Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a nice scoping narrative review of the medical management of secreting syndromes in functioning pancreatic neuroendocrine tumors, which has been registered on the Open Science framework (ref 15). It focussed on medical treatment which is an important aspect of managing functional neuroendocrine tumors for functional and symptomatic control and treatment. The conduction of this scoping review has been strong, robust and thorough, as reflected throughout the paper. The authors have also rightly acknowledged the limitation that the studies collected are heterogeneous and the a scoping review has not addressed formal risk-of-bias, certainty assessment and cannot rank comparative effectiveness of the medical therapies.  I have the following suggestions to made.

1. As the readership may be more interested in  the "clinical" aspect rather than the methodological aspect of the study, I suggest that the methodological aspect in the conduction of this study be appropriately simplified and shortened. The tables had done very good jobs in reflecting the robustness of the methodology.
2. Though the authors have specified in the Introduction that control of biochemical and clinical symptoms of these syndromes is essential, often necessary before performance of tumor-directed therapy, I suggest that the importance of anti-tumor treatment should not be "played down" as this is the ultimate basis for the final control or resolution of the disease. If possible, the authors may also consider including some information on the contribution of ant-tumor treatment in the "Medical" control of these syndromes.
3. A  minor point on defining acronyms on first mention: ENETS, NANETS, ASCO in the text and in Table 2; SSA, PrPI/H2, PRRT, even though most readers would understand these acronyms.

Author Response

Thank you for your valuable feedback. Below is our detailed response to each comment.

Comment 1. The methodological detail could be simplified and shortened, since the readership is primarily clinical, and the tables already convey the rigor.

Response1. Thank you for your comment. Methodological detail that documents process rather than informing clinical interpretation has been moved to Supplementary File S1, which already carries the search strategy, selection flow and adjudication records. The Methods in the main text now retain only what a clinical reader needs in order to judge what was included and why. We have kept in the main text the two charting conventions that materially affect the results, because the numbers cannot be read correctly without them: the definition of a syndrome-control outcome, and the rule that syndrome and therapy categories are non-exclusive.

Comment 2. The importance of antitumor treatment should not be played down; please consider adding information on its contribution to medical control of these syndromes.

Response2. We agree, and we are grateful for the correction of emphasis. A new paragraph has been added to the Discussion. It states explicitly that separating hormonal control from antitumor activity is an analytical device rather than a claim about relative importance, and that reducing tumor burden is the basis on which durable syndrome control ultimately rests. It then shows that this is visible in our own corpus: chemotherapy contributed to 23 studies, peptide receptor radionuclide therapy to 19 and everolimus to 11, and in many of these the hormonal endpoint improved alongside the radiological one, everolimus in insulinoma being a clear example. The paragraph also explains why the two axes are nonetheless not interchangeable, citing hypercalcemia recurring before radiological progression, symptomatic relief preceding measurable tumor change, and somatostatin analogs aggravating hypoglycemia in insulinoma while tumor burden is unchanged.

Comment 3. Define acronyms at first mention: ENETS, NANETS, ASCO in the text and in Table 2; SSA, PPI/H2, PRRT.

Response3. ENETS, NANETS and ASCO are now expanded at first mention in the Abstract and Introduction; proton pump inhibitors and histamine H2-receptor antagonists are expanded in the Methods; somatostatin analog and peptide receptor radionuclide therapy are expanded at first use. All abbreviations are now defined in the table footnotes and figure legends as well as in the text.

Reviewer 2 Report

Comments and Suggestions for Authors

Manuscript ID life-4626599

Type Review

Title: Medical Management of Secreting Syndromes in Functioning Pancreatic Neuroendocrine Tumors: A Scoping Review and Evidence Map

 

Feedback for the Authors

This manuscript will answer a clinically relevant question and provide a helpful framework of the evidence for the medical management of functioning pancreatic neuroendocrine tumor syndromes. The separation of hormonal control from the antitumor activity is appropriate, and the attempt to supply some additional extraction and mapping data is a strong point. However, these are the issues to consider before acceptance:

Major Revisions

1. Explain the classification of a syndrome. Ten ENETS syndromes are mentioned in the Introduction, but ACTH- and CRH-secreting tumors are discussed separately. A CRH category does not appear in the evidence map, and there are ten rows. If ACTH and CRH are to be considered as a single category, please explain that and change the terminology as well as the tables, figures, abstract, and denominators accordingly.

