WO2008040841A1

WO2008040841A1 - A desiccant system for the regulation of humidity in inhalation powders

Info

Publication number: WO2008040841A1
Application number: PCT/FI2007/050486
Authority: WO
Inventors: Vesa-Pekka Lehto
Original assignee: Lab Pharma Ltd.
Priority date: 2006-10-04
Filing date: 2007-09-12
Publication date: 2008-04-10
Also published as: FI20065636A0

Abstract

A desiccant system for controlling the humidity of inhalation powder, said system comprising an air-tight container containing a dry desiccant, and a drug chamber containing the inhalation powder, the air-tight container being arranged inside the drug chamber or in its vicinity, said desiccant being capable of maintaining a fixed point of humidity as a saturated solution of at least one salt

Description

A DESICCANT SYSTEM FOR THE REGULATION OF HUMIDITY IN INHALATION

POWDERS

The invention concerns a desiccant system for controlling the humidity of a powder formulation intended to be inhaled from a powder inhaler. The desic- cant preferably consists of magnesium chloride, which may be anhydrous or a hexahydrate. Both forms of salts show substantially greater water absorption capacity than silica gel. Moreover, they are able to maintain a very stable relative humidity around the inhalation powder. Other salts having a fixed relative humidity point lower than 50% RH as saturated water solutions can also be used.

The invention also concerns a powder inhalation apparatus wherein the desiccant system described above is utilized.

The invention further provides a method for efficiently regulating the humidity inside the medicament chamber of a powder inhalation apparatus.

The invention additionally concerns the use of the new form of desiccant for controlling the humidity of inhalable powder formulations.

Powder inhalers are used for producing inhalable drug particles to be inhaled into the lungs. The clinical effect depends on the amount of inhalable particles, which is normally much smaller than the total drug dose delivered. The diameter of these particles is some micrometers. Powder inhalers can be di- vided roughly into two categories: reservoir-based multi-dose inhalers (MDPIs) and devices using pre-metered doses (capsules, blisters). When MDPI s are used, each dose must be metered from the drug reservoir by a mechanism and the dose must be transferred out of the reservoir for inhalation. Therefore, because of the moving/sliding parts, it is difficult to construct an airtight reservoir. Snce the device should be usable for months, or even up to one year or longer, even small clearances in critical parts are hazard- ous in respect of moisture penetration if no desiccant is used inside or outside the drug reservoir. I n this case, the plastic to plastic sliding parts do not hinder the penetration of water vapor and the powder formulation can ad(ab)sorb moisture freely. Not only MDPIs, but also capsules and blisters for inhalation powders often need protection against moisture. Surprisingly, it has been shown that even aluminium blisters may leak, especially in warm and humid conditions {Journal of Aerosol Medicine, Volume 18, Number 3, 2005, pp. 304-310).

A well-known way of protecting any inhaler or powder formulation from moisture is to place it in an impermeable bag or container together with a conventional desiccant pack This pack is normally a plastic container filled with silica gel and provided with a permeable membrane to facilitate the penetration of moisture into the desiccant.

The powder formulation may consist of plain micronized drug particles, or the particles may be mixed with other particulate excipients, usually coarser lactose monohydrate. I n most cases, the micronization of the drug substance is carried out by air jet milling. I n this method, the larger drug particles are violently crushed through collisions and the particle size distribution is wide, typically from 0.5 to 10 micrometers. The particle morphology is difficult to control and in most cases a more or less amorphous (unordered = noncrystalline) substance is created during the process even if the original drug was completely crystalline. Unordered mass is thermodynamically unstable and tends to recrystallize, especially at higher temperatures and humidity. Other micronization methods are spray drying, ball milling, supercritical precipitation and microcrystallization which may produce unstable material as well.

According to pharmacopoeal requirements, the amount and mean particle size in the dose delivered from powder inhalers must be within a certain range during storage and shelf-life of the product. Correspondingly, the physical and chemical stability of the product must be tested by the manufacturer. I n accelerated stability studies, the product is stored at an elevated temperature and humidity where the chemical and physical changes are also accelerated. The most common reason for the poor stability of powder formulations is the decrease in the amount of inhalable particles and an increase in the variation of unit dose content (dose accuracy). This means that the clinical effect decreases and the variation between individual doses increases. This deterioration of the formulation is caused by physical changes that lead to the agglomeration of the drug particles with each other and the carrier sugar. Drug particles are less prone to separate during inhalation and the flow properties are poorer, thus making dose metering inaccurate.

The effect of moisture is essential in the stability of inhalation powders. It may cause recrystallization of amorphous substances (and agglomeration), affect the static electricity of the powder and inhaler and cause formation of liquid bridges between the particles. Hydrophilic components are more prone to form liquid bridges than hydrophobic components. Hygroscopic drugs are almost impossible to formulate as inhalation powders by conventional means. An increase in temperature and/or humidity also reduces the chemical stability. The phenomena mentioned above are commonly accepted in the field of the art and documented in numerous published studies.

Theoretically, regarding the stability of the formulation, it would be most ad- vantageous to keep the powder formulation at a low and constant humidity. However, this is difficult over a prolonged period because the drug reservoirs/chambers are not completely tight and moisture will penetrate into them through any plastic walls and through joints and clearances in the dose metering mechanism. Moisture protection of the powder formulation in the drug reservoir of a powder inhaler is very difficult. Ideally, the desiccant should maintain the humidity around the powder constant at different humidity levels and temperatures for several months, preferably several years. No such construction has as yet been described. One solution is to improve moisture control around the powder formulation.

DESI CCANT MATERI ALS

Slica αel

Slica gel is a reversible drying agent, which is in equilibrium with ambient relative humidity. When totally dry, it can adsorb 17 w-% at 35% RH and 30 w-% at 80% RH at 25⁰C. If silica gel is brought from 80% RH back to 35% RH, it will release 13 w-% of its dry weight as moisture in order to reach an equilibrium under the new conditions. The adsorption kinetics of silica gel depends on temperature and the capacity decreases when the temperature increases.

Silica gel is the only desiccant used in the powder inhalers on the market. It is considered very safe and its use has been approved by the medical authorities. It is, however, not the ideal desiccant for the purpose. Its capacity is relatively low and it is not able to maintain the internal humidity stable (Lehto and Lankinen, Int. J. Pharm. 275, 155 (2004)).

Molecular sieve desiccants

These are practically irreversible desiccants, which adsorb moisture from the ambient air very effectively until they have used up their capacity of ca. 20 w-%. These are not good desiccants for powder inhalers. Their capacity is low and they maintain the internal humidity too dry in most cases. The dry- ness may cause strong triboelectric charging of the powder and plastic parts, thus disturbing the dose metering. These materials have not been used with powder inhalers.

Chemical absorbents

Several chemicals are able to bind water chemically. Some absorb water as crystalline water like calcium sulphate and some react with water like phosphorous pentoxide. Regarding the present invention, the salts which are ca- pable of binding crystalline water as anhydrates, after which they act as saturated salt solution desiccants, are extremely interesting due to their very high water absorption capacity. .Examples of such salts are magnesium and calcium chloride. Chemical absorbents are not used with powder inhalers.

Saturated salt solutions

Several salts show an interesting moisture regulation effect as saturated solutions. They are able to maintain the ambient humidity stable as long as there are undissolved crystals left. They are, therefore, in an equilibrium with the relative humidity of the surrounding air in a different way than silica gel. However, if the humidity of the ambient air falls under the equilibrium humidity, the salt solution releases moisture in order to maintain the humidity stable. The kinetics is only slightly dependent on temperature between 10 and 3O⁰C. The capacity can be calculated directly from the solubility of the salt. Saturated potassium acetate solution regulates the relative humidity of the ambient air to 23.4% RH at 1 O⁰C and to 21.6% RH at 3O⁰C. The corresponding values for magnesium chloride are 33.5 and 32.4% RH. Saturated salt solutions have not been used or suggested for use with powder inhalers. Comparison between the water absorption capacities of magnesium chloride and silica αel

The aim of the present invention is to develop a system that can maintain the maximum relative humidity around the powder formulation below 50% RH for a prolonged period.

