Efficacy and Safety of JAK1 Inhibitor Abrocitinib in Atopic Dermatitis
Mohamed A. Kamal
Y. Fujita
Round 1
Reviewer 1 Report
Iznardo et al present a comprehensive review of the efficacy/safety of Abrocitinib to treat moderate to severe AD. While the review presents interesting insights, there are major concerns with interpretation of the data presented by the authors. In particular, this relates to assessment of the risk of JAK inhibitors for thrombosis and PE events based on a recent meta-analysis and proper assessment of the risk/benefit profile of Abrocintib. A major revision is recommended before publication.
Major concerns:
Abstract (Page 1)
Line 17: Interpretation of the meta- analysis is misleading by the authors. The meta-analysis resulted in a Hazard Ratio point estimate of 0.95 with a confidence interval (0.62,1.45) which includes 1 (line 444 , Discussion section). Therefore, the results are equivocal, and lack of risk associated with development of thrombosis and PE cannot be ruled out.
Pharmacokinetics and Pharmacodynamics (Page 3)
Throughout the document, authors point out the advantages of Abrocitinib over dupilumab, but do not conversely point out the advantages of large molecule therapies over small molecules particularly as it relates to lack of Drug-Drug interactions and the lack of need for dose adjustment in hepatic and renal impairment. This contrast should be discussed thoroughly in this section.
Table 1.
The incidence of DVT/PE should be listed for all studies in this table particularly long -term extension studies.
Typo on 3rd row, 6th column, EASI-5 should be EASI-75
Section 3.2
In JADE-COMPARE, Percent of patients with moderate versus severe AD should be stated for each treatment group to ensure treatment arms are balanced with respect to baseline disease severity.
Line 257: Interestingly among patients….
This comparison is flawed because patients from JADE compare who entered JADE Extend after 6 weeks off drug (from week 14 to week 20 on placebo) have still not fully washed out Dupilumab. The time to washout for Dupilumab is 12 weeks on average, this in addition to the indirect response of dupilumab where dubilumab will continue to show pharmacodynamic and clinical effects for several weeks even after concentrations have fallen below the limit of detection. Therefore, the statement that patients who achieved EC90 when switched to Abrocitinib in JADE EXTEND is likely inclusive of the combined drug effect of Dupilumab+Abrocitinib and not Abrocitinib alone.
Section 3.4
Line 301. The comparative incidence of AE’s lack parallelism to number of AE’s stated (5 AEs vs 4 comparative incidences)
Discussion:
Line 362: Authors do not discuss the fact that the majority of patients who develop conjunctivitis while on dupilumab are treated for conjunctivitis and usually can remain on the biologic. Are authors suggesting that patients on dupilumab who develop conjunctivitis switch to a JAK inhibitor which carries other more severe safety risks? The sentence as written is misleading.
Line 444
Authors present results of a large Meta Analysis assessing risk of developing deep vein thrombosis and pulmonary embolism. This analysis included data from several studies and resulted in a hazard ratio with a point estimate of 0.95. However, the confidence interval includes 1 (0.62,1.45) which means that the results are equivocal. This should be clearly pointed out be the authors.
In general, in the discussion, there should be more discussion on the therapeutic index of Abrocitinib and more robust discussion on the risk/benefit particularly as it related to the 200 mg dose. While clearly there is an increased clinical efficacy with increase in dose, this high dose carries the risk of thrombocytopenia observed in short-term studies and to a lesser extent DVT/PE observed only in longitudinal studies. The relationship between patients on Abrocitinib developing a reduction in platelet counts (thrombocytopenia) and the likelihood to develop DVT/PE in a longitudinal (long term follow up) assessment should be explained.
Author Response
Iznardo et al present a comprehensive review of the efficacy/safety of Abrocitinib to treat moderate to severe AD. While the review presents interesting insights, there are major concerns with interpretation of the data presented by the authors. In particular, this relates to assessment of the risk of JAK inhibitors for thrombosis and PE events based on a recent meta-analysis and proper assessment of the risk/benefit profile of Abrocintib. A major revision is recommended before publication.
We would like to thank the reviewer for their detailed review of our manuscript and providing their valuable input
Major concerns:
Abstract (Page 1)
Line 17: Interpretation of the meta- analysis is misleading by the authors. The meta-analysis resulted in a Hazard Ratio point estimate of 0.95 with a confidence interval (0.62,1.45) which includes 1 (line 444 , Discussion section). Therefore, the results are equivocal, and lack of risk associated with development of thrombosis and PE cannot be ruled out.
We agree with the conclusions of the authors of the referenced meta-analysis: HR 0.95 with 95%CI 0.62–1.45 indicates no significant association of AD with incident VTE; furthermore, in their meta-analysis of clinical trials they found no significant difference in the risk of VTE between patients with AD receiving JAK inhibitors, dupilumab, or placebo. We have reworded our Discussion to make this clearer. Of course, lack of risk cannot be ruled out, but association has by no means been demonstrated.
Pharmacokinetics and Pharmacodynamics (Page 3)
Throughout the document, authors point out the advantages of Abrocitinib over dupilumab, but do not conversely point out the advantages of large molecule therapies over small molecules particularly as it relates to lack of Drug-Drug interactions and the lack of need for dose adjustment in hepatic and renal impairment. This contrast should be discussed thoroughly in this section.
We have added this paragraph to the Discussion: The potential for drug interactions and the need for dose adjustment in patients with significant renal or hepatic impairment are shared to some extent by all JAK inhibitors and may determine the choice of monoclonal antibodies for treatment of AD in some patients.
Table 1.
The incidence of DVT/PE should be listed for all studies in this table particularly long -term extension studies.
Rather than detailing the incidence of VTE for each study, we have expanded the reference to the 5 VTE AEs in the integrated safety analysis.
Typo on 3rd row, 6th column, EASI-5 should be EASI-75
Thank you, we have corrected the typo.
Section 3.2
In JADE-COMPARE, Percent of patients with moderate versus severe AD should be stated for each treatment group to ensure treatment arms are balanced with respect to baseline disease severity.
The corresponding percentages at baseline were
|
|
Total |
Abro 200 |
Abro 100 |
Dupi |
Pbo |
|
IGA 3 |
64.6 |
61.1 |
64.3 |
66.9 |
67.2 |
|
IGA 4 |
35.4 |
38.9 |
35.7 |
33.1 |
32.8 |
|
EASI |
30.9 |
32.1 |
30.3 |
30.4 |
31.0 |
|
BSA |
48.5 |
50.8 |
48.1 |
46.5 |
48.9 |
|
SCORAD |
67.9 |
69.3 |
66.8 |
67.9 |
67.9 |
We cannot see any evident imbalance between groups…
Line 257: Interestingly among patients….
This comparison is flawed because patients from JADE compare who entered JADE Extend after 6 weeks off drug (from week 14 to week 20 on placebo) have still not fully washed out Dupilumab. The time to washout for Dupilumab is 12 weeks on average, this in addition to the indirect response of dupilumab where dubilumab will continue to show pharmacodynamic and clinical effects for several weeks even after concentrations have fallen below the limit of detection. Therefore, the statement that patients who achieved EC90 when switched to Abrocitinib in JADE EXTEND is likely inclusive of the combined drug effect of Dupilumab+Abrocitinib and not Abrocitinib alone.
We have included this interesting and sensible insight in our manuscript: ‘It is worth mentioning that in patients who switched from dupilumab to abrocitinib, after 6 weeks of placebo washout and thereafter some residual effect of dupilumab might have contributed to the efficacy of abrocitinib in JADE EXTEND.’
Section 3.4
Line 301. The comparative incidence of AE’s lack parallelism to number of AE’s stated (5 AEs vs 4 comparative incidences)
In the most recent integrated safety analysis of abrocitinib, 5 VTE are reported and described; this number may differ from figures reported in previous meta-analyses. We have expanded the corresponding description in this section.
Discussion:
Line 362: Authors do not discuss the fact that the majority of patients who develop conjunctivitis while on dupilumab are treated for conjunctivitis and usually can remain on the biologic. Are authors suggesting that patients on dupilumab who develop conjunctivitis switch to a JAK inhibitor which carries other more severe safety risks? The sentence as written is misleading.
The suggestion implied by the reviewer is not made explicit, and it is up to each prescriber to balance the efficacy/safety profile of the different alternatives we mention in this paragraph. Anyhow, we have changed the wording: ‘Dupilumab-associated conjunctivitis can usually be managed with ophthalmologic treatment, but some patients must interrupt dupilumab due to development of AE (mainly conjunctivitis and head and neck erythema).’
