JAK INHIBITORS AS A NEW MODALITY FOR TREATING ATOPIC DERMATITIS: A BETTER UNDERSTANDING OF ITS EFFICACY AND SAFETY

AD is a chronic relapsing disease characterized by pruritis and chronic inflammatory skin changes, (1) which affects approximately 10Ã¢Â€Â“30% of children and as much as 10% of adults worldwide.(2) This condition has a great significant impact on morbidity and presents an outstanding social economic burden.(3) AD is a multifactorial disease that develop by interaction between these factors in a positive feedback cycle. Treatment of AD interrupts the causal pathway. Management with conventional therapies has been a challenge, but a novel biological treatment called dupilumab was recently approved for the treatment of moderate-to-severe AD, but this drug only achieved 40% clear skin in combination with TCs.(4). JAK inhibitors are another new drug family that inhibit JAK-signaling pathways, which involve JAK1, JAK2, JAK3 and TYK2. JAK inhibitors have been approved to treat inflammatory diseases like rheumatoid arthritis, and high attention is currently being focused on the clinical development of JAK inhibitors for the treatment of AD. Which are a possible treatment for certain disease that are related to lymphocyte activation, such as psoriasis, alopecia areata, vitiligo and AD. JAK inhibitors are available in topical and oral forms, thereby allowing more administration routes depending on the severity of AD, which ranges from mild to severe. Since JAK inhibitors are a new treatment modality in dermatology, the efficacy of this new medicine and the safety thereof are still being debated. A systematic review and metaÃ¢Â€Âanalysis were done for all randomized clinical trials that evaluated JAK inhibitors for Atopic dermatitis to investigate their pooled efficacy and safety compared to placebo. Results might be useful as a milestone to develop a more accurate view of this medication and provide direction for further research.


Data Extraction
After screening selected studies and assessments, key information from these studies is extracted into an Excel spreadsheet that is derived from the standardized data-extraction tool for quantitative studies devised by the Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI). In order to increase accuracy of extracted data, two independent reviewers (AN and HM) were assigned to extract data using a standardized electronic data collection form. Discrepancies are resolved by consensus.

Included Identification
Records identified through database searching (n = 838)

Records excluded (n = 436)
Full-text articles assessed for eligibility (n = 54) Full-text articles excluded with reasons (n = 45) 15: Duplicated study 4: Wrong study design 10: No full text 6: Wrong outcomes 1: Wrong population 1: Wrong setting 3: Biopsy-based study, no clinical outcome 1: Conference presentation 2: Not completed study 2: Phase one study Studies included in synthesis (meta-analysis) (n = 9) 6 Extracted information includes: 1. Study characteristics of reviewed papers, such as authors, years of publication and the name of the publishing journal 2. Methods of the study, including study design (RCT, quasi-RCT, Longitudinal, retrospective) and research purpose and/or questions 3. Participant characteristics, country where the study took place, setting, population, sample size, age and sex 4. Interventions used to treat AD in the experimental and control groups, including the dosages (if applicable) 5. Outcome measures and results 6. Conclusions of reviewed papers and any comments from reviewers.
For primary outcomes, clinical outcomes are assessed using the EASI scoring method. Clinical response defined by a EASI score ≥ 50 are extracted and compared to the baseline. Primary studies that made few modifications for EASI way of measurement and were not expected to demonstrate significant difference from regular EASI were included. Clinical response defined by an IGA 0-1 or ≥ 2-point reduction from baseline are also extracted. Patient-Oriented Eczema Measure (POEM) showing a reduction of four or more points from the baseline are indicated as a primary outcome and are extracted from one study. Finally, the Peak Pruritus Numerical Rating Scale-11  showing an improvement of four or more points from the baseline are extracted from three studies that assessed oral JAK inhibitors. The DLQI with an improvement of ≥ 4 points reduction from baseline are not measured in any of the included studies. Different definitions for outcomes were found in primary studies of atopic dermatitis. To overcome this problem, It was standardized through the use of above definitions and extracted data based on that to allow for comparisons between these trials.
For secondary outcomes, safety was assessed by extracting the proportions of patients who got any of the following events: adverse Events (AE) or serious AEs (SAE) or quit the study due to AE or infections related adverse events or worsening Atopic dermatitis. In addition, We collected the number of cases who experienced herpes zoster (HZ), serious events such as cancer or death. AEs that occurred during parts of studies were placebo was not used as a comparator was not included.
Serious adverse events (SAEs) are defined as any untoward medical occurrence that results in death, is lifethreatening, requires inpatient hospitalization or prolongation of existing hospitalization, results in persistent or significant disability and/or incapacity, results in a congenital anomaly and/or birth defect or if appropriate medical judgment deems it as an important medical event that may jeopardize the health of the research participant or may require medical intervention to prevent one of the outcomes listed above. (9) It is also important to define serious infectious AEs, as these are infections that lead to death or hospitalization or require intravenous antibiotics. (10)