 

2. Arrange the study-selection numbers in a study-selection list. The manuscript contains numerous totals: 5,757 records, 2,679 retained records, 77 original-review reports, 83 adjudicated reports, 321 contextual records, and 13,378 update records. Please give a clear explanation of the reconciliation of these numbers in the PRISMA flow diagram and explain the terms record, report, study, cohort, and patient precisely. A clearer explanation of the difference between 65 primary studies, 60 map-contributing studies, and five mixed records is needed.

 

3. Improve methodological reproducibility. Indicate if data extraction was done by more than one reviewer, how disagreements were resolved, and whether data extraction was checked against the original reports. Include a comprehensive list of full-text exclusions, with reasons, and explain the language limitations, database coverage, deduplication method, and lack of forward citation searching.

 

4. Describe in more detail what the outcome is. The biochemical or physiological response related to the 58/65 studies would seem to be more convincing than the evidence underlying it, since the endpoints reported were very heterogeneous. Please report these outcomes by syndrome, specify the response categories, and don't report changes in hormones, symptom improvement, electrolyte correction and glucose requirements, or tumor response as equivalent measures.

 

5. Make tentative decisions about the effectiveness of treatment. Due to the lack of risk of bias or certainty assessment and some historical and related cohorts may overlap, conclusions regarding strength of evidence should be kept cautious. No primary study” should be used as words but not as an indication that there is no clinical evidence.

 

7. Discuss mixed populations and cohort overlap. Describe the treatment of pancreatic and non-pancreatic primaries, MEN1-associated disease, mixed tumor grades, and overlapping institutional cohorts. Inclusion of a supplementary sensitivity analysis without studies that could not be attributed to the pancreas or have possibly been duplicated would support the interpretation.

Minor Revisions

There are some minor revisions to be made:

Correction of typographical and spacing errors, e.g., the missing space after reference [47]; standardization of terminology and abbreviations; definition of all abbreviations in tables and in the figure legends; clearly stating the denominator for all percentages. The following editorial fields (such as the Placeholder DOI and the Firstname Lastname fields) should be filled in with content before publication. Care must be taken to ensure that the references and supplementary files are consistent with the final data set.

Author Response

Thank you for your valuable feedback. Below is our detailed response to each comment.

Comment 1. Explain the classification of a syndrome. Ten ENETS syndromes are mentioned, but ACTH- and CRH-secreting tumors are discussed separately and no CRH category appears in the evidence map, which has ten rows.

Response1. Thank you for your comment. The reviewer has identified a real inconsistency in our terminology. ACTH-secreting and CRH-secreting tumors were treated throughout the analysis as a single category, because both present clinically as ectopic Cushing syndrome and the medical management recorded in the eligible studies is the same. The ten rows of the evidence map were therefore already correct and the denominator is unchanged, but the wording did not say so. The Introduction now states explicitly that tumors secreting adrenocorticotropic hormone or corticotropin-releasing hormone present as ectopic Cushing syndrome and are treated here as a single category, and the category label has been harmonized across the text, the abstract, the tables, the figures and the supplementary files. We note that one CRH-secreting pancreatic tumor report was identified and is listed in the contextual inventory in Supplementary File S3; no CRH-secreting report met the criteria for the primary corpus.

Comment 2. Arrange the study-selection numbers clearly and define record, report, study, cohort and patient. Explain the difference between 65 primary studies, 60 map-contributing studies and five mixed records.

Response2. A new paragraph in the Methods defines each term following PRISMA-ScR and reconciles every total in sequence: the parent search screened 5,757 records and retained 2,679 across all primary sites; the functioning pancreatic candidate set was drawn from that pool together with guideline-reference snowballing and a dedicated search, giving 77 reports at full text, of which 16 were excluded before charting and 61 were charted; the 2026 update contributed 13,378 records, 12,335 after within-update deduplication, from which 22 further reports were assessed, 16 entering the primary corpus and 6 the contextual tier; across both streams 83 reports were adjudicated, comprising 65 primary, 13 contextual and 5 excluded after charting; and the 321 contextual and secondary records are a separate inventory that is counted rather than charted. The paragraph also explains that 60 of the 65 primary studies contributed at least one full-text-confirmed assignment to the evidence map, while the remaining five are mixed-syndrome or mixed-therapy reports in which no individual syndrome-therapy pairing could be attributed from the text, so they are counted in the corpus but generate no map cell. The PRISMA flow diagram has been redrawn to display this reconciliation.

Comment 3. Improve methodological reproducibility: indicate whether extraction was duplicated, how disagreements were resolved, whether extraction was checked against the original reports, and describe full-text exclusions, language limitations, database coverage, deduplication and the absence of forward-citation searching.

Response3. The Methods now state the process in full. Data were extracted in duplicate and independently by two reviewers (A.P. and S.P.P.), each working from the full text of the source report; the two extractions were compared field by field, discrepancies were resolved by discussion between the two extractors, and any that remained were adjudicated by a third reviewer (I.Kat.). Title and abstract screening and full-text screening were likewise performed by three reviewers (S.P.P., C.Than. and A.P.) with disagreements resolved by a fourth (I.Kat.). Searches covered PubMed/MEDLINE, Embase and Cochrane CENTRAL from 1970 onward. Records were deduplicated in Rayyan by digital object identifier and PubMed identifier, with residual duplicates removed manually on title, first author and year. Reports in languages the review team could not assess were excluded and are listed with that reason. We also state plainly that forward-citation searching was not performed, and that the exact final search strings, duplicate counts and source-specific yields of the original parent search were not preserved; both limitations are declared in the Discussion and in Supplementary File S1. A complete list of the sixteen reports excluded after full-text assessment, each with its reason, has been added as Section S1.9 of Supplementary File S1.

Comment 4. Describe the outcome in more detail. The biochemical or physiological response in 58/65 studies appears more convincing than the evidence underlying it, because the endpoints were heterogeneous. Report outcomes by syndrome, specify the response categories, and do not treat hormone change, symptom improvement, electrolyte correction, glucose requirement and tumor response as equivalent.

Response4. We accept this entirely; it is the most substantive point raised and it has improved the paper. Outcomes are now classified rather than pooled. Of the 65 studies, 37 reported both a biochemical or physiological and a clinical outcome, 21 a biochemical or physiological outcome only, and 7 a clinical outcome only; the previously quoted 58/65 is the sum of the first two groups, and 44/65 reported a clinical outcome. The report-level distribution by syndrome is now given in the Results (syndrome assignments are non-exclusive, and multisyndrome reports may contribute to more than one row): Insulinoma 33 (17 both, 13 biochemical only, 3 clinical only); VIPoma 13 (8 both, 2 biochemical only, 3 clinical only); Gastrinoma/ZES 12 (6 both, 4 biochemical only, 2 clinical only); Glucagonoma 11 (8 both, 2 biochemical only, 1 clinical only); PTHrPoma 4 (4 both, 0 biochemical only, 0 clinical only); Ectopic ACTH 2 (1 both, 1 biochemical only, 0 clinical only); Ectopic GHRH 1 (0 both, 1 biochemical only, 0 clinical only); Somatostatinoma 1 (0 both, 0 biochemical only, 1 clinical only); Multiple/mixed 5 (4 both, 1 biochemical only, 0 clinical only). A further paragraph distinguishes, within the biochemical and physiological group, studies that measured the hormone causing the syndrome from those reporting a downstream consequence such as glycemia, serum potassium or serum calcium, and states that these are not interchangeable and that neither is equivalent to a radiological response. Radiological endpoints are recorded separately and are not counted as syndrome control. Study-level outcome detail, which was previously not populated, has been added for all 65 studies in Supplementary File S2.

Comment 5. Keep conclusions about treatment effectiveness tentative; “no primary study” should not be read as an indication that no clinical evidence exists.

Response5. Agreed. Every statement of this kind now reads as absence of eligible evidence under our criteria rather than evidence of absence, in the Abstract, the Results and the Discussion. The existing statement that this scoping review did not undertake risk-of-bias or certainty assessment and cannot rank comparative effectiveness has been retained and is now referred to at each point where a syndrome-therapy combination is reported as empty.

Comment 6. Discuss mixed populations and cohort overlap, and consider a supplementary sensitivity analysis excluding studies that could not be attributed to the pancreas or may be duplicated.

