Immunohistochemical Analysis of Nicotinamide Phosphoribosyltransferase Expression in Gastric and Esophageal Adenocarcinoma (AEG)

Round 1

Reviewer 1 Report

The authors provide a brief report on a study of NAMPT expression in esophagogastric junction and stomach adenocarcinomas. Overall, the authors provide a nice, concise report utilizing a large, clinically annotated cohort with one major  issue being the way the results are reported and a couple of small questions that may help improve the manuscript (see below). 

1A. "Regarding the prognostic relevance of NAMPT expression for both age 70 groups, the survival analysis showed a better prognosis for NAMPT positive compared 71 to negative patients with an age <65 years (p=0.263). In patients with older age (>65) 72 NAMPT showed the opposite effect with better prognosis for NAMPT negative compared 73 to positive patients (47.4 vs 27.4; p=0.129)."  Given the insignificant P values, it does not really make sense to claim better or worse prognosis. One could suggest a trend but that is about it.  

1B. "The differences in overall survival between 74 these four groups reached level of significance (p=0.02) " How was this calculated and what does the suggest or mean (i.e.what exactly is being tested)? 

1C. "Furthermore, we see that in patients with older age (>65) and positive NAMPT ex-111 pression the prognosis is worse compared to NAMPT negative cases."  Is this accurate (ie is there a statistical difference? 

2. Can the authors expand a little on the exact definition of the samples used? i.e. How was AGE defined? Were all gastric cancers included or were only proximal included? How proximal into the esophagus were included?

3. How does the patient cohort characteristics and survival compare to published numbers for the region? i.e. Was the cohort representative of the AGE population?

Author Response

1A. "Regarding the prognostic relevance of NAMPT expression for both age  groups, the survival analysis showed a better prognosis for NAMPT positive compared  to negative patients with an age <65 years (p=0.263). In patients with older age (>65) NAMPT showed the opposite effect with better prognosis for NAMPT negative compared to positive patients (47.4 vs 27.4; p=0.129)."  Given the insignificant P values, it does not really make sense to claim better or worse prognosis. One could suggest a trend but that is about it.

The reviewer is right, that the reported p values do not reach a level of significance, so that we can only suggest a trend in this age dependent comparison. According changes have been made:

Results, 4^th paragraph:

Regarding the prognostic relevance of NAMPT expression for both age groups, the survival analysis showed a numerical longer survival that did not reach statistical significance for NAMPT positive compared to negative patients with an age <65 years (p=0.263). In patients with older age (>65) NAMPT showed the opposite effect with a trend towards a better prognosis for NAMPT negative compared to positive patients (47.4 vs 27.4; p=0.129).

1B. "The differences in overall survival between these four groups reached level of significance (p=0.02) " How was this calculated and what does the suggest or mean (i.e.what exactly is being tested)?

We are very sorry, that the description of the statistical analysis was omitted in the manuscript. Following section was added:

Materials and Methods, 4^th paragraph:

Statistical analysis was performed using IBM SPSS Version 24. Overall survival was defined as time from diagnosis to death or last follow-up and was compared using Kaplan-Meier method with the log-rank test for assessment of statistical significance. The associations of NAMPT expression with clinic-pathologic characteristics were tested by using the X2 test.

1C. "Furthermore, we see that in patients with older age (>65) and positive NAMPT expression the prognosis is worse compared to NAMPT negative cases."  Is this accurate (ie is there a statistical difference?

The reviewer already made the observation in the Results section, that the p values of the subgroup survival analyses do no reach levels of significance. Besides the previous changes the following modifications were included:

Discussion, 3^rd paragraph:

Furthermore, we find numerical longer survival that did not reach statistical significance in NAMPT negative compared to positive patients (47.4 vs 27.4; p=0.129) in patients with older age (>65). One might expect an association between higher age, NAMPT linked obesity and prognosis, but previous studies have suggested a potential protective effect on mortality in overweight and mild obese patients in gastric cancer, which was called “obesity paradox” [22, 23]. In this context the reasons for the trend towards a worse prognosis of NAMPT positive, elderly patients remains unclear. However, there are also statistical limitations to these findings as only 9.8% of our cohort showed a NAMPT expression which leads to very small subgroups for further analysis

2. Can the authors expand a little on the exact definition of the samples used? i.e. How was AGE defined? Were all gastric cancers included or were only proximal included? How proximal into the esophagus were included?

