CFW Newsletter - New Approaches to Therapy

MEDICAL/SCIENTIFIC TOPIC

New Approaches to Therapy

I hope that all of those affected by cystic fibrosis will take the opportunity to read the following summary of the article by Chris Boyd, (and the full article on line) even if it may be a bit complex for non-scientists, because it discusses all of the new approaches being studied worldwide. Published in this issue is a summary of these approaches and CFW will be publishing updates on developments as they become available.

We, like you, live in the "world of cystic fibrosis" all the time, and are totally involved in trying to make things better for those with CF. I am sure that every adult living with CF knows that each group of doctors and scientists is committed to its own favourite approach. The article by Boyd reminds us that there are lots of approaches, not just "gene therapy" or "ion transport modifiers" or "better antibiotics". Indeed, in the 25 years that I have worked on CF, the improvements have come from small advances in many different areas, each of which can be more or less important for any individual living with CF. Many of the children I knew with CF are now adults, working productively, and while no one would pretend that life with CF is a bed of roses (even 65 Roses), it is much improved on what it was, both in quality and in time.

As the initiator of the first liposome CF gene therapy trial, carried out in London in the early 1990s, I am (of course) disappointed that all of us underestimated the problems that we would encounter introducing a healthy gene into the lungs of a patient with CF. This "gene therapy" approach is only just beginning to yield positive results, and probably will have to be combined with new stem cell and genomic approaches, perhaps in infancy or even in the womb. However, we all can be pleased at the fact that better mucolytics, better antibiotics, better treatment for immuno-rejection and better anti-inflammatory drugs all help with therapy in the here and now. The CF charities and research groups should be complimented and supported for trying such a wide range of approaches, because this ensures that things will continue to improve for everyone living with CF, perhaps not dramatically, but steadily over the years to come.

This article is dedicated to the memory of Sam Hillyard, who originally commissioned it.

A Christopher Boyd PhD
UK CF Gene Therapy Consortium
Medical Sciences (Medical Genetics)
University of Edinburgh
Molecular Medicine Centre
Western General Hospital
Crewe Rd
Edinburgh
EH4 2XU UK
Updated January 2005

Summary
Great strides have been made in our understanding of the underlying basis of CF at the cellular and molecular level. This article examines, from the point of view of a scientist, how these advances are being translated into the clinic to treat CF lung disease, both through conventional drugs and gene therapy.

Although the life expectancy for patients has increased over thirty-fold since 1940, we still lack a truly effective treatment for CF in the airways. Why is this? One reason is that the root cause of the disease was unknown until the discovery of the cystic fibrosis trans-membrane conductance regulator (CFTR) gene by Lap Chee Tsui and colleagues in Toronto in 1989. It then became apparent that the primary molecular defect in CF was a faulty protein which normally acts as a chloride channel. But that was only the beginning: it has taken all the intervening years to build our present understanding of how this microscopic fault leads to the infected and damaged CF lung over time. Our knowledge is by no means complete, and many more details need to be worked out. The prevailing idea is that defective CFTR leads to absorption of water from the liquid covering the airways: this thin but sticky layer stops the cilia (tiny hairs that cover the inner lining of the air sacs) from beating and thus prevents mucus clearance mechanisms from operating effectively. As a result, inhaled particles become trapped on the airway surface for a long time, and bacteria are able to colonise and initiate lung disease.
There are two basic therapeutic approaches. We can either target the lung problems (eg. attack microbial infection or attempt to modify the mucus layer); or we can target the basic causes (eg. either coax the faulty CFTR protein into working or bypass it by gene therapy or other means). Therapies targeting the basic causes are more desirable for the treatment of CF, but it is undeniable that such treatments are difficult to develop.

Concept to clinic: a long haul

The greatest obstacle to the emergence of new drugs for any disease is time. The template for pharmaceutical research has evolved over many decades and falls into three stages: (i) drug development and testing, (ii) clinical trials (Phases I, II, III and IV), and (iii) approval by regulatory bodies. To translate a single drug from concept to clinic has traditionally taken fifteen years and cost around US$500 M. In addition, it must always be borne in mind that only a tiny proportion of drugs that enter development and testing pass muster and enter clinical practice. The good news is that modern techniques such as high throughput screening (HTS), and better coordination of clinical trials internationally may soon reduce the time span of the process to 7-10 years and reduce the cost several fold. Indeed, the US Cystic Fibrosis Foundation (CFF) is pioneering this approach in its Therapeutics Development Program started in 19981. Technological advances enabled the rapid testing of thousands of compounds per day. The CFF began to foster this technology in 2000 with grants direct to pharmaceutical companies (eg. Aurora Biosciences) and also direct grants to academic sites Alan Verkman's group, supporting major drug discovery and development programs. At the beginning of 2004, the CFF had an interest in 9 pre-clinical studies using animal models and 14 clinical trials with human subjects.

Symptomatic therapy in the lung can either be direct, targeting the bacteria themselves with antibiotics or vaccines, or indirect, limiting the effects of damage caused by bacteria or the inflammatory process. Here we consider both lines of attack.

The bacterium Pseudomonas aeruginosa is a major pathogen in the CF lung. Pseudomonas management is crucial in CF care, and many antibiotic-based approaches have been used to treat it. One of the problems of delivering antibiotics by normal routes is that undesirably high doses are required. In recent years, TOBI (Chiron Corporation), a nebulised form of tobramycin, has entered clinical practice and provides more effective Pseudomonas control by targeting the drug directly to the lung. Another antibiotic product, Colomycin (Pharmax), has also been used successfully in this way. The concern with all antibiotics is the emergence of bacterial resistance. Recent UK data suggest that resistance to the commonly used antibiotics in Pseudomonas from CF lungs is worryingly high: we need to prescribe antibiotics more prudently and monitor resistance patterns routinely in order to conserve the benefits of antibiotics in the long term.

Another interesting new approach to tackling Pseudomonas in CF -- still in its preclinical phase -- is DNA-based vaccination. If proven, infants could be injected with a DNA vaccine and gain long-lasting immunity to the capacity of Pseudomonas to infect the lungs.

Dealing with CF mucus
Viscosity or stickiness of the airway liquid layer is a key feature of CF, favouring the colonisation of bacteria and other pathogens. Ironically, the presence of large amounts of DNA released by damaged neutrophils (cells involved in lung defence that are present in large numbers in CF) worsens viscosity in the CF airways. The belief that therapies aimed at destroying DNA from damaged cells would reduce this viscosity and hence improve mucus clearance has led to the development and widespread clinical use of Pulmozyme (Genentech Corporation). Pulmozyme is a recombinant (DNA that has been created artificially.from two or more sources and incorporated into a single recombinant molecule) form of human DNAse delivered by aerosol inhalation.

A more recent contender for this type of treatment is Nacystelyn (NAL), a derivative of the food supplement N-acetyl cysteine which has had a history of therapeutic use in the CF lung. In 2002, an international group led by Dr App reported on a trial of the effectiveness of NAL (inhaled as a dry powder) for treating impaired mucus clearance. NAL was seen to increase the water content of mucus -- a desirable property -- as well as making it less sticky. There were no serious side effects but it remains to be seen how well NAL performs in more extensive trials. This approach, of making slight modifications to existing drugs to improve their effectiveness, is an important strategy in pharmaceutical research that can short-circuit some of the lengthy development times.

Stimulating alternative ion channels
Dr Richard Boucher's group in North Carolina is researching drugs that are able to stimulate chloride channels other than CFTR and hence substitute for its defective function. Results of a study of one candidate drug, INS37217, were reported in 2002. This drug is the latest of a series of so-called P2Y receptor activators which have the desired effect. What is new about INS37217 is that it appears to be more stable (and hence can maintain its activity for longer) in CF sputum than previous drugs of this kind. Intriguingly, it increases both mucus production and ciliary motion. Clinical evaluation of INS37217 is under way.