Response6. A sensitivity analysis has been added. Twelve of the 65 primary studies drew on cohorts that were not exclusively pancreatic, reported a pancreatic subgroup within a larger series, or came from centers and registries whose patients may overlap with another included study; each is listed with its reason in Supplementary File S3. Removing all twelve reduces the populated cells from 35 of 80 to 29 of 80. No combination that was empty in the main analysis becomes populated, and every cell that becomes empty had been supported by a single flagged study, so the conclusion that the evidence is unevenly distributed is unchanged. Patient numbers were not summed across reports, so no pooled patient denominator is presented. We now state explicitly that possible repeated representation may nevertheless affect report-level study counts and map cells. MEN1-associated disease and mixed tumor grades were charted where reported and were not used to restrict eligibility.

Minor points

Response. The missing space after reference [47] has been corrected, as has the hyphenation of “ten functioning pan-NET syndromes”. Abbreviations have been standardized and are defined in the text, the table footnotes and the figure legends. Every percentage now states its denominator. The editorial template fields identified by the reviewer are assigned by the journal at production. We removed the visible placeholder text (including ‘Firstname Lastname’ and the placeholder DOI) from the revised author file rather than inventing production metadata. The supplementary files have been regenerated from the frozen dataset so that they are consistent with the final figures; in doing so we identified and corrected an error in the previously submitted Supplementary File S3, in which twenty syndrome-therapy pairings that had been rejected during full-text adjudication were inadvertently retained, causing that file to show 41 populated cells against the 35 reported in the manuscript. The corrected file contains 138 confirmed PMID–syndrome–therapy assignments, reproduces 35 of 80 populated cells, and contains no unresolved assignment values.

Additional author correction. We corrected the compound surname in reference 10 from ‘van Djikum, E.N.’ to ‘Nieveen van Djikum, E.’ The detailed outcome paragraphs were also moved from the Abstract to Section 3.4, leaving a concise abstract-level summary.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Manuscript ID life-4626599

Type Review

Title: Medical Management of Secreting Syndromes in Functioning Pancreatic Neuroendocrine Tumors: A Scoping Review and Evidence Map

The authors have improved this manuscript considerably, and the revised manuscript explicitly includes ACTH- and CRH-producing tumors in the same group and updates the tables and figures. It talks about independent duplicate extraction, discrepancy resolution, and deduplication. It also describes the differences between the 65-study corpus and the 60 studies included in the evidence map. The manuscript now separates clinical and biochemical/physiological outcomes, without including radiological response in the syndrome control. Introduces a sensitivity analysis in which 12 potentially problematic studies are excluded; map coverage goes down to 29/80 cells from 35/80 cells. It presents tentative conclusions and distinguishes a lack of available studies from a lack of clinical experience.

However, two points must be addressed before acceptance:

1. Check the supplementary files; no files S1–S3 were provided
2. The analysis counts distinct PMIDs while acknowledging possible overlapping cohorts. These should not imply fully independent investigations; Figure 1's "Primary evidence map, n = 65" and Figure 5's heading should clearly distinguish the full corpus from the 60 map-contributing studies.

Author Response

Comment 1. Check the supplementary files; no files S1–S3 were provided.

Response. We apologize for the upload omission. Supplementary Files S1–S3 are now provided as three separate files: S1 contains the search, selection and full-text adjudication material; S2 contains the 65-report primary-study index; and S3 contains the evidence-map assignments, contextual inventory, reproducibility checks and sensitivity analysis. Their filenames and internal references have been checked against the revised manuscript.

Comment 2. The analysis counts distinct PMIDs while acknowledging possible overlapping cohorts. These should not imply fully independent investigations; Figure 1 and Figure 5 should distinguish the 65-report corpus from the 60 map-contributing studies.

Response. Thank you for your comment. Figure 1 now shows the 65-report primary corpus together with its split into 60 map-contributing reports and five non-grid mixed reports. Its legend states that the 65 reports are identified by distinct PMIDs, that PMIDs are report identifiers, and that they do not establish that all cohorts are independent. Figure 5 is now headed “60 map-contributing reports,” and its legend states that the counts are based on distinct PMIDs, that related publications may contain overlapping cohorts, and that the counts should therefore not be interpreted as fully independent investigations. Both legends have been revised and are highlighted in the manuscript.