Impact of doxofylline compared to theophylline in asthma: A pooled analysis of functional and clinical outcomes from two multicentre, double-blind, randomised studies (DOROTHEO 1 and DOROTHEO 2)

Please cite this article as: Calzetta L, Hanania NA, Dini FL, Goldstein MF, Fairweather WR, Howard WW, Cazzola M, Impact of doxofylline compared to theophylline in asthma: A pooled analysis of functional and clinical outcomes from two multicentre, double-blind, randomised studies (DOROTHEO 1 and DOROTHEO 2), Pulmonary Pharmacology & Therapeutics (2018), doi: https://doi.org/10.1016/ j.pupt.2018.09.007.


Introduction
Xanthines are structurally related drugs, used in clinical management of patients with chronic obstructive respiratory disorders, including asthma and chronic obstructive pulmonary disease (COPD). However, commonly used xanthines (theophylline, aminophylline) have a major drawback in that they have a very narrow therapeutic window and propensity for many pharmacological interactions [1][2][3][4][5][6]. The advent of other classes of drugs, such as the new inhaled corticosteroids (ICSs) and long-acting bronchodilator agents, has limited the use of xanthines, despite their clear clinical benefit in the treatment of some patients with chronic obstructive respiratory disorders [3].
Doxofylline is a newer generation xanthine with bronchodilating and antiinflammatory actions [7][8][9][10][11]. In experimental animals, this drug has also been shown to have anti-inflammatory activity in a rat pleurisy model and to inhibit human eosinophil activation by affecting Ca ++ activated K + channels [12,13].
Doxofylline is also able to provide prophylactic effects against bronchoconstriction induced by platelet-activating factor and methacholine in guinea-pigs and dogs [13][14][15].
In human, two multicenter, double-blind, randomized trials, carried out in 21 Italian pulmonary clinics to investigate the therapeutic efficacy and tolerability of doxofylline compared to slow-release theophylline or aminophylline, demonstrated that doxofylline is an effective and well tolerated agent in patients suffering from chronic reversible airways obstruction [16,17]. A number of other studies confirmed the beneficial clinical effects of doxofylline in asthma and COPD [1,5,18].
In summary, bronchodilatory effects of doxofylline have been demonstrated in patients suffering from asthma or COPD [1,2,15,[19][20][21], but, to the best of our knowledge, large scale data comparing the efficacy and safety of doxofylline to those of the more commonly used theophylline in the treatment of asthma are still lacking. To explore this, we examined data from two independent, double-blind, randomized, placebo-controlled trials that aimed to investigate the Impact of DOxofylline compaRed tO THEOphylline (DOROTHEO 1 and DOROTHEO 2 studies) [22,23] in asthma. Data were pooled in order to perform a pre-specified analysis of the two studies, which together are powered to provide more reliable estimates of the effect of the investigated drugs on lung function, asthma events rate, use of salbutamol to relieve asthma symptoms, and adverse events (AEs).

Study design
DOROTHEO 1 and DOROTHEO 2 were Phase III, multicentre, double-blind, randomized, parallel-group, placebo-controlled, clinical trials, conducted in 37 centres in the United States [22,23]. Each study had one week run-in period during which the subject took placebo followed by 12-week treatment period and a 1-week single-blind placebo run-out phase at the end of the study.
Patients were permitted to use inhaled salbutamol as rescue medication.
Treatments with other oral xanthines, inhaled β 2 -agonists or antimuscarinic agents and inhaled corticosteroids were withheld between 72 hours and one week before the beginning of the study and during the study period.

Study endpoints
The primary endpoint of this pooled analysis was the change from baseline in The patients used diary cards to record the date and time that each dose of study medication was taken, date and time of asthma episodes, date and time of each salbutamol use, and date and time of any AEs.

Assessment of safety and drop-outs
All clinical AEs were recorded and graded as mild, moderate or severe. Their relationships with active treatments were classified as follows: 1) not related, 2) possibly related, 3) definitely related or 4) unknown. Also recorded for each event were the duration of the symptoms and the action taken (none, reduction of the dose or discontinuation of treatment). Subjects were removed from therapy or assessment for: 1) non-adherence to treatment, 2) persistent drug-related AEs with patient's willingness to discontinue treatment, 3) serum theophylline levels exceeding 20 µg/mL or 4) elevated doxofylline concentrations (>2 standard deviations above the mean) in the presence of any drug-related AE.

Statistical analysis
Data are expressed as mean ± standard error of the mean (SEM). The statistical analysis compared FEV 1 measured by spirometry obtained at baseline (immediately prior to the start of double-blind treatment) with the results obtained 2 hours after administration of study medication (2-h postdose FEV 1 ) at each visit during double-blind treatment. The derived variable was the percent change between these two assessments. Absolute changes were calculated for the asthma events rate (total number of events divided by total number of days on study medication) and salbutamol use rate (total number of puffs divided by total number of days on study medication).
Baseline for the latter two variables was defined as the value obtained from the diaries during the placebo run-in phase. The safety analysis was performed by calculating the risk of AEs compared to placebo, and data were reported as risk ratio (RR) and 95% confidence interval (95%CI).
The DOROTHEO 1 and DOROTHEO 2 studies [22,23] had similar protocol, with the same treatment groups, except that DOROTHEO 1 included also a low dose doxofylline group (200 mg tid).
The analysis of the changes from baseline was performed by using t-test, whereas the two-way analysis of variance (ANOVA) was used for evaluating the difference across the treatments during the study period. All differences were considered significant for P<0.05. Data analysis was performed by using Prism 5 software (GraphPad Software Inc, CA, USA) and OpenEpi software [24].

Patient population
Six hundred sixty six patients were screened for participation in the pooled studies [22,23] and entered the placebo wash-out. One hundred patients did investigated only in DOROTHEO 1 study [22]. Patient enrolment and the reasons for discontinuation are presented by treatment group in Figure 1. found elsewhere [22].
Patient demographics and baseline characteristics were similar with no significant differences (P>0.05) across the treatment groups, as reported in Table 1. The summary of prior asthma medications is reported in is reported in supplementary data file (Table S2).

Asthma events
Doxofylline

Safety profile
The safety profile of doxofylline 400 mg (the overall risk of AEs) was not significantly higher that what was encountered with placebo (RR 1.16 CI95%  Three serious AEs (SAEs) were recorded in the doxofylline group (1 asthma exacerbation, 1 systemic reaction to immunotherapy, 1 cold leading to asthma M A N U S C R I P T

A C C E P T E D ACCEPTED MANUSCRIPT
13 exacerbation), 1 SAE was detected in the theophylline group (severe asthma exacerbation), and 1 SAE was reported in the placebo group (degenerative disk disease). Among these SAEs, only the severe asthma exacerbation in the theophylline group was classified by the investigator to be possibly related to study medication, the other SAEs were classified to be not drug related.
No subjects died during the studies or within 30 days after finishing the studies.

Discussion
This pooled analysis of DOROTHEO 1 and DOROTHEO 2 studies [22,23] showed that over 12 weeks, treatment with doxofylline 400 mg and theophylline 250 mg both significantly increased 2-h postdose FEV 1 compared to baseline in asthmatic patients, and that such improvements exceeded the MCID from week 2 onwards. The overall improvement in the change from baseline in 2-h postdose FEV 1 was similar between doxofylline 400 mg and theophylline 250 mg, and significantly greater than that elicited by placebo. Similarly, both doxofylline 400 mg and theophylline 250 mg were effective in reducing the rate of asthma events and use of salbutamol to relieve asthma symptoms compared to baseline. In this regard, both medications were significantly more effective than placebo during the study period, and no difference in efficacy was detected between the active treatments. Interestingly, the trend of improvement elicited by doxofylline 400 mg and theophylline 250 mg with respect to the rate of asthma events was similar to that of salbutamol use. This evidence suggests that metrics of salbutamol used as rescue medication may predict the rate of asthma events, and further supports previous findings indicating that higher daily salbutamol use was associated with future severe exacerbations, poor asthma control, and increased risk of future extreme salbutamol overuse [27].
Indeed, we cannot omit that also placebo had a certain effect on the outcome assessed in this pooled analysis, although placebo was always significantly less effective than both doxofylline 400 mg and theophylline 250 mg during M A N U S C R I P T

