Will pulmonary drug delivery for systemic application ever fulfill its rich promise?

KEYWORDS:

• Inhalation
• systemic delivery
• insulin
• aerosol delivery
• lung

1. Introduction

While oral delivery of tablets and capsules remains the preferred treatment modality for most patients, inhaled therapeutics have dominated the landscape for managing many common lung diseases including asthma and chronic obstructive pulmonary disease and orphan lung diseases like cystic fibrosis (CF) and pulmonary hypertension [1]. The inhalation route allows drugs to be delivered directly to the site of disease, leading to improved efficacy, and reduces the potential for side effects due to lower systemic exposure [1]. These inhaled treatments are available in a variety of formats, including jet and mesh nebulizers, soft-mist inhalers, metered dose inhalers (MDIs), and dry powder inhalers (DPIs), and new technologies are becoming available to improve convenience and adherence, e.g. breath-actuation, electronic monitoring, and feedback [1]. The potential to use the inhalation route to treat systemic disease has long been debated, especially with the emergence of biotechnology products [1].

The application of recombinant DNA technology to produce biopharmaceuticals, and the subsequent approval of recombinant human insulin in 1982, ushered in the era of biotechnology. The initial focus on peptides for hormone replacement rapidly expanded to large-molecular-weight proteins and antibodies, as understanding of the fermentation, cell culture, and purification sciences improved. Biologics for systemic application could not be given orally due to degradation in the gastrointestinal tract and so have relied upon injection or infusion for efficacy. For as long as these products have been on the market, scientists have sought innovative technologies to overcome the discomfort and stigma associated with injections, the ‘crude’ insertion of metal shafts into the body to allow proteins access to the subcutaneous (SC) tissue [1].

Noninvasive administration of biologics remains an elusive goal with the exception of the inhalation route: the lung with its large surface area and the thin, alveolar epithelium makes such delivery possible and this attracted intense focus from the early 1990s. Early clinical success spurred optimism that the lung could be utilized for systemic uptake of insulin and other biologics [2]. Inhaled products were conceived to replace injections for diseases as diverse as diabetes (insulin and GLP-1), osteoporosis (parathyroid hormone), growth hormone (GH) deficiency, anemia (erythropoietin), hepatitis (interferon alpha [IFNα]), multiple sclerosis (interferon beta [IFNβ]), pain management (morphine and fentanyl), migraines (triptans and dihydroergotamine), and anxiety disorder (loxapine) [1,2]. However, significant technical hurdles had to be overcome to develop formulations that possessed adequate stability, ideally at room temperature for convenience, and inhalation systems that were efficient, reproducible, and simple to use correctly. Interestingly, safety concerns about delivery of proteins to the lung were much reduced after the approval of the first protein delivered by inhalation – recombinant human DNase – for the treatment of CF [1]. However, while conception occurred for many of these products more than two decades ago, few have reached the market and none has achieved commercial success.

2. Period of rapid technology innovation

Twenty years ago, the challenges were great. The central airways of the lung evolved to filter out and rapidly remove deposited particles by mucociliary clearance which would compete with the desired drug absorption pathway for biologics [1,2]. None of the existing inhalers were able to deliver a large, or reproducible, fraction of the drug payload to the lung, let alone the peripheral lung, where the highly vascularized alveolar region could facilitate transport of the therapeutic into the bloodstream. Expensive biologics necessitated highly efficient delivery to achieve affordable therapy and drugs like insulin with a narrow therapeutic window mandated low variability in the clinically effective lung dose to achieve efficacy and safety targets. The barriers to systemic absorption from the lung have been reviewed along with the mechanisms for absorption of peptides and proteins [2,3].

