Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This narrative review covers a clinically important and expanding area (paraneoplastic neurological syndromes, PNS) and appropriately highlights the impact of novel antibody discovery, the 2021 PNS-Care criteria, and the growing intersection with immune checkpoint inhibitor (ICI) associated neurotoxicity. The manuscript is generally well structured and clinically oriented. However, minor revisions are needed, primarily to correct multiple citation / claim mismatches and to strengthen the framing of evidence.

 

1. Please see lines 449–451, 'commercial line blots are prone to false positive ... ', while ref [62] is 'Advances in Antibody Phage Display Technology' (Drug Discov Today 2022), which does not support a claim about commercial line‑blot false positivities. I think the best option is using your own ref list, replace [62] -> [4,5], which you already cite for line‑blot pitfalls in the Intro.

 

2.  In lines 463–465, the problematic text is 'PhIP‑Seq was instrumental in the discovery of anti‑KLHL11 ...[64].' Your Ref [64] is anti‑Hu/anti‑Yo epitope mapping by programmable phage display, not KLHL11 discovery. The primary KLHL11 discovery paper is already in your reference list as Ref [58] (NEJM 2019). So please fix by citing [58] for KLHL11 discovery, and keep [64] only for the broader point that programmable phage display can map epitopes in onconeural autoimmunity.

3. In lines 465–470, Neurofilament light chain (NfL), references don’t match the PNS claim. Ref [65] describes hereditary transthyretin amyloidosis, not PNS. Ref [66] reports serum NfL in cancer patients with acute neurological complications, not a PNS‑specific validation, and it does not clearly support 'guide decisions on escalating immunotherap' in PNS as written. Please fix by either tone down the claim to 'general neuroaxonal injury marker studied in autoimmune encephalitis/cancer‑neuro complications' or add proper AE/PNS‑relevant NfL references and reframe as 'emerging/under evaluation' rather than established in PNS.

 

4. In '7.3. Emerging Therapies for Refractory Disease', multiple citation mismatches and/or overstated. 

In 503–507 and Table 1, Bortezomib: stiff‑person claim not supported by the cited ref. Problematic text is 'bortezomib shown efficacy in refractory anti‑NMDAR encephalitis and severe stiff‑person syndrome ... [72]'. Your Ref [72] is the anti‑NMDAR bortezomib report (Shin et al., Ann Clin Transl Neurol 2018), and it does not support stiff‑person syndrome unless explicitly shown. Can remove 'stiff‑person syndrome' here or add an SPS‑specific source.

Lines 507–510 and Table 1. Tocilizumab: ICI‑CRS/neurotoxicity claim can not be supported by the cited ref. Ref [73] is tocilizumab in autoimmune encephalitis refractory to rituximab (Lee et al., 2016) not ICI‑CRS. Please fix by either splitting into two claims with different citations, or tone down and cite ICI‑CRS literature.

Lines 510–513 and Table 1. FcRn inhibitors (Efgartigimod): citation is outdated and irrelevant. Ref [74] is a 2018 MuSK‑MG immunotherapy review (predates pivotal Efgartigimod phase 3 data and does not support the stated investigation claims in LGI1/IgLON5). Please cite pivotal Efgartigimod trials for MG and present LGI1/IgLON5 with specific studies/trials.

Lines 513–516 and Table 1. Daratumumab: wrong disease cited in the reference. Ref [75] is a POEMS syndrome case report, not NMDAR encephalitis. Please replace [75] with an actual daratumumab for refractory anti‑NMDAR encephalitis report, or rewrite to match POEMS.

Lines 519–522 and Table 1. CAR‑T claim 'including severe PNS': needs rewording. 
Problematic text is 'CAR‑T is being tested including severe PNS [77]' and Table 1 cites [80] for 'refractory antibody‑mediated PNS'. The cited evidence appears to be related to autoimmune encephalitis trial landscape [77] and autoimmune neuropathies [80], not clearly PNS. Please rephrase to avoid implying established PNS evidence. Can be described such as, 'CD19‑targeted CAR‑T approaches are being explored experimentally for severe, treatment‑refractory autoimmune neurologic diseases, while evidence in classical PNS remains minimal and should be considered investigational'.

