University of Groningen
Resistance is Futile
Sibum, Imco
DOI:
10.33612/diss.143830475
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Sibum, I. (2020). Resistance is Futile: The targeted delivery of antibiotics to the lungs. University of Groningen. https://doi.org/10.33612/diss.143830475
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
7
GENERAL DISCUSSION &
FUTURE PERSPECTIVES
INTRODUCTION
The pulmonary tract is liable to infections, which might be bacterial, fungal, parasitic, or viral in origin [1,2]. In fact, as I am writing this the world has ground to a standstill as a result of the severe acute respiratory syndrome coronavirus 2 (CoV-2) pandemic. SARS-CoV-2 is colloquially known as the coronavirus. However, it is important to note that while we do not know how this pandemic will progress, the number one deadliest infectious disease worldwide is still tuberculosis (TB), which is bacterial in origin [3]. As of 2018 a quarter of the world’s population is latently infected by it. In that same year 10.0 million people developed the active disease and 1.5 million people died because of it [3].
To compound on this problem, pathogens resistant to treatment have increased in incidence, also for TB. As a result, in some Eastern European countries over half of the patients developing TB are resistant to the first line treatment of isoniazid and rifampicin [4]. They have developed multi-drug resistant TB (MDR-TB). To counter resistance, higher concentrations of antibiotics are required at the site of infection. However, the maximum dose that can be administered systemically (orally or parenterally) for antibiotics is usually determined by their safety profile. Increasing the dose above this is untenable, as more severe side effects can be expected. Local targeting may be the answer to this [5–9]. If done well, it allows for an increase in drug concentration at the site of infection but without an increase in systemic exposure and the associated higher risk of severe side effects. Since TB is predominantly a pulmonary disease [5], delivery of antibiotics via the pulmonary route seems logical.
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
Pulmonary delivery can be divided in low and high doses. To facilitate discussion, we have defined high doses as inhaled drug doses larger than 2.5 mg [10]. We have established this definition on the fact that the often-used adhesive mixture formulation technique, developed for low doses, cannot be reliably used for drug doses above 2.5 mg. This is based on currently marketed dry powder inhalers (DPIs), which usually can contain around 10 to 25 mg of powder, and on the maximum drug content that can be formulated in adhesive mixtures, which is between 5 and 10% [11]. This definition of high doses is based on DPIs as we deemed other devices unsuitable for the delivery of antibiotics, or other high dose drugs for that matter [10]. However, evidenced by the lack of high dose dry powder inhaler products on the market, it is a challenge to formulate them.
Powders for inhalation have to meet certain criteria to be suitable for pulmonary administration. First of all, the aerodynamic particle size distribution of the powders should be in the range of 1 - 5 μm, preferably between 1 – 3 μm, to be able to realize central and deep lung deposition [12]. Powders should be easily dispersed in an aerosol, preferably in the first 0.5 to 1.5 liters of air during inhalation, and retention of powder in the inhaler should be low. It should do all these things in a consistent manner. Lastly, powders should be physically and chemically stable. Suitable powders should be married with suitable DPIs. In standard practice, a powder is formulated to allow it to function with an already existing DPI. However, more efficient products can be made when inhaler and formulation are developed in conjunction. This minimizes the need for excipients, which lowers the powder bulk needed to be dispersed which lowers the powder burden on the patient [10].
FORMULATING ISONIAZID
To satisfy these requirements can be a challenge, as is apparent in the chapters describing the formulation of isoniazid. Milling, the cornerstone of particle preparation techniques, resulted in blockages in the Twincer® DPI even when channels were widened [13]. Jet-milled isoniazid had an unexpected morphology, which was deemed the cause of the poor results. The unexpected morphology is probably caused by local increases in temperature during milling, which sublimates part of the isoniazid particles [13,14]. Sublimated isoniazid deposits and fuses particles together when the temperature decreases again. This was corroborated when after the addition of a lubricant, which should decrease the local increases in temperature, the expected morphology was seen after milling. Unfortunately, the lubricant is not a suitable excipient for pulmonary administration.
It would be interesting to determine if the phenomenon described is more wide spread, and other compounds with a relative low sublimation temperature show a deviating morphology after milling as well. Further research is needed to determine if this phenomenon can be prevented by optimization of the milling process, or if the addition of a second mill step can break up the fused particles, resulting in a usable formulation. This may be aided by an experimental setup which is able to show, and maybe even to quantify, the local increases in temperature during the milling process. However, such a setup is not known to exist yet.
