Round 1

Reviewer 1 Report

Comments and Suggestions for Authors The manuscript addresses a genuine unmet need in veterinary regenerative medicine, where standardized potency assays for canine MSCs are lacking, and the single-platform HPLC concept is conceptually sound and novel in this context. The directional findings are internally consistent across all three donor lines, and the authors are appropriately cautious in framing conclusions at the feasibility level. The clinical motivation is well-grounded, and the writing is generally clear and well-organized. However, several concerns need to be addressed before the manuscript can be considered for publication.

 

1. The authors need to provide representative chromatograms for all four analytes under both licensed and unlicensed conditions.

2. Please provide spike recovery data and matrix interference assessment for IL-10 and TGF-β at the reported concentration range. Either validate HPLC protein quantification with an orthogonal method (e.g., ELISA) or explicitly reframe the protein measurements as exploratory/semi-quantitative with appropriate caveats throughout the manuscript.

3. The authors need to clarify and justify the difference in collection timepoints (48h vs. 72h) between analyte classes, and acknowledge this as a limitation on integrated panel interpretation.

4. Please confirm whether FBS-free conditions applied equally across analyte classes, and if not, discuss the potential impact on cross-analyte comparison.

5. The authors have to justify the IFN-γ and TNF-α concentrations selected relative to published literature and physiologically relevant concentrations in canine IMHA.

6. The authors need to soften language in the abstract and conclusions where pilot data (n=3) is presented in a manner that may imply broader generalizability.

7. Please explicitly distinguish mechanistic evidence derived from human/murine MSC literature from validated canine-specific biology throughout the Discussion.

8. Please add a dedicated limitations paragraph that consolidates all methodological constraints.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

SUMMARY

This manuscript describes a compact HPLC-based potency panel intended to quantify four immunomodulatory outputs, adenosine, kynurenine, IL-10, and TGF-β, from conditioned media of canine adipose-derived mesenchymal stromal cells (MSCs). The authors tested three independent MSC lines under baseline and cytokine-licensed conditions (IFN-γ + TNF-α) and reported directionally consistent increases for all four analytes following licensing.The topic is relevant. Potency testing is a recognized bottleneck in MSC manufacturing, and the idea of anchoring multiple analytes to a single chromatographic platform is worth exploring. That said, the manuscript in its current form reads more as a feasibility report from a single laboratory than as a validated method paper. The analytical evidence for detecting cytokines by RP-HPLC-UV without separation from complex biological matrix is thin, and the statistical analysis is insufficient to support the conclusions as written. Major revisions are required before this work can be recommended for publication.

MAJOR ISSUES

1. The core analytical claim for IL-10 and TGF-β is not substantiated

This is the most serious problem in the manuscript. The authors quantify IL-10 and TGF-β from conditioned media by reversed-phase HPLC with UV detection at 214 nm. However, no evidence is provided that the detected peaks actually correspond to these cytokines rather than to co-eluting matrix components.

The authors acknowledge that 214 nm detects all peptide bonds non-selectively. Conditioned media, even after concentration and 0.22 μm filtration, contains dozens of secreted proteins that could co-elute at similar retention times. Without mass spectrometry confirmation, immunoprecipitation, or at minimum a spike-recovery experiment in conditioned media matrix (not blank medium), the assignment of peaks to IL-10 and TGF-β is not credible.

The reported concentrations (IL-10: ~0.8–1.3 ng/mL; TGF-β: ~1.1–1.8 ng/mL) are also well below what most RP-HPLC-UV setups can reliably detect without enrichment to near-purity. The 20-fold concentration step helps, but concentrating complex media 20x also concentrates all matrix interferents by the same factor.

The authors need to either: (a) provide orthogonal confirmation of peak identity (ELISA, mass spectrometry, or immunodepletion) for at least one cell line and condition, or (b) substantially reframe the protein claims as preliminary and exploratory, with explicit acknowledgment that the peaks have not been confirmed to represent the intended analytes.

2. Three biological replicates with no inferential statistics

The dataset consists of three MSC lines, one measurement per line per condition. This is a legitimate sample size for a pilot study, but the manuscript presents fold-changes and CV values without any statistical test. No p-values, no confidence intervals, no non-parametric analysis. The reader cannot assess whether the observed differences are consistent with chance.

For a pilot with n = 3, a paired Wilcoxon signed-rank test or a sign test would be more appropriate than parametric testing, and these should be applied and reported. If the authors maintain that formal testing is premature at this n, that position should be stated explicitly in the limitations, and the language in the abstract and conclusions should be revised accordingly. Phrases such as “directionally consistent” and “low analytical variability” should not stand in for statistical evidence.

3. Selectivity of the adenosine and kynurenine methods is not demonstrated

Adenosine is measured in the filtrate from a 10 kDa MWCO filter, and kynurenine is measured after TCA precipitation. These preparation steps reduce protein content but do not eliminate low-molecular-weight metabolites that absorb at 260 nm (nucleotides, nucleosides, NAD+, NADH, FAD) or at 360 nm (other kynurenine pathway intermediates such as 3-hydroxykynurenine or anthranilic acid).

The manuscript states that retention time matching to reference standards was used for peak identity confirmation, and that standards were run in blank medium. This is a necessary but not sufficient control. The authors should show representative chromatograms from conditioned media with the assigned peaks labeled, demonstrate that unlicensed media does not show interfering co-elution, and ideally include a spiked recovery in conditioned media matrix for at least one analyte. Without this, the reader cannot evaluate whether the measured peaks are clean analyte signal or overlapping metabolite background.

4. Conflict of interest and study independence

All authors are employees or affiliates of Safari Stem Cell, Inc. or Safari Veterinary Care Centers. The corresponding author is the founder of Safari Stem Cell, Inc. and is developing this panel for commercial application. The MSC clinical outcomes reference that anchors the biological relevance argument (ref. 31) is also a publication by the same group, using the same cell source, with two overlapping authors.

This does not invalidate the data, but it does raise the question of independent corroboration. The editors should consider whether the conflict-of-interest statement is adequately visible and whether the manuscript requires disclosure language beyond what appears in its current form. The authors should also be asked explicitly whether the canine donors, the MSC lines, and the clinical cases in ref. 31 are from the same animal population or overlapping cohorts.

MINOR ISSUES

5. Differing collection timepoints for metabolites and proteins are not justified

Small molecules were collected at 72 hours and proteins at 48 hours. The rationale given is reduction of albumin-mediated matrix interference for the protein samples, which is a practical consideration. However, using different timepoints for the same panel means the reported fold-changes are not directly comparable within a single secretory snapshot. This limitation should be stated clearly in the discussion. The authors might also consider whether a single timepoint could be adopted in future experiments if matrix concerns are addressed through alternative sample preparation.

6. System suitability and method performance metrics are incomplete

The manuscript reports calibration linearity (R² ≥ 0.9939), LOQ by SNR ≥ 10, and back-calculated accuracy within 10% for most standards. These are appropriate starting points, but several standard analytical validation parameters are missing: precision (intra-day and inter-day CV at multiple concentrations), accuracy across the calibration range, matrix effects, and recovery. For a manuscript making the case that this panel supports standardized manufacturing, the absence of these data is notable. The authors acknowledge that further method refinement is needed, but they should be more explicit about which validation components remain outstanding.

7. The TGF-β calibration has an unaddressed outlier

One mid-range TGF-β calibration standard showed 14% deviation from nominal, which is outside the commonly accepted 10% threshold. The authors note this and call it acceptable for a pilot study, which is a reasonable position — but the statement that “strong overall curve linearity” justifies this deviates from standard practice. The sentence should be revised to acknowledge that this point falls outside typical acceptance criteria and that the TGF-β calibration should be revisited in subsequent method development.

8. The biological interpretation of IMHA is overstated

The manuscript devotes considerable space to canine IMHA as a motivating use case. The connection is reasonable at the level of disease biology: IMHA involves dysregulated innate and adaptive immunity, and the mediators measured here are relevant to those processes. However, the paper goes further and uses the clinical outcomes data from ref. 31 (76% hematologic recovery) to frame the secretory profile described here as clinically meaningful. This is a stretch. The present study measures conditioned media outputs from three MSC lines in vitro, under artificial licensing conditions, with no direct link to the clinical cases. The authors should dial back the clinical framing and present the IMHA rationale as motivational context rather than evidence of therapeutic relevance.

9. Figure quality and statistical annotation

Figure 1 shows bar plots with error bars representing SD. At n = 3, bar plots with SD are difficult to interpret and can be misleading. The authors should consider presenting individual data points alongside the mean, or using dot plots with mean and range. If bar plots are retained, individual values should at least be overlaid. No statistical comparisons are annotated on the figures, which is consistent with the absence of formal testing but should be addressed as noted in Major Issue 2.

10. Writing and language

The manuscript is generally readable, but several sections are repetitive. The introduction restates the same framing (multi-analyte rationale, licensing justification, IMHA context) multiple times, and portions of the discussion closely mirror the results section. A tighter revision would improve readability. The phrase “a controllable upstream manufacturing lever” appears five or more times across the manuscript; it should appear once or twice, stated clearly, and then the authors can use less loaded language in subsequent references.

Additionally, the abbreviation list defines “SD” as “Standard of Deviation” (should be “Standard Deviation”) and “MSC” is defined as Mesenchymal Stem Cell in the abbreviation table but as “Mesenchymal Stromal/Stem Cell” throughout the text — the inconsistency should be resolved, and the authors should adopt the stromal/stem framing consistently given current field nomenclature (Viswanathan et al., 2019, ref. 13).

MAJOR REVISION REQUIRED.

The paper addresses a real need in the MSC field and the HPLC-first approach has conceptual merit. However, the protein analyte identification is not supported by sufficient analytical evidence, the statistical analysis is absent, and the clinical framing oversteps what the data can support. These issues are addressable and do not require new cell line experiments: orthogonal validation data for the protein peaks, application of appropriate non-parametric statistics to the existing dataset, revised language in the abstract and conclusions, and additional method characterization data would substantially strengthen the manuscript.

 

Author Response

Please see the attachment. 

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript "A Compact HPLC-Based Potency Panel for Canine Adipose-Derived Mesenchymal Stromal Cells: Quantification of Adenosine, Kynurenine, IL-10, and TGF-β in Conditioned Media" by Garner and colleagues addresses an important gap in standardized analytical readouts for canine MSC secretome characterization. The concept of combining metabolite and protein outputs is interesting, and the manuscript is generally well-written and scientifically sound. However, in its current form it remains a preliminary feasibility study, and several central claims exceed what the data robustly support. In particular, the work does not yet establish a compact single-platform potency assay, the analytical specificity of the protein measurements is insufficiently demonstrated, and the statistical analysis is too limited for the interpretive scope. Substantial revision is therefore required.

