Robert Lins, MD

	Respiratory diseases are a growing health problem, with as much as 7% of the world’s population now thought to suffer from one or another. Yet, in contrast to many other diseases for which the prognosis has improved, the mortality rate for respiratory illnesses is rising, having doubled in the past four decades.

 The two most common respiratory diseases, asthma and chronic obstructive pulmonary disease (COPD), have experienced an increase in prevalence, partly because the population is ageing, but also as a result of the fact that exposure to common risk factors is becoming more common.

	Although there have been advances in treatment in recent years, largely because the mechanisms underpinning the disease are more widely understood, for many patients, asthma and COPD still represent a significant area of unmet medical need. However, patient stratification techniques such as endotyping and phenotyping are beginning to make targeted treatment choices a reality, holding out hope that in the future, chronic respiratory diseases will become much more manageable.

	While that improved understanding of the causes of disease, particularly the inflammation pathways involved, has provided novel targets for drug therapies, significant difficulties in establishing successful clinical trial endpoints remain. Clinicians still largely have to rely on a combination of pharmacodynamics and surrogate outcome parameters. This is a problem, as sensitivity and quantitative accuracy are lacking, particularly for surrogate endpoints.

	Commonly used surrogate measurements include FEV-1, which involves the measurement of dynamic lung volume after one second of forced expiration, and clinical exacerbations.  There are also a large number of subjective scales used, including the St. George respiratory questionnaire for COPD, but also validated in other respiratory diseases, which assesses symptoms, activity and the impact of the disease in a qualitative rather than quantitative way. The correlation between measurements and overall survival is not good for any of these techniques, and there is a general lack of sensitivity that makes it difficult to assess an intervention with any great accuracy.

	As a result, clinical trials in respiratory diseases are particularly difficult to run and manage, both in terms of recruiting the numbers of patients required, and the ability to measure outcomes accurately. However, by applying different thinking to the problem, better measures may be possible, and several alternative techniques that are already used by clinicians when treating patients may ultimately prove successful in assessing patients in clinical trials.

	Spirometry is used to quantify relative volumes of air in the lungs, and while it can be used to produce FEV-1 measurements, it does not give values for absolute volumes. This is a problem, as residual volume, functional residual capacity and total lung capacity are also important factors in respiratory disease. Clinicians are increasingly using body plethysmography, or the ‘body box’, to assess lung function, as it is better controlled and permits more sensitive measurements of dynamic lung volumes. 

	Furthermore, this technique is capable of providing greater insight into lung function, such as measurements of residual volume and airway resistance, as well as the diffusion capacity for carbon monoxide and oscillometry. It is even able to assess the reversibility of bronchoconstriction after the patient has inhaled a bronchoconstrictor, and this ability to give sensitive and repeatable measurements in bronchial challenge tests could be of great value in a clinical trial setting.

	It is possible to gain an insight into the concentration of inhaled drugs and outcome biomarkers in the lower airways via sputum induction and the measurement of inflammatory cytokines as biomarkers.

 The method involves the inhalation of increasing concentrations of saline, and its simplicity means its use has become commonplace in drug development programs. 

	The technique is not perfect, however, and its reproducibility has been questioned. Furthermore, the preparation of sputum samples for analysis is non-trivial, even for experienced analysts. There remains no recommended procedure, and the repeatability of sputum induction and collection is unclear.

	A study carried out by SGS however has shown good results to be achievable. In comparison to reported results of 10% and 70% for healthy non-smoking patients and asthma patients respectively

, success rates of 29% and 74% were achieved in patient groups of 175 and 35. A high success rate was also achieved in healthy volunteers who smoke, albeit with too few subjects to allow firm conclusions to be made. With care and precision, the technique can be operated reliably.

	Local bronchial pharmacokinetics is another method that can be employed on collected bronchoalveolar lavage fluid, to establish the time–concentration profile of both drugs and cytokines when looking to predict therapeutic efficacy. Its utility lies in the ability to assess both the local and systemic pharmacokinetics of inhaled drugs, and it can be used with either single doses or repeated administration. This method however, is invasive. To make the necessary series of measurements, the bronchoscope needs to be wedged into multiple different points of the bronchial tree, and the saline has to be infused distant from the scope before samples are collected. It must therefore be carried out by a pulmonologist who has expertise in bronchoscopy.

