You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 3 Next »

  1. Definition:
    1. The Microscopic Findings (MI) domain captures the microscopic evaluations/histopathology of the study.
    2. This domain should contain at least 1 record for every protocol-scheduled tissue for all subjects in the study (e.g., if an organ was examined and no pathological changes were present, it should have a record indicating "UNREMARKABLE"). Unscheduled tissues that were examined should also have a record. Subjects that were not scheduled for examination should not have records unless they were examined. This assumption supports the creation of incidence tables and statistical analysis on histopathological data.
    3. The MI dataset provides a record for each microscopic finding observed throughout the study.
  2. The date/time of the subject disposition in DS is the most relevant date for interpretation of microscopic observations and is used to populate MIDTC.
  3. Specimen definition:
    1. The protocol-scheduled organ/tissue for examination is described by up to 5 fields: MISPEC, MIANTREG, MILAT, MIDIR, and FOCID.
    2. MISPEC defines the base organ or tissue examined.
    3. MIANTREG should be used where applicable and further specifies a part or section of the organ/tissue specified in MISPEC, when that subregion is the targeted area for examination. Examples include the cortex of the kidney, when separated from the kidney medulla, or a study-specific sectioning of the organ (e.g., top section of left liver lobe), but not a case where the liver is examined as a whole, but a specific finding is found for one of the lobes.
    4. For a paired organ, the organ used for the specimen should be specified as left, right, or bilateral, using the MILAT variable.
  4. Result definition:
    1. In MIORRES, a finding should comprise only 1 base pathological process and its modifiers (e.g., severity, chronicity, distribution, characteristics). However, it is recognized that data may not have been captured in this way; Example 2 shows a way to handle this situation.
    2. When MIORRES is populated, there must be an entry in MISTRESC. Other relevant components of the MIORRES finding must be parsed into 1 or more of these variables: MISEV, MIDISTR, MICHRON, and the supplemental qualifier MIRESMOD, which are used in combination to standardize the value in MIORRES.
    3. MISTRESC: This variable is important for standardizing the value in MIORRES and where possible must use the controlled lists NONNEO and NEOPLASM. If a microscopic finding in a tissue includes 2 related processes, then it can be described by a combination term that has both terms entered, separated by a "/" with no spaces. For example, features of degeneration and regeneration may be observed in a tissue as a continuum of the pathology. The processes can be identified separately or as part of a combined process of degeneration/regeneration. When the process of degeneration/regeneration is used to describe both components, this base process should be recorded in MISTRESC in a single row. The most common combination terms have been included on the NONNEO codelist. Other combination terms that represent 2 related processes can be constructed using preferred terms on the NONNEO codelist, separated by a "/" with no spaces. Terms should not be combined for processes that are unrelated (e.g., NECROSIS and CYST). Unrelated processes should be presented in 2 separate rows.
    4. The variables MISTRESC, MIDISTR, and MICHRON use CDISC Controlled Terminology derived from INHAND (International Harmonization of Nomenclature and Diagnostic Criteria for Lesions in Rats and Mice), a collaboration among international societies of toxicological pathologists.
    5. The supplemental qualifier MIRESMOD is used to further qualify the finding recorded in MISTRESC. MIRESMOD must be populated if 1 or more modifiers were part of the result in MIORRES and not otherwise reported in the modifier variables part of the MI domain structure (e.g., MISEV, MIDISTR, and MICHRON). In addition to describing observed characteristics of a lesion (e.g., cell type), this variable may be used to identify a specific region affected by the finding within the specimen. For example, when the liver is examined as a whole, but a finding is noted for a specific lobe, this location goes into MIRESMOD. Note that MIANTREG qualifies the excised specimen, whereas MIRESMOD qualifies the particular base pathological process identified within that excised specimen. In general, values for MIRESMOD should be separated by a semicolon unless they only make sense together. An example of an anatomic term is "bile duct"; this cannot be split into "bile" and "duct." An example of a term that should be separated is "centrilobular hepatocytic," which contains 2 terms that stand independently ("centrilobular" and "hepatocytic"). See the SUPPMI examples for further guidance on the use of MIRESMOD. 
      Unable to render {include} The included page could not be found.
    6. The use of MIRESMOD does not preclude sponsors from creating other supplemental qualifiers containing specifically defined modifiers. It is currently expected that MIRESMOD will contain the complete list of modifiers not contained in standard variables (e.g., MISEV, MIDISTR, MICHRON) regardless of their being part of sponsor-defined supplemental qualifiers.
    7. Tumor findings should have a record in this domain, even if they also have records in the TF domain. It is, however, not required to populate MIRESCAT for tumor findings in the MI domain.
      1. When MIORRES contains a tumor finding the corresponding term from NEOPLASM (CDISC Controlled Terminology list) should be used to populate MISTRESC.
      2. For TF domain: When MISTRESC contains a tumor finding, the corresponding term from the NEOPLASM CT list should be used to populate TFSTRESC. Additional variables populated in the MI domain (e.g., MICHRON, MIDISTR, MIRESMOD) are not populated in the TF domain, since they are not needed to create the tumor.xpt file.
    8. Expectations of when MISTAT is "NOT DONE":
      1. If an organ (scheduled for histopathology or introduced later because of adverse findings) for some reason was not examined, the record will have a blank value in MIORRES and MISTAT will be "NOT DONE". NOTE: Failure to adhere to this assumption will preclude generation of the tumor.xpt dataset.
      2. Use of MISTAT and MIREASND: Whenever MISTAT is "NOT DONE", MIREASND should provide the reason for not completing the evaluation as described in the study plan.
    9. For microscopic evaluations that have numeric results (e.g., specific cell count tests), MISTRESN and MISTRESU should be included, as well as other applicable variables for the Findings observation class to reflect the data accurately.
  5. The MISPID variable is intended to reflect the mass identification. This variable should be used to link in-life findings (e.g., mass identification) with pathology findings. The mass identifier in --SPID should be consistent across domains (Clinical Observations, Palpable Masses, Macroscopic Findings, MI, and TF).
  6. The value in FOCID should have semantic value; that is, although "1" is not considered adequate, "Injection site 1" is acceptable.
  • No labels