The poor prognosis of acute myeloid leukemia (AML) and the highly heterogenous nature of the disease motivates targeted gene therapeutic investigations. Rho-associated protein kinases (ROCKs) are crucial for various actin cytoskeletal changes, which have established malignant consequences in various cancers, yet are still not being successfully utilized clinically towards cancer treatment. This work establishes the therapeutic activity of ROCK inhibitor (5Z)-2–5-(1H-pyrrolo[2,3-b]pyridine-3-ylmethylene)-1,3-thiazol-4(5H)-one (DJ4) in both in vitro and in vivo preclinical models of AML to highlight the potential of this class of inhibitors. Herein, DJ4 induced cytotoxic and proapoptotic effects in a dose-dependent manner in human AML cell lines (IC50: 0.05–1.68 μM) and primary patient cells (IC50: 0.264–13.43 μM); however, normal hematopoietic cells were largely spared. ROCK inhibition by DJ4 disrupts the phosphorylation of downstream targets, myosin light chain (MLC2) and myosin-binding subunit of MLC phosphatase (MYPT), yielding a potent yet selective treatment response at micromolar concentrations, from 0.02 to 1 μM. Murine models injected with luciferase-expressing leukemia cell lines subcutaneously or intravenously and treated with DJ4 exhibited an increase in overall survival and reduction in disease progression relative to the vehicle-treated control mice. Overall, DJ4 is a promising candidate to utilize in future investigations to advance the current AML therapy.

Background Risk stratification for patients undergoing hepatectomy can be attempted using established models. This study compares the platelet‐albumin‐bilirubin (PALBI) score with albumin‐bilirubin (ALBI) and model for end‐stage liver disease sodium (MELD‐Na) for predicting posthepatectomy liver failure (PHLF) and 30‐day mortality. Methods The 2014–2018 NSQIP database was queried for patients who underwent elective hepatectomy. Multivariable logistic regressions assessed associations of posthepatectomy outcomes with patient and clinical characteristics. Predictive accuracy of the grading systems was evaluated using receiver operator characteristic (ROC) curves and calculating area under the curve (AUC). Results Severe PHLF (Grade B/C) and mortality were present in 2.58% (N = 369) and 1.2% (N = 171) of patients who underwent hepatectomy (N = 13 925), respectively. ALBI Grade 2/3 had a stronger association with severe PHLF (odds ratio [OR] = 1.62, p < 0.01) and mortality (OR = 2.06, p < 0.005) than PALBI Grade 2/3 (OR = 1.14, p = 0.43 for PHLF and OR = 2.01, p < 0.005 for mortality) or MELD‐Na ≥10 (OR = 1.29, p = 0.25 for PHLF and OR = 1.84, p < 0.03). ALBI had a higher AUC (0.671) than PALBI (0.625) and MELD‐Na (0.627) for predicting severe PHLF. ALBI had a higher AUC (0.695) than PALBI (0.642) for predicting 30‐day mortality. Conclusions ALBI was a more accurate predictor of severe PHLF and 30‐day mortality than MELD‐Na and PALBI for patients who underwent hepatectomy.

No abstract