Objective Semapimod, a small molecule known to inhibit proinflammatory cytokine activity, was studied to determine the optimal dose necessary to achieve a response in patients with moderate to severe Crohn's disease (CD).
Methods A randomised, double-blind, placebo-controlled trial (CD04) was carried out followed by an open-label extension study (CD05). The trial was conducted in international multicentre outpatient clinics and included patients with moderate to severe CD (Crohn's Disease Activity Index (CDAI) 250–400). Placebo was administered for 3 days; 60 mg semapimod intravenously for 1 day with placebo for 2 days; or 60 mg semapimod intravenously for 3 days. Participants who completed CD04 could participate in the open-label extension study, CD05, to receive up to five additional semapimod HCl 60 mg daily doses three times every 6–8 weeks. The main outcome measures were CDAI, Inflammatory Bowel Disease Questionnaire (IBDQ), Crohn's Disease Endoscopic Inflammation Score (CDEIS) and serum C-reactive protein (CRP) concentration.
Results 152 patients were randomised in CD04. Responses for 1 and 3 day regimens were similar to placebo for CDAI (p=0.82), IBDQ (p=0.85), CDEIS (p=0.57) and CRP (p=0.40). The only noteworthy treatment-related safety finding was infusion reaction (phlebitis): 7.3, 34.8 and 62.7% for the placebo and 1 and 3 day semapimod treatment groups, respectively (p<0.001). In the open-label CD05 study (included=119 patients) a posthoc analysis showed that the mean CDAI improved in patients receiving 6 compared with ≤3 cumulative doses (204.1±83 vs 251.4±103.05, p=0.006).
Conclusions Single and 3 day dosing of semapimod (≤180 mg) was ineffective for the treatment of moderate to severe CD. However, cumulative dosing ≥360 mg was associated with decreased CDAI in a limited number of patients.
- Clinical trials
- Crohn's disease
- MAPK inhibitor
Statistics from Altmetric.com
Tumour necrosis factor α (TNFα) plays a central role in the initiation and amplification of the inflammatory reaction seen in Crohn's disease (CD).1 2 Anti-TNFα monoclonal antibodies have been extensively used for the treatment of CD during the last decade,3–6 but significant concerns of loss of response and the risk of side effects may limit the use of this approach.7–9 TNFα secretion by T cells and macrophages is under the control of mitogen-activated protein kinases (MAPKs), intracellular signal transduction proteins involved in transcriptional and post-transcriptional regulation of many proinflammatory cytokines and T helper 1 (Th1) differentiation, and increased activation of MAPK has been demonstrated in inflamed CD mucosa.10–12 MAPK inhibitors are small molecules that block MAPK phosphorylation/ATP binding, and several of these compounds have been effective in experimental models of inflammation.13 14 Semapimod, a synthetic guanylhydrazone compound, selectively inhibited TNFα synthesis in lipopolysaccharide (LPS)-stimulated monocytes and macrophages.15 16 In a pilot study, 8 or 25 mg/m2 of semapimod intravenously once daily for 10 days over a 12 day period showed clinical and endoscopic benefit in patients with moderate to severe CD.17
The current study was designed to examine the effect of short-course intravenous administration. An open-label extension evaluated the long-term safety and efficacy of cumulative doses of semapimod on CD activity.
The randomised CD04 study was conducted at 28 centres in five countries. The local Investigational Review Board or Ethics Committee at each site approved the study, and informed consent was obtained from all participating subjects. Inclusion criteria were: age ≥18 years; diagnosis of CD based on accepted clinical radiological or endoscopic criteria; and a Crohn's Disease Activity Index (CDAI) of 250–400 points.18 Permitted concomitant treatment included: methotrexate or azathioprine/6-mercaptopurine at a stable dose for at least 4 or 10 weeks, respectively, prior to study entry; steroids (no more than 20 mg/day prednisone or the equivalent) for at least 2 weeks and on a stable dose for those 2 weeks, and antibiotics for CD for at least 2 weeks and on a stable dose for those 2 weeks. The washout period for CD medication was 4 weeks prior to screening, except for infliximab, which had an 8 week washout requirement.