Line 444
Authors present results of a large Meta Analysis assessing risk of developing deep vein thrombosis and pulmonary embolism. This analysis included data from several studies and resulted in a hazard ratio with a point estimate of 0.95. However, the confidence interval includes 1 (0.62,1.45) which means that the results are equivocal. This should be clearly pointed out be the authors.
We do not understand this comment. Any HR including 1 within 95%CI must be considered not significant as regards the association in consideration.
In general, in the discussion, there should be more discussion on the therapeutic index of Abrocitinib and more robust discussion on the risk/benefit particularly as it related to the 200 mg dose. While clearly there is an increased clinical efficacy with increase in dose, this high dose carries the risk of thrombocytopenia observed in short-term studies and to a lesser extent DVT/PE observed only in longitudinal studies. The relationship between patients on Abrocitinib developing a reduction in platelet counts (thrombocytopenia) and the likelihood to develop DVT/PE in a longitudinal (long term follow up) assessment should be explained.
We have expanded the discussion of the laboratory abnormalities associated with abrocitinib treatment in AD patients.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear Authors,
I read with interest your review. It is an interesting and well structured review on every issue about Abrocitinib for atopic dermatitis. I will recommend creating a "Methods" section to explain how the review was performed, to improve the scientific quality of the paper. As it was not a systematic review, just explain that it was a review of the papers considered more important by the authors.
Author Response
Thank you for your kind review. A short Methods section has been added as per your recommendation:
Methods
A PubMed search with the term ‘abrocitinib’ yielded 117 results; all the abstracts were screened, and the papers considered more important by the authors, as well as relevant general references on JAKs were included in this review.
Reviewer 3 Report
The authors reviewed the comprehensive properties and results of clinical trials of abrocitinib, a novel JAK1 inhibitor for atopic dermatitis. The review is well-constructed and well-written. Although it would be better if the authors described the pharmacokinetic comparison among other JAK inhibitors available for atopic dermatitis, the reviewer enjoyed this review article.
Author Response
Thank you for your kind review and insightful comment. A comparative pharmacologic review with the other JAK inhibitors approved for atopic dermatitis would probably require some efficacy/safety correlations and would probably exceed the scope and expected extent of this manuscript.
Reviewer 4 Report
Methods used for the review were not presented. Conflicts of interest are not explicit.
Author Response
Thank you for your kind review. We have added a short Methods section and our conflict of interests disclosure, which was inadvertently omitted in the original submission.
Methods
A PubMed search with the term ‘abrocitinib’ yielded 117 results; all the abstracts were screened, and the papers considered more important by the authors, as well as relevant general references on JAKs were included in this review.
Round 2
Reviewer 1 Report
There is still one major concern with respect to interpretation of the meta analysis assessing the association between increased risk of DVT/PE incidence in AD and treatment with JAK inhibitors. While the point estimate of the Hazard Ratio is 0.95, the Confidence Interval is (0.62, 1.45). The lower bound of the CI is 0.62 indicating decreased risk of DVT/PE with JAK inhibitor therapy, while the upper bound of 1.42 indicates increased risk of DVT/PE with JAK therapy. Therefore because the CI around 1 (an HR of 1 would indicate no risk) is wide, the authors cannot make the conclusion stated that there is no association of JAK inhibitor therapy and incidence of DVT/PE based on this analysis. This messaging needs to be reflected in the paper and is misleading as written.
Author Response
We respectfully disagree with the reviewer's interpretation of the meaning of hazard ratio and confidence intervals.
A HR of 0.95, with 95%CI between 0.62 and 1.45 precisely means that the HR is not significant, and at any rate the central estimate is close to 1 (we are not talking abot 1.42 with 95%CI 0.95–1.90, where one cold argue about statistical power or whatever).
The meta-analysis has not demonstrated either an increased or an increased risk of DVT/PE in association with AD or JAK inhibitor therapy or dupilumab, for that matter. One might use this wording, but it is self-evident to any reader with some statistical knowledge, and would be detrimental to the perceived quality of our writing.
At any rate, we have modified the Abstract in an attempt to allay any concerns this reviewer may have with our wording:
'Even though recent meta-analyses of clinical trials have not shown that atopic dermatitis, or its treatment with JAK inhibitors or dupilumab, modify the risk of deep venous thrombosis or pulmonary embolism, long-term follow-up studies will better define the safety profile of abrocitinib.'
Round 3
Reviewer 1 Report
The authors have addressed all comments and the manuscript can be recommended for publication.
