Critical Path Institute Project Pipeline

The hippocampus is a small brain structure key to storing and retrieving memories. Patients with Alzheimer's Disease (AD) show early and progressive loss or atrophy of the hippocampus. Quantitative magnetic resonance imaging (MRI) is a noninvasive way to measure hippocampal atrophy and is used to assess disease progression in AD. Reductions in hippocampal volume represent the only correlate of cognitive impairment over time amongst all biomarkers studied to date.

It is believed that effective disease modifying therapies will need to be administered to patients very early in the course of the illness. To date, clinical trials in patients with Mild Cognitive Impairment have posed significant challenges since many patients that present with cognitive impairment do not have true Alzheimer's disease. CAMD is advancing a qualification regulatory biomarker application that proposes that baseline measurement of low hippocampal volume by MRI in patients with episodic memory deficit is a useful means of predicting those patients more likely to evolve to AD dementia during the course of an AD clinical trial. This biomarker would aid in the selection of a cohort which will yield a relatively uniform clinical trial population more likely to progress to AD dementia.

Therapeutic intervention at an early, asymptomatic or mildly symptomatic stage of Alzheimer's Disease (AD) is likely required to preserve cognitive function and delay disease progression. Attempts to identify patients at increased risk for progression to AD dementia based on clinical assessment have lacked desired diagnostic sensitivity and specificity. Accordingly, three National Institute on Aging and Alzheimer's Association (NIA-AA)-sponsored workgroups recently recommended revised diagnostic guidelines for AD dementia, mild cognitive impairment (MCI) due to AD, and the definition of preclinical stages of AD. The proposed guidelines rely upon combined clinical and biomarker-based patient assessment and offer a framework for research on the relationship between the pathophysiology and the emergence of clinical symptoms.

In alignment with these guidelines, cerebrospinal Fluid (CSF), which may be acquired via lumbar puncture, is important as a biofluid signature of what is happening in the brain. The CSF biomarkers being advanced for qualification by CAMD are as follows:

-cerebrospinal fluid (CSF) 42 amino acid-containing isoform of amyloid beta(Aβ42)

-total tau (t-tau)

-tau phosphorylated at threonine 181 (P-tau181)

CAMD's qualification effort provides evidence to support the use of CSF Aβ42, t-tau and/or P-tau181 as biomarkers that can identify cognitively impaired subjects who are likely to progress to AD pathology for inclusion in research and drug registration clinical trials of candidate therapeutics for AD.

A quantitative model that describes the natural history of cognitive change in Alzheimer's Disease (AD) would be an enormously valuable tool in the development of novel therapies. Utilizing the extensive historical data available in the literature and from participating members, CAMD has developed such a model which can be used to simulate and test relevant characteristics of clinical trial designs for different hypothesized expected effects of a drug. The model is expected to yield useful information that can be incorporated into trial design such as dose selection, population inclusion, sample size estimates, and study duration.

In recent years, it has become clear that there is a very long prodromal period of Alzheimer's Disease (AD) pathology that precedes onset of clinical symptoms by as much as 20 years. It is currently believed that novel disease modifying therapies will be efficacious only if one can initiate treatment early in the course of disease. To date, significant challenges have occurred in assessing effects of therapies in mild cognitive impairment (MCI) patients, largely due to the difficulty of identifying which patients will progress to true AD as opposed to other forms of dementia. Advances in pharmacometrics pose an exciting and promising avenue to address this issue. Following the successful development of its disease model of mild and moderate AD, CAMD is currently investigating the development of a modeling and simulation tool based on a quantitative model of early MCI subjects. Analysis of data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database to date suggests that patients at very early stages of dementia/MCI progress at a slower rate and show greater interindividual variability. As a result, the disease model would be essential in demonstrating the number of subjects and duration of treatment that is needed to test new therapies in newly diagnosed patients. This tool would be used to design and execute clinical trials in this area. Specifically, the tool would be used to inform clinical trial design in subjects with amnestic mild cognitive impairment likely to progress to AD dementia. The goal is for this work to serve as a resource for teams designing clinical trials across the industry. It is intended that teams will utilize this simulation tool to allow them to provide a quantitative rationale for selection of study design and inclusion criteria for specific compounds and for internal decision making during all clinical stages of development.