The CD04 part of this investigation was a multicentre, double-blind, placebo-controlled, parallel-group study. All patients were randomly assigned to receive intravenous infusion of one of the following treatments once daily for 3 days: (1) placebo; (2) semapimod 60 mg on day 1 followed by placebo daily for 2 days; or (3) semapimod 60 mg daily for 3 days. Each 6 ml vial was diluted in 100–250 ml of 5% dextrose and water, and administered over a period of 1–3 h. Consenting patients were offered endoscopy at baseline and at ∼4 weeks after the start of therapy, but this was not required. Patients who complied with the protocol were invited to participate in an open-label study, CD05. In this follow-up study, all participants received semapimod 60 mg daily three times every 6–8 weeks for up to five courses of therapy. Responders could enter the follow-up study 6–8 weeks after the initial treatment CD04 course. CD04 non-responders at the day 29 visit were allowed to enter the CD05 open-label study at that point or at any time up to the final visit (day 57). Response in these two studies was defined as a decrease of at least 70 points in the individual's CDAI from baseline to day 29. Remission was defined as a CDAI of ≤150 at day 29.
The CDAI score was calculated and the Inflammatory Bowel Disease Questionnaire (IBDQ)19 was completed at baseline, and on days 8, 14, 29, 48 and 57 of the CD04 study. C-reactive protein (CRP) was measured at the same time points. Safety was assessed at each visit by questioning the patients about adverse events and concomitant medications, and by performing a physical examination and assessment of laboratory studies.
The primary end point for the CD04 study was the proportion of patients responding at day 29. Secondary end points for the CDAI included remission rate (percentage of patients with a CDAI ≤150 at day 29 and, independently, at day 57), the proportion of patients with a decrease of at least 100 points in the CDAI, and a quantitative change in CDAI from baseline at both day 29 and day 57. Other secondary end points included change in IBDQ, change in the Crohn's Disease Endoscopic Inflammation Score (CDEIS) (for patients with follow-up endoscopy) and change in CRP concentration.
The primary end point for the CD05 study was originally designed to be the duration of response. However, the results of CD04 (reported below) became known while CD05 was ongoing, and the decision was made to terminate CD05 before most patients had completed all five courses. This resulted in modifying the CD05 end point to assess efficacy based on the effect of cumulative dose of semapimod on CD activity. Thus, at baseline and subsequent time points in CD05, the three treatment groups from study CD04 were compared with respect to response to treatment, as well as mean and median CDAI and IBDQ scores over time. Secondary end points included median CDAI change over time, intrapatient change in CDAI over time, the remission rate and its duration, and changes in median IBDQ and CRP over time.
At baseline, the CD05 study patients had to meet the same concomitant medication criteria as those for study CD04. There were to be no changes in CD medications between the time of CD05 screening and the start of study treatment. If at all possible, all the patients were to be kept on the same anti-CD medications at constant doses up to day 29.
Adverse events in both studies were defined as: grade 1 (mild), no interference with function; grade 2 (moderate), some interference with function; grade 3 (severe), marked interference with function and daily activities; and grade 4 (very severe/life threatening), may result in extremely marked interference with function and daily activities.
The sample size was based on the expected proportion of patients with CDAI reduction of at least 70 points from baseline up to day 29 (primary efficacy variable). Previous studies indicated that the expected response rate in placebo patients with this severity of CD is ∼25%.3 The response rate with at least one of the active treatments was expected to be >30% higher than the placebo response rate. Fifty patients were required in each treatment group in order to detect a 30% difference among the three treatment groups using 80% power and a two-sided test of hypothesis at a significance level of 0.05, and allowing for a drop-out rate of ∼10%.
Paired sample methods were used to evaluate change from baseline within CD04 treatment groups and analysis of variance (ANOVA) for change from baseline among treatment groups. All efficacy analyses used an intent-to-treat approach patient population which included all patients who received any study medication and had at least one postbaseline assessment for the parameter evaluated. End point parameters in the CD04 study were assessed for changes from baseline at both day 29 and day 57, and at various times over the 8 week study period, except for CDEIS, which was measured post-treatment at day 29 only for those patients who underwent endoscopy.
For CD05, individuals collected the CDAI parameters for 7 days prior to the next 3 day treatment course, and this CDAI score was compared with the score at study baseline to assess intrapatient changes. In a posthoc analysis, mean and median CDAI and IBDQ scores over time were also compared between groups, with groups sorted according to the previous CD04 group.
