WHO recommend to divide AR into 4 subgroups according to the symptom frequency (intermittent or persistent) and severity (mild or moderate-to-severe). For the persistent moderate-to-severe AR subgroup, the guideline suggests to treat with intranasal corticosteroid (INS) plus antihistamines (AH1) for 2-4 weeks. If the symptom is controlled then degrade the treatment (usually with INS) and maintenance for more than 4 weeks. However, up to 70% of patients suffering from AR do not follow treatment recommendation, they stopped medication when they feel better. This behavior always leads to uncontrolled AR, which has been identified as a high-risk factor of induction and exacerbation of asthma and chronic rhinosinusitis. A recent survey showed that AR patients prefer to an on-demand treatment rather than continuous treatment. In general, poor adherence is always a considerable issue for all long-term treatments. Previous studies have shown that as dosing frequency increases, the adherence rate decreases. Thus, less medication frequency is an important factor to optimize the management of chronic diseases including AR.

Intranasal AH1 can relieve AR symptoms including sneezing, rhinorrhea and nasal itching in 3 to 5 minutes, while INS can inhibit the underlying mucous allergic inflammation and is recommended as the first-line medication for moderate-to-severe AR. INS combined AH1 have shown a synergic effect on control AR inflammation and provide rapid AR symptom relief. Investigators hypothesis that the on-demand administration of INS combined AH1 can achieve similar AR control level with less dosing frequency as compared to the daily INS maintenance in controlled moderate-to-severe AR patients.

Background/Significance The prevalence of AR is increasing worldwide. A large number of studies have shown that AR has a significant adverse impact on the quality of life especially in the moderate-to-severe AR population. However, as AR is not a fatal disease, only a small proportion of AR patients will go to hospital to seek professional medical advice. They have a tendency to do so only when their AR-related symptoms become intolerable, which led to a dramatic low level of AR control. It is well known that uncontrolled AR has been identified as a high-risk factor of exacerbation of asthma and chronic rhinosinusitis. In AR patients, 40% will develop asthma and up to 30% will develop chronic sinusitis. Adequate treatment of AR always help to improve patients' quality of life and also reduce the risk of AR related complications.

To achieve the goal of AR control, long-term treatment is always needed. AR is an IgE-mediated inflammatory disease induced by allergen exposure. Our previous study had shown that more than 90% of AR patients in central China were induced by house dust mite (HDM), a perennial indoor allergen. HDM exposure leads to persistent allergic inflammation in nasal mucous and causes periodic exacerbation of AR symptoms. INS is currently the cornerstone medication to inhibit the allergic inflammation and is recommended as the first-line medication for moderate-to-severe AR. However, the anti-inflammation effect can only be achieved after administrate for several days to two weeks. For this reason, the ARIA guideline suggests to use INS regularly for 2-4 weeks, then step down gradually. In another side, antihistamine belongs to symptom-relief medication, which has been proven to act within 5min for nasal spray and 30min for oral route. Antihistamine is also recommended as first-line medication for mild AR patients and combined with INS for moderate-to-severe AR patients. Our previous studies had shown that if AR patients were treated according to the current guideline for 15 days, only 72.3% of them were controlled. However, if the treatment extended to 60 days, more than 95% would achieve AR controlled.

Low adherence is always a considerable issue for long-term treatment. In asthma patients, more than 80% of patients taken β2 agonist as needed to relieve their symptom instead of daily using inhalant corticosteroids. For AR, the adherence to treatment recommendation was below 30% in the real world. Most AR patients wouldn't take their medication continuously for 14 days as they usually suffered from several episodes of short-term AR symptoms. They have a strong preference for a treatment which can provide both a rapid onset of action and clinically-relevant symptom relief.

In this study, investigators will compare the AR control level of on-demand treatment versus daily maintenance treatment in moderate-to-severe AR patients. If the on-demand treatment can achieve similar or non-inferior AR control rate as daily maintenance treatment, it will be very helpful to optimize current management of AR patients.

Preliminary Data A recent study in 150 pollen-induced AR children found that the percentage of symptom-free-day was in favor of INS on-demand (30%) compared with low dosage INS daily (22%), although the statistical significance was not achieved. It implied that on-demand fluticasone propionate treatment was at least as effective as a regular treatment at a lower dose. The authors concluded that an on-demand INS strategy has the advantage of a lower overall corticosteroid exposure and less costs.

Two large randomized controlled studies in mild asthma patients (4215 and 3849 patients were randomized respectively) also shown that budesonide-formoterol used as needed was non-inferior to twice-daily budesonide with respect to the rate of severe asthma exacerbations during 52 weeks of treatment.

Investigators have validated the Allergic Rhinitis Control Test (ARCT) questionnaire in 255 AR Chinese patients, the ARCT score ≥20 was regard as controlled AR. Investigators also validated ARCT questionnaire in guiding stepwise treatment in 510 AR patients. Thus, the ARCT questionnaire will be used to calculate the AR control level in our study. In our randomized controlled trial, 129 AR patients received INS without dosage adjustment for 60 days, 85.8% were controlled AR (ARCT score≥20). In a hospital based study, investigators found that only 4.8% of the AR patients were controlled without taken medication or taken medication occasionally. However, the data could not reflect the real-life situation as usually only the intolerable symptoms would drive AR patients to the hospital. There is still lack of data of AR control rate in on-demand treatment patients.

Research Design and Methodology Study Design This is an open-labeled, randomized, controlled clinical study with HDM-related AR which will be conducted in four multicenter in Central China.. The symptom uncontrolled moderate-severe AR patient were enrolled with ARCT˂20. In the run-in period, AR patients will be treated with INS plus AH1 twice daily for 2 weeks and assessed with ARCT questionnaire and Quality-of-Life Questionnaires. If AR is controlled, the patients will be randomized into on-demand INS combine intranasal AH1 group, on-demand INS or daily INS maintenance group and treat for 4 weeks, then the treatment will be stopped and the patients will be followed up for another 8 weeks. During the study period, the patients have to fill in the diary card every day to record TNSS and daily ARCT every day. During the run-in period, the treatment and follow-up period, the patients were followed up every 4 weeks in the hospital to finish ARCT, RQLQ and IPQ questionnaires to compare the AR control and recurrence time among the three groups.

Study Population This is a multi-center study. Patients with persistent moderate-to-severe AR will be enrolled after sign the informed and content form, adolescents need guardian consent.

Inclusion Criteria (1) 18-65 years old（2) AR patients (according to ARIA guidelines)（3) Mono-sensitized to HDM and had AR symptoms after HDM exposure; （4) had at least 2 nasal symptoms(sneezing, runny nose, itching, nasal congestion) in the screening period（4) ARCT score ≥20 after 2-week INS combined with AH1 treatment at run-in period（5) had adequent informed and given their consent to participate in the study.

Exclusion Criteria

(1) Pregnant or lactating women, patients with malignant tumor, patients with congenital or acquired immunodeficiency disease, patients with mental illness（2) Acute upper respiratory infection in the run-in period（3) History of chronic sinusitis with nasal polyps（4) Severe deviation of nasal septum（5) received allergen immunotherapy in the past 5 years（6) received biological therapy in the past 6 months （7)Acute upper respiratory tract infection in run-in period（8) Patients who failed to achieve AR control in the run-in period.(9) Patients who are participating in other clinical trials; (10) patients who are not suitable for this clinical trial due to other reasons (evaluated by investigators).

Study protocol

1. Enroll HDM mono-sensitized AR patients.
2. Diagnosis and differential diagnosis were made according to ARIA. The other related diseases that may increase nasal symptoms were excluded according to ARIA.
3. Evaluate ARCT score, RQLQ, and Brief IPQ questionnaire in the baseline. Treat with intranasal Azelastine and Fluticasone Propionate and re-evaluate ARCT score to assure ARCT≥20 after 2 weeks, before randomization. During the run-in period the patients need to complete the daily TNSS and the daily ARCT on-line.
4. Randomize into On-demand intranasal Azelastine and Fluticasone Propionate group, on-demand Fluticasone Propionate group or daily intranasal Fluticasone maintenance group. The patients have to complete the dairy online to record the TNSS and ARCT.
5. The pharmacotherapy will be stopped after 4 weeks and the patients will be followed up for another 8 weeks. During the follow-up period, the patients filled the diary card every day to recorded TNSS and ARCT. During the treatment and follow-up period, the patients have to revisit every 4 weeks in the hospital. ARCT, RQLQ and IPQ questionnaires were filled in to compare the AR control and recurrence time among the three groups.

Statistical Approach and sample size calculation Referring to the relevant literature, investigators assume the AR control rate of continuous NCS treatment group will be 85% and the AR control rates of on-demand NCS group and on-demand NCS + AH1 group both will be 75%, investigators also set the non-inferiority margin value as 10%, and the enrollment ratio as 1:1:1. The sample size of each group is 49, plus 20% drop-out rate, a total of 180 cases are required in this trial.

Possible risks and preventive measures Fluticasone propionate nasal spray and azelastine hydrochloride nasal spray used in this study have been approved and widely used in AR patients for many years in China. The common local adverse reactions include epistaxis, nasal cavity and throat dryness, local nasal pain and irritation, bitter taste or nausea, etc. Systemic adverse reactions are rare. Any adverse events will be recorded and managed by the investigators according to Good Clinical Practice guidelines.

Data collection and statistical analysis Data collection: the patients filled in the diary on-line every day to record the nasal symptom (TNSS) score and ARCT. The symptoms included nasal itching, sneezing, runny nose and nasal congestion. The scoring criteria were: 0: asymptomatic, 1: mild, 2: moderate and 3: severe. The ARCT score was used to assess the control of AR symptoms. During the treatment and follow-up period, the patients were followed up every 4 weeks on-site. ARCT, RQLQ and IPQ questionnaires were filled in to compare the AR control and recurrence among the three groups.

Statistical analysis: the primary endpoint was AR control level at 4-week among the treatment groups. The secondary endpoint included TNSS scores of different treatment groups, AR recurrence time after treatment, RQLQ and IPQ questionnaire score comparison and adverse events among different treatment groups. The data of continuous variables were tested for normality, and the variables in accordance with normal distribution were expressed by mean ± standard deviation. T test was used to compare the differences between groups; The variables of non-normal distribution were expressed by median ± quartile interval. Mann Whitney U test was used to compare the differences between groups; For binary variables, Chi square test was used to determine the difference between groups. Kaplan Meier survival curve was used to analyze the recurrence time of AR in different treatment groups. A p value less than 0.05 was considered statistically significant.

Inclusion Criteria:

• 18-65 years old
• AR patients (according to ARIA guidelines)
• Mono-sensitized to HDM and had AR symptoms after HDM exposure;
• Participants had at least 2 nasal symptoms(sneezing, runny nose, itching, nasal congestion) in the screening period
• ARCT score ≥20 after 2-week INS combined with AH1 treatment at run-in period
• had adequent informed and given their consent to participate in the study.

Exclusion Criteria:

• Pregnant or lactating women, patients with malignant tumor, patients with congenital or acquired immunodeficiency disease, patients with mental illness
• Acute upper respiratory infection in the run-in period
• History of chronic sinusitis with nasal polyps
• Severe deviation of nasal septum
• received allergen immunotherapy in the past 5 years
• received biological therapy in the past 6 months
• Acute upper respiratory tract infection in run-in period
• Patients who failed to achieve AR control in the run-in period
• Patients who are participating in other clinical trials
• patients who are not suitable for this clinical trial due to other reasons (evaluated by investigators).

We plan to share the periodic study reports with peers, share the data, problems we encounter in the study and problem solutions, and modification of the scheme if it is necessary.

The Clinical Study Report (CSR) will be available in six months and for 1 year.

The peer investigators who contact with the Central Contact Person by email or phone call.