As a prototype of genomics-guided precision medicine, individualized thiopurine dosing based on pharmacogenetics is a highly effective way to mitigate hematopoietic toxicity of this class of drugs. Recently, NUDT15 deficiency was identified as a genetic cause of thiopurine toxicity, and NUDT15-informed preemptive dose reduction was quickly adopted in clinical settings. To exhaustively identify pharmacogenetic variants in this gene, we developed massively parallel NUDT15 function assays to determine the variants’ effect on protein abundance and thiopurine cytotoxicity. Of the 3,097 possible missense variants, we characterized the abundance of 2,922 variants and found 54 hotspot residues at which variants resulted in complete loss of protein stability. Analyzing 2,935 variants in the thiopurine cytotoxicity-based assay, we identified 17 additional residues where variants altered NUDT15 activity without affecting protein stability. We identified structural elements key to NUDT15 stability and/or catalytical activity with single amino acid resolution. Functional effects for NUDT15 variants accurately predicted toxicity risk alleles in patients treated with thiopurines with far superior sensitivity and specificity compared to bioinformatic prediction algorithms. In conclusion, our massively parallel variant function assays identified 1,152 deleterious NUDT15 variants, providing a comprehensive reference of variant function and vastly improving the ability to implement pharmacogenetics-guided thiopurine treatment individualization.

The folate metabolism enzyme MTHFD2 (methylenetetrahydrofolate dehydrogenase/cyclohydrolase) is consistently overexpressed in cancer but its roles are not fully characterized, and current candidate inhibitors have limited potency for clinical development. In the present study, we demonstrate a role for MTHFD2 in DNA replication and genomic stability in cancer cells, and perform a drug screen to identify potent and selective nanomolar MTHFD2 inhibitors; protein cocrystal structures demonstrated binding to the active site of MTHFD2 and target engagement. MTHFD2 inhibitors reduced replication fork speed and induced replication stress followed by S-phase arrest and apoptosis of acute myeloid leukemia cells in vitro and in vivo, with a therapeutic window spanning four orders of magnitude compared with nontumorigenic cells. Mechanistically, MTHFD2 inhibitors prevented thymidine production leading to misincorporation of uracil into DNA and replication stress. Overall, these results demonstrate a functional link between MTHFD2-dependent cancer metabolism and replication stress that can be exploited therapeutically with this new class of inhibitors.

Extracellular adenosine (ADO), present in high concentrations in the tumor microenvironment (TME), suppresses immune function via inhibition of T cell and NK cell activation. Intratumoral generation of ADO depends on the sequential catabolism of ATP by two ecto-nucleotidases, CD39 (ATP → AMP) and CD73 (AMP → ADO). Inhibition of CD73 eliminates a major pathway of ADO production in the TME and can reverse ADO-mediated immune suppression. Extensive interrogation of structure−activity relationships (SARs), structure-based drug design, and optimization of pharmacokinetic properties culminated in the discovery of AB680, a highly potent (K i = 5 pM), reversible, and selective inhibitor of CD73. AB680 is further characterized by very low clearance and long half-lives across preclinical species, resulting in a PK profile suitable for long-acting parenteral administration. AB680 is currently being evaluated in phase 1 clinical trials. Initial data show AB680 is well tolerated and exhibits a pharmacokinetic profile suitable for biweekly (Q2W) iv-administration in human.