Demographic and disease-related variables at baseline were summarised and differences between the treatment groups were analysed using the Fisher exact test for qualitative variables and ANOVA for quantitative variables. Pairwise comparisons were performed if the overall comparison among groups was significant. A logistic regression model was used to evaluate the primary end point and compare treatment groups. A two-sided p value <0.05 was regarded as significant. All computations were performed using SAS, version 6.12 or higher. Results are expressed as mean±SD.
A total of 199 patients were screened and 152 patients were enrolled in the CD04 study between October 2002 and May 2004. Compliance was excellent, with 96.7% of patients having all three planned infusions. Accordingly, all analyses for both safety and efficacy were conducted on the total treated population. Figure 1 outlines the disposition of patients and table 1 provides their demographics and baseline disease characteristics.
The majority of patients had CD for a mean of 9–11 years, and the baseline CDAI scores were comparable among the various treatment groups. The proportion of patients with arthritis/arthralgias at baseline was lower in the 3 day treatment group compared with the placebo and 1 day treatment groups (p=0.05).
All except four patients (2.6%) had received prior treatment for CD and 130 patients (86%) were taking concomitant CD treatment at the time of study entry. Overall, 85 patients (56%) had received prior anti-TNFα treatment, principally infliximab. The number and nature of CD medications being used at the time of study entry were similar among the treatment groups.
As shown in table 2, one course of neither of the two tested semapimod regimens was more effective in treating CD than placebo for any parameter tested.
There were no significant differences among the treatment groups for response (70- and 100-point response rates), remission or mean/median CDAI values throughout the CD04 study. Response rates and CDAI scores were also similar when analysed separately for patients who had or had not undergone prior anti-TNFα treatment.
The IBDQ scores were 115±27, 121±31 and 122±30 at baseline and 136±38, 138±34 and 140±33 at day 29 for the placebo and for the 1 and 3 day semapimod groups, respectively (p=0.78). The CDEIS scores (n=22 for the endoscopic substudy) were 6.12±6.46 (n=5), 17.25±11.45 (n=10) and 14.81±10.54 (n=7) at baseline and 4.27±7.31, 21.01±23.59 and 12.70±9.18 at day 29 for the placebo, 1 day and 3 day semapimod, respectively (p=0.57).
Almost all patients experienced one or more adverse event. Table 3 presents adverse events considered to be related to treatment occurring in at least 10% of patients in any treatment group.
The per patient incidence of having any adverse event and of having a grade 2 or higher adverse event was slightly higher for patients who received 3 days of treatment (36, 70%) than for either the placebo group (30, 54.5%) or for patients who received 1 day of semapimod (29, 63%), a difference primarily due to infusion site reactions. Semapimod HCl clearly demonstrated dose-related vasotoxic effects, with infusion site reactions such as phlebitis, infusion site pain and infusion site redness occurring in 4 (7.3%) 16 (34.8%) and 32 (62.7%) in the placebo, 1 and 3 day semapimod groups, respectively (p <0.001). Two patients, both in the 3 day treatment group, chose to discontinue treatment due to infusion site reactions. Infusion site reactions were generally managed by using as large a vein as possible and slowing the rate for subsequent infusions.
Two patients in the 1 day (4.3%) semapimod group, five (9.8%) patients in the 3 day semapimod group and no patients in the placebo group had oral paraesthesias. These were mild and self-resolving and none resulted in discontinuation of the study medication.
Laboratory abnormalities were mild and not clinically or statistically significantly different among the groups. Alanine aminotransferase (ALT) elevations occurred in 7.3, 8.7 and 11.8%, creatinine increases in 3.6, 4.3 and 15.7%, and lymphopenia in 16.4, 21.7 and 29.4% of patients in the placebo, 1 day and 3 day semapimod groups, respectively.
The CD05 open-label extension study enrolled 119 patients or 78.3% of those entered into the CD04 study, with each of the three dosing groups equally represented. As with CD04, the CD05 study also measured CDAI scores and determined treatment efficacy by analysing the rate of responders at day 29 after each 3 day treatment course. Once the results of CD04 were available and showed no benefit of the medication compared with placebo, the CD05 study was halted before all patients had gone on to complete all five courses of therapy. A total of 111 patients received at least one 3 day course in CD05, and 16 patients completed all five courses. Thus, the CD05 study represents more than two hundred 3 day courses of exposure to semapimod, with some patients having had six courses (CD04 + CD05) over ∼48 weeks. In a posthoc analysis of the CD05 results, we focused our comparisons on patients who had been treated with placebo and 3 days of semapimod in the CD04 study, since we expected that these two groups would have the greatest potential of showing a difference due to the treatment effect of cumulative doses and repeated courses of treatment, if such an effect existed. At the end of the second CD05 course, the response rate was 81% (17/21) compared with 73.7% (14/19), and the remission rate was 42.9% (9/21) for the CD04 3 day course compared with 31.6% (6/19) for the placebo group (p=NS (non-significant)).
For patients who entered the CD05 study, there was no difference at baseline in mean CDAI score for those who had been exposed to 3 days of semapimod in the CD04 study compared with those exposed to placebo (p=0.496) (table 4).
Following course 1, the mean intrapatient change from baseline in CDAI at day 29 for the group with 6 days cumulative exposure to semapimod (3 days in CD04 plus 3 days in CD05) showed a significant improvement compared with each patient's own baseline scores. After the completion of course 2 (9 days and 6 days cumulative dosing for the former CD04 3 day and placebo groups, respectively), the difference from baseline in CDAI for the 9 day group compared with their own baseline scores remained significant (p=0.022, n=21). Unlike the changes seen in the 3 day CD04 group, the difference was not significant for patients in the 6 day exposure group. At the end of course 2, three of 19 patients (15.8%) in the 6 day exposure group were responders based on a scale of a ≥100 point drop in CDAI from baseline versus 9/21 (42.9%) in the 9 day exposure group (p=0.089), again indicating that 9 days of cumulative or repeated exposure to semapimod HCl appeared beneficial in these patients. All patients with nine exposures (three courses of therapy) who had a decrease in their CDAI of least 100 points at the end of course 2 (9/21, 42.9%) were also in remission, with a CDAI of ≤150.
With all patients now receiving 3 day courses of semapimod 60 mg every 6–8 weeks in an open-label manner, the CRP values across all treatment groups declined over time, parallelling the decline in CDAI scores (figure 2).
Similarly to the safety results seen in the CD04 randomised study, common treatment-related adverse events in the CD05 study were infusion site reactions, occurring in 55 out of 119 (46.2%) of the CD05 study participants. Additionally, mild, transient circumoral paraesthesias or dysaesthesias occurred in 24 (20.2%) of them. Seven CD05 patients (5.9%) discontinued the study because of infusion site reactions, none due to any other adverse event. Paraesthesias generally occurred during infusion and, when the start and stop times were recorded, lasted from 0.5 to 9 h (median 3 h). All of these events were grade 1, and none resulted in interruption or discontinuation of the study medication.
Transaminase and creatinine were mild (grade 1 (<2.5× ULN (upper limit of normal) for transaminase, <1.5× ULN for creatinine) except for one grade 2 (2.5–5× ULN for transaminase, 1.5–3× ULN for creatinine) in each group) and transiently elevated in 21 (17.6%) and 8 (6.7%) patients at some time during the study. No patient discontinued treatment due to a transaminase or creatinine elevation. Moreover, there was no trend towards increasing creatinine or transaminase levels with increasing time among patients on study CD05. Common, generally low-grade infectious adverse events occurred among 32 (26.9%) study participants. There were no opportunistic infections suggesting immunosuppression.
