Paclitaxel Hypersensitivity: Is Titrated Dosing in Gynecologic Oncology Patients Necessary?

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript addresses the extremely interesting and important issue for practical oncology and chemotherapy: the influence of administration routes (specifically, dose titration) of hydrophobic, water-insoluble chemotherapeutic agents (using paclitaxel as an example) on the manifestation of hypersensitivity reactions. However, this manuscript requires significant improvement.

The work is purely descriptive and establishes a fairly clear conclusion: dose titration is necessary for the successful use of highly toxic anticancer combination drugs such as Taxol® (Cremophor® EL (Polyoxyl 35 castor oil) + paclitaxel) with minimal risk to patients' overall well-being. A very limited statistical sample was used, so to obtain a more accurate, generalized quantitative conclusion, the manuscript needs to be supplemented by a meta-analysis of the literature. Furthermore, the novelty and originality of the work under review are unclear to me, as the authors have already published similar data (https://doi.org/10.1200/JCO.2024.42.16_suppl.e23240).

The manuscript also requires careful editing and stylistic correction. For example, the captions to all the tables are completely uninformative; Table 1 (line 149) and Table 4 (line 188) contain a neologism ("Latinx") that is inappropriate for scientific literature, etc.

Author Response

Comment 1: The work is purely descriptive and establishes a fairly clear conclusion: dose titration is necessary for the successful use of highly toxic anticancer combination drugs such as Taxol® (Cremophor® EL (Polyoxyl 35 castor oil) + paclitaxel) with minimal risk to patients' overall well-being. A very limited statistical sample was used, so to obtain a more accurate, generalized quantitative conclusion, the manuscript needs to be supplemented by a meta-analysis of the literature. Furthermore, the novelty and originality of the work under review are unclear to me, as the authors have already published similar data (https://doi.org/10.1200/JCO.2024.42.16_suppl.e23240).

Response 1: Thank you for this observation. Yes, we have previously had an abstract accepted to the annual ASCO conference, but have not published this data in a journal. This is the experience of one institution which underwent realtime changes to the way that paclitaxel was administered; this is not a multi-institutional trial that accounts for thousands of patients receiving paclitaxel. We do discuss the limitations of our study in terms of sample size and overall limits of generalizability given that data was collected from just our own institution. 

Comment 2: The manuscript also requires careful editing and stylistic correction. For example, the captions to all the tables are completely uninformative; Table 1 (line 149) and Table 4 (line 188) contain a neologism ("Latinx") that is inappropriate for scientific literature, etc.

Response 2: The tables and figures were submitted for improvement through author services and changes were made using the provided service from MDPI. We have also edited the tables ourselves to include more appropriate scientific terminology and have added captions and descriptions to each table.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript evaluates whether titrated paclitaxel infusion reduces hypersensitivity reactions compared with non-titrated administration in gynecologic oncology patients. The topic is clinically relevant because paclitaxel remains a cornerstone therapy and infusion reactions still represent a common safety concern in oncology practice. The study takes advantage of an institutional protocol change to retrospectively compare outcomes before and after removal of titration. Overall the manuscript is clear and easy to follow, and the authors provide practical clinical observations that may be useful for infusion centers considering workflow modifications.

The study design is straightforward. The authors retrospectively analyzed 150 patients and 282 infusion visits and reported a higher rate of hypersensitivity reactions in the non-titrated group compared with the titrated group (20.8% vs 11.9%, RR 1.73). The effect appears more pronounced during the second infusion, where non-titrated administration showed a significantly higher reaction rate (22.4% vs 8.4%).These observations are interesting and potentially clinically meaningful because many centers have moved toward simplified infusion protocols to improve efficiency. The manuscript also provides useful descriptive information on timing of reactions and CTCAE grading.

There are however several issues that should be clarified. First, the retrospective design and institutional protocol change introduce potential confounding factors. Titration was discontinued at the same time that the tubing priming strategy also changed. As written, it is difficult to separate whether the observed difference is due to titration itself or to the absence of drug priming. This point deserves stronger discussion because the manuscript frequently interprets the results as an effect of titration alone.

Second, the definition of hypersensitivity reactions relies on EMR documentation and retrospective CTCAE grading. While understandable in this design, this approach may introduce misclassification bias. It would be helpful if the authors clarify how consistently infusion reactions were documented by nursing staff and whether standardized reaction reporting exists in the infusion unit.

Third, the statistical analysis is relatively simple for a dataset that includes repeated infusions from the same patient. Each infusion appears to be treated as an independent event, although some patients contributed two observations. A brief justification for this approach or a sensitivity analysis accounting for within-patient correlation would strengthen the analysis.

The manuscript is generally well written, although a few sections could be tightened. The introduction contains several paragraphs summarizing infusion practices that could be shortened. The discussion occasionally speculates beyond the presented data, particularly when interpreting priming practices or proposing optimized protocols.

Overall this is a useful observational study addressing a practical clinical question. With clarification of the methodological limitations and a more balanced interpretation of titration versus priming effects, the manuscript could make a valuable contribution to discussions on paclitaxel infusion protocols in gynecologic oncology.

Author Response

Comment 1: There are however several issues that should be clarified. First, the retrospective design and institutional protocol change introduce potential confounding factors. Titration was discontinued at the same time that the tubing priming strategy also changed. As written, it is difficult to separate whether the observed difference is due to titration itself or to the absence of drug priming. This point deserves stronger discussion because the manuscript frequently interprets the results as an effect of titration alone.

Response 1: This is a valid point. Given prior data regarding titration playing a larger role in HSRs regardless of priming, we suspect that the change in HSRs that we saw at our institution is a result of titration changes - this was added to the discussion.

Comment 2: Second, the definition of hypersensitivity reactions relies on EMR documentation and retrospective CTCAE grading. While understandable in this design, this approach may introduce misclassification bias. It would be helpful if the authors clarify how consistently infusion reactions were documented by nursing staff and whether standardized reaction reporting exists in the infusion unit.

Response 2: You are correct in identifying this - we did rely heavily on documentation from infusion center staff, as we were not present for these HSRs. However, all infusion reactions had dual documentation by both nursing staff and advanced practice providers/physicians. We do not have a standard reporting note type for HSRs in our EMR, although it would be a great idea for us to implement one. We do discuss these limitations in the discussion. 

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

Your study on paclitaxel titration and the incidence of hypersensitivity reactions (HSRs) is of great clinical relevance, particularly in demonstrating that titrated infusion significantly reduces these reactions, especially during the second administration.

However, the introduction seems to lack sufficient references to support the statements presented. I also would like to suggest an addition that could further strengthen the scientific rationale for your study, particularly regarding the transition to non-titrated protocols and the use of different priming techniques. The current text focuses heavily on traditional IgE-mediated pathways, but recent literature emphasizes the importance of non-IgE mechanisms, such as Complement Activation-Related Pseudoallergy (CARPA). This mechanism explains why reactions often occur upon the very first exposure, as the solvent (Cremophor EL) can directly trigger the complement cascade. This biological 'trigger' provides a robust scientific justification for your focus on titration and priming strategies, as these methods physically control the initial surge of the excipient.

Furthermore, the introduction does not clearly explain the scientific basis for why saline priming may be superior. Adding that saline priming creates an initial concentration gradient, thereby avoiding the immediate Cremophor EL peak that triggers the complement cascade, would lend much greater support to your hypothesis.

In my view, the manuscript’s main weaknesses relate to the methodology and the clarity/presentation of the results. Upon a detailed review of the Methods section, I noted that the data collection was strictly divided into two distinct timeframes (April 2022 to March 2023 for the titrated group and April 2023 to November 2023 for the non-titrated group). This raises a critical question regarding the consistency of the data presented in Table 1: How were patients classified if they began their treatment in March 2023 (titrated 1st dose) but completed it in April 2023 (non-titrated 2nd dose)?

