatment of immune evasion in cancer patients with DKK1 upregulation. Therefore, a more nuanced discussion should be added by the authors on the contextual significance of their findings pertaining Wnt inhibitors.  

Authors response.  Thank you for your comments related to this deficiency in current manuscript. We are in full agreement with your comments. We also acknowledge, that although DDK1 and its relationship with immune evasion is addressed in our original submission (line # 195-196), but its therapeutic utility in relation of inducing host immunity against cancer was not explicatively addressed. To address this deficiency, discussion section is modified to include this contextual significance of DKK1 expression, which is divergent from being a Wnt inhibitor. We have added some key references in this context. The inclusion of this perspective of Wnt-pathway inhibitors, not only add noteworthy merit to our manuscript but also plug in the gap in current literature related to the spectrum of the clinical application of Wnt-pathway inhibitors in old age AML.

Round 2

Reviewer 1 Report

The Authors have made the necessary modifications for the improvement of the manuscript which is suitable for publication in the present form.

Reviewer 2 Report

I thank the authors for their clarification regarding the nature and scope of their study (i.e., "Communication" versus "Original Research Article") and for their thoughtful responses to my comments. The further editing efforts made by the authors, especially to the Discussion section of their manuscript, provide additional clarity to their findings and a better understanding on the intricacies pertaining therapeutic targeting of aberrant Wnt signaling in old-age AML.