Skip to content

Advertisement

  • Poster presentation
  • Open Access

A new therapy (MP29-02*) effectively treats patients with seasonal allergic rhinitis who suffer most from the bothersome nasal symptom of congestion

  • 1,
  • 2,
  • 3,
  • 4,
  • 5,
  • 6 and
  • 7
Clinical and Translational Allergy20133 (Suppl 2) :P39

https://doi.org/10.1186/2045-7022-3-S2-P39

  • Published:

Keywords

  • Allergic Rhinitis
  • Allergic Rhinitis
  • Fluticasone
  • Fluticasone Propionate
  • Nasal Congestion

Background

In clinical practice, allergic rhinitis (AR) patients frequently present with a predominant or particularly bothersome symptom, most frequently nasal congestion.

Objective

To assess the efficacy of MP29-02* (a novel intranasal formulation of azelastine hydrochloride [AZE] and fluticasone propionate [FP]) in patients with seasonal AR (SAR) suffering predominantly from nasal congestion, compared to commercially available AZE or FP nasal sprays and placebo.

Methods

610 patients (≥12 years old) with moderate-to-severe SAR were randomized into a double-blind, placebo-controlled, 14-day, parallel-group trial to MP29-02*, AZE or FP nasal sprays and placebo (all given as 1 spray/nostril bid; total daily dose: 548µg AZE, 200µg FP]. Patients were defined as 'nasal congestion predominants' if their maximum symptom score at baseline was the nasal congestion score (n=368). Both reflective total nasal symptom score (rTNSS; max score =24) and nasal congestion symptom score (max score =6) reduction were assessed in these patients to show effect on their overall nasal symptom burden, as well as specific relief from nasal congestion.

Results

MP29-02* induced the greatest reduction in rTNSS in patients complaining of nasal congestion (-5.64), compared to -3.93 for FP (Diff -1.71; 95% CI -3.00, -0.43; p=0.0093), -3.28 for AZE (Diff -2.36; 95% CI -3.51, -1.21; p<0.0001) and -2.63 for placebo (Diff -3.01; 95% CI -4.14, -1.88; p<0.0001), corresponding to a relative treatment difference of 57% to FP and 79% to AZE. These nasal congestion-predominant patients treated with MP29-02* also experienced a significantly greater reduction in their nasal congestion score; -1.41 vs -0.90 for FP (Diff: -0.51; 95% CI -0.83, -0.19; p=0.0018), -0.83 for AZE (Diff: -0.58; 95% CI -0.88, -0.29; p=0.0001) and -0.69 for placebo (Diff -0.72; 95% CI -1.02, -0.42; p<0.0001), with a relative treatment difference of 71% to FP and 81% to AZE. Neither AZE nor FP significantly differed from placebo in terms of nasal congestion reduction in these patients.

Conclusion

Unlike currently available first line therapy, MP29-02* effectively reduced nasal congestion and the overall nasal symptom burden of patients suffering predominantly from nasal congestion. This indicates that for nasal congestion predominant patients a decongestant might not be required prior to MP29-02* administration, and further supports the position of MP29-02*as the drug of choice for the treatment of AR.

*Dymista

Authors’ Affiliations

(1)
Dept of Oto-rhinolaryngology, Ghent University Hospital, Ghent, Belgium
(2)
Dept of General Practice & Primary Care, University of Aberdeen, Abderdeen, UK
(3)
The Royal National Throat, Nose and Ear Hospital, London, UK
(4)
Academic Medical Center, Dept of Otorhinolaryngology, Amsterdam, the Netherlands
(5)
Dept of Otorhinolaryngology, Head and Neck Surgery, University Hospitals Leuven, Leuven, Belgium
(6)
Meda Pharma, Biostatistics & Market Access, Bad Homburg, Germany
(7)
Hopital Arnaud de Villeneuve University Hospital, Montpellier, France

Copyright

Advertisement