TY - JOUR
T1 - A Crowdsourcing Approach to Developing and Assessing Prediction Algorithms for AML Prognosis
AU - DREAM 9 AML-OPC Consortium
AU - Noren, David P.
AU - Long, Byron L.
AU - Norel, Raquel
AU - Rrhissorrakrai, Kahn
AU - Hess, Kenneth
AU - Hu, Chenyue Wendy
AU - Bisberg, Alex J.
AU - Schultz, Andre
AU - Engquist, Erik
AU - Liu, Li
AU - Lin, Xihui
AU - Chen, Gregory M.
AU - Xie, Honglei
AU - Hunter, Geoffrey A M
AU - Boutros, Paul C.
AU - Stepanov, Oleg
AU - Norman, Thea
AU - Friend, Stephen H.
AU - Stolovitzky, Gustavo
AU - Kornblau, Steven
AU - Qutub, Amina A.
N1 - Publisher Copyright:
© 2016 Noren et al.
PY - 2016/6
Y1 - 2016/6
N2 - Acute Myeloid Leukemia (AML) is a fatal hematological cancer. The genetic abnormalities underlying AML are extremely heterogeneous among patients, making prognosis and treatment selection very difficult. While clinical proteomics data has the potential to improve prognosis accuracy, thus far, the quantitative means to do so have yet to be developed. Here we report the results and insights gained from the DREAM 9 Acute Myeloid Prediction Outcome Prediction Challenge (AML-OPC), a crowdsourcing effort designed to promote the development of quantitative methods for AML prognosis prediction. We identify the most accurate and robust models in predicting patient response to therapy, remission duration, and overall survival. We further investigate patient response to therapy, a clinically actionable prediction, and find that patients that are classified as resistant to therapy are harder to predict than responsive patients across the 31 models submitted to the challenge. The top two performing models, which held a high sensitivity to these patients, substantially utilized the proteomics data to make predictions. Using these models, we also identify which signaling proteins were useful in predicting patient therapeutic response.
AB - Acute Myeloid Leukemia (AML) is a fatal hematological cancer. The genetic abnormalities underlying AML are extremely heterogeneous among patients, making prognosis and treatment selection very difficult. While clinical proteomics data has the potential to improve prognosis accuracy, thus far, the quantitative means to do so have yet to be developed. Here we report the results and insights gained from the DREAM 9 Acute Myeloid Prediction Outcome Prediction Challenge (AML-OPC), a crowdsourcing effort designed to promote the development of quantitative methods for AML prognosis prediction. We identify the most accurate and robust models in predicting patient response to therapy, remission duration, and overall survival. We further investigate patient response to therapy, a clinically actionable prediction, and find that patients that are classified as resistant to therapy are harder to predict than responsive patients across the 31 models submitted to the challenge. The top two performing models, which held a high sensitivity to these patients, substantially utilized the proteomics data to make predictions. Using these models, we also identify which signaling proteins were useful in predicting patient therapeutic response.
UR - http://www.scopus.com/inward/record.url?scp=84978888838&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84978888838&partnerID=8YFLogxK
U2 - 10.1371/journal.pcbi.1004890
DO - 10.1371/journal.pcbi.1004890
M3 - Article
C2 - 27351836
AN - SCOPUS:84978888838
SN - 1553-734X
VL - 12
JO - PLoS computational biology
JF - PLoS computational biology
IS - 6
M1 - e1004890
ER -