Phenotype-Oriented Characterization of NSC828786 Identifies Convergent HPN-AMACR-Associated Transcriptomic Signatures in Prostate Adenocarcinoma and Broad-Spectrum Antiproliferative Activity

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents an integrative analysis of NSC828786 using NCI-60 pharmacological profiling, public transcriptomic datasets, network analysis, molecular docking, computational ADMET prediction, and zebrafish embryo testing. The supplementary materials improve the reporting of several methods, including GEO dataset processing and zebrafish exposure conditions. However, important concerns remain regarding the biological interpretation, methodological rigor, and internal consistency of the computational analyses. Substantial analytical correction, methodological clarification, and experimental validation or major reframing are required.

 

Major comments

1. HPN and AMACR were identified by intersecting tumor-versus-normal differential-expression lists, whereas NSC828786 activity was evaluated independently using NCI-60 pharmacological data. Docking the compound into predicted regions of these proteins does not establish that HPN or AMACR contributes to compound sensitivity.The Supplementary Methods state that the NCI COMPARE gene-expression module was used to examine associations between baseline gene expression and NSC828786 response, but the corresponding results are not presented. The authors should report the complete COMPARE gene-expression output and specifically quantify the correlations between NSC828786 sensitivity and HPN or AMACR expression across the full NCI-60 panel. Effect sizes, p values, multiple-testing correction, probe identifiers, and sensitivity metrics should be provided. More importantly, direct functional validation is needed. Appropriate studies could include HPN or AMACR knockdown, knockout, overexpression, rescue experiments, enzymatic assays, or direct target-engagement measurements such as CETSA, DARTS, thermal proteome profiling, or biophysical binding assays. In the absence of such evidence, the title, highlights, abstract, and conclusions should describe HPN and AMACR as exploratory candidate associations rather than molecular programs linked to NSC828786 responsiveness.

 

2. The GEO analyses compare prostate or breast tumors with normal tissues. These comparisons can identify tumor-associated genes but do not establish associations with castration resistance, AR independence, treatment exposure, lineage plasticity, or TNBC. TCGA-PRAD primarily represents localized primary prostate adenocarcinoma and should not be treated as a surrogate for AR-independent or metastatic CRPC. Similarly, the breast datasets contain broadly classified breast cancers rather than clearly defined TNBC cohorts. The authors should provide receptor status, treatment status, disease stage, and other available clinical characteristics for every dataset. Analyses should preferably be repeated using datasets that directly compare hormone-sensitive versus castration-resistant disease, AR-high versus AR-low/negative disease, or TNBC versus receptor-positive breast cancer. Otherwise, terminology such as “hormone-independent,” “therapy-resistant,” and “adaptive resistance program” must be substantially moderated.
3. The NCI-60 prostate panel contains only PC-3 and DU-145 cells, both of which are AR-negative. There is therefore no AR-positive prostate comparator within this dataset. In addition, several breast cell lines classified as relatively sensitive are not TNBC; for example, T-47D and MCF-7 are hormone-receptor-positive models. The reported GI50 values are broadly similar across the prostate and breast lines, with some breast models appearing more sensitive than the prostate models. The authors should avoid claiming subtype selectivity unless they perform a quantitative comparison across all NCI-60 lineages and receptor-defined subgroups. Independent concentration-response studies should include AR-positive prostate models such as LNCaP, VCaP, or 22Rv1; additional AR-negative models; receptor-defined breast models; and nonmalignant epithelial controls. Comparison with niclosamide under identical experimental conditions would help determine whether NSC828786 offers improved potency or selectivity. Figure 1C and 1D are described as comparative GI50 values, but their axes show cell-growth percentage and appear to represent single-dose screening results. The caption and Results text should be corrected.

 

4. Supplementary Table S4 requires extensive verification and correction.

First, the molecular formula for NSC828786 is listed as C19H11F4NO5, whereas the main manuscript gives C19H11F4NO2. The reported molecular weight of 361.29 g/mol is consistent with the latter formula, not the former. The chemical identity and all calculated descriptors should therefore be rechecked using the correct structure and SMILES.

Second, several comparator molecular weights are displayed incorrectly, including values of 7.43, 1.5, and 3.9 g/mol for apalutamide, tamoxifen, and paclitaxel. Compound identifiers also appear duplicated or potentially incorrect. All compound identities, identifiers, structures, and outputs should be audited.

Third, the interpretation is more favorable than the reported predictions support. NSC828786 has a consensus logP of 5.20 and predicted logS values of −6.08 and −7.93, indicating very high lipophilicity and extremely poor predicted aqueous solubility. These are meaningful development liabilities and should not be summarized merely as “moderate lipophilicity” or as unlikely to limit pharmacology.

Fourth, the toxicity table reports scores of 0.88 for hERG blockade and 0.89 for hepatotoxicity. If these values represent probabilities of positive classification, they indicate potentially important safety alerts and directly conflict with the statement that no overt high-risk liabilities were identified. The authors must identify the platform and model used for each endpoint, define score directionality and classification thresholds, and interpret the results accordingly.

Finally, log Kp from SwissADME represents predicted skin permeation, not blood–brain barrier permeability. It should not be used to infer limited CNS exposure. BBB predictions should be reported separately using the relevant BBB model, with model confidence and applicability-domain information.

The developability conclusions should be rewritten conservatively after correcting these issues.

5. The supplementary information now reports embryo numbers, replicate structure, daily medium renewal, temperature, and DMSO concentration. However, the authors should clarify whether biological replicates represent independent clutches, identify the experimental unit, report exact group sizes, and state whether assessments were blinded. Because NSC828786 is predicted to have low solubility, precipitation and actual exposure concentrations should also be addressed. The statistical statement remains inadequate. One-way ANOVA is not generally appropriate for longitudinal survival and hatching proportions. Survival, hatching, and continuous morphometric endpoints should each have a prespecified analysis. Generalized mixed models, survival methods, or other analyses accounting for repeated observations and clustering by well or clutch may be appropriate. The exact post hoc test, assumption checks, multiplicity correction, effect sizes, confidence intervals, and exact p values should be reported. The manuscript states that exposure began at 4 hpf, whereas the Figure 4 timeline begins at 0 hpf. This inconsistency must be corrected. Figure 4B should also include scale bars.
6. HPN and AMACR are already established prostate cancer-associated genes. The principal novelty would therefore depend on demonstrating a functional relationship with NSC828786, which is not currently established. The manuscript should be reframed as an exploratory, hypothesis-generating study unless direct mechanistic validation is added.

Minor comments

1. Reconcile the R versions reported in the supplementary information: version 4.3.2 is stated for GEO processing, whereas version 4.2.2 is stated in the statistical section.
2. Explain the multiple colored RMSF traces in Figure 3 and state exactly what each trace represents.
3. Use consistent terminology for “growth percentage,” “growth inhibition,” GI50, IC50, TGI, and LC50.
4. Replace general statements such as “appropriate normalization,” “default parameters,” “standard workflows,” and “post hoc tests” with exact procedures.
5. Correct editorial errors including duplicated words in the correspondence line, duplicated periods in section headings, inconsistent manuscript year information, and punctuation and spacing errors throughout.