Spray-drying of isoniazid without excipients also does not result in the correct particle size distribution for inhalation [13]. The primary particle size distribution obtained after spray-drying was
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
too coarse. Bizarre enough, the particles were even larger than the droplets generated by the spray-driers’ nozzle. This was likely caused by crystallization induced particle fusion after powder collection. When the particles were measured in the air stream they were small enough to be inhaled. Crystallization was unavoidable with a Tg of -3.99 ± 0.18 °C [13]. Furthermore, particle engineering was severely hampered by the fact that isoniazid starts to sublimate at 80 °C. Isoniazid particles could thus never be exposed to these or higher temperatures.
It was hypothesized that creating a physical barrier between isoniazid particles would allow them to crystalize while preventing the fusion issue [13]. This physical barrier can be formed by coating the individual particles with a compound, such as L-leucine, which will enrich at the surface of the droplet during the spray-drying process [21]. The addition of 5% L-leucine (w/w) prevented fusion between particles, and the primary particle size distribution was suitable for inhalation [13].
Coating of particles during the spray-drying process has been done before. However, this was always performed for two reasons. Either to provide protection against moisture to the core of the particle, or to increase the dispersability of the powder. Indeed, the isoniazid with 5% L-leucine formulation dispersed well from the Twincer®. Furthermore, it allowed for a nominal dose of 50 mg, where only 25 mg of the milled isoniazid could fit in the blister cartridge. However, the main reason for coating the particles was to prevent particle fusion due to crystallization, and thus to obtain a suitable primary particle size distribution for inhalation. To our knowledge this is the first time a coating has been applied for this reason. It is likely that this approach can also be used for other compounds which have a Tg below room temperature and are
liable to crystallization and particle fusion. Furthermore, crystallized powders are in general more stable than their amorphous counterparts. Spray-dried powders are often amorphous but if an increase in stability is required a possible approach would be to induce crystallization of the powder, and use a coating to prevent particle fusion. It was thus a surprised when stability of the isoniazid particles coated with 5% L-leucine was poor. When exposed to moisture particles again fused together, likely caused by dissolution-crystallization [15].
By improving the coating, the stability may be improved by either delaying fusion or by preventing fusion at higher humidities. L-leucine has been known to provide moisture protection [16,17]. Improvement of the coating may be achieved in several ways. By optimizing the amount of excipient used, by optimizing the spray-drying parameters, or by optimizing the used excipient itself [18]. For high dose drugs minimization of the excipient content is paramount. So, while an increase in content might be beneficial from a stability point of view, optimization was primarily focused on decreasing it. Changing the spray-drying parameters seemed to have a considerable impact on the coating, with suitable and reproducible particle size distributions only being found at inlet temperatures of 120 °C and above. However, no spray-drying conditions were found that resulted in sufficient stability. Again, the fact that isoniazid sublimates at 80 °C severely limited the spray-drying conditions that could be tested [13].
Changing the excipient to trileucine resulted in a substantial increase in stability [18]. Powder could now be stored at 75% RH for at least 3 months without any signs of dissolution-crystallization being evident. Trileucine consists of three L-leucines connected with peptide
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
bonds and has a higher molecular weight (358 versus 131 g/mol) and a lower solubility (8 mg/mL versus 22 mg/mL) [19,20]. Both the higher molecular weight and the lower solubility improve the coating efficiency. However, not only an improvement in coating efficiency explains the increase in stability. Samples of trileucine spray-dried under conditions that resulted in lower leucine to isoniazid surface ratios than L-leucine still had an improvement in stability [18]. A difference in solid state may explain these results. L-leucine is known to separate to the solid phase as a crystal during spray-drying, while trileucine separates into the amorphous form [20]. How this fact exactly translates to the results described remains unclear and further research is required to determine the effect of the solid state of the coating on the moisture protection and stability enhancement it provides. However, with the addition of 3% trileucine to isoniazid (w/w), a suitable and stable isoniazid formulation was made.
