Development and characterization of an Indacaterol/Tiotropium Bromide composite dry powder inhalation formulation for COPD

Abstract

As a major public health issue, chronic obstructive pulmonary disease (COPD) causes approximately 2.7 million deaths worldwide each year and has a significant negative impact on patients’ health. We have developed a novel long-acting fixed-dose combination dry powder inhalation formulation with synergistic therapeutic effects, containing indacaterol maleate (IND, a long-acting β₂ agonist) and tiotropium bromide (TIO, a long-acting muscarinic receptor antagonist), aimed at improving the clinical management of COPD and providing patients with superior bronchodilator effects. We prepared composite powder formulations using different ratios of fine lactose powder and magnesium stearate, combined with active pharmaceutical ingredients of varying particle sizes. We conducted preliminary studies on the total mixture content, mixing uniformity, delivered dose (DD), and fine particle dose (FPD) of each formulation using high-performance liquid chromatography (HPLC). Further screening and characterization were performed using scanning electron microscopy (SEM), laser particle size analyzers, and powder X-ray diffraction (XRD); simultaneously, we conducted aerodynamic studies and accelerated stability testing on the selected optimal formulation. This formulation (F7) demonstrated superior performance compared to other formulations. It exhibited excellent mixing uniformity, appropriate content and optimal aerodynamic properties with FPD values near the median standard range. Accelerated stability testing confirmed excellent stability with no significant changes in crystalline form or aerodynamic properties after 6 months. The developed IND/TIO composite inhalation powder represents an innovative dual-bronchodilator DPI formulation. The dual-stage lactose carrier system significantly enhanced drug detachment efficiency and in vitro pulmonary deposition. The introduction of magnesium stearate effectively addressed challenges of poor flowability and particle aggregation in micronized APIs. The formulation exhibits simple preparation process, excellent stability, and superior in vitro pulmonary drug delivery efficacy, making it a promising therapeutic candidate for COPD treatment.

1. Introduction

Chronic obstructive pulmonary disease (COPD) is a highly prevalent condition with significant morbidity and mortality rates [1], as a respiratory syndrome characterized by progressive, partially reversible airway obstruction and pulmonary hyperinflation [2,3]. It leads to progressive dyspnea, limitations in daily activities, reduced quality of life, and increased frequency and severity of acute exacerbations, while also imposing a significant burden on healthcare providers and society as a whole [2,4]. Therefore, in treating COPD, seeking targeted and highly effective medications is particularly crucial.

Inhaled powder medications are widely used to treat lung diseases such as asthma, COPD, or bacterial infections [[5], [6], [7]]. Unlike oral or injectable formulations, it delivers therapeutic drugs in minute doses directly to the airways, enabling localized effects in the lungs while minimizing unnecessary systemic reactions [6]. The dry powder inhaler (DPI) essentially consists of an air inlet, a capsule/blister/dosing chamber, a cyclonic chamber or dispersion-enhancing component, an air outlet, and a mouthpiece [8,9]. Patients can control the inhalation flow rate to disaggregate the powdered drug formulation into particles most likely to deposit in the lungs, achieving a delivery rate of up to 40% of the administered dose [10,11]. DPI formulations contain micronized drug particles with a aerodynamic diameter ranging from 1 to 5 μm. These formulations may also incorporate additional excipients and/or carrier particles to enhance powder flowability and aerosol dispersion, while improving drug stability. They represent an ideal choice for delivering sensitive molecules such as biologics [8,12].

The primary medications used to treat COPD are bronchodilators, including beta-2 agonists, anticholinergic drugs, theophylline, and corticosteroids [13]. Indacaterol maleate (IND) is a long-acting beta-2 agonist (LABA) that takes effect rapidly within minutes of administration and provides 24-h bronchodilation for patients. However, monotherapy carries the risk of exacerbating the condition [14,15]. As a long-acting muscarinic antagonist (LAMA), tiotropium bromide (TIO) can improve lung function and quality of life in COPD patients while reducing the frequency of acute exacerbations [16]. Previous studies have demonstrated that the combination of two bronchodilators with distinct mechanisms of action (LABA + LAMA) not only enhances therapeutic efficacy but also reduces the use of rescue medications in patients previously treated with a single bronchodilator, without significantly increasing adverse effects [[17], [18], [19], [20], [21]].