Author Response

Response to the Reviewers

Manuscript ID: cells-4373891

Title: Phenotype-Oriented Characterization of NSC828786 Identifies Convergent HPN–AMACR-Associated Transcriptomic Signatures and Broad-Spectrum Antiproliferative Activity across Cancer Models

 

Dear Editor and Reviewers,

We sincerely thank the Editor and both reviewers for their careful evaluation of our manuscript and for their thoughtful and constructive comments. We greatly appreciate the time and effort devoted to reviewing our work. The reviewers' suggestions have substantially improved the scientific rigor, clarity, and overall presentation of the manuscript. In response, we have carefully addressed every comment and have comprehensively revised the manuscript.

The revision extends well beyond textual editing and includes substantial analytical and conceptual improvements. The principal changes include:

• completion of the requested CellMiner/NCI-60 COMPARE gene-expression analysis;
• comprehensive revision of the Title, Highlights, Graphical Abstract, Abstract, Introduction, Results, Discussion, and Conclusions to consistently present the study as phenotype-oriented and hypothesis-generating;
• moderation of biological interpretations to avoid unsupported mechanistic or subtype-specific claims;
• expanded annotation of all GEO datasets together with clarification of the limitations of tumor-versus-normal transcriptomic comparisons;
• re-analysis and reinterpretation of the NCI-60 pharmacological data, including correction of Figure 1 and removal of claims regarding AR-negative or TNBC selectivity;
• comprehensive verification and correction of the computational developability analyses and Supplementary Information;
• revision of the zebrafish experimental methodology, statistical analyses, and reporting;
• extensive editorial revision throughout the manuscript to improve consistency, reproducibility, and scientific clarity.

For ease of review, all revisions have been highlighted using Track Changes in the revised manuscript. Below we provide a detailed point-by-point response to every reviewer comment. Reviewer comments are reproduced in black, and our responses are provided immediately below each comment.

Reviewer 1

The manuscript presents an integrative analysis of NSC828786 using NCI-60 pharmacological profiling, public transcriptomic datasets, network analysis, molecular docking, computational ADMET prediction, and zebrafish embryo testing. The supplementary materials improve the reporting of several methods, including GEO dataset processing and zebrafish exposure conditions. However, important concerns remain regarding the biological interpretation, methodological rigor, and internal consistency of the computational analyses. Substantial analytical correction, methodological clarification, and experimental validation or major reframing are required.

Response:

 Major comments

1     HPN and AMACR were identified by intersecting tumor-versus-normal differential-expression lists, whereas NSC828786 activity was evaluated independently using NCI-60 pharmacological data. Docking the compound into predicted regions of these proteins does not establish that HPN or AMACR contributes to compound sensitivity. The Supplementary Methods state that the NCI COMPARE gene-expression module was used to examine associations between baseline gene expression and NSC828786 response, but the corresponding results are not presented. The authors should report the complete COMPARE gene-expression output and specifically quantify the correlations between NSC828786 sensitivity and HPN or AMACR expression across the full NCI-60 panel. Effect sizes, p values, multiple-testing correction, probe identifiers, and sensitivity metrics should be provided. More importantly, direct functional validation is needed. Appropriate studies could include HPN or AMACR knockdown, knockout, overexpression, rescue experiments, enzymatic assays, or direct target-engagement measurements such as CETSA, DARTS, thermal proteome profiling, or biophysical binding assays. In the absence of such evidence, the title, highlights, abstract, and conclusions should describe HPN and AMACR as exploratory candidate associations rather than molecular programs linked to NSC828786 responsiveness.

Response:

We sincerely thank the reviewer for this fundamental and insightful comment. We fully agree that the original manuscript overstated the relationship between HPN/AMACR and NSC828786, and we have substantially revised the manuscript to address this concern. Rather than presenting HPN and AMACR as validated molecular determinants of NSC828786 activity, the revised manuscript now consistently frames these genes as exploratory candidate molecular associations identified through independent transcriptomic and computational analyses.

(1) COMPARE gene-expression analysis.

As requested, we performed an NCI-60 COMPARE gene-expression analysis using the CellMiner platform by correlating the GI50 activity profile of NSC828786 (NSC 828786) with baseline microarray expression of AMACR and HPN across the NCI-60 panel. Neither gene showed a statistically significant association with NSC828786 sensitivity (AMACR: r = 0.096, p = 0.477; HPN: r = 0.112, p = 0.405; Benjamini–Hochberg FDR-adjusted p = 0.477 for both genes). The complete COMPARE output, including gene identifiers, correlation coefficients, p values, and FDR-adjusted p values, has been added and is summarized in the revised Results section.

(2) Revision of the manuscript interpretation.

In light of these findings, we agree that neither differential-expression analysis, COMPARE correlation analysis, nor molecular docking establishes a functional or causal relationship between HPN/AMACR and the antiproliferative activity of NSC828786. Accordingly, we have comprehensively revised the Title, Highlights, Graphical Abstract, Abstract, Results, Discussion, and Conclusions to moderate our interpretation. Throughout the revised manuscript, HPN and AMACR are consistently described as exploratory candidate molecular associations identified through cross-cohort transcriptomic integration and structure-based computational analyses, rather than molecular programs linked to or mediating NSC828786 responsiveness. The revised title has therefore been changed to: "Phenotype-Oriented Characterization of NSC828786 Identifies Convergent HPN–AMACR-Associated Transcriptomic Signatures and Preferential Antiproliferative Activity in Prostate Cancer Models"

Title: A Niclosamide-Like Salicylanilide Derivative Associated with HPN–AMACR-Linked Metabolic–Proteolytic Programs in Hormone-Independent Cancer Models

• → Phenotype-Oriented Characterization of NSC828786 Identifies Convergent HPN–AMACR-Associated Transcriptomic Signatures and Broad-Spectrum Antiproliferative Activity across Cancer Models

(3) Functional validation and study limitations.

We fully acknowledge that direct functional validation—including HPN or AMACR knockdown/knockout, overexpression or rescue experiments, enzymatic inhibition assays, and direct target-engagement approaches such as CETSA, DARTS, thermal proteome profiling, or biophysical binding assays—was not performed in the present study. Moreover, because baseline expression of neither AMACR nor HPN correlated with NSC828786 sensitivity in the COMPARE analysis, the current data do not support a causal or mechanistic relationship between these genes and compound responsiveness. We now explicitly state this limitation in the revised Discussion and identify direct target validation as the principal direction for future mechanistic investigation.

Highlights

What are the main findings?

• NSC828786, a niclosamide-like salicylanilide derivative, showed preferential antiproliferative activity in hormone-independent prostate and breast cancer models.
• →NSC828786, a niclosamide-like salicylanilide derivative, showed preferential antiproliferative activity in AR-independent prostate cancer and hormone-independent breast cancer models.
• Cross-cohort analyses identified HPN-AMACR-associated metabolic and proteolytic programs linked to NSC828786 responsiveness.
• →Cross-cohort transcriptomic analyses identified convergent HPN–AMACR-associated metabolic and proteolytic programs in hormone-independent cancer models.

What are the implications of the main findings?

• Phenotype-oriented transcriptomic integration may help prioritize adaptive tumor vulnerabilities beyond single-target drug discovery.
• →Phenotype-oriented integration of pharmacological and transcriptomic analyses may facilitate the identification of adaptive molecular contexts beyond conventional single-target drug discovery.
• NSC828786 represents a hypothesis-generating salicylanilide lead requiring further target engagement and functional validation.
• →NSC828786 represents a hypothesis-generating salicylanilide lead compound that warrants further target-engagement and functional validation.

2     The GEO analyses compare prostate or breast tumors with normal tissues. These comparisons can identify tumor-associated genes but do not establish associations with castration resistance, AR independence, treatment exposure, lineage plasticity, or TNBC. TCGA-PRAD primarily represents localized primary prostate adenocarcinoma and should not be treated as a surrogate for AR-independent or metastatic CRPC. Similarly, the breast datasets contain broadly classified breast cancers rather than clearly defined TNBC cohorts. The authors should provide receptor status, treatment status, disease stage, and other available clinical characteristics for every dataset. Analyses should preferably be repeated using datasets that directly compare hormone-sensitive versus castration-resistant disease, AR-high versus AR-low/negative disease, or TNBC versus receptor-positive breast cancer. Otherwise, terminology such as “hormone-independent,” “therapy-resistant,” and “adaptive resistance program” must be substantially moderated.

Response:

We sincerely thank the reviewer for this insightful comment and appreciate the opportunity to clarify the scope and interpretation of our transcriptomic analyses. We agree that tumor-versus-normal differential expression analyses identify tumor-associated molecular features but, by themselves, cannot establish associations with castration resistance, androgen receptor (AR) independence, treatment exposure, lineage plasticity, or triple-negative breast cancer (TNBC) biology. The primary objective of the present study was to identify genes that were reproducibly dysregulated across independent prostate adenocarcinoma cohorts and subsequently evaluate their biological context through pathway enrichment, network analysis, and structure-based computational analyses. Accordingly, our transcriptomic analyses were intended as an exploratory, phenotype-oriented, and hypothesis-generating framework, rather than a comparison of clinically defined resistant disease states.

To address the reviewer's concerns, we have substantially revised the manuscript in several ways.

(1) Clinical annotation of transcriptomic datasets.

We have expanded Supplementary Table S1 to summarize all available clinical and pathological information for each GEO dataset, including pathological stage, histological grade, receptor status (ER/PR/HER2), TMPRSS2 fusion status, treatment information, and other available metadata whenever reported in the original datasets. Annotation availability varied considerably across studies. For example, pathological stage and histological grade were available for GSE69223, TMPRSS2 fusion status for GSE55945, and receptor status (ER/PR/HER2) together with histological grade for GSE29044, with partial receptor annotation available for GSE42568. Variables not reported in the original datasets are now explicitly designated as "NR (not reported)" rather than being inferred.

(2) Moderation of biological interpretation.

We have systematically moderated the terminology throughout the manuscript to avoid overinterpretation. Expressions such as "hormone-independent cancer programs," "therapy-resistant adaptive programs," and related mechanistic statements have been replaced with more conservative descriptions, including "tumor-versus-normal transcriptomic signatures in prostate adenocarcinoma cohorts," "candidate metabolic-proteolytic associations," and "exploratory molecular contexts." Corresponding revisions have been made throughout the Title, Highlights, Abstract, Introduction, Results, Discussion, and Conclusions.

(3) Interpretation of AR-independent disease.

References to AR-independent disease are now restricted exclusively to the experimentally established AR-negative prostate cancer cell lines (PC-3 and DU145) included in the NCI-60 pharmacological analyses. We no longer extrapolate AR-independent or hormone-independent characteristics to the transcriptomic cohorts analyzed.

(4) Dataset scope and future validation.

We agree that transcriptomic datasets directly comparing hormone-sensitive versus castration-resistant prostate cancer, AR-high versus AR-low/negative disease, or receptor-defined TNBC versus receptor-positive breast cancer would provide substantially stronger biological evidence. However, appropriately annotated datasets with sufficient sample sizes were not available within the public cohorts included in the present cross-cohort analysis. Therefore, we did not repeat the analysis using incompletely annotated datasets that would not adequately address the reviewer's concern. Instead, we now explicitly acknowledge this limitation in the revised Discussion and Limitations section and state that future validation using receptor-defined and resistance-stratified clinical cohorts will be required to determine whether the identified HPN–AMACR-associated transcriptomic signatures are specifically enriched in therapy-resistant disease.

Collectively, these revisions clarify that our transcriptomic findings should be interpreted as exploratory molecular associations identified in independent prostate adenocarcinoma cohorts, rather than evidence of CRPC-, AR-independent-, therapy-resistant-, lineage plasticity-, or TNBC-specific molecular programs.

3     The NCI-60 prostate panel contains only PC-3 and DU-145 cells, both of which are AR-negative. There is therefore no AR-positive prostate comparator within this dataset. In addition, several breast cell lines classified as relatively sensitive are not TNBC; for example, T-47D and MCF-7 are hormone-receptor-positive models. The reported GI50 values are broadly similar across the prostate and breast lines, with some breast models appearing more sensitive than the prostate models. The authors should avoid claiming subtype selectivity unless they perform a quantitative comparison across all NCI-60 lineages and receptor-defined subgroups. Independent concentration-response studies should include AR-positive prostate models such as LNCaP, VCaP, or 22Rv1; additional AR-negative models; receptor-defined breast models; and nonmalignant epithelial controls. Comparison with niclosamide under identical experimental conditions would help determine whether NSC828786 offers improved potency or selectivity. Figure 1C and 1D are described as comparative GI50 values, but their axes show cell-growth percentage and appear to represent single-dose screening results. The caption and Results text should be corrected.

Response:

We sincerely thank the reviewer for this thoughtful and constructive comment. We agree that the original presentation overstated subtype-specific interpretations based on the available NCI-60 dataset, and we have substantially revised the manuscript accordingly.

(1) Revision of Figure 1 and pharmacological data presentation.

We thank the reviewer for identifying the inconsistency in the original description of Figure 1C and 1D. These panels represent the single-dose (10 μM) percent growth obtained from the NCI-60 primary screening rather than comparative GI50 values. Accordingly, the figure legend, axis labels, and corresponding Results text have been corrected to accurately describe the data presented. Quantitative GI₅₀, TGI, LC₅₀, and IC₅₀ values are now presented separately in Figure 1.

(2) Interpretation of prostate and breast cancer models.

We agree that the NCI-60 prostate cancer subpanel contains only the AR-negative cell lines PC-3 and DU145, and therefore does not permit direct comparison between AR-positive and AR-negative prostate cancer models. Likewise, we acknowledge that several breast cancer cell lines included in the NCI-60 panel (e.g., MCF-7 and T-47D) are hormone receptor-positive rather than triple-negative breast cancer (TNBC) models. Accordingly, we have revised the manuscript to distinguish receptor-defined breast cancer subtypes and no longer describe the breast cancer panel collectively as TNBC.

To further address the reviewer's concern, we quantitatively re-examined the NCI-60 pharmacological data across all evaluable cell lines. This analysis confirmed that the available data do not support subtype-selective activity toward AR-negative prostate cancer or TNBC. Consequently, we have removed all statements implying subtype selectivity or preferential activity toward hormone-independent disease throughout the Title, Highlights, Abstract, Results, Discussion, and Conclusions. Instead, the revised manuscript now describes NSC828786 as exhibiting phenotypic antiproliferative activity within the available NCI-60 cancer models, without attributing this activity to specific receptor-defined tumor subtypes.

