Efficacy and safety of intralymphatic immunotherapy in allergic rhinitis: A systematic review and meta‐analysis

Abstract Background Intralymphatic immunotherapy (ILIT) is a potential treatment option for allergic rhinitis (AR). We aimed to determine the efficacy (primary outcomes) and safety (secondary outcomes) of ILIT in treating patients with AR. Methods An electronic literature search was performed using MEDLINE and Cochrane Central Register of Controlled Trials CENTRAL (from their inception to December 2020). A random‐effects model was used to estimate the pooled prevalence with 95% confidence intervals. This study is registered with PROSPERO (CRD42019126271). Results We retrieved a total of 285 articles, of which 11 satisfied our inclusion criteria. There were 452 participants with age ranged from 15 to 58 years old. Intralymphatic immunotherapy was given in three doses with intervals of four weeks between doses in 10 trials. One trial gave three and six doses with an interval of two weeks. Both primary and secondary outcomes showed no difference between ILIT and placebo for all trials. There was no difference in the combined symptoms and medication score (SMD ‐0.51, 95% CI −1.31 to 0.28), symptoms score (SMD −0.27, 95% CI −0.91 to 0.38), medication score (SMD −6.56, 95% CI −21.48 to 8.37), rescue medication (RR 12.32, 95% CI 0.72–211.79) and the overall improvement score (MD −0.07, 95% CI −2.28 to 2.14) between ILIT and placebo. No major adverse events noted. Conclusions Intralymphatic immunotherapy possibly has a role in the treatment of AR patients. This review found it is safe but not effective, which could be contributed by the high variation amongst the trials. Future trials should involve larger numbers of participants and report standardized administration of ILIT and outcome measures.


| BACKGROUND
The prevalence of self-reported allergic rhinitis (AR) has been estimated to be approximately 25% in children and more than 40% in adults. 1 Allergic rhinitis is frequently associated with asthma which is found in 15%-38% of AR patients while nasal symptoms present in 6%-85% of patients with asthma. 2 The management of AR involves patient education, allergen (and pollutant) avoidance, pharmacotherapy, and allergen-specific immunotherapy (AIT) when appropriate. 3 Pharmacologic options for the treatment of AR include intranasal corticosteroids, oral and topical antihistamines, decongestants, intranasal cromolyn, intranasal anticholinergics and leukotriene receptor antagonist, amongst others. 4,5 A recent advance has been the combination of an antihistamine and a corticosteroid in the same nasal spray. 3 Allergen-specific immunotherapy should be considered for patients with moderate or severe persistent AR that is not responsive to pharmacological treatments. 6 The most widely employed form of AIT involves the administration, subcutaneously or sublingually, of increasing doses of the causative allergen to induce clinical and immunologic tolerance. However, the conventional subcutaneous immunotherapy (SCIT) requires 30 to 80 allergen injections over three to five years and maybe associated with allergic side effects. 3,6,7 Sublingual immunotherapy (SLIT) is more patient-friendly, but treatment duration could not be shortened. 8 Intralymphatic immunotherapy (ILIT) is a potential treatment option to overcome these limitations.

Intralymphatic immunotherapy is a form of AIT where the allergen is directly delivered to B-and T-cells within the lymph nodes.
This procedure is reported to induce a stronger cytotoxic T-cell response and higher immunogenicity than other routes, even with the delivery of smaller amounts of allergen as typically employed in ILIT. 9 Allergenic extract is usually injected into inguinal lymph nodes using ultrasound guidance. Therapy is complete after only three injections. When given four weeks apart, ILIT is reported to be safe and effective for up to three years. [10][11][12][13] In patients with allergic rhinitis that require immunotherapy treatment, ILIT might be an option to treat their symptoms. This meta-analysis aimed to evaluate the efficacy and safety of ILIT in the treatment of AR.