Based on the clinical benefits of combining bronchodilators—where 1 + 1>2—we have innovatively combined IND and TIO to develop a compound dry powder inhalation formulation. We anticipate that this formulation will provide patients with superior bronchodilatory effects, effectively alleviate fluid retention, and offer them additional treatment options. This compound dry powder inhalant combines two active ingredients with lactose as an excipient. In addition to the carrier lactose, a certain proportion of fine-powdered lactose was added. This facilitates the easier dissociation of micronized drugs from the lactose system, promoting drug deposition in the lungs [22,23]. Compared to the original formulation, we selected magnesium stearate as an additional dispersant for this formulation for the first time, as the finely powdered raw material has poor flowability and tends to form large agglomerates, which could adversely affect the particle size distribution of the DPI [24,25]. Magnesium stearate enhances powder flowability and improves atomization performance in DPI [26]. Hydroxypropyl methylcellulose (HPMC)capsules have relatively low moisture content, effectively reducing the risk of hydrolysis of the active ingredient. Therefore, HPMC capsules are selected for powder filling [27].

Currently, the development of novel, improved medications for the treatment of COPD is essential. We ingeniously combined two bronchodilators, IND and TIO, and innovatively selected magnesium stearate as a dispersing agent and excipient to prepare a series of novel compound inhalation powders. After comprehensive evaluation, we identified the optimal formulation, F7, which demonstrates superior stability and a well-defined manufacturing process. This formulation represents a novel, highly effective, and safe treatment for COPD.

2. Materials and methods

2.1. Materials

Indacaterol Maleate, Tiotropium Bromide Monohydrate purchased from Shandong Rui Shun Pharmaceutical Co. Ltd. Carrier lactose (Inhalac 140) and fine powder lactose (Inhalac 500) purchased from MEGGLE GmbH & Co.KG. Magnesium stearate (inhalation grade) purchased from Peter Greven Nederland C.V. HPMC Empty Capsule Shells (3#) purchased from Qualicaps Co. Ltd. Sodium dihydrogen phosphate monohydrate (ACS) was purchased from Shengong Bioengineering (Shanghai) Co. Ltd. Phosphoric acid (GR), disodium ethylenediaminetetraacetate dihydrate (AR), acetonitrile (AR), glycerol (AR), and ethanol (AR) were all purchased from Sinopharm Chemical Reagent Co. Ltd.

2.2. Preparation of inhalation powder

First, the two raw materials IND and TIO were micronized using an air jet mill (Micron JETMILL Pilot, Nuozhe Fluid Technology Co., Ltd.). The grinding pressures were set to 5.0 bar and 3.0 bar, respectively, with feed pressures of 6.0 bar and 4.0 bar, and feed speeds of 9.0 rpm and 7.0 rpm. Next, lactose, fine powdered lactose, and magnesium stearate were alternately layered into the TRV mixer (PcM CONTROL UNIT MOBILE, PcM TRV1, GEA Process Engineering Ltd) for pre-mixing. Mix at a speed of 3 m/s for 6 min. Collect a portion of the pre-mixed material for later use. Subsequently, add the remaining pre-mixed material alternately with micronized IND and TIO into the TRV mixer for the first total mixing. Mix at a speed of 8 m/s for 2 min. Subsequently, the reserved premixed material was added to the TRV mixer at a speed of 8 m/s and mixed for 2 min to perform the second total mixing.

After completion, the total mixture was manually filled into hydroxypropyl methylcellulose capsule shells to prepare the formulation. Each capsule was filled with a weight of 25 mg, containing 143 μg of IND and 22.5 μg of TIO per capsule. Solid pharmaceutical composite sheets and pharmaceutical aluminum foil made by cold stamping polyamide/aluminum/polyvinyl chloride are used to package capsules. The total mixtures and formulations prepared using the aforementioned methods were subjected to subsequent studies.

2.3. Formulation screening

The total mixture content, mixture homogeneity, delivered dose (DD), and fine particle dose (FPD) of each formulation were evaluated using HPLC. High-Performance Liquid Chromatography Instrument (1260, Agilent Technologies) Parameters: Stationary Phase: Octadecylsilane-bonded silica gel (Agilent Poroshell EC-C₁₈, 4.6 mm × 100 mm, 2.7 μm), Mobile phase A: Phosphate buffer-acetonitrile (85:15), Mobile phase B: phosphate buffer-acetonitrile (30:70). Phosphate buffer is prepared by dissolving 3.73 g of sodium dihydrogen phosphate monohydrate in 2000 mL of water using an ultrasonicator, then adjusting the pH to 2.5 with phosphoric acid. Gradient elution was performed according to Table 1. Flow rate: 1.5 mL/min; column temperature: 40 °C; detection wavelength: 237 nm; injection volume: 30 μL.

