Predictive test for patient benefit from antibody drug blocking ligand activation of the T-cell programmed cell death 1 (PD-1) checkpoint protein and classifier development methods

Inventors

Röder, Joanna • Meyer, Krista • Grigorieva, Julia • Tsypin, Maxim • Oliveira, Carlos • Steingrimsson, Arni • Röder, Heinrich • Asmellash, Senait • Sayers, Kevin • Maher, Caroline

Assignees

Biodesix Inc

Abstract

A method is disclosed of predicting cancer patient response to immune checkpoint inhibitors, e.g., an antibody drug blocking ligand activation of programmed cell death 1 (PD-1) or CTLA4. The method includes obtaining mass spectrometry data from a blood-based sample of the patient, obtaining integrated intensity values in the mass spectrometry data of a multitude of pre-determined mass-spectral features; and operating on the mass spectral data with a programmed computer implementing a classifier. The classifier compares the integrated intensity values with feature values of a training set of class-labeled mass spectral data obtained from a multitude of melanoma patients with a classification algorithm and generates a class label for the sample. A class label “early” or the equivalent predicts the patient is likely to obtain relatively less benefit from the antibody drug and the class label “late” or the equivalent indicates the patient is likely to obtain relatively greater benefit from the antibody drug.

Core Innovation

A method detects a class label in a non-small cell lung cancer patient by conducting mass spectrometry on a blood-based sample and obtaining mass spectral data. It obtains integrated intensity values for a multitude of mass-spectral features, including features listed in Appendix A, Appendix B, or Appendix C, and a programmed computer operates on the mass spectral data with a classifier.

The classifier compares the integrated intensity values with feature values of a reference set of class-labeled mass spectral data obtained from blood-based samples from cancer patients treated with an antibody drug blocking ligand activation of programmed cell death 1 (PD-1). The document further grounds classification in survival-based class definitions, including Early and Late, and overall survival on treatment with nivolumab.

In hierarchical or ensemble implementations, the classifier is described as a hierarchical ensemble of tumor classifiers, including an ensemble implementation that compares outputs from multiple subgroup-specific tumor-size classifiers and applies decision logic to assign final labels. The document also describes classifier construction that uses iterative label refinement with time-to-event data, filtering based on hazard ratio thresholds, and master-classifier combination using regularized logistic regression with extreme dropout.

Claims Coverage

The covered inventive features focus on blood-based mass spectrometry with integrated intensities for predefined mass-spectral features, comparing those values to PD-1 antibody-treated reference class-labeled mass spectra to detect a class label, and classifier architecture and construction refinements including a hierarchical tumor ensemble, regularized combination with extreme dropout, and feature selection tied to Acute Response and Wound Healing.

The covered inventive features focus on blood-based mass spectrometry with integrated intensities for predefined mass-spectral features, comparison to PD-1 antibody-treated reference class-labeled mass spectra to detect a class label, and classifier refinements including a hierarchical tumor ensemble, regularized combination with extreme dropout, survival-based Early/Late labeling, and feature selection tied to Acute Response and Wound Healing.

Stated Advantages

Documented Applications