The Changing Landscape of Cystic Fibrosis Clinical Trials

Less than 70 years ago, cystic fibrosis (CF) was uniformly fatal in the first year of life. The inherited genetic disease, which affects between one in 2,500 and one in 3,200 live births in Caucasians, causes the body to produce a thick, sticky mucus. In the lungs, the mucus can cause potentially life-threatening infections; in the pancreas, it blocks the absorption of food and nutrients.

Today, thanks to the development of drugs that improve nutrition, mucociliary clearance, and antimicrobial therapy, CF patients usually live to between 40 and 50 years of age, with some patients surviving into their 70s. Yet, the disease remains chronically debilitating and eventually life-threatening. Patients survive much longer, but they are plagued by bronchiectasis and chronic pulmonary disease, which negatively impact their quality of life.

Organizations like the Cystic Fibrosis Foundation (CFF) have been instrumental in raising awareness for the disease, which helps generate funding for new research and furthers development of clinical practice guidelines. Thus, clinical trials continue to optimize the clinical management of CF patients, which address the complications but not the underlying cause of the disease, and to find new treatment modalities.

Before setting out to conduct a CF trial, it is important for sponsors to recognize the key challenges, as well as understand the latest standards and guidelines so they can design an appropriate study that takes into account all the variables and unique considerations associated with CF patients.

Improving CF standards

Pancreatic enzyme supplements were developed before the US Food and Drug Administration (FDA) required standardized testing in clinical trials. Like a number of older medications, they were later "grandfathered" onto the market and continued to be sold despite the fact that they had never been formally approved.

In April 2004, the FDA issued a new rule requiring makers of existing pancreatic enzyme products to obtain the agency's approval. FDA established a date of April 28, 2010 for the makers of pancreatic enzyme products to stop manufacturing and distributing unapproved products. The European Medicinal Agency (EMA) has also stated in its guidelines on the clinical development of medicinal products for the treatment of cystic fibrosis that there is a need for the development of age-appropriate formulations of pancreatic enzymes preparations and of bile salts, including acid-resistant ones.¹ EMA considers that the clinical data on the addition of bile salt therapy and improvement of mucosal transport are still needed and would be of great interest for CF patients with on-going steatorrhoea despite optimal standard treatment.

The reason for the rule was to ensure that the older medicines met the same standards of testing as any other new drug. In particular, the FDA wanted to ensure that the supplements contained the right amount of active ingredients to digest food. There were a variety of pancreatic enzyme products on the market, and they differed in what they contained, how they were made, and how many pills must be taken. The FDA did not know exactly how much of the active ingredient was in each product, which was critical because inconsistencies in enzyme formulation can cause problems with digestion.

To gain FDA approval, manufacturers of pancreatic enzyme supplements must confirm their products' safety and effectiveness by testing them in clinical trials in patients with pancreatic diseases, including CF. Manufacturing processes must be standardized, ensuring the consistency of the capsules from batch to batch.

Once the clinical trials are complete, the FDA determines if each company's pancreatic enzyme products should be approved. Because of the importance of pancreatic enzymes for patients with CF, the FDA allowed the old medicines to be sold until the new rule went into effect. On March 1, 2012, two new pancreatic enzyme products used to aid food digestion, Ultresa (pancrelipase) and Viokace (pancrelipase) became the fourth and fifth pancreatic enzyme products to be approved by the FDA. Other FDA-approved pancreatic enzyme products include Creon (2009), Zenpep (2009), and Pancreaze (2010). Approved pancreatic enzyme products meet FDA standards for safety, efficacy, and product quality.

While some companies work to gain FDA approval of their enzyme replacement therapies, other companies are focused on developing new treatments for CF. The discovery of the CF gene in 1989 has led to an explosion of research in the field.

Clinical trials in cystic fibrosis investigate a widening array of novel therapeutics addressing the basic underlying defect in CF. According to CFF there are approximately 30 potential drugs in development today for the treatment of cystic fibrosis. The Table 1 summarizes new approaches in CF drug development.

Designing CF clinical trials

Clinical studies have been an essential mechanism for amplifying knowledge of the disease's pathophysiology and response to therapies. However, trials for both existing and investigational CF drugs offer unique challenges.

According to the CFF, there are only about 30,000 CF patients in the United States and perhaps 70,000 worldwide.² The limited number of patients make enrollment into a clinical trial difficult. CF patients are generally in favor of participating in clinical trials, but have become more selective due to the number of trials offered to them. Additionally, some regions lack sophisticated equipment needed for these types of clinical trials.

There are some obvious steps a sponsor can take to optimize enrollment into a clinical trial, such as focusing trial sites in geographic areas where CF is most prevalent, including the United States, Europe, and some parts of Latin America. Phase II and III usually require a global reach to CF patients that can only be achieved through global international studies, but most patients who are diagnosed can be located on a global basis since they are locked into secondary, tertiary, and quaternary care.

Best practices to attract the best patients

Advocacy group support. Especially in developed countries, sponsors can turn to patient support and advocacy groups, such as the CFF in United States or the ECFS-Clinical Trial Network (ECFS-CTN) in Europe, for assistance with the site identification and protocol review. The main value of CF clinical trial networks is to identify the centers that fulfill a certain set of criteria, such as the number of patients, human resources, experience in clinical trials, and the infrastructure necessary to support a clinical trial. These groups may be able to provide site recommendations, as well as provide a way to increase awareness of a trial directly with patients through their websites and newsletters.

As with every chronic disease, networks and patient groups are increasingly viewed as valued partners. Patient associations are very influential since they provide information on the nature and progression of the disease and on the latest treatments and research on fighting it directly to patients, as well as to parents of patients with CF. Apart from serving as educational resources, they also facilitate forums for patients with CF to communicate with one another.

Building an online presence. Another strategy for direct patient outreach is to build an online presence for the trial. CF patients are extremely knowledgeable about their disease, and many spend significant time on the Internet researching their treatment options. Thus online outreach through one of several good, CF clinical trial networks can be an effective way to increase trial awareness among these patients, particularly in regions where Internet usage is high. Advertisements, reaching out to sites directly or working with a CRO that has investigator contacts in the region provide alternate strategies for attracting patients in less developed countries.

Pediatric considerations. One difficulty affecting CF trials is that about half of CF patients are children, requiring a special pediatric approach and oversight in the clinical trial process. Sponsors need to understand that in pediatric trials, the patient is the entire family. The trial design must accommodate their various work and school schedules, or it will have difficulties with accrual.

In general, the single most important step a sponsor can take to increase enrollment in a CF trial is to make the design of that trial attractive to patients and their families. While this is not always possible due to the need to appropriately address key safety and efficacy issues, the protocol should provide a rationale for some specific design, and the appropriate information should be entered into the informed consent documents. Additionally, the protocol schedule of events and assessments should be accommodating to school and work schedules, the study assessments should be minimally invasive and whenever possible, the research assessments should coincide with the usual standard of care assessments.

Sponsors must also take into account that there are limits on the amount of blood that can be drawn from pediatric patients, not to mention that trials with fewer blood draws and less invasive assessment procedures tend to be more appealing to patients. Another unique consideration that sponsors need to be aware of is that CF patients cannot skip enzyme replacements, which are often taken with meals, so they cannot be kept waiting in a doctor's office for extended periods of time. CF patients must also be isolated from other patients waiting at the doctor's office or hospital to minimize the risk of being exposed to infections.

The essential part of a successful enrollment begins with a process of informed consent. Special considerations should be taken for pediatric clinical trials in cystic fibrosis including disclosing information to parents and providing age-appropriate information to children. In general, children with cystic fibrosis are often very familiar with the background and medical terminology used to describe their disease. However, parents need details about the study requirements and time commitment, while children need materials that accurately reflect what will be asked of them in language they will understand. Recognizing the perspective of families is critical to successful support of site efforts to recruit and retain children in trials.

Since in the majority of cases, CF trials are international multicenter studies, there should be clear and meticulous protocol requirements that do not allow any type of different interpretations. In particular, the inclusion/exclusion criteria as well as specific study assessment procedures should be very precise. The induced sputum procedures or standards for pulmonary function tests might differ between sites. The same applies for any type of nebulizers or physical therapy.