Inflammation is a key manifestation of CF, though its root causes remain unclear. Neutrophils (white blood cells and a primary defence against bacterial infection) are involved in the inflammatory response that produces enzymes called proteases whose main function is to damage bacteria. The airway epithelium (membranes lining the airways) produces anti-proteases to prevent proteases from biting back and damaging the lung itself. The huge excess of neutrophils in CF generates too much protease for the lung's anti-proteases to deal with, and lung damage unfortunately results. A therapy based on adding more of a natural anti-protease called alpha-1-antitrypsin (AAT) has been mooted as a way of restoring the balance. Recombinant AAT has been shown to work when delivered as an aerosol but this approach has many drawbacks. A more refined approach is being developed by Dr Pam Davis and colleagues, in which AAT is attached to a targeting protein such that (when delivered either as an aerosol or intravenously) the AAT is ferried to the surface of epithelial cells and secreted at the site of desired action. So far this has been tested successfully in laboratory culture of human tracheal sections. Other groups are working on alternative anti-protease strategies attacking the same inflammatory defect. This mode of therapy has considerable promise, but, should it reach the clinic, it is likely to be beneficial only in individuals with early-stage disease.

Recently there has been much interest in non-steroidal treatments of CF lung inflammation that avoid the side effects associated with corticosteroid treatment. The prominent non-steroidal anti-inflammatory drug ibuprofen has undergone a trial in CF patients with encouraging results: the rate of decline in lung function was reduced. However, there has been a report of undesirable toxicity when ibuprofen was used for CF in conjunction with an antibiotic of the aminoglycoside class. Clinicians will remain wary of using ibuprofen for CF until a means of avoiding such toxic interactions is thoroughly understood.
Finally, the anti-inflammatory properties of antibiotics called macrolides have been increasingly evaluated in CF. These drugs may work at several different levels. For example, the macrolide clarithromycin has been shown to disrupt the biofilm environment proposed to favour Pseudomonas growth in the infected lung. There has been a surge of interest in azithromycin in particular, whose use in CF was inspired by its success in treating diffuse panbronchiolitis. Several trials showing promising results on CF lung function have taken place. However, the mode of action of macrolides in CF is still not clear, though their anti-inflammatory properties are thought to dominate. The results with azithromycin have been encouraging enough that we are beginning to see its use recommended under certain conditions in CF patients as an adjunct to more established therapies. We are still some way from a consensus on its efficacy and long term randomised controlled trials with large numbers of patients are required to determine its efficacy and optimal dosage with precision.

The attraction of therapies that target CFTR is that they have the potential to compensate for all the symptoms associated with the defective protein. Almost all therapies of this kind are still in their early developmental stages. It is important to remember that many compounds that work well in laboratory experiments and simple model systems in many cases do not translate to useful drugs in living animals, due to toxicity and other undesirable effects.

CFTR as a drug target
Our detailed knowledge of CFTR is being applied in the search for therapeutic drugs that improve the function of defective CFTR or (as we have seen) stimulate the activity of other proteins that may substitute for it. Some drugs emerged early on in the CFTR story and are still invaluable for laboratory research, but have not progressed to clinical use. Genistein, an activator of CFTR, is a well-known example. Genistein could yet play a clinical role: a proposed trial of a combination therapy comprising phenylbutyrate and genistein at the Children's Hospital of Philadelphia is recruiting patients (NLM identifier NCT00016744).
In the last few years, Dr Freedman and others in the US raised the possibility that docosahexaenoate (DHA) which is a polyunsaturated fatty acid, given as a dietary additive, could be used to treat CF. This followed published evidence that the DHA/arachidonic fatty acid imbalance in CF cells could be corrected by this means. However, the work remains in the pre-clinical phase and initial excitement about this approach has declined significantly.

Most CF patients have the F508del variant of CFTR, so it is not surprising that a great deal of attention has been focused on the properties of this protein. So-called protein-mobilising drugs that mimic the beneficial effects of growth in the cold are being actively sought. Such a drug may not be sufficient, however, because when F508del protein does reach the cell surface in patients it is less functional than normal CFTR. We may need to find F508del activity-enhancing drugs as well.

There are several experimental drugs that appear to mobilise F508del-CFTR. These include MPB members of the benzo(c)quinolizinium family. Dr Bob Dormer and colleagues in Cardiff in collaboration with Dr Fred Becq's group in Poitiers have shown that treatment of F508del nasal cells with MPB compounds results in better delivery of CFTR to the cells' upper surfaces. Under certain conditions the same compounds were also able to stimulate chloride channel function.

