Skip to main content
NCBI home page
JCI Insight logoLink to JCI Insight
. 2018 Aug 23;3(16):e124015. doi: 10.1172/jci.insight.124015

An early-biomarker algorithm predicts lethal graft-versus-host disease and survival

Matthew J Hartwell, Umut Özbek, Ernst Holler, Anne S Renteria, Hannah Major-Monfried, Pavan Reddy, Mina Aziz, William J Hogan, Francis Ayuk, Yvonne A Efebera, Elizabeth O Hexner, Udomsak Bunworasate, Muna Qayed, Rainer Ordemann, Matthias Wölfl, Stephan Mielke, Attaphol Pawarode, Yi-Bin Chen, Steven Devine, Andrew C Harris, Madan Jagasia, Carrie L Kitko, Mark R Litzow, Nicolaus Kröger, Franco Locatelli, George Morales, Ryotaro Nakamura, Ran Reshef, Wolf Rösler, Daniela Weber, Kitsada Wudhikarn, Gregory A Yanik, John E Levine, James LM Ferrara
PMCID: PMC6141176  PMID: 30135313

Original citation: JCI Insight. 2017;2(3):1–9. https://doi.org/10.1172/jci.insight.89798

Citation for this corrigendum: JCI Insight. 2018;3(16):e124015. https://doi.org/10.1172/jci.insight.124015

In the statistics section, the equation to generate a final prediction model from the training set was incorrect. The correct sentence is below.

We then created a training set at random and repeated the entire process to generate a final model: log[–log(1 – p̂)] = –11.263 + 1.844(log10ST2) + 0.577(log10REG3α), where p̂ = predicted probability of 6-month NRM.

The authors regret the error.

Version 1. 08/23/2018

Electronic publication

Footnotes

Articles from JCI Insight are provided here courtesy of American Society for Clinical Investigation

RESOURCES