This stable isoniazid formulation can be well dispersed by the Twincer® DPI [18]. However, retention of the powder by the inhaler was high and inconsistent. A different inhaler design, more suited to the physicochemical properties of the isoniazid trileucine formulation, lowered inhaler retention and improved consistency. This inhaler is called the Cyclops®. Both the Twincer® and the Cyclops® use air classifier technology as their dispersion mechanism [6,21]. In an air classifier, particles are collided against the classifier wall, which generates inertial impaction forces. These inertial forces break up agglomerates and disperses the powder. The Cyclops® lowers the impact angle between the particles and the classifier wall, lowering the inertial forces generated during dispersion [6]. The classifier also gains 8 bypass channels, for a total of 11 air channels. These extra air channels form a sheet of air
against the classifier wall, slowing small particles down and reducing inertial forces further. Only larger agglomerates have enough inertia to break through this sheet of air and impact against the classifier wall [22]. As a result from these reductions, deposits of powder on the classifier walls are less likely to form. Furthermore, the addition of bypass channels lowers the total surface area of the classifier, lowering the places where powder deposits can form and thus lowering retention further. However, as a result of the decrease in inertial forces generated, dispersion power of the inhaler decreases. But, in front of the bypass channels shear forces are generated as a result from impinging airflows [10,22]. Shear forces may be better suited for high dose cohesive powders, and may improve dispersion. All these changes put together mean that for the isoniazid 3% trileucine formulation dispersion was similar while retention was decreased and more consistent [18].
To determine if shear forces are better suited for dispersion of high dose cohesive powders it may be interesting to develop an experimental inhaler which only generates shear forces. This would allow for direct comparison between an inhaler which generates only inertial impaction forces (Twincer®), an inhaler which generates only shear forces, and an inhaler which generates both (Cyclops®).
Both the Twincer® and Cyclops® use a new single dose cartridge (SDC) as their dose compartment. In this SDC 50 mg of the isoniazid with 5% L-leucine formulation could fit. This increased to 100 mg for the isoniazid with 3% trileucine formulation. The Cyclops® is capable of dispersing this nominal dose, but with reduced efficiency. A nominal dose of 80 mg was found to be optimum [18].
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
Determination of powder properties other than the primary particle size distribution and the dispersibility from an inhaler might be interesting in the future. Such as the density and flowability of the powder and the surface energy of the particles. For example, the bulk density and flowability will likely explain why 25 mg of milled isoniazid, 50 mg of isoniazid spray-dried with L-leucine and 100 mg when spray-dried with trileucine could fit in an SDC and be emptied. Determining the surface energy of the particles may give an indication of their dispersibility. Especially when coupled with Time-of-Flight secondary ion mass spectrometry (TOF-SIMS) data about the coating, this would add to the mechanistic understanding of the isoniazid formulation described here, and would help in the rational design of new formulations.
Further optimization may be achieved by increasing the fraction < 5 μm, which was around 90% for the isoniazid 3% trileucine formulation. A higher fraction < 5 μm may result in a higher fine particle fraction (FPF) when dispersed from an inhaler. Smaller particles may be formed during spray-drying by decreasing the droplet size generated by the nozzle. This may be achieved by increasing the atomizing airflow or by decreasing the feed rate of the solution. The feed rate was already low and it is unlikely that smaller droplets would be generated by decreasing it even further. The atomizing airflow could not be increased further, it resulted in an unstable airflow through the nozzle and larger particles. If the droplets cannot be made smaller another possibility is to decrease the total concentration of solutes in the feed solution. As the amount of solute per droplet is lower smaller particles will be formed. However, this would likely decrease the coating efficiency [19]. A nozzle which generates smaller droplets with the same atomizing airflow may be able to increase the fraction < 5 μm.
A nozzle capable of this has already been designed, but has not been constructed yet at the moment of writing.
However, based on the results described it can be concluded that trileucine as a coating material is superior compared to L-leucine. It provides substantially more moisture protection, it improved dispersibility and, likely a result from a higher bulk density and flowability, the nominal dose could be doubled. Applying a coating of trileucine may be of interest for other high dose drugs as well. Such as for drugs suffering from high hygroscopicity, drugs which are liable to crystallization and fusion, or drugs which are poorly stable. Furthermore, drugs which show poor dispersability as a result of the cohesiveness encountered with small particles may benefit as well.
AUTOMATIC FILLING AND ITS EFFECT ON THE
FORMULATION
In literature new products are often described as the dispersion performance of a formulation and its inhaler. The suitability of the formulation and inhaler for automatic filling, which is necessary to obtain the production numbers required, is often ignored.
To illustrate, the standard treatment for drug susceptible TB recommends a daily oral administration of 300 mg isoniazid for 6 months [23,24]. If we assume that the same dose is wanted via the pulmonary route, and that the delivered dose of an inhaler is 50 mg, 1098 inhalers would be necessary for the 6-month treatment for one single person. Given that in 2018 10.0 million people developed active TB it would require the yearly production of 11.0 billion inhalers if everyone would receive this treatment. An amount that can hardly be produced by
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
hand. This of course neglects the higher efficiency expected of targeted delivery.