	Functional respiratory imaging is much less invasive, although significant investment in computer technology and scanning capabilities is required. This method uses 3D segmented computer models of human organs. First, data must be generated from multiple different imaging techniques, including ultrasound, magnetic resonance imaging and high-resolution computed tomography. These are then put together using advanced computational tools such as those that were originally developed for the aerospace industry, such as finite element analysis and computational fluid dynamics.

	Once the models have been generated, it is possible to measure airway resistance in a variety of different airways. Other measurements that can be made include changes in lobar hyperinflation and lobar perfusion. It is also possible to make assessments of local drug deposition. Pharmacodynamic measurements are also possible, if the models are used in combination with the administration of inhaled radiolabelled drugs that are tracked using positron emission tomography.

	This method has already been used for early stage clinical trials in COPD and asthma, as well as cystic fibrosis, idiopathic pulmonary fibrosis and sleep disorders. Its non-invasive nature means it could be applicable to the later stages of clinical development, too. 

	Respiratory challenge tests have great potential in the clinical development of drugs for respiratory diseases, as they allow novel therapies to be assessed in healthy volunteers, as well as those with mild or moderate forms of disease. One of the longest established test types, bronchoprovocation, involves the subject inhaling rising concentrations of a solution of histamine before the responsiveness of the airway is measured. The test allows bronchial hyperreactivity to be diagnosed and, importantly, quantified. The measurement, PC20, is the concentration of challenge agent required to generate a 20% reduction in FEV-1.

	Commonly used to diagnose asthma, bronchoprovocation can be used in addition to, or instead of, supplement reversibility tests as these frequently give false negatives, particularly in mild to moderate asthma that is well controlled. It also has potential in assessing asthma–COPD overlap syndrome. In trials, it can increase the number of eligible subjects in pre-trial screening.

	It is not perfect, however. The histamine challenge agent can cause bronchoconstriction, headache and even tachycardia. Alternatives such as metacholine and adenosine can be used instead, but while these have a lower incidence of side-effects, they are not so commonly used. All three have potential in proof-of-concept trials, particularly in asthma, where bronchoprotection or bronchodilation measurements are required.

	Another substance that can be administered via inhalation in a challenge test is the TLR-4 agonist, lipopolysaccharide (LPS), an endotoxin which initiates an acute inflammatory response, and the production of cytokines. This mimics one of the mechanisms involved in COPD and asthma, and enables the effectiveness of antiinflammatory drugs to be assessed. 

	There are two ways in which LPS can be administered. Either a solution of up to 50µg/ml in isotonic saline can be inhaled using an ultrasonic nebulizer, or 4ng/kg can be given via intravenous infusion over two minutes. Several parameters can be assessed in induced sputum, including the levels of a range of cytokines including IL-1beta, IL-6, IL-8, IL-10 and TNF-alpha, and the number of granulocytes.

	An alternative technique, the viral challenge model, has been used in the early stages of clinical development for some time. It is particularly valuable in proof-of-concept studies in infections of the upper respiratory tract. The method can be used to establish clear correlates of protection for both antiviral drugs and vaccines, as well as informing go/no-go decisions and the up-or-down selection of the different arms of a trial. As asthma is often made worse by viral infections, it is increasingly being used in trials for asthma therapies.

	In order to run a viral challenge trial, a dedicated isolation suite is required. Ideally, this will be situated in a clinical pharmacology unit, where the staff are experienced in running this specific type of study. The trials involve inoculation with a virus to cause infection ahead of the administration of the study drug or vaccine, and its efficacy measured. They can be used with healthy volunteers, and also in asthma patients to assess the effectiveness of drugs in treating or preventing asthma exacerbations if the virus used in the challenge is one that typically heightens asthma symptoms. Healthy subject or patient recruitment can be problematic, particularly if there are co-infection, antibody level or other inclusion/exclusion criteria that need to be met. But in the right circumstances, a viral challenge study can be an important addition to the clinical trials repertoire for the development of treatments or preventers for respiratory disease.

