Documented use cases with analysis approach

From PHUSE Wiki
Jump to: navigation, search

What is our goal?
Do we want to predict findings in early clinical development or post-marketing findings?

Potential Use Cases / Scenarios

  • Drug-Induced Liver Injury (DILI) - A Use Case Scenario for Hepatoxicity: DILI (drug induced liver injury) is the most prominent adverse events to cause warnings or withdrawals, or to cause a compound to fail in late stage clinical trials (eg. A wide variety of compounds).
  • Cardiovascular Toxicity – A Use Case for Evaluating Cardiac Toxicity and Safety Pharmacology: Cardiac toxicity is the 2nd most common reason for revised labeling or recall decisions in the post-marketing phase; eg withdrawals or safety warnings of approved drugs (Avandia, Avastin, statins, etc).
  • Drug-Induced Acute Kidney Injury – A Use Case Scenario for Nephrotoxicity: Drug toxicity is an important cause of acute kidney injury and has been implicated in upto 25 % of acute renal failure in critically ill patients.
  • Most predictive animal models - Determine more specifically which animal models are most predictive of human outcomes, across e.g. particular therapeutic areas , or compound classes, or target classes
  • Pathology predictors - Use current and historic data, and translational informatics methods, to evaluate non-clinical pathology outcomes (adverse event, adverse clinical chemistry, etc.) as predictors of clinical outcome
  • Understand class effects - Use the database in the drug development process to answer questions of what others have seen in certain classes of drugs. This investigation could done at any point in the process to verify or compare results; acting as a guidepost. This could be done from a non-clinical and clinical standpoint.
  • Male reproductive toxicity - Translation of male reproductive effects from rodents to humans, particularly when the effects in rodents are slight or when there are testicular/sperm effects in rodents that don’t appear to impact rodent fertility.
  • Bone marrow toxicity - Explore the predictability of subtle but significant changes in bone marrow (WBC and RBD production) observed in preclinical species and the translation of those effects to humans.
  • Rodent bone findings - Do findings in bone growth/structure in rats translate to humans? Does the age of the rat or the human matter?
  • False positives - Identify nonclinical endpoints (or panels of endpoints) that regularly or consistently are considered false positives once human data is available
  • False negatives - Which clinical findings do nonclinical studies regularly / consistently miss? Are there signals in the nonclinical data that we are missing? This analysis could also focus on clinical endpoints for which there is currently no known nonclinical measure (e.g. headaches, taste changes).
  • Nonvalue-added studies - Are there nonclinical study types that are consistently negative, or when they are positive, it is regularly a false positive?
  • Vaccines - Extrapolate the results based on animal models to design Phase I human trials for adjuvanted vaccines
  • Update Olsen methodology
  • Neurotoxicity - A lot of nonclinical and clinical data is available, but the correlation is questionable
  • Dermal Toxicity - Limited nonclinical and clinical data and correlation is unclear
  • Biologics vs. Small Molecules - Questionable relevance of nonclinical models; however, it would be useful to determine if toxicity is usually on-target or off-target


  • Jon Kimball
  • Jane Reed
  • Tim Gartner
  • Dave Seyler
  • Paul Cornwell
  • Paul Bradley
  • Martha Lee
  • Jeremy Wally
  • Dave Seyler


FDA comments are an informal communication and represent the individual's best judgment. These comments do not bind or obligate FDA. The contents of this wiki are from the individual contributors and do not necessarily reflect the view and/or policies of the Food and Drug Administration, the employers of the individuals involved or any of their staff.

Last revision by Pacorn,07/16/2012