One of the possible reasons why little research has been performed on automatic filling is because it does not qualify as fundamental research. Coming to a new formulation and inhaler requires fundamental research, as is discussed above, but also a healthy dose of applied science. It is the combination of the two that results in a successful product. During the formulation of isoniazid challenges were encountered, described, and overcome by researching fundamentally what were the phenomena playing a role in the difficulties and why. The solutions found for these problems are not only applicable to isoniazid, but for a multitude of other drugs with similar dose requirements and physicochemical properties as well. However, if these solutions cannot be used in practice they are of little value except for scientific curiosity. Only the step from fundamental to applied science, which includes the determination if a model formulation using the solutions described is suitable for automatic filling, may tell us if they are more than a scientific curiosity and if they may be used in real life applications.
Automatic filling can have a substantial impact on the performance of an inhaler-formulation product. Furthermore, the formulation likely has a substantial impact on the functioning of the automatic filling machine. Several automatic filling techniques are described in literature. However, almost all of them are designed to be used with low dose drugs formulated in adhesive mixtures. To our knowledge, automatic filling of high dose dry powders in inhalers is not described in literature at all. A vacuum drum filler does not require adhesive mixtures, and thus may be capable of administering high
dose drugs. Indeed, pure micronized colistimethate sodium could be consistently and reproducibly dosed [25]. However, a vacuum drum filler works by forming pellets of a defined volume. As a result of the formed pellets during dosing the Twincer® was again prone to blockages in the already wider powder channel. Furthermore, some pellets were retained by the blister cartridge. A redesign of the Twincer® was needed [25]. The redesign got an even wider powder channel, and the classifiers were changed with an increase in size of the classifier inlets and with improved circulation in the classifiers themselves. The SDC was introduced, which features improved air flow, eliminating pellet retention in the dose compartment. With this improved Twincer® pellets could be successfully dispersed with low inhaler retention [25]. The Cyclops® got a similar treatment. This emphasizes the need to keep automatic filling in mind when designing an inhaler.
The used vacuum drum filler could also reliably and consistently dose the isoniazid with 3% trileucine formulation over a wide range of product pressures, ranging from 200 to 600 mbar [26]. A spray-dried amikacin formulation, also an antibiotic used in TB treatment, containing 1% L-leucine could also be dosed reliably and consistently over the same wide range of product pressures, but a small adaptation to the vacuum drum filler was needed.
The formation of pellets had no effect on the dispersion of the isoniazid formulation. It did lower the inhaler retention, this effect being most pronounced in the product pressure range of 400 to 600 mbar. Furthermore, while for loose powder the optimal nominal dose was 80 mg, with pellets a nominal dose of 105 mg seemed superior. More than 105 mg could not physically fit in the SDC, so a wider SDC, named
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
the L-SDC, was made. In the L-SDC 150 mg of the automatically filled isoniazid formulation could fit. Of this nominal dose 107.35 ± 13.52 mg was dispersed by the inhaler in the right particle size distribution. At 4 kPa, the average pressure drop people can attain, this dose was dispersed in the first 0.39 liters of air [26].
Amikacin pellets were also well dispersed by the Cyclops® inhaler, with again the optimum product pressure range found between 400 to 600 mbar. Also here an improvement in retention was seen with pellets. Bulk dispersion of the 57 mg nominal dose took place during the first 0.6 liters of air for the 400 mbar pellets and in the first 0.9 liter of air for the 600 mbar pellets. Quick enough to be able to realize deep lung deposition [26].
These data have shown that while there are significant challenges in formulating a high dose antibiotic for pulmonary administration, they can be overcome by developing the formulation and its inhaler in unison, while keeping the automatic filling technique to be used in mind. This approach allows for the DPI design to be optimized to the properties of the formulation. By use of spray-drying and particle engineering technologies, coupled with an inhaler designed to handle the physicochemical properties of the drug in question as efficient as possible, an isoniazid and amikacin product was developed. The isoniazid formulation is spray-dried with 3% trileucine, can be exposed to 75% RH for three months without deteriorating, can be automatically filled, and can be quickly and efficiently dispersed by the Cyclops® inhaler with nominal doses of 150 mg. The amikacin formulation is spray-dried with 1% L-leucine, can also be automatically filled, and is well dispersed by the Cyclops®, with a maximum nominal dose of 57 mg being tested.
A stability study on the amikacin product is ongoing, with stability up to 27 months shown so far.
FUTURE PERSPECTIVES
Regulatory approval for a clinical trial with the amikacin product is obtained and the product is produced, and released, under GMP conditions. The trial itself has just started. The trial is a single center, active control, ascending dose response study with 8 patients with drug-susceptible TB, testing doses of 400, 700, and 1000 mg administered with several inhalers containing nominal doses of 50 mg.
A clinical trial with the isoniazid formulation is unlikely in the short term. No safety data is available on the pulmonary administration of trileucine, an essential component to the formulation. A toxicology assessment of trileucine as an excipient for inhalation was performed by the Toxicology Knowledge Team Sweden. They determined that local toxicity in the lungs was unlikely with a daily dose of 6 mg, which translates to a dose of 200 mg of the isoniazid formulation. This was based on multiple facts, chief among them the fact that other clinical trials have been conducted with trileucine as excipient for pulmonary administration [27,28]. Furthermore, it was deemed appropriate to use read-across data of L-leucine, which itself has been considered to have a low risk for local lung toxicity. Systemic exposure to trileucine is no issue. Trileucine is found in casein and soy protein and is absorbed at least partly as a tripeptide by the intestine. Systemic exposure to trileucine is thus likely naturally occurring as part of a normal diet. They concluded that a systemic long-term toxicology study is likely not needed but that it can be expected that authorities will request a short-term study on an animal model before a clinical trial with the isoniazid formulation is
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
approved. However, based on the superiority of trileucine over L-leucine in the isoniazid formulation, which also may be translated to other drugs, it would be prudent to get trileucine approved as an excipient for pulmonary administration.
An often-overlooked aspect of pulmonary administration is the difficulty of measuring the exposure of the patient to the drug at the target site. For the amikacin clinical trial systemic exposure (i.e. serum concentration) will be measured as an indication for lung exposure. However, as systemic exposure is likely lower after pulmonary administration, measuring the low concentrations can be challenging. For the trial an amikacin serum immunoassay was optimized for lower concentrations by increasing the sample volume and decreasing the reagent volumes. However, more direct methods allow for a more localized description of the deposition pattern in the lung and the resulting concentrations. They can be noninvasive, such as external imaging with hyperpolarized noble gases. But resolution is often inadequate with these techniques [29]. Quantification of local drug deposition via gamma scintigraphy, single-photon-emission computed tomography, and positron-emission tomography is also subject to challenges. The major challenges are poor image resolution, fast radiotracer clearance, the difficult and time consuming process of labeling the drug without changing the morphology and size of the powder particles, and the expertise and expensive hardware needed [30,31]. Data obtained via bronchoalveolar lavage is more useful in that regard. During bronchoalveolar lavage a bronchoscope is lowered into the lungs via the mouth or nose [32]. This is quite invasive and it is unlikely that this could be standard practice during clinical trials. Lung microdialysis suffers from the same issue [33,34]. It requires surgery
to place the microdialysis probe. Its use is thus limited as a result of ethical objections. Exhaled breath condensate, which is collected saliva and lung lining fluid from exhaled air, has been hypothesized to allow for determination of drug concentrations in the lungs. However, high variability is encountered and its use for pharmacokinetics discouraged [35]. Furthermore, it does not allow for localized description of drug deposition as it is likely the exhaled saliva contains the drug as well. A technique not subject to these disadvantages would be a welcome addition.
The Twincer® and Cyclops® are made out of polycarbonate, which has been standard for inhalers in the last few decades. However, with the advent of new polymers, it would be interesting to determine their effect on the retention. Two forces are dominant in the adhesion of powder particles to the inhaler. These are Van der Waals forces and electrostatic forces [10]. Different polymers results in different Van der Waals forces between the particles and the inhaler [36,37]. Furthermore, polymers may be antistatic, reducing the electrostatic forces and thus the electrostatic attraction present. A small experiment with Cyclopses made out of Novodur and the anti-static Permastat polymers did not show any substantial improvement for the isoniazid and amikacin formulations (Figure 7.1). However, for these two formulations retention seems to be most predominant in the SDC, which was kept the same and is made out of low-density polyethylene (LDPE). Making the SDC out of other polymers may decrease the already low retention further for isoniazid and amikacin. However, the used polymer should not only result in a lower retention compared to LDPE. It should be as non-reactive as LDPE for stability reasons and it should also be weldable to allow it to be sealed with a lidding foil after filling. It is also still interesting to
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
test other polymers for the inhalers itself. While the two tested polymers did not have an effect on isoniazid and amikacin, other polymers might result in an improvement. Furthermore, the two tested polymers may have an effect on other formulations. Especially powders which are statically charged, something that is often introduced during jet milling, may benefit. Particles may also be charged by the airstream during inhalation and by the repeated particle to particle and particle to wall contacts [10]. Particles prone to this may also benefit from anti-static inhalers.