Data Synthesis
In order to compare treatment effect between intervention groups and placebo groups We used the relative risk (RR) for efficacy and safety as well. We used intention to treat principles with calculation of risk ratio with 95% confidence interval. A RR<1 will represent a lower rate of outcome among the group of patients who were treated with JAK inhibitors. For studies that assessed effects of different doses of JAK inhibitors on patients, We pooled all the data together. Because of the expected heterogeneity of data collected, We used a random-effects model to pool these data. The random effect is based on what is called the DerSimonian and Laird method which consider different types of heterogeneity within-and between-studies. (11) This will account for the heterogeneity between study populations given the pooling of different therapies and doses. Chi-square test was used to test the heterogeneity. In addition, between study heterogeneity was evaluated using the I 2 metric. It indicates the percentage of the total variation across studies. The following thresholds for I 2 interpretation were used in accordance with the Cochrane Handbook for Systematic Reviews of Interventions: 0%-40% (may not be important), 30%-60% (moderate heterogeneity), 50%-90% (substantial heterogeneity), 75%-100% (considerable heterogeneity). Potential publication bias was not formally evaluated by funnel plots as less than 10 studies were identified for both of Oral and topical JAK inhibitors.
All analyses were conducted in review manager 5.3. All P-values are two-tailed and P < 0.05 indicates statistical significance.

Systematic Review and Meta-Analysis Results Database Search Resultsand Characteristicsof Selected Studies
A total of 838 records were identified through searching different databases. Removing duplicates was done and ended up with 490 records. These records were screened and 436 studies were found to be ineligible and excluded based on reviewing the title and abstract; in final step which is full text assessment (see Figure 2) 44 records were excluded for the following reasons: 1. 14 were duplicated 2. The full text for 10 studies could not be found 3. Four had the wrong study design 4. Nine showed the wrong outcomes with three of using biopsy for their outcome assessment 5. One had the wrong population (pediatric patients) 6. One was in the wrong setting 7. One was a conference presentation 8. Two were phase-1 studies 9. Two were incomplete 10 placebo controlled RCTs were eligible for inclusion in the systematic review and meta-analysis, (118-125) but two independent trials of these RCTs were conducted in one study (20) 10. Six evaluated patients on oral JAK inhibitors (13)(14)(15)(16)20) 11. Four trials evaluated patients on topical JAK inhibitors (118,(123)(124)(125) These trials and their baseline patient characteristics are summarized in Table 2.
The included studies were conducted in countries located in Europe, North and Latin America, Asia and Australia. Participants in the studies were vitally stable adult patients (18-65 years of age). Details about the studies are also summarized in Table 2. A total of 2612 AD patients, including 1085 (41.5%) females, were randomized: 1795 to intervention groups with a JAK inhibitor, and 817 to control groups. Only two trials were found to assess JAK1selective inhibitors (Upadacitinib (16) and Abrocitinib (14) ) and seven trials that assessed pan-JAK inhibitors (tofacitinib, (12) Gusacitinib, (13) Baricitinib, (121) RUX, (17) Delgocitinib (18,19) and Baricitinib. (20) Table2 Baseline characteristics of included randomized controlled trials assessing efficacy and safety of JAK inhibitors in treating atopic dermatitis disease. * Mean age is reported in years (Y) ± standard deviation (SD), unless described as median age. ** Int. is the intervention groups pooled together, and Ctl. is the control group. *** TCs is topical corticosteroids; TAC is tacrolimus.

Assessment of methodological quality and risk of bias
The quality of the nine selected studies was independently assessed by researchers AN and HM. The standard critical appraisal tools for quantitative studies devised by the JBI-MAStARI Methodological was used to determine the validity of the selected papers for retrieval, as shown in Table 3. All disagreements between the two reviewers were solved through discussions.
Assessment of study validity revealed few sources of bias. All studies reported treatment randomisation and blinding of participants. All studies concealed allocation to treatment groups. Withdrawn participants were considered in the analysis of outcomes in six studies, but this was unclear in Nakagawa et al. (2020); not reported in Nakagawa et al. (2018); and incomplete in Simpson et al. (2020). A majority of the included studies-five of nine-did not clearly report that outcome assessors were blind to treatment. All studies except for Simpson et al. (2020) showed comparable intervention groups to placebo group in which this criterion was partially met. All the studies that fulfilled the inclusion criteria were included after a methodological quality assessment, as they all achieved > 75% in the JBI-MAStARI criteria.
Table3:-Quality assessment of selected studies using JBI-MAStARI. The result of the meta-analysis of pooled data from the four included studies was highly homogenous-I 2 = 0%, and showed a statistically significant effect of the topical use of JAK inhibitors to treat AD, as shown by the two recorded primary outcomes: EASI 50 and IGA 0-1 with RR = {3.29 (2.46, 4.39) χ 2 = 0.9 P = 0.83} and {2.93 (1.22, 7.04), χ 2 = 0.17 P = 0.68}, respectively, compared to the placebo (see Figures 3 and 4).

