Effectiveness and safety of omalizumab in chronic spontaneous or inducible urticaria: evaluation of 154 patients

Dear Editor, Out‐of‐trial data on long‐term safety and predictors of effectiveness of omalizumab (anti‐IgE) in chronic urticaria are lacking. We evaluated the outcome of treatment with omalizumab in a large cohort of patients with chronic spontaneous or chronic inducible urticaria (CSU and CINDU, respectively). Data were retrieved via retrospective review of patient records from a tertiary dermatological referral centre (Department of Dermatology, Bispebjerg Hospital, Copenhagen, Denmark). Complete ascertainment of patient records was possible as the department keeps track of all administered biological therapy. A consultant dermatologist (S.F.T.) verified the urticaria diagnosis and assessed response to treatment. Two authors (M.N.G. and S.F.T.) extracted the data and, in cases of disagreement, consensus was obtained by discussion with all the authors. According to Danish law, scientific ethics approval is not needed for retrospective chart reviews. Approval for the study was obtained from the National Board of Health (to examine patient records) and from the Data Protection Agency (to save patient data on file). All patients diagnosed with CSU or CINDU who had received at least one injection of omalizumab from 30 June 2010 to 31 December 2014 were included in the study. Response to omalizumab treatment was graded according to the degree of relief of symptoms reported by the patient and according to the physician's assessment. It was possible to score the clinical response to treatment with omalizumab for each patient as a ‘complete or almost complete response’, ‘partial response’ and ‘no/limited response’ after 3–6 months of treatment. This grading is in accordance with the modified physician global assessment used in other dermatological diseases, where a complete or almost complete response corresponds to ≥ 90% reduction of symptoms; partial response is reduction in symptoms of between 30% and 89%; and no response or a limited response is a reduction in symptoms of < 30%.1 In total, 154 patients were identified; 110 (71·4%) women and 44 (28·6%) men (Table 1). The majority of the patients were diagnosed with CSU (89·0%). Of these, 30·7% had concomitant angioedema, 34·3% had a positive histamine release (HR) test and 22·6% had significant comorbidities, the most common being asthma, diabetes, hypertension and hypothyroidism. The 17 patients with CINDU had delayed‐pressure urticaria (n = 5), cold urticaria (n = 5), cholinergic urticaria (n = 4), symptomatic dermographism (n = 2) and solar urticaria (n = 1). The average age at disease onset was significantly higher for patients with CSU compared with those with CINDU. Patients with CINDU had, on average, a longer duration of symptoms before initiation of treatment with omalizumab than patients with CSU. Of all patients, 39·6% were previously treated with prednisolone and 26·0% had been treated with dapsone or one or more immunosuppressant drugs such as ciclosporin A, azathioprine, mycophenolate mofetil, methotrexate and/or tumour necrosis factor (TNF) inhibitors. We have previously reported that among a subset of 25 patients with chronic urticaria who were naïve to biological drugs or who had received omalizumab, 60% obtained complete or almost complete resolution of urticaria and angioedema after onset of therapy with either adalimumab or etanercept, whereas another 15% of patients experienced a partial response to treatment with these TNF inhibitors.2 3 Characteristics of 154 patients with chronic spontaneous urticaria (CSU) or chronic inducible urticaria (CINDU) treated with omalizumab Data are n (%) unless otherwise indicated. aNumbers are calculated from available data. HR, histamine release. Characteristics of 154 patients with chronic spontaneous urticaria (CSU) or chronic inducible urticaria (CINDU) treated with omalizumab Data are n (%) unless otherwise indicated. aNumbers are calculated from available data. HR, histamine release. The initial dose of omalizumab given was 150 mg every 2 weeks in 45·5% of cases and 300 mg every 4 weeks in 55% of cases (Table 1). At the time of treatment, use of omalizumab for CSU was still off‐label. Therefore, the choice of start dose and duration was based on physician preference and not on a predefined belief of superiority of one treatment regimen over the other. However, of the patients who started on 150 mg every 2 weeks, 83% stayed on the same dose and interval throughout the observation period, whereas 6% were changed to 150 mg every 4 weeks owing to symptomatic remission, and 11% were changed to 300 mg given at intervals every 2–4 weeks owing to symptomatic deterioration. Conversely, of the patients who started on 300 mg every 4 weeks, 87% stayed on the same dose and interval throughout the observation period, whereas 6% were changed to 150 mg every 2–4 weeks or to 300 mg every 6 weeks owing to symptomatic remission, and 7% were changed to 300 mg given at intervals every 2–3 weeks owing to symptomatic deterioration. Unfortunately, we did not collect enough systematic data to provide estimates of time to response or time to relapse after discontinuation of omalizumab. A total of 67% of patients with CSU responded completely or almost completely to treatment; 22% responded partially, whereas 10% had no or limited response to treatment. These numbers are somewhat higher than those observed in the randomized controlled phase III trials of omalizumab in CSU but very much in accordance with other real‐life studies of response to omalizumab in CSU that have used various measures of effectiveness such as reduction in symptoms, need for rescue medication and validated scoring systems of symptomatic activity.4 5 6 7 8 9 10 Among the patients with CSU, there was a larger fraction of complete or almost complete responders to omalizumab among patients with a negative HR test [77·3% vs. 27·3%; odds ratio (OR) 9·07, 95% confidence interval (CI) 2·80–29·32 (P < 0·01)]; among patients without angioedema [72·3% vs. 56·1%; OR 