The market perception that a noninvasive inhalation product would be superior to injections drove significant financial investment into pulmonary delivery to overcome these hurdles. With many players in the race to be first to market with inhaled insulin, technology innovations developed rapidly. The initial pioneering focus was to develop new highly efficient devices for powder dispersion that used external energy supply (batteries or compressed air) to form the aerosols. Later, sophisticated particle engineering strategies emerged allowing for the use of simpler and less expensive DPIs with superior performance [4–6]. One approach involved the generation of low-density porous particles with larger geometric sizes which reduced particle cohesion and improved powder dispersibility [5,6]. These dispersible powders required more effective manufacturing and packaging technology and good manufacturing practices (GMP) facilities which could conduct these operations under well-controlled temperature and humidity conditions, e.g. Exubera inhaled insulin [4,5]. In parallel, a new class of sophisticated liquid inhalers were developed, termed soft-mist inhalation systems, like the Aeroneb Go vibrating mesh device and the AERx System [1,7]. The AERx System combined breath-control features with a disposable nozzle that produced nearly monodisperse droplets ideal for alveolar deposition that practically eliminated deposition in the oropharyngeal region and thus achieved intra-subject variability equivalent or superior to SC injection (Figure 1) [1,7].

Figure 1. Improved Inhalation Technologies like AERx lead to Reduced Inter-subject Variability and Greater Lung Deposition. Yellow circles are for AERx, green triangles are for DPIs, blue diamonds are for MDIs and red squares are for nebulizers. Adapted with permission from Cipolla D et al., [8]. Copyright 2010 Future Science. Full colour available online

With new technologies developed to deliver peptides and proteins reproducibly to the deep lung, scientists set out to determine their bioavailability and assess commercial potential in both animal and human studies. Small peptides are rapidly absorbed from the lung with 20–50% bioavailability relative to that of SC injection (Figure 2). Proteins about 6–50 kDa have moderate systemic absorption, ranging from 10% to 40% [2]. However, systemic absorption of GH and IFNα following aerosol administration in animals was 45% and 70% [2], respectively, versus 3–10% for inhaled GH and IFNα [1] in humans; thus, caution is advised when using pulmonary absorption studies in animals to make estimates in humans. Large-molecular-weight antibodies are not absorbed across the lung to a significant extent (≪10%) unless an active transport system is utilized such as the neonatal Fc receptor; e.g. the bioavailability of erythropoietin with a molecular weight of 30 kDa was improved from 15% to 35% by attaching monomeric erythropoietin to a fusion protein with a combined molecular weight of ~82 kDa [9]. Somewhat surprisingly, the lower pulmonary bioavailability relative to SC injection for many of these peptides, including insulin, did not spell the end to development of an inhaled insulin product. In the intervening years, the insulin API manufacturing capacity had expanded, addressing potential supply concerns and affordability of an inhaled product. The initial clean safety record for inhaled proteins, including recombinant human DNase that had been widely used since 1993 to treat CF, also supported further clinical development [1,10].

Figure 2. Systemic Absorption of Peptides and Proteins from the Lung. Following intratracheal instillation in rats and dogs, the bioavailability relative to SC injection and the Tmax are reported. Adapted from [2].

3. Inhaled insulin clinical development and market reception

Large-scale clinical trials with the inhaled insulin products were initially delayed while device performance and feature set were optimized and GMP dosage form manufacturing capacity was established for the three most advanced systems: Exubera (Nektar, San Carlos, CA, USA), the AERx insulin diabetes management system (AERx iDMS, Aradigm), and the AIR insulin system (Alkermes, Waltham, MA, USA). The Exubera inhaled insulin program was also delayed to address regulatory concerns associated with an observed increase in non-neutralizing insulin antibodies and a clinically insignificant decline (~0.1 L) in lung function (FEV₁) in diabetic patients which returned to normal upon cessation of therapy [11]. Both issues were ultimately addressed to the satisfaction of the regulators. Clinical trials with all three inhaled insulins demonstrated pharmacological profiles with more rapid onset and a shorter duration of action than for SC regular insulin, thus more closely mimicking the endogenous prandial insulin profile [11]. Inhaled insulins were well tolerated with no increase in the rates of hypoglycemia compared to SC insulin in both Type 1 (T1D) and Type 2 diabetes (T2D) [11,12]. In T2D, Exubera had a lower incidence of hypoglycemia and was associated with reduced weight gain than SC insulin [12]. Furthermore, in all studies that assessed patient satisfaction, inhaled insulin was preferred over SC insulin [11,12].