Another drug with reputedly similar beneficial effects on F508del cells is thapsigargin, a calcium pump inhibitor, as reported in 2002 by Dr Marie Egan and colleagues at Yale. This work is somewhat at odds with other published data that suggest thapsigargin acts to inhibit proper processing of proteins. Dr Egan argues that some proteins, including F508del-CFTR, are exceptions to this rule. We await clarification of the effects of thapsigargin on F508del-CFTR with interest. A related compound, curcumin, which is a component of the natural food additive turmeric, was reported by Dr Egan and colleagues in 2004 to correct F508del-CFTR defects in mice. Unfortunately, in a Phase 1 study in patients, no effect on F508-del CFTR was seen. Furthermore, scientists in other laboratories have been unable to replicate the effects of curcumin in mice.

High throughput screening methodologies (see below) offer the possibility of vastly increasing our repertoire of protein mobilizing drugs.

Other mutant forms of CFTR
Not all attention is focused on F508del-CFTR: much research is devoted to discovering and analysing drugs that activate whole classes of mutant forms of CFTR. For example, Dr David Sheppard and colleagues in Bristol are studying phloxine B and similar compounds that belong to a family of commercially-used food dyes. In early studies, these compounds have been shown to activate and inhibit CFTR in cell culture systems. This class of drugs is particularly attractive because it is already in use in food, its lack of toxicity is already known which therefore offers the prospect that future trials may be less protracted.

As an example of HTS, Dr Alan Verkman and colleagues in San Francisco are undertaking a CFF-funded drug discovery program. In their first study, hundreds of flavone and benzoquinolizinium derivatives were screened in a special cell system designed to detect drugs that affect CFTR activity. Certain 7,8-benzoflavones were shown to be most active; one in particular was found to be a potent inhibitor of CFTR activity. While this drug is not applicable to CF patients, it will be a useful addition to the armory of drugs available for laboratory research. There is also a possibility that it could be used in the clinic to treat conditions such as cholera. After this promising start, Dr Verkman's group and others are scaling up HTS efforts by screening over 100,000 compounds, to search for drugs that can activate or mobilise defective CFTR. Much progress has been made, and an inhibitor of CFTR has been identified that may have a role in the treatment of cholera. Lead compounds that activate or mobilize CFTR in vitro are currently being evaluated in vivo. It is too early to say whether any of them will find clinical application.

As with many other genetic diseases, CF progression varies between individuals. The environment has a major effect on variability, but the genetic make-up of the patient is also important. Why, for example, do some F508del patients become severely ill early in life and others remain healthy into middle age and beyond? We know that there are genes for proteins that :
(i) interact directly with CFTR,
(ii) act as ion channels,
(iii) take part in the inflammatory response and
(iv) have direct anti-bacterial activity.
Such genes and others are potential modifier genes, affecting the severity of disease progression. As a hypothetical example, suppose there is a gene X responsible for an ion channel P. X might exist in two forms, one producing a more a more beneficial version of P than the other. Patients who happen to have inherited the more beneficial version of X might then experience less severe symptoms because the more active P is able to compensate in part, for the CFTR defect.

The mouse has proved to be an invaluable tool by use of which modifier genes can be rapidly identified using carefully-designed breeding programs. In addition, genomics and proteomics -- whole disciplines devoted to new ways of looking at the expression patterns of large numbers of genes and proteins simultaneously, facilitate the identification of such genes. Once a modifier gene has been identified in the mouse, its relevance to the disease in humans can be tested by analysing the equivalent human gene in families and especially in twins with CF. Any CF modifier genes and proteins confirmed as human-relevant will provide new drug targets and thus significantly widen the scope of CF drug development.

CF gene therapy consists of delivering normal CFTR genes (usually in the form of DNA) into the affected airway cells of patients where they can produce undamaged CFTR protein and hence restore cells to normal working. As soon as the CFTR gene was discovered, the concept of gene therapy for CF was enthusiastically promoted and many groups worldwide took up the challenge of developing it. It was thought that CF was an ideal disease to treat by gene therapy for two main reasons. Firstly, the affected tissue, the lining or epithelium of the lung airways, was readily accessible, and secondly, it was believed that replacement of only about 5% of normal CFTR function would be sufficient to correct the defect. Both these reasons remain valid, but early attempts to transfer the normal gene into patients (delivered either by a modified virus or by a synthetic carrier) proved disappointing. Trials provided evidence that the process was safe, that the gene was getting into lung cells, and that functional CFTR was being made: but production levels were low and the normal gene stopped working after a few days.
It is now clear that the lung epithelium presents a series of impressive barriers to gene therapy, barriers that have evolved over millions of years as defense systems to ward off invading pathogens and keep the lungs clear and healthy for gas conductance and exchange. As far as the epithelium is concerned, gene therapy complexes represent just another class of invaders that need to be repelled. The short-lived gene expression seen in early trials revealed one aspect of lung defense that has evolved to switch off the genes of malevolent intruders such as viruses. The CF lung is an even more formidable target, because the sticky mucus layer physically traps many kinds of gene therapy complex and prevents their efficient entry into cells.

The greatest challenge for CF gene therapy research therefore involves efficiently breaching these barriers. We have learnt much from viruses such as adenovirus, which have evolved efficient means of entering cells. Indeed, modified viruses have been the most used carriers (or vectors) for delivering normal CFTR in gene therapy. However, until we solve the problem of keeping the introduced gene working for a long time, we are going to need to deliver gene therapy complexes at regular intervals. The efficiency of the immune system (even in CF patients) at building up immunity over time means that viruses cannot currently be used as carriers for this purpose. Also, serious adverse effects have been reported recently in (non-CF) gene therapy trials involving viruses. Because of these and other perceived drawbacks of viruses, more effort is currently being expended towards developing non-viral or synthetic means of delivering the CFTR gene. The main drawback of non-viral delivery methods is their relative inefficiency, but new methods, including the addition of small pieces of protein that allow the complex to home in on epithelial cells, are being developed. By such means and others, we are beginning to increase delivery efficiency and expect to devise effective formulations in the foreseeable future.

Two gene therapy trials in the US have reported recently. In one, a Phase II trial, a recombinant adeno-associated virus carrying CFTR was administered to the maxillary sinuses of 23 patients with antrostomies (an opening in the sinus for the purpose of drainage). While the treatment was safe and well tolerated, most of the endpoint measurements showed no statistically significant changes, and it is difficult to conclude that the intervention had a positive effect.
In the other trial (Phase I), a proprietary non-viral formulation containing compacted DNA nanoparticles, was used to deliver CFTR to the noses of 12 patients. There were no significant side effects and interestingly, as in some earlier nasal trials, there was a trend towards correction of the potential difference defect in a proportion of treated patients. The results have just been published in full, confirming partial reconstitution of CFTR function. Further studies of this interesting formulation are being carried out.

The CF Trust has set up the UK CF Gene Therapy Consortium (UKCFGTC) which combines the research efforts of three leading CF gene therapy centres in Oxford University, London (Imperial College) and Edinburgh University with the express aim of developing effective formulations for clinical use. This ambitious initiative has allowed us to combine forces to evaluate and devise new approaches to CF gene therapy. We are focused on initiating a clinical trial in 2006, and pre-clinical testing of a range of gene transfer agents is under way. Other approaches being studied by the UKCFGTC, some more futuristic, are described below.

How can long term CFTR production be achieved?
As mentioned above, the first generation of gene therapy vectors functioned for only a short time. Consequently, improvement in vector design is necessary in order to establish long-term clinical benefit. My group in Edinburgh is constructing and testing large vectors with much more of the natural CFTR gene than is present in conventional vectors, in the expectation that more sustained expression (with a higher degree of cell and tissue specificity) will result. Currently we are beginning to improve delivery efficiency, and plan to test these vectors in vivo.
An important finding from early non-viral gene therapy trials was that the DNA used, being of bacterial origin, was recognised as alien by lung cells and triggered a specialised defence mechanism resulting in inflammation. We now understand what features of the DNA induce this mechanism. For immediate application, UKCFGTC scientists in Oxford are therefore developing small vectors with improved CFTR gene sequences that are designed to produce CFTR in a sustained and non-inflammatory fashion.

One idea being investigated is stem cell based therapy. Stem cells are progenitor (ancestor) cells that are believed to exist in most organs, including the lung, and divide when called upon to replace dead or dying cells. The aim is to isolate and grow stem cells from patients' lungs in the laboratory, during which time the normal CF gene can be inserted. Corrected cells could then be put back into the patient where they would establish themselves in the appropriate airway region. If sufficient numbers of corrected stem cells could be induced to bed down in this way, they could provide a permanent supply of epithelial cells producing normal CFTR, and hence alleviate the disease. We do not presently know how we might achieve this in practice, but spectacular recent advances in stem cell research lead us to believe that the approach is worth pursuing, although it is likely to be a longer term goal. We are encouraged by the success of bone marrow based gene therapy, which has proved efficacious in the treatment of babies suffering from a fatal condition known as X-linked SCID.

Electroporation
As we have seen, delivering DNA to lung cells is difficult. One way of providing the energy to force DNA into cells is a method called electroporation, in which short pulses of electric current are used to punch tiny holes in cells allowing DNA to enter. This has been used in the laboratory setting for years, but now researchers have shown that this method can work in vivo. The challenge is to develop suitably safe and effective means of carrying out procedures such as electroporation in the lungs of patients. The UKCFGTC is investigating whether electroporation is a feasible way of enhancing gene therapy delivery to the lung. This is just one of a variety of physical methods to aid in gene delivery being assessed by the UKCFGTC.

Rather than supplying an entire working copy of the CFTR gene, there has been much effort recently towards repairing the mutant gene itself. In brief, the concept is to deliver a small "patch" of normal CFTR DNA which could provide the template for the cell machinery to repair the mutated part of a resident CFTR gene. This method works to varying extents in the laboratory, but UKCFGTC and other scientists have found its efficiency in epithelial cells to be immeasurably low. While the method remains attractive in principle, a clear demonstration that efficiency in the target cells can be vastly improved must be found before we can revisit this approach to CF gene therapy.
A parallel approach by Dr Engelhardt and colleagues in Iowa has been to correct CFTR mRNA rather than DNA. (The cell makes a message or mRNA copy of the DNA of each gene which it then processes to make protein.) Using a highly complex procedure, they were able to correct the F508del mutation both in cells and in mice. The efficiency is currently too low for in vivo clinical application. However, it may find clinical application as part of a stem-cell based therapy.

Persuading big pharmaceutical companies to invest large sums into developing therapies for CF (which is, after all, a relatively rare disease) is difficult. CF's orphan status means that charity-funded and academic-based groups are bearing much of the burden of necessary work. Nevertheless, all the high-tech paraphernalia of drug and gene therapy development are being harnessed enthusiastically for the treatment of CF. The US CFF is actively promoting intensive research into new drug discovery. Candidate compounds with exciting properties have already emerged and more are certain to be produced in the near future. Drugs that mobilise F508del-CFTR in an active form to the cell surface in vivo are most eagerly awaited.

For all the reasons discussed above, gene therapy has proved to be more difficult to develop than initially hoped. In the last few years, however, there have been notable gene therapy successes for other diseases, and more focused research such as that being undertaken by the UKCFGTC will accelerate progress. Achieving efficient delivery to lung cells is the trickiest challenge, but one or more of the new physical delivery methods being assessed may help us reach the threshold of efficiency required to bring clinical benefit.

It is impossible to predict whether effective treatment of CF lung disease will eventually be drug-based or gene-based: both approaches have advantages and difficulties. Many believe that a multi-pronged treatment, combining drug and gene therapy elements, will succeed.

Organisations such as Cystic Fibrosis Worldwide and national CF societies have a major role to play in encouraging the multi-centre coordination of clinical trials that will be necessary to speed the assessment of all new therapies. They also provide valuable forums in which clinicians, scientists and patients can meet and exchange views on how to advance our treatment of CF.

Acknowledgements
Thanks are due to the UK CF Trust for supporting my research, and to colleagues in the CF research community (especially those in the UK CF Gene Therapy Consortium) for helpful discussions .
A Christopher Boyd PhD
UK CF Gene Therapy Consortium
Medical Sciences (Medical Genetics)
University of Edinburgh
Molecular Medicine Centre
Western General Hospital
Crewe Rd
Edinburgh
EH4 2XU UK
Updated January 2005

Editor’s notes:
For a list of references for this article please contact us at
editor@cfww.org

The following web pages are supplied for readers interested in further information about specific drug trials:

[click here]
for a printable version of this article

Article translated by: NAME
Contact:

See the complete list of translators who helped in this newsletter