A rare case report: multiple intrahepatic masses in a pediatric patient with citrin deficiency.

Deficiency of citrin, the liver-type aspartate-glutamate carrier, arises from biallelic mutations of the gene SLC25A13. Although citrin deficiency (CD) is associated with higher risk of hepatocellular carcinoma (HCC) in adult patients, this association remains inconclusive in pediatric cases. The patient in this paper had been diagnosed to have CD by SLC25A13 analysis at the age 10 months, and then in response to dietary therapy, her prolonged jaundice and marked hepatosplenomegaly resolved gradually. However, she was referred to the hospital once again due to recurrent abdominal distention for 2 weeks at her age 4 years and 9 months, when prominently enlarged liver and spleen were palpated, along with a strikingly elevated serum alpha-fetoprotein (AFP) level of 27605 ng/mL as well as a large mass in the right liver lobe and a suspected tumor thrombus within the portal vein on enhanced computed tomography. After 4 rounds of adjuvant chemotherapy, right hepatic lobectomy and portal venous embolectomy were performed at her age 5 years and 3 months, and metastatic hepatoblastoma was confirmed by histopathological analysis. Afterwards, the patient underwent 5 additional cycles of chemotherapy and her condition remained stable for 7 months after surgery. Unfortunately, hepatoblastoma recurred in the left lobe at the age 5 years and 10 months, which progressed rapidly into liver failure, and led to death at the age 6 years and 1 month. As far as we know, this is the the first case of hepatoblastoma in a patient with CD, raising the possibility of an association between these two conditions.

[Corrigendum] MicroRNA­222­3p promotes tumor cell migration and invasion and inhibits apoptosis, and is correlated with an unfavorable prognosis of patients with renal cell carcinoma.

Following the publication of the above article, an interested reader drew to the attention of the Editorial Office that, in Fig. 3A on p. 530, two pairs of data panels were overlapping, such that certain of the panels appeared to have been derived from the same original sources where the results from differently performed experiments were intended to have been portrayed. The authors have examined their original data, and realize that errors associated with data handling/labelling during the preparation of the representative images in Fig. 3A had occurred. The revised version of Fig. 3, showing the correct data for the 'NC/ACHN/Invasion and Migration' data panels, the 'Inhibitor NC/786­O' panel and the 'Inhibitor NC/ACHN/Invasion' panel, is shown on the next page. The authors can confirm that the errors associated with this figure did not have any significant impact on either the results or the conclusions reported in this study, and all the authors agree with the publication of this Corrigendum. The authors are grateful to the Editor of International Journal of Molecular Medicine for giving them the opportunity to publish this Corrigendum; furthermore, they apologize to the readership of the Journal for any inconvenience caused. [International Journal of Molecular Medicine 43: 525­534, 2019; DOI: 10.3892/ijmm.2018.3931].

[Retracted] Oncogenic miR-23a-5p is associated with cellular function in RCC.

Following the publication of this article, a concerned reader drew to the Editor's attention that, for the invasion and migration assay data shown in Fig. 4 on p. 2314, three pairs of data panels were overlapping, such that data which were intended to show the results from differently performed experiments were obtained from a smaller number of original sources. Moreover, after having conducted an internal investigation, the Editorial Office also observed that some of the flow cytometric data shown in Fig. 6 were duplicated in Fig. 7. Considering the number of overlapping data panels that have been identified in this published paper, the Editor of Molecular Medicine Reports has concluded that the article should be retracted from the publication on account of a lack of confidence in the integrity of the data. Upon contacting the authors about this matter, they accepted the decision to retract this paper. The Editor apologizes to the readership for any inconvenience caused, and thanks the interested reader for drawing this matter to our attention. [Molecular Medicine Reports 16: 2309-2317, 2017; DOI: 10.3892/mmr.2017.6829].

[Corrigendum] Identification of lncRNA EGOT as a tumor suppressor in renal cell carcinoma.

Subsequently to the publication of the above paper, an interested reader drew to the authors' attention that, concerrning the Transwell cell migration and invasion assay data shown in Fig. 6A and B for the 786­O cell line on p. 7206, the pcDNA3.1­EGOT 'Migration' and 'Invasion' (a­1 and b­1) data panels appeared to contain overlapping sections of data, such that they were potentially derived from the same original source, where these panels were intended to show the results from differently performed experiments. The authors have re­examined their original data, and realize that the 'Invasion' (b­1) panel in Fig. 6B was inadvertently chosen incorrectly. The revised version of Fig. 6, now featuring the correct data for the 'Invasion' experiment (B1 in the replacement figure) in Fig. 6B, is shown on the next page. Note that this error did not adversely affect either the results or the overall conclusions reported in this study. All the authors agree with the publication of this corrigendum, and are grateful to the Editor of Molecular Medicine Reports for allowing them the opportunity to publish this. They also wish to apologize to the readership of the Journal for any inconvenience caused.[Molecular Medicine Reports 16: 7072­7079, 2017; DOI: 10.3892/mmr.2017.7470].

Multi-contrast learning-guided lightweight few-shot learning scheme for predicting breast cancer molecular subtypes.

Invasive gene expression profiling studies have exposed prognostically significant breast cancer subtypes: normal-like, