Background

Acute myeloid leukaemia (AML) is a blood cancer that originates from immature myeloid lineage cells. Metabolic re-programming is a hallmark of cancer cells (1, 2). AML features metabolic heterogeneity, but limited translation from lab to clinic necessitates direct and precise metabolic examining evidence. ¹⁸F-fluorodeoxyglucose-(FDG)-positron emission tomography (PET) imaging have been wildly used to detect cancer and monitor treatment responses with higher SUV_max indicative of more aggressiveness(3, 4). However, since both leukaemia cells and other hematopoietic cells propagate in bone marrow, it’s difficult to distinguish AML cells from other non-leukaemia cells inbone marrow micro-environment (BMME) through PET imaging. Moreover, it has been recognized that SUV values from PET images can be misinterpreted by confounding factors, hindering their accuracy and specificity in measuring glucose uptake capacity of BMNCs. Reinfeld et al. developed a solid method to quantify the incorporation of metabolites in vivo at single-cell resolution based on PET tracer(4). Adapted from his approach, we previousely reported leukaemia cells compete over BMME cells on glucose uptake in a MLL-AF9-driven AML mouse model(5).

But here come the question: What’s the glucose partitioning pattern in the bone marrow of AML patients? Does glucose uptake by leukaemia cells correlate with response to induction chemotherapy?

Results

We recruited 16 newly-diagnosed untreated subjects with AML and measured glucose uptake in BM cells obtained by BM puncture after FDG-PET probing. Intriguingly, subjects achieving histological complete remission had higher per-cell ¹⁸F-FDG uptake than non-responders. We also ruled out the heterogeneous co-variates that might contribute to the response difference including detection time across patients, cell yields, treatment option bias and baseline clinic parameters. Notably, all subjects with low glucose uptake capacity were ELN poor-risk compared with 3 of 9 subjects with high glucose uptake capacity who were ELN poor-risk.

One might postulate that the extent of disease saturationor relative glucose sufficiency in bone marrow might affect glucose uptake at the per-cell level. Interestingly, there was no association between the blast percentage and the glucose uptake capacity. We sought to investigate which cell sub-population was the main component responsible for the glucose uptake pattern. There was more ¹⁸F-FDG uptake by leukaemia cells compared with other cells in responders. In contrast, leukaemia cells from non-responders showed decreased glucose uptake capacity compared with other cells. PDX models enriched for pure leukemia cells phenocopied these results. Therefore, leukaemia cells determine the glucose uptake pattern we observed for the total bone marrow.

Last, using a computational metabolic flux model, single-cell flux estimation analysis (scFEA) which characterizes stochiometric relationships of metabolic networks, we found that responders had higher inferred-flux in the glucose import module, whereas inferred-flux of glutamine uptake was enriched in non-responders. This finding requires further direct evidence.

Take home message

Based on a cell-wise quantitative FDG-PET tracer, our work suggests that responders had higher per-cell ¹⁸F-FDG uptake (“Hot” AML) than non-responders (“Cold” AML). We provide a rationale for evaluating ex vivo glucose uptake for AML patients as an approach to predict response to induction chemotherapy. Obviously, our conclusions require further validation with a larger sample size.It also warrants further study to determine the underlying mechanism between nutrient selection and chemo-sensitivity and opens up the opportunity to target the specific metabolic vulnerability of subgroups of AML patients for precision medicine.

Reference

1. Vander Heiden MG, DeBerardinis RJ. Understanding the Intersections between Metabolism and Cancer Biology. Cell. 2017;168(4):657-69.
2. Fernandez-de-Cossio-Diaz J, Vazquez A. Limits of aerobic metabolism in cancer cells. Sci Rep. 2017;7(1):13488.
3. Ye H, Adane B, Khan N, Alexeev E, Nusbacher N, Minhajuddin M, et al. Subversion of Systemic Glucose Metabolism as a Mechanism to Support the Growth of Leukemia Cells. Cancer Cell. 2018;34(4):659-73 e6.
4. Reinfeld BI, Madden MZ, Wolf MM, Chytil A, Bader JE, Patterson AR, et al. Cell-programmed nutrient partitioning in the tumour microenvironment. Nature. 2021;593(7858):282-8.
5. Deng S, Du J, Gale RP, Wang L, Zhan H, Liu F, et al. Glucose partitioning in the bone marrow micro-environment in acute myeloid leukaemia. Leukemia. 2023;37(7):1407-12.