Neuroendocrine Neoplasms of the Esophagus and Esophagogastric Junction in Germany, 2009–2023
, Ina Wellmann 2, Bernd Holleczek 3, Alice Nennecke 4, Guido Schumacher 5 and Hiltraud Kajüter 2
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis manuscript provides an assessment of register-based data on a rare disease obtained in a large European population. It contributes with important basic and outcome data, highly useful for further research. Results largely confirm those of a few prior database studies, mostly from the US, but some are divergent. As patients are included over many years and histology is not reviewed or reclassified according to recent recommendations it has some limitations, but these are discussed by authors.
My main concern (besides data quality mentioned above) is the number of missing variables that are not accounted for in detail and might affect the validity of results. I suggest that authors construct a table presenting baseline data on NEN patients that includes the number and percentages of missing variables for each parameter included in the calculations.
The manuscript is very data dense and and readability could be improved by reducing details, including minor results already shown in tables and figures.
A short description of the data and data source used for assessing the population of adenocarcinomas and squamous cell carcinomas is needed to evaluate, whether these are comparable to the population of NENs. However, the section from line 243 and onwards “Adenocarcinoma and squamous cell carcinoma of the esophagus and esophagogastric junction” is out of scope (as it does not relate to NEN) and should be deleted.
Specific comments
Line 130-131: “we excluded cancer registries that had problems with vital status monitoring 130 from the survival analyses”. From the supplementary file it is found that 1320 of 2025 patients were included in the analysis. These numbers should be included in the abstract and the main manuscript text.
Line 144-145: “Overall, 95.2% were histologically verified”. Did you include not histologically verified cases and, if so, how were these diagnosed as NEN?
I suggest that Table 2 and Figure 1 are placed in the Supplementum as results are easily described in text. On the other hand, I suggest that Suppl. Figure 1 is included in the text as it is very illustrative.
Author Response
REVIEWER 1
Comment 1
This manuscript provides an assessment of register-based data on a rare disease obtained in a large European population. It contributes with important basic and outcome data, highly useful for further research. Results largely confirm those of a few prior database studies, mostly from the US, but some are divergent. As patients are included over many years and histology is not reviewed or reclassified according to recent recommendations it has some limitations, but these are discussed by authors.
My main concern (besides data quality mentioned above) is the number of missing variables that are not accounted for in detail and might affect the validity of results. I suggest that authors construct a table presenting baseline data on NEN patients that includes the number and percentages of missing variables for each parameter included in the calculations.
The proportion of missing values was presented for all variables in the original submission.
Table 1: „NET, G unknown“
Table 2: „Unspecified localization within esophagus”
Table 3: “Unspecified cancers”
Suppl. Table 1: “Unknown grade”, “unknown stage”
Suppl. Table 5: “Unspecified localization”
Comment 2
The manuscript is very data dense and and readability could be improved by reducing details, including minor results already shown in tables and figures.
We removed Figure 1 from the manuscript to the supplementary file to reduce the density of data presentation. We deleted the complete result paragraph related to adenocarcinomas and squamous cell carcinomas from the main text.
Comment 3
A short description of the data and data source used for assessing the population of adenocarcinomas and squamous cell carcinomas is needed to evaluate, whether these are comparable to the population of NENs. However, the section from line 243 and onwards “Adenocarcinoma and squamous cell carcinoma of the esophagus and esophagogastric junction” is out of scope (as it does not relate to NEN) and should be deleted.
To compare the epidemiology of NENs with adenocarcinomas and squamous cell carcinomas, we extracted not only the NENs but also the adenocarcinomas and squamous cell carcinomas from the same data set. We have added this information to the methods section.
Tables 2 and 3 in the results section thus enabled a direct comparison of the anatomical distribution and incidence of NENs, adenocarcinomas, and squamous cell carcinomas.
Comment 4
Line 130-131: “we excluded cancer registries that had problems with vital status monitoring from the survival analyses”. From the supplementary file it is found that 1320 of 2025 patients were included in the analysis. These numbers should be included in the abstract and the main manuscript text.
We added these numbers in the abstract and main manuscript text.
Comment 5
Line 144-145: “Overall, 95.2% were histologically verified”. Did you include not histologically verified cases and, if so, how were these diagnosed as NEN?
The term “histologically verified” is cancer registry jargon. It simply means that the cancer registry has been informed that the cancer has been histologically verified. The remaining 4.8% have in all likelihood also been histologically verified, but the cancer registry has not been informed of this.
We have described this situation in more detail in the revision:
“Overall, the cancer registries had information that 95.2% of all NENs had been histologically verified.”
Comment 6
I suggest that Table 2 and Figure 1 are placed in the Supplementum as results are easily described in text. On the other hand, I suggest that Suppl. Figure 1 is included in the text as it is very illustrative.
As suggested, we have moved Figure 1 to the supplement. As mentioned above, we have reduced the amount of results relating to adenocarcinomas and squamous cell carcinomas in the main text.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors analyzed the clinicopathological features of neuroendocrine neoplasms (NENs) of the esophagus and esophago-gastric junction in Germany population.
This article is interesting and can provide important information in the field of oncology.
There are some concerns.
- As mentioned in the limitation section of the Discussion, the classification and diagnostic criteria of NENs, such as neuroendocrine tumor (NET) G3 and small cell neuroendocrine carcinoma (NEC), is very important. It may be useful to divide the population into patients with NEN before and after the introduction of the recent WHO Classification.
2. In this cohort, the prognosis was poorer in patients with NET G3 than with NEC. It must be discussed.
Author Response
REVIEWER 2
Comment 1
- As mentioned in the limitation section of the Discussion, the classification and diagnostic criteria of NENs, such as neuroendocrine tumor (NET) G3 and small cell neuroendocrine carcinoma (NEC), is very important. It may be useful to divide the population into patients with NEN before and after the introduction of the recent WHO Classification.
After consulting with NEN pathologists in Germany, the problem is that there is no clear or sharp calendar year from which the new WHO classification was adopted by pathologists. Years before 2019, a gradual shift in thinking began in terms of the classification of NEN. Nevertheless, we explored the suggested approach by the reviewer and stratified the analyses by calendar year: 2009-2018 versus 2019-2023.
The comparison of the estimated hazard ratios for NEC and MiNEN compared to NET in 2009-2018 and 2019-2023 showed barely any difference. The comparison of the estimated hazard ratios for NET G2 and NET G3 compared to NET G1 in 2009-2018 and 2019-2023 showed that the precision of the hazard ratio estimates was so low that a meaningful interpretation of the results was not possible.
Comment 2
- In this cohort, the prognosis was poorer in patients with NET G3 than with NEC. It must be discussed.
The relative 5-year survival for NET G3 was 9.0% with a standard error of 7.6%, while the relative 5-year survival rate for NEC was 12.5% with a standard error of 1.3%. In our view, the imprecision (standard error 7.6%) for NET G3 does not allow us to conclude that NET G3 has a worse prognosis than NEC.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsAuthors have addressed reviewers criticism and the paper now stands out as scientifically important and easily readable.