MAJOR points
1.    The manuscript repeatedly presents the method as a "compact, single-platform HPLC potency panel" and implies that all four analytes are quantified from the same conditioned-medium sample in one unified workflow. This is not accurate. Adenosine and kynurenine were measured after 72 h in serum-containing medium, while IL-10 and TGF-Beta were measured after 48 h in serum-free medium. The study also employs different sample-preparation procedures, different mobile phases, different chromatographic programs, and two distinct columns. These should be described more precisely as a set of analyte-specific HPLC assays run on the same instrument. The title, abstract, introduction, discussion, and conclusions must be revised accordingly. The present wording overstates the degree of analytical integration and could mislead readers about the level of method harmonization achieved.
2.    The primary methodological concern is the analytical specificity of the IL-10 and TGF-beta measurements. Quantification of low-abundance proteins in complex conditioned medium by reverse-phase HPLC with UV detection at 214 nm, relying solely on retention-time matching, does not adequately establish analyte identity. Conditioned medium contains numerous UV-absorbing proteins and peptides, yet the manuscript provides no orthogonal evidence (e.g., spike-recovery experiments in matrix, peak-purity assessment, chromatographic resolution from matrix peaks, fraction collection with confirmation, immunological cross-check, or mass-spectrometric verification) that the assigned peaks are uniquely attributable to the target analytes. This is especially critical for TGF-beta, where the assay does not clarify whether active or total TGF-beta is measured or how latent complexes are handled. Without such evidence, the protein component of the panel remains insufficiently validated.
3.    Analytical validation is incomplete for a study that emphasizes standardization and potential use as a potency assay. While linearity and approximate LOQs are reported, key parameters such as selectivity, matrix effects, extraction recovery, repeatability, intermediate precision, carryover, retention-time reproducibility, robustness, and system-suitability criteria are not addressed. Representative chromatograms (standards, blank medium, conditioned medium, and spiked conditioned medium) should be included. In addition, the IL-10 and TGF-β concentrations in Table 2 fall below the stated 5 ng/mL LOQ unless they have been corrected for pre-concentration; this relationship must be explicitly clarified.
4.    Statistical treatment is insufficient. With only three donor-derived MSC lines and paired licensed/unlicensed conditions, no inferential statistics, effect sizes, or confidence intervals are presented. The abstract and discussion should clearly state that the dataset is underpowered for strong claims and frame the work as a feasibility study rather than evidence of robust manufacturing control or assay readiness.
5.    Use of the term “potency assay” is premature. The panel measures secreted analytes that may relate to MSC immunomodulation, but no data demonstrate that the HPLC outputs predict or correlate with a relevant biological function (e.g., T-cell suppression or macrophage polarization). Without such functional correlation, the work is better described as an analytical characterization panel or candidate surrogate readout. The title and conclusions should be revised unless functional data are added.
6.    The framing around canine IMHA is overly assertive relative to the evidence provided. IMHA serves as the motivating clinical context, but the manuscript contains no patient samples, disease-specific model, or link between assay output and therapeutic performance. The translational discussion should be shortened and presented more cautiously.

MINOR points
7.    Clarify whether the mid-range TGF-beta calibration standard (14 % back-calculation error) overlaps with the effective injected concentration range of the study samples after pre-concentration.
8.    Table 1 lists protein calibration ranges starting at 0.5 ng/mL, yet the text states an LOQ of 5 ng/mL. Distinguish clearly between the lowest calibrator and the validated LOQ.
9.    Replace or supplement Figure 1 with donor-level paired plots. Bar graphs with only three biological replicates obscure data structure and hinder evaluation of paired consistency.
10.    Provide additional donor and manufacturing metadata for the three MSC lines (donor characteristics, passage/culture status consistency, and whether analyses were performed in a single batch or across multiple days).

Author Response

Please see the attachment. 

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

The authors presented a platform for determining mesenchymal stem cell activity using HPLC. The language of the manuscript is very difficult to understand.

Unfortunately, the presented results were not validated using alternative approaches.

The Introduction section needs to be completely rewritten. It is very long and does not allow for a focus on the study. Please include a drawing of all the described pathways and interactions.

The Results section does not include a single chromatogram. Without them, all the presented data appears unsubstantiated.

Line 41. "Historically, MSCs have been utilized for their ability to differentiate into chondrocytes and osteocytes within orthopedic medicine." This statement is controversial. Please refrain from unsubstantiated statements.

Line 45. The relationship between autoimmune diseases, veterinary medicine, and MSC regulation is unclear.

Line 51. "These effects are mediated through both cell contact-dependent interactions and secreted factors." This is unclear. Describe these interactions in more detail.

Line 58. Explain "proposed dominant mechanisms and preferred potency markers."

Line 61. What is "inflammatory priming"?

Line 66. "IDO." Please explain. Provide the reaction scheme.

Lines 114-115. "Rather than interpreting these pathways in isolation, a clinically relevant MSC potency readout should reflect the coordinated, synergistic nature of MSC immune regulation." A synergistic effect can also be determined by multiplex analysis.

Line 122. "Secretome fingerprint." This term is unclear.

