Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: A method is disclosed for predicting in advance whether a melanoma patient is likely to benefit from high dose IL2 therapy in treatment of the cancer. The method makes use of mass spectrometry data obtained from a blood-based sample of the patient and a computer configured as a classifier and making use of a reference set of mass spectral data obtained from a development set of blood-based samples from other melanoma patients. A variety of classifiers for making this prediction are disclosed, including a classifier developed from a set of blood-based samples obtained from melanoma patients treated with high dose IL2 as well as melanoma patients treated with an anti-PD-1 immunotherapy drug. The classifiers developed from anti-PD-1 and IL2 patient sample cohorts can also be used in combination to guide treatment of a melanoma patient.

Abstract: A method of predicting whether an MDS patient has a good or poor prognosis uses a general purpose computer configured as a classifier and mass-spectrometry data obtained from a blood-based sample. The classifier assigns a classification label of either Early or Late (or the equivalent) to the patient's sample. Patients classified as Early are predicted to have a poor prognosis or worse survival whereas those patients classified as Late are predicted to have a relatively better prognosis and longer survival time. The groupings demonstrated a large effect size between groups in Kaplan-Meier analysis of survival. Most importantly, while the classifications generated were correlated with other prognostic factors, such as IPSS score and genetic category, multivariate and subgroup analysis showed that they had significant independent prognostic power complementary to the existing prognostic factors.

Abstract: A method of generating a classifier includes a step of classifying each member of a development set of samples with a class label in a binary classification scheme with a first classifier; and generating a second classifier using a classifier development process with an input classifier development set being the members of the development set assigned one of the two class labels in the binary classification scheme by the first classifier. The second classifier stratifies the members of the set with an early label into two further sub-groups. We also describe identifying a plurality of different clinical sub-groups within the development set based on the clinical data and for each of the different clinical sub-groups, conducting a classifier generation process for each of the clinical sub-groups thereby generating clinical subgroup classifiers.

Abstract: A method of predicting whether an MDS patient has a good or poor prognosis uses a general purpose computer configured as a classifier and mass-spectrometry data obtained from a blood-based sample. The classifier assigns a classification label of either Early or Late (or the equivalent) to the patient's sample. Patients classified as Early are predicted to have a poor prognosis or worse survival whereas those patients classified as Late are predicted to have a relatively better prognosis and longer survival time. The groupings demonstrated a large effect size between groups in Kaplan-Meier analysis of survival. Most importantly, while the classifications generated were correlated with other prognostic factors, such as IPSS score and genetic category, multivariate and subgroup analysis showed that they had significant independent prognostic power complementary to the existing prognostic factors.

Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: A method for predicting whether an early stage (IA, IB) non-small-cell lung cancer (NSCLC) patient is at a high risk of recurrence of the cancer following surgery involves subjecting a blood-based sample from the patient (obtained prior to, at, or after the surgery) to mass spectrometry and classification with a computer implementing a classifier. If the patients blood sample is classified as “high risk”, highest risk“or the equivalent, the patient can be guided to more aggressive treatment post-surgery. The classifier, or combination of classifiers, can be arranged in a hierarchical manner to make intermediate classifications, such as intermediate/high or intermediate/low, as well as low risk” or “lowest risk” classifications. Such additional classifications may guide clinical decisions as well.

Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: Shapley values (SVs) have become an important tool to further the goal of explainability of machine learning (ML) models. However, the computational load of exact SV calculations increases exponentially with the number of attributes. Hence, the calculation of SVs for models incorporating large numbers of interpretable attributes is problematic. Molecular diagnostic tests typically seek to leverage information from hundreds or thousands of attributes, often using training sets with fewer instances. Methods are described for evaluate SVs using Monte Carlo sampling or exact calculation in polynomial time (i.e., reasonably quickly and efficiently) using the architecture of a ML model designed for robust molecular test generation, and without requiring classifier retraining.

Abstract: A method of generating a classifier includes a step of classifying each member of a development set of samples with a class label in a binary classification scheme with a first classifier; and generating a second classifier using a classifier development process with an input classifier development set being the members of the development set assigned one of the two class labels in the binary classification scheme by the first classifier. The second classifier stratifies the members of the set with an early label into two further sub-groups. We also describe identifying a plurality of different clinical sub-groups within the development set based on the clinical data and for each of the different clinical sub-groups, conducting a classifier generation process for each of the clinical sub-groups thereby generating clinical subgroup classifiers.

Abstract: A blood-based sample from a cancer patient is subject to mass spectrometry and the resulting mass spectral data is classified with the aid of a computer to see if the patient is a member of a class of patients having a poor prognosis. If so, the mass spectral data is further classified with the aid of the computer by a second classifier which identifies whether the patient is nevertheless likely to obtain durable benefit from immunotherapy drugs, e.g., immune checkpoint inhibitors, anti-CTLA4 drugs, and high dose interleukin-2.

Abstract: The invention provides systems and methods for determining and predicting the effect of providing a population of tumor infiltrating lymphocytes (TILs) on a condition associated with an entity, for example the effect of providing a population of tumor infiltrating lymphocytes (TILs) on a subject having cancer. The systems and methods rely on acquiring a computer readable analytical signature from a sample of the entity, obtaining a trained model output value for the entity by inputting the computer readable analytical signature into a tier trained model panel, and classifying the entity based upon the trained model output value with a time-to-event class in an enumerated set of time-to-event classes, each of whom is associated with a different effect of providing a population of TILs to the entity.

Abstract: A blood-based sample from a cancer patient is subject to mass spectrometry and the resulting mass spectral data is classified with the aid of a computer to see if the patient is a member of a class of patients having a poor prognosis. If so, the mass spectral data is further classified with the aid of the computer by a second classifier which identifies whether the patient is nevertheless likely to obtain durable benefit from immunotherapy drugs, e.g., immune checkpoint inhibitors, anti-CTLA4 drugs, and high dose interleukin-2.

Abstract: A laboratory test apparatus for conducting a mass spectrometry test on a blood-based sample of a cancer patient includes a classification procedure implemented in a programmed computer that generates a class label for the sample. In one form of the test, “Test 1” herein, if the sample is labelled “Bad” or equivalent the patient is predicted to exhibit primary immune resistance if they are later treated with anti-PD-1 or anti-PD-L1 therapies in treatment of the cancer. In another configuration of the test, “Test 2” herein, the Bad class label predicts that the patient will have a poor prognosis in response to treatment by either anti-PD-1 or anti-PD-L1 therapies or alternative chemotherapies, such as docetaxel or pemetrexed. “Test 3” identifies patients that are likely to have a poor prognosis in response to treatment by either anti-PD-1 or anti-PD-L1 therapies but have improved outcomes on alternative chemotherapies.

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.

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.

Abstract: Classifier generation methods are described in which features used in classification (e.g., mass spectral peaks) are selected, or deselected using bagged filtering. A development sample set is split into two subsets, one of which is used as a training set the other of which is set aside. We define a classifier (e.g., K-nearest neighbor, decision tree, margin-based classifier or other) using the training subset and at least one of the features (or subsets of two or more features in combination). We apply the classifier to a subset of samples. A filter is applied to the performance of the classifier on the sample subset and the at least one feature is added to a “filtered feature list” if the classifier performance passes the filter. We do this for many different realizations of the separation of the development sample set into two subsets, and, for each realization, different features or sets of features in combination.

Abstract: A method and system for predicting in advance of treatment whether a cancer patient is likely, or not likely, to obtain benefit from administration of a yeast-based immune response generating therapy, which may be yeast-based immunotherapy for mutated Ras-based cancer, alone or in combination with another anti-cancer therapy. The method uses mass spectrometry of a blood-derived patient sample and a computer configured as a classifier using a training set of class-labeled spectra from other cancer patients that either benefited or did not benefit from an immune response generating therapy alone or in combination with another anti-cancer therapy.

Abstract: A programmed computer functioning as a classifier operates on mass spectral data obtained from a blood-based patient sample to predict indolence or aggressiveness of prostate cancer. Methods of generating the classifier and conducting a test on a blood-based sample from a prostate cancer patient using the classifier are described.