(3) Additional experimental validation.

We appreciate the reviewer's recommendations regarding independent validation using AR-positive prostate cancer models (e.g., LNCaP, VCaP, and 22Rv1), additional AR-negative models, receptor-defined breast cancer models, nonmalignant epithelial controls, and direct comparison with niclosamide under identical experimental conditions. We fully agree that these experiments would substantially strengthen the pharmacological characterization of NSC828786. However, these validation studies were beyond the scope of the present phenotype-oriented investigation integrating public pharmacological resources with transcriptomic and computational analyses. We have now explicitly included these experimental approaches as important future directions in the revised Discussion and Limitations.

Collectively, these revisions substantially moderate the interpretation of the NCI-60 data and clarify that the present study does not establish subtype-selective activity, but rather reports the exploratory phenotypic antiproliferative profile of NSC828786 within the currently available NCI-60 cancer models.

4     Supplementary Table S4 requires extensive verification and correction. First, the molecular formula for NSC828786 is listed as C19H11F4NO5, whereas the main manuscript gives C19H11F4NO2. The reported molecular weight of 361.29 g/mol is consistent with the latter formula, not the former. The chemical identity and all calculated descriptors should therefore be rechecked using the correct structure and SMILES. Second, several comparator molecular weights are displayed incorrectly, including values of 7.43, 1.5, and 3.9 g/mol for apalutamide, tamoxifen, and paclitaxel. Compound identifiers also appear duplicated or potentially incorrect. All compound identities, identifiers, structures, and outputs should be audited. Third, the interpretation is more favorable than the reported predictions support. NSC828786 has a consensus logP of 5.20 and predicted logS values of −6.08 and −7.93, indicating very high lipophilicity and extremely poor predicted aqueous solubility. These are meaningful development liabilities and should not be summarized merely as “moderate lipophilicity” or as unlikely to limit pharmacology. Fourth, the toxicity table reports scores of 0.88 for hERG blockade and 0.89 for hepatotoxicity. If these values represent probabilities of positive classification, they indicate potentially important safety alerts and directly conflict with the statement that no overt high-risk liabilities were identified. The authors must identify the platform and model used for each endpoint, define score directionality and classification thresholds, and interpret the results accordingly. Finally, log Kp from SwissADME represents predicted skin permeation, not blood–brain barrier permeability. It should not be used to infer limited CNS exposure. BBB predictions should be reported separately using the relevant BBB model, with model confidence and applicability-domain information. The developability conclusions should be rewritten conservatively after correcting these issues.

Response:

We sincerely thank the reviewer for this careful and technically detailed evaluation of Supplementary Table S3. We agree that several errors and overly optimistic interpretations were present in the original supplementary information. Accordingly, we have comprehensively re-audited the entire developability analysis using the verified chemical structure of NSC828786 and have substantially revised Supplementary Table S3, Supplementary Figure S4, and the corresponding Results and Discussion.

(1) Chemical identity and molecular descriptors.

We confirmed that the correct molecular formula of NSC828786 is C₁₉H₁₁F₄NO₂, which is fully consistent with its verified molecular weight (361.29 g/mol). The incorrect molecular formula originally reported in Supplementary Table S4 resulted from a transcription error and has now been corrected. In addition, the chemical structure, SMILES representation, and all calculated physicochemical descriptors have been regenerated and verified using the correct molecular structure.

(2) Verification of comparator compounds.

All comparator compounds (including apalutamide, tamoxifen, paclitaxel, niclosamide, and other reference compounds) have been comprehensively re-audited. Incorrect molecular weights, duplicated identifiers, and formatting inconsistencies have been corrected following verification against the corresponding reference structures. The revised Supplementary Table S4 has been carefully checked to ensure consistency among compound identity, molecular structure, molecular weight, and all calculated descriptors.

(3) Conservative interpretation of developability predictions.

We agree that the original interpretation overstated the developability profile of NSC828786. The revised manuscript now explicitly states that the predicted consensus LogP (5.20) and LogS values (−6.08 and −7.93) indicate high lipophilicity and poor predicted aqueous solubility, representing potential formulation and developability challenges rather than favorable physicochemical properties. Accordingly, both the Results and Discussion have been rewritten to provide a more balanced and conservative interpretation of these computational predictions.

(4) Toxicity prediction.

We appreciate the reviewer's comments regarding the interpretation of the toxicity prediction results. In the revised Supplementary Table S4, we now clearly identify the prediction platform used for each endpoint together with the corresponding model, score interpretation, and classification criteria. The toxicity section has been revised to avoid overstating compound safety, and we now emphasize that these computational predictions should be regarded as preliminary in silico assessments that identify potential safety concerns requiring subsequent experimental validation rather than definitive evidence of safety.

(5) Blood–brain barrier prediction.

We agree that the original manuscript incorrectly interpreted SwissADME logKp as a predictor of blood–brain barrier permeability. This statement has been removed. Blood–brain barrier permeability is now reported separately using an appropriate dedicated BBB prediction model, and the corresponding Results and Discussion have been revised accordingly. We no longer draw conclusions regarding central nervous system exposure based on skin-permeation predictions.

Collectively, these revisions substantially improve the accuracy, transparency, and scientific rigor of the computational developability assessment. The revised manuscript now presents a considerably more conservative interpretation of the in silico predictions while emphasizing that comprehensive pharmacokinetic, toxicological, and experimental validation studies will be required before the developability profile of NSC828786 can be fully established.

5     The supplementary information now reports embryo numbers, replicate structure, daily medium renewal, temperature, and DMSO concentration. However, the authors should clarify whether biological replicates represent independent clutches, identify the experimental unit, report exact group sizes, and state whether assessments were blinded. Because NSC828786 is predicted to have low solubility, precipitation and actual exposure concentrations should also be addressed. The statistical statement remains inadequate. One-way ANOVA is not generally appropriate for longitudinal survival and hatching proportions. Survival, hatching, and continuous morphometric endpoints should each have a prespecified analysis. Generalized mixed models, survival methods, or other analyses accounting for repeated observations and clustering by well or clutch may be appropriate. The exact post hoc test, assumption checks, multiplicity correction, effect sizes, confidence intervals, and exact p values should be reported. The manuscript states that exposure began at 4 hpf, whereas the Figure 4 timeline begins at 0 hpf. This inconsistency must be corrected. Figure 4B should also include scale bars.

Response:

We sincerely thank the reviewer for these constructive comments regarding the design, reporting, and statistical analysis of the zebrafish embryo experiments. We agree that additional methodological details improve the transparency, reproducibility, and interpretability of the study. Accordingly, we have substantially revised the Supplementary Methods, Figure 4, the corresponding Results, and the Discussion.

(1) Experimental design and biological replicates.

We now explicitly describe the experimental design in the revised Supplementary Methods. Embryos used in this study originated from a single breeding pair (one clutch) obtained from the institutional zebrafish facility. Individual embryos were randomly allocated to control and treatment groups prior to exposure, constituting a completely randomized experimental design at the level of the individual embryo rather than a clustered design involving multiple independent clutches. We have also clarified the experimental unit, exact group sizes for each endpoint, daily renewal of exposure medium, incubation temperature, and final DMSO concentration. Morphometric measurements were performed using predefined ImageJ-assisted criteria. We further acknowledge that the study was conducted using a single clutch and that future validation using multiple independent clutches will be required to strengthen the biological reproducibility of these findings.