According to the reviewer`s comment we expanded the information of the used samples:

Materials and Methods, 1^st paragraph:

Tissue samples were collected from the archive of the Department of Pathology, Charité–Universitätsmedizin Medicine Berlin. Paraffin-embedded tumor samples were available from surgically treated chemotherapy-naive patients. All samples were reevaluated according to histological diagnosis, tumor stage, and grade, and classified by the histological architecture of AGE/S carcinoma using Lauren and Ming classification by a specialist in gastrointestinal pathology. The classification of adenocarcinoma of the esophagogastric junction (AEG) was applied as defined by Siewert and Stein and later approved at the second International Gastric Cancer Congress in Munich in April 1997 (29,30). In our cohort all gastric cancers were included in the AEG type III.

3. How does the patient cohort characteristics and survival compare to published numbers for the region? i.e. Was the cohort representative of the AGE population?

According to the reviewers comment we included further information of the cohort in comparison to published series.

Materials and Methods, 1^st paragraph:

As previously reported, the clinical cohort characteristics is comparable with those of other studies populations (31).

Reviewer 2 Report

The article is well-written, clear, and concise.

The major limitation is that the study is retrospective. Due to this situation, normal mucosa biopsies are missing. Is important to know if there are differences in the expression of NAMPT in normal and pathological biopsy samples. Can you obtain normal gastric mucosa tissue samples from your pathology department fur further immunohistochemical assessment?

Due to the small number of patients with high NAMPT expression, statistical analysis cannot be very accurate. 

Larger studies have to be taken into account. 

Author Response

1. The major limitation is that the study is retrospective. Due to this situation, normal mucosa biopsies are missing. Is important to know if there are differences in the expression of NAMPT in normal and pathological biopsy samples. Can you obtain normal gastric mucosa tissue samples from your pathology department fur further immunohistochemical assessment?

The reviewer is right about the limitations of a retrospective study.

However, the TMA tumor samples included adjacent normal mucosa, which was negative for NAMPT expression. According to the reviewer`s comment we included that fact:

Results, 2^nd paragraph:

A positive NAMPT expression was detected in 26 of 296 tumor samples (9.8%) while the adjacent normal gastric mucosa was negative for NAMPT expression.

2. Due to the small number of patients with high NAMPT expression, statistical analysis cannot be very accurate. 

We agree with the reviewer`s observation. However, we already addressed these limitations:

Discussion, 3^rd paragraph:

However, there are also statistical limitations to these findings as only 9.8% of our cohort showed a NAMPT expression which leads to very small subgroups for further analysis.

3. Larger studies have to be taken into account. 

The reviewer is right that larger studies might help in elucidating the role of NAMPT in adenocarcinoma of the esophagogastric junction and stomach (AEG/S), especially due to the low expression rate. However, we would like to present the data of a single institution study with a cohort comparable to previous studies on NAMPT expression in other solid cancers.

Reviewer 3 Report

This study investigated immunohistochemical analysis of NAMPT expression in gastric and esophageal adenocarcinomas. It is interesting, but there are some issues that need to be addressed.

1.      Introduction (line47): Please insert a reference after Bi et al. Moreover, there are several missing references after et al.

2.      Introduction (line48): gastric cancer seems not to be prevalent in Germany. Please describe why you authors investigate specifically gastric cancers: there needs to be an explanation of the incidence, and prevalence of gastric or esophageal cancers in Germany compared to other countries and their risk factors.

3.       Introduction (line50): Since gastric cancer and esophageal adenocarcinoma have different pathogenesis, why do authors investigate these two diseases? 

4.      Table 1: There are cautions to classifying esophageal and gastric junction adenocarcinoma to use staging classification between two cancer categories.

5.      Methods Material and methods: (Study cohort) you should explain detailed information about the respective distance from EG junction within esophageal gastric junction adenocarcinomas. The definition of EG junction adenocarcinoma is missing in the methods.

6.      Description of statistical analyses is missing in the manuscript.

 

7.      Authors investigated immunohistochemical staining according to IRS. Do you have any reference that is standard for scoring NAMPT expressions?

Author Response

1. Introduction (line47): Please insert a reference after Bi et al. Moreover, there are several missing references after et al.