Exubera was approved in 2006 but a year later was discontinued from the market due to ‘commercial reasons’ even though the sales ramp for Exubera was on the same trajectory as that for the injectable long-acting insulin (Lantus) which went on to become one of the top three selling pharmaceutical products [12]. The speculated reasons for the claimed ‘slow adoption’ were the large size of the device, requirements for periodic lung function tests, labeling differences (Exubera was dosed in milligrams versus the conventional international units [IUs] for SC insulin), lack of dose linearity (1 and 3 mg strips equated to 3 and 8 IUs of insulin), and fears of long-term safety [11–13]. Shortly after pulling Exubera from the market, Pfizer (New York, USA) notified regulators that there were six cases of lung cancer in their controlled clinical trials, all in heavy ex-smokers, five on Exubera and one on SC insulin [12]. This difference was not statistically significant and was notably lower than the expected rate of lung cancer in the general population of ex-smokers [12]. Following this announcement, Novo Nordisk (Copenhagen, Denmark) and Eli Lilly (Indianapolis, IN, USA) terminated all Phase 3 development activities for AERx iDMS and AIR insulin programs, respectively, presumably because their SC insulin franchises were now no longer at risk [12].

MannKind (Valencia, CA, USA) entered the inhaled insulin race late but was unfazed by the Pfizer termination of Exubera and continued development of Afrezza inhaled insulin in a small, palm-sized device, addressing the most obvious shortcoming of Exubera, its large size. Afrezza’s formulation adsorbs insulin to fumaryl diketopiperazine crystals, an excipient that increases insulin transcytosis in the lung, resulting in even more rapid absorption of insulin compared to the other inhaled insulins, a claimed potential advantage even to SC faster-acting insulin analogs [14]. The clinical development program was subsequently delayed due to a change from the MedTone to the Dreamboat inhaler requiring an additional Phase 3 trial in both T1D and T2D [15]. The clinical efficacy data from these trials for Afrezza met the primary end point of non-inferiority to SC rapid-acting insulin (insulin aspart) in prandial blood glucose lowering in T1D and had superior blood glucose lowering in T2D compared to inhaled placebo powder [14,15]. Additionally, in T1D, treatment with Afrezza resulted in significantly lower fasting plasma glucose levels, less hypoglycemia, and lower bodyweight gain [14,15]. For the dry powder insulin formulations, including both Exubera and Afrezza, cough was the most common adverse event compared to SC insulin [12,14,15]. Afrezza was approved by FDA in June 2014 with Sanofi (Bridgewater, NJ, USA) as the marketing partner [12]. Due to poor sales, Sanofi recently returned Afrezza marketing rights to MannKind who continues to market Afrezza. Does this spell the end for inhaled insulin?

4. Expert opinion

The future of inhaled insulin certainly looks bleak from the commercial perspective; however, many diabetic patients on inhaled insulin continue to be strong advocates for the product. I believe that an opportunity exists not only for inhaled insulin, but also for use of the inhaled route to deliver systemic therapies, and I am not alone in that belief [12,13]. The perception that an opportunity exists is based not on projections of near-term commercial profit calculations, but on patients with medical needs who currently have inadequate treatment options (Table 1).

Table 1. Systemic inhaled products in development to treat patient populations with unmet needs.

Unmet need	Drug	Product and latest stage of development	Manufacturer	Reference
T2D not on injectable insulin	Insulin	Afrezza, marketed	MannKind	[11–13]
		AERx iDMS, Phase 3	Aradigm	[7,13]
		Dance-501, Phase 2	Dance Biopharm	[13]
Anxiety^a	Loxapine	Adasuve, marketed	Alexza Pharma	[1,13]
Migraine^b	Dihydroergotamine	Semprana, NDA submission	Allergan	[1,13]
	Zolmitriptan	CVT-427, Phase 1	Acorda Ther.	-
Parkinson’s disease	Levodopa	CVT-301, Phase 3	Acorda Ther.	[13]
Breakthrough pain management	Morphine/fentanyl	AERx PMS, Phase 2	Aradigm	[1,7,13]
Smoking cessation	Nicotine	AERx nicotine, Phase 2	Aradigm	[1,7,13,16]
Insomnia	Zaleplon	AZ-007, preclinical	Alexza Pharma	[1]
Acute coronary syndrome	Relaxin	Unavailable	Unavailable	–
Idiopathic pulmonary fibrosis	IFNγ	Preclinical	InspirRx Pharma/Nostrum Pharm	[1,13]