Given the unusual frequency of citation / claim mismatches identified, kindly recommend that authors conduct a re-verification of all citations throughout the manuscript to ensure accuracy before resubmission, but no new analyses are required.

 

Author Response

Dear Reviewer,

Thank you for your insightful comments, which are addressed below:

Comment 1: We agree that references [4] and [5] are the appropriate sources for this statement. We have removed reference [62] and replaced it with references [4] and [5] in the specified sentence to correctly support the statement regarding the diagnostic pitfalls of commercial line blots (Line 378).

Comment 2: We appreciate the correction regarding the attribution of the KLHL11 discovery. We have revised the citation. We now cite reference [58] specifically for the discovery of anti-KLHL11 (Line 391). We have retained reference O’Donovan et al. (now [63]) only to support the general utility of PhIP-Seq for epitope mapping (Line 392).

Comment 3: We have restructured this section to ensure the citations strictly align with the reported data. We have retained reference [64] (Gragera-Martinez et al.) to support the general utility of NfL as a biomarker of disease onset and progression in neurological disorders. Regarding reference [65] (Gottiparthy et al.), we have revised the text to accurately reflect their findings in cancer patients with acute neurological complications. We have toned down the statement regarding "guiding immunotherapy" to instead emphasize its utility in "gauging clinical recovery," as supported by their case series data linking longitudinal trends to outcomes. Finally, we have incorporated a new reference [66] (Farina et al., 2025) to provide specific, high-quality evidence regarding the prognostic value of serum NfL in anti-IgLON5 disease (Lines 392-400).

Comment 4: We have revised sub-section 7.3 to ensure that all claims are precisely supported by the corresponding citations. We have updated the text and citations for each agent as follows:

Bortezomib: We have retained the mention of Stiff-Person Syndrome (SPS) but clarified that this is based on "single case reports" rather than broad trial evidence, adding the specific citation [74] (Lines 435-436).

Tocilizumab: We have removed the reference to ICI-associated cytokine release syndrome to maintain the focus on paraneoplastic mechanisms. We elaborated on its use in autoimmune encephalitis to highlight the findings of reference [75] (Lines 437-440).

FcRn Inhibitors (Efgartigimod): We updated the reference by replacing [74] Vakrakou et al. with the pivotal Phase 3 ADAPT trial for MG. We refined the text to explicitly include anti-CASPR2 disease alongside anti-LGI1 and anti-IgLON5 as potential candidates for this therapy based on a recent case report by Fabry et al. describing a patient positive for anti-CASPR2 antibodies in the context of Efgartigimod treatment, providing specific clinical precedent for discussing this antibody phenotype in relation to FcRn inhibition. We clarified that this potential is based on the shared IgG4 immunopathogenesis of these syndromes (Lines 441-445).

Daratumumab: We included a reference with a case report documenting successful use in refractory anti-NMDAR encephalitis while retaining Leung et al. (Ref. 78) to provide a wider breadth of evidence for refractory management (Line 447).

JAK Inhibitors: We replaced the previous citations (Cezar et al. and Yang et al.) with Picca et al. (2021) and Shin (2021) to provide more specific and mechanistically relevant evidence. Picca et al. were selected to substantiate the use of JAK inhibitors specifically for severe ICI-associated neurotoxicity, which aligns directly with the expanded scope of Table 1 (described below). Shin (2021) was incorporated to illustrate the immunological rationale (cytokine suppression in inflammatory epilepsy/NORSE) that drives the experimental use of these agents in neuroinflammation. Consequently, we have refined the manuscript text to explicitly characterize the application of JAK inhibitors in classic PNS as 'limited' and 'experimental,' ensuring the strength of the recommendation accurately reflects the current paucity of direct clinical data (Lines 448-451).

CAR-T Therapy: We rephrased the text to characterize CAR-T as experimental. We replaced the earlier paper [80] Motte et al. on CAR-T therapy with the more recent and clinically specific study by Wickel et al. (2025). The new reference describes a case of Stiff-Person Syndrome treated with CAR-T, which directly aligns with our updated text discussing the potential of CAR-T in high-burden, antibody-mediated phenotypes like SPS. This swap ensures the manuscript reflects the most current and phenotype-specific proof-of-concept evidence available (Lines 453-459).

Comment 5: We have taken this recommendation seriously and have conducted a complete re-verification of the bibliography to ensure that every citation in the revised manuscript accurately supports the preceding statement.

1) We have rephrased the statement regarding Cyclophosphamide use for T-cell mediated intracellular syndromes in 7.2 to align more precisely with the conclusions of Abboud et al. We replaced 'is often favored... due to its broader mechanism' with 'is typically considered for T-cell mediated intracellular syndromes... or refractory disease.' While Cyclophosphamide is mechanistically distinct from Rituximab, the cited consensus guideline positions it primarily as an alternative or escalation therapy for intracellular antigens/refractory cases, rather than establishing it as the universally 'favored' standard. This change maintains clinical accuracy while faithfully representing the cited evidence (Lines 426-428).

2) We revised the heading of Section 7.3 to 'Emerging and Experimental Therapies for Refractory Disease.' This adjustment was necessary to accurately categorize the developmental stage of the interventions discussed. While some agents are nearing standard off-label use, others are currently applied on a strictly experimental or compassionate basis in PNS. This distinction manages clinical expectations and aligns the title with the text's cautionary language regarding the current level of evidence (Line 429).

3) We have revised Table 2 (previously Table 1) to include updated citations and have modified the title to 'Emerging therapies in PNS and ICI-mediated neurotoxicity.' This change acknowledges that several agents discussed are increasingly relevant for ICI-associated neurological adverse events, as highlighted in the updated text. We ensured the citation numbers in the table correspond accurately to the revised reference list. We simplified the fourth column header from 'Potential Utility in PNS' to 'Potential Utility.' This change eliminates the inaccuracy of labeling conditions such as ICI-associated cytokine release syndrome as strictly 'PNS,' ensuring the column header aligns with the revised, broader table title (Line 469).

4) To ensure maximum accuracy and avoid citing a specific percentage range that may vary across small cohorts, we have amended the sentence in Section 7.4 to read: “Re-challenging with ICIs after a neurological adverse event is high-risk, with significant relapse rates, particularly in patients with pre-existing paraneoplastic antibodies.” (Line 467).

5) We have carefully curated the reference list to support the manuscript's claims with the most relevant and comprehensive evidence. We retained Leung et al. [78] (Line 447) and Abboud et al. [82] (Line 456) to provide a wider breadth of evidence for refractory management and the clinical trial landscape, respectively. Additionally, we incorporated a new reference regarding CSF liquid biopsy (Hickman et al., Text citation [68]) to strengthen the discussion and improve accuracy on diagnostic advances in Section 6.3 (Line 402).

6) In sub-section 3.2 we have updated the terminology in Section 3.2 from 'small-cell lung cancer (SCLC)' to 'lung cancer.' This adjustment ensures strict alignment with the data presented in the cited population-based study [2] Vogrig et al., which correlates the median age of PNS onset with the peak incidence of lung cancer generally, without restricting the observation exclusively to the small-cell subtype (Line 116).

7) In sub-section 3.3 we have added [18] Dubey et al. to the sentence discussing breast cancer associations. While the previous references covered anti-Yo and anti-Ri, Reference 18 specifically provides robust clinicopathologic evidence linking anti-amphiphysin antibodies to breast cancer. This addition ensures that every antibody phenotype listed in the statement is directly supported by specific primary data (Line 149).

8) We modified the phrase 'neural tissue... organized into germinal centers' to 'neural tissue... infiltrated by immune cells organizing into tertiary lymphoid structures with germinal centers.' This change more accurately reflects the histological findings reported in [25] Dabner et al. (Lines 150-153).

9) In sub-section 4.3 we have revised the immunogenetics section to ensure precise nomenclature and alignment with the latest GWAS data. We corrected the formatting of all HLA alleles (e.g., changing 'DRB113:01' to 'HLA-DRB1*13:01') to adhere to standard scientific nomenclature. We refined the description of the anti-Yo susceptibility haplotype to explicitly cite HLA-DRB1*13:01 and HLA-DQA1*01:03, consistent with the findings of Hillary et al. (Ref [37]). We clarified the statement regarding the PTPRD gene. While previously linked to synaptic adhesion, we now explicitly state it is a novel non-HLA risk locus for anti-LGI1 encephalitis as identified in recent large-scale GWAS (Ref [38]), ensuring the distinction between the HLA and non-HLA genetic contributions is clear (Lines 224-234).

10) In sub-section 4.4 we have refined the discussion on tumor genetics to accurately reflect the findings of Small et al. (2018). We corrected the statement that CDR2L mutations are 'frequent,' clarifying instead that while somatic mutations/amplifications can occur, they are found in a minority of anti-Yo positive tumors. We emphasized that the overexpression of the onconeural protein (CDR2L) and the inflammatory tumor microenvironment are the universal drivers described in the study, rather than genetic mutations being the sole mechanism. This nuance is critical for accurately describing the pathogenesis of paraneoplastic cerebellar degeneration (Lines 240-250).

11) We have updated Figure 1 and its corresponding legend to ensure strict alignment with the revised evidence presented in Section 4.4. The figure legend now explicitly references 'Ectopic Antigen Expression' and 'Antigen Overexpression' as the primary drivers of T-cell priming, rather than 'neoantigen formation.' This change reflects the incorporation of recent genomic data (Small et al., 2018; Vogrig et al., 2024) into the manuscript, ensuring that the visual illustrations accurately represent the current understanding of intratumoral tolerance breakdown. Additional revisions of the description density were done to comply with Reviewer 2’s comments (Lines 254-259).

12) In sub-section 6.1 we have refined the description of the validation data for the 2021 PNS-Care Score to ensure maximum statistical precision. We specified that the reported 93% sensitivity applies to the combined cutoff of 'Definite or Probable' (Score ≥6), rather than the 'Definite' category alone. This distinction is critical because, as noted in the cited study (Zhao-Fleming et al., 2021), the sensitivity of the 'Definite' category alone actually decreased compared to previous criteria. Highlighting the combined Score ≥6 metric accurately reflects the functional clinical utility of the new system (Lines 348-357).

13) 6.2 We removed the reference to 'anti-NfL antibodies' as a specific paraneoplastic entity, as current evidence primarily supports NfL as a biomarker of axonal injury rather than a distinct paraneoplastic antibody target. Anti-NfL antibodies appear in various conditions but are not still a defined paraneoplastic antibody (Line 373).

14) We have added a new reference to support the statement regarding liquid biopsy in plasma and CSF. The original reference [67] (Netti et al., 2025) primarily discusses liquid biopsy in the context of renal cell carcinoma and serum biomarkers. Since our statement explicitly highlights the utility of CSF analysis for detecting occult CNS malignancies ("...in plasma or CSF..."), we found it necessary to add a complementary citation that specifically validates this approach in the central nervous system. We have included Hickman et al. (2023) [68] to clearly support the claim regarding cerebrospinal fluid liquid biopsy, ensuring the citation profile matches the scope of the clinical recommendation (Line 402).

The reference list has been updated to 86 citations to accommodate the new evidence requested by the reviewers and to support the revised content.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presented a comprehensive narrative review of paraneoplastic neurological syndromes (PNS), with a focus on recent advances in epidemiology, immunopathogenesis, diagnostics, and management. The authors outlined the shift from classical/non-classical definitions to the 2021 PNS-Care risk-stratified framework, emphasizing distinctions between T-cell–mediated syndromes associated with intracellular antigens and antibody-mediated disorders targeting neuronal surface proteins. The review integrates emerging concepts such as molecular mimicry, tumor genetics, HLA susceptibility, ICI-associated neurotoxicity, and novel diagnostic tools including PhIP-Seq, neurofilament light chain, and liquid biopsy. Therapeutic strategies are discussed with emphasis on early tumor control, immunotherapy, and newer targeted approaches for refractory disease. However, in its current form, it is overly long, insufficiently critical in parts, and occasionally overconfident in emerging diagnostic and therapeutic advances. I recommend major revision focused on tightening the narrative, clearly distinguishing established evidence from evolving hypotheses, tempering therapeutic optimism, and improving conceptual clarity, particularly regarding ICI-associated syndromes and prognostic limitations.  The following are the major concerns:

1. Although the review is extensive and factually rich, it is overly long and densely written, with excessive detail that often obscures key messages. The manuscript would benefit from substantial condensation, particularly in the epidemiology and clinical phenotype sections, where similar concepts are reiterated multiple times with marginal added value.
2. The search strategy is described in detail, but the review remains narrative rather than systematic. Despite reporting numbers of retrieved and included studies, there is no clear critical appraisal of evidence quality, study design heterogeneity, or risk of bias. This creates an impression of methodological rigor that is not fully supported by the synthesis approach.
3. The manuscript frequently blurs the line between well-established evidence and emerging or speculative concepts. For example, molecular mimicry, tumor neoantigens, and HLA associations are presented as near-universal mechanisms, whereas in reality these remain incompletely validated and apply strongly only to selected antibody–tumor pairs.
4. The discussion of immune checkpoint inhibitor–associated syndromes is valuable but conceptually underdeveloped. The manuscript repeatedly states that ICI-related neurotoxicity overlaps with classic PNS, yet does not sufficiently delineate where immunopathogenesis clearly diverges, risking conceptual conflation rather than clarification.
5. The diagnostic section strongly endorses the 2021 PNS-Care criteria, but limitations and real-world implementation challenges (e.g., antibody-negative cases, limited access to advanced testing, inter-laboratory variability) are underemphasized relative to their clinical importance.
6. The review implicitly assumes that earlier diagnosis and immunotherapy will necessarily improve neurological outcomes, which is not always the case, particularly for intracellular antigen–associated syndromes where irreversible neuronal loss dominates. Prognostic pessimism for these entities should be more explicitly acknowledged.
7. The management section is therapeutically ambitious, presenting multiple emerging agents (bortezomib, tocilizumab, FcRn inhibitors, CAR-T cells) in a manner that may overstate their readiness for clinical use. Evidence for many of these approaches is limited to case series or anecdotal reports, yet this limitation is not consistently emphasized.
8. There is limited discussion of adverse effects, cost, accessibility, and risk–benefit considerations of aggressive immunotherapies, especially relevant for an often elderly oncology population.
9. Figures are visually appealing but function more as didactic illustrations than critical syntheses. Some schematics (e.g., molecular mimicry pathways) risk oversimplifying complex immunological processes and may unintentionally imply mechanistic certainty where none exists.
10. The manuscript lacks a clear prioritization of future research directions. While many gaps are mentioned, they are listed rather than ranked, leaving readers without a sense of which unresolved questions are most pressing or tractable.

Minor concerns

• Several sections contain redundant background information that could be shortened without loss of clarity.
• Figure legends are overly long and sometimes repeat information already stated in the main text.
• Some epidemiological figures are cited repeatedly; a concise summary table could improve readability.
• Abbreviations are occasionally redefined unnecessarily or introduced late.
• Language is generally strong but occasionally verbose, with long sentences that reduce clarity.
• The conclusion largely reiterates earlier sections and could be sharpened to deliver a more focused take-home message.

Author Response

Dear Reviewer,

Thank you for your constructive advice. Our response to your comments is written below:

Comment 1:

We have addressed the concern that the manuscript was 'overly long and densely written' by significantly condensing Sections 3 and 5.

Section 3. We have executed a broad revision to condense the text and improve the clarity of the key messages without losing scientific accuracy. As suggested in Comment 13, we have removed the repetitive citation of raw incidence figures in the text and consolidated this data into a new Table 1 ("Population-Based Epidemiology of PNS and Antibody-Mediated Encephalitis"). This allows for a direct, at-a-glance comparison of the pivotal Italian, US, and French cohorts, improving readability. Consequently, the previous Table 1 ("Emerging Therapies...") has been renumbered to Table 2 throughout the manuscript (Lines 98-109).

Section 5. We significantly condensed the descriptions of individual syndromes. Instead of exhaustive symptom lists, we focused strictly on the defining semiological features and specific cancer associations that are critical for clinical recognition. We eliminated repetitive introductory transitions and overlapping definitions, particularly in the comparison between high-risk and intermediate-risk phenotypes. By sharpening the language - especially regarding complex presentations like 'Triple M' syndrome and the distinction between paraneoplastic vs. idiopathic disorders - we have improved the manuscript's flow. These changes ensure that the key clinical messages stand out clearly without being obscured by excessive detail (Lines 275-310).

Section 6.1 We streamlined the discussion of the 2021 PNS-Care Score validation data. By focusing strictly on the composite sensitivity metric (Score ≥6) and removing speculative interpretations regarding the 'Probable' category, we reduced the text density while enhancing the statistical clarity of the section (Lines 348-357).

Comment 2:

We have substantially revised and condensed the Methods section to align with the critique regarding methodological transparency. We explicitly defined the manuscript as a narrative review and condensed the previous subsections into a single text. While we retained the specific Boolean search strings to ensure the reproducibility of our search results (1,942 records), we removed the granular study flow metrics that may have implied a systematic review or meta-analysis. We added a specific statement clarifying that a formal risk of bias assessment was not performed. We justified this omission by citing the inherent nature of the PNS literature, which relies heavily on retrospective cohorts and case series rather than randomized controlled trials, making standard quality grading less applicable (Lines 78-95).

Comment 3:

We have substantially revised Section 4.4 to address this and ensure mechanistic accuracy. We have expanded the opening definition to explicitly include 'ectopic antigen expression' alongside molecular mimicry. This acknowledges that in most PNS (unlike post-infectious autoimmunity), the trigger is the aberrant expression of the bona fide neuronal antigen by the tumor, rather than cross-reactivity with a foreign pathogen. We corrected the discussion on tumor genetics to align with the findings of Small et al. We clarified that while somatic mutations in antigen genes (e.g., CDR2L) occur, they are rare. Instead, we now emphasize that the intense overexpression of onconeural proteins within a pro-inflammatory tumor microenvironment is the universal driver of tolerance breakdown. We retained the concept of 'neoantigens' but contextualized it within the scope of peptide presentation in an inflammatory context (Ref. 32), rather than attributing it solely to somatic mutations. These changes provide a more nuanced and scientifically accurate description of the immunopathogenesis (Lines 240-250).

Comment 4 regarding the immunopathogenesis of ICI-associated syndromes

We have revised Section 5.3 to clearly delineate the mechanistic divergence between ICI-neurotoxicity and classic PNS, addressing the reviewer's concern regarding conceptual clarity. We added explicit language stating that while classic PNS is driven by antigen-specific cross-reactivity, many n-irAEs result from a generalized 'non-specific T-cell disinhibition,' distinct from the tumor-driven antibody response (Refs. 3, 54). We refined the discussion of the 'unmasking' phenomenon to present it as a specific intersection where these distinct mechanisms converge, ensuring the text accurately reflects the unique immunobiology of checkpoint inhibitor toxicity (Lines 312-323).

Comments 5-8:

We have significantly expanded Section 9 (Strengths and Limitations) to provide a comprehensive critical appraisal that addresses mentioned concerns. We explicitly added the challenges of inter-laboratory variability and false positives in commercial line blots. We inserted a specific acknowledgment that for intracellular syndromes, early diagnosis often fails to reverse established neuronal loss, directly addressing the concern regarding the limits of immunotherapy. We qualified the discussion of emerging agents by emphasizing that current evidence is largely restricted to case series and that cost/toxicity risks require careful consideration in elderly oncology cohorts (Lines 502-516).

Comments 3, 9 and 12:

We sincerely appreciate the detailed feedback regarding the mechanistic precision and presentation of our figures. We have made several revisions to Figure 1 and Section 4.4 to address these points. We agree that 'molecular mimicry' implies a pathogen-host cross-reactivity that does not fully capture the tumor-driven nature of PNS. We have updated the Figure 1 title to 'Ectopic Antigen Expression and Tolerance Loss' and revised the legend to emphasize that the tumor expresses the bona fide self-antigen (e.g., CDR2L overexpression) rather than a mimic. This aligns the figure with our revised text in Section 4.4 and the genomic data from Small et al. (2018). To address the concern that the schematics might imply 'mechanistic certainty where none exists,' we have revised the figure legend to explicitly frame the diagram as a 'schematic overview' of 'proposed models.' We incorporated hedging language (e.g., 'hypothesized to facilitate') to clarify that this is a conceptual framework rather than a definitive pathway. In response to the minor comment regarding legend length, we have significantly condensed the description. We removed the extensive explanatory text that is now covered in the main manuscript, focusing the legend strictly on the sequential steps illustrated in the panels (Lines 254-259).