2 kPa 4 kPa 6 kPa
0 10 20 30
Pressure drop (kPa)
R et en tio n (% )
2 kPa 4 kPa 6 kPa
0 10 20 30
Pressure drop (kPa)
R et en tio n (% ) PolycarbonateNovodur Permastat A B
Figure 7.1 - Retention found for the isoniazid 400 mbar pellets (A) and the amikacin
400 mbar pellets (B) in the Cyclops® made of three different polymers at a pressure drop of 2, 4, and 6 kPa (average ± SD, n = 5). Retention determined gravimetrically during inhaler dispersion analysis with a HELOS diffraction unit equipped with an INHALER2000 disperser.
Larger SDCs may allow for the administration of higher fine particle doses (FPD). The SDC used in the Twincer® and Cyclops® has an internal volume of 350 mm3. A L-SDC has been tested with an internal volume of 520 mm3, which resulted in a substantial higher FPD for isoniazid. SDCs with an internal volume of 660 mm3 are possible. It would be interesting to determine if a further increase in internal volume, and subsequently a higher nominal dose, would increase the FPD further. Furthermore, an SDC which would release its dose in a more
gradual manner may increase the dispersion efficiency of the inhaler. As was encountered during the testing of the L-SDC, where during one of the measurements the dose was released gradually on accident, the FPF increased with 15%. An SDC which does this in a consistent manner and within the first 0.5 to 1.5 liters of air is thus of interest.
Improving the modeling for the Peclet number, the number which tells if a compound at a specific evaporation rate will enrich at the surface or in the center of a particle, is required. The models described in literature are able to calculate the Peclet numbers when both compounds are in solution. However, they are unable to calculate the Peclet number when a compound precipitates during spray-drying [19]. This compound, which is either amorphous or crystalline, is phase separated and has a much lower diffusivity, increasing the Peclet number. Models able to accurately calculate this increase in Peclet number, and which are able to calculate the Peclet number at different stages during spray-drying, may provide a more accurate estimation. Indeed, Peclet numbers calculated for isoniazid and L-leucine showed that for the spray-drying settings tested no coating should form, TOF-SIMS analysis showed that it did. Because of this discrepancy, formulating spray-dried particles with a coating asks for a more trial-and-error approach than a quality-by-design approach.
While the delivery of antibiotics via pulmonary delivery has been our focus, it is likely that the techniques and findings described herein can be applied to other high dose drugs delivered via the pulmonary route as well [10]. It may be especially of benefit for large biopharmaceuticals. Many of the recent drug developments take place in this field. However, it has been a considerable challenge to administer them, as the human
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
intestinal lining cannot effectively absorb them. However, it has been shown that the lungs may be able to do so [12]. It is likely that DPIs are also for this group the optimal choice. Lastly, biopharmaceuticals usually require excipients to stabilize them [38]. The biopharmaceutical together with its required excipient should be considered as the dose, which taken together is often above 2.5 mg. Antivirals, especially with the issues experienced in the last few months, may also be of interest.
CONCLUSIONS
In conclusion, in this thesis opportunities and challenges for high dose drug delivery via the pulmonary route are described. Methods are developed to meet these challenges, with the focus on antibiotics used in TB treatment. However, it is likely that these techniques are also applicable to other high dose drugs. This thesis hopefully serves as a stepping stone for further development of high dose pulmonary products and techniques to formulate them, and it positively allows for product opportunities in this field to transform to success stories.
REFERENCES
1. Cheepsattayakorn, A.; Cheepsattayakorn, R. Parasitic pneumonia and lung involvement. Biomed Res. Int. 2014, 2014.
2. Lanks, C.W.; Musani, A.I.; Hsia, D.W. Community-acquired Pneumonia and Hospital-acquired Pneumonia. Med. Clin. North Am. 2019, 103, 487–501.
3. World Health Organization Global Tuberculosis Report; 2019;
4. Global tuberculosis report 2016. World Heal. Organ.
5. Pandey, R.; Khuller, G.K. Antitubercular inhaled therapy: Opportunities, progress and challenges. J. Antimicrob. Chemother. 2005, 55, 430–435.
6. Hoppentocht, M.; Akkerman, O.W.; Hagedoorn, P.; Frijlink, H.W.; De Boer, A.H. The Cyclops for pulmonary delivery of aminoglycosides; A new member of the Twincer family. Eur. J. Pharm. Biopharm. 2015, 90, 8–15.
7. Hoppentocht, M.; Hagedoorn, P.; Frijlink, H.W.; De Boer, A.H. Developments and strategies for inhaled antibiotic drugs in tuberculosis therapy: A critical evaluation. Eur. J. Pharm. Biopharm. 2014, 86, 23–30.
8. Kadota, K.; Senda, A.; Tagishi, H.; Ayorinde, J.O.; Tozuka, Y. Evaluation of highly branched cyclic dextrin in inhalable particles of combined antibiotics for the pulmonary delivery of anti-tuberculosis drugs. Int. J. Pharm. 2017, 517, 8–18.
9. Parumasivam, T.; Chang, R.Y.K.; Abdelghany, S.; Ye, T.T.; Britton, W.J.; Chan, H.-K. Dry powder inhalable formulations for anti-tubercular therapy. Adv. Drug Deliv. Rev. 2016, 102, 83–101.
10. Sibum, I.; Hagedoorn, P.; de Boer, A.H.; Frijlink, H.W.; Grasmeijer, F. Challenges for pulmonary delivery of high powder doses. Int. J. Pharm. 2018, 548, 325–336.
11. Grasmeijer, F.; Grasmeijer, N.; Hagedoorn, P.; Boer, H.W.F. and A.H. de Recent advances in the fundamental understanding of adhesive mixtures for inhalation. Curr. Pharm. Des. 2015, 21, 5900–5914.
12. Labiris, N.R.; Dolovich, M.B. Pulmonary drug delivery. Part I: Physiological factors affecting therapeutic effectiveness of aerosolized medications. Br. J. Clin. Pharmacol. 2003, 56, 588–599.
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
13. Sibum, I.; Hagedoorn, P.; Frijlink, H.W.; Grasmeijer, F. Characterization and Formulation of Isoniazid for High-Dose Dry Powder Inhalation. Pharmaceutics 2019, 11, 233.
14. Chikhalia, V.; Forbes, R.T.; Storey, R.A.; Ticehurst, M.; Pharmaceutical, R.; R, P.G.; Ct, K. The effect of crystal morphology and mill type on milling induced crystal disorder. 2005, 7, 19–26.
15. Zafar, U.; Vivacqua, V.; Calvert, G.; Ghadiri, M.; Cleaver, J.A.S. A review of bulk powder caking. Powder Technol. 2017, 313, 389–401.
16. Li, L.; Sun, S.; Parumasivam, T.; Denman, J.A.; Gengenbach, T.; Tang, P.; Mao, S.; Chan, H.-K. l-Leucine as an excipient against moisture on in vitro aerosolization performances of highly hygroscopic spray-dried powders. Eur. J. Pharm. Biopharm. 2016, 102, 132–141.
17. Chang, Y.X.; Yang, J.J.; Pan, R. Le; Chang, Q.; Liao, Y.H. Anti-hygroscopic effect of leucine on spray-dried herbal extract powders. Powder Technol. 2014, 266, 388–395.
18. Sibum, I.; Hagedoorn, P.; Kluitman, M.P.G.; Kloezen, M.; Frijlink, H.W.; Grasmeijer, F. Dispersibility and Storage Stability Optimization of High Dose Isoniazid Dry Powder Inhalation Formulations with L-Leucine or Trileucine. Pharmaceutics 2020, 1–14.
19. Vehring, R. Pharmaceutical particle engineering via spray-drying. Pharm. Res. 2008, 25, 999–1022.
20. Wang, H.; Nobes, D.S.; Vehring, R. Particle Surface Roughness Improves Colloidal Stability of Pressurized Pharmaceutical Suspensions. Pharm. Res. 2019, 36, 1–17.
21. de Boer, A.H.; Hagedoorn, P.; Westerman, E.M.; Le Brun, P.P.H.; Heijerman, H.G.M.; Frijlink, H.W. Design and in vitro performance testing of multiple air classifier technology in a new disposable inhaler concept (Twincer®) for high powder doses. Eur. J. Pharm. Sci. 2006, 28, 171–178.