Specifics of CF patients and diverse patient profiles. Another challenge of CF trials is that patients vary widely in their genetic and phenotypic profiles as well as in their disease stage. There is no "ideal patient profile" upon which to base decisions. Patients with CF also have a different pharmacokinetic profile compared to other patient populations. Studies have shown that CF patients have an increased volume of distribution and faster elimination of drugs, mainly due to an increased renal clearance, or a more rapid metabolism in the liver. This means that patients with CF usually require higher doses of medicines, and extrapolation of data from other patient groups is usually not possible.

Subjects with CF frequently have adverse events (AEs) during clinical trials at relatively high rates, even in the placebo arms of the studies, which can complicate the interpretation of the safety profile of the therapy being tested. This is especially true for small Phase I and Phase II trials, which sometimes do not even include a placebo arm. Some authors have published tables that can assist in safety monitoring in CF clinical trials involving novel inhaled drugs. These tables can be used to determine the expected number of AEs for a study and whether the number of AEs observed in a particular category is unusually large.³

Patients with CF frequently have laboratory values outside the normal range and have significant longitudinal variability of laboratory values. However, clinical trials in CF currently use laboratory specific reference ranges that are based on a non-CF population. Interpretation of AEs in the clinical trial setting may be complicated by the underlying high rates of some laboratory abnormalities in the CF population.

The baseline laboratory values of otherwise clinically stable CF subjects are often abnormal. That is why caution should still be maintained when interpreting abnormal safety laboratories obtained during the conduct of clinical trials in CF patients. Hepatotoxicity remains a potential significant safety concern for any novel therapeutic agent. CF patients may be at increased risk of hepatotoxicity given the high rate of underlying liver disease related to CF.⁴

The selection of the equation references for the pulmonary function tests represents another challenge in CF clinical trials. In one cross-sectional analysis, the choice of reference equation affected the distribution of subjects classified as having mild, moderate, or severe lung disease. CF physicians should be aware of the impact of choice of reference equation in both clinical care and research.⁵ Pulmonary function tests in preschool children represent another difficulty due to lack of active cooperation in this age group. Some authors suggest the use of pulmonary function tests that do not require active cooperation, such as expiratory interrupter resistance (Rintexp) values and functional residual capacity measurements since these measurements might help to follow young children with CF who are unable to perform reproducible forced expiratory maneuvers.⁶

Additionally, pediatric CF patients can be a diverse group in and of themselves. Trials might focus on newborns, infants, or adolescents, and each group will have a different metabolism to consider. Beyond the biological and scientific differences, there are regulatory differences to consider as well. For example, in addition to the informed consent forms that parents must sign, children may receive an assent form that varies depending on the age group and may need to be translated into pictures for very young CF patients. Assent is defined as an agreement by an individual not competent to give legally valid informed consent to participate in research.

Finally, the therapeutic successes in CF have created some additional challenges to the development of future CF therapies. In particular, statistically significant changes in pulmonary function and exacerbation rate become more difficult to demonstrate as baseline population values improve. A systematic approach to outcome measures development is needed to provide regulators with the tools necessary for evaluating new therapies. That is why the EMA has recently organized a workshop with a focus on outcome measures for CF lung disease and exocrine pancreatic insufficiency, with a view to informing regulatory decisions.

Recent CF guidelines for trial design

There is no single optimal trial design for CF, but useful guidelines were recently published in the Journal of Cystic Fibrosis.⁷ The guidelines stated that a parallel group design is often preferable to a crossover design in CF clinical trials. Crossover trials have the benefit of fewer patients, where each patient can act as his/her own control, and the data are generally less variable. However, this design can take longer based on the length and number of treatments and washout periods.

There is also the risk of a patient having an exacerbation during one of the arms of the study and discontinuing participation in the trial. Additional arguments against crossover designs include carry-over effects, seasonal variation, and patient inconvenience of multiple visits, as well as the inherent instability of CF lung disease.

More ethics committees and IRBs are requiring that CF clinical trials have an active control. If a placebo-controlled trial is mandated, or a medication has to be taken out of the patient's normal regimen, enrollment could take longer.

Non-inferiority studies are often not appropriate for CF trials because the number of subjects necessary in such trials is usually larger than the number of subjects required for conventional parallel group trials. Depending on the drug profile, the three arm non-inferiority study design with the placebo, might not be ethical for CF patients and not feasible for enrollment, therefore it is important that the difference from the comparator be carefully defined. For most drug developers, such a risk is not a good use of limited resources. The exception would be if a trial is comparing changes in formulation for a directly competing treatment-for example, new inhaled versus oral antibiotics.

A true equivalence study is usually set up as a parallel group randomized, controlled trial, though it can be a crossover design. Shorter term, placebo-controlled trials, allowing establishment of efficacy, followed by longer open label experience for safety evaluation, may be an acceptable compromise.

Additional useful guidelines for designing CF trials were published in May 2008 by the EMA.¹ The document was the first official guideline on developing products to treat cystic fibrosis and contains many recommendations that are applicable in enhancing the design of all CF trials.

For example, the EMA advises drug makers to use forced expiratory volume in one second (FEV1) as the primary lung-function endpoint when designing clinical trials of drugs aimed at improving CF lung function. Such drugs might include those intended to prevent, thin, or clear mucus from the airways; those intended to hydrate mucus; or those intended to reduce lung inflammation. For slowing or stopping pulmonary disease progression, a 12-month FEV1 endpoint is recommended.

For drug makers focused on fighting acute and chronic lung infections, the EMA recommends stratifying patients at inclusion according to the severity of their pulmonary impairment based upon respiratory function tests, and evaluating a minimum of a six-month primary endpoint assessing the respiratory function through FEV1 measurement, with a 12-month follow-up for safety. A microbiological primary endpoint at 28 days is acceptable for confirmatory trials in the treatment of early lung colonization or of chronic infection exacerbations.

Randomized active-controlled confirmatory trials are mandatory when a reference treatment exists. When no reference treatment exists, a placebo-controlled study in mild to moderate patients on top of best supportive care is recommended.

There is increasing evidence that CT scoring should also be used and is more sensitive than pulmonary function tests for the detection of relevant disease progression in CF.⁶ Bronchiectasis, which is progressive and irreversible in CF, is probably the most relevant structural change on CT scans that can be scored reliably. CT measurement of airway wall thickening is also possible. Airway wall thickening is related to inflammation; thus, this endpoint is of significance for interventional studies that include anti-inflammatory drugs.⁸

For drug makers focused on improving pancreatic endpoints, such as those making pancreatic enzyme replacement therapies, the EMA recommends standardization of the patient's specific diet, on a patient-per-patient basis. Placebo-controlled superiority confirmatory trials in the frame of add-on studies are mandatory (on top of standard therapy). The primary efficacy criterion should be clinical: target height at 12 months and normal weight at six months in children, weight gain or nutritional status at six months (changes in body weight, weight/height and lean body mass) in adults.

A biological endpoint (steatorrhoea or protein synthesis) can be accepted as a short-term primary endpoint in confirmatory trials, preferably in the frame of cross-over design due to the high level of inter-individual variability.

Some drugs currently in development are aimed at overcoming the CFTR mutation that underlies CF, such as with gene therapy, to replace the defective gene or by correcting the function of the defective CFTR protein. The EMA notes that such drugs would be expected to demonstrate clinical improvement in both pulmonary and pancreatic disease. Both the FDA and the European Commission have approved Kalydeco (ivacaftor)-the first medicine to treat the underlying cause of cystic fibrosis, in people with a specific genetic mutation (G551D)-to treat a specific subgroup of patients with CF. Kalydeco targets a gene mutation that only occurs in about 4% of CF patients. Academia also invests a lot in a gene CF therapy. For example, a consortium of researchers from Oxford University, Imperial College London, and the University of Edinburgh launched a clinical trial where the gene therapy they have developed will be delivered using an inhaler.

Funding CF clinical trials

The nonprofit CFF has contributed hundreds of millions of dollars to support clinical trials of new drugs for CF. The group pioneered a shift in the role of patient advocacy organizations, which previously gave a majority of their money to academic research grants. The CFF has instead focused on working with drug companies to fund clinical trials.

Between 1998 and 2007, the CFF committed more than $250 million to biotechnology companies. Projects funded by the organization include Vertex Pharmaceuticals' VX-770, a CFTR potentiator in Phase III clinical trials; Alnara Pharmaceuticals' liprotamase, a porcine-free pancreatic enzyme replacement therapy that has completed Phase III trials and led to a recent acquisition by Eli Lilly and Co.; and PTC Therapeutics' ataluren, a Phase III drug designed to override the nonsense mutation in CF.

Drug makers get the added benefit from CFF funding because it is non-dilutive. Additionally, the CFF provides access to its clinical trial network, patient tissue samples, assays, and expertise in the field.

Probably the largest CF study and a worldwide first in terms of the length of the study, the number of patients involved, and the number of doses of gene therapy that is ongoing in the United Kingdom is funded by the British Medical Research Council (MRC) and the National Institute for Health Research (NIHR).⁹

Clinical Research Organizations CROs, too, are starting to work with drug companies to make CF trials more affordable. Many CROs have instituted risk-sharing business models, tying their fees to the delivery of value-creating milestones, offering volume discounts, and sometimes taking a financial stake in the development of the product. Additionally, CROs can provide deep therapeutic expertise in CF and access to a broad network of CF investigators and thought-leaders. Some global CROs are able to provide therapeutic expertise in clinical trials for cystic fibrosis. An internal team of scientists and drug development professionals often have proven success with a broad range of study designs and indications with various CF treatment modalities.

A bright future

Cystic fibrosis is one disease that clearly demonstrates the value and benefit of clinical research as evidenced by the significant improvement in life expectancy and quality of life. CF clinical trials are complex and require a multidisciplinary approach involving different stake holders from an early stage in the drug development program.

Due to a limited number of patients and a relatively large number of competing trials, it is essential to make the appropriate protocol design and have the right site selection. In addition, the changing demographics of patient population and evolving standards of care bring additional challenges when creating protocol designs. The authorities in the United States and the European Union are providing new requirements and guidelines that should help further development of new drugs for patients with cystic fibrosis.

Alexander Cvetkovich-Muntanola, MD, is a pediatrician and has worked in CF clinics, and is the Executive Director, Clinical Development at INC Research, 3201 Beechleaf Court, Suite 600, Raleigh, NC, e-mail: alexandar.cvetkovichmuntanola@incresearch.com.

References

1. EMA, "Guideline on the Clinical Development of Medicinal Products for the Treatment of Cystic Fibrosis," (2009), http://bit.ly/mDCwlr.

2. Cystic Fibrosis Foundaion, "About Cystic Fibrosis," http://www.cff.org/AboutCF/.

3. H. Sucharew et al., "Respiratory Adverse Event Profiles in Cystic Fibrosis Placebo Subjects in Short- and Long-Term Inhaled Therapy Trials," Contemp Clin Trials, 27 (6) 561-570 (2006).

4. C. Goss et al., "Laboratory Parameter Profiles Among Patients With Cystic Fibrosis," Am J Rep Crit Care Mes, 165 (8) A283 (2002).

5. M. Rosenfeld et al., "Effect of Choice of Reference Equation on Analysis of Pulmonary Function in Cystic Fibrosis Patients," Pediatr Pulmonol, 31 (3) 227-237 (2001).

6. N. Beydon, et al., "Pulmonary Function Tests in Preschool Children with Cystic Fibrosis," Am J Respir Crit Care Med, 166 (8) 1099-1104 (2002).

7. G. Döring et al., "Clinical Trials in Cystic Fibrosis," Journal of Cystic Fibrosis, 6, 85–99 (2007).

8. H. A. Tiddens and P. A. de Jong, "Imaging and Clinical Trials in Cystic Fibrosis," Proc Am Thorac Soc, 4 (4) 343-346 (2007).

9. Medical Xpress, "Largest Gene Therapy Trial for Cystic Fibrosis Begins," (2012), http://bit.ly/yPZHCE.