We are sorry for the missing references after et. al. According changes have been made.

2. Introduction (line48): gastric cancer seems not to be prevalent in Germany. Please describe why you authors investigate specifically gastric cancers: there needs to be an explanation of the incidence, and prevalence of gastric or esophageal cancers in Germany compared to other countries and their risk factors.

The reviewer is right about a lower prevalence of gastric carcinoma in comparison to other countries. However, this the incidence of this disease is rising in Western countries affecting also Germany.

As a we expected that a detailed description might go beyond the scope of a brief report we omitted to depict the differences between the incidence and prevelance of gastric or esophageal cancers in Germany compared to other countries.

According to the reviewer`s comment we included this topic:

Discussion, 2^nd paragraph:

Although major genetic and environmental differences seem obvious between Western and Eastern countries, the role of these specific genetic mutations remains unclear (17). Moreover, the high seroprevalence of H. pylori infection in Asian countries has been linked to differences in the incidence of gastric carcinoma (18).

3. Introduction (line50): Since gastric cancer and esophageal adenocarcinoma have different pathogenesis, why do authors investigate these two diseases? 

We agree with the reviewer about different the different pathogenesis of gastric cancer and esophageal adenocarcinoma. However, the recent findings showed corresponding molecular features/groups within the TCGA studies [1, 2], so that it seems reasonable to us to investigate a molecular feature like NAMPT in both of these entities.

4. Table 1: There are cautions to classifying esophageal and gastric junction adenocarcinoma to use staging classification between two cancer categories.

The reviewer is right, that these two cancer categories with different TNM classifications have been combined. In order to take this in account, we included a note in the table captions.

Table 1: Patient characteristics of the analyzed patient cohort (combined TNM classification of AEG & gastric carcinoma) and distribution of NAMPT high and low expressing primary tumors. Significance calculated by X2-Test.

5. Methods Material and methods: (Study cohort) you should explain detailed information about the respective distance from EG junction within esophageal gastric junction adenocarcinomas. The definition of EG junction adenocarcinoma is missing in the methods.

According to the reviewer`s comment we included the named definitions:

Materials and Methods, 1^st paragraph:

The classification of adenocarcinoma of the esophagogastric junction (AEG) was applied as defined by Siewert and Stein and later approved at the second International Gastric Cancer Congress in Munich in April 1997 (29, 30). In our cohort all gastric cancers were included in the AEG type III.

6. Description of statistical analyses is missing in the manuscript. 

We are very sorry, that the description of the statistical analysis was omitted in the manuscript. Following section was added:

Materials and Methods, 4^th paragraph:

Statistical analysis was performed using IBM SPSS Version 24. Overall survival was de-fined as time from diagnosis to death or last follow-up and was compared using Kaplan-Meier method with the log-rank test for assessment of statistical significance. The associations of NAMPT expression with clinic-pathologic characteristics were tested by using the X2 test.

7. Authors investigated immunohistochemical staining according to IRS. Do you have any reference that is standard for scoring NAMPT expressions?

Unfortunately a standard scoring of NAMPT expression is lacking so far. Previous studies have applied different semiqualitative scores ranging from 0-3/0-4  [3, 4].  For this reason we think that an assessment of the expression according to IRS is more reasonable.

 

 

1. Integrated genomic characterization of oesophageal carcinoma. Nature 2017, 541(7636):169-175.
2. Comprehensive molecular characterization of gastric adenocarcinoma. Nature 2014, 513(7517):202-209.
3. Li H, Bai E, Zhang Y, Jia Z, He S, Fu J: Role of Nampt and Visceral Adiposity in Esophagogastric Junction Adenocarcinoma. J Immunol Res 2017, 2017:3970605-3970605.
4. Davis K, Dunseth CD, Mott SL, Cramer-Morales KL, Miller AM, Ear PH, Mezhir JJ, Bellizzi AM, Chan CHF: Nicotinamide phosphoribosyltransferase expression and clinical outcome of resected stage I/II pancreatic ductal adenocarcinoma. PLoS One 2019, 14(3):e0213576.

Round 2

Reviewer 2 Report

no comments

Reviewer 3 Report

The authors should submit by replying more clearly and writing down the revised contents in the reply sheet as well. Hard to check.