Respiratory Disease: Considerations For Conducting Early Phase Trials In Orally Inhaled Products
Over the last 20 years, the rising rate of respiratory diseases has been driving new drug research and development in this therapeutic area and required the development of specialized knowledge and expertise to address the specific needs of respiratory early and late phase clinical trials. Anticipating this rise and need for specialization, SGS’s Clinical Pharmacology Units invested in training nurses and technicians to perform high quality pulmonary function testing, using different inhalers (pMDI, DPI, aerolizers...) as well as in equipment and patient access.
Over the years SGS continued these efforts to keep pace with the ever-changing drug development arena. In this article we will provide a backdrop for the rising demand of these specialized respiratory techniques as well as provide guidance for conducting early phase respiratory disease clinical trials based on our experience and in-depth knowledge of the field.
In the early 1990’s, chronic obstructive pulmonary disease (COPD) ranked fifth on the list of causes of death in the Western world, after ischemic heart disease, cerebrovascular diseases, lung cancer and pneumonia. Ever since, its deadly impact steadily grew in importance. COPD is now listed in a fourth position and it is estimated that by 2030 it will become the third leading cause of death, hereby even surpassing lung cancer (1) (4).
While causalities due to cardiovascular diseases and stroke have progressively dwindled over the last three decades, COPD related fatalities have increased at a similar rate.
COPD, asthma, lung cancer, pneumonia and tuberculosis together make up the five most common respiratory conditions. They are often classified together and count for 20% of global mortality (2) (3).
The annual cost evolving from these respiratory diseases is estimated to be 102 billion EUR in Europe alone. COPD is responsible for 34.1 billion EUR, of which 28.5 billion EUR due to indirect costs such as productivity losses and loss of working days (3). This strongly indicates that without improved treatments the chronic respiratory diseases, and COPD in particular, will have an enormous socio-economic impact (4) (5).
As a result of this increased need for improved treatments, the pharmaceutical market for respiratory conditions has expanded enormously. During pipeline review of all top 15 pharmaceutical companies, about 25 new drug candidates have been identified that are being tested in early phase studies. This number is expected to rise since new potential targets have been discovered in recent times. In the past, treatments for respiratory disease typically concerned small molecules. This changed by 2005-2006 and now includes an array of options: nanobodies, monoclonal antibodies, peptides, and various modes of administration such as intravenous injection, inhalation via Dry Powder Inhaler (DPI) or pressurized Metered Dose Inhaler (pMDI), nebulisation, oral administration, etc.
In response to the increased demand for respiratory clinical trials, SGS’s Clinical Pharmacology Units committed themselves to providing the industry with the required resources and experience. Throughout the last two decades, SGS has conducted a compelling number of early phase studies in this therapeutic area, asthma and COPD being the most common indications.
The first respiratory clinical trials performed at SGS, investigated the long-acting B2-mimetics in the mid-nineties. By the end of the 20th century, as a result of the European ban on chlorofluorocarbon (CFC) in aerosols, SGS experienced another boost of trials, this time researching various propellants and DPI devices. In total, more than 60 trials in this therapeutic area were executed by SGS over the past 15 years. Since 2008, the Antwerp clinic alone, conducted at least 16 early phase clinical trials in asthma and COPD, of which eight trials included at least 1 panel of patients. The GMP Certification of the site Pharmacy has allowed SGS to hand-prepare dry powder capsules on-site thus overcoming the difficulties often faced by sponsors to supply ready-prepared IMP for early phase trials
In addition to the early phase trials, the Antwerp Phase I unit functions as an important contributing site in many phase 2 and phase 3 respiratory disease trials.
Most Common Early Phase Study Designs
First-In-Human Trials (FIH)
When designing a FIH clinical trial for orally inhaled products (OIPs) it is important to take into account the probability of low systemic exposures due to the nature of drug delivery as well as the possibility of systemic reactions. This should be addressed in the protocol by means of an adequate PK/PD sampling schedule and appropriate cardiovascular and biochemistry monitoring.
