Methods and systems for detecting usual interstitial pneumonia
Inventors
Kennedy, Giulia C. • Huang, Jing • CHOI, Yoonha • Pankratz, Daniel • Walsh, Patric Sean
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11976329-B2
Publication Date
2024-05-07
Expiration Date
Abstract
The present disclosure provides systems, methods, and classifiers for differentiating between samples as usual interstitial pneumonia (UIP) or non-UIP.
Core Innovation
The invention relates to a method for identifying whether a subject is positive for usual interstitial pneumonia (UIP) or non-UIP. The method includes pooling lung tissue samples from a subject suspected of having UIP to produce a pooled lung tissue sample, assaying the pooled lung tissue sample for a level of expression of one or more markers associated with UIP or non-UIP, and processing the assayed expression to generate a UIP versus non-UIP classification that is then output.
The assaying is performed for expression of gene markers selected from a specified set of markers associated with UIP or non-UIP, including gene marker panels described in the document and transcript/gene selection sets referenced to tables and enumerated gene/ENSG identifiers. The processed level of expression is used to generate the classification of the pooled lung tissue sample as positive for UIP or non-UIP.
The disclosed approach also accounts for smoker status and smoker-status bias in relation to classification by determining an additional expression level of one or more markers associated with a subject’s smoker status and using that additional expression level to determine the classification. Further refinements include use of a weighted algorithm that excludes smoker-status-bias-associated marker expression and tiered use of smoker versus non-smoker classification before UIP versus non-UIP classification.
Claims Coverage
The claim coverage indicates one independent method claim directed to pooling lung tissue samples and classifying UIP versus non-UIP based on expression of UIP or non-UIP-associated gene markers, followed by outputting the classification. It also includes dependent inventive features involving performance thresholds and smoker-status-related marker handling.
Pooling and UIP/non-UIP classification from pooled lung tissue expression
Pooling lung tissue samples from a subject suspected of having UIP to produce a pooled lung tissue sample; assaying the pooled lung tissue sample for a level of expression of one or more markers associated with UIP or non-UIP; processing the level of expression to generate a classification of the pooled lung tissue sample as being positive for UIP or non-UIP; and outputting the classification.
Marker panel gene expression for UIP versus non-UIP
Assaying comprises measuring expression of one or more markers associated with UIP or non-UIP, wherein the one or more markers comprise a gene selected from the enumerated set including MPO, GGNBP2, SELE, FMO3, SLC6A13, EXTL3, and the other listed genes.
Performance constrained classification
The classification is generated with a specificity of at least about 90% and, in another refinement, with a sensitivity of at least about 70%.
Smoker-status expression integration and smoker-status bias handling
Determining an additional expression level of one or more markers associated with a subject’s smoker status and using that additional expression level to determine the classification; and using a weighted algorithm that excludes the additional level of expression of smoker-status-bias-associated markers.
Across the provided claim set, the inventive concept centers on producing a pooled lung tissue sample, assaying pooled-sample gene marker expression for UIP versus non-UIP, processing that expression into a UIP/non-UIP classification, and outputting the result, with refinements that constrain classification performance and incorporate smoker-status-related expression and smoker-status bias handling.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?