A comparison of the capacities of silica gel and some saturated salt solutions can be easily done.

For anhydrous magnesium chloride, the solubility is 54.3 g / 100 cm³. Correspondingly, 1 gram may absorb 1.84 g of water as a saturated solution, maintaining the ambient relative humidity below 33%.

For magnesium chloride hexahydrate, the solubility is 302 g / 100 cm³, and 1 gram may absorb 0.33 g of water.

For potassium acetate, the solubility is 258 g / 100 cm³ and thus 1 gram of the salt may absorb 0.39 g of water; with this salt the ambient humidity generated is below 23 %.

Dried silica gel is capable of absorbing ca. 11 w-% at 23% RH and ca. 20 w- % at 33% RH. Thus 1 g of silica gel may absorb 0.11 g water to reach 23% RH and 0.20 g to reach 33 % RH. Consequently, the capacity of anhydrous magnesium chloride is 9.2-fold compared with silica gel, 1.7-fold for hexahydrate and 3.5-fold for potassium acetate.

I n the present invention, salts which are able to maintain a fixed relative humidity as saturated water solutions, are used as desiccants. Salts which are non-toxic and have a fixed point humidity lower than 50% RH are preferred. Examples of such salts include magnesium chloride, calcium chloride and potassium acetate. Some of the salts may also contain crystalline water, such as magnesium chloride and calcium chloride. When these salts are used in dehydrated form, they first absorb the crystalline water as chemical desiccants and then subsequently form the saturated salt solution. The highest water absorption capacity can be obtained in this way. When used in hydrated form, the capacity is correspondingly lower but the benefit is very stable humidity, which stays very close to the fixed point hu- midity of the solution. Test results are given in examples 1 -4.

The desiccant must be able to have vapor contact with the powder formulation without any risk of contamination. However, vapor contact with ambient air should be avoided as effectively as possible. This can be achieved by placing the desiccant in a closed plastic container, the plastic material being permeable to water vapor. Vapor contact with ambient air can be minimized by:

1. placing the desiccant container inside the drug powder chamber of an inhaler, the chamber being made of material which is substantially impermeable or having poor permeability to water vapor;

2. placing the desiccant container inside a protective cap of a powder inhaler, both the cap and the rest of the body of the inhaler being substantially impermeable or having poor permeability to water vapor and the cap being tightly sealable to the body; the desiccant container having high permeability to water vapor;

3. placing both the inhaler provided with a drug powder chamber, or a separate dose chamber, and the desiccant container inside a separate cover/bag which is substantially impermeable or having poor perme- ability to water vapor; the desiccant container having high permeability to water vapor. PRI OR ART

The characteristics of several saturated salt solutions in humidity regulation are widely known. The fixed point humidity for some solutions at 25⁰C are:

Salt %RH cesium fluoride 3.4 lithium bromide 6.4 lithium chloride 11.3 potassium acetate 22.5 calcium chloride 31 magnesium chloride 32.8 potassium carbonate 43.2 sodium bromide 57.6 potassium iodide 68.9 sodium chloride 75.3 potassium chloride 84.2 potassium sulphate 97.3

Such solutions are used when a certain humidity should be maintained, for example in the calibration of instruments, in corrosion studies and many kinds of humidity boxes for storing cigars, guns or musical instruments. I n the present work, potassium chloride was used to maintain the storage conditions of 84.2% RH in a desiccator.

US 6 244 432 discloses a gun case humidity control device for use in maintaining a desired humidity in the gun case, a water vapor permeable pouch and a thickened saturated solution having a suitable humidity control point. Materials and the water permeabilities of the pouch have been specified. The water solution/suspension contains water, 20-70% of the selected salt and a thickener. US 5 936 178 discloses the same principle of humidity regulation: a thickened saturated salt solution in a water vapor permeable pouch is used in storage containers. Examples include moisture regulation for stringed instruments, cigars, gummy bears/ licorice, dried fruit, electronic devices, fine jewelry, fire arms, painting, sculptures, tapestries as well as the objects themselves and whatever else is best stored under conditions of constant humidity.

US 6 921 026 B2 discloses an identical solution for maintaining a desired humidity in food packages, but in addition to containing an oxygen scavenger.

US 2002/0014305 A1 discloses a desiccant composition mix which comprises salt and modified starch. A large number of salts having different fixed points of humidity are named. The desiccant is placed in a special water- permeable film container. Modified starch forms a gel with water and a deliquescent salt, thus diminishing the risk of leaks through permeable film container.

Deliquescent salts as desiccants are also mentioned in US 4 749 392, where a general purpose desiccant pack comprises a permeable film.

There are numerous patents concerning different powder inhalers, but only a few of them comprise a desiccant system. Three patents are relevant in re- spect of the present invention. I n the launched version of the device described in US patents 4 524 769 and 4 907 583 (Turbohaler), the desiccant (silica gel) is placed inside the powder inhaler, in a container provided with a permeable membrane. The drug reservoir with the dose metering system is not tightly sealed, but contains moving/sliding plastic parts. Hence, silica gel is able to desiccate the drug formulation and also all other inner parts and spaces as well. The system works well when handled according to the pa- tient instructions. There are, however, some drawbacks in the desiccant system. If the protective cap is not placed correctly, or not used at all, the capacity of silica gel is exhausted in a short time, especially in a humid environment. The capacity of silica gel is relatively low and each time a dose is taken, the protective cap must be removed and water vapor enters inside. A relatively large amount of silica gel is needed to ensure correct functioning. Due to the silica gel, the internal humidity will increase as a function of time in humid conditions. Several moisture sensitive drugs cannot be used in the device.

A very similar device and desiccant function as in the Turbohaler are described in US 5,829,434. The device can be sealed tightly by a protective cap, but there is a conventional desiccant pack inside the cap. The performance and drawbacks are very similar to those of the Turbohaler.

I n US 6, 132,394, the desiccant is placed in a tightly sealed plastic container, which is placed inside a drug reservoir. Moisture permeability through the desiccant container is higher than that of the drug reservoir. I n this system, the desiccant (silica gel) only desiccates the inside of the reservoir. However, because each dose must be metered and taken outside the reservoir for inhalation, the dose metering system must be carefully constructed. The drawback is that the invention was designed for silica gel, which is not ideal for the purpose due to its sorption behavior and the limited adsorption capacity of silica gel as these issues were discussed above.

When comparing these three patents, US 6, 132,394 protects the formulation best as the desiccant is inside the tightly sealed drug reservoir. However, sealing the reservoir is demanding because moving parts are needed for metering the powder dose. Due to the low capacity of silica gel, a fairly large amount is needed to ensure long protection. I n the two other devices, extra humidity penetrates inside the device every time the cap is opened and a dose is inhaled. I nserting the cap tightly is not foolproof.

DESCRI PTI ON OF THE I NVENTI ON

The desiccant system according to the invention is characterized in the subsequently presented independent claim 1 , and preferred embodiments in the dependent claims 2 - 8.

The invention also concerns an inhalation apparatus utilizing the desiccant system. The apparatus is defined in the subsequently presented independent claim 9, and its preferred embodiments in the dependent claims 10 - 15.

The invention further provides a method for regulating the humidity in an inhalation apparatus, and the characteristics of this method are presented in the subsequent independent claim 15.

Finally, the invention also concerns the use of the desiccant for controlling the humidity of inhalation powders, the desiccant being dry when placed in the container but capable of maintaining a fixed point of humidity as a saturated solution of at least one salt.

As was concluded above, the saturated salt solution having a relatively low fixed relative humidity would be an ideal desiccant for powder inhalers be- cause it might be able to maintain a stable relative humidity around the powder formulation and would have a higher absorption capacity than silica gel or other conventional solid desiccant materials. However, the system would have to be absolutely safe with respect to patient safety and no contamination between the powder formulation and the salt solution should be possible. Yet the solution should have water vapor contact with the formulation for desiccant purposes. All prior art desiccant packs based on saturated salt solutions are based on permeable membranes, microporous laminates or water repellent fabrics such as Gore- Tex and Tyvek. These kinds of materials are not considered suitable for the present purpose due to the safety issue. They are difficult to seal tightly by any means to a container made of any material, they are thin and vulnerable to mechanical stress, and they the may leak when damaged or due to differences in pressure, for example during flights. A safe solution is disclosed in US 6,132,394, where the desiccant container is made of a single plastic material without any membranes. Such a container of proper plastic can be safely sealed by melting or by ultrasonic or laser welding and 100 % tightness test can be carried out economically by modern automation technology.

The results (RH and humidity absorption as a function of time) of the tests described below are graphically presented in the accompanying drawings wherein the used desiccants are:

Figure 1 saturated solution of magnesium chloride hexahydrate Figure 2 silicagel Figure 3a moist magnesium chloride hexahydrate Figure 3b moist potassium acetate Figure 4a dry magnesium chloride hexahydrate Figure 4b dry potassium acetate Figure 5 dry and anhydrous magnesium chloride and Figure 6 : calculated moisture regulation using silicagel and dry, anhydrous magnesium chloride under test circumstances

I n the first tests, the aim was to study qualitatively the absorption kinetics of magnesium chloride hexahydrate as a saturated salt solution. The solution with some undissolved crystals (2.5 g) was placed in a silicone rubber pouch and the pouch was sealed with silicone. The pouch was placed in a 60 ml polycarbonate bottle provided with a humidity sensor. The water permeabilities of the pouch and the bottle were measured as 1300 and 104 ng/(min-%RH), respectively. The closed bottle was placed in a closed desiccator containing saturated sodium chloride solution at the bottom inducing 75.3% RH as fixed humidity. The relative humidity and mass increase (absorbed water) were measured for 200 days. As shown in Figure 1 , the relative humidity inside the bottle is 35 - 36% RH for140 days and after that starts to increase due to the dissolution of the crystals (ca. 2.5 g salt / 0.85 g H₂O). The absorption capacity is thus 34 w-%. However, the non-saturated solution still acts as a desiccant and the mass increase is quite linear during the monitoring. When extrapolated linearly, it can be approximated that it would take at least 300 days to reach 50% RH and 500 - 600 days to reach the ambient humidity (75.3%).

When a similar test was carried out using silica gel as the desiccant, the results were different, as can be seen in Figure 2. The increase in mass indicates that 2.3 g silica gel was capable of absorbing 500 mg at 75.3% RH, and the absorption capacity was 22 w-%. According to the theory, the relative humidity inside the bottle increases regularly until the absorption capac- ity is exhausted and the environmental humidity reached.

The following tests were carried out using the drug reservoirs of a commercial MDPI (Taifun^®) according to US 6,132,394. The reservoirs, reservoir lids and empty desiccant containers and lids were obtained from LAB Pharma Oy, Turku, Finland. The important features of the Taifun drug reservoir are that the desiccant container is made of injection moulded polycarbonate and located inside the reservoir, the reservoir is made of polypropylene, and there are compliant seals around the round dose pin which slides through the seals and transfers the dose into a dose slot for inhalation. The desiccant container is relatively permeable to moisture but the reservoir is not. When the moisture ingress is very slow, a saturated salt solution in a polycarbonate con- tainer may be able to regulate the internal humidity within a narrow RH- range for a long time.

Magnesium chloride hexahydrate (J.T. Baker), potassium acetate (J.T. Baker) and anhydrous magnesium chloride (J.T. Baker) were the model salts. Ca. 300 mg of the salts were enclosed in the Taifun polycarbonate desiccant containers, the polycarbonate lid being glued with Araldite Standard. The total volume of the container was 1 ml and the wall thickness 0.5 mm. Dry and moist magnesium chloride hexahydrate and potassium acetate were used to find out the possible differences. Dry anhydrous magnesium chloride was also used to find out its performance. The desiccant containers were placed in the Taifun polypropene drug reservoirs provided with calibrated humidity sensors (SHT 15, Sensirion AG, Switzerland; the specified accuracy being ± 2 % RH). The sensor pins were fixed on the reservoir walls through small holes, glued with TiO₂ doped Araldite Standard. The reservoirs were closed with a polypropylene lid as in a normal assembly. The two holes for the dose pin were closed with metal knobs to ensure minimal moisture penetration. The drug reservoirs were then placed in a desiccator containing saturated potassium chloride solution for generating the storage conditions of 84.2% RH at room temperature. The increase of mass (absorbed water) in the drug reservoirs and the sensor indicating the relative humidity inside the drug reservoirs were monitored for several months. Moist and dry magnesium chloride hexahydrate and moist and dry potassium acetate were also tested in conditions where the external humidity was decreased from 84.2 to 5% RH by placing the reservoirs into a desiccator containing dried silica gel for six days. The results are shown in Figures 3, 4, and 5. EVALUATI ON OF TH E RESULTS

Exam ple 1

Figure 3a. Adding 60 mg of extra water certainly produces a saturated salt solution inside the desiccant container. I nside the drug reservoir, the humidity changes from an initial value of 27% to 35% within one day, to 37% within 51 days and to 37% RH within 1 12 days. The corresponding values for mass increase (water absorption) were 0.7, 6.3 and 12.7 mg, respectively. The increase in mass is 0.109 mg/day and is presumed to be stable until all of the salt is dissolved. The constant humidity inside the desiccant container must be ca. 33% RH while the ambient humidity around the drug reservoir is very close to 84.2% RH. However, the humidity inside the drug reservoir is very stable during the observation time and seems to settle at ca. 37% RH. This is only ca. 4% RH higher than the generated humidity of saturated magnesium chloride hexahydrate solution.

As the solubility of magnesium chloride hexahydrate is 302 g / 100 cm³, the capacity of the used desiccant can be calculated. I n this case, the exact amount of the salt was 308 mg with 60 mg of extra water. The salt is able to absorb 102 mg of water, of which 60 mg was used and 42 mg of the capacity remained. With an increase in mass of 0.109 mg/day, the solution remains saturated for more than one year and regulates the humidity inside the drug reservoir to ca. 37% RH. It is noteworthy, that the capacity can be increased by decreasing the amount of the extra water, and thus the time for regulating the humidity may be prolonged.

Figure 3b. Moisturized potassium acetate behaves in a very similar manner. However, as the humidity generated by the saturated salt solution is lower (22.5% RH), the corresponding equilibrium humidity inside the drug reservoir is lower and settles at ca. 30% RH. Exam ple 2

Figure 4a. The test was carried out for two purposes: to prove that dry magnesium chloride hexahydrate (with higher capacity) behaves like the moistur- ized salt, and also to show the influence of low ambient humidity on the humidity equilibrium inside the drug reservoir. No clear differences can be seen when comparing the results in Figures 3a and 4a with each other. However, the total capacity is higher because no extra water was added. After storage for 50 days at 84.5% RH, the reservoir was transferred to 5% RH of ambient humidity for six days. The internal drug reservoir humidity decreased from 39% RH to 35% RH within two days and to 33% RH in six days. When returned to 84.5% RH, the internal humidity increased to 38% RH in one day and to 39% RH in three days. Accordingly, the change in mass followed the humidity changes. As according to the theory, the saturated solution was able to release water through the walls of the desiccant container into the drug reservoir and further to the ambient air through the reservoir walls. The most important information is that in utmost humidity, the change in the internal humidity is minimal and in this case only ca. 7% RH.

It is an important finding, that dry magnesium hexahydrate can be used when filling the desiccant container. Obviously, the very small amount of water absorbed by the salt during handling and filling is capable of initiating the fixed point humidity control. The salt becomes moist and liquid later on inside the air-tight container when a considerable amount of water is ab- sorbed.

Figure 4 b. The results correspond to those of dry magnesium chloride hexahydrate (Figure 4a). Dry potassium acetate is, therefore, able to regulate the internal humidity between 20% RH and 25% RH in the range of 5% RH to 84.5% RH of ambient humidity. Exam ple 3

Figure 5. When dry anhydrous MgQ₂ was enclosed in the desiccant container and the drug reservoir was provided with the lid, the relative humidity inside the reservoir decreased from 40% RH to 14% RH within one day and settled at ca. 10% RH within a few days for the monitoring period of 34 days. The salt is obviously very susceptible to absorbing the crystalline water (6 molecules). The amount of absorbed water was 5.00 mg and water permeation 0.100 mg/day when considering the linear part of the mass increase curve.

When magnesium chloride hexahydrate contains 53 w-% of crystalline water, the amount of anhydrous salt (303 mg) used is capable of absorbing 344 mg of crystalline water. This means that it takes more than 9 years to turn the anhydrate form into the hexahydrate form. After that, the salt will act as a saturated solution and the permeating water will dissolve the crystals of MgQ₂. The behavior of that phase is depicted in Figure 4a.

Exam ple 4

Figure 6. The theoretical calculations are presented in Figure 6 to clarify the difference in the desiccant behaviour of the dry anhydrous salt and the dry silica gel. The calculations have been made for 300 mg of dry anhydrous magnesium chloride and for 300 mg of dry silica gel. For silica gel, it is supposed that water permeation into the desiccant is 1.67 ng/(min-%RH). The results show unambiguously the overwhelming superiority of magnesium chloride over silica gel.

The invention gives new opportunities for controlling the humidity of powder formulations for inhalation. The superior water absorption capacity makes it possible to control the humidity surrounding the powder formulation for a very long period of time. I n most cases, it is beneficial that the variation in internal humidity is very slight and almost independent of ambient humidity. Hence, even hygroscopic drug formulations can be stored and used in a powder inhaler provided with this invention, which minimizes physical changes and agglomeration in the drug formulation caused by humidity.

The following special case is further disclosed. Several polypeptide/protein drugs, such as insulin, are being developed for systemic delivery via the lungs. Most of them are prone for chemical degradation as water solutions and nebulization as a spray are difficult or impossible. I nhalation of powder is preferred. However, many such drugs are extremely sensitive to moisture, and chemical degradation may take place in normal atmosphere. If the main factor is the atmospheric humidity, this can be controlled by means of the present invention.

A modern approach for protecting polypeptides from chemical and physical changes is to hide the molecules in a protective "glassy" matrix. These well- known matrices may consist of sugars or polyalcohols such as mannitol and trehalose, for instance. When spray-dried with the polypeptide drug, the additives may form amorphous (glassy) particles of suitable size for inhalation. Changes in the active configuration or even chemical degradation are prevented when the drug molecules are in the rigid matrix. However, because the glassy matrix is amorphous, it is thermodynamically unstable and tends to recrystallize. If this occurs, the system is disturbed and the physical/chemical stability of the formulation is questionable. This recrystallization depends on the glass transition temperature (T₉), at which the mobility of the molecules increases so that recrystallization can take place. This phenomenon is dependent on ambient humidity, so that increased humidity decreases the Tg according to a certain (Gordon-Taylor) equation. Therefore, increasing relative humidity increases the rate of physical changes in this kind of formulations. At room temperature, the T₉ temperature for trehalose is 79⁰C in dry conditions, but much lower at higher levels of humidity. Trehalose- based glassy formulations must, therefore, be protected from high humidity.

Also glassy lactose and glucose are sensitive to moisture. Mannitol has a low Tg temperature and it is easily crystallized during processing and hence, it may produce particles which are more resistant to moisture.

It is necessary that the permeation of water vapor into the desiccant con- tainer is greater than the permeation of water vapor into the space surrounding the desiccant container. It is not possible to specify the permeation difference precisely because it depends on the construction and material choices of the system. However, the difference is reflected in the equilibrium humidity outside the desiccant container. I n test 1 , permeation through the desiccant container walls was 12.5 times greater than permeation through the plastic bottle walls. The equilibrium humidity inside the bottle was 35- 36% RH or 3-4% higher than the fixed point humidity of the desiccant (magnesium chloride). I n this case, ambient humidity was 75.3%.

I n EΞxamples 1 -2, ambient humidity was 84.2% and the desiccant container and the drug reservoir were injection-moulded parts of a commercial powder inhaler. The equilibrium humidity was quite close to the fixed point humidity of the desiccants (magnesium chloride and potassium acetate) and the permeation differences were within a proper range.

From a practical point of view, the equilibrium humidity around the inhalation powder should not exceed 50% RH at which most powder formulations remain stable. It is estimated, that when magnesium chloride or potassium acetate are used as desiccants, permeation into the desiccant container should be at least five times greater than permeation into the space outside (powder space) surrounding the desiccant container. Finally, the sensitivity of the powder formulation to moisture is decisive for the selection of proper desiccants, constructions and construction materials.

It has been shown that the desiccant container for the purpose of this inven- tion can be made of solid plastic and no membranes or porous materials are necessary. This finding eliminates the risk of contamination between the inhalation powder and the desiccant material.

It is surprising, that a liquid desiccant can be used to control the humidity of an inhalation powder and yet have a more precise control of humidity and higher capacity compared with the conventional desiccant silica gel.

The preferred deliquescent salts are safe and have been approved for oral and/or intravenous use for humans. The salts are low-priced and they can be handled and filled in normal atmosphere without becoming excessively moist. Some absorption of moisture is beneficial to initiate the desiccant performance as saturated solution.

From a practical point of view, it is very important that the salts do not need to be in the form of saturated solutions during packing, but that dry salts can be used.

If maximum capacity is required, an anhydrous salt may be used, but then the initial humidity may be very low, like with anhydrous magnesium chlo- ride, acting as a chemical absorbent in the beginning.

If very stable humidity is required, good choices are magnesium chloride hexahydrate, calcium chloride hexahydrate or potassium acetate, which is anhydrous but not a chemical absorbent. Finally, the desiccant container may be placed in or outside the drug chamber, or outside the capsule or blister, if the prerequisites for construction and materials are met.

Claims

Clai m s

1. A desiccant system for controlling the humidity of inhalation powder, said system comprising an air-tight container containing a dry desiccant, and a drug chamber containing the inhalation powder, the air-tight container being arranged inside the drug chamber or in its vicinity, said desiccant being capable of maintaining a fixed point of humidity as a saturated solution of at least one salt.

2. A desiccant system according to claim 1 , wherein the air-tight container for the desiccant is permeable to water vapor.

3. A desiccant system according to claim 2, wherein the water vapor permeability of the air-tight container is at least five times greater than that of the wall(s) separating the air tight container from ambient air or outer atmosphere.

4. A desiccant system according to any of the preceding claims, wherein the metal is an alkali metal or alkali earth metal.

5. A desiccant system according to any of the preceding claims, wherein the metal is selected from Li, Na, K, Cs, Mg and Ca.

6. A desiccant system according to any of the preceding claims, wherein the salt is an acetate, halide, sulphate or carbonate.

7. A desiccant system according to any of the preceding claims, wherein the desiccant is MgQ₂.

8. A desiccant system according to claim 7, wherein the desiccant is MgQ₂ hexahydrate.

9. An inhalation apparatus with a tightly sealed drug chamber including a desiccant system for an inhalation apparatus, said system comprising an airtight container containing a desiccant, the air-tight container being arranged inside the drug chamber or in its vicinity, said desiccant being able to maintain a fixed point of humidity as a saturated solution of at least one salt.

10. An inhalation apparatus as set forth in claim 9, characterized in that the desiccant is an alkali metal or alkali earth metal.

11. An inhalation apparatus as set forth in claim 10, characterized in that the metal is selected from Li, Na, K, Cs, Mg and Ca.

12. An inhalation apparatus as set forth in any of the preceding claims, characterized in that the salt is an acetate, halide, sulphate or carbonate.

13. An inhalation apparatus as set forth in claim 12, characterized in that the desiccant is MgQ₂.

14. An inhalation apparatus as set forth in claim 13, characterized in that MgQ₂ is in form of its hexahydrate salt.

15. A method for the regulation of moisture in an inhalation apparatus with a tightly sealed drug chamber including a desiccant system for an inhalation apparatus, said system comprising an air-tight container containing a desiccant, the air-tight container being arranged inside the drug chamber or in its vicinity, said desiccant being capable of maintaining a fixed point of humidity as a saturated solution of at least one salt.

16. The use of a desiccant for controlling the humidity of inhalation powder, said desiccant being dry when placed in the container but able to maintain a fixed point of humidity as a saturated solution of at least one salt.