Two Comprehensive Liquid Chromatography High-Resolution Mass Spectrometry (UPLC-MS/MS) Multi-Methods for Real-Time Therapeutic Drug Monitoring (TDM) of Five Novel Beta-Lactams and of Fosfomycin Administered by Continuous Infusion
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis manuscript describes according to the authors one LC-MS method for the quantification of several antibiotics. Reading the method section of the manuscript it becomes obvious that the authors use two (2) different HPLC columns for the measurement of the different drugs, so that they effectively present two(2) different methods using the same mobile phases and the same preaalytical procedures. This fact must be clarified in the title of the manuscript and throughout the text.
In the last years several methods have been published describing the concomitant quantification of several beta-lactam drugs and beta-lactamase inhibitors. In the introduction part of the manuscript the authors should cite this papers and describe the advantages of their new methods.
Line 20: In the abstract the authors describe that standardized analytical methods for the quantification of different antibiotics. Unfortunately they do not refer within the manuscript to this and also do not describe standardized analytical methods.
Line 172: „the carry-over was evaluated soon after the injection of calibrator 6….“. What did they really do and what means soon?
Author Response
Response to Reviewer’s Comments (pharmaceutics-4037051; A comprehensive liquid chromatography high - resolution mass spectrometry (UPLC - MS/MS) multi-method for real-time therapeutic drug monitoring (TDM) of five novel beta-lactams and of fosfomycin administered by continuous infusion)
Dear Editor,
We would like to thank you for the opportunity to resubmit a revised version of this manuscript. We appreciated the reviewer’s constructive comments. All have been carefully considered and incorporated, where and whenever possible, in the revision. Furthermore, as as suggested we carefully reviewed the English language in order to improve the readability.
Our point-by-point responses are provided below.
Q= QUERY; A= ANSWER
Reviewer #1:
Q1. This manuscript describes according to the authors one LC-MS method for the quantification of several antibiotics. Reading the method section of the manuscript it becomes obvious that the authors use two (2) different HPLC columns for the measurement of the different drugs, so that they effectively present two(2) different methods using the same mobile phases and the same preanalytical procedures. This fact must be clarified in the title of the manuscript and throughout the text.
A1. Thank you for this clarification. We corrected both the title and the text of the manuscript accordingly.
Q2. In the last years several methods have been published describing the concomitant quantification of several beta-lactam drugs and beta-lactamase inhibitors. In the introduction part of the manuscript the authors should cite this papers and describe the advantages of their new methods.
A2 Thank you for this suggestion. We added a comparison with previously published articles in the Introduction section (refer to Lines 91-97).
Q3. Line 20: In the abstract the authors describe that standardized analytical methods for the quantification of different antibiotics. Unfortunately they do not refer within the manuscript to this and also do not describe standardized analytical methods.
A3. We thank the reviewer for this comment, allowing us to clarify this misunderstanding. In the abstract, we mentioned that currently no standardized analytical methods exist for the simultaneous quantification specifically of the novel different antibiotics to which we referred, thus the aim of our study is to fill the gap of this unmet need.
Q4. Line 172: „the carry-over was evaluated soon after the injection of calibrator 6….“. What did they really do and what means soon?
A4. We thank the reviewer for this comment, allowing us to better clarify this issue. The carryover was assessed by injecting a blank sample immediately after finishing the run of the highest calibrator (calibrator 6) and it was considered negligible whenever the signal intensity was less than 20% of that of the LOQ, as recommended by the EMA guidelines. We revised this sentence in order to improve clarity (refer to Line 204-207).
Reviewer 2 Report
Comments and Suggestions for AuthorsDear Editor,
Thank you for the opportunity to review this manuscript. The authors present a multi-analyte UPLC-qTOF-MS/MS method designed for real-time therapeutic drug monitoring (TDM) of several novel β-lactams, β-lactamase inhibitors, and fosfomycin. The topic is clinically relevant and fits well within the scope of Pharmaceutics, especially considering the increasing demand for rapid, accurate, and clinically implementable TDM platforms for critically ill patients.
The manuscript is generally well written, and the analytical procedure appears technically sound. However, several major scientific and methodological issues substantially limit the clinical applicability and robustness of the proposed assay. Additional clarification and experimental evidence are needed before the manuscript can be considered for publication. My specific concerns are summarized below.
Major Concerns
- The authors propose the method for “real-time TDM“. However, how long does it take from sample preparation to obtaining the test results?
- The calibration ranges are tailored exclusively to continuous infusion (CI) regimens. Can this analytical method be applied to other dosing regimens? Have you considered how to address the potentially lower or higher drug concentrations that may occur with alternative administration strategies?
- The stability data indicate substantial degradation for several analytes following freeze–thaw cycles. Given that TDM samples are frequently transported, stored, or reprocessed, this finding has significant implications for laboratory feasibility. The manuscript does not provide practical guidance on:
- recommended maximum time-to-analysis,
- potential stabilizing conditions,
- handling recommendations for remote or batch-processing laboratories.
Without such information, the assay’s implementation in routine TDM remains unclear.
4. Table 8 reports matrix effects exceeding ±30% for several analytes. Although isotopically labeled internal standards were employed, the manuscript does not provide:
-
- IS-normalized matrix factor values,
- CV% across different plasma sources (EMA requirement <15%),
- pre- and post-correction data demonstrating successful compensation.
Given the heterogeneity of critically ill patient plasma, more robust matrix-effect characterization is required to support the method’s reliability.
As a suggestion, presenting the matrix effect and extraction recovery in Table 8 by listing each drug with its low-, medium-, and high-concentration levels grouped together, would make the table easier to read.
5. Several analytes elute within narrow retention windows (e.g., RT 3.2–3.9 min). Has the potential impact of co-administered drugs—commonly prescribed in critically ill populations—on chromatographic separation or MRM signal specificity been assessed, particularly regarding peak co-elution or ion suppression/enhancement?
6. Are the chromatograms shown in Figure 1a and 1b derived from real clinical samples or from spiked samples?
- It appears that non-weighted linear regression was used to construct calibration curves. For assays with broad dynamic ranges and low-end sensitivity requirements, EMA guidelines typically recommend 1/x or 1/x² weighting to ensure accuracy at the LOQ. The authors should justify the regression model used and provide evidence of improved or comparable accuracy relative to weighted models.
- Several β-lactamase inhibitors demonstrate LOQs (e.g., tazobactam 3.13 µg/mL, avibactam 1.56 µg/mL) that exceed typical trough concentrations in patients with high renal clearance or at the end of dosing intervals. Additional discussion is needed regarding the clinical relevance of the LOQs reported.
Minor Concerns
- Inconsistency of decimal notation (comma vs. period) should be standardized prior to publication.
- Inclusion of chromatograms from clinical samples would strengthen evidence of robustness.
- The manuscript would benefit from a clearer comparison against existing single-analyte LC-MS/MS methods from the same group, highlighting improvements in workflow efficiency or turnaround time.
- Some typographical errors and formatting inconsistencies should be corrected during revision.
The manuscript is generally well written, and the analytical procedure appears technically sound. However, several major scientific and methodological issues substantially limit the clinical applicability and robustness of the proposed assay.
Author Response
Response to Reviewer’s Comments (pharmaceutics-4037051; A comprehensive liquid chromatography high - resolution mass spectrometry (UPLC - MS/MS) multi-method for real-time therapeutic drug monitoring (TDM) of five novel beta-lactams and of fosfomycin administered by continuous infusion)
Dear Editor,
We would like to thank you for the opportunity to resubmit a revised version of this manuscript. We appreciated the reviewer’s constructive comments. All have been carefully considered and incorporated, where and whenever possible, in the revision. Furthermore, as as suggested we carefully reviewed the English language in order to improve the readability.
Our point-by-point responses are provided below.
Q= QUERY; A= ANSWER
Reviewer #2:
Thank you for the opportunity to review this manuscript. The authors present a multi-analyte UPLC-qTOF-MS/MS method designed for real-time therapeutic drug monitoring (TDM) of several novel β-lactams, β-lactamase inhibitors, and fosfomycin. The topic is clinically relevant and fits well within the scope of Pharmaceutics, especially considering the increasing demand for rapid, accurate, and clinically implementable TDM platforms for critically ill patients.
The manuscript is generally well written, and the analytical procedure appears technically sound. However, several major scientific and methodological issues substantially limit the clinical applicability and robustness of the proposed assay. Additional clarification and experimental evidence are needed before the manuscript can be considered for publication. My specific concerns are summarized below.
We thank the reviewer for appreciating our manuscript.
Major concerns
Q1. The authors propose the method for “real-time TDM“. However, how long does it take from sample preparation to obtaining the test results?
A1. We thank the reviewer for this comment, allowing us to better clarify this important issue. The turnaround time elapsing between preparing sample preparation and providing TDM result is approximately of 3 hours. Consequently, the expert clinical pharmacological advice may be performed in real-time in the same day in which the sample is collected. We better clarified this issue in the Discussion section (refer to Line 391-394).
Q2. The calibration ranges are tailored exclusively to continuous infusion (CI) regimens. Can this analytical method be applied to other dosing regimens? Have you considered how to address the potentially lower or higher drug concentrations that may occur with alternative administration strategies?
A2. We thank the reviewer for this relevant comment, allowing us to better clarify this issue. It should be noticed that at our center, CI of traditional and novel beta-lactams is always comprehensively adopted hospitalwide, as recently reported (please refer to doi: 10.1097/FTD.0000000000001334). The rationale is that of maximizing the likelihood of attaining an aggressive PK/PD target of beta-lactams, namely a target that was recently shown to be associated with significantly higher clinical cure and microbiological eradication rates, and with lower resistance development (please refer to doi: 10.1186/s13054-024-04911-5). Obviously, this analytical method could be applied also to other modes of administering the dosing regimens (namely by intermittent and/or extended infusion), provided that the linearity intervals be adjusted according to the respective ranges of the expected concentrations. Specifically, if measuring low trough concentrations compared to those adopted for continuous infusion, the caliberation curve should be tailored accordingly in order to maintain a proper linearity range and avoiding the risk of inaccuracy related to our calibration range. Conversely, if measuring high peak concentrations, samples should be diluted during the preparatory phase so that the concentrations would fall within the validated linearity range and then corrected by the dilution factor. We added this issue among limitations of our manuscript (refer to Line 435-447).
Q3. The stability data indicate substantial degradation for several analytes following freeze–thaw cycles. Given that TDM samples are frequently transported, stored, or reprocessed, this finding has significant implications for laboratory feasibility. The manuscript does not provide practical guidance on:
- a) recommended maximum time-to-analysis,
- b) potential stabilizing conditions,
- c) handling recommendations for remote or batch-processing laboratories.
Without such information, the assay’s implementation in routine TDM remains unclear.
A3. We thank the reviewer for these important comments, allowing us to better discuss these relevant issues. Stability data indicate that most analytes may withstand up to 2 or 3 freeze/thaw cycles, thus allowing the delivery, storage, and reprocessing of samples. In case of unstable analytes exhibiting a withstand of maximum 1 freeze/thaw cycle, it is recommended to analyze them immediately after collection. Overall, in order to maximize the stability of analytes and avoiding substantial degradation, the following recommendations should be implemented:
- a) samples should be processed and analyzed on the same day of collection, taking care to keep the blood tube at +4°C up to processing;
- b) whether processing cannot be carried out in the same day or delivery is from a remote hospital, blood samples should be centrifuged and the supernatant stored in dry ice up to processing, so that may undergo maximum one freeze/thaw cycle.
We added all the required information in the revised version (refer to Line 421-429).
Q4. Table 8 reports matrix effects exceeding ±30% for several analytes. Although isotopically labeled internal standards were employed, the manuscript does not provide:
IS-normalized matrix factor values,
CV% across different plasma sources (EMA requirement <15%),
pre- and post-correction data demonstrating successful compensation.
Given the heterogeneity of critically ill patient plasma, more robust matrix-effect characterization is required to support the method’s reliability.
As a suggestion, presenting the matrix effect and extraction recovery in Table 8 by listing each drug with its low-, medium-, and high-concentration levels grouped together, would make the table easier to read.
A4. Thank you for this suggestion. We implemented the values of the IS normalized matrix effect in Table 8, in order to report the improvement of the signal correction. Regarding data coming from different plasma sources, unfortunately we did not find any commercial formulation of plasma reproducing conditions similar to those of the critically ill patients, considering that commercial drug-free human plasma is usually collected from healthy volunteers. However, it should not be overlooked that critically ill patients may exhibit a wide intra- and inter- individual variability due to peculiar pathophysiological alterations potentially resulting in a large heterogeneity of plasma composition between each other. This may make challenging and inappropriate using plasma collected from healthy subjects for method validation.
Q5. Several analytes elute within narrow retention windows (e.g., RT 3.2–3.9 min). Has the potential impact of co-administered drugs—commonly prescribed in critically ill populations—on chromatographic separation or MRM signal specificity been assessed, particularly regarding peak co-elution or ion suppression/enhancement?
A5. We thank the reviewer for this comment, allowing us to better clarify this issue. Although several analytes may have similar retention times, this does not represent a clinical concern in practice, because agents like meropenem/vaborbactam, ceftazidime/avibactam, ceftolozane/tazobactam, and cefiderocol are always used one at a time in a patient. We better clarify this aspect in the Discussion section (refer to Line 384-391).
Q6. Are the chromatograms shown in Figure 1a and 1b derived from real clinical samples or from spiked samples?
A6. We thank the reviewer for this comment, allowing us to better clarify this issue. The chromatograms shown in Figures 1a and 1b refer to samples spiked at the detection limit of the method. We specified this issue in the Figure 1 legend.
Q7. It appears that non-weighted linear regression was used to construct calibration curves. For assays with broad dynamic ranges and low-end sensitivity requirements, EMA guidelines typically recommend 1/x or 1/x² weighting to ensure accuracy at the LOQ. The authors should justify the regression model used and provide evidence of improved or comparable accuracy relative to weighted models.
A7. We thank the reviewer for this comment, allowing us to better clarify this issue. We used a linear regression with 1/x weighting for determining the calibration curves, as recommended by the EMA guidelines. We better specified this issue in the Methods section (refer to Line 212) and in the Table 6 legend (refer to Line 321).
Q8. Several β-lactamase inhibitors demonstrate LOQs (e.g., tazobactam 3.13 µg/mL, avibactam 1.56 µg/mL) that exceed typical trough concentrations in patients with high renal clearance or at the end of dosing intervals. Additional discussion is needed regarding the clinical relevance of the LOQs reported.
A8. We thank the reviewer for this relevant comment, allowing us to better clarify this important issue. We agree with the fact that the LOQs reported for some beta-lactamase inhibitors may potentially exceed the typical trough concentrations particularly in patients showing augmented renal clearance. However, it should be noticed that administering the beta-lactam/beta-lactamase inhibitor combinations by continuous infusion may greatly reduce the risk of having steady-state concentrations of beta-lactamase inhibitors below the LOQs. To this regard, our recent study conducted among 263 patients receiving a TDM-guided therapy with CI ceftazidime-avibactam and ceftolozane-tazobactam (please refer to doi: 10.1097/FTD.0000000000001334), showed that median avibactam and tazobactam steady-state concentrations were as high as 10.25 mg/L (IQR 5.33-16.78 mg/L) and 10.6 mg/L (IQR 6.5-17.6 mg/L), respectively. Only 1.7% and 3.9% of avibactam and tazobactam concentrations were below the LOQs. We better discuss this relevant topic in the Discussion section (refer to Line 447-453).
Minor Concerns
Q1. Inconsistency of decimal notation (comma vs. period) should be standardized prior to publication.
A1. Thank you for your suggestion. We standardized decimal notation throughout the text.
Q2. Inclusion of chromatograms from clinical samples would strengthen evidence of robustness.
A2. Thank you for your suggestion. We included chromatograms coming from clinical samples for each analyte (refer to Figure 2/Line 286 and 311).
Q3. The manuscript would benefit from a clearer comparison against existing single-analyte LC-MS/MS methods from the same group, highlighting improvements in workflow efficiency or turnaround time.
A3. Thank you for this relevant suggestion. We compared our multi-method approach with existing single-analyte methods in order to highlights the points of strengthen of our approach in the Discussion section (refer to Line 394-401).
Q4. Some typographical errors and formatting inconsistencies should be corrected during revision.
A4. Thank you for your suggestion. We carefully revised the text in order to correct the different typographical errors and formatting inconsistencies.
Reviewer 3 Report
Comments and Suggestions for AuthorsThanks for the efforts done by the authors in this work. I have some minor comments as below:
1. Section 2.1, line 97: Mention City & Country for the company where the authors got the internal standards
2. Why are QC samples prepared in MilliQ water and not in plasma?
3. It was mentioned in section 2.2 that the calibrators were prepared in MilliQ water, whereas, in line 118, it was mentioned that “Calibrators were obtained by spiking drug-free plasma ….”. Please clarify
4. There is a typo error at line 174
5. Line 173: Please mention reference “… was considered negligible if the signal was lower than 20% of the method’s LOQ”
6. For Meropenem borderline stability at one cycle, how can the authors interpret this behavior.
Author Response
Response to Reviewer’s Comments (pharmaceutics-4037051; A comprehensive liquid chromatography high - resolution mass spectrometry (UPLC - MS/MS) multi-method for real-time therapeutic drug monitoring (TDM) of five novel beta-lactams and of fosfomycin administered by continuous infusion)
Dear Editor,
We would like to thank you for the opportunity to resubmit a revised version of this manuscript. We appreciated the reviewer’s constructive comments. All have been carefully considered and incorporated, where and whenever possible, in the revision. Furthermore, as as suggested we carefully reviewed the English language in order to improve the readability.
Our point-by-point responses are provided below.
Q= QUERY; A= ANSWER
Reviewer #3:
Thanks for the efforts done by the authors in this work. I have some minor comments as below:
We thank the reviewer for appreciating our manuscript.
Q1. Section 2.1, line 97: Mention City & Country for the company where the authors got the internal standards.
A1. Thank you for this suggestion. We added City and Country of the company accordingly (refer to Line 109).
Q2. Why are QC samples prepared in MilliQ water and not in plasma?
A2. Thank you for your comment, allowing us to better clarify this issue. We unfortunately performed a typo, namely the QC samples were effectively prepared in plasma. We revised the text accordingly (refer to Line 115-118).
Q3. It was mentioned in section 2.2 that the calibrators were prepared in MilliQ water, whereas, in line 118, it was mentioned that “Calibrators were obtained by spiking drug-free plasma ….”. Please clarify
A3. Thank you for your comment, allowing us to better clarify this issue. As reported in comment No. 2, unfortunately we performed a typo, considering that the calibrators used for developing the calibration curves were prepared in MilliQ water at different concentrations as serial dilutions from the concentrated stock solution, spiking drug-free plasma in each of them.
Q4. There is a typo error at line 174
A4. Thank you for your suggestion. We revised accordingly (refer to Line 209).
Q5. Line 173: Please mention reference “… was considered negligible if the signal was lower than 20% of the method’s LOQ”.
A5. Thank you for your comment. We added the specific reference as suggested (refer to Reference No.31).
Q6. For Meropenem borderline stability at one cycle, how can the authors interpret this behavior.
A6. We thank the reviewer for this important comment. It is well-known that meropenem is highly sensitive to water, pH, temperature, and light. Unfortunately, all of these may cause the hydrolysis and opening of the b-lactam ring. This may explain why freezing/thawing is a major process favoring its degradation, so that it would be much reliable analyzing samples immediately after collection. We discussed this issue in the Discussion section (refer to Line 423-429).
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe questions of the reviewer have been answered satisfactory. I have no further comments.
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for the replies. No more comments.
