Expression of Hypoxia-Inducible Factors in Different Stages of Pancreatic Tumor Progression

Round 1

Reviewer 1 Report

The MS provides novel approaches to the pancreatic neoplasms that would facilitate further investigations. The paper, however, needs some additional data.

To get correct conclusions, the studied groups should be as homogeneous as possible. The authors generalize about the given diseases, but this view may lead to false interpretation. This is the major issue.

It should be clearly stated what kind of pancreatic cancers were studied. The results may significantly vary when the well-moderately-poorly differentiated forms, or the different stages are grouped together. Similarly, the "NET" terminology can be misleading if the G1/G2/G3 tumors are not separated. One also cannot generalize about the PanIN, because PanIN-1-2-3 cases harbor different genetic alterations, therefore their biological behavior is dissimilar. At least the low grade/high grade PanINs should be distinguished.

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The study "Expression of Hypoxia-Inducible Factors in Different Stages of Pancreatic Tumor Progression" provides the analysis of HIF-1α and HIF-2α protein staining in pancreatic tissues ranging from PanIN and IPMN to NET and PDAC as well as matching non-tumor tissues.

As authors pointed out, a limited number of samples was available; however, there was an increased staining intensity with increased severity of the disease.

There are some major concerns:

1. Explain if PanIN and IPMN samples came from the same patients as PDAC or NET? If different patients, how come there are 29 samples with pancreatic changes and only 21 samples matching non-tumor samples?
2. In Figure 1 please include PanIN as well.
3. Please take pictures with higher magnification so that nuclear staining can be assessed.
4. In Figure 1 both HIF-1α and HIF-2α IHC staining seem quite intense in cytoplasm, which does not match the text (page 3, line 91-93).
5. Include scale bars in Figure 1.
6. Show the representative pictures of "strength of immunoreactivity" ranging from I to IV that corresponds to Table 3, and/or show representative pictures of "proportional score" and "intensity score".
7. Table 3 - Does the "strength of immunoreactivity" corresponds to staining intensity of nucleus or cytoplasm or it accounts for staining of both (nucleus + cytoplasm)?
8. Perhaps Table 2 is not necessary since Figure 2 shows the exactly same data in addition to statistics. However, from the Figure 2 it appears that statistics is calculated as a comparison to PanIN instead of non-tumor tissues which is suggested in the Figure 2 legend. (page 5, line 119).
9. Instead of bar graph in Figure 2, represent data with the scatter plot.
10. Please label the x and y axes in Figure 3 - what are the numbers referring to? Final score?
11. Discussion is somewhat redundant with the introduction - especially the second paragraph of Discussion (page 6, lines 144-149).
12. Mentioning of SIAH in Discussion has no direct relevance to this manuscript without performing additional experiments (e.g. IHC for SIAH) on the same set of samples. The discussion paragraph from line 159 to 169 (page 7) is not really useful, since numerous other connections based on HIF pathway can be made as well. 

Minor corrections:

1. Table 2 shows "final scores" - this term has to be defined earlier in text.
2. Page 5, line 123 - "As described before,..." - it was not described before so please remove that statement.
3. It would be preferred to show number of analyzed pictures instead of % in Table 3.
4. Table 3 legend - there is no mention of SEM in it so please remove it from page 6, line 133.
5. In discussion, please expand on the outcome of the clinical trials mentioned on page 6, line 155,156 (e.g. which diseases were included in those trials, shortly explain what is tirapazamine, etc.)
6. Page 8, line 206 - it says "X anti-rabbit" - please replace the X with species.

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Thank you for making suggested corrections to the manuscript.

A few minor changes: 

• the cytoplasm still looks clearly positive for HIF-1α and HIF-2α
• double check if scale bar really corresponds to 100 μm
• figure 3/supplementary figure 3 - I apologize for confusion - I didn't ask for the column mean graph that you showed in supplementary figure S3, but scatter plot - this would plot individual samples and allow a reader to assess the distribution of samples within each group (I understood there were more than one measurement in each group. Alternatively, it is unlikely that many measurements all fall at the exactly same spot)
  • here is a link to an example of a scatter plot: https://communities.sas.com/t5/Graphics-Programming/Categorical-scatterplots-similar-to-PRISM/td-p/140103
  • I am not suggesting the use of any specific software to make this graph, just using this as an example of what I meant by scatter plot
• Figure 2 - for example if you are comparing NT and PDAC, and calculate the statistics for that particular comparison, then the line should stretch from NT to PDAC. Otherwise, a reader understands that statistics was calculated for comparison between PanIN and PDAC.
• Figure 2 legend: "*p<0.0001 versus NT." This statement is unclear. Remove or describe WHAT group versus NT.
• page 8, line 209: I believe you meant "goat" rather than "gout".

Author Response

Please see attachment.

Author Response File: Author Response.pdf