A C C E P T E D ACCEPTED MANUSCRIPT
14 the study period and did not reach the MCID for the change from baseline in FEV 1 . Abnormally high responses to placebo are not infrequently reported with no apparent explanation in randomized controlled trials [28]. However, we can explain such an unexpected level of placebo effect by considering that better adherence to drug regimens in the context of a clinical trial may have contributed to this observation, as recently reported in severe asthmatic patients with respect to the percentage reduction in oral glucocorticoid dose and change from baseline in FEV 1 [29,30].
Doxofylline 400 mg showed a favourable safety profile that was significantly superior to that of theophylline 250 mg. In fact, AEs leading to drop-outs occurred more frequently with theophylline 250 mg than doxofylline 400 mg.
Furthermore, while the only significant AE in patients receiving doxofylline 400 mg was insomnia, in subject treated with theophylline a significant higher risk of nausea, nervousness, insomnia, and overdose was detected compared to placebo.
The findings of this pooled analysis confirm the recent evidence rose from meta-analyses aimed to assess the functional and clinical impact of xanthines in COPD [1,2]. Specifically, doxofylline produced a large to very large improvement in lung function and reduced dyspnoea similarly to theophylline, and the use of doxofylline was associated with a significantly better safety profile than theophylline [1,2]. Overall, in COPD patients doxofylline showed a more favourable efficacy/safety profile than theophylline [1]. Results of this pooled analysis are also in line with a previous study reporting the efficacy and safety profile of doxofylline compared to theophylline in asthmatic patients [19].
Although this pooled analysis and previous studies demonstrated that the treatment with xanthines might significantly improve lung function and symptoms in asthmatic patients and subjects with COPD, their therapeutic use has declined substantially in recent years [1,2,19]. The development of ICSs has significantly altered the therapeutic approach to asthma in most part of the world [31]. Nonetheless, nowadays, new evidence appears to support a reappraisal in the use of these drugs [32].
Theophylline is usually less effective than ICS, but, when used in combination with low-to-medium doses of an ICS, the same effects as those obtained with an increased inhaled corticosteroid dose can be achieved in asthmatic patients [33]. Therefore, theophylline can be recommended when asthma control cannot be achieved by using ICSs. With the introduction of doxofylline a new window of opportunity has opened up as a result of a similar efficacy and better tolerability than theophylline to treat patients suffering from chronic obstructive respiratory disorders, such as asthma and COPD [1,2,7,[34][35][36][37][38][39][40].
Doxofylline demonstrated significant anti-inflammatory activity in the lung which can result in significant steroid sparing activity [41]. Rajanandh and colleagues [42] have recently demonstrated that doxofylline was as effective as montelukast and tiotropium, all in association with low-dose inhaled steroids, in relieving airways obstruction asthmatic patients. The use of xanthines, and particularly of doxofylline, could be particularly valuable in elderly patients with asthma or smoker subjects, where other drugs, especially ICSs, are less likely to work due to the inhibition of histone deacetylase 2 (HDAC2) activity [43]. Finally, Mennini and colleagues [44] have provided evidence of the cost-effectiveness of doxofylline. Although, theophylline has an average base price lower than doxofylline, physicians can be recommended to prescribe it instead of theophylline not only for its favourable efficacy/safety profile, but also because it has been demonstrated that doxofylline can reduce the costs associated with the management of respiratory disorders [44]. Some limitations should be acknowledged with respect to this pooled analysis. A remarkably high dropout rate of approximately 40% in the theophylline group can be considered a major drawback [19,45], although this highlights the problems related with the tolerability of theophylline.
Nevertheless, the average adherence of clinical trials reported in the literature is between 43% to 78% [46], and the overall adherence of this pooled analysis was about 70%. As far as the question whether the better tolerability profile of doxofylline was real or "wishful thinking", it is important to note that the occurrence of AEs typical of xanthine medications, such as dyspepsia, M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 16 nausea, vomiting, dizziness, and insomnia [1,2], were more frequent in patients who were given theophylline than in those who received doxofylline.

Conclusion
In line with the assumption that xanthines still play an important role in treatment of asthma, from the results of this pooled analysis we can conclude that doxofylline is a xanthine bronchodilator with valuable characteristics by functional and clinical viewpoint. Since doxofylline was associated with improved bronchodilatory response, reduced rate of asthma events, reduced use of salbutamol as rescue medication and fewer AEs, it seems to offer a promising alternative to theophylline in the management of patients with asthma.

M A N U S C R I P T
A C C E P T E D ACCEPTED MANUSCRIPT 17

Funding
This study was funded by ABC Pharmaceuticals, Turin, Italy and by Roberts, Inc, Eatontown, NJ, USA. The sponsors had no role in study design, data collection, data analysis, data interpretation, or writing of the report.