^aDue to risk evaluation and mitigation strategies, the marketing restrictions with this product make it difficult to address the true unmet need in bipolar disorder and schizophrenia.

^bSemprana was formerly named Levadex; NDA: new drug application; PMS: pain management system.

In diabetes, like many diseases, it takes time to educate patients and health-care personnel about the benefits of new therapies. Most patients already on injectable insulin are unlikely to switch to an alternative product, like inhaled insulin, until a long-term comfort level has developed, and that takes time. However, many T2D patients are not on injectable insulin, with an average delay of 5–10 years from diagnosis to insulin prescription, even though the recent guidelines recommend that insulin treatment be initiated within 2–3 months of diagnosis [12]. Delaying insulin treatment in T2D has profound health-care implications for both the individual, leading to increased morbidity and mortality, and society at large [12]. To change the established treatment paradigm, the focus should instead be on transitioning insulin-naive T2D patients onto inhaled insulin, if they are reluctant to transition to SC insulin which is a real concern for many patients [12]. But this strategy requires patience and a long-term focus [12].

Many times in the history of pharmaceutical development, a new formulation/delivery technology (e.g. liposomes) or therapeutic mechanism (e.g. immuno-oncology) has been associated with initial euphoria only to see the market-driven enthusiasm ebb as unanticipated technical hurdles beset development. In a race to beat out their like-minded brethren, large pharma and the investment community instead become captivated by the next ‘big thing’ (e.g. gene therapy in the 1990s, stem cell therapy and siRNA in the 2000s). Remarkably, when interest reaches a nadir is often when most of the technical challenges have been identified and de-risked, and those with insight and persistence can make the most of the market opportunity! For example, many liposomal products are now on the market and liposomes have reemerged as a promising therapeutic differentiator, even in the inhalation space [5]. Similarly, the immuno-oncology field emerged as the 2013 Science ‘Breakthrough of the Year’; however, since initial efforts in immuno-oncology began in the 1970s, there were 30 years of adversity before the recent successes and enthusiasm.

We now have many elegant inhalation device and formulation technologies but the interest for noninvasive systemic delivery is out of favor. While much of the initial driving force was to avoid injections, the low-hanging fruit may be using the inhaled route to improve the pharmacokinetic profile relative to oral, buccal, or transdermal therapy; e.g. more rapid absorption from the lung may provide benefit for the large population of smokers who want to quit but are addicted to nicotine and fail current cessation therapy [16]. Many smokers turn to e-cigs to provide a safer alternative to cigarettes because there are no marketed nicotine replacement therapies that provide rapid nicotine absorption through the lung [16]. Inhaled medications to treat people with migraines, insomnia, anxiety, Parkinson’s disease, and cancer patients with episodic breakthrough pain have the potential to provide faster relief or differentiated effectiveness and ultimately improvements in productivity and quality of life (Table 1). Migraine sufferers continue to await approval of inhaled fast-acting dihydroergotamine which demonstrated clinical benefit in Phase 3 trials but is mired with chemistry, manufacturing and controls issues (Table 1). Development of inhaled systemic treatments will flourish again; the only question is when.

Declaration of interest

D Cipolla is an employee of Aradigm Corporation and participated in the development of their AERx inhaled insulin diabetes management system. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Acknowledgments

The author is indebted to fruitful discussions with Wilbur de Kruijf, Igor Gonda, David Lechuga, Gerhard Scheuch, and Jeff Weers regarding the future of pulmonary delivery.

Additional information

Funding

This paper was not funded.