2·05, 95% CI 0·95–4·40 (P = 0·07)]; and among patients with no previous history of treatment with systemic immunosuppressant drugs [71·4% vs. 56·8%; OR 1·91, 95% CI 0·87–4·17 (P = 0·10)]. Among patients with these characteristics, complete or an almost complete response was observed in > 85% of patients. There was a statistically nonsignificant tendency for a better response to treatment among patients with a late age at onset and a shorter duration of urticaria. There was no significant effect of sex or smoking on treatment response. Also, the effect of omalizumab was not significantly different in patients receiving 300 mg every 4 weeks compared with patients receiving 150 mg every 2 weeks (65·8% vs. 68·9%; OR 1·15, 95% CI 0·56–2·38 (P = 0·70)]. Of the 17 patients with CINDU, nine (53%) had a complete or almost complete response to treatment: three of the five patients with delayed pressure urticaria; two of the five patients with cold urticaria; three of the four patients with cholinergic urticaria; and one of the two patients with symptomatic dermographism. The patient with solar urticaria had no response to treatment. Fourteen percent reported adverse effects during treatment with omalizumab. Reported adverse effects were, in descending order of frequency, suspected allergic reactions, gastrointestinal symptoms, injection site reactions, worsening urticaria, headache, dizziness, arthritis and fatigue. The drug was administered in the outpatient clinic, with the exception of a few patients who were treated at home by autoinjection. Patients were monitored 30 min after each injection. The suspected allergic reactions occurred during or immediately after administration of the drug but after retrospective evaluation were probably vasovagal syncopes and not suspected to be anaphylaxis. Worsening urticaria was only observed in two patients in the weeks after one of the two‐first injections, and was probably related to slow or absence of response to treatment rather than to genuine worsening of urticaria. All observed suspected side‐effects were reversible. One year after initiation of omalizumab, 56% of the patients with CSU were still on continuous treatment, while 44% had discontinued treatment owing to any cause. The median overall drug survival for the first round of treatment (discontinuation owing to any cause) in patients with CSU was 406 days (range 325–487); 462 days (range 305–619) for women and 364 days (range 332–496) for men (P = 0·41 for the difference between men and women). The most common causes of discontinuation were symptomatic remission or lack of effect, whereas a smaller proportion discontinued because of nonadherence to scheduled ambulatory visits, adverse effects or pregnancy (Table 2). Of the patients who discontinued treatment with omalizumab owing to symptomatic remission, 58% (n = 14/24) resumed treatment one or more times at a later stage owing to recurrence of symptoms. Among these, the effect of omalizumab was comparable with the first round of treatment. Of interest, 10 patients with CSU did not resume treatment with omalizumab owing to long‐term symptomatic remission and were eventually discontinued from the department. This corresponds to an overall remission rate of 7% out of the 137 patients with CSU, which is lower than the inferred 1‐year remission rates observed in most but not in all previous studies of the natural history of CSU.11 12 13 14 Outcome of treatment 1 year after initiation of omalizumab in 154 patients with chronic spontaneous urticaria (CSU) or chronic inducible urticaria (CINDU) aOf the patients on continuous treatment (n = 20), 17 with CSU and three with CINDU had been on treatment for < 1 year by the end of the observation period but had good effect of the treatment and no apparent side‐effects. bOf the 24 patients who discontinued treatment within the first year owing to symptomatic remission, 14 patients resumed treatment ≥ 1 times at a later stage owing to recurrence of symptoms. Outcome of treatment 1 year after initiation of omalizumab in 154 patients with chronic spontaneous urticaria (CSU) or chronic inducible urticaria (CINDU) aOf the patients on continuous treatment (n = 20), 17 with CSU and three with CINDU had been on treatment for < 1 year by the end of the observation period but had good effect of the treatment and no apparent side‐effects. bOf the 24 patients who discontinued treatment within the first year owing to symptomatic remission, 14 patients resumed treatment ≥ 1 times at a later stage owing to recurrence of symptoms. This real‐life study of the effectiveness and safety of treatment with omalizumab in patients with CSU or CINDU showed that around two‐thirds had a complete or an almost complete response to treatment, with no significant difference in effect between the sexes. Predictors of favourable response to treatment were presence of CSU, no angioedema, negative HR test, older age, shorter duration of symptoms and no history of treatment with systemic immunosuppressant drugs. In patients with these characteristics complete response to treatment was observed in > 85% of the cases. It is of novel interest that response to omalizumab is better in patients with a negative HR test, i.e. in patients with nonautoreactive CSU. However, this finding is in accordance with previous studies that have associated longer duration and severity of CSU, with the presence of a positive autologous serum test.12 In a clinical setting, documentation of autoreactivity in CSU should therefore be accommodated in the decision of when to deem a patient a nonresponder. This study was retrospective and examined the effect of omalizumab in a single dermatological centre. Furthermore, no validated instruments for measuring urticaria severity, such as the Urticaria Activity Score or the Urticaria Control Test,15 16 were employed to more objectively estimate the effect of omalizumab treatment. Future prospective studies of chronic urticaria should examine in more detail the clinical and paraclinical predictors of response, and the adverse effects of omalizumab.