Oral Janus kinase inhibitors for treating hospitalized patients with COVID-19: An updated systematic review and meta-analysis of randomized controlled trials

Objectives This study investigated the clinical efficacy and safety of oral Janus kinase inhibitors (JAKis) in the treatment of hospitalized patients with COVID-19. Methods The PubMed, Embase, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov databases were searched for relevant articles written before January 29, 2022. Only randomized controlled trials (RCTs) that assessed the clinical efficacy and safety of oral JAKis in patients with COVID-19 were included. Results In the pooled analysis of the 7 RCTs, the all-cause 28-day mortality rate in the study group receiving JAKis was significantly lower than that in the control group (9.4% [183/1941] vs. 10.9% [184/1687], risk ratio [RR] = 0.69, 95% confidence interval [CI], 0.58–0.81, I2 = 0%). In addition, the risk of 14-day mortality was in the study group was lower than that in the control group (RR = 0.65, 95% CI, 0.46–0.92, I2 = 0%). Finally, the study group and the control group exhibited similar risks of any adverse events (RR = 0.96, 95% CI, 0.89–1.04, I2 = 0%). Conclusions Oral JAKis can significantly reduce the risk of death among patients with COVID-19. In addition, JAKis are tolerable for hospitalized patients with COVID-19.


Introduction
Since the first outbreak of coronavirus disease 2019 (COVID- 19) toward the end of 2019, the COVID-19 pandemic has rapidly become one of the most devastating global problems. 1,2 As of May 3, 2022, 511,965,711 cases of SARS-CoV-2 infections and 6,240,619 COVID-19-related deaths have been confirmed and reported by the World Health Organization. 3 Patients with SARS-CoV-2 infections could exhibit protean clinical manifestations, ranging from asymptomatic to severe acute respiratory distress syndrome (ARDS). 4 Moreover, some patients who have recovered from acute SARS-CoV-2 infections continue to suffer from long COVID-19 or post-acute sequelae of COVID-19. 5 The severity of COVID-19 is significantly associated with a dysregulated immune response and excessive release of proinflammatory agents, resulting in cytokine storm. 6,7 Therefore, in addition to limited direct antiviral agents against SARS-CoV-2, such as remdesivir, molnupiravir, and nirmatrelvir plus ritonavir, 8e10 repurposing anti-inflammation therapy is considered a promising therapeutic approach for patients with COVID- 19.
Several systemic anti-inflammatory and immunomodulatory agents, including corticosteroids and interleukin-6 blockade, exhibit a survival benefit for hospitalized patients with COVID-19. 11,12 Additionally, Janus kinase inhibitors (JAKis), which can ameliorate the inflammatory response and enhance antibody production, have been predicted (with the use of artificial intelligence algorithms) to be a potential therapeutic option against  Several randomized controlled trials (RCTs) have investigated the clinical efficacy of JAKis, including baricitinib, ruxolitinib, tofacitinib, and nezulcitinib, in the treatment of patients with COVID-19, 16e20 and several meta-analyses have assessed their effects based on these RCTs. 21e23 Since the publication of the last of these meta-analyses, one more RCT 24 and two more clinical trials 26,26 have reported relevant findings. Therefore, we conducted this study to provide updated information regarding the usefulness of JAKis in the treatment of hospitalized patients with COVID-19.

Material and methods
This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines, 27 and the study protocol is registered in PROSPERO under number CRD42022307914. This meta-analysis did not require the approval of an institutional review board because the sources of the analyzed data are public and the analysis does not make the data individually identifiable.

Study selection
Two investigators (TSW and YHC) independently screened the titles and the abstracts for eligibility and reviewed the full texts of potentially eligible studies. In the case of disagreement, a third investigator (HJT) was responsible for establishing a consensus. Only RCTs comparing the clinical efficacy and safety of JAKis with placebo or standard of care in hospitalized patients with COVID-19 were included in this review. Clinical efficacy included mortality, clinical improvement, and the requirement of mechanical ventilation (MV) or extracorporeal membrane oxygenation (ECMO). Safety was defined as the risk of adverse events (AEs). We manually searched for additional eligible articles cited in the reference lists of the identified articles. The inclusion criteria was (1) examination of hospitalized patients with COVID-19; (2) use of oral JAKis as the experimental drug; (3) use of placebo or standard of care for comparison; and (4) availability of clinical data for analysis. Exclusion criteria included in vitro studies, case reports, case series, post hoc analysis studies, posters or conference abstracts, and studies without available data for outcome analysis.

