After the diagnosis of UC, individualized treatment according to the severity of the disease is the key point of disease management. The short-term goal of UC treatment is to remission symptoms, and the long-term goal is to achieve sustained clinical remission, endoscopic improvement, even histological healing. The first-line drug for induction and maintenance treatment of mild to moderate UC is 5-ASA. Different 5-ASA preparations do not differ in efficacy and safety. Moderate to severe UC should be treated with corticosteroids during the induction period. If there is no response, biological agents or small molecule drugs should be selected for induction. Available biological agents and small molecule drugs include: (1) anti-TNF-α drugs: Infliximab, Adalimumab and Golimumab can effectively induce remission with moderate to severe UC[12–15]. Among them, Infliximab is the earliest and most widely used biological agent and it works rapidly. The SUCCESS[16] trial showed that in terms of inducing clinical remission, the combination of AZA and Infliximab was more effective than AZA or Infliximab alone (39.7% vs. 23.7% vs. 22.1%, p = 0.032). The endoscopic improvement rate was also similar at week 16 (62.8% vs. 36.8% vs. 54.6%, p = 0.001). (2) Leukocyte adhesion molecule antibody: Vedolizumab blocks the binding of α4β7 integrin and MAdCAM-1 in the intestine and is approved by the FDA for moderate to severe UC. (3) Anti-IL-12/23 antibody: Ustekinumab, which has been approved by the FDA for moderate to severe UC that fails to respond adequately to TNF inhibitors or is intolerant and contraindicated, can block IL-12/23-mediated inflammation. Anti-IL-12/23 antibody also has a large number of mature phase III pipelines in the clinic[17]. (4) JAK inhibitor: Tofacitinib is the first non-selective JAK inhibitor approved by FDA. Upadacitinib was also approved as a selective JAK inhibitor in March 2022[18]. (5) S1P receptor modulator: Ozanimod isolates lymphocytes from peripheral lymphatic organs to prevent inflammation, and has been approved for the treatment of moderate to severe UC in 2021[19]. Patients with steroid-induced remission can be transitioned to mercaptopurine or biological therapy, and patients with biological agents or small molecule drugs should continue to use the same drug. In summary, there are many drugs for the treatment of moderate to severe UC, but the number of "head-to-head" clinical trials between these drugs is very limited. We used network meta-analysis to evaluate the efficacy and safety of drugs for the treatment of moderate to severe UC. There were no restrictions on the species of aminosalicylic acids, hormones, and immunosuppressants. Biological agents and small molecule drugs were included in FDA-approved drugs, and there were no restrictions on the use of drugs alone or in combination. A total of 18 studies were found to meet the inclusion and exclusion criteria, including 7873 patients with moderate to severe UC and the following 11 interventions: Infliximab, Infliximab/AZA, Adalimumab, Vedolizumab, Ustekinumab, Golimumab, Tofacitinib, Upadacitinib, Ozanimod, tacrolimus, Placebo.
The study Lawrance 2019[20] involving tacrolimus was not included in the network meta-analysis due to the small sample size, which is described separately here: In this study, patients with ulcerative proctitis with Mayo score of 6–12 were treated with tacrolimus suppository induction therapy, and 11 patients in the control group and 10 patients in the experimental group. After 8 weeks, 8, 5 and 8 patients respectively achieved clinical response, clinical remission and endoscopic improvement in the experimental group, and 1, 0 and 1 patient in the control group. We calculated the clinical response rate ratio (p = 0.040), clinical remission rate ratio (p = 0.103) and endoscopic improvement rate ratio (p = 0.040). There were significant differences in clinical response rate and endoscopic improvement rate between the tacrolimus group and the control group, but not in clinical remission rate. No information on the safety of tacrolimus was provided. This study showed a good effect of tacrolimus suppository in the treatment of ulcerative proctitis, but it has the defects of small sample size. RCTS with larger sample sizes are still needed to prove its efficacy and safety.
Network meta-analysis of induction of clinical response showed that all interventions had superior induction clinical response rates than placebo. The top three interventions were Upadacitinib, Infliximab/AZA, and Infliximab, and Adalimumab was the last. This may be related to the differences in pharmacokinetics, bioavailability, drug's structure, different doses and routes of administration. Upadacitinib is a selective JAK1 inhibitor, which is a small-molecule drug. Compared with TNF inhibitors and integrin inhibitors, the small-molecule drugs have many advantages: (1) no immunogenicity, no neutralizing antibodies; (2) It can resist the decomposition of gastric acid after oral administration, rapidly reach the systemic circulation, has high bioavailability, and reaches the peak blood concentration within 1 ~ 2h; (3) Age, sex, body weight and disease severity did not seem to have any effect on the mean plasma concentration; (4) Biological agents can produce long-lasting inhibitory effects on the corresponding targets, while small molecule drugs can regulate the inflammatory response in a progressive and reversible manner[21–23]. Because the JAK family mediates a variety of cytokine signaling, they can cause a variety of adverse reactions. Upadacitinib is more selective than Tofacitinib, and its inhibitory effect on JAK1 is superior than JAK2 and JAK3. In addition to reducing some adverse events related to JAK2 and JAK3 inhibition, Upadacitinib improved efficacy, with a 96.0% probability of ranking first in the clinical response rate during the induction phase. Infliximab, a human-mouse chimeric monoclonal antibody with stronger immunogenicity, binds to TNF-α at the site of inflammation by intravenous administration and inhibits inflammation with 100% bioavailability. Therefore, it has a rapid onset of action and can quickly induce clinical responses in patients with moderate to severe UC[24]. Anti-drug antibodies leading to decreased serum Infliximab concentration is the main reason for the primary or secondary loss of respond to Infliximab in some patients. Immunomodulators combined with Infliximab can reduced anti-drug antibody production and increase Infliximab trough serum concentrations in patients with Crohn's disease or rheumatoid arthritis[25, 26]. However, some studies have shown that the combination of immunosuppressants in patients with moderate to severe UC has no significant change in the pharmacokinetic parameters of Infliximab[27], and the reason for this difference is still unclear. Adalimumab is a recombinant humanized monoclonal antibody containing only human peptide sequence. The route of administration is subcutaneous injection. The process of absorption and distribution is slow, so it is not dominant in the treatment of induction period.