Line 124. It is unclear why IMHA is described in the context of MSC and cytokine profiling.

Lines 133-139. This text is unclear.

Line 144. The terms "licensing" and "manufacturing lever" are unclear. Please use generally accepted scientific terminology.

Line 156. Please indicate how the MSCs were obtained. Describe all samples analyzed in more detail. How was it demonstrated that the cells obtained were MSCs? Provide histological photographs.

Line 157. Please provide chromatograms and calibration curves.

Table 1. What are "Points," "Slope," and "Intercept"?

Line 168. Explain Licensed vs. Unlicensed Conditioned Media.

Line 171. Please provide chromatograms for analyte comparison.

Line 221. Explain "biological and matrix-driven variability."

Author Response

Please see the attachment. 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed the reviewer's comments.

Author Response

We thank you for your careful evaluation of the revised manuscript and are pleased that all concerns have been addressed to your satisfaction. No further changes were made in response to this review.

Reviewer 2 Report

Comments and Suggestions for Authors

 

The authors have addressed most of the concerns in a transparent and thoughtful way. The manuscript is substantially improved.

there are three remaining minor issues before recommending acceptance.

Title inconsistency
The title uses "Mesenchymal Stromal Cells" but the field standard (ISCT) recommends "Mesenchymal Stromal/Stem Cells" to reflect current nomenclature. Please update the title to match the terminology used in the rest of the manuscript.

Missing documentation of the IL-10 calibration outlier
In your cover letter you noted that one IL-10 standard (25 ng/mL) showed a 14% back-calculation error. This is not clearly stated in the manuscript. Please add a brief note in Table 1 or in Section 2.1 acknowledging this deviation and stating that this calibration point will be revisited in future method refinement.

Table 2 footnote clarity
The footnote symbols in Table 2 are slightly confusing. The dagger (†) is used for the sign test result, but the footnote text could be clearer. Please separate the sign test note from the exploratory protein note for readability.

Author Response

We thank you for your careful reading of the revised manuscript and your helpful final comments. All three points have been addressed as described below.

1. Comment: "Title inconsistency — The title uses 'Mesenchymal Stromal Cells' but the field standard (ISCT) recommends 'Mesenchymal Stromal/Stem Cells' to reflect current nomenclature. Please update the title to match the terminology used in the rest of the manuscript."

Response: The title has been corrected to "Mesenchymal Stromal/Stem Cells" to align with ISCT nomenclature (Viswanathan et al. 2019, ref. [13]) and the terminology used throughout the manuscript.

2. Comment: "Missing documentation of the IL-10 calibration outlier — In your cover letter you noted that one IL-10 standard (25 ng/mL) showed a 14% back-calculation error. This is not clearly stated in the manuscript. Please add a brief note in Table 1 or in Section 2.1 acknowledging this deviation and stating that this calibration point will be revisited in future method refinement."

Response: A note has been added to Section 2.1 specifying that the 25 ng/mL IL-10 calibration standard back-calculated at 14% deviation from nominal. The text now clarifies that this point was retained in the current calibration model and will be revisited in subsequent method refinement. A corresponding § footnote has been added to the IL-10 row in Table 1.

3. Comment: "Table 2 footnote clarity — The footnote symbols in Table 2 are slightly confusing. The dagger (†) is used for the sign test result, but the footnote text could be clearer. Please separate the sign test note from the exploratory protein note for readability."

Response: The combined footnote has been separated into two distinct notes: † for the sign test result and ‡ for the exploratory/semi-quantitative status of the protein measurements, improving readability.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors fixed their manuscript egregiously.

Author Response

We thank you for your careful evaluation of the revised manuscript and are pleased that all concerns have been addressed to your satisfaction. No further changes were made in response to this review.

Reviewer 4 Report

Comments and Suggestions for Authors

The new additions to the manuscript made a big difference. The quality of the paper had improved, and all my questions were addressed. No more comments.

 

Author Response

We thank you for your careful evaluation of the revised manuscript and are pleased that all concerns have been addressed to your satisfaction. No further changes were made in response to this review.