(2) Solubility and compound exposure.

We appreciate the reviewer's comment regarding the predicted low aqueous solubility of NSC828786. The compound was prepared as a DMSO stock solution and freshly diluted into E3 embryo medium immediately before use, with a final DMSO concentration of 0.1% in all treatment groups. No visible precipitation was observed throughout the exposure period under the experimental conditions used. We have added this information to the revised Supplementary Methods. However, because dissolved drug concentrations were not analytically quantified, the reported concentrations should be interpreted as nominal exposure concentrations. This limitation is now explicitly acknowledged in the revised Discussion.

(3) Statistical analyses.

We agree that the original statistical description was insufficiently detailed. Because embryos originated from a single clutch, mixed-effects models incorporating clutch as a random effect were not appropriate for the present dataset. We have therefore revised the Statistical Analysis section to clearly specify the analytical approach used for each endpoint, including the experimental unit, statistical tests, post hoc procedures where applicable, assumption checks, and significance criteria. Continuous morphometric endpoints were analyzed separately from survival and hatching outcomes, and the corresponding statistical methods are now described explicitly in the revised Methods. Exact p values are reported whenever available, and additional statistical details have been incorporated into the revised Supplementary Materials.

(4) Figure revisions.

We thank the reviewer for identifying the inconsistencies in Figure 4. The experimental timeline has been corrected to indicate that compound exposure began at 4 hours post-fertilization (hpf), consistent with the Methods section. In addition, scale bars have been added to all representative embryo images in Figure 4B, and the corresponding figure legend has been revised accordingly.

Collectively, these revisions substantially improve the methodological transparency and reporting quality of the zebrafish experiments while emphasizing that these data provide preliminary organism-level tolerability information. We now explicitly acknowledge that confirmation of the developmental safety profile of NSC828786 will require future studies incorporating multiple independent clutches together with more comprehensive pharmacological and toxicological evaluation.

6     HPN and AMACR are already established prostate cancer-associated genes. The principal novelty would therefore depend on demonstrating a functional relationship with NSC828786, which is not currently established. The manuscript should be reframed as an exploratory, hypothesis-generating study unless direct mechanistic validation is added.

Response:

We sincerely thank the reviewer for this important comment. We agree that HPN and AMACR are well-established prostate cancer–associated genes, and this concern has also been addressed in our response to Major Comment 1. We do not claim that these genes represent newly identified biomarkers or previously unrecognized drivers of prostate cancer biology. Rather, the novelty of the present study lies in the integration of independent pharmacological profiling, cross-cohort transcriptomics, systems biology, and structure-based computational analyses to place the phenotypic activity of NSC828786 into a biologically relevant molecular context.

We fully agree that, in the absence of direct mechanistic validation, the present data do not establish a functional relationship between NSC828786 and HPN or AMACR. Accordingly, we have substantially reframed the manuscript as an exploratory, phenotype-oriented, and hypothesis-generating study, rather than a mechanistic investigation or target-validation study.

To reflect this revised scope, we have comprehensively revised the Title, Highlights, Graphical Abstract, Abstract, Introduction, Results, Discussion, and Conclusions. Throughout the revised manuscript, HPN and AMACR are consistently described as exploratory candidate molecular associations identified through cross-cohort transcriptomic integration and computational analyses, rather than validated molecular targets or direct mediators of NSC828786 responsiveness. We further emphasize that the present study confirms the recurrent upregulation of HPN and AMACR across independent prostate cancer cohorts and explores their potential biological context in relation to the phenotypic activity of NSC828786, rather than claiming a causal mechanistic relationship.

Furthermore, we have expanded the Discussion to explicitly acknowledge that direct target-engagement and mechanistic validation—including genetic perturbation, biochemical inhibition assays, and orthogonal target-validation approaches (e.g., CETSA, DARTS, or thermal proteome profiling)—will be required to determine whether HPN, AMACR, or related adaptive pathways functionally contribute to the observed antiproliferative activity of NSC828786.

We believe these revisions more accurately define both the scope and the novelty of the present work as a hypothesis-generating framework for prioritizing candidate molecular contexts associated with phenotypically active compounds, while avoiding mechanistic claims that are not directly supported by the current data.

Minor comments

1     Reconcile the R versions reported in the supplementary information: version 4.3.2 is stated for GEO processing, whereas version 4.2.2 is stated in the statistical section.

Response:

We thank the reviewer for identifying this inconsistency. All analyses were performed using R version 4.3.2. The discrepant reference to R version 4.2.2 in the Statistical Analysis section of the Supplementary Methods resulted from a typographical error and has now been corrected. The manuscript and Supplementary Methods have been revised to consistently report R version 4.3.2 throughout.

2     Explain the multiple colored RMSF traces in Figure 3 and state exactly what each trace represents.

Response:

We thank the reviewer for this helpful suggestion. We agree that the original Figure 3 legend did not sufficiently explain the multiple colored RMSF traces. Accordingly, we have revised both the Figure 3 legend and the corresponding Results section to clearly define the meaning of each RMSF profile. The colored traces represent residue-wise root-mean-square fluctuation (RMSF) profiles of the target protein backbone obtained from molecular dynamics simulations of complexes with five different ligands: NSC828786 (red), apalutamide (green), enzalutamide (blue), niclosamide (purple), and honokiol (yellow). These comparator compounds were included as clinically approved drugs or structurally related reference compounds to facilitate comparative evaluation of ligand-associated protein dynamics. In the revised manuscript, we further clarify that Figure 3D presents the RMSF comparison for the HPN complexes, whereas Figure 3E corresponds to the AMACR complexes (please confirm the protein assignment). This comparative analysis illustrates how different ligands influence local protein flexibility throughout the simulations. We now explicitly describe in the revised Results that the NSC828786-bound complex exhibited relatively increased residue fluctuations in specific regions compared with the comparator ligands, particularly within residues approximately 160–190 and 240–260 in the corresponding RMSF profile, while emphasizing that these observations represent comparative computational dynamics rather than evidence of differential biological activity or target engagement.

3     Use consistent terminology for “growth percentage,” “growth inhibition,” GI50, IC50, TGI, and LC50.

Response:

We thank the reviewer for this helpful suggestion. We agree that the original manuscript used several pharmacological response terms inconsistently, which could lead to confusion. Accordingly, we have carefully reviewed and standardized the terminology throughout the manuscript, including the Results, Figure 1, Figure legends, Methods, and Supplementary Materials. Specifically, "growth percentage" is now used exclusively to describe the single-dose (10 μM) NCI-60 screening results, whereas GI₅₀, TGI, LC₅₀, and IC₅₀ are reserved exclusively for the corresponding five-dose pharmacological response parameters provided by the NCI Developmental Therapeutics Program (NCI-DTP). In addition, the terms "growth inhibition" and "cell growth" are now used consistently to describe the biological response measured in the NCI-60 assay. These revisions improve the consistency and clarity of the pharmacological terminology throughout the revised manuscript.

4     Replace general statements such as “appropriate normalization,” “default parameters,” “standard workflows,” and “post hoc tests” with exact procedures.