If such patients were allocated to the 'Titrated' column in Table 1 based solely on their first exposure, the demographic characterization of that group becomes imprecise, as it overlooks the fact that the same individual also experienced the non-titrated protocol during their second visit. This temporal overlap suggests that while the infusion-based analysis in Table 2 remains robust, the patient-based analysis in Table 1 requires further clarification on how these transition cases were managed.

To address this gap and enhance the study's transparency, we strongly suggest the inclusion of a flowchart (diagram) detailing the patient journey from enrollment to final analysis. This would clearly visualize how many patients received both doses under the same protocol versus those who crossed the institutional policy change in April 2023. Without this clarification, the interpretation of Table 1 remains ambiguous, and the methodology may appear incomplete regarding the management of protocol transitions. Furthermore, providing this level of detail is essential to fully support the conclusions drawn from the comparative analysis.

Your proposal to prospectively evaluate titration rates and 'saline priming' is an important step toward refining paclitaxel administration and optimizing patient safety. I recognize that the focus of your work lies in the immediate safety of administration, effectively addressing HSRs which, as discussed, are often linked to the Cremophor EL vehicle and infusion rate, rather than directly to drug metabolism.

Nevertheless, I would like to suggest that future research considers the integration of pharmacogenetics in this context. Although pharmacogenetics is primarily used to predict long-term toxicities, such as neuropathy and myelosuppression, through the analysis of genes like CYP2C8 and CYP3A4, which influence paclitaxel metabolism, it would be valuable to explore whether any correlation exists, even if secondary, between an individual's metabolic profile and their susceptibility to HSRs. If the presentation of the results and the methodological description leave room for uncertainty, these aspects should be clarified and strengthened to ensure that the conclusions are fully supported and clearly understood by readers. Addressing these points would, in my view, significantly improve the manuscript and enhance its impact.

Kind Regards

Author Response

Comment 1: However, the introduction seems to lack sufficient references to support the statements presented. I also would like to suggest an addition that could further strengthen the scientific rationale for your study, particularly regarding the transition to non-titrated protocols and the use of different priming techniques. The current text focuses heavily on traditional IgE-mediated pathways, but recent literature emphasizes the importance of non-IgE mechanisms, such as Complement Activation-Related Pseudoallergy (CARPA). This mechanism explains why reactions often occur upon the very first exposure, as the solvent (Cremophor EL) can directly trigger the complement cascade. This biological 'trigger' provides a robust scientific justification for your focus on titration and priming strategies, as these methods physically control the initial surge of the excipient.

Furthermore, the introduction does not clearly explain the scientific basis for why saline priming may be superior. Adding that saline priming creates an initial concentration gradient, thereby avoiding the immediate Cremophor EL peak that triggers the complement cascade, would lend much greater support to your hypothesis.

Response 1: Thank you for these suggestions! We have included further information in the introduction regarding complement-mediated paclitaxel reactions (including references) and additionally included more information related to how this can be avoided with saline priming.

Comment 2: In my view, the manuscript’s main weaknesses relate to the methodology and the clarity/presentation of the results. Upon a detailed review of the Methods section, I noted that the data collection was strictly divided into two distinct timeframes (April 2022 to March 2023 for the titrated group and April 2023 to November 2023 for the non-titrated group). This raises a critical question regarding the consistency of the data presented in Table 1: How were patients classified if they began their treatment in March 2023 (titrated 1st dose) but completed it in April 2023 (non-titrated 2nd dose)?

If such patients were allocated to the 'Titrated' column in Table 1 based solely on their first exposure, the demographic characterization of that group becomes imprecise, as it overlooks the fact that the same individual also experienced the non-titrated protocol during their second visit. This temporal overlap suggests that while the infusion-based analysis in Table 2 remains robust, the patient-based analysis in Table 1 requires further clarification on how these transition cases were managed.

Response 2: We have now clarified that there were no patients who had crossover from titrated to non-titrated paclitaxel administration. All patients who had two treatments with paclitaxel either had both treatments titrated or both treatments non-titrated. 

Reviewer 4 Report

Comments and Suggestions for Authors

The study is retrospective and compares two time periods (before and after protocol change). Potential temporal confounders such as changes in clinical practice, nursing workflow, or patient characteristics should be discussed more clearly.

• The titrated group used paclitaxel-primed tubing, whereas the non-titrated group used diluent-primed tubing. Therefore, two variables changed simultaneously. The manuscript should clarify that the observed effect may partly reflect priming differences rather than titration alone.
• The manuscript sometimes uses hypersensitivity reactions (HSR) and infusion related reactions (IRR) interchangeably. Since the introduction distinguishes them, the authors should clearly define which events were included in the analysis.
• The statistical methods section is brief. The authors should clarify whether additional analyses (e.g., multivariable adjustment for confounders such as age, cancer type, or dosing schedule) were considered.
• The total number of reactions is relatively small. This limitation should be emphasized when interpreting subgroup analyses, particularly for first vs second infusion comparisons.
• The title could be more specific to reflect the study design and patient population.
• The abstract implies causality (“HSR incidence increased following cessation”). This should be phrased more cautiously to reflect an observational association.
• The Methods section could be improved by organizing it into clearer subsections (Study Design, Patient Population, Infusion Protocol, Outcome Definitions, Statistical Analysis).
• Tables could be improved by clearly reporting sample sizes and percentages in a consistent format.

Overall, this manuscript addresses a clinically relevant question regarding paclitaxel administration protocols. The findings suggest that titrated infusion may reduce hypersensitivity reactions. With clarification of methodological details and minor revisions, the manuscript would provide useful information for oncology infusion practices.

 

Author Response

Comment 1: The titrated group used paclitaxel-primed tubing, whereas the non-titrated group used diluent-primed tubing. Therefore, two variables changed simultaneously. The manuscript should clarify that the observed effect may partly reflect priming differences rather than titration alone.

Response 1: This is a valid observation that some of the other reviewers also noted. We have edited the manuscript to clarify this.

Comment 2: The manuscript sometimes uses hypersensitivity reactions (HSR) and infusion related reactions (IRR) interchangeably. Since the introduction distinguishes them, the authors should clearly define which events were included in the analysis.

Response 2: We have deleted all mention of IRR, as this study focused on HSRs.

Comment 3: The total number of reactions is relatively small. This limitation should be emphasized when interpreting subgroup analyses, particularly for first vs second infusion comparisons.

Response 3: We discussed this in the discussion portion of the manuscript.

Comment 4: The title could be more specific to reflect the study design and patient population.

Response 4: The title has been edited.

Comment 5: The abstract implies causality (“HSR incidence increased following cessation”). This should be phrased more cautiously to reflect an observational association.

Response 5: This has been edited to reflect that we are not implying causality given that this is an observational study.

Comment 6: The Methods section could be improved by organizing it into clearer subsections (Study Design, Patient Population, Infusion Protocol, Outcome Definitions, Statistical Analysis).

Response 6: We have edited the methods section to include clearer subsections.

Comment 7: Tables could be improved by clearly reporting sample sizes and percentages in a consistent format.

Response 7: The tables have been edited and the journal's editing service has been employed. Descriptions have been added to each table.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I have no more questions, but the manuscript needs to be brought into compliance with the journal's guidelines. Also, the neologism "Latinx" in Table 4 needs to be corrected.

Author Response

Comment 1: I have no more questions, but the manuscript needs to be brought into compliance with the journal's guidelines. Also, the neologism "Latinx" in Table 4 needs to be corrected.

Response 1: Table 4 has been edited. The manuscript has been submitted with the journal's template and all spelling/grammar has been assessed and addressed.