Approximately a million U.S. patients suffer from PD, a number that is expected to increase by 40% over the next 30 years. As the scientific community and pharmaceutical industry move to develop innovative disease modifying agents to augment the symptomatic treatment currently available, the need for accurate diagnosis of PD at a stage earlier than is currently feasible becomes a public health imperative. One of the main challenges for the disease modification approach is that by the time the disease is clinically apparent, the neurodegenerative process is far advanced. CAMD proposes to address this need through the qualification of a biomarker for PD, specifically, a reduction of dopamine transporter protein (DAT) activity. The SPECT (single photon emission computed tomography) ligand, N-xfluoropropyl-2b-carbomethoxy-3b-[4-iodophenyl] nortropane (FP-CIT) (DaTSCAN™ GE Healthcare) is currently approved for the diagnosis of PD to distinguish patients from those with essential tremor. The current proposed context of use is to expand the label to exclude all patients that have SWEDDS (scans without evidence of dopamine deficiency) from clinical trials of new candidate PD therapies. This imaging biomarker will be used to identify at the earliest stages of motor impairment patients with dopamine deficiency for the purposes of enrichment in clinical trials.

Following review of the first ever biomarker qualification data submission by PSTC, the FDA (2008), EMA (2007), and PMDA (2010) issued a formal regulatory opinion that urinary beta2-microblobulin, urinary total protein, urinary albumin, urinary KIM-1, urinary clusterin, urinary cystatin c, and urinary trefoil factor 3 (TFF-3) can be utilized on a voluntary basis, in addition to serum creatinine and blood urea nitrogen, in GLP rat toxicology studies to monitor drug-induced kidney injury.

In Phase II of PSTC's nonclinical qualification, the seven previously qualified biomarkers, plus serum cystatin c, urinary retinol-binding protein 4 (RBP-4), GST-alpha, GST-pi/mu, NAG, NGAL, osteoactivin, and osteopontin, are being further investigated for their mechanistic specificity, their potential to detect very early kidney injury (i.e. before microscopic evidence of injury to the tissue), their utility in monitoring injury reversibility, and their specificity to drug-induced renal injury.

Several putative drug-induced kidney biomarkers, including those currently under investigation in the rat model, are being evaluated for their utility in monitoring patient renal safety in a series of studies that will characterize inter- and intra-subject variability, as well as the impact of gender, age, and diurnal variation in healthy volunteers. In collaboration with the FNIH Biomarkers Consortium, PSTC will also determine thresholds for drug-induced renal tubular injury. Biomarkers to be evaluated include: urinary albumin, urinary creatinine, urinary alpha-1-microglobulin, urinary beta-2-microglobulin, urinary calbindinD28, urinary clusterin, urinary cystatin c, urinary kidney injury molecule-1 (KIM-1), urinary IL-18, urinary total protein, urinary trefoil factor 3 (TFF3), serum cystatin c, urinary NAG, urinary GST-alpha, urinary GST-pi, urinary RBP4, urinary NGAL, urinary osteopontin, urinary Tamm-Horsfall Protein (uromodulin), urinary VEGF, and urinary connective tissue growth factor (CTGF). Newly qualified biomarkers will enable safer drugs to be developed in order to reduce drug-induced injury in patients.

Muscle pain is an unfortunate common side effect of treatment in popular classes of drugs, including cholesterol-lowering statins. AST and creatinine kinase activity (CK), traditional measures of drug-induced skeletal muscle injury, lack both specificity and sensitivity. Skeletal troponin I (Tnni1, Tnni2), skeletal troponin T (Tnnt1, Tnnt3), creatinine kinase protein M (Ckm), parvalbumin (Pvalb), myosin light chain 3 (Myl3), fatty acid binding protein 3 (Fabp3), aldolase A (Aldoa), and myoglobin - all measured in serum or plasma - are proposed as more sensitive and specific biomarkers of drug-induced skeletal muscle injury, that additionally may distinguish between injury to fast and slow twitch muscle fiber types in rats. Adoption of these newer biomarkers should allow therapeutics with fewer muscle side-effects to be developed.

Drug-induced skeletal muscle injury is an undesirable side effect of some valuable medications, including cholesterol-lowering statins and fibrates. Serum ALT and creatinine kinase, as well as clinical outcomes reported by patients and physicians such as muscle pain and weakness, are imprecise and nonspecific. New biomarkers including skeletal troponin I (Tnni1, Tnni2), skeletal troponin T (Tnnt1, Tnnt3), creatinine kinase protein M (Ckm), parvalbumin (Pvalb), myosin light chain 3 (Myl3), fatty acid binding protein 3 (Fabp3), aldolase A (Aldoa), and myoglobinare are under investigation for their potential to more specifically and accurately detect and monitor drug-induced skeletal muscle injury.

The most common traditional biomarkers of drug-induced liver injury, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), have several limitations. In the course of testing a therapeutic for potential to cause liver injury, transaminase increases are commonly observed in the absence of evidence of injury to tissue, and, conversely, sometimes do not increase even when tissue injury is observed. PSTC is pursuing qualification of several biomarkers that improve upon this lack of concordance, as well as biomarkers that may truly predict whether a drug-induced injury will progress to a serious outcome (i.e. liver failure) or whether adaptation may occur.