Despite significant improvements in CD treatment, specifically the biological agents targeting TNFα, a substantial percentage of patients remain suboptimally treated. This may be due to primary non-response to biological treatments, to loss of response, to the potential for significant side effects or to the cost of these agents. Targeting TNFα and additional proinflammatory cytokines using small molecules that block intracellular signalling pathways involved in cytokine secretion thus seems to be a rational approach. Semapimod is a small molecule with anti-inflammatory effects that have been demonstrated in several models of inflammation and inflammatory diseases.12–14 17 The mechanism of action of semapimod may be post-transcriptional, targeting proinflammatory cytokine production (TNFα, interleukin 1 β (IL-1β), IL-6, macrophage inflammatory protein-1α,β (MIP-1α,β)) regulated by MAPK (p38 and JNK).20 Semapimod-induced impaired MAPK signalling and decreased cytokine production in murine macrophages was demonstrated by Lowenberg et al.21 In vitro kinase assays revealed macrophage c-Raf as the direct molecular target of semapimod. Moreover, expression of phospho-MEK, the substrate of Raf, significantly decreased in patients with CD who had a good clinical response to semapimod, but not in a clinically non-responsive patient.
Promising positive results from an open-label phase II study had been reported by Hommes et al,17 supporting further clinical assessment of semapimod. The 10 day administration protocol used in the pilot study, however, was uncomfortable and inconvenient for patients and time-consuming for physicians. Thus, we embarked on a shorter course, double-blind placebo-controlled trial in which patients with CD were treated with one or three doses of semapimod given over the course of 3 days and compared the results with placebo (CD04). We then conducted an open-label extension study (CD05).
In the CD04 randomised study, neither a 1 nor a 3 day course of 60 mg intravenous semapimod (a 180 mg cumulative dose) was effective in inducing any response or remission in active CD compared with placebo as measured by changes in the CDAI, and no significant differences were noted for the secondary end points of CDEIS, IBDQ or CRP. The major side effect was a local response at the infusion site. While no significant benefit was associated with this single, short-term exposure, a posthoc analysis of the longer term CD05 study demonstrated that greater cumulative exposure to semapimod appeared to be associated with an improvement in CDAI. The differences in mean change in CDAI between patients with more exposures was greater when compared with those with fewer exposures at day 29 of course 1 in the CD05 study, and intrapatient improvements were significantly different from baseline at this time point (p=0.006), indicating that additional courses of treatment may have been effective, although statistical considerations have modest significance in an open-label study with a low number of patients.
A greater cumulative dose of semapimod in the CD05 study was also associated with a significant decrease in CRP levels. These results are consistent with the previously published results of a pilot open-label study.17 The similarity in efficacy between that investigation and our long-term treatment study may be the higher cumulative doses used.
One unanswered question is whether it was the repeated days of exposure, the cumulative drug dose or the combination of these two factors that led to the observed improvement in some of our study patients. Perhaps it is necessary to load at least 360 mg (6 days×60 mg/day) in the initial induction phase in order to observe an efficacious effect of semapimod HCl. While it may have been beneficial to test higher daily doses of semapimod, the occurrence of phlebitis was dose limiting in the CD04 study and so we were unable to test above the daily 60 mg dose by the intravenous route. The phlebitis may be the result of irritation resulting from the strong acidity of the HCl counterion, which might be potentially overcome using an alternative intravenous or an oral formulation. In addition, >3 days of treatment per course is considered impracticable.
In addition to its potential clinically beneficial effect, the advantages of using semapimod are that it is a small molecule, it is not associated with immunogenicity, its defined mechanism of action enables targeted modulation of proinflammatory cytokine production and, potentially, it may reduce risk of treatment-induced infections and malignancies.
There are several factors that need to be considered when interpreting the results of our study. Trials involving biological agents have shown that a drug loading strategy is required to optimise clinical response, and our not having one could have affected the observed response rate. A limitation of the CD05 study was the open-label design and the considerable number of drop-outs (the latter could indicate that the sicker patients went on to another treatment), which weaken any statistical significance of its results. Unfortunately, we were unable to continue the CD05 study long enough to gather data to determine its therapeutic outcome because the lack of efficacy in the CD04 study led to its premature termination.
In conclusion, semapimod in a single course of 60 mg, or 180 mg given over 3 days was ineffective in ameliorating active CD. The CD05 study showed that cumulative dosing ≥360 mg was associated with significant clinical improvement, although this finding was observed during open-label testing with analyses performed posthoc in a limited number of subjects. Phlebitis was a significant and dose-limiting side effect, and future clinical studies performed with intravenous semapimod HCl need to address the issue of phlebitis. Oral daily dosing using a new salt form, currently being developed, may allow a higher daily dose, which would be more convenient and potentially more effective for the treatment of CD.
