High TLR6 Expression Status Predicts a More Favorable Prognosis after Esophagectomy for Locally Advanced Thoracic Esophageal Squamous Cell Carcinoma

Round 1

Reviewer 1 Report

I thank the editor for suggesting that I review this manuscript.

 

This study is very interesting and highlights, for the first time, the role of TLR6 and PNG in the prognosis of ESCC.

 

The objectives of the study were achieved and the results are useful to readers also because they highlight the importance of a healthy oral microbiota.

 

However some changes shoul be made:

 

It is necessary to briefly describe the microarray procedure.

 

The authors should report a table with the results obtained for TLR4 even if published in their other study, readers should have the results available in the text.

 

 

Author Response

It is necessary to briefly describe the microarray procedure.

Response: We really appreciate the constructive suggestion. Based on this suggestion, we added sentences to explanation about TMA in Material and Methods as follows. “An ESCC tissue microarray (TMA) was constructed at the Pathology Institute, Toyama, Japan, as previously described [14, 19-21]. According to many validations of TMA to overcome cancer tissue heterogeneity, assessment of double cores measuring 0.6 mm in diameter sufficiently reflect the whole section [22]. We therefore employed triplicate cores measuring 0.6 mm in diameter for further enhance its reliability. Triplicate cores were collected randomly from separate carcinoma areas and transferred to the TMA. The TMA block contained 531 cores (3 cores each from 177 paraffin blocks of main tumor from enrolled ESCC patients) in total.” 

 

The authors should report a table with the results obtained for TLR4 even if published in their other study, readers should have the results available in the text.

Response: Based on this constructive suggestion, we added Supplement Table1 summarizing TLR4-high group and TLR4-low group in the same cohort. If Supplement Table is not appropriate, we could change it as Table3 and Table3 as Table4. Our previous table about TLR4 was based on UICC7th, but this Supplement Table1 is based on UICC8th.

We also added sentences about this table in the Results 3.4. as follows. “Our previous data of the two groups, TLR4-high (n=132) and TLR4-low (n=45) are summarized in Supplement Table 1 [14]. Exhibiting combined TLR6-low/TLR4-low were 25 patients (14.1%), TLR6-low/TLR4-high were 87 patients (49.2%), TLR6-high/TLR4-low were 20 patients (11.3%) and TLR6-high/TLR4-high were 45 patients (25.4%). There was no statistical correlation between TLR6 and TLR4 expression status (p=0.21). Patients exhibiting combined TLR6-low/TLR4-high expression had much worse 5-year OS and DSS than other patients (p=0.0038 and p=0.0285, respectively) (Figure 4). Consistent with the results of 5-year OS and DSS, the patient survival rate was twice as high among TLR6 high/TLR4-low patients (60.0%) than among TLR6 low/TLR4-high patients (33.3%) (Table 3).”

 

Reviewer 2 Report

High TLR6 expression status predicts a more favorable prognosis after esophagectomy for locally advanced thoracic esophageal squamous cell carcinoma by Yusuke Sato, Akiyuki Wakita, Eri Maeda , Yushi Nagaki , Ryohei Sasamori , Kohei Kemuriyama  , Shu Nozaki , Satoru Ito, Kaori Terata , Kazuhiro Imai , Hiroshi Nanjo, Kyoko Nomura, and Yoshihiro Minamiya,  is a paper aimed to explore the correlation of TLR6 expression status with prognosis after curative esophagectomy and to examine  whether PGN influence cell proliferation activity of ESCC lines on 177  patients. High TLR6 expression (categorized as 3+ and 2+) correlated with significantly more favorable 5-year overall survival (OS) and disease-specific survival (DSS) after esophagectomy than lower TLR6 expression (1+ and 0). Authors found that TLR6 expression status was an independent prognostic factor affecting 5-year OS. PGN significantly inhibited cell proliferation activity of ESCC lines.

I think that the paper could be of good quality but it needs to be improved.