First set: GLDH, MDH, PON-1, PNP
Second set: ARG-1, SDH, GST-alpha
Third set: miR-122

Drug-induced liver injury (DILI) is one of the most frequent adverse events and can sometimes result in liver failure. Rare (e.g. < 1/10,000) liver safety events have resulted in the withdrawal of drugs from the market that were efficacious for the overwhelming majority of patients. Identifying which patients might be susceptible or succumb to DILI is an active and complex area of research. PSTC is focused specifically on biomarkers that would predict early in the course of therapeutic treatment (i.e. earlier than meeting a Hy's Law definition) whether a patient might progress to serious liver injury or adapt, and thus safely continue with therapeutic treatment. These biomarkers would be enormously useful in deciding whether to continue or alter the course of treatment, while simultaneously improving patient safety and enabling continued therapy for patients in need.

Set 1: GLDH, MDH, SDH, GST-alpha, ARG-1, miR122

Some therapies can cause cardiac hypertrophy leading to left ventricular systolic dysfunction and possibly heart failure. These drug-induced changes are most commonly monitored in patients utilizing echocardiography (imaging), and more recently, natriuretic peptides (NPs), cardiac hormones synthesized and secreted in response to myofiber stretch. The non-invasive, rapid, and inexpensive nature of NP measurements make them attractive candidates for both nonclinical species and patients. PSTC is investigating whether NT-proANP, the active, cleaved form of atrial NP, is an appropriate marker for measurement of drug-induced changes in heart weight and decreased ejection fraction that, when elevated in nonclinical safety studies, could trigger utilization of clinical BNP measurements to monitor drug-induced hemodynamic stress leading to decreased ejection fraction and/or cardiac hypertrophy. This work will protect patients from potentially serious cardiac side-effects from therapeutic treatment.

Drug-induced vascular injury (DIVI) is common in preclinical species such as the rat, dog, or monkey and is classically associated with a class of drugs called PDE inhibitors, including many given safely and effectively to humans for a variety of conditions and diseases. However, observations of histopathology are dissimilar among these species and even more different than histopathology observed in humans. Moreover, compounds known to cause DIVI in preclinical species do not appear to cause vascular injury in humans. PSTC is pursuing biomarkers in rats that are both sensitive and specific to drug-induced injury to the epithelial or smooth muscle cells of the vasculature, and can be demonstrated of mechanistic relevance for humans.

Set 1: VEGF, GRO/CINC-1, TIMP-1, vWGpp, NGAL, TSP-1, smooth muscle alpha actin, calponin, transgelin

PSTC is working to qualify non-invasive biomarkers for drug-induced vascular injury (DIVI) that can be used to monitor safety in nonclinical and clinical drug development. A major confounding factor is that DIVI seen in preclinical species is very dissimilar to histopathological observations in man. Additionally, sensitivity/susceptibility to DIVI from the same therapeutic varies widely from animals to humans. Thus observations of DIVI in nonclinical safety studies usually lead to termination of the program, as no mechanism for monitoring this injury currently exists for use in human clinical trials. This likely leads to the loss of many promising therapies that may have posed no risk for DIVI to humans. PSTC's new biomarkers should enable the initiation of clinical trials where patient safety can be confidently monitored.

Check back soon

The clinical course of PKD is marked by a long period of stable kidney function (as measured by glomerular filtration rate (GFR) and serum creatinine, the currently accepted endpoints for clinical trials in drug development) during enormous expansion of the kidney due to cyst growth. Serum creatinine levels only rise once the kidneys have incurred serious, irreversible damage. Use of serum creatinine or GFR as an endpoint in clinical trials has thus impeded the development of therapeutics that have the potential to slow or stop cyst growth at an early stage of disease. The PKD Outcomes Consortium is developing an extensive database and creating a disease model that will support the use of imaging of total kidney volume (TKV) 1) as a prognostic biomarker of the rate of disease progression, 2) to select patients likely to respond to therapy into clinical trials, and 3) as an endpoint to predict clinical benefit of drug therapy. Qualification will mean that TKV can be used in proof of concept studies to identify therapeutics of likely benefit to PKD patients.

There is an acute need for therapeutics that can slow or halt the progression of Alzheimer's disease, and it is currently believed that therapeutic intervention is likely necessary at a very early stage of the disease. Assessment of the patient's experience of the subtle cognitive changes that have functional impact on daily activity may be a critical tool for measuring effects of a potential therapeutic being tested in a clinical trial. The Cognition Working Group seeks to develop outcome measures that improve upon the current measurement of mild levels of cognitive impairment (MCI) due AD, as well as that capture the patient's perspective on relevant outcomes that would contribute to detection of disease, description of disease progression, and the measurement of treatment effect. These instruments would be used to support labeling claims of new AD therapies.

