Morphometric detection of dna mismatch repair deficiency
AssigneeVisiongate, Inc.
InventorsDaniel J. Sussman, Michael G. Meyer, Randall Mastrangelo, Alan C. Nelson
Life of this patent
Estimating the term…
Abstract
A method to develop one or more morphometric classifiers to identify a mismatch repair deficiency (MMRD). The method provides a non-invasive method of characterizing MMRD that is responsive to a tumor in its early stages of development and irrespective of the tumor size. The method allows targeting cancer therapy to the specific characteristics of the cancer that the patient may have, allowing more efficient cancer management with far fewer side effects.
Claims17
A method for developing at least one more morphometric classifier to identify cells exhibiting MMRD, the method comprising:
- deriving selected clones from transduced cells;
- analyzing the selected clones for MLH1 expression to identify a subset of MLH1 expression clones:
- screening the subset of MLH1 expression clones to identify those with MLH1 expression levels reduced in comparison to a parental cell line to identify a second subset of the selected clones;
- expanding the second subset of the selected clones in culture;
- harvesting the second subset of the selected clones;
- determining an MMRD level;
- analyzing the selected clones on a 3D microscopy optical tomography system;
- and
- comparing the selected clones to a set of control transduced cell lines.
5-28. (canceled)
A classification cascade method for training classifiers adapted to identify specific mutations associated with different cancer types comprising:
- training a first classifier to group data from malignant cell lines and assign the data to a first class including a set of malignant cells;
- training a second classifier to separate malignant subtypes by first isolating small cell lung cancer from adenocarcinoma and squamous cancer and then isolating adenocarcinoma cell types from squamous cancer;
- training a third classifier to isolate Adeno:
- SW900 from the adenocarcinoma cell types and pass a first set of remaining cell types to a fourth classifier,
- training the fourth classifier to isolate Adeno:
- ALK+, NCI-H2228 cell types from the remaining cell types and pass a second set of remaining cell types to a fifth classifier;
- training the fifth classifier to isolate Adeno:
- Wild type, A549 from EGFR+ Adeno cell types and pass EGFR+ Adeno subtypes from the second set of remaining cell types to a sixth classifier, and
- training the sixth classifier to isolate Adeno:
- T790M, NCI-H1975 from the Adeno:
- EGFR-p.E746_A750del.
A classifier for identifying specific mutations associated with different cancer types comprising:
- a first classifier trained to group data from malignant cell lines and assign the data to a first class including a set of malignant cells;
- a second classifier trained to separate malignant subtypes by first isolating small cell lung cancer from adenocarcinoma and squamous cancer and then isolating adenocarcinoma cell types from squamous cancer;
- a third classifier trained to isolate Adeno:
- SW900 from the adenocarcinoma cell types and pass a first set of remaining cell types to a fourth classifier,
- the fourth classifier trained to isolate Adeno:
- ALK+, NCI-H2228 cell types from the remaining cell types and pass a second set of remaining cell types to a fifth classifier;
- the fifth classifier trained to isolate Adeno:
- Wild type, A549 from EGFR+ Adeno cell types and pass EGFR+ Adeno subtypes from the second set of remaining cell types to a sixth classifier, and
- the sixth classifier trained to isolate Adeno:
- T790M, NCI-H1975 from the Adeno:
- EGFR-p.E746_A750del.
Dependencies are inferred from the claim text.