Oral JAK inhibitors
The research for primary studies detected five RCTs that evaluated the efficacy and safety of orally administered JAK inhibitors to treat AD patients. In an RCT conducted by Bissonnette et al. (2019), a total of 36 patients were randomised. The proportion of patients achieving EASI 50 at day 29 was significantly higher for patients receiving gusacitinib in 40mg (100%, P = 0.003) and 80mg (83%, P = 0.03) doses, but not 20mg (20%, P = 0.93) doses, compared with placebo (22%), with all pooled groups achieving RR = 3.17 (0.91, 11.01), which is statistically insignificant. The proportion of patients achieving IGA 0-1 with at least a two-point reduction from baseline at day 29 was not statistically significant for different intervention groups compared with placebo: 43% (P = 0. 16 Two studies conducted by Guttman-Yussky et al. (15,16) investigated the efficacy of the oral JAK inhibitors baricitinib and upadacitinib, respectively. The collective result of the proportions of EASI 50 in the intervention groups to placebo group were statistically significant 64% to 12   *All included studies that measured NRS were assessing the efficacy of oral JAK inhibitors. Table 4 summarizes the safety of JAK inhibitors in intervention groups compared to control groups. There is no significant increase in overall risk for adverse events in atopic dermatitis patients who got JAK inhibitors (orally and topically treated pooled together) with RR= 1.05 [95% CI 0.87-1.26] P = 0.0009, χ 2 = 26.47, I 2 = 70% (see Figure  6) TEAE are undesirable events that were not present prior to starting a medical treatment. The meta-analysis for TEAE showed that the pooled-RR for orally administered JAK inhibitors RR= 1.12 (95%, CI 0.95-1.34) P = 0.02, χ 2 = 12.07, I 2 = 67% is slightly higher than in the topically applied groups RR = 0.91 (95%, CI 0.51-1.65) P = 0.01, χ 2 = 10.93, I 2 = 73%; however, none of them are statistically significant. A total of seven studies evaluated serious AEs, with a pooled-RR 0.75 ([95%, CI 0.41-1.38] P = 0.84, χ 2 = 2.08, I 2 = 0%), which denotes a protective effect but is statistically insignificant (see Figure 9). Study quit and withdrawals due to adverse events in intervention groups showed roughly similar risks compared to placebo groups (RR 1. It should be noted that only one case of herpes zoster has been reported in placebo groups compared to two cases in intervention groups of the same study BREEZ1 with no statistically significant importance. Nakagawa et al. (2020) noticed that a majority of reported AEs were not related to the study drug, with only five of the 23 reported AEs having been found to be related to the treatment; this is not significantly different from the placebo group. Another study done by Gooderham et al. (2020) (14) found a total of 184 of 267 patients (68.9%) experienced 402 TEAEs that were mostly mild; of these, 125 events reported by 64 of 267 (24.0%) patients were considered to be related to treatment, specifically abrocitinib and placebo. The most frequently reported of these are TEAE, and intestinal disorders were found to be significantly related to abrocitinib treatment. Only two out of seven reported serious adverse events in intervention groups were considered to be related to treatment. Significant infectious AEs were only reported only in 10 cases out of a total of 1422 participants in all intervention groups of reporting studies, compared to zero out of 733 in placebo groups. Upper respiratory tract infection and gastrointestinal manifestations including diarrhea, nausea and headache were the most commonly reported treatment-emergent AEs by most of the included studies. The most commonly reported infections were upper respiratory tract infections or nasopharyngitis. The most commonly reported AEs that were related to the study drug was application site pain and infection. However, a meta-analysis of different secondary outcomes indicated that applying JAK inhibitors did not result in a statistically significant increased risk of AE compared to placebo groups.

Safety of Topical and Oral JAK Inhibitors
Table4:-Summaryofsafetyoutcomes.     Figure 10). It is also important to note that JAK1 selective inhibitor showed an increased incidence of TEAE (RR 1.37 [95% CI 1.06-1.77], P = 0.14, n = 2 studies) (see Figure 11), compared to the statistically insignificant decrease in TEAE associated with treatment with non-selective JAK inhibitors (RR 0.98 [95% CI 0.83-1.15], P = 0.04, n = 8 studies). EASI 50 and TEAE were used to assess the effectiveness of JAK inhibitor selectivity on the efficacy and safety of the study drugs, as these two outcomes were measured by a large number of included studies and yielded high accuracy expectations  AD is a disease that has a large burden; it affects physical, psychological, social and economic parameters. (21) Many drugs have been used to treat AD that mainly depend on emollients and topical anti-inflammatory medications; in most moderate and severe cases, however, these drugs are insufficient. (22) For such cases, systemic corticosteroids have become more effective, but safety factors cause their use to be restricted. (23) Other drugs, including systemic immunosuppressants, can be prescribed for AD, but these drugs are not approved in many countries and carry a high risk of side effects that minimize their application time. (22) One of the most recently approved treatment for moderate-to-severe AD is dupilumab. (24) This new drug has many issues related to its application and inadequate responses that vary across patients. As a result of all these factors, there is a need for alternative AD therapies. JAK inhibitors are orally and topically administered, and the small molecular size causes it to be eligible for consideration as a treatment for AD.

