Saturday, February 27, 2016

Rare Disease Clinical Trials: Fast, Small and Decisive

"The subphenotyping of COPD [chronic obstructive pulmonary disease] into separate groups based on mechanism sets the stage for the rational development of therapeutics." - Stephen Rennard (1)

Modern clinical trials had great success in late 1960s and early 1970s, when highly effective chemotherapeutic agents were found to be effective against a wide range of rare, childhood cancers. The prospective randomized control trial, performed on children with cancer, was so very successful that it served as a requirement and a standard for drug testing, for the past half-century.

Today, large, randomized prospective clinical trials are the standard for common diseases, such as cancer. The problem has been that none of the drugs tested on adults with cancer have had the kind of curative successes that we saw with the childhood tumors. Larger, longer and increasingly expensive studies were conducted to demonstrate incremental improvements in chemotherapeutic regimens. Though there have been successes in clinical trials for the common cancers occurring in adults, no trial on common cancers has yielded the spectacular successes witnessed for the rare childhood cancers.

Rule - Clinical trials are the best method ever developed to determine whether a drug is safe and effective for a particular purpose, in a particular target population. Nonetheless, aazzclinical trials cannot provide the clinical guidance we need to develop all of the new medications that will be needed to conquer the common diseases.zzaa
Brief Rationale - We simply do not have have the money, time, and talent to perform all the anticipated clinical trials for the common diseases.

Modern clinical trials are long and expensive. It takes about 10 to 15 years for an experimental drug to be developed (2). Only 5 in 5,000 compounds that have preclinical testing will enter clinical trials (2). The cost of developing a drug and bringing it to market is about $1 billion (3).

Clinical trials can be very large. In the realm of cancer trials, the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO, NIH/NCI trial NO1 CN25512) serves as an example. The PLCO is a randomized controlled cancer trial. Between 1992, when the trial opened, and 2001, when enrollment ended, 155,000 participants were recruited (4). The study will end in 2016.

It can be difficult or impossible to enroll all the patients required for a clinical trial. In an analysis of 500 planned cancer trials, 40% of trials failed to accrue the minimum necessary number of patients. Of cancer trials that have passed through preclinical, phase I clinical and phase II clinical trials, three out of five failed to achieve the necessary patient enrollment to move into the final phase III clinical trial (5). Most clinical trials for cardiovascular disease, diabetes, or depression are designed to be even larger than cancer trials (5).

Overall, about 95% of drugs that move through the clinical trial gauntlet will fail (6). Of the 5% of drugs that pass, their value may be minimal. To pass a clinical trial, a drug must have proven efficacy. It need not be curative; only effective. Of the drugs that pass clinical trials, some will have negligible or incremental benefits. After a drug has reached market, its value to the general population might be less than anyone had anticipated. Clinical trials, like any human endeavor, are subject to error (7), (8), (9). Like any human endeavor, clinical trials need to be validated in clinical practice (3). It may take years or decades to determine whether a treatment that demonstrated a small, but statistically significant effect in a clinical trial, will have equivalent value in everyday practice.

Funders of medical research are slowly learning that there simply is not enough money or time to conduct all of the clinical trials that are needed to advance medical science at a pace that is remotely comparable to the pace of medical progress in the first half of the twentieth century.

Rule - Clinical trials for common diseases have limited value if the test population is heterogeneous; as is often the case.
Brief Rationale - Abundant evidence suggests that most common diseases are heterogeneous, composed of genotypically and phenotypically distinct disease populations, with each population responding differently with the clinical trial.

The population affected by a common disease often consists of many distinct genetic and phenotypic subtypes of the disease; essentially many different diseases. A successful clinical trial for a common disease would require a drug that is effective against different diseases that happen to have a somewhat similar phenotype. One-size-fits-all therapies seldom work as well as anticipated, and more than 95% of the clinical trials for common diseases fail (6).

Rule - Clinical trials for the rare diseases are less expensive, can be performed with less money, and provide more definitive results than clinical trials on common diseases.
Brief Rationale - Common diseases are heterogeneous and produced a mixed set of results on subpopulations. This in turn dilutes the effect of a treatment and enlarges the required number of trial participants. Rare diseases are homogeneous, thus producing a uniform effect in the trial population, and lowering the number of trial participants required to produce a statistically convincing result.

