Spiekermann, K. | Search bioRxiv

bioRxiv2019

Functional plasticity and reversible growth behavior of patients acute myeloid leukemia stem cells growing in mice

Resistance against chemotherapy remains a major obstacle in treating patients with acute myeloid leukemia (AML). Novel therapeutic concepts are especially desired to target and eliminate resistant AML stem cells. Here we show that AML stem cells harbor the plasticity to switch from a low-cycling, chemotherapy resistant state into an actively proliferating state associated with treatment response. We used patient-derived xenograft (PDX) cells from patients with high risk or relapsed AML, which were lentivirally transduced for marker expression, stained with the proliferation-sensitive dye Carboxyfluorescein succinimidyl ester (CFSE), and re-transplanted into next-recipient mice. A rare subpopulation of AML cells displayed reduced proliferation in vivo, associated with increased treatment resistance. The proportion of AML cells with stem cell potential was identical in both, the highly and lowly proliferative sub-fraction. In re-transplantation experiments, proliferation behavior proved reversible, and AML stem cells were able to switch between a high and low proliferation state. Our data indicate that AML stem cells display functional plasticity in vivo, which might be exploited for therapeutic purposes, to prevent AML relapse and ultimately improve the prognosis of patients with AML.

cancer biology↗

bioRxiv2019

Human-level recognition of blast cells in acute myeloid leukemia with convolutional neural networks

Reliable recognition of malignant white blood cells is a key step in the diagnosis of hematologic malignancies such as Acute Myeloid Leukemia. Microscopic morphological examination of blood cells is usually performed by trained human examiners, making the process tedious, time-consuming and hard to standardise. We compile an annotated image dataset of over 18,000 white blood cells, use it to train a convolutional neural network for leukocyte classification, and evaluate the networks performance. The network classifies the most important cell types with high accuracy. It also allows us to decide two clinically relevant questions with human-level performance, namely (i) if a given cell has blast character, and (ii) if it belongs to the cell types normally present in non-pathological blood smears. Our approach holds the potential to be used as a classification aid for examining much larger numbers of cells in a smear than can usually be done by a human expert. This will allow clinicians to recognize malignant cell populations with lower prevalence at an earlier stage of the disease.

pathology↗