Data extraction and outcome measurement
The following data were extracted from each included study: year of publication, study design, study patients, JAKi regimen, clinical outcomes, and risk of adverse events (AEs). Data extraction and quality assessment were conducted independently by TSW and YHC. Any discrepancies were resolved through discussion and, if necessary, evaluation by third investigator (HJT).
The primary outcome was all-cause 28-day mortality, and the secondary outcomes were 14-day mortality, rate of clinical improvement, time of clinical improvement, requirement of MV or ECMO, and risk of AEs.

Assessment of risk of bias
Two investigators (SHL and CMC) independently assessed the risk of bias for each of the included studies using the Cochrane risk-of-bias tool 2.0. 28 Disagreements were resolved through discussion and consensus with a third investigator (CMC).

Statistical aanalysis
Statistical analyses were performed with Review Manager (version 5.3; Nordic Cochrane Centre, Copenhagen, Denmark). Statistical heterogeneity in the results of the trials was assessed using the c2 test and is expressed in terms of I 2 index values. 29 The pooled risk ratios (RRs) or mean differences (MDs) and 95% confidence intervals (CIs) were calculated for outcome analysis using a randomeffects model.

Search results
The search of the online databases yielded a total of 683 studies (n Z 428 from PubMed; n Z 130 from Embase; n Z 71 from Cochrane library; n Z 54 from ClinicalTrial. gov) (Appendix 1). Among these studies, 152 duplicate studies were excluded. We deemed 472 of the studies irrelevant after screening the titles and abstracts and discovering that we could not access the full texts of the publications. After screening the full texts of 59 of the articles, we excluded 52 of the studies. Finally, seven studies 16e19,24e26 fulfilling the inclusion criteria were included in our meta-analysis (Fig. 1).

Study characteristics
Among the seven included RCTs, five 17e19,25,26 employed double-blind designs. Four of the RCTs 17,18,25,26 were multinational studies, and the remaining three were conducted in India, 24 Brazil, 19 and China. 16 One of the studies 24 was a single center study; the other six 16e19,25,26 were multicenter studies. Three JAKis (baricitinib, ruxolitinib, and tofacitinib) were used as interventions in the included studies. Three, two, and two studies evaluated the clinical efficacy of ruxolitinib, 16,25,26 baricitinib, 17,18 and tofacitinib, 19,24 respectively. All of the studies involved hospitalized patients with COVID-19. One of the studies focused on moderate COVID-19, 24 three focused on moderate-tosevere COVID-19, 17e19 and three focused only on severe COVID-19. 16,25,26 A total of 3631 patients (n Z 1944 in the study group, n Z 1687 in the control group) were included in the meta-analysis (Table 1) 24 was open-label design, however, it did not affect the outcome analysis. Therefore, we considered this study 24 exhibited some concerns of randomization and the associated overall bias. All the other six included RCTs 16e19,25,26 exhibit a low risk of bias.

Primary outcome
In the pooled analysis of the RCTs, the all-cause 28-day mortality rate in the study group receiving JAKis was significantly lower than that in the control group, and low heterogeneity was observed among the studies ( Fig. 3). In the leave-one-out sensitivity test to assess the confounding effect of each study, these results remained unchanged.
In the subgroup analysis of the two RCTs that used baricitinib as the study drug, 17,18 the mortality rate in the study group was significantly lower than that in the control group (RR Z 0.63, 95% CI, 0.48e0.81, I 2 Z 0%). In the subgroup analysis of the studies using either of the other two JAKis as the study drug, the mortality rate in the study group was also lower than that in the control group; however, the difference did not reach statistical significance (tofacitinib: RR Z 0.50, 95% CI, 0.16e0.61; ruxolitinib: RR Z 0.78, 95% CI, 0.46e1.30, I 2 Z 15%). In the subgroup analysis of the three RCTs 17e19 focusing on patients with moderate-to-severe COVID-19, the mortality rate in the study group was significantly lower than that in the control group (RR Z 0.62, 95% CI, 0.48e0.80, I 2 Z 0%). In the subgroup analysis of the studies focusing on patients with severe COVID-19, the mortality rate in the study group was lower than that in the control group; however, the

Discussion
This meta-analysis was conducted to assess the clinical efficacy and safety of using JAKis (baricitinib, tofacitinib, and ruxolitinib) to treat hospitalized patients with COVID-19. The results revealed that the study group (patients who   received JAKis) exhibited more favorable clinical outcomes than did the control group, which was supported by the following evidence. First, the all-cause 28-day mortality rate was significantly (31%) lower in the study group than in the control group. Second, this trend remained unchanged in the leave-one-out sensitivity test. Third, the trend was similar across various subgroups defined according to the type of JAKis used, disease severity of COVID-19, and study design. Fourth, the 14-day mortality rate was 35% lower in the study group than in the control group. Finally, the study group had a higher clinical improvement rate, a shorter time to clinical improvement, and a lower rate of MV or ECMO use than did the control group. In addition, the included RCTs exhibited low heterogeneity (I 2 < 50%). Our findings are in line with previous meta-analyses 21,22 of four or five RCTs involving less than 3000 patients. However, our analysis included seven RCTs with a total of 3631 patients and could therefore more effectively verify the clinical effectiveness of JAKis. In summary, our findings indicate that JAKis can significantly reduce the risk of mortality and improve the clinical outcomes of patients with COVID-19, thereby aiding patients and health-care professionals in the fight against COVID-19.
The clinical benefit of JAKis in the treatment of patients with COVID-19 can be explained as follows. Severe COVID-19 may be caused by a dysregulated immune response or cytokine storm, which is characterized by the uncontrolled secretion of many proinflammatory cytokines, including interleukin (IL)-I, IL-2, IL-6, IL-10, IL-18, and IL-33; tumor necrosis factor-a; interferon-g; and macrophage inflammatory protein 1a. 30e32 JAKis can target and downregulate JAK1, JAK2, JAK3, and tyrosine kinase-2 (TYK2), thereby attenuating hyperinflammatory responses. 33 In this metaanalysis, three JAKis e baricitinib, tofacitinib, and ruxolitinib e were evaluated. Baricitinib and ruxolitinib are selective inhibitors of JAK1/JAK2, and tofacitinib is a selective inhibitor of JAK 1/JAK3, with functional selectivity for JAK2. Baricitinib exhibits additional potential to impair the cellular viral entry of SARS-CoV-2. 15,34 Although the effectiveness of JAKis was consistently demonstrated in this meta-analysis, the clinical efficacy of different JAKis could vary, and additional RCTs are warranted to clarify their differences in clinical practice.
Beyond clinical efficacy, we also assessed the tolerability of JAKis. Overall, the patients who received JAKis did not exhibit a higher risk of any AEs or serious AEs than did the patients in the control group. Furthermore, the use of JAKis did not increase the patients' risk of infection, sepsis, or septic shock. These findings are consistent with those of previous studies 11,12,21e23 and indicate that JAKis are safe for use in the treatment of patients with COVID-19.
However, this study has some limitations. First, the numbers of studies on each JAKi, as well as the numbers of patients in each study, were limited. Only 289 and 674 of the patients received tofacitinib and ruxolitinib, respectively. By contrast, 2558 patients were involved in the study focusing baricitinib. Therefore, we found that only baricitinib was associated with a significant survival benefit, whereas tofacitinib and ruxolitinib could have been associated with marginal effects. Second, because the disease severity of COVID-19 varied among the included RCTs, we could not determine which populations may consistently benefit from JAKis. Therefore, more largescale RCTs are required to evaluate the clinical efficacy of each inhibitor in the treatment of patients with COVID-19 of different severities.
In conclusion, JAKis can significantly reduce the risk of mortality, enhance clinical improvement, and prevent the need for MV or ECMO among patients with COVID-19. In addition, JAKis are tolerable for hospitalized patients with COVID-19.

Ethical approval
Not required.

Funding
This paper was not funded.

Declaration of competing interest
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants, patents received or pending, and royalties.