Table 1.

Gradient program elution table.

Time (min)	Mobile phase A (%)	Mobile phase B (%)
0	100	0
8	50	50
8.1	100	0
10	100	0

2.3.1. Study on the dosage of fine lactose powder and magnesium stearate

First, prepare a master stock solution (25 mg/mL) for each formulation to assess the master stock's content (n = 6) and homogeneity (n = 12). We assembled the apparatus by connecting the inhaler (Breezhaler®, Novartis) to the adapter, then adjusted the flow rate on the flow controller (TPK2000, Copley Scientific) to 40 L/min. The inhalation flow rate of the Breezhaler inhaler is 30 L/min to 90 L/min. Place one capsule vertically into the inhaler with the capsule cap facing upward. Quickly press the inhaler button to puncture the capsule. Activate the vacuum pump (HCP5, Copley Scientific). After suctioning for 6 s, remove the inhaler and collection tube. Then measure 20 mL of the diluent (prepared by dissolving 250 mg of disodium ethylenediaminetetraacetate dihydrate in 990 mL of water, adding 10 mL of hydrochloric acid solution, and shaking well) into each collection tube. Shake to completely dissolve the sample in the collection tube, yielding a solution for testing the DD of the formulation (n = 10).

The Next Generation Cascade Impactor (NGI) serves as a critical method for evaluating the aerodynamic size of inhaled products [28]. We first coated the collection cup with a 1% glycerol-ethanol solution, assembled the NGI device (842, Copley Scientific), and checked for airtightness. We added 15 mL of diluent to the pre-separator and adjusted the flow rate to 60 L/min. Connect the inhaler to the adapter. Take one capsule, place it vertically in the inhaler with the cap facing upward, and quickly press the inhaler button to puncture the capsule. Maintain the button-pressed position while connecting to the impactor inlet. Release the button, activate the vacuum pump for 4 s, then shut off the pump and wait for 5 s. Take 5 additional capsules and repeat the above procedure for each. Dismantle the apparatus and clean each component separately. Obtain one set of FPD solution for testing the formulation, with a total of three sets tested (n = 3).

Using the aforementioned HPLC method, the amounts of lactose fine powder and magnesium stearate in the formulations were investigated. The studied formulations (F1-F8) are shown in Table 2.

Table 2.

Formulation dosage review table (300 g/batch).

Formulation Code	IND (g)	TIO (g)	Inhalac140 (g)	Inhalac500 (g)	Magnesium stearate (g)	Inhalac500/Lactose (w/w%)	Magnesium stearate/Total (w/w%)	Batch number
F1	1.716	0.27	298.014	/	/	/	/	IND-02/TIO-02
F2	1.716	0.27	289.074	8.940	/	3	/	IND-02/TIO-02
F3	1.716	0.27	280.134	17.880	/	6	/	IND-02/TIO-02
F4	1.716	0.27	271.194	26.820	/	9	/	IND-02/TIO-02
F5	1.716	0.27	253.311	44.703	/	15	/	IND-02/TIO-02
F6	1.716	0.27	279.711	17.853	0.45	/	0.15	IND-02/TIO-02
F7	1.716	0.27	279.288	17.826	0.90	/	0.30	IND-02/TIO-02
F8	1.716	0.27	278.724	17.790	1.50	/	0.50	IND-02/TIO-02
F9	1.716	0.27	279.288	17.826	0.90	/	0.30	IND-01/TIO-01
F10	1.716	0.27	279.288	17.826	0.90	/	0.30	IND-03/TIO-03

2.3.2. Investigation of the particle size distribution of active pharmaceutical ingredient

The particle size distribution of the active pharmaceutical ingredient (API) was analyzed using a laser particle size analyzer (HELOS BR, Sympatec GmbH) with a RODOS/M dispersing unit (dispersion pressure: 4 bar, feed rate: 80%, measured with an R₂ lens, valid for measurements with optical concentration greater than 5%). We further investigated the effects of IND and TIO with different particle size distributions on the DD and FPD of the formulation, employing the same testing method as described in Section 2.3.1. The studied formulations (F7, F9, F10) are presented in Table 2.

2.4. Morphological characterization

Using a scanning electron microscope (SIGMA 300, ZEISS), images of IND, TIO, and samples of different formulations were captured before and after pulverization to observe and analyze particle size, morphology, and the mixing state of the drug and carrier.

2.5. Particle size distribution analysis

Using a laser particle size analyzer, the particle size distribution of the prepared samples was measured. The D₁₀, D₅₀, and D₉₀ values of samples from different formulations were analyzed to screen for DPI meeting the standards, ensuring optimal aerodynamic performance.

2.6. Powder X-ray diffraction analysis

Powder X-ray diffraction analysis (Empyrean, Malvern Panalytical) was performed on IND, TIO, and prepared samples before and after grinding. The 2θ shift values of each sample were evaluated to analyze the crystalline phase transformation patterns of IND and TIO before and after processing, as well as the crystalline states of different DPI components.

2.7. In vitro aerodynamic assessment

We analyzed the aerodynamic properties of the DPI through in vitro deposition testing to evaluate the trend of drug delivery to the lungs. Using the same NGI apparatus and detection method as described in Section 2.3.1, a total of three groups were tested.

2.8. Stability study

Next, we examined the phase changes and aerodynamic stability of the samples during storage to analyze and determine the quality variation of the DPI over time. The samples were placed in a pharmaceutical stability testing chamber (SHH-500SD-2T; Chongqing Yongsheng Experimental Instruments) for six months under storage conditions of 40 ± 2°C and 75% ± 5% RH.

2.9. Statistical analysis

Statistical analysis data are expressed as Mean ± SD. Statistical analysis was performed using GraphPad Prism 9.5.0 software. One-way ANOVA was employed to indicate significant differences. ns indicates no significance; *P < 0.05 is considered statistically significant; **P < 0.01, ***P < 0.001, and ****P < 0.0001 denote extremely significant differences.

3. Results and discussion

3.1. Formulation screening

3.1.1. Screening of optimal dosages for fine lactose powder and magnesium stearate

The results of the fine powder lactose dosage formulation study are shown in Table 3: Except for F1, the total mixture content of all formulations remained within standard limits. Additionally, the mixing uniformity results indicate that as the mass proportion of fine powdered lactose in the system increased, the RSD for mixing uniformity improved from >5% in F1 to <1% in F3. However, when the fine powdered lactose proportion reached 15% (F5), the RSD deteriorated again to >5%. This occurs because during mixing, without fine powdered lactose (F1), density differences between the drug and lactose excipient cause uneven mixing or even separation. Conversely, with a high proportion of fine powdered lactose (F5), the friction coefficient of the total mixture powder increases, reducing flowability and weakening the shear dispersion effect of the mixing equipment, also leading to uneven mixing [29]. Therefore, both F1 and F5 fail to meet requirements.

Table 3.

Results of the study on the dosage of fine powder lactose and magnesium stearate.

Formulation Code		Content (%) (n = 6)	Mixing uniformity (RSD %)(n = 12)	DD (μg) (n = 10)	FPD(μg) (n = 3)
F1	IND	94.7 ± 1.3	5.71	78.9 ± 3.9	38.0 ± 1.3
	TIO	94.2 ± 1.4	6.19	9.1 ± 0.5	1.8 ± 0.1
F2	IND	98.9 ± 0.7	1.90	83.1 ± 4.3	44.0 ± 0.9
	TIO	98.2 ± 0.8	1.69	9.8 ± 0.6	2.6 ± 0.1
F3	IND	99.5 ± 0.5	0.80	85.0 ± 5.1	46.0 ± 0.8
	TIO	99.2 ± 1.0	0.60	10.1 ± 0.6	2.7 ± 0.1
F4	IND	99.3 ± 0.8	1.10	83.8 ± 3.8	45.0 ± 1.2
	TIO	99.0 ± 0.5	1.40	10.0 ± 0.5	2.6 ± 0.1
F5	IND	93.7 ± 1.6	5.40	77.8 ± 4.6	40.0 ± 0.2
	TIO	92.5 ± 1.3	5.90	9.0 ± 0.5	1.9 ± 0.1
F6	IND	99.3 ± 0.6	1.00	85.3 ± 5.5	48.0 ± 0.4
	TIO	99.2 ± 0.6	1.10	10.2 ± 0.6	2.9 ± 0.04
F7	IND	99.6 ± 0.1	0.70	84.8 ± 3.8	51.0 ± 1.0
	TIO	99.3 ± 0.5	0.60	10.1 ± 0.5	3.1 ± 0.1
F8	IND	99.3 ± 0.6	1.30	86.1 ± 3.9	55.0 ± 0.9
	TIO	99.1 ± 0.7	1.20	10.2 ± 0.5	3.5 ± 0.1

Note: Standard limits: Content (95.0%–105.0%); Homogeneity of mixture (RSD ≤5.0%); DD (IND: 73–97 μg, TIO: 8.5–11.5 μg); FPD (IND: 40–60 μg, TIO: 2.0–4.2 μg).

By comparing the specifications of the reference inhalation powder tiotropium bromide and indacaterol maleate, it was found that when the fine powder lactose content ranged from 3% to 9% w/w in formulations F2, F3, and F4, all specifications of the total mixture met requirements. Therefore, a fine powder lactose content of 6% w/w (F3) was selected for subsequent formulation studies.

Table 3 also shows that in F3, when magnesium stearate was not added, although the content of both active ingredients and the uniformity of mixture in the total mixture met the standard limits, the FPD of both active ingredients in the formulation still tended toward the lower limit of the standard range. Conversely, when magnesium stearate concentrations in F6, F7, and F8 ranged from 0.15% to 0.50% w/w, all parameters in the total mixture met standard limits. It is worth noting that in the F7 formulation, which contains 0.30% (w/w) magnesium stearate, the FPD values for both active ingredients were close to the median of the standard limits, yielding the best results and ensuring batch-to-batch consistency. Furthermore, compared to Formulations F6 and F8, Formulation F7 has a lower RSD value. Studies have shown that highly uniform mixtures (with low RSD values) represent a thermodynamically stable “ordered mixture,” which reduces agglomeration of fine drug particles, allows them to adhere tightly to the lactose carrier surface, and preserves the original particle morphology and aerodynamic properties, thereby maintaining dose uniformity during storage. In contrast, the higher RSD values of Formulations F6 and F8 indicate greater sensitivity to environmental conditions than those of Formulation F7, which may lead to potential fluctuations in DD and FPD after long-term storage [30,31]. Therefore, we selected 0.30% w/w magnesium stearate for subsequent studies.

3.1.2. API granularity assessment

The particle size of the API is a key factor determining DPI performance. Optimal pulmonary delivery is achieved when the aerodynamic diameter of the API is between 1 and 5 μm, and peak fine particle deposition is typically observed at around 2–3 μm [32]. When the API is smaller than this range, it exhibits stronger interparticle cohesion and greater adhesion to the carrier, making it difficult for the particles to detach during inhalation. Conversely, APIs larger than 5 μm are increasingly cleared from the airflow due to inertial impaction in the oropharynx, resulting in a sharp decline in the inhalable fraction [33]. The particle size distribution of the API was analyzed using a laser particle size analyzer to ensure that the majority of drug particles exhibited an aerodynamic diameter distribution within the 1–5 μm range. As shown in Table 4, the particle size distributions of the IND-02 and TIO-02 API batches were more consistent with the requirements compared to the other groups. This is because the median particle size of this group is closer to 2 μm, which falls precisely within the optimal aerodynamic diameter range for the API and limits the presence of fine particles that could hinder dispersion. Therefore, the API from the IND-02 and TIO-02 batches was selected for subsequent studies.

Table 4.

API particle size distribution test results (n = 6).

API	Batch number	Particle size (μm)
		D10	D50	D90
IND	IND-01	0.50 ± 0.01	1.49 ± 0.02	3.56 ± 0.04
	IND-02	0.65 ± 0.01	2.01 ± 0.02	4.75 ± 0.03
	IND-03	0.93 ± 0.01	2.75 ± 0.04	6.58 ± 0.05
TIO	TIO-01	0.59 ± 0.01	1.60 ± 0.02	3.86 ± 0.04
	TIO-02	0.68 ± 0.02	2.12 ± 0.02	5.06 ± 0.03
	TIO-03	0.85 ± 0.01	2.65 ± 0.02	6.33 ± 0.03

The particle size distribution of the active pharmaceutical ingredient affects the DD and FPD of the formulation. In DPI, fine active pharmaceutical ingredients (APIs) are prone to uncontrolled particle agglomeration due to electrostatic forces and extremely high surface energy, resulting in increased variability in fine particle output and posing a risk to dose uniformity. Therefore, the FPD value must be maintained within an appropriate range [34]. The results of the API granularity distribution analysis are shown in Fig. 1As the API particle size increases sequentially from F9, F7, to F10, the DD and FPD of the formulation exhibit a decreasing trend. Compared with F7 and F10, F9—which has a smaller particle size—has the highest FPD value, but it is already close to the upper limit of the standard, making it more prone to agglomeration during large-scale production and long-term storage, and thus more difficult to maintain stability. In contrast, F7, with its moderate particle size, has an FPD value closer to the median of the standard, ensuring better physical stability and batch-to-batch reproducibility for the formulation within the specified limits. Therefore, we selected F7 as the final formulation.

Fig. 1.

Statistical analysis of API particle size for F7, F9, and F10. (A) Delivery dose (DD) distribution chart. (B) Fine particle dose (FPD) distribution chart.

3.2. Morphological characterization

Samples of IND, TIO, and different formulations before and after micronization were analyzed using scanning electron microscopy (SEM). Both pre- and post-micronization IND exhibited irregular flake-like morphology with particles stacked upon each other, showing no obvious particle aggregation (Fig. 2A–B). Before micronization, TIO appeared as irregularly shaped particles of varying sizes with smooth edges and no agglomeration. After micronization, TIO particles exhibited agglomeration (Fig. 2C–D), indicating that micronized TIO possesses strong cohesive properties. This necessitates the addition of magnesium stearate in the formulation to disperse and prevent agglomeration of TIO.

Fig. 2.

SEM images. (A) IND before pulverization. (B) IND at 0 h after pulverization. (C) TIO before pulverization. (D) TIO at 0 h after pulverization. (E) F1. (F) F3. (G) F5. (H) F7.

Morphological characterization and comparison were performed on formulations F1, 3, 5, and 7 to analyze the reasons for significant differences among the formulations. The morphology of samples F1, 3, and 5 revealed an increasing number of fine powder particles within the field of view (Fig. 2E–G), with these fine particles exhibiting a tendency to agglomerate or adhere to the surface of coarse lactose granules. Comparing F3 (Fig. 2F) with F7 (Fig. 2H), the presence of the dispersing agent magnesium stearate in F7 resulted in uniform distribution of the visible fine lactose powder throughout the powder system without agglomeration. Therefore, F7 is the optimal formulation.

3.3. Particle size distribution analysis

The laser particle size analyzer calculates particle size and distribution by measuring the scattered light signal generated when particles are exposed to laser irradiation. We therefore employed this instrument to examine the particle size distribution of formulations F1, F3, F5, and F7. When the mass fraction of fine lactose powder in the system increased to 15%, the D₁₀, D₅₀, and D₉₀ values of the formulation exhibited significant changes compared to F1 (Fig. 3A), demonstrating that fine lactose powder substantially influences the particle size distribution of the formulation. This is because fine lactose powder, as the third component of DPI, preferentially adheres to high-energy active sites on the surface of coarse lactose particles, forming a “satellite” structure. This directly increases the proportion of fine particles, causing D₁₀ and D₅₀ to shift toward smaller particle sizes. These fine lactose particles can also fill the voids between coarse carrier particles, improving the powder's packing state and acting as “solid lubricants,” which further reduces the D₉₀ value [[35], [36], [37]]. Subsequent analysis comparing the D₁₀ and D₉₀ particle size distributions of F3 and F7 demonstrated that the inclusion of magnesium stearate in the formulation did not cause significant changes in the D₁₀ or D₉₀ values (Fig. 3A–i, iii). After comprehensive evaluation, F7 was still selected as the optimal formulation.

Fig. 3.

A: Statistical distribution of particle sizes for F1, F3, F5, and F7. (A-i) D₁₀ statistics, (A-ii) D₅₀ statistics, (A-iii) D₉₀ statistics; B: XRD patterns of raw materials and formulations. (B-i) IND before grinding and after grinding at 0 h, (B-ii) TIO before grinding and after grinding at 0 h, (B-iii) XRD patterns of formulations F1, F3, F5, and F7.

3.4. Powder X-ray diffraction analysis

During the preparation of DPI, we use air-jet milling to reduce the particle size of the active pharmaceutical ingredient (API); however, the intense mechanical forces exerted by air-jet milling may lead to the accumulation of crystal defects and polymorphic transformations in the API, which directly affect the drug's stability [38,39]. Therefore, maintaining the consistency of the API's crystal form before and after milling, as well as during the formulation process, is essential for ensuring the quality of DPI. Powder X-ray diffraction is a technique that infers the internal structure of crystals by measuring the angles and intensities at which X-rays are diffracted by the crystals. This method was employed to detect and analyze the crystal phases of the samples [40]. The results indicate that the XRD patterns of IND and TIO showed no significant changes before and after pulverization (Fig. 3B–i, ii). Furthermore, compared to the pre-pulverized IND and TIO, the 2θ shifts of the XRD diffraction peaks in the post-pulverized samples were all within ±0.2°. This demonstrates that the jet milling process reduced the particle size of the raw materials without altering their crystalline morphology during pulverization. Similarly, the XRD diffraction results (Fig. 3B–iii) reveal that all formulations exhibit identical characteristics, with the 2θ values of the XRD diffraction peaks showing deviations within ±0.2°. This indicates that the components of different formulations are all in the same crystalline state, and that no solid-state transformations or mutual reactions occurred between the components during the formulation process [38].

3.5. In vitro aerodynamic assessment

The optimal formulation F7 was selected for DPI preparation. In vitro deposition was measured using NGI to analyze the effective particle deposition rate of both drugs in the lungs. Table 5 and Fig. 4A and B demonstrate that this DPI exhibits optimal inhalation and deposition characteristics. The recovery rates of both active ingredients in the samples were excellent, approaching 100%.Under the same experimental conditions, compared with the original formulation, Spiriva® capsules, the FPD value of TIO in formulation F7 was closer to the median of the standard range (2.0–4.2 μg), ensuring that the amount of TIO delivered per inhalation remained at an optimal level, reflecting the superior delivery accuracy of this combination formulation [41]. Formulation F7 delivers two synergistic drugs with a single inhalation, eliminating the hassle of using two inhalers sequentially and minimizing user-related variability, thereby offering greater convenience. The fraction of fine particles (FPF) reached a high level. The aerodynamic particle size distribution is a key factor determining the airway deposition behavior of drugs. It is generally accepted that a mass median aerodynamic diameter (MMAD) < 5 μm is a prerequisite for particles to reach the lower respiratory tract [42]. In this study, the MMAD of both drugs in formulation F7 was approximately 3 μm, comparable to that of the reference formulation Onbrez® Breezhaler®, and fully met the particle size requirements for efficient pulmonary deposition [43]. The geometric standard deviation (GSD) was low, indicating uniform particle size distribution. The impactor sized mass (ISM) further confirmed that the actual drug dose delivered to the patient per inhalation was appropriate. Overall, these results suggest favorable in vitro aerodynamic properties that are consistent with deep lung deposition characteristics.

Table 5.

Statistical results on extracellular deposition of F7 and Spiriva® capsules (n = 3).

Parameter	F7		Spiriva® capsules
	IND	TIO	TIO
Recovery (%)	96.84 ± 0.510	95.95 ± 1.375	/
FPD (μg)	51.67 ± 2.082	3.12 ± 0.021	5.13 ± 0.001 [41]
FPF (%)	56.07 ± 1.842	34.83 ± 0.038	/
MMAD (μm)	3.05 ± 0.046	3.59 ± 0.061	/
GSD	1.83 ± 0.012	1.79 ± 0.012	/
ISM (μg)	58.21 ± 1.130	4.04 ± 0.059	/

Note: Recovery is the ratio of measured total quantity to theoretical labeled quantity. FPF represents the proportion of cumulative drug particles with aerodynamic diameters less than 5 μm. MMAD denotes the diameter at which 50% of the drug particles in the sample are smaller than this value. GSD indicates the width of the particle size distribution. ISM is the sum of drug mass collected by impactors of grades 2–7 and the MOC-grade impactor per sample.

Fig. 4.

Settlement of inhaled aerosol particles at different heights for IND(A) and TIO(B) in F7, (C) XRD patterns of samples at Day 0 stability. (D) Comparison of drug deposition levels for IND in both samples. (E) XRD patterns of samples after accelerated 6-month stability testing. (F) Comparison of drug deposition levels for TIO in both samples.

3.6. Stability study

We further conducted accelerated stability studies on F7. The data show that the XRD patterns of samples stored for 0 days and accelerated for 6 months were essentially identical, with 2θ value shifts within ±0.2°, indicating no significant alteration in the drug's crystal form during storage (Fig. 4C–E). These stability results are superior to those observed for some DPI formulations using lactose as a carrier, which exhibited drug recrystallization under accelerated conditions. This is attributed to the crystal form-protective effect of magnesium stearate in the formulation, as well as the fact that fine lactose powder preferentially occupies high-energy active sites, thereby inhibiting degradation-promoting interactions between the drug and the carrier [35,44,45]. Concurrently, in vitro deposition tests were conducted on both samples. Results revealed no significant changes in the DPI's FPD, fine particle fraction, geometric standard deviation, or median mass aerodynamic diameter throughout the stability assessment. This confirms that the aerodynamic properties of the samples remained stable during the accelerated 6-month period (Fig. 4D, F, Table 6), demonstrating excellent stability.

Table 6.

Statistical results of F7's in vitro deposition after accelerated testing for 6 months (n = 3).

Parameter	IND	TIO
Recovery (%)	96.18 ± 0.405	94.45 ± 0.709
FPD (μg)	50.33 ± 0.577	3.10 ± 0.100
FPF (%)	55.52 ± 0.470	34.53 ± 0.876
MMAD (μm)	3.05 ± 0.042	3.68 ± 0.080
GSD	1.83 ± 0.012	1.77 ± 0.036
ISM (μg)	56.51 ± 0.053	4.10 ± 0.100

4. Limitation

Although we have demonstrated that this DPI is capable of effective deposition in the lungs, there are still limitations. We did not simulate the impact and deposition of the inhaled drug in the human oropharyngeal region, and thus cannot accurately and comprehensively reflect the in vitro evaluation results of this inhaled powder formulation, which affects the assessment of its consistency with the original formulation. Therefore, it is particularly important for future studies to use models that closely resemble the actual human anatomy to simulate and evaluate this compound inhalation formulation.

5. Conclusion

This study successfully developed and systematically characterized a novel IND/TIO combination inhalation powder formulation for the treatment of chronic obstructive pulmonary disease (COPD). The optimized F7 formulation, utilizing a two-stage lactose carrier and magnesium stearate, effectively improved the flowability and addressed agglomeration issues of the micronized active pharmaceutical ingredient (API), while demonstrating excellent aerodynamic performance and accelerated stability. This formulation offers a new and effective option for the clinical treatment of COPD, while its design approach provides valuable insights for the development of bicomponent dry powder inhalers. We will conduct further in vivo efficacy and safety evaluations to validate its potential for clinical translation.

Data availability

All data supporting the findings of this study are included in this published article.

Ethics approval

Not applicable.

Declaration of generative AI in scientific writing

Not applicable.

Funding information

This study was supported by the State Key Laboratory of Drug Regulatory Science (2025SKLDRS0323), Taishan Industry Leading Talent Project Special Fund (tscx202306073) and Tai'an “Double Ten” Sci-Tech Innovation Project (23JSGG12).

CRediT authorship contribution statement

Xiaobing Wang: Conceptualization, Funding acquisition, Project administration, Resources. Xiaoxi Xu: Formal analysis, Investigation, Writing – original draft. Chunlei Cheng: Formal analysis, Investigation, Writing – original draft. Jing Zheng: Formal analysis, Writing – review & editing. Wenkun Liu: Investigation, Writing – original draft. Yuwen Xu: Conceptualization, Funding acquisition, Project administration. Jihang Xu: Writing – original draft. Lihua Wu: Writing – review & editing. Xianhai Sun: Formal analysis, Investigation. Weijian Wang: Conceptualization, Funding acquisition, Project administration, Resources. Hao Fang: Conceptualization, Funding acquisition, Project administration, Resources.

Declaration of interests

The authors declare no competing interests.