Rare diseases often have a single genetic aberration, driving a single metabolic pathway, that results in the expression of a rather uniform clinical phenotype. This means that a drug that succeeds in one patient will likely succeed in every patient who has the same disease. Likewise, a drug that fails in one patient will fail in all the other patients. This phenomenon has enormous consequences for the design of clinical trials. When the effects of drugs are obvious, the number of patients enrolled in clinical trials can be reduced, compared with the size of clinical trials wherein the effects of drugs are highly variable among the treated population. In general, clinical trials targeted on rare diseases or on genotypically distinct subsets of common diseases require fewer enrolled participants than trials conducted on heterogeneous populations that have a common disease (6).

Consider the following story:
Molybdenum cofactor deficiency type A is a metabolic disorder leading to seizures and progressive brain damage in affected infants. There was no known treatment, and about 100 babies died of this disease, worldwide, each year. Gunther Schwarz, a plant biologist in Germany, developed a compopund which could, in theory, compensate for the cofactor deficiency. The compound had been successfully tested in mice, but not in humans. When a an infant with the disease was born in Australia, Dr. Scwharz sent all his avaialable compound to the baby's doctor, Alex Veldman, at Monash Medical Center. Taking into consideration the infant's worsening condition, the bioethical board approved the experimental treatment. Within minutes of treatment, the baby's condition greatly improved (10), (11), (12).

It is easy to assume that because rare diseases affect fewer individuals than do the common diseases, it would be difficult to recruit a sufficient number of patients into an orphan drug trial. Due to the energetic and successful activities of rare disease organizations, registries of patients have been collected for hundreds of different conditions. For the most part, patients with rare diseases are eager to enroll in clinical trials. The rare disease registries, made available to clinical trialists, eliminates the hit-or-miss accrual activities that characterize clinical trials for common diseases.

In an effort to increase the scientific and clinical value of clinical trials, trialists often include ancillary studies in their trial designs. These ancillary studies may consist of molecular studies on tissue biopsies obtained from trial subjects. Using biopsy samples, different responses to a treatment can be correlated with a genetic marker or a genetic profile measured on tissues. In instances for which rare disease organizations collect and store biopsies obtained from their registered members, ancillary studies for orphan drug trials can be performed quickly, and with less expense than comparable studies on common diseases.

In the U.S. several laws have been passed to encourage and facilitate clinical trials for the rare diseases:

Public Law 105-115, FDA Modernization Act of 1997, grants an exemption for orphan drugs from drug approval application fees that would otherwise apply (13). Amendments to the Act, in 2007, include the Best Pharmaceuticals for Children Act (Public Law 110-85), which encourages the recruitment of children into clinical trials.

Public Law 111-80, the Agriculture, Rural Development, Food and Drug Administration, and Related Agencies Appropriations Act of 2010zzaa authorized the FDA to appoint a review group to recommend design improvements for preclinical and clinical trials aimed at preventing, diagnosing and treating rare diseases (13).

Thee U.S. Food and Drug Administration, is poised to provide guidance to organizations and corporations conducting clinical trials on orphan drugs (14). It is crucial that trial sponsors stay in close touch with FDA staff during the planning stages of drug trials. A little advice from a regulator can avoid the heartbreak that comes when an effective drug fails approval due to poor trial design.
Trials on orphan drugs commonly accrue human subjects from vulnerable populations (e.g., children, mentally impaired subjects, subjects with multiple life-threatening conditions). In such cases, human subjects may not be able to provide informed consent, and a parent or guardian will need to be consulted (See Glossary item, Informed consent). Trialists must be sensitive to the special needs of their subjects and their families. Recruiting an independent clinical safety board or institutional review board, with no financial ties to the trialists or their sponsors, is a prudent measure (14).

Rule - Clinical trials on common disease can be reduced to one or more trials of a subtype of the disease.
Brief Rationale - The heterogeneity of populations with a common disease allows trialists the freedom to design small trials, for subsets of individuals who have a particular genotype (i.e., a gene marker or a gene expression profile), a particular mode of inheritance (i.e., Mendelian), or a distinguishing clinical phenotype (e.g., early onset disease).