Response:

We thank the reviewer for this valuable suggestion. We agree that several methodological descriptions in the original manuscript were overly general and did not provide sufficient detail for reproducibility. Accordingly, we have revised the Materials and Methods and Supplementary Methods to replace general expressions such as "appropriate normalization," "default parameters," "standard workflows," and "post hoc tests" with explicit descriptions of the analytical procedures used. Specifically, we now report the exact normalization methods applied to each transcriptomic dataset, the software versions and computational parameters used for molecular docking and molecular dynamics simulations (including AutoDock Vina settings), and the precise statistical tests and multiple-comparison procedures used for each experimental endpoint. Where applicable, software versions, package names, and analytical parameters are now explicitly provided to improve reproducibility. These revisions substantially enhance the transparency, reproducibility, and methodological rigor of the study.

5     Correct editorial errors including duplicated words in the correspondence line, duplicated periods in section headings, inconsistent manuscript year information, and punctuation and spacing errors throughout.

Response:

We thank the reviewer for carefully identifying these editorial issues. All identified editorial errors have been corrected in the revised manuscript. Specifically, the duplicated text in the correspondence line has been removed, duplicated punctuation in section headings (e.g., "3.7.") has been corrected, manuscript year and date references have been reconciled for consistency, and the entire manuscript has been carefully proofread to correct punctuation, spacing, formatting, and other minor typographical inconsistencies throughout.

 

Reviewer 2

Comments and Suggestions for Authors

The submission by Wen et al. deals with the anticancer properties of the new salicylanilide derivative NSC828786. The described results are meaningful, but the manuscript style and experiments have some flaws. Thus, I recommend major revision for the following reasons:

Response:

We sincerely thank the reviewer for the careful evaluation of our manuscript and for the constructive comments and suggestions. We appreciate the reviewer's insightful feedback, which has helped us improve both the scientific presentation and the overall clarity of the manuscript. Below, we provide a detailed point-by-point response to each comment. All corresponding revisions have been incorporated into the revised manuscript and are highlighted using track changes.

Abstract: The abbreviations AMACR and HPN should be explained in the abstract.

Response:

We thank the reviewer for this helpful suggestion. The Abstract has been revised to define both abbreviations at their first appearance. Specifically, AMACR is now introduced as alpha-methylacyl-CoA racemase, and HPN as hepsin, consistent with the terminology used throughout the manuscript.

Introduction: The roles and functions of AMACR and HPN should be described in more detail in the introduction.

Response:

We thank the reviewer for this valuable suggestion and agree that additional biological background would improve the rationale for the study. Accordingly, we have expanded the Introduction to provide a more comprehensive description of the biological functions and clinical relevance of both AMACR and HPN. Specifically, AMACR (alpha-methylacyl-CoA racemase) is now described as a peroxisomal and mitochondrial enzyme that catalyzes the racemization of branched-chain fatty acid and bile acid intermediates, thereby supporting lipid metabolism and β-oxidation. We also note its well-established role as a diagnostic immunohistochemical marker of prostate adenocarcinoma. In addition, HPN (hepsin) is now introduced as a type II transmembrane serine protease involved in extracellular matrix remodeling through the proteolytic processing of extracellular substrates. We further describe its reported roles in epithelial remodeling, basement membrane disruption, and tumor invasion when aberrantly overexpressed.

Finally, we have added a concluding paragraph emphasizing that although AMACR and HPN have traditionally been investigated independently, both are frequently overexpressed in prostate cancer and participate in complementary metabolic and extracellular remodeling processes. This expanded background provides the biological rationale for exploring these genes together as candidate metabolic–proteolytic molecular associations in the context of the present exploratory study.

3.1.: Please compare the activities of NSC828786 with those of niclosamide.

Response:

We thank the reviewer for this valuable suggestion. As recommended, we have included a direct comparison between NSC828786 and niclosamide (NSC188817) using pharmacological response data obtained from the NCI-60 Developmental Therapeutics Program database. Specifically, comparative GI50, TGI, and LC50 values for both compounds are now presented for the prostate cancer cell lines (PC-3 and DU145) and representative breast cancer cell lines (MDA-MB-231 and BT-549) in the revised Table 1 and Supplementary Table S2. The corresponding Results section and Discussion have also been revised to incorporate this comparison. To avoid overinterpretation, we do not conclude that NSC828786 is more potent or more selective than niclosamide. Rather, the comparison demonstrates that NSC828786 exhibits a pharmacological activity profile broadly comparable to that of niclosamide within the evaluated NCI-60 models, supporting its characterization as a niclosamide-like salicylanilide derivative while acknowledging that definitive comparisons of potency and selectivity will require side-by-side experimental evaluation under identical conditions. These revisions provide the direct pharmacological comparison requested by the reviewer while maintaining a conservative interpretation of the available NCI-60 data.

3.1.: What positive control (approved anticancer drug) was applied for the NCI-60 antiproliferation experiments?

Response:

We thank the reviewer for this important question. The antiproliferative data analyzed in the present study were obtained from the National Cancer Institute Developmental Therapeutics Program (NCI-DTP), which performs compound evaluation using its standardized NCI-60 five-dose sulforhodamine B (SRB) screening platform. Because the screening experiments were conducted by the NCI-DTP rather than in our laboratory, the publicly available dataset does not specify a compound-specific positive control for individual NSC compounds, including NSC828786. Instead, all compounds are evaluated using the same standardized experimental protocol, allowing direct comparison of pharmacological response parameters (GI₅₀, TGI, and LC₅₀) across compounds within the NCI-60 database. To clarify this point, we have revised the Materials and Methods to explicitly state that the NCI-60 pharmacological data were retrieved from the NCI-DTP database and generated using the standardized NCI-60 screening platform. We have also added the appropriate reference describing the NCI-DTP assay methodology.

3.1.: A table with the IC50 values should be added to the main manuscript and not to the supplementary materials file. Dose-response graphs of Figure 1E and 1F can be moved to the supplementary materials instead.

Response:

We thank the reviewer for this helpful suggestion. Following the reviewer's recommendation, the quantitative pharmacological response parameters (GI₅₀, TGI, LC₅₀, and IC₅₀) have been transferred from the Supplementary Materials to the main manuscript and are now presented as Table 1, allowing readers to directly evaluate the antiproliferative activity of NSC828786 within the main text. In addition, the five-dose concentration–response curves previously shown in Figure 1E and Figure 1F have been moved to the Supplementary Materials, where they continue to provide representative pharmacological response profiles without interrupting the flow of the main Results section. We believe these revisions improve the organization and accessibility of the pharmacological data while presenting the key quantitative results more prominently in the main manuscript.

3.5.: The structures of NSC828786 (amide?!) shown in Figure 3 are wrong. Thus, the described docking and MD results are doubtful, too.

