Relapse is the leading cause of mortality in children with acute lymphoblastic leukemia (ALL). Among chemotherapeutics, thiopurines are key drugs in the backbone of ALL combination therapy. Using whole-exome sequencing, we identified relapse-specific mutations in phosphoribosyl pyrophosphate synthetase 1 (PRPS1), a rate-limiting purine biosynthesis enzyme, in 24/358 (6.7%) relapse B-ALL cases. All individuals who harbored PRPS1 mutations relapsed early on-treatment, and mutated ALL clones expanded exponentially prior to clinical relapse. Our functional analyses of PRPS1 mutants uncovered a new chemotherapy resistance mechanism involving reduced feedback inhibition of de novo purine biosynthesis and competitive inhibition of thiopurine activation. Notably, the de novo purine synthesis inhibitor lometrexol can effectively abrogate PRPS1 mutant-driven drug resistance. Overall these results highlight the importance of constitutive activation of de novo purine pathway in thiopurine resistance, and offer therapeutic strategies for the treatment of relapsed and resistant ALL.

Studies have shown inflammation is involved in the development of acute aortic dissection (AAD). The hypothesis that white blood cell count (WBCc) on admission may have an impact on the short- and long-term outcomes of type A AAD was tested in a large-scale, prospective observational cohort study.From 2008 to 2010, a total of 570 consecutive patients with type A AAD in Fuwai hospital were enrolled and were followed up. Baseline characteristics and WBCc on admission were collected. The primary outcomes were 30-day and long-term all-cause mortality.During a median of 1.89 years of follow-up, the 30-day and long-term all-cause mortality were 10.7% and 6.5%, respectively. Univariate Cox regression analysis identified admission WBCc as an independent predictor of 30-day mortality when considered as a continuous variable or as a categorical variable using the cutoff of 11.0  × 109 cells/L (all P < 0.05). After adjustment for age, sex, C-reactive protein, d-dimer, and surgical intervention, elevated admission WBCc (>11.0 × 109 cells/L) remained an independent predictor of 30-day mortality of AAD (hazard ratio = 3.31, 95% confidence interval 1.38–7.93, P = 0.007). No impact of admission WBCc was observed on the long-term all-cause mortality.In conclusion, elevated admission WBCc may be valuable as a predictor of 30-day mortality, and may be useful in the risk stratification of type A AAD during hospitalization.