22. De Boer, A.H.; Hagedoorn, P.; Gjaltema, D.; Goede, J.; Frijlink, H.W. Air classifier technology (ACT) in dry powder inhalation: Part 3. Design and development of an air classifier family for the Novolizer® multi-dose dry powder inhaler. Int. J. Pharm. 2006, 310, 72–80.
23. Who; The World Health Organization Treatment of tuberculosis: guidelines. 4Th Ed. 2010, 160.
24. World Health Organization Guidelines for treatment of drug-susceptible tuberculosis and patient care 2017 update.
25. Grasmeijer, F.; Hagedoorn, P.; de Boer, A.H. Re-designing the colistimethate sodium Twincer® for large-scale production. DDL 2017 2017.
26. Sibum, I.; Hagedoorn, P.; Botterman, C.O.; Frijlink, H.W.; Grasmeijer, F. Automated Filling Equipment Allows Increase in the Maximum Dose to Be Filled in the Cyclops ® High Dose Dry Powder Inhalation Device While Maintaining Dispersibility. Pharmaceutics 2020, 1–14.
27. Billiard, J.; Baker, S.; Chandrasekaran, V.; van den Berg, F.; Yang, S.; Dumont, E. Dry powder inhaled ribavirin in healthy volunteers: safety, tolerability, lung and systemic pharmacokinetics 2017, PA4125.
28. Bodier-Montagutelli, E.; Mayor, A.; Vecellio, L.; Respaud, R.; Heuzé-Vourc’h, N. Designing inhaled protein therapeutics for topical lung delivery: what are the next steps? Expert Opin. Drug Deliv. 2018, 15, 729–736.
29. Patton, J.S.; Brain, J.D.; Davies, L.A.; Fiegel, J.; Gumbleton, M.; Kim, K.-J.; Sakagami, M.; Vanbever, R.; Ehrhardt, C. The Particle has Landed— Characterizing the Fate of Inhaled Pharmaceuticals. J. Aerosol Med. Pulm. Drug Deliv. 2010, 23, S-71-S-87.
30. Newman, S.; Fleming, J. Challenges in assessing regional distribution of inhaled drug in the human lungs. Expert Opin. Drug Deliv. 2011, 8, 841–855.
31. Corcoran, T.E. Imaging in Aerosol Medicine. Respir. Care 2015, 60, 850–857.
32. Rodvold, K.A.; Yoo, L.; George, J.M. Penetration of Anti-Infective Agents into PulmonaryEpithelial Lining Fluid. Clin. Pharmacokinet. 2011, 50, 689–704.
33. Kempker, R.R.; Barth, A.B.; Vashakidze, S.; Nikolaishvili, K.; Sabulua, I.; Tukvadze, N.; Bablishvili, N.; Gogishvili, S.; Singh, R.S.P.; Guarner, J.; et al. Cavitary penetration of levofloxacin among patients with multidrug-resistant tuberculosis. Antimicrob. Agents Chemother. 2015, 59, 3149–3155.
34. Zeitlinger, M.; Müller, M.; Joukhadar, C. Lung microdialysis - A powerful tool for the determination of exogenous and endogenous compounds in the lower respiratory tract (mini-review). AAPS J. 2005, 7, 600–608.
7
G
en
er
al D
iscu
ssio
n a
nd F
utu
re P
er
spe
ctiv
es
35. Kruizinga, M.D.; Birkhoff, W.A.J.; van Esdonk, M.J.; Klarenbeek, N.B.; Cholewinski, T.; Nelemans, T.; Dröge, M.J.; Cohen, A.F.; Zuiker, R.G.J.A. Pharmacokinetics of intravenous and inhaled salbutamol and tobramycin: An exploratory study to investigate the potential of exhaled breath condensate as a matrix for pharmacokinetic analysis. Br. J. Clin. Pharmacol. 2020, 86, 175–181.
36. Gingell, D.; Parsegian, V.A. Prediction of van der waals interactions between plastics in water using the Lifshitz theory. J. Colloid Interface Sci. 1973, 44, 456–463.
37. Leite, F.L.; Bueno, C.C.; Da Róz, A.L.; Ziemath, E.C.; Oliveira, O.N. Theoretical models for surface forces and adhesion and their measurement using atomic force microscopy; 2012; Vol. 13; ISBN 5501532295902.
38. Hinrichs, W.L.J.; Prinsen, M.G.; Frijlink, H.W. Inulin glasses for the stabilization of therapeutic proteins. Int. J. Pharm. 2001, 215, 163–174.