| METHODS
A systematic review and meta-analysis were undertaken in accordance with the preferred reporting items for systematic reviews and meta-analyses statement (PRISMA) guideline. 14 This study is registered with PROSPERO, number CRD42019126271. The search was conducted for randomized clinical trials (RCTs) and case-control studies comparing ILIT with placebo or conventional AIT in patients diagnosed with AR (with or without allergic conjunctivitis or urticaria or asthma). The search was not limited by age, gender, race, or ethnicity. Allergic rhinitis required clinician diagnosis. Types of interventions included AIT with single or multiple allergens involving recombinant, synthetic or natural allergens. The principle of the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) approach for evaluating the quality of evidence was used. 15 The search was conducted using the following MeSH terms "allergic rhinitis" OR "rhinoconjunctivitis" AND "intralymphatic immunotherapy" OR "intralymphatic allergen" AND "sublingual immunotherapy" OR "subcutaneous immunotherapy" in MEDLINE Two review authors (NRA, BA) screened the titles and abstracts independently from the searches and obtained full-text articles when they appeared to meet the eligibility criteria or when there was insufficient information to assess the eligibility. The eligibility of the trials was assessed independently and the reasons for exclusion were documented. Any disagreements between the review authors were resolved by discussion. The study authors were contacted when any clarification was required.
Using data extraction form, the study setting, participant characteristics (age, sex, ethnic), methodology (number of participants randomized and analyzed, duration of follow-up), type of allergen used, method of diagnosing AR, perennial or seasonal group, type of comparison group which was either a placebo or conventional AIT, assessment of duration, technique and number of injection, the occurrence of related adverse events such as local swelling or systemic symptoms such as fatigue and at last the requirement of rescue medication were extracted. The effects of ILIT were assessed at two weeks following completion of treatment with three doses of ILIT.
The primary outcomes were combined symptoms medication score, symptoms score, medication score, rescue medication and overall Any disagreements were resolved by discussion.
The meta-analyses were done using Review Manager 5.3.5 software. 17 The statistical analyses were performed using the random-effects model and the results expressed as risk ratio (RR) for dichotomous outcomes and mean difference (MD) for continuous outcomes with 95% confidence intervals (CI). The included trials were checked for unit of analysis errors. If any cluster-RCTs were encountered, the results from trials were adjusted to show a unit of analysis errors based on mean cluster size and intracluster correlation coefficient. 16 If studies have non-extractable or missing data, analyses of the available data were performed. The presence of heterogeneity was assessed in two steps. First, obvious heterogeneity at face value was assessed by comparing populations, settings, interventions and outcomes. Second, statistical heterogeneity was assessed through the I 2 statistic. 16 The heterogeneity was interpreted as outlined: 0% to 40% might not be important; 30% to 60% may represent moderate heterogeneity; 50% to 90% may represent substantial heterogeneity, and 75% to 100% would be considerable heterogeneity. 16

| RESULTS
A total of 285 records were retrieved and screened from the search of electronic databases ( Figure 1). Fifteen full-text articles were assessed for eligibility and four records were excluded. Three trials [19][20][21] excluded from the review due to no control group while one trial 22 did not meet the inclusion criteria. 11 articles 10-13,18,23-28 were included in the review. The characteristics of the included trials are shown in Table 2.

| Outcomes
Three trials reported the symptoms score. 13,25,27 Symptoms score was scored on a 0-3 points scale corresponding to no symptoms, mild, moderate or severe symptoms for two trials 13,27 while one trial 25 scored on 0-4 points scale based on the frequency of symptoms: daily (4 points); every second day (3 points); 1-3 days per week (2 points); occasionally (1 point); never (0 points). Medication score was reported in three trials. 13,25,27 The medication score was calculated based on the use of daily medications (consisting of an oral antihistamine, antihistamine eye drop, nasal corticosteroid and oral corticosteroid). The maximum daily points were 20 in one trial 27 and 24 in another trial. 13 Five trials 13,18,[26][27][28] reported the combined symptoms medication score. The combined symptoms medication score is the sum of the daily symptoms score and medication score. Eight trials 10,11,13,[23][24][25]28 reported the overall improvement score. The overall improvement score was reported in a visual analog scale ranging from zero (unchanged symptoms or no improvement) to 10 (total symptoms relief or complete recovery) when asked to compare allergic symptoms during recent pollen season versus the previous experienced season before the treatment started. Three trials 10,23,24 reported the rescue medication in AINI ET AL. The adverse events include local swelling at site of injection, 10

| Risk of bias in included studies
The assessment of the risk of bias for individual trials is presented in Figure 2. Seven trials 10,12,23-27 described the method of randomization used. Four trials 10,23-25 randomized the participants using simple randomization with a ratio of one placebo to one patient who received the intervention. Three trials 12,25,27 randomized the participants according to a pre-printed allocation list generated by a private analytical company (IKFE GmbH, Germany), 12 randomizer.org 26 and REDCap Cloud EDC system 27 using block randomization. The randomization method was not reported in five trials 10,11,18,23,28 ; thus, we considered random sequence generation as an unclear risk bias. Allocation concealment was not clear in four trials. 11,18,26,28