Network meta-analysis of induction of clinical remission showed clinical remission rates were superior for all interventions than placebo. The top two were Upadacitinib and Infliximab. Upadacitinib still has a prominent position in inducing clinical remission. Because the Panaccione 2014 study did not report clinical remission of Infliximab/AZA, it was not possible to compare the ranking of Infliximab/AZA and Infliximab. However, based on the ranks of clinical response and endoscopic improvement in the induction phase and for the reasons described in preceding part of the text, we speculate that if the clinical response outcome had been reported, Infliximab/AZA would have ranked superior than Infliximab. Vedolizumab selectively blocks the binding of α4β7 integrin to MAdCAM-1 and inhibits the migration of lymphocytes into the lamina propria and lymphoid tissue of the intestine, but it cannot block the lymphocytes that have already played a role in the inflammatory site. Therefore, Vedolizumab has a poor efficacy in the induction phase.
Network meta-analysis of induction of endoscopic improvement showed that all interventions were superior than placebo, and the top three were still Upadacitinib, Infliximab/AZA, and Infliximab. The highly selective JAK1 inhibitor Upadacitinib is very effective in inducing clinical response, clinical remission, and endoscopic improvement, while Infliximab, as the first biological agent for the treatment of UC, still plays a very important role in inducing remission. Combination of immunosuppressive agents may be more effective. Ozanimod, Tofacitinib, and Ustekinumab also showed promising efficacy in inducing endoscopic improvement. Although Adalimumab and Vedolizumab are better than placebo in the induction period, they are far less effective than the other interventions.
Network meta-analysis of maintenance of clinical remission showed that all interventions had superior clinical remission rates than placebo. The top three were Upadacitinib, Tofacitinib and Vedolizumab. It can be seen that small-molecule drugs such as JAK inhibitors are more effective in maintaining clinical remission, and selective inhibitors are more effective than widely-inhibitors. Although the efficacy of Vedolizumab in the induction phase was not outstanding, it showed superior ability to maintain remission, which is related to the mechanism of action of Vedolizumab described in preceding part of the text. Vedolizumab targets the upstream of the entire inflammatory pathway, so it has a slower onset of action. Once it takes effect, it exhibits a sustained anti-inflammatory effect. Ustekinumab, Golimumab, Tofacitinib, Upadacitinib, Ozanimod, and Vedolizumab other than the VARSITY 2019 study patients who had a clinical response after induction were rerandomly assigned, and those who did not have a clinical response were assigned to the open-label or other study phase. This approach would have excluded patients with a primary nonresponse from the maintenance phase, potentially amplifying the effect in the maintenance phase. In the subgroup analysis according to rerandomization during the maintenance period, Upadacitinib, Tofacitinib, and Vedolizumab were still the top three, suggesting that this difference of design did not have a large effect on the probability ranking.
Network meta-analysis of maintenance of endoscopic improvement showed that all interventions had superior endoscopic improvement rates than placebo, and the top four were Upadacitinib, Tofacitinib, Infliximab, and Vedolizumab. After subgroup analysis, Upadacitinib still ranked first, and the order of Infliximab, Tofacitinib, and Vedolizumab changed slightly but the probability was not significantly different. In conclusion, Upadacitinib, Infliximab, Vedolizumab, and Tofacitinib all showed promising efficacy in the maintenance phase. Ustekinumab, Golimumab, and Ozanimod performed reasonably well in both induction and maintenance periods.
In addition to efficacy, safety is also an important aspect of evaluating the drugs. Network meta-analysis of AE incidence showed that all interventions were no statistical differences compared to placebo. Ozanimod and Golimumab ranked in the top two and Vedolizumab ranked last. S1P receptors mediate a wide range of biological effects. Ozanimod belongs to S1P1 and 5 receptor modulators. Because the sequence of these 5 isoforms is highly homologous, it is easy to cause side effects due to off-target effects. Golimumab, the newest member of the TNF inhibitor family, is effective in inducing and maintaining UC patients with a more convenient route of administration than Infliximab (subcutaneous vs. intravenous), and longer dosing cycle than Adalimumab (4 weeks vs. 2 weeks)[28]. However, the FDA issued a black box warning that Golimumab may lead to an increased chance of serious infections that may cause hospitalization or even death, especially in patients with combined immunosuppressors or hormones. Vedolizumab has a favorable safety, which may be related to its unique mechanism of action, and its excellent ability to maintain clinical remission and endoscopic mucosal improvement makes it highly competitive with Infliximab in patients with primary and secondary non-responsive, or refractory UC. Infliximab combined with immunosuppressants may improve the efficacy but also reduce the safety, which should be considered comprehensively in clinical selection. Ustekinumab and Tofacitinib also had good efficacy and safety in the induction and maintenance phases.
All pairwise comparisons showed no statistically significant difference in the incidence of SAE. The ranking of probabilities of SAE incidence suggested that Golimumab had the highest incidence and Upadacitinib had the lowest. Taken together with the efficacy, Upadacitinib may have the best combined effect.