Predictive test for prognosis of myelodysplastic syndrome patients using mass spectrometry of blood-based sample

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.

We claim: 
     
         1 . A method for predicting prognosis of a myelodysplastic syndrome (MDS) patient comprising the steps of:
 (a) performing MALDI-TOF mass spectrometry on a blood-based sample obtained from the MDS patient by subjecting the sample to at least 100,000 laser shots and acquiring mass spectral data;   (b) obtaining integrated intensity values in the mass spectral data of a multitude of pre-determined mass-spectral features; and   (c) operating on the integrated intensity values of the mass spectral data with a programmed computer implementing a classifier;   wherein in the operating step 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 other MDS patients with the values obtained in step (b) with a classification algorithm and generates a class label for the sample, wherein the class label is associated with a prognosis of the MDS patient.   
     
     
         2 . The method of  claim 1 , wherein the classifier is configured as a combination of filtered mini-classifiers using a regularized combination method. 
     
     
         3 . The method of  claim 1 , wherein the obtaining step (b) comprises obtaining integrated intensity values of at least 50 features listed in Appendix A. 
     
     
         4 . The method of  claim 3 , wherein the obtaining step comprises obtaining integrated intensity values of at least 100 features listed in Appendix A. 
     
     
         5 . The method of  claim 3 , wherein the obtaining step comprises obtaining integrated intensity values of at least 300 features listed in Appendix A. 
     
     
         6 . A classifier for predicting the prognosis of a MDS patient, comprising in combination:
 a memory storing a reference set of mass spectral data obtained from blood-based samples of a multitude of MDS patients;   a programmed computer configured to implement a classifier configured as a combination of filtered mini-classifiers with drop-out regularization;   wherein the reference set of mass spectral data includes feature values of at least some of the m/z features listed in Appendix A.   
     
     
         7 . The classifier of  claim 6 , wherein the reference set of mass spectral data includes feature values of at least 50 features listed in Appendix A. 
     
     
         8 . The classifier of  claim 6 , wherein the reference set of mass spectral data includes feature values of at least 100 features listed in Appendix A. 
     
     
         9 . The classifier of  claim 6 , wherein the reference set of mass spectral data includes feature values of at least 300 features listed in Appendix A. 
     
     
         10 . A laboratory testing system for conducting tests on blood-based samples from MDS patients and predicting the prognosis of the MDS patients, comprising:
 a MALDI-TOF mass spectrometer configured to conduct mass spectrometry on a blood-based sample from a patient by subjecting the sample to at least 100,000 laser shots and acquire resulting mass spectral data;   a memory storing a reference set of mass spectral data obtained from blood-based samples of a multitude of MDS patients and associated class labels; and   a programmed computer configured to implement a classifier operating on the reference set and the resulting mass spectral data obtained from the blood-based sample from the patient;   wherein the reference set of mass spectral data includes feature values of at least some of the m/z features listed in Appendix A; and   wherein the programmed computer is programmed to generate a class label for the sample as an output of the classifier, wherein the class label is associated with a prognosis of the MDS patient.   
     
     
         11 . The system of  claim 10 , wherein the classifier is configured as a combination of filtered mini-classifiers with drop-out regularization.