Rescue Therapy for Supratherapeutic Concentrations of Calcineurin Inhibitors Using Potent Cytochrome P450 Inducers

Round 1

Reviewer 1 Report

Dear Authors,

I read with interest your article that explores drugs that boost specific enzymes (CYP inducers) as potential treatments for CNI (tacrolimus and ciclosporin) overdoses. Data analysis from the WHO database and case studies showed rare cases of CNI overdoses, often due to interactions with other drugs. Review of published cases where CYP inducers like phenobarbital, phenytoin, and rifampicin were used showed varied success in treating CNI overdoses.

General concept comments

Overall, the subject is of interest, but I have several questions regarding the methodology.

First regarding paragraph 3.2:

Could you clarify the search done in Vigibase. You correctly define what an SMQ is in MedDRA, a set of terms. Ciclosporin, tacrolimus, overdose (correctly mentioned as a PT) and drug intoxication (a LLT) are not SMQs, these are single active ingredients and reaction terms. Overdose does include the LLTs that are underneath it like overmedication, acute overdose like all PTs in MedDRA. Drug intoxication is an LLT. In the same group of LLTs there is for example “drug toxicity”, “acute drug intoxication” and even “cyclosporine toxicity”. Could you explain why these terms were not included in your search ? Would you have extracted more cases to analyse ?

Regarding paragraph 3.3. : was the aim of the search to be systematic ? If so, could you provide the exact method of combination of the search words and maybe a flow diagram of extracted studies.

Regarding conclusion : Maybe you could slip a word on prevention of the interaction in the patients with drug intoxications that would avoid the adverse reactions and having to treat them with other drugs.

Regarding references(1, 2, 3, 32 and 33) : I am not sure that citing global databases is useful. For exemple, reference 32 does not explain what an SMQ is .

Specific comments

Throughout the text, the denomination VigiBase, WHO global pharmacovigilance database and WHO pharmacovigilance database are mentioned. Being the same “entity”, it would be easier if the same word was used throughout the document.

Line 178 : 4.1. Demographics of CNI overdose and intoxication…. In Vigibase ? I would add this to the title.

Line 179 : “Out of 143’710 documented ICSRs since January 1992”. What is this number ? It is not the result of your search “Ciclosporin, tacrolimus, overdose and drug intoxication” because you mention afterwards that you have 405 cases. Could you explain what it is ?

Line 191 : same comment as for line 178

Line 210-211 : do you have an idea of the doses ? Would that explain the difference ?

Line 202 : same comment as for line 178

Line 221 : The references of these 24 articles seem to be missing

Line 302-304 : underreporting exists in all pharmacovigilance data, not only because there was a bias with fewer reports from “less developed” countries (the correct term would be developing and least developed countries). There is no way to conclude from the data extracted from Vigibase that “very few cases of overdose culminate in intoxication”.

Line 317-319 : this seems weird because with older patients, polypharmacy increases and so do drug-drug interactions. For example, in the article from Amkreutz that you cite, the mean age of the patients in the study was 48.6 +/- 13.4, so no old patients, allowing to conclude that DDIs occur in patients between 45-64 y.o.

Line 332 : I would avoid using “side” effect

Line 361 : “case safety report” can be replaced by ICSR

Line 361-362: there are…. no cases where the treatment is described in the database.

Kind regards

Author Response

Response to Reviewer 1

Thank you very much for taking the time to review this manuscript. Please find as outlined below the responses to your comments. The corresponding revisions are highlighted in the re-submitted manuscript (version 2).

General impression of reviewer:

I read with interest your article that explores drugs that boost specific enzymes (CYP inducers) as potential treatments for CNI (tacrolimus and ciclosporin) overdoses. Data analysis from the WHO database and case studies showed rare cases of CNI overdoses, often due to interactions with other drugs. Review of published cases where CYP inducers like phenobarbital, phenytoin, and rifampicin were used showed varied success in treating CNI overdoses. Overall, the subject is of interest, but I have several questions regarding the methodology.

General concept comments:

Paragraph 3.2

Comment: Could you clarify the search done in Vigibase. You correctly define what an SMQ is in MedDRA, a set of terms. Ciclosporin, tacrolimus, overdose (correctly mentioned as a PT) and drug intoxication (a LLT) are not SMQs, these are single active ingredients and reaction terms. Overdose does include the LLTs that are underneath it like overmedication, acute overdose like all PTs in MedDRA. Drug intoxication is an LLT. In the same group of LLTs there is for example “drug toxicity”, “acute drug intoxication” and even “cyclosporine toxicity”. Could you explain why these terms were not included in your search? Would you have extracted more cases to analyse?

Response:

Thank you for the explanatory comments on SMQs. (L. 145,146)

The active ingredients “ciclosporin” and “tacrolimus” were applied against the search queries: "overdose PT" and "drug intoxication LLT" (as reaction terms) due to the following reasons:

• The only PT terms related to intoxication in the database were: “salt intoxication”, “water intoxication” and “calcium intoxication”. It would have been optimum if “drug intoxication PT” was applicable in the database.
• In order to avoid case duplications from the various reaction terms, we decided to use the most reliable term that shows case details in line with supratherapeutic concentrations of CNIs. Although “cyclosporine toxicity” yielded relatively more results, these cases were reported due to suspected adverse reactions, which were not explicitly related to increased cyclosporine blood concentrations. In addition, there was no reaction term for “tacrolimus toxicity” in the database. The case details derived from the term “drug intoxication LLT” described cases with increased concentrations of CNIs.
• Due to the above considerations, “Drug intoxication LLT” was the most suitable term. Kindly find below a summary of the search results:

Paragraph 3.3

Comment: Was the aim of the search to be systematic? If so, could you provide the exact method of combination of the search words and maybe a flow diagram of extracted studies?

Response:

Yes, we used Boolean operators with the aid of the following key words for the search: “calcineurin inhibitors”, “tacrolimus”, “ciclosporin”, “cytochrome 3A4 inducers”, “rifampicin”, “phenobarbital”, “carbamazepine”, “phenytoin”, “management of acute toxicity, overdose or intoxication” and “pharmacoenhancement”.

The flow diagram below has been added to the manuscript as figure 8:

Regarding conclusion:

Comment: Maybe you could slip a word on prevention of the interaction in the patients with drug intoxications that would avoid the adverse reactions and having to treat them with other drugs.

Response:

We agree and have emphasized on this point accordingly (L. 375-379)

“The relatively high occurrence of polypharmacy in organ transplant patients under treatment with CNIs require that health care providers consider drug-drug interactions as potential causes of severe acute CNI intoxication. In this category of patients, it is highly recommended to critically review and consider dose adjustment of relevant concomitant medications to prevent the need of using other drugs to reverse toxicity.”

Regarding references:

Comment: (1, 2, 3, 32 and 33), I am not sure that citing global databases is useful. For example, reference 32 does not explain what an SMQ is.

Response:

The clinical pharmacology of the CNIs and incidences of adverse reactions based on previous pharmacoepidemiological studies are stated in these databases (1, 2, 3).  

The specific link in reference 32 in which SMQs are explained has been added (https://www.meddra.org/how-to-use/tools/smqs)

Specific comments:

Comment: Throughout the text, the denomination VigiBase, WHO global pharmacovigilance database and WHO pharmacovigilance database are mentioned. Being the same “entity”, it would be easier if the same word was used throughout the document.

Response:

After mentioning VigiBase™ as the WHO global pharmacovigilance database for the first time in Line 28, VigiBase™ is now maintain throughout the manuscript. (L. 90, 284, 311, 364)

Line 178, 4.1.

Demographics of CNI overdose and intoxication…. In Vigibase? I would add this to the title.

Response:

Yes, in VigiBase™. The title has been updated accordingly (L. 166)

Line 179:

“Out of 143’710 documented ICSRs since January 1992”. What is this number? It is not the result of your search “Ciclosporin, tacrolimus, overdose and drug intoxication” because you mention afterwards that you have 405 cases. Could you explain what it is?

Response:

“143,710” is the cumulative number of suspected adverse reactions regarding ciclosporin and tacrolimus, out of which 405 were documented as overdose (375) and intoxication (30). (L. 167 updated)

Line 191:

Same comment as for line 178

Line 210-211:

Do you have an idea of the doses? Would that explain the difference?

Response:

Unfortunately, the doses in majority of the cases were not documented in the VigiLyze case details. Yes, we agree that this would have provided more information to explain the difference.  

Line 202:

Same comment as for line 178

Line 221:

The references of these 24 articles seem to be missing

Response:

The articles can be found in reference: 13-23, 26, 53-59 and 64-67 (highlighted). These are also indicated in the supplementary table. 

Line 302-304:

Underreporting exists in all pharmacovigilance data, not only because there was a bias with fewer reports from “less developed” countries (the correct term would be developing and least developed countries). There is no way to conclude from the data extracted from VigiBase that “very few cases of overdose culminate in intoxication”.

Response:

The statements have been revised using the terms “developing and least developed countries” (L. 316-317).

The statement regarding “culmination of overdose in intoxication” has been removed. (L. 312-314)

Line 317-319:

This seems weird because with older patients, polypharmacy increases and so do drug-drug interactions. For example, in the article from Amkreutz that you cite, the mean age of the patients in the study was 48.6 +/- 13.4, so no old patients, allowing to conclude that DDIs occur in patients between 45-64 y.o.

Response:

Thank you for pointing this out. We agree with this comment. Polypharmacy generally increases with increasing age in the overall patient population. However, majority of transplant patients who receive immunosuppressants fall within the age group 45 – 64 years.  

 Line 332:

I would avoid using “side” effect

Response: “side effects” has been removed from the statement (L. 338).

Line 361:

“Case safety report” can be replaced by ICSR

Response:

After defining ICSR as case safety report for the first time in Line 38, ICSR is now maintain throughout the manuscript. (L. 317, 364)

Line 361-362:

There are…. no cases where the treatment is described in the database.

Response:

Yes, although details on treatment are not described in the VigiLyze case details, in majority of cases, immediate drug withdrawal or dose reduction was undertaken. It would have been interesting to have further information regarding treatment to compare the outcomes after drug withdrawal or dose reduction alone versus drug withdrawal or dose reduction with specific treatment strategies such as pharmacoenhancement. 

“Treatments” has been changed to “actions taken” (L. 201, 212) 

Reviewer 2 Report

The manuscript entitled "Rescue Therapy for Supratherapeutic Concentrations of Calcineurin Inhibitors Using Potent Cytochrome P450 Inducers" addresses the unintentional CNI overexposure, a serious risk in immunosuppressed inpatients and outpatients with severe nephrological, neurological and immunological ADRs. The provision of a well-evaluated and fast rescue intervention that does not pose additional hazards may be of benefit to these patients. With reference to a case report, the authors aim to review the available data from VigiBase™ and the literature on this topic to determine the most effective and safe countermeasure.
However, there are some very important aspects that need to be considered and further clarified in the Results, Discussion and Conclusions section.
1. Most important missing: Laboratory data on 19 / 20 Dec.
2. Liver function may already have been impaired by previous drugs, e.g. vancomycin, etc., so that the dose of tacrolimus should have been reduced already before fluconazole was added because of reduced hepatic metabolic capacity, itself requiring caution with dosage in this case. Please refer to
the cumulative risk of liver injury by the predisposing additive drugs (refer to each of the antibiotics and caspofungin in the presented list) prior to fluconazol already .
3. If so, the addition of rifampicin would have posed an even greater cumulative risk of liver injury in addition to fluconazol itself. Fortunately it did not happen.
4. It is not possible to conclude that it was rifampicin that lowered the drug level. It could also be the TAC withdrawal itself (as we effectively perform in case of severe TAC or CsA overexposure) and, in parallel, the improvement in liver function and its associated enhanced metabolic capacities.
5. Since there obviously are no data, not even in the literature? to compare the time to normalisation of TAC blood levels between reduction/withdrawal alone and the addition of an inducer such as rifampicin, the conclusion cannot be to recommend rifampicin.
6. The onset of the CYP3A4-inducing effect is known to be delayed for several days. This is another point to be discussed. The induction of CYP3A4 by rifampicin takes around six days to develop and 11 days to disappear (Hukkanen J. 2012. Induction of cytochrome P450 enzymes: a view on human in vivo findings. Expert Review of Clinical Pharmacology 5: 569–585).
7. Furthermore, rifampicin was almost contraindicated in severe liver injury at time of its onset.
8. Tables etc: Please give the name of the active ingredient first, followed by the trade name in brackets.
9. Although not relevant in this manuscript, since fluconazol was stopped and rifampicin added afterwards for rescue therapy only, but generally important to know concerning rifampicin: In the case of a therapeutic option, alternative antifungal agent should be used because loss of efficacy of
fluconazole possible, requires close monitoring, dose adjustment of fluconazole if necessary.
10. Did you intend to apply caspofungin and fluconazol in parallel or was it just an overlap on the first day of the single shot fluconazol?
11. You discontinued fluconazol after the single dose for liver injury and TAC overexposure. Besides coiling, what antifungal therapy was used for the final tailored treatment of the multiple mycotic aneurysms described before patient discharged? Or was the only treatment the interventional
coiling?
12. Time course of white blood cells and platelets (the latter partly indicating severity of hepatic disease and additive vancomycin toxicity prior to fluconazol - vancomycin levels from therapeutic drug monitoring?)
13. Cave: The required transient dose increase of TAC with discharge needs to be reduced again with close monitoring after cessation of the almost 14 days lasting inducer effect of rifampicin.
14. The illustrations of the epidemiological data presented are very interesting. However, they also lack information on the above-mentioned aspects, which may not be available from the resources. If so, this should be stated.
15. Conclusion: Not generalisable because of all the aspects mentioned above: hepatotoxicity of rifampicin itself can worsen hepatic condition, especially, if already pre-impaired (fortunately did not happen here). Any countermeasure other than the reduction/temporary discontinuation of the CNI should always be tailored to the individual high-risk situation, its clinical course and the drug elimination capacity of the organs involved.
It is not possible to conclude whether these countermeasures are responsible for the reduced CNI blood levels (see delayed onset of the CYP3A4 inducer effect), because in parallel with the CNI cessation.
With these aspects being considered, the manuscript on this important topic, which provides a good representation of the epidemiological data studied and presented in detail, is interesting and may become more critical and accurate.