Comment 10:

We have completely restructured Section 8 to explicitly prioritize the field's challenges. We identified reducing diagnostic latency as the single most pressing priority, citing the irreversibility of neuronal loss as the driver for this ranking. We highlighted 'basket trial' designs as the primary solution for the 'tractability' problem in rare disease research, explaining how mechanistic grouping (surface vs. intracellular) solves the statistical power issue. We clearly delineated survivorship and health economics as the subsequent, yet critical, frontier for research once acute mortality is addressed. This hierarchical approach provides readers with a clear roadmap of where the field needs to move next (Lines 471-489).

Comment 11:

We have systematically streamlined the manuscript to remove redundant background information and improve readability, addressing the concern regarding text density while also addressing additional Comments (1 and 2). We merged the previous subsections (2.1–2.4) into a single, concise narrative and removed granular study flow metrics that did not add value to a narrative review. We removed the repetitive citation of raw incidence figures in the text of Section 3, moving this data to the new Table 1. The text now focuses solely on the interpretation of trends rather than listing statistics. We eliminated repetitive introductory transitions and overlapping definitions from Section 5, particularly in the comparison between high-risk and intermediate-risk phenotypes, to focus strictly on defining semiological features. We streamlined the discussion of the PNS-Care Score validation data in Section 6.1, removing speculative interpretations to enhance statistical clarity.

Comments 12:

In response to the feedback regarding the length and redundancy of figure legends, we have significantly condensed the caption for Figure 1 (described earlier), Figures 2 and 3. We removed the exhaustive textual list of every specific syndrome from the legend, as these are already visually presented in Figure 2 and detailed in the main text. The revised legend now focuses on the clinical implication of the stratification (i.e., the necessity of antibody testing for intermediate-risk groups vs. immediate screening for high-risk groups), ensuring the legend functions as a functional guide rather than a repetitive list. We have condensed the description of Figure 3. We removed the redundant enumeration of specific novel antibodies that are already detailed in the main text. The revised legend now focuses strictly on the functional integration of the new methodologies (PhIP-Seq, NfL, Liquid Biopsy) into the diagnostic workflow. We also revised the title to 'Integrated Diagnostic Workflow for PNS' to more accurately reflect that the current framework incorporates clinical phenotype and oncological screening alongside biomarkers, rather than relying on biomarkers alone.

Comment 13:
Described above in response to Comment 1.

Comment 14:
We have carefully reviewed the manuscript to ensure consistent usage of abbreviations throughout the text. Specifically, we removed the unnecessary re-definitions of ICIs and n-irAEs in Section 5.3 (Line 312). We ensured SCLC is defined at its first occurrence in Section 3.3 (Line 134), as previous mentions were removed during the revision process. We confirmed that cerebrospinal fluid (CSF) is defined in Section 6.3 (Line 403), as this represents its first appearance in the main body text (excluding Figure 3).

Comment 15:

We have revised the manuscript to enhance clarity and precision. In addition to the substantial condensation of the text detailed in our responses to Comments 1 and 11, we specifically tightened sentence structures in the Pathophysiology (Section 4.1 – Lines 183-185), Diagnostics (Section 6.1 – Lines 335-336), and Management (Section 7 – Lines 406-407) sections to eliminate verbosity.

Comment 16:

We have rewritten the conclusion to address the feedback regarding focus and clarity. We sharpened the text to deliver three core conceptual messages without reiterating previous sections. We emphasized that for intracellular syndromes, the primary barrier to improved outcomes is the latency of diagnosis relative to irreversible neuronal loss, highlighting the need for earlier screening. We argued that treatment selection and future trial design (specifically 'basket trials') should be grounded in the mechanistic distinction between antibody-mediated and T-cell mediated pathogenesis. We positioned the differentiation between PNS and ICI-toxicity as the defining challenge for the next era of neuro-oncology (Lines 518-536).

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

the authors have adequately addressed raised concerns and the manuscript has been improved to be accepted