See 1 citation found by title matching your search:
Deep learning using tumor HLA peptide mass spectrometry datasets improves neoantigen identification.
Bulik-Sullivan B1,
Busby J1,
Palmer CD1,
Davis MJ1,
Murphy T1,
Clark A1,
Busby M1,
Duke F1,
Yang A1,
Young L1,
Ojo NC1,
Caldwell K1,
Abhyankar J1,
Boucher T1,
Hart MG1,
Makarov V2,
Montpreville VT3,
Mercier O3,
Chan TA2,
Scagliotti G4,
Bironzo P4,
Novello S4,
Karachaliou N5,
Rosell R6,
Anderson I7,
Gabrail N8,
Hrom J9,
Limvarapuss C10,
Choquette K11,
Spira A11,
Rousseau R1,
Voong C1,
Rizvi NA12,
Fadel E3,
Frattini M12,
Jooss K1,
Skoberne M1,
Francis J1,
Yelensky R1.
- 1
- Gritstone Oncology, Inc., Emeryville, California and Cambridge, Massachusetts, USA.
- 2
- Memorial Sloan Kettering Cancer Center, New York, New York, USA.
- 3
- Centre Chirurgical Marie Lannelongue, Le Plessis-Robinson, France.
- 4
- University of Turin, Department of Oncology at San Luigi Hospital, Orbassano (Turin), Italy.
- 5
- Instituto Oncologico Dr. Rosell - Hospital Universitari Quiron Dexeus Location, Barcelona, Spain.
- 6
- Catalan Institute of Oncology, Barcelona, Spain.
- 7
- St Joseph Heritage Healthcare, Santa Rosa, California, USA.
- 8
- Gabrail Cancer Center, Canton, Ohio, USA.
- 9
- Hattiesburg Clinic/Forrest General Cancer Center, Hattiesburg, Mississippi, USA.
- 10
- Solano Hematology Oncology, Vallejo, California, USA.
- 11
- Virginia Cancer Specialists, Fairfax, Virginia, USA.
- 12
- New York Presbyterian/Columbia University Medical Center, New York, New York, USA.
Abstract
Neoantigens, which are expressed on tumor cells, are one of the main targets of an effective antitumor T-cell response. Cancer immunotherapies to target neoantigens are of growing interest and are in early human trials, but methods to identify neoantigens either require invasive or difficult-to-obtain clinical specimens, require the screening of hundreds to thousands of synthetic peptides or tandem minigenes, or are only relevant to specific human leukocyte antigen (HLA) alleles. We apply deep learning to a large (N = 74 patients) HLA peptide and genomic dataset from various human tumors to create a computational model of antigen presentation for neoantigen prediction. We show that our model, named EDGE, increases the positive predictive value of HLA antigen prediction by up to ninefold. We apply EDGE to enable identification of neoantigens and neoantigen-reactive T cells using routine clinical specimens and small numbers of synthetic peptides for most common HLA alleles. EDGE could enable an improved ability to develop neoantigen-targeted immunotherapies for cancer patients.