Author Response

Response to Reviewer 2

Thank you very much for taking the time to review this manuscript. Please find as outlined below the responses to your comments. The corresponding revisions are highlighted in the re-submitted manuscript (version 2).

General impression of reviewer:

The manuscript entitled "Rescue Therapy for Supratherapeutic Concentrations of Calcineurin Inhibitors Using Potent Cytochrome P450 Inducers" addresses the unintentional CNI overexposure, a serious risk in immunosuppressed inpatients and outpatients with severe nephrological, neurological and immunological ADRs. The provision of a well-evaluated and fast rescue intervention that does not pose additional hazards may be of benefit to these patients. With reference to a case report, the authors aim to review the available data from VigiBase™ and the literature on this topic to determine the most effective and safe countermeasure. However, there are some very important aspects that need to be considered and further clarified in the Results, Discussion and Conclusions section.

Comment 1: Most important missing: Laboratory data on 19 / 20 Dec

Response: Unfortunately, there was no laboratory investigation carried out on 19/Dec.

The available values from 20/Dec have been added to table 3 accordingly.

Comment 2: Liver function may already have been impaired by previous drugs, e.g. vancomycin, etc., so that the dose of tacrolimus should have been reduced already before fluconazole was added because of reduced hepatic metabolic capacity, itself requiring caution with dosage in this case. Please refer to the cumulative risk of liver injury by the predisposing additive drugs (refer to each of the antibiotics and caspofungin in the presented list) prior to fluconazol already.

Comment 3: If so, the addition of rifampicin would have posed an even greater cumulative risk of liver injury in addition to fluconazol itself. Fortunately, it did not happen.

Response: The dose of tacrolimus was drastically reduced on the day of administration of fluconazole.

Considering an oral bioavailability of approximately 20%, 2 mg oral formulation is equivalent to 0.4 mg intravenous. Hence, the administered intravenous dose of 0.072 mg/d was just about 18% of the expected plasma exposure from the baseline oral dose of 2 mg/d. The sudden derangement of the liver and renal parameters as well as acute onset encephalopathy demonstrated a close temporal correlation to the supratherapeutic TAC trough level, which was predominantly due to cytochrome enzyme inhibition by fluconazole. Hepatotoxicity, nephrotoxicity and neurotoxicity are well documented manifestations of acute TAC intoxication. Nonetheless, a pre-existing reduction in the overall hepatic metabolic capacity due to polypharmacy may be partly responsible. Since drug-induced liver injury (cumulative risk of the anti-infective agents) is a diagnosis of exclusion, the supratherapeutic TAC level is more suggestive of a concentration-dependent hepatic injury in this case. (L. 272-275 and 289-300)  

Comment 4: It is not possible to conclude that it was rifampicin that lowered the drug level. It could also be the TAC withdrawal itself (as we effectively perform in case of severe TAC or CsA overexposure) and, in parallel, the improvement in liver function and its associated enhanced metabolic capacities.

