BackgroundWe previously reported that repeated low-dose grass pollen intradermal allergen injection suppresses allergen-induced cutaneous late-phase responses, comparable with conventional high-dose subcutaneous and sublingual immunotherapy.ObjectiveTo evaluate the efficacy and mechanism of grass pollen intradermal immunotherapy for treatment of allergic rhinitis.DesignA Phase II, double-blind, randomised controlled parallel-group trial.SettingSingle-centre UK study.ParticipantsAdults aged 18–65 years, with grass pollen-induced allergic rhinoconjunctivitis.InterventionsSeven 2-weekly intradermal injections were given into the forearm, containing eitherPhleum pratensesoluble grass pollen extract (7 ng of the major allergen Phl p 5) or histamine control.Main outcome measuresThe primary outcome was a combined symptom and medication score (CSMS) during the 2013 grass pollen season. Secondary clinical outcomes were overall symptom scores; individual symptoms scores for nose, mouth, eyes and lungs; overall medication scores; CSMSs during the peak season; visual analogue scale (VAS) scores for nose and eye symptoms; Mini Rhinitis Quality of Life Questionnaire scores; health-related quality-of-life scores (European Quality of Life-5 Dimensions, 5-levels); a global evaluation of symptoms, number of symptom-free and medication-free days; number of days when prednisolone was used; and adverse events. Mechanistic studies included measurement of late-phase skin response sizes, allergen-specific antibody titres, analysis of skin biopsies and basophil activation tests.ResultsThere was no significant difference in CSMSs between treatment arms [difference in median area under curve (AUC) 14, 95% confidence interval (CI) –172.5 to 215.1;p = 0.80]. Paradoxically, among the secondary outcomes, nasal symptoms measured with daily scores were higher in the active arm (difference in median AUC 35, 95% CI 4.0 to 67.5;p = 0.03), with a trend for higher nasal symptoms measured by VASs (difference in median AUC 53, 95% CI –11.6 to 125.2;p = 0.05). No differences were seen in other clinical outcomes in the main intention-to-treat analysis. In mechanistic studies, active treatment increasedP. pratense-, Phl p 1- and Phl p 5-specific immunoglobulin E (allp = 0.001) compared with the control. T cells cultured from skin biopsies of active intradermal immunotherapy subjects showed higher T helper type 2 cell (Th2) marker CRTH2 (chemoattractant receptor-homologous molecule expressed on Th2 cells) expression (p < 0.05) and lower T helper type 1 cell marker CXCR3 [chemokine (C-X-C Motif) receptor 3] expression (p < 0.05), respectively. Interleukin 5 messenger ribonucleic acid, measured by microarray, was more highly expressed by cultured skin T cells in the active arm (p < 0.05). Late-phase skin responses to grass pollen were still inhibited up to 7 months after intradermal immunotherapy (p = 0.03), but not at 10–13 months’ time points.LimitationsGrass pollen doses were not increased during the course, as our proof-of-concept trial showed that repeating the same doses was sufficient to achieve almost complete late-response suppression. Injections were not continued throughout the season, as previous subcutaneous grass pollen immunotherapy trials have demonstrated preseasonal regimen efficacy.ConclusionsIntradermal immunotherapy suppressed late-phase skin responses to allergen, but was not clinically effective. The intervention appeared to have an immunological priming effect and exacerbated certain seasonal symptoms, notably in the nose.Future workFurther studies on low-dose intradermal grass pollen immunotherapy are not recommended because of our demonstrated worsening of allergic rhinitis symptoms and immunological priming. The findings are of great significance for other novel immunotherapies targeting the skin, such as epicutaneous techniques.Trial registrationCurrent Controlled Trials ISRCTN78413121.FundingThis project was funded by the Efficacy and Mechanism Evaluation programme, a Medical Research Council and National Institute for Health Research partnership.
Read full abstract