Back

Measurable Residual IDH1 before Allogeneic Transplant for Acute Myeloid Leukemia

Gui, G.; Dillon, L. W.; Ravindra, N.; Hegde, P. S.; Andrew, G.; Mukherjee, D.; Wong, Z.; Auletta, J.; El Chaer, F.; Chen, Y.-B.; Corner, A.; Devine, S. M.; Iyer, S.; Jimenez Jimenez, A. M.; De Lima, M. J. G.; Litzow, M. R.; Kebriaei, P.; Spellman, S. R.; Zeger, S. L.; Page, K. M.; Hourigan, C. S.

2023-08-01 oncology
10.1101/2023.07.28.23293166 medRxiv
Show abstract

Measurable residual disease (MRD) in adults with acute myeloid leukemia (AML) in complete remission is an important prognostic marker, but detection methodology requires optimization. The persistence of mutated NPM1 or FLT3-ITD in the blood of adult patients with AML in first complete remission (CR1) prior to allogeneic hematopoetic cell transplant (alloHCT) has been established as associated with increased relapse and death after transplant. The prognostic implications of persistence of other common AML-associated mutations, such as IDH1, at this treatment landmark however remains incompletely defined. We performed testing for residual IDH1 variants (IDH1m) in pre-transplant CR1 blood of 148 adult patients undergoing alloHCT for IDH1-mutated AML at a CIBMTR site between 2013-2019. No post-transplant differences were observed between those testing IDH1m positive (n=53, 36%) and negative pre-transplant (overall survival: p = 0.4; relapse: p = 0.5). For patients with IDH1 mutated AML co-mutated with NPM1 and/or FLT3-ITD, only detection of persistent mutated NPM1 and/or FLT3-ITD was associated with significantly higher rates of relapse (p = 0.01). These data, from the largest study to date, do not support the detection of IDH1 mutation in CR1 blood prior to alloHCT as evidence of AML MRD or increased post-transplant relapse risk.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.