Response:

We sincerely thank the reviewer for identifying this important error. After carefully re-examining our computational workflow, we confirmed that the incorrect chemical structure shown in the original Figure 3 resulted solely from an error in the graphical representation of the ligand used to prepare the figure. Specifically, the salicylanilide linkage was incorrectly depicted in the interaction diagrams. The chemical structure shown in the revised Figure 3 has now been corrected to the verified structure of NSC828786 (C19H11F4NO2), consistent with the synthesized compound described in Section 2.1, its NMR characterization, and the structure presented in Figure 1B. Importantly, we carefully verified the ligand files used throughout the computational study, including the docking input structure, molecular dynamics topology, and simulation coordinates. These files all correspond to the correct chemical structure of NSC828786. Therefore, the error was limited exclusively to the graphical illustration shown in the original figure and did not affect the molecular docking calculations, molecular dynamics simulations, binding analyses, or the conclusions derived from these computational studies. To avoid any ambiguity, the revised Figure 3 has been regenerated using the correct ligand structure, and all interaction diagrams have been updated accordingly. The Results and corresponding figure legend have also been carefully reviewed to ensure complete consistency with the corrected chemical structure. We appreciate the reviewer for identifying this graphical error, which has improved the accuracy and clarity of the revised manuscript.

3.7.: The authors should also investigate the toxicity of niclosamide and compare it with the toxicity of NSC 828786. Maybe there is also some published literature about niclosamide toxicity in zebrafish the authors might cite instead.

Response:

We thank the reviewer for this valuable suggestion. We agree that comparison with the parent salicylanilide compound, niclosamide, would provide useful context for interpreting the developmental tolerability of NSC828786. Because the present study was designed to evaluate the preliminary developmental effects of NSC828786, we did not perform parallel zebrafish toxicity experiments using niclosamide, and such experiments were beyond the scope of the current work. Instead, following the reviewer's recommendation, we have expanded the Discussion to include published zebrafish studies describing the developmental toxicity profile of niclosamide, including its reported developmental and cardiotoxic effects at low-micromolar concentrations. We now discuss these published findings alongside our observations for NSC828786, noting that although both compounds belong to the salicylanilide family, direct quantitative comparison should be interpreted cautiously because the studies were performed under different experimental conditions. Accordingly, we present the literature comparison as contextual information rather than evidence of superior or equivalent safety. The relevant references have been incorporated into the revised manuscript and are cited in Section 3.7 and the Discussion.

Discussion: Please discuss the potential of NSC828786 as a new drug for the therapy of prostate cancer, also in comparison with currently applied drugs for prostate cancer therapy. Literature about the effects of niclosamide on prostate cancers should also be discussed and cited.

Response:

We thank the reviewer for this valuable suggestion. In response, we have substantially expanded the Discussion to better position NSC828786 within the current therapeutic landscape of prostate cancer and to place our findings in the context of the existing literature on niclosamide. Specifically, we now summarize the current standard-of-care therapies for advanced prostate cancer, including androgen deprivation therapy, androgen receptor (AR) pathway inhibitors (e.g., enzalutamide, abiraterone, and apalutamide), and taxane-based chemotherapy (docetaxel and cabazitaxel). We discuss their established mechanisms of action and the clinical challenge posed by the emergence of therapeutic resistance. We have also expanded the discussion of niclosamide, highlighting published studies demonstrating its antitumor activity in prostate cancer, including its reported effects on AR-V7, Wnt/β-catenin, STAT3, and mitochondrial signaling pathways, as well as its evaluation in combination with standard therapies in preclinical and early clinical studies. Appropriate references have been added to support these statements. Importantly, we have revised the Discussion to position NSC828786 as an exploratory salicylanilide lead compound with a pharmacological profile that warrants further investigation, rather than as a validated therapeutic candidate or defined molecular inhibitor. We now explicitly state that direct comparisons with currently approved prostate cancer therapies—including efficacy, mechanism of action, pharmacokinetics, and safety—will require dedicated experimental studies in receptor-defined prostate cancer models and appropriate in vivo systems before any conclusions regarding therapeutic potential can be drawn. These revisions provide broader clinical context while maintaining a balanced and appropriately cautious interpretation of the present findings.

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The submission by Wen et al. deals with the anticancer properties of the new salicylanilide derivative NSC828786. The described results are meaningful, but the manuscript style and experiments have some flaws. Thus, I recommend major revision for the following reasons:

Abstract: The abbreviations AMACR and HPN should be explained in the abstract.

Introduction: The roles and functions of AMACR and HPN should be described in more detail in the introduction.

3.1.: Please compare the activities of NSC828786 with those of niclosamide.

3.1.: What positive control (approved anticancer drug) was applied for the NCI-60 antiproliferation experiments?

3.1.: A table with the IC50 values should be added to the main manuscript and not to the supplementary materials file. Dose-response graphs of Figure 1E and 1F can be moved to the supplementary materials instead.

3.5.: The structures of NSC828786 (amide?!) shown in Figure 3 are wrong. Thus, the described docking and MD results are doubtful, too.

3.7.: The authors should also investigate the toxicity of niclosamide and compare it with the toxicity of NSC 828786. Maybe there is also some published literature about niclosamide toxicity in zebrafish the authors might cite instead.

Discussion: Please discuss the potential of NSC828786 as a new drug for the therapy of prostate cancer, also in comparison with currently applied drugs for prostate cancer therapy. Literature about the effects of niclosamide on prostate cancers should also be discussed and cited.

Author Response

Reviewer 2

Comments and Suggestions for Authors

The submission by Wen et al. deals with the anticancer properties of the new salicylanilide derivative NSC828786. The described results are meaningful, but the manuscript style and experiments have some flaws. Thus, I recommend major revision for the following reasons:

Response:

We sincerely thank the reviewer for the careful evaluation of our manuscript and for the constructive comments and suggestions. We appreciate the reviewer's insightful feedback, which has helped us improve both the scientific presentation and the overall clarity of the manuscript. Below, we provide a detailed point-by-point response to each comment. All corresponding revisions have been incorporated into the revised manuscript and are highlighted using track changes.

Abstract: The abbreviations AMACR and HPN should be explained in the abstract.

Response:

We thank the reviewer for this helpful suggestion. The Abstract has been revised to define both abbreviations at their first appearance. Specifically, AMACR is now introduced as alpha-methylacyl-CoA racemase, and HPN as hepsin, consistent with the terminology used throughout the manuscript.

Introduction: The roles and functions of AMACR and HPN should be described in more detail in the introduction.

Response:

We thank the reviewer for this valuable suggestion and agree that additional biological background would improve the rationale for the study. Accordingly, we have expanded the Introduction to provide a more comprehensive description of the biological functions and clinical relevance of both AMACR and HPN. Specifically, AMACR (alpha-methylacyl-CoA racemase) is now described as a peroxisomal and mitochondrial enzyme that catalyzes the racemization of branched-chain fatty acid and bile acid intermediates, thereby supporting lipid metabolism and β-oxidation. We also note its well-established role as a diagnostic immunohistochemical marker of prostate adenocarcinoma. In addition, HPN (hepsin) is now introduced as a type II transmembrane serine protease involved in extracellular matrix remodeling through the proteolytic processing of extracellular substrates. We further describe its reported roles in epithelial remodeling, basement membrane disruption, and tumor invasion when aberrantly overexpressed.

Finally, we have added a concluding paragraph emphasizing that although AMACR and HPN have traditionally been investigated independently, both are frequently overexpressed in prostate cancer and participate in complementary metabolic and extracellular remodeling processes. This expanded background provides the biological rationale for exploring these genes together as candidate metabolic–proteolytic molecular associations in the context of the present exploratory study.

3.1.: Please compare the activities of NSC828786 with those of niclosamide.

Response:

We thank the reviewer for this valuable suggestion. As recommended, we have included a direct comparison between NSC828786 and niclosamide (NSC188817) using pharmacological response data obtained from the NCI-60 Developmental Therapeutics Program database. Specifically, comparative GI50, TGI, and LC50 values for both compounds are now presented for the prostate cancer cell lines (PC-3 and DU145) and representative breast cancer cell lines (MDA-MB-231 and BT-549) in the revised Table 1 and Supplementary Table S2. The corresponding Results section and Discussion have also been revised to incorporate this comparison. To avoid overinterpretation, we do not conclude that NSC828786 is more potent or more selective than niclosamide. Rather, the comparison demonstrates that NSC828786 exhibits a pharmacological activity profile broadly comparable to that of niclosamide within the evaluated NCI-60 models, supporting its characterization as a niclosamide-like salicylanilide derivative while acknowledging that definitive comparisons of potency and selectivity will require side-by-side experimental evaluation under identical conditions. These revisions provide the direct pharmacological comparison requested by the reviewer while maintaining a conservative interpretation of the available NCI-60 data.

3.1.: What positive control (approved anticancer drug) was applied for the NCI-60 antiproliferation experiments?

Response:

We thank the reviewer for this important question. The antiproliferative data analyzed in the present study were obtained from the National Cancer Institute Developmental Therapeutics Program (NCI-DTP), which performs compound evaluation using its standardized NCI-60 five-dose sulforhodamine B (SRB) screening platform. Because the screening experiments were conducted by the NCI-DTP rather than in our laboratory, the publicly available dataset does not specify a compound-specific positive control for individual NSC compounds, including NSC828786. Instead, all compounds are evaluated using the same standardized experimental protocol, allowing direct comparison of pharmacological response parameters (GI₅₀, TGI, and LC₅₀) across compounds within the NCI-60 database. To clarify this point, we have revised the Materials and Methods to explicitly state that the NCI-60 pharmacological data were retrieved from the NCI-DTP database and generated using the standardized NCI-60 screening platform. We have also added the appropriate reference describing the NCI-DTP assay methodology.

3.1.: A table with the IC50 values should be added to the main manuscript and not to the supplementary materials file. Dose-response graphs of Figure 1E and 1F can be moved to the supplementary materials instead.

Response:

We thank the reviewer for this helpful suggestion. Following the reviewer's recommendation, the quantitative pharmacological response parameters (GI₅₀, TGI, LC₅₀, and IC₅₀) have been transferred from the Supplementary Materials to the main manuscript and are now presented as Table 1, allowing readers to directly evaluate the antiproliferative activity of NSC828786 within the main text. In addition, the five-dose concentration–response curves previously shown in Figure 1E and Figure 1F have been moved to the Supplementary Materials, where they continue to provide representative pharmacological response profiles without interrupting the flow of the main Results section. We believe these revisions improve the organization and accessibility of the pharmacological data while presenting the key quantitative results more prominently in the main manuscript.

3.5.: The structures of NSC828786 (amide?!) shown in Figure 3 are wrong. Thus, the described docking and MD results are doubtful, too.

Response:

We sincerely thank the reviewer for identifying this important error. After carefully re-examining our computational workflow, we confirmed that the incorrect chemical structure shown in the original Figure 3 resulted solely from an error in the graphical representation of the ligand used to prepare the figure. Specifically, the salicylanilide linkage was incorrectly depicted in the interaction diagrams. The chemical structure shown in the revised Figure 3 has now been corrected to the verified structure of NSC828786 (C19H11F4NO2), consistent with the synthesized compound described in Section 2.1, its NMR characterization, and the structure presented in Figure 1B. Importantly, we carefully verified the ligand files used throughout the computational study, including the docking input structure, molecular dynamics topology, and simulation coordinates. These files all correspond to the correct chemical structure of NSC828786. Therefore, the error was limited exclusively to the graphical illustration shown in the original figure and did not affect the molecular docking calculations, molecular dynamics simulations, binding analyses, or the conclusions derived from these computational studies. To avoid any ambiguity, the revised Figure 3 has been regenerated using the correct ligand structure, and all interaction diagrams have been updated accordingly. The Results and corresponding figure legend have also been carefully reviewed to ensure complete consistency with the corrected chemical structure. We appreciate the reviewer for identifying this graphical error, which has improved the accuracy and clarity of the revised manuscript.

3.7.: The authors should also investigate the toxicity of niclosamide and compare it with the toxicity of NSC 828786. Maybe there is also some published literature about niclosamide toxicity in zebrafish the authors might cite instead.

Response:

We thank the reviewer for this valuable suggestion. We agree that comparison with the parent salicylanilide compound, niclosamide, would provide useful context for interpreting the developmental tolerability of NSC828786. Because the present study was designed to evaluate the preliminary developmental effects of NSC828786, we did not perform parallel zebrafish toxicity experiments using niclosamide, and such experiments were beyond the scope of the current work. Instead, following the reviewer's recommendation, we have expanded the Discussion to include published zebrafish studies describing the developmental toxicity profile of niclosamide, including its reported developmental and cardiotoxic effects at low-micromolar concentrations. We now discuss these published findings alongside our observations for NSC828786, noting that although both compounds belong to the salicylanilide family, direct quantitative comparison should be interpreted cautiously because the studies were performed under different experimental conditions. Accordingly, we present the literature comparison as contextual information rather than evidence of superior or equivalent safety. The relevant references have been incorporated into the revised manuscript and are cited in Section 3.7 and the Discussion.

Discussion: Please discuss the potential of NSC828786 as a new drug for the therapy of prostate cancer, also in comparison with currently applied drugs for prostate cancer therapy. Literature about the effects of niclosamide on prostate cancers should also be discussed and cited.

Response:

We thank the reviewer for this valuable suggestion. In response, we have substantially expanded the Discussion to better position NSC828786 within the current therapeutic landscape of prostate cancer and to place our findings in the context of the existing literature on niclosamide. Specifically, we now summarize the current standard-of-care therapies for advanced prostate cancer, including androgen deprivation therapy, androgen receptor (AR) pathway inhibitors (e.g., enzalutamide, abiraterone, and apalutamide), and taxane-based chemotherapy (docetaxel and cabazitaxel). We discuss their established mechanisms of action and the clinical challenge posed by the emergence of therapeutic resistance. We have also expanded the discussion of niclosamide, highlighting published studies demonstrating its antitumor activity in prostate cancer, including its reported effects on AR-V7, Wnt/β-catenin, STAT3, and mitochondrial signaling pathways, as well as its evaluation in combination with standard therapies in preclinical and early clinical studies. Appropriate references have been added to support these statements. Importantly, we have revised the Discussion to position NSC828786 as an exploratory salicylanilide lead compound with a pharmacological profile that warrants further investigation, rather than as a validated therapeutic candidate or defined molecular inhibitor. We now explicitly state that direct comparisons with currently approved prostate cancer therapies—including efficacy, mechanism of action, pharmacokinetics, and safety—will require dedicated experimental studies in receptor-defined prostate cancer models and appropriate in vivo systems before any conclusions regarding therapeutic potential can be drawn. These revisions provide broader clinical context while maintaining a balanced and appropriately cautious interpretation of the present findings.

 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have made a substantial and constructive revision, which addresses the main conceptual concerns of the previous review.

Reviewer 2 Report

Comments and Suggestions for Authors

The revised manuscript is suitable for publication now.