Development of therapies to treat irritable bowel syndrome (IBS) depends upon critical communication from the patient about symptoms such as abdominal pain and bloating that cannot be measured any other way. There is an abundance of literature on available outcome measures in IBS, however, the development and validation methods used to define these measures are not rigorous enough to meet FDA's PRO Guidance (2009). While there are many questionnaires that have been used in IBS research, there is not one instrument that fulfills all the criteria of the PRO Guidance in its development or validation, and the IBS Working Group has undertaken the additional work needed to create a new PRO for labeling purposes.

Despite a number of effective, safe therapies to treat asthma available, a high proportion of patients remain symptomatic and at risk for exacerbations. No standard PRO instrument currently exists for asthma that is fit for measuring important patient-experienced aspects of the disease. Such an instrument could be used in addition to clinical measures like spirometric assessment of lung function in the development of drugs for the treatment of persistent asthma. The Asthma Working Group aims to develop an asthma diary that would enable capture of asthma symptoms such as coughing, wheezing, trouble breathing, and chest tightness during both day and night.

Depression is a highly prevalent and under-treated condition in the U.S., and worldwide. Despite the availability of safe and efficacious medications, a high proportion of patients do not achieve remission of symptoms even after switching treatments. Effective measurement of symptom improvement from a patient's perspective is imperative for the development of novel therapies. The Depression Working Group will evaluate existing, and potentially develop (as needed), PRO measures for their appropriateness as an efficacy endpoint in clinical trials for major depressive disorder. Depression symptoms likely to be captured include those related to dysphoria (e.g. sadness and irritability), physical/somatic symptoms (e.g. appetitie change, low energy), sleep-related symptoms (e.g. insomnia and somnolence), and cognitive symptoms (e.g. difficulty concentrating, restlessness).

Lung cancer is the most common cancer worldwide and non-small cell lung cancer (NSCLC) is the most common type of lung cancer. Substantial unmet medical need exists in this patient population, and the survival advantage of different regimens may be, unfortunately, small. Thus, detection of therapeutic impact that can palliate debilitating symptoms such as persistent cough, chest pain, and recurring pneumonia or bronchitis is critical. Additionally, non-pulmonary symptoms include appetite loss, sleep disturbance, lack of energy, and pain. The objective of the NSCLC Working Group is to develop a symptom inventory that can be used to support labeling claims.

Functional dyspepsia is defined by the absence of any organ, systemic, or metabolic disease likely to explain symptoms thought to originate in the gastroduodenal region. It is estimated that up to 40% of the general population experience this condition, however estimates range from 3%-40% - this broad range reflecting the variability in diagnostic criteria used in clinical studies. The Functional Dyspepsia Working Group is developing instruments to assess symptoms related to the two functional dyspepsia subtypes: Postprandial Distress Syndrome (PDS) and Epigastic Pain Syndrome (EPS). These instruments will be utilized to assess therapeutic relief of patient symptoms and to support labeling claims.

Rheumatoid Arthritis (RA) affects about 1% of the population in Western countries and is characterized by painful inflammation of the joints that leads to physical disability. The objective of the RA Working Group is to develop PRO instruments that will support the evaluation of treatment benefit in medical product clinical trials for patients with mild to severe RA, with the intention to support claims in product labeling. Current instruments were developed prior to FDA's 2009 PRO Guidance and lack inclusion of concepts deemed important to patients with RA, such as fatigue or physical tiredness.

The CPTR Biomarkers & Clinical Endpoints Workgroup is evaluating all relevant research regarding this tool and will identify potential opportunities for qualification by regulatory agencies.

Pooled data from clinical trials performed by CPTR Drug Coalition member companies and other sources will be used to create robust clinical quantitative disease-progression models for TB regimen development. Additional data on relevant biomarkers (e.g. laboratory tests, imaging parameters, microbiologic assays, etc.) and clinical endpoints will be progressively incorporated into the models.
The primary use of the clinical disease-progression model will be to inform clinical trial design. Such an approach can be used to characterize and quantify natural disease progression, placebo and drug effects, and inform dose selection as well as trial execution variables (e.g., patient discontinuation rates, dosing schedules, compliance, specific design, etc.) from multiple trials using patient-level data. Models can increase efficiency and decrease risk of errors in drug development decisions by overcoming the complexity and uncertainties of the disease-drug-treatment interaction.

The Preclinical Sciences work group will proceed with qualification of the hollow fiber model as a major advancement for early and effective evaluation of drug combinations. This in vitro model allows for the evaluation of drug combinations under relevant conditions that mimic the various disease states of the TB organism. Drugs and their combinations can be evaluated under dynamic dosing conditions consistent with human paradigms allowing for the rational design of confirmatory in vivo studies and support early clinical dosing strategies when partnered with in silico PKPD modeling approaches.