	These problems can be illustrated by highlighting a Phase I randomized, double-blind, placebo-controlled trial that was undertaken, first in 12 healthy normal volunteers, and then, in the second part of the trial, in 60 asthmatic subjects, on a monoclonal antibody targeting human TLR-3 for the prevention of asthma exacerbations. Subjects were given intravenous doses of the antibody ahead of inoculation with human rhinovirus type 16. The goals of the trial were initially safety and tolerability in the first part, and then the efficacy, as measured by pulmonary function testing and patient-reported outcomes, within the follow-up second part. A range of secondary endpoints included pharmacokinetics, pharmacodynamics assessed via additional pulmonary function tests, the cold symptom assessment score, and fractional excretion of nitric oxide (a parameter for inflammation), biomarkers in nasal lavage, immunogenicity and pharmacogenomics. 

	The initial phase of the trial was successfully executed in four months with no delays, including in the potential hurdle of gaining regulatory approval to test a drug with a new mechanism of action combined with the viral challenge. However, recruitment to the program was a major challenge, as to enroll just 12 individuals, over 150 healthy, consenting volunteers had to be tested, because of a higher than expected positive antibody status. Greater numbers of inclusion and exclusion criteria for the second part of the study led to a 100% failure rate for recruitment after the assessment of 80 asthmatic patients. As a consequence, the second part of the trial could not be completed.

	It is early days, but many of these techniques are showing a good deal of promise as potential endpoints in clinical trials for respiratory diseases. It is undeniable that classical primary respiratory endpoints have only limited success when carrying out clinical trials in respiratory diseases, whether the studies are exploratory or confirmatory in nature. 

	Although most applications for the newer techniques have, thus far, been in Phase I and II trials designed to establish safety and proof of concept, a number could translate to be of clinical use in later stage trials. Respiratory clinical trials remain challenging to run and endpoints create many problems that need to be overcome. However, with some imaginative use of modern methodologies, this situation could be improved, to the great benefit of patients with these serious diseases that are not adequately treated or controlled.

	Robert Lins, MD, is Project Director, Respiratory Diseases at SGS Clinical Research.

		Jiemin Ma J, Ward EM, Siegel RL, Jemal A. Temporal Trends in Mortality in the United States, 1969-2013. JAMA 2015; 314: 1731-1739 	

		Cazzola M, MacNee W, Martinez FJ, Rabe KF, Franciosi LG, Barnes PJ, Brusasco V, Burge PS, Calverley PM, Celli BR, Jones PW, Mahler DA, Make B, Miravitlles M, Page CP, Palange P, Parr D, Pistolesi M, Rennard SI, Rutten-van Mölken MP, Stockley R, Sullivan SD, Wedzicha JA, Wouters EF, Society AT, COPD ERSTFooo. Outcomes for COPD pharmacological trials: from lung function to biomarkers. The European Respiratory Journal 2008; 31: 416-469	

		Chanez P, Holz O, Ind PW, Djukanovic R, Maestrelli P, Sterk PJ. Sputum induction. Eur Respir J 2002; 20 (Suppl 37): 3s–8s	

		Vos W, De Backer J, Poli G, Devolder A, Ghys L, Van Holsbeke C, Vinchurkar S, De Backer L, De Backer W. Novel Functional Imaging Of Changes In Small Airways Of Patients Treated With Extrafine Beclomethasone/Formoterol. Respiration 2013; 86: 393-401

		Bergmann KC, Sehlinger T, Boelke G, Suberbier T. Clinical validation of a mobile environmental allergen exposure chamber. Allergy 2015; 70 (issue S101): 540  

Articles

Respiratory disease trials are difficult to initiate, both in terms of patient recruitment and accurately collecting data. Alternative methods are required to improve outcomes and develop new treatments for respiratory diseases.

Robert Lins, MD

Articles

Early Phase Pharmacodynamic Models for Respiratory Drug Candidates