First, why did the authors employ the HER2 IHC scoring system (ASCO & CAP guidelines)? Is the criterion of assigning a score of 3+ if there was intense TLR6 staining in the cytoplasm or nuclei in more than 30% cells based on an objective method? Why didn't the Authors perform a ROC curve to evaluate a prognostic cutoff value? In my opinion, adopting the criteria used for another marker such as HER2 IHC, seems rather arbitrary and biased.

Furthermore, did the authors evaluate the value of the agreement among the 3 physicians blinded to the clinical data?

 

 

 

Author Response

First, why did the authors employ the HER2 IHC scoring system (ASCO & CAP guidelines)? Is the criterion of assigning a score of 3+ if there was intense TLR6 staining in the cytoplasm or nuclei in more than 30% cells based on an objective method? Why didn't the Authors perform a ROC curve to evaluate a prognostic cutoff value? In my opinion, adopting the criteria used for another marker such as HER2 IHC, seems rather arbitrary and biased.

Response: We really appreciate the constructive suggestion. As this reviewer mentioned, we performed ROC curve to evaluate a prognostic cutoff value in our previous studies. However, this method varies widely by physicians and complicated. Therefore, we employed simpler HER2 IHC scoring system and it worked well in our previous studies, Ref14,20,21. So, we employed this method in the present study as well.

 

Furthermore, did the authors evaluate the value of the agreement among the 3 physicians blinded to the clinical data?

Response: We really appreciate the important question. We answered to this question in Material and Methods as follows. “Three physicians blinded to the clinical and the prognostic data assigned a staining score.” And “If the scoring was not unanimous, the score assigned by 2 of the 3 physicians was adopted.”

Reviewer 3 Report

In this study, Sato et al found that TLR6 expression status can be used for an independent prognostic factor affecting 5-year OS of ESCC patients. The authors found that High expression of TLR6 was associated with better 5-year OS and DSS after esophagectomy compared to low expression of TLR6. Further, the authors also found that TLR6 high/TLR4 low patients showed better survival than TLR6 low/TLR4 high patients. These data may indicate the contribution of oral gram-positive bacteria. All studies are well-planned, the results are remarkable, and the conclusions are convincing.  

Minor comments,

1. TLRs indeed recognize bacterial-derived PAMPs, but they are also known to react to self cell-derived components at the same time. Since the authors did not actually use gram-positive bacteria in this study, it would be better to mention other possibilities in addition to the involvement of beneficial bacteria.

2. In Fig 5, it would be nice to show or argue why PGN treatment to KYSE190 and OE21 cells reduces cell proliferation. Does PGN induce cell death or cell-cycle arrest in the cell lines?

Author Response

1. TLRs indeed recognize bacterial-derived PAMPs, but they are also known to react to self cell-derived components at the same time. Since the authors did not actually use gram-positive bacteria in this study, it would be better to mention other possibilities in addition to the involvement of beneficial bacteria.

Response: We really appreciate the constructive and important suggestion. Based on this suggestion, we added sentences about DAMPs in Discussion as follows. “Another limitation causes cognitive ability of TLRs. Recent evidence indicates that in addition to PAMPs, TLRs recognize damage-associated molecular patterns (DAMPs), which are endogenous molecular patterns released from injured or dying host cells (9-11). Therefore, there is a possibility that TLR6 recognizes not only PGN but also DAMPs from host cells. Validation of this possibility also needs further studies.”

 

 

2. In Fig 5, it would be nice to show or argue why PGN treatment to KYSE190 and OE21 cells reduces cell proliferation. Does PGN induce cell death or cell-cycle arrest in the cell lines?

Response: We really appreciate the constructive suggestion. As we mentioned in Discussion, we could find only a couple of studies about tumor inhibitory effect of TLR6 signaling, Ref25, 26 and 27. However, these mechanisms are still unclear even in these studies. We added sentences about this in limitations as follows. “The mechanisms of this inhibitory effect and relation to prognosis through TLR6 signaling mediators and cytokines are remains to be determined in further studies.”

Round 2

Reviewer 2 Report

The quality of the paper has improved and it can be accepted for publication.