| Blinding
Placebo was used as the control group in 10 trials, 10,12,13,18,[23][24][25][26][27][28] while SCIT was used as the control group by Senti et al. 11 and in a pilot study by Hylander et al. 10 Both patients and investigators were blinded in eight trials. 10,12,13,[23][24][25][26][27] The control group in one trial 11 received subcutaneous injection as immunotherapy; thus, both patient and investigator were not blinded throughout the study. Two trials 18,28 did not describe blinding. One trial 24 had detection bias due to an error during the outcome assessment.
For ILIT versus SCIT group, one trial 10 showed no difference in symptoms improvement during the next pollen season after the treatment (MD −0.60, 95% CI −1.74 to 0.54; p = 0.300; 1 trial; 13 participants). One trial 11 comparing ILIT and SCIT, reported the reduction of rescue medication was higher in SCIT group compared to ILIT (RR 0.64, 95% CI 0.45 to 0.93; p = 0.020; 1 trial, 107 participants).

F I G U R E 3 Primary outcomes of intralymphatic immunotherapy versus placebo
AINI ET AL.  participants). Three trials 13,24,25 reported no difference in the quality of life between ILIT and placebo (SMD −0.10, 95% CI −0.86 to 0.67; p = 0.800; I 2 = 69%; 3 trials, 88 participants; low certainty evidence) ( Figure 5, Table 3). The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). b Small number of participants (<400). c substantial heterogeneity.

| DISCUSSION
This review found the efficacy and safety of ILIT are not different from placebo in AR management. Most results showed no difference between ILIT and control for the primary and secondary outcomes.
Despite most trials reported swelling at the injection site as being the most common adverse event in their patients, the present review found no significant difference in the overall adverse events between F I G U R E 4 Secondary outcomes (adverse events) of intralymphatic immunotherapy versus placebo AINI ET AL. ILIT and placebo. Whilst most trials did not report severe reaction, a trial by Lee et al. 20 reported two patients experienced anaphylaxis following ILIT. Remarkably, the events occurred after the first injection in one patient and second injection in another patient. Both events were attributed to the use of non-standardized allergen extract as compared to standard allergen extract used in other trials.
Albeit the limited availability of trials comparing ILIT with other AIT, the results of two trials 10,11 suggest that ILIT has the advantage of less cutaneous reaction and dyspnea to SCIT in addition to the benefit of faster reduction of serum specific IgE level.
We noted the inconsistency of the administration of ILIT in a 2-week interval in one trial 13 compared with a 4-week interval for others which can be considered as an outlier. The general recommendation for ILIT is three doses given in a 4-week interval for the development of successive waves of antigen-specific immune responses 25 but different doses and interval between injection were used among the trials. Although, the impact of this incongruity on the outcome is arguable, standardization of the injection interval is required to ensure comparable outcomes. If further trials are to be conducted to analyze ILIT, using a standardized interval and number of doses to administer ILIT coupled with a suitable outcome measurement such as combined symptoms and medication score would represent a better reflection of the actual potential of ILIT. In trials involving AIT, symptom and medication scores are often assessed independently. The World Allergy Organization taskforce 31 recommended the standardization of primary endpoint in AIT trial by using the combined symptoms and medication score. As the severity of symptoms entails a higher frequency of medication consumption, this link will be better replicated by combining the symptoms and medication score to indicate their equivalent contribution.
Interestingly, the results of our review contradict the findings of a meta-analysis recently published by Hoang et al. 32 which found ILIT confers short term improvement of combined symptoms medication score. The discrepancy can be explained by the different statistical models applied for the meta-analysis, the fixed-effect model in their review and the random-effects model in the present review. We used the random-effects model as it was advocated to account for the heterogeneity across the studies. 16 For the analysis results to be generalized to a population, it should consider the varying resources available to the investigators from study to study which makes it unlikely that the intervention effects across the studies to be identical.

| Limitations
High heterogeneity among the trials could be contributed by different allergens exhibiting different immunogenicity effect. One trial 24 treated AR patients with polysensitization using grass and birch allergen demonstrated an opposite effect on serological IgE levels.
While grass-specific IgE showed a transient increase, birch-specific IgE did not exhibit any alteration. The transient increase of IgE level is attributed to an early phase of desensitization, similarly seen during conventional AIT. Hence, the different immunogenicity effect of different allergen used for ILIT needs to be further investigated.

| CONCLUSIONS
Intralymphatic immunotherapy possibly has a role in the treatment of AR patients. This review found it is safe but not effective, which could be contributed by the high variation amongst the trials. Future trials should involve larger numbers of participants and report standardized administration of ILIT and outcome measures.