NR : not reported
To my knowledge, this is the first systematic review and meta-analysis reporting on the efficacy and safety of JAK inhibitors in the treatment of AD. All randomized controlled trials of JAK inhibitors targeting adult patients with moderate-to-severe atopic dermatitis were included, and it was found that these JAK inhibitors are effective treatments for such patients. The pooled data also showed that intervention groups were not accompanied with significant increase in overall risk of adverse events or serious adverse events when compared to control groups. However, an increased risk of infection, mainly upper respiratory tract infections, was reported in most of the included RCTs but was not referred to in the study drug. In addition, gastrointestinal manifestations, application site pain and headaches were recognized as treatment-related AEs, although they were still not at a level of significance (14,15,17) The patients that were evaluated in these trials demonstrated a clinically meaningful efficacy for the treatment. This has been detected through the increase in the proportion of patients achieving EASI-50 to reach 63.3% as shown by the pooled analysis. JAK inhibitor treatment was also associated with an almost three-fold increase in the chances for improvement of patients, as detected by achieving IGA 0-1,or two point improvement from the baseline and a 3.6-fold increase in the proportion of patients who achieved a four-or-more-point improvement below baseline in their Pruritus NRS score compared to the placebo groups. All these results were statistically significant.
Importantly, limitations of the included primary studies must be considered. Because the analysis was dependent on data that was exclusively retrieved from short-term trials, these trials ranged in duration from four weeks, as in the Bissonnette and Nakagawa trials, (12,13,18,19) to 16 weeks in the Guttman-Yassky and Simpson (16,20) to confirm these results, which are considered to be limited. It was also found that all the included trials described similar limitations that can be resolved by including larger trials of longer duration and patient demographic characteristics in both adult and paediatric patients. The inclusion of active comparator and biomarker measures will aid in elucidating the relative efficacy, safety and mechanisms of action. (12) It is also important to note that efficacy and safety outcomes reported by Nakagawa et al. (2020) in the second openlabel 52-week extension part of their study confirmed that improvement in a modified EASI score was maintained and most adverse events were mild and unrelated to delgocitinib (19) ; however, this part of the study was excluded from the analysis because of the absence of a comparator. Another limitation in the study is that there is a very limited number of studies assessing the efficacy and safety of the wide range of JAK inhibitors, which made it impossible to determine which JAK inhibitor could be more effective that the others or to even determine the best dosage for each drug. Anther limitation of the study is the differences within and between included primary studies that raised the heterogenicity of the data.
Drugs that inhibit cytokine signaling were found to be effective in improving AD. It was found that dupilumab suppresses both IL-4 and IL-13 signaling when is used in clinical settings. (24) Nemolizumab, which is an inhibitor of IL-31 signaling, was also of interest because of its antipruritic effect. (25) As with other cytokines, IL-4, IL-13 and IL-31 exert their biological effects via the JAK/STAT pathway. Clinical evidence show that these two drugs support the efficacy of JAK inhibitors, even though they vary in route of administration (systemic versus topical). Additionally, non-selective, pan-JAK inhibitors can broadly inhibit other cytokine signaling, which is considered to be a better profile compared to JAK1 inhibitors, given that many cytokines are involved in AD pathophysiology. This was confirmed by the results of the meta-analysis with statistically significant efficacy for the pooled-RR of pan-JAK inhibitors compared to no significant efficacy with selective JAK1 inhibitors (RR = 3.37 [95% CI 2.6-4.37] P = 0.95, χ 2 = 1.14, I 2 = 0%) and (RR = 3.92 [95% CI 0.66-23.44] P = 0.003, χ 2 = 9.04, I 2 = 89%), respectively.
All studies that moved from phase-2 to phase-3 showed similar results, which denotes that JAK inhibitors are considered to be a safe and effective treatment with no significant risk of AEs from their use.

Conclusion:-
In conclusion, this study showed that JAK inhibitors are effective novel medications with significant improvement of moderate-to-severe AD with overall tolerated safety in both the topical and oral forms. It is recommended that further studies with longer durations and larger sample sizes that cover a wider range of demographics are conducted