Significance of this study
What is already known about this subject?
Anti-TNFα monoclonal antibodies are used for the treatment of CD.
MAP kinases (MAPKs) are intracellular signal transduction proteins controlling TNFα secretion by T cells and macrophages. Increased activation of MAPKs has been demonstrated in inflamed CD mucosa.
Semapimod, a synthetic guanylhydrazone compound, selectively inhibits TNFα synthesis in stimulated monocytes/macrophages.
Intravenous semapimod was clinically and endoscopicaly beneficial in an open-label pilot study in patients with moderate to severe CD.
What are the new findings?
One course of semapimod was not more effective than placebo in treating CD.
Cumulative dosing of semapimod in an open-label extension study was associated with significant clinical improvement compared in a posthoc analysis. This was accompanied by decline in CRP values.
Intravenous semapimod HCl has significant dose-related vasotoxic effects.
How might it impact on clinical practice in the foreseeable future?
Semapimod, a small molecule targeting TNFα and additional proinflammatory cytokines by blocking MAPK, may be a potential therapeutic strategy in CD.
Participataing Centres in CD04/CD05: Iris Dotan, Tel Aviv Sourasky Medical Center Tel Aviv, Israel; Daniel Rachmilewitz, Shaare Zedek Hospital, Jerusalem, Israel; Stefan Schreiber, University of Kiel, Kiel, Germany; Rami Eliakim, Rambam Health Care Campus, Haifa, Israel; C Janneke van der Woude, Erasmus Medical Center, Rotterdam, The Netherlands; Asher Kornbluth, The Henry D Janowitz Division of Gastroenterology, The Mount Sinai Medical Center, New York, NY, USA; Alan L Buchman, Division of Gastroenterology and IBD Center, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA; Shimon Bar-Meir, Chaim Sheba Medical Center, Tel Hashomer, Israel; Bernd Bokemeyer, Gastroenterologische Fachpraxis, Minden, Germany; Eran Goldin, Hadassah Medical Center, Jerusalem, Israel; Christian Maaser, University of Munster, Munster, Germany; Uma Mahadevan, UCSF Center for Colitis and Crohn's Disease, San Francisco, CA, USA; Ursula Seidler, Medizinischen Hochschule-Hannover, Hannover, Germany; Jörg C Hoffman, Charité, Campus Benjamin Franklin, Berlin, Germany; Douglas Homoky, Gastroenterology Associates, Kingsport, TN, USA; Paul Rutgeerts, Academic Hospital Gasthuisberg, Leuven, Belgium; Daniel Hommes, Department of Gastroenterology and Hepatology, Leiden University Medical Center, Leiden, The Netherlands; Seymour Katz, Long Island Clinical Research Associates, Great Neck, NY, USA; R Stockbrugger, Academisch Ziekenhuis Maastricht Maastricht, The Netherlands; W Stremmel, Universitats Klinikum Heidelberg, Heidelberg, Germany; Douglas Wolf, Atlanta Gastroenterology Associates, Atlanta, GA, USA; Kevin Casey, Rochester General Hospital, Rochester, NY, USA; Oliver DeWit, Cliniques Universitaires Saint-Luc, Brussels, Belgium; AA van Bodegraven, Free University (Vrije Universiteit), Amsterdam, The Netherlands; Jeffrey Levine, Gastroenterology Associates, Bristol, TN, USA; Donald Lipkis, Institute of HealthCare Assessment, San Diego, CA, USA; C Pehl, Stadtisches Krankenhaus Munchen-Bogenhausen, Munchen, Germany; Brian Hudes, Advanced Gastroenterology Associates, Suwanee, GA, USA. Esther Eshkol is thanked for editorial assistance.
Funding The CD04 and CD05 studies were funded by Cytokine PharmaSciences.
Competing interests Dan Hommes has received funding from Cytokine Pharmasciences, Inc. as a consultant for the semapimod program. Terry Plasse was an employee of Cytokine Pharmasciences, Inc. at the time the study was conducted. Barbara Powers is Vice President of Clinical Development for Cytokine Pharmasciences, Inc., the study sponsor.
Patient consent Obtained.
Ethics approval The local Investigational Review Board or Ethics Committee at each site apporved the study.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.