A recurring theme in these blogs is that common diseases are collections of genotypically distinct diseases that share a common phenotype and common disease pathways (1), (15), (16). If there is some reason to expect a drug to be particularly effective against a defined subset of individuals with a common disease, it may be worthwhile to design the trial for these individuals. The pharmaceutical company Genentech employed this strategy when it developed the breast cancer drug trastuzuab (trade name herceptin). Trastuzumab is a monocloncal antibody against the HER2 receptor ). In this case, preclinical evidence indicated that trastuzumab might be effective against breast cancers that had high levels of HER2. By limiting their study to individuals with HER2-positive breast cancers, the company achieved success, with a relatively small number of trial participants (6).

It is easy to find rare diseases that pose as variant subsets of common diseases (e.g., B-K mole syndrome patients composing a subset of individuals at high risk of developing melanoma; BRCA gene positive individuals as a subset of individuals at high risk of breast cancers; patients with alpha-1-antitrypsin deficiency as a subset of emphysema cases). A clinical trial specifically aimed at a rare subset of a common disease might facilitate later trials directed at other subsets of the same disease.

Such clinical trials are in progress. The I-SPY 2 trial matches treatments against subgroups of breast cancer patients whose tumor cells match particular molecular profiles (6). In the I-SPY 2 trial, multiple drugs are tested on relatively small, selected subgroups of cancer patients. As results are collected, unsuccessful drugs are phased out and replaced by other drug candidates, all within the same trial (6).

Rare Disease Day is coming up February 29 (a rare day for rare diseases). In honor of the upcoming event, I'll be posting blogs all month, related to the rare diseases and to rare disease funding.

- Jules Berman (copyrighted material)

key words: rare disease, orphan drugs, orphan diseases, zebra diseases, rare disease day, disease complexity, common diseases, clinical trials, improving clinical trials, ancillary clinical trials, jules j berman


[1] Rennard SI, Vestbo J. The many "small COPDs", COPD should be an orphan disease. Chest 134:623-627, 2008.

[2] Orphan Drugs in Development for Rare Diseases; 2011 Report. America's Biopharmaceutical Research Companies. Available from, viewed July 14, 2013.

[3] Berman JJ. Principles of Big Data: Preparing, Sharing, and Analyzing Complex Information. Morgan Kaufmann, Waltham, MA, 2013.

[4] Prostate, lung, colorectal & ovarian cancer screening trial (PLCO) Available from:, viewed August 22, 2013.

[5] English R, Lebovitz Y, Griffin R. Forum on Drug Discovery, Development, and Translation. Institute of Medicine, 2010.

[6] Leaf C. Do clinical trials work? The New York Times. July 13, 2013.

[7] Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, et al. The STARD statement for reporting studies of diagnostic accuracy: explanation and elaboration. Clin Chem 49:7-18, 2003.

[8] Ioannidis JP. Why most published research findings are false. PLoS Med 2:e124, 2005.

[9] Ioannidis JP. Some main problems eroding the credibility and relevance of randomized trials. Bull NYU Hosp Jt Dis 66:135-139, 2008.

[10] Schwahn BC, Van Spronsen FJ, Belaidi AA, Bowhay S, Christodoulou J, Derks TG, et al. Efficacy and safety of cyclic pyranopterin monophosphate substitution in severe molybdenum cofactor deficiency type A: a prospective cohort study. Lancet 15:00124-00125, 2015.

[11] Schwarz G, Santamaria-Araujo JA, Wolf S, Lee HJ, Adham IM, Grone HJ, et al. Rescue of lethal molybdenum cofactor deficiency by a biosynthetic precursor from Escherichia coli. Hum Molec Genet 13:1249-1255, 2004.

[12] Donovan S. Dying baby cured in world first. ABC News 5 Nov 5, 2009.

[13] Field MJ, Boat T. Rare Diseases and Orphan Products: Accelerating Research and Development. Institute of Medicine (US) Committee on Accelerating Rare Diseases Research and Orphan Product Development. 2010. The National Academics Press, Washington, D.C. Available from:

[14] Wizemann T, Robinson S, Giffin R. Breakthrough Business Models: Drug Development for Rare and Neglected Diseases and Individualized Therapies Workshop Summary. National Academy of Sciences, 2009.

[15] Crow YJ. Lupus: how much "complexity" is really (just) genetic heterogeneity? Arthritis and Rheumatism 63:3661-3664, 2011.

[16] Wade N. Many Rare Mutations May Underpin Diseases. The New York Times May 17, 2012.