Response: Under fluconazole-induced CYP inhibition, the effective half-life of TAC reduced from 30.4 hours to 27.7 hours after 4 days of initiating rifampicin. This was less likely to occur after drug withdrawal alone, in the presence of CYP inhibition. (L. 289-292)

Moreover, in the conclusion and limitations, we stated that the analysis of the case reports did not include cases of CNI intoxication whereby the management was solely restricted to drug withdrawal or dose reduction; therefore, it was not possible to evaluate the actual benefit of pharmacoenhancement. As a result, multicenter clinical investigations comparing drug withdrawal or dose reduction with and without pharmacoenhancement are required to validate the safety and effectiveness of this approach.

 Comment 5: Since there obviously are no data, not even in the literature? to compare the time to normalization of TAC blood levels between reduction/withdrawal alone and the addition of an inducer such as rifampicin, the conclusion cannot be to recommend rifampicin.

Response: In some selected patient groups with pre-existing or evolving indications, such as underlying bacterial infections (rifampicin can be considered as antibiotic of choice) or CNI-induced seizures (phenytoin or phenobarbital can be considered as antiepileptic of choice), the benefits of pharmacoenhancement outweigh the risks. This is why we recommended that a critical patient-oriented risk-benefit analysis regarding the choice of CYP-inducer should be considered. 

Comment 6: The onset of the CYP3A4-inducing effect is known to be delayed for several days. This is another point to be discussed. The induction of CYP3A4 by rifampicin takes around six days to develop and 11 days to disappear (Hukkanen J. 2012. Induction of cytochrome P450 enzymes: a view on human in vivo findings. Expert Review of Clinical Pharmacology 5: 569–585).

Response: Yes, after initiating treatment with rifampicin, a gradual enzyme inducing effect develops over the course of approx. 2 weeks, which gradually decreases after drug cessation. The initial CYP-inducing effect of rifampicin, even within the first few days of treatment can result in clinically significant increase in the total clearance of tacrolimus. This has been demonstrated in previous pharmacokinetic studies. (289-292):

(Hebert MF, Fisher RM, Marsh CL, Dressler D, Bekersky I. Effects of rifampin on tacrolimus pharmacokinetics in healthy volunteers. J Clin Pharmacol. 1999 Jan;39(1):91-6. doi: 10.1177/00912709922007499. PMID: 9987705)

Comment 7: Furthermore, rifampicin was almost contraindicated in severe liver injury at time of its onset.

Response: After critical risk-benefit analysis, rifampicin was started. The benefits of pharmacoenhancement and antimicrobial action of rifampicin outweighed the risk of further liver injury in this patient who had recurrent bacterial infections and sepsis. (359-361)

 Comment 8: Tables etc: Please give the name of the active ingredient first, followed by the trade name in brackets.

Response: The medications in table 1 and 2 have been rearranged accordingly.

Comment 9: Although not relevant in this manuscript, since fluconazol was stopped and rifampicin added afterwards for rescue therapy only, but generally important to know concerning rifampicin: In the case of a therapeutic option, alternative antifungal agent should be used because loss of efficacy of fluconazole possible, requires close monitoring, dose adjustment of fluconazole if necessary.

Response: This was taken into account by augmenting the antifungal treatment with caspofungin. 

Comment 10: Did you intend to apply caspofungin and fluconazol in parallel or was it just an overlap on the first day of the single shot fluconazol?

Response: It was an overlap with the single shot fluconazole.

Comment 11: You discontinued fluconazol after the single dose for liver injury and TAC overexposure. Besides coiling, what antifungal therapy was used for the final tailored treatment of the multiple mycotic aneurysms described before patient discharged? Or was the only treatment the interventional
coiling?

Response: Since previous microbiological investigations demonstrated multiple resistance to antifungal agents, the patient was maintained on caspofungin.

 Comment 12: Time course of white blood cells and platelets (the latter partly indicating severity of hepatic disease and additive vancomycin toxicity prior to fluconazol - vancomycin levels from therapeutic drug monitoring?)

Response: The vancomycin trough levels and corresponding blood cell counts have been added to table 3 accordingly.

Comment 13:  Cave: The required transient dose increase of TAC with discharge needs to be reduced again with close monitoring after cessation of the almost 14 days lasting inducer effect of rifampicin.

Response: This was implemented during follow up visits. The dose after hospital discharge was reduced from 2 mg 2x/d to the baseline dose of 1 mg 2x/d. (L. 124)

Comment 14: The illustrations of the epidemiological data presented are very interesting. However, they also lack information on the above-mentioned aspects, which may not be available from the resources. If so, this should be stated.

Response: Unfortunately, cases obtained from the database do not give detailed description of treatments undertaken apart from drug withdrawal or dose reduction. In 43% of cases the exact actions taken were not reported. This is a major limitation of the analyses of the epidemiological data.

Comment 15: Conclusion: Not generalizable because of all the aspects mentioned above: hepatotoxicity of rifampicin itself can worsen hepatic condition, especially, if already pre-impaired (fortunately did not happen here). Any countermeasure other than the reduction/temporary discontinuation of the CNI should always be tailored to the individual high-risk situation, its clinical course and the drug elimination capacity of the organs involved.
It is not possible to conclude whether these countermeasures are responsible for the reduced CNI blood levels (see delayed onset of the CYP3A4 inducer effect), because in parallel with the CNI cessation.
With these aspects being considered, the manuscript on this important topic, which provides a good representation of the epidemiological data studied and presented in detail, is interesting and may become more critical and accurate.

Response: Yes, we stated this as a limitation. Since the analysis of the case reports did not include cases of CNI intoxication whereby the management was solely restricted to drug withdrawal or dose reduction, it was not possible to evaluate the actual benefit of pharmacoenhancement. This is why we stated in the conclusion that multicenter clinical investigations comparing drug withdrawal with and without pharmacoenhancement are required to validate the safety and effectiveness of this strategy.

We avoided generalization by stating that: For individual cases, a critical patient-oriented risk-benefit analysis regarding the choice of CYP inducer as well as the dosage and route of administration is highly recommended.   

Round 2

Reviewer 1 Report

Dear Authors,

I read with interest your revised article, and thank you for considering my suggestions.

I still have a few comments :

Paragraph : Extraction of data from VigiBase™ : Thank you for your explanation, but it still feels you must have missed some cases. First, for the eight cases of “Drug Intoxication LLT” with ciclosporin, it is not clear for all of them that there was an increased concentration of CNIs. And conversely, just scrolling a bit randomly, there is a case with "tacrolimus” and “drug toxicity”, clearly also labelled “drug level above therapeutic”. Difficult however to assess how many cases could have been added.

Line 146-147 : The mention of SMQs can be removed, as you did not use SMQs.

Line 223 : I would add the references here also as to be able to have them easily. The supplementary table does not send back to the reference number of the article and I would add this also in the table.

Fig 8 : Flow diagram : please add numbers also in the excluded boxes

Line 314 : I still don’t agree with this sentence. “There was a reporting bias with respect to the countries…”. If you look at the global data, it’s the whole Vigibase that is biased, due to the known flaw of spontaneous pharmacovigilance. For me, there is no point in adding this remark.

Kind regards

Author Response

Response to Reviewer 1

Thank you for your much appreciated comments and suggestions that further enhance the quality of our article. Kindly find below our responses.

 General impression of reviewer:

I read with interest your revised article, and thank you for considering my suggestions.

I still have a few comments:

Paragraph: Extraction of data from VigiBase™:

Thank you for your explanation, but it still feels you must have missed some cases. First, for the eight cases of “Drug Intoxication LLT” with ciclosporin, it is not clear for all of them that there was an increased concentration of CNIs. And conversely, just scrolling a bit randomly, there is a case with "tacrolimus” and “drug toxicity”, clearly also labelled “drug level above therapeutic”. Difficult however to assess how many cases could have been added.

Response:

Yes, we agree. It was very challenging to extract cases that explicitly reported supratherapeutic concentrations of CNIs from other reaction terms without duplication from “Drug Intoxication LLT”. The availability of a PT for “Drug Intoxication” would have been optimum. The general observation in the database was that the specification of increased blood concentrations of CNIs was more frequently indicated for cases of tacrolimus compared to ciclosporin. 

“Drug Intoxication LLT” with ciclosporin: The cases that do not explicitly specify “drug level above therapeutic” however report a constellation of adverse reactions that were most likely due to severe ciclosporin intoxication.       

Comment: Line 146-147

The mention of SMQs can be removed, as you did not use SMQs.

Response:

Statements regarding SMQs have been removed.

Comment: Line 223

I would add the references here also as to be able to have them easily. The supplementary table does not send back to the reference number of the article and I would add this also in the table.

Response:

The references have been added (L. 223-224). These have also been indicated in the supplementary table accordingly.

Comment: Fig 8

Flow diagram, please add numbers also in the excluded boxes.

Response:  N=447 and N=27 have been added accordingly.

Comment: Line 314

I still don’t agree with this sentence. “There was a reporting bias with respect to the countries…”. If you look at the global data, it’s the whole Vigibase that is biased, due to the known flaw of spontaneous pharmacovigilance. For me, there is no point in adding this remark.

Response:

This statement has been removed.

Reviewer 2 Report

I suggest that the following aspects be addressed in final:

Reason different: Based on its high molecular weight, poor aqueous solubility, and extensive erythrocyte and plasma protein binding, it is anticipated that tacrolimus will not be dialysable (SmPC).

TAC-measurements are blood trough levels

Mode of i.v. application: exact duration of i.v. TAC infusion? (e.g. over 2 hours, 4 hours or per continuitatem?)

Loss of efficacy of caspofungin due to P-pg induction by rifampicin to be considered.

Please add literature in line 292 (references?)

Fig 11: Indication of simultaneous TAC withdrawal essential

Important: Refer to lack of comparison of TAC withdrawal alone in the Conclusion session

Line 359: Only 3 days of rifampicin application cannot be declared as a supportive  antibacterial therapy (see also Conclusion incorrect: benefited from merely 3 days antibacterial rifampicin?)

Please concretise the risk of rifampicin use in severely pre-existing hepatic impairment: It remains  an off-label use with further risk of toxicity (see contraindications from its SmPC). This is important because in other cases the strategy could be copied with fatal outcomes.

Congratulations on reviewing the literature on this topic.

Please indicate that the literature numbers are not representative. It must be assumed that the number of unreported cases is much higher.

Author Response

Response to Reviewer 2

Thank you for your much appreciated comments and suggestions that further enhance the quality of our article. Kindly find below our responses.

General impression of reviewer:

Congratulations on reviewing the literature on this topic. I suggest that the following aspects be addressed in final.

Comment: Reason different:

Based on its high molecular weight, poor aqueous solubility, and extensive erythrocyte and plasma protein binding, it is anticipated that tacrolimus will not be dialysable (SmPC).

Response:

Thank you, we have revised this statement (L. 80-81)

Comment:

TAC-measurements are blood trough levels

Response:

All statements regarding TAC-measurements have been amended accordingly.

Comment: Mode of i.v. application

Exact duration of i.v. TAC infusion? (e.g. over 2 hours, 4 hours or per continuitatem?)

Response:

TAC infusion over 2.5 hours has been indicated.

Comment:

Loss of efficacy of caspofungin due to P-pg induction by rifampicin to be considered.

Response:

This statement has been added (L. 291-292).

Comment:

Please add literature in line 292 (references?)

Response:

Literature has been added (reference 51)

Comment: Fig 11

Indication of simultaneous TAC withdrawal essential

Response:

This has been indicated (L. 304)

Comment:

Important: Refer to lack of comparison of TAC withdrawal alone in the Conclusion session

Response:

Due to the lack of clinical cases whereby the effect of TAC withdrawal alone compared to that of withdrawal and pharmacoenhancement, multicenter clinical investigations are required to validate the safety and effectiveness of the various pharmacoenhancement strategies (L. 368-371).

Comment: Line 359

Only 3 days of rifampicin application cannot be declared as a supportive antibacterial therapy (see also Conclusion incorrect: benefited from merely 3 days antibacterial rifampicin?)

Response:

“The benefit of antimicrobial action of rifampicin” has been removed in the conclusion. 

Comment:

Please concretise the risk of rifampicin use in severely pre-existing hepatic impairment: It remains  an off-label use with further risk of toxicity (see contraindications from its SmPC). This is important because in other cases the strategy could be copied with fatal outcomes.

Response:

The conclusion has been modified with the statement: “Although rifampicin is contraindicated in patients with severe pre-existing or acute liver injury, it was administered as an off-label therapy after rigorous risk-benefit analysis”

Comment:

Please indicate that the literature numbers are not representative. It must be assumed that the number of unreported cases is much higher.

Response:

This has been indicated as part of the limitations (L. 387-389).