In practice, we do recommend a relatively classical three-part combined protocol for the FIH trial, where the first part would be the single ascending dose (SAD) part followed by a multiple ascending dose (MAD) in part two, both conducted in healthy volunteers.
For the SAD performed in the first part of the combined protocol, SGS has had good experience with an alternating panel design where two panels are given single ascending doses over different periods. Escalation schemes need to be discussed based on the emerging preclinical data.
|Period 1||Period 2||Period 3||Period 4|
The MAD part has a sequential panel design with a dosing schedule based on the results seen in the SAD. To reduce overall study time, the MAD part can run with a partial lag to SAD part rather than fully sequential.
An early proof-of-concept trial can be added to this protocol as a third part, using a sufficient number of patients (usually in the order of magnitude of 18). SGS does recommend testing a low dose and a high dose as well as placebo in a 3-way crossover design in an effort to eliminate high inter-individual variability to the drug-response by letting the patients act as their own control.
For substances with long half-lives (e.g. monoclonal antibodies) some adaptations of this basic design are necessary to avoid carry-over effects between periods. A parallel group design could be recommendable here.
It’s evident that this type of design does not only apply for OIPs, but also works for other modes of administration.
Also for monoclonal antibodies, at least the single ascending dose part can often be run in healthy volunteers, speeding up the first knowledge on single dose pharmacokinetics and possible pharmacodynamic effects, without “wasting” valuable patients (and maybe even recruitment time) in this first stage.
Bioequivalence (BE) – Therapeutic Equivalence (TE) trials
For orally or IV administered drugs, BE trials are usually quite straightforward and the assessment of BE is mainly based on the coefficient of variance (CV%) and on primary PK parameters (Cmax, Tmax and AUC). For such compounds, BE is usually established if the 90% confidence interval for the ratio of the test and reference products are contained within the acceptance interval of 80-125% for the parameters like Cmax and AUC.
For OIP the situation is much more complex, and a lot of debate still exists on this matter: measured plasma concentrations (PK) do not reflect differences in airway disposition between test and reference product, nor do they reflect equivalent local activity.
For that reason, it can be needed to perform two different clinical trials (i.e., a PK study for evaluating the systemic exposure and a TE study showing that the local efficacy is the same) to demonstrate BE between two different formulations of the same locally acting compound.
Although a dose-response relationship is recommended by most regulatory instances for demonstrating equivalence of local delivery, most orally inhaled products exhibit a shallow dose-response for standard clinical parameters of efficacy, such as lung function and symptom scores (e.g. inhaled corticosteroids).
Secondly, whilst doses on the steep part of the dose response curve are most sensitive to detect product differences and are hence ideal to compare test and reference products, the lowest doses of most marketed reference OIPs lie already on the flat part of the dose-response curve, thereby complicating equivalence comparisons with test formulations (6) (7) (8). Those two factors make the dose-response criterion for OIPs very difficult to achieve.
a. BE testing
The in vitro comparison between the different products remains one of the pillars to demonstrate bioequivalence. However, in addition to this comparison, several other types of studies have been proposed by the European guidance to deal with the problem of showing bio/therapeutic equivalence.
For bronchodilators SGS’s current preference is the four-way crossover design using PK with and without charcoal block, as a first pivotal trial. Charcoal blocks are used to differentiate local absorption through the lungs from absorption in the GI tract due to swallowed drug that would be adhering to the palatine-pharyngeal walls after inhalation. When absorption of drugs is blocked by the charcoal, the systemic availability of an inhaled drug is directly proportional to the amount of drug delivered to the airways (6) (7) (8). It is recommendable to use a charcoal block when the anticipated gastro-intestinal absorption of drug would be > 1%.
This approach can be used for OIPs, both for bronchodilators as well as inhaled corticosteroids (ICS) and/or combination products. Systemic (equivalent) safety needs to be assessed by introducing the necessary safety parameters in the clinical trial protocol. For ICS this comprises measuring possible effects on the hypothalamic pituitary adrenocortical axis (e.g. 24h cortisol levels).
The standard population would ideally be healthy volunteers. They can perfectly act as surrogate for patients, usually showing slightly higher systemic concentrations than patients. In cases where the disease state might impact the relative bioavailability and lung disposition, patients should be considered as an additional arm (e.g. cystic fibrosis).
b. Therapeutic equivalence testing
As stated before, obtaining dose-response relationships for OIPs can be very difficult to achieve. For bronchodilators it can be possible to obtain a dose –response by using a challenge model with histamine or metacholine. Different methods have been suggested: bronchoprotection, bronchodilation and the improved airways paradigm.
b.1. Bronchoprotection trials
In this type of trial, patients are submitted to a brochoconstrictive challenge agent (usually histamine or metacholine) at the estimated peak pharmacodynamic effect of the supposed OIP. A crossover design is the best choice, using placebo, low dose and high dose of comparator/reference and test drug in a 6x6 latin square balanced allocation sequence (9). A medium dose level can be added for better defining dose-response.
At baseline patients are chosen based on their medical history, pulmonary function testing and the initial concentration of challenging agent where the forced expiratory volume in one second (FEV1) drops with 20% of baseline (PC20) for metacholine or histamine (12). This PC20 should be ideally < 4mg/mL at baseline to increase the robustness of data obtained by the bronchoprovocation. After recovery of the baseline challenge, patients are administered a short acting beta-agonist and the PC20 is assessed once more at 10 to 15 minutes post bronchodilation with the reference product (or salbutamol). At our CPU we use the method as described by Juniper and Hargreave (11).
Subsequent visits are performed with at least a 3 day washout period (maximum 7 days), where patients are randomized to the different treatment regimes. At each of these visits PC20 is assessed at the same time interval relative to drug administration.
The timing of the PC20 assessment needs to be aligned with the anticipated pharmacodynamic effect profile. For long-acting bronchodilators the assessments of PC20 can be placed at 1 hour postdose, and at 12 hours and/or 24h postdose.
Dose-response curves of test and reference products are plotted against the (log2) PC20 values. Without going into details, a repeated-measures analysis of variance (ANOVA) is one of the cornerstones of the data-analysis.
Heart rates, blood pressure and pulmonary function testing are measured at several timepoints. Adverse effects of the challenging agent are noted on a 3-point Likert scale. These adverse effects are: throat irritation, hoarseness, flushing, nose irritation, cough, wheezing, headache, tightness of chest and shortness of breath.
As previously mentioned, the PC20 should ideally be situated < 4mg/mL, so mild-to-moderate stable patients should be used in this type of trial.
Under ideal conditions, the placebo-arm can be omitted if one would choose to use the results of the challenge test at screening as baseline, shortening the trials by 2 periods per subject/patient.
Bronchoprotection trials for ICS have been described, however there is not much experience with this model for the effects of corticosteroids.
b.2. Bronchodilation trials
Bronchodilation trials are fairly straightforward designs where the subjects are randomized to a crossover design, using low dose and high dose of reference and test product, again using a 4x4 latin square balanced sequence.
Placebo control can be inserted as option, but good and careful baselining and including stable patients will prevent the need for a placebo-arm.
On each study visit the Peak Expiratory Flow (PEF), FEV1 and respective Area Under the Curve (AUC) are measured at selected time points and the values are compared product to product and dose to dose.
As touched earlier: it can be extremely difficult to show any dose-response, as the marketed strengths of reference product are usually situated around the near-flat part of the dose-response curve.
b.3. Assessment of ‘improved airway function’ trials for ICS
For ICS testing the paradigm of therapeutic equivalence is even more difficult because the dose-response for clinical outcome parameters with ICS is shallow and reaction to treatment is delayed. There are methods published based on eosinophil counts, exhaled fraction of nitric oxide (FeNO) or on spirometry. But none of them is recognized as optimal.
It is known that patients with mild eosinophilic asthma do respond to treatment with ICS (eosinophilia defined as > 3% eosinophils in white blood cell counts). Hence eosinophils in expectorated sputum are used as biomarker for response to ICS treatment in asthmatic patients. Quantitative sputum cell counts are regarded as a valid and discriminative biomarker in patients with moderate-to-severe asthma. As spontaneous obtained sputa are usually of low quality with regard to sputum plugs that can be recovered (we need at least 100 mg of sputum plugs per sampling timepoint), sputum is best obtained during an induction sequence with increasing concentration of hypertonic saline. Sputum induction is repeatable and the eosinophilic inflammation is responsive to ICS treatment, whereas neutrophilic (non-eosinophilic) inflammation is less responsive to treatment with ICS (13) (14).
This method is often combined with FeNO. FeNO increases with increasing sputum eosinophilia. FeNO alone is not specific enough because it is influenced by many different factors and becomes less specific and discriminative than sputum cell counts.
However this biomarker can be of great interest when used in combination with other ones. FeNO assessment is easy to implement, fast, non invasive and well-tolerated by the patients.
Another way of testing therapeutic efficacy of ICS can be done by a randomized, double-blind, parallel group design for both test and reference product. The primary parameters of interest here are then FEV1 and PEF, measured daily at home (ideally). At least 2 doses should be tested of both test and reference product for four weeks.
Population under study would ideally be stable but undertreated patients, who are responsive to ICS treatment. In these trials the aim is then to demonstrate an equally improved airway function between test and reference product. This remark also applies to the other study designs: in Western Europe physicians tend to overtreat patients rather than undertreat. The art is to find those patients that are stable, but that will improve if treatment would be optimized.
What about the practical aspects : tips’n’tricks
|Site staff trained for inhalation technique and device?||Site staff trained for pharmacodynamic parameters?||Does the site have performance indicators or quality controls in place for both drug administration and pharmacodynamics?|
|Is the volunteer/patient trained for device and inhalation technique?||Is the blood sampling room protected from contamination due to drug administration?||30Is there a written instruction for the drug administration available, describing all the different steps like priming, holding breath…?|
|Calibration equipment?||Formula’s used for calculating predicted values? Which Author has been chosen?||Is there a discrepancy between expectations of sponsor and understanding of them at the site?|
|Patient access?||Stability of lung function ? Reversibility criteria ?||Is there a discrepancy between expectations of sponsor and understanding of them at the site?|
|Availability of IMP ?||Can IMP be made “on site” in GMP environment ?||Ability to manufacture IMP just before administration ?|
Techniques and tests available at SGS
SGS has invested a lot in training our technicians to meet the applicable standards in pulmonary function testing. The following are techniques and tests available at SGS, performed by dedicated SGS staff:
|Performed in house by trained SGS staff||In collaboration with the local department Pulmonology|
|Spirometry||Fraction Exhaled NO (FeNO)||Broncho provocation testing|
|Vitalograph® Training for use of DPI/MDI
||Sputum Induction (cytospins and RNA)
||Broncho Alveolar Lavage|
|Exhaled Breath Condensate||Whole Body Plethysmography||Bronchoscopy|
Designing trials for OIPs is not always as easy as it might seem. Study design needs to be adapted on a case-by-case basis, bearing in mind the ultimate goal of the trial, as well as the characteristics of reference products available, the characteristics of the target population and the applicable regulatory guidances.
Before going into complicated and expensive biomarker program, we should try to see if there are other ways to reach the goals of the development program. We need to carefully think about the added value of bronchoprotection/bronchodilation trials.
The question ‘Are there other biomarkers or techniques that can add value to my trial?’ should be answered well in advance of starting the clinical part, allowing the clinical site to train and retrain themselves.
The specifics of the patient population to be included needs to reflect current clinical practice and needs to be realistic. The impact of “recruitability” on quality of study data is not to be underestimated.
As a lot of debate is still ongoing, the regulatory guidances and visions from both EMA and FDA will be updated/changed in the near future.
Ultimately, when a doubt exists whether or not a development program would fulfil the expectations of the regulatory bodies, a study design should be submitted for regulatory advice to EMA/FDA.
Steven Ramael, MD, FBCPM
Medical Director Early & Late Phase - Clinical Research
SGS Life Science Services
(1) Eurostat data September 2011 (online data code : hlth_cd_asdr).
(2) Chronic Disease Alliance. A Unified Prevention Approach. European Society of Cardiology, European Society of Hypertension, European Cancer Organisation, European Association for the Study of the Liver, International Diabetes Federation (Europe), European Heart Network, European Respiratory Society, European Society for Medical Oncology, European Kidney Health Alliance, Foundation of European Nurses in Diabetes, 2010.
(3) Loddenkemper R, Gibson GJ, Sibille Y. “European Lung White Book: the First Comprehensive Survey on Respiratory Health in Europe.” Sheffield, European Respiratory Society/European Lung Foundation, 2003; 16–23.
(4) M. Decramer, Y. Sibille, A. Bush, K-H. Carlsen, K.F. Rabe, L. Clancyf, A. Turnbull, B. Nemery, A. Simonds and T. Troosters. “The European Union conference on chronic respiratory disease: purpose and conclusions” Eur Resp J April 1, 2011 vol. 37 no. 4 738-742.
(5) Jemal A, Ward E, Hao Y, et al. “Trends in the leading causes of death in the United States,” 1970–2002. JAMA 2005; 294: 1255–1259.
(6) Guideline on the requirements for clinical documentation for orally inhaled products (OIP) including the requirements for demonstration of therapeutic equivalence between two inhaled products for use in the treatment of asthma and chronic obstructive pulmonary disease (COPD) in adults and for use in the treatment of asthma in children and adolescents. CPMY/EWP/4151/00 Rev. 1, 1 August 2009.
(7) S. Dissanayake. “Application of the EU Guidelines for Pharmacokinetic Studies of Locally Acting Orally Inhaled Drug Products” Respiratory Drug Delivery 2010, Vol 1, pp 293-304.
(8) Dennis O’Connor, B.S. et al. “Role of Pharmacokinetics in Establishing Bioequivalence for Orally Inhaled Drug Products: Workshop Summary Report” Journal of Aerosol Medicine and Pulmonary Drug Delivery Volume 24, Number 3, 2011, 119-135.
(9) Jones B, Kenward, MG. (1989) “Design and Analysis of Crossover Trials.” Chapman-Hall, London, England.
(10) American Thoracic Society. Standardization of spirometry - 1994 update. Am J Respir Crit Care Med 1995; 152:1107-36.
(11) Juniper, E.F. et al. (1994) “Histamine and Metacholine Inhalation Test: A Laboratory Tidal Breathing Protocol.” 2nd Ed. Astro Draco AB, Lund.
(12) Jokic, R, et al. “Methacholine PC20extrapolation.” Chest 1998; 114, 1796-1797.
(13) Lemière C, et al. “Airway inflammation assessed by invasive and noninvasive means in severe asthma: eosinophilic and noneosinophilic phenotypes.” J Allergy Clin Immunol 2006;118(5):1033-9.
(14) Pizzichini E. et. al. “Indices of airway inflammation in induced sputum: reproducibility and validity of cell and fluid phase measurements.” Am J. Respir. Crit. Care Med. 1996; 154: 308-17.
(15) Taylor DR, et al. “Exhaled nitric oxide measurements: clinical application and interpretation.” Thorax 2006;61:817 827.
(16) “ATS/ERS Recommendations for Standardized Procedures for the Online and Offline Measurement of Exhaled Lower Respiratory Nitric Oxide and Nasal Nitric Oxide, 2005.” Am J Respir Crit Care Med 2005; 17: 912 930.