Introduction: Pegaspargase (PEG) is key to the treatment of pediatric Acute Lymphoblastic Leukemia. PEG depletes blood asparagine (Asn), killing leukemic but not healthy cells. Guidelines on optimum dosing of PEG are ambiguous. Dietary and gut Asn can diffuse from the gut into the bloodstream through epithelial cells. We also know that some patients have bacteria in the gut that produce asparaginase. It is unclear whether dietary Asn impacts PEG efficacy. In this study we sought to determine if dietary Asn can impact PEG-mediated plasma Asn depletion in a pre-clinical model.

Methods: Pre-diet blood and stool samples from 10 healthy, non-tumor-bearing C57BL/6 mice were collected. The mice were then divided into 2 cages and given an Asn rich (4% Asn, cage A) or depleted (0%, cage B) diet. Blood and stool were sampled again 35 and 72 days after commencing the diet. The mice were then injected with 200 IU/kg PEG intraperitoneally. 5-days post-PEG blood and stool samples were collected. Blood PEG activity was immediately quenched following collection using a solution of 20% formic acid to ensure that Asn was not catabolized post collection. Mouse A2 unfortunately received roughly half the PEG in comparison to the other mice and was excluded from the final data summaries. Over 200 blood and stool metabolites were analyzed using LC-MS/MS at each timepoint. ANCOVA was used to test for between-diet significance. MetaboAnalyst 5.0 was used to identify pathways that were different between diets. A p value of ≤0.05 was considered significant.

Results:

Blood Asn levels were similar between the 2 diets at all time points. Blood and stool metabolomic PCA plots each showed two distinct clusters, with the day 0 samples clustering together and all other sample timepoints in a second cluster. There were some outliers present in the blood PCA plot with mice of both diets on day 77 and one mouse on high Asn on day 35 positioned outside of the clusters. In the stool plot, 3 mice on the 0% diet on day 77 did not cluster with other samples, as well as 1 mouse on the 4% diet, on day 77 (Figure 1).

Blood Asn levels were depleted below the level of detection in all the 5-day post PEG samples, the last timepoint of the experiment. There were 5 pathways in pre-diet, 10 on day 35, 12 on day 72, and 4 on day 77 that were significantly different between diets. One pathway of interest that was significantly different on day 72 was aspartate (Asp) metabolism. Asp was significantly higher in the Asn rich group compared to the Asn depleted group. L-acetylaspartylglutamate was higher in the Asn depleted group compared to the Asn rich group (Figure 2).

Stool Asn levels were significantly lower in the Asn rich group after 72 days on diet. There were 20 pathways in pre-diet, 5 on day 35, 4 on day 72, and 1 on day 77, that were significantly different between diets. None of the pathways that were different in the blood were also affected at the same time in the stool.

Conclusions and Interpretations:

Our outcome showing differences late in the diet period demonstrate that diet can have a profound effect on metabolites, not necessarily including Asn. Nevertheless, we provide no evidence that controlling dietary Asn will impact blood Asn levels and PEG efficacy. We found corresponding low stool Asn and high blood Asp in the Asn rich diet after 72 days on the diet. One possible explanation for this is the high Asn diet caused an increase in asparaginase-producing bacteria, which converted the dietary Asn into Asp. The Asp then diffused into the blood from the gut. Further studies examining the genomes of the bacteria present to confirm this theory are underway. Despite changes in metabolites, PEG depleted Asn in mice on both diets. PCA plots revealed that some post-PEG samples did not cluster with the others, indicating that PEG, like diet, could have a profound impact on metabolites. Further studies should examine the microbial and blood changes leading to these outliers. It is also important to take a closer look at the replenishment of Asn post-PEG to see if diet can affect the rate at which blood Asn is restored. The changes we observed and the consequences of these changes, will need to be confirmed in mice with leukemia.

Acknowledgement: This research was funded from a grant from the Beatrice Hunter Cancer Research Institute, Dalhousie Medical Research Institute, and ZF was funded by an IWK Health Graduate Studentship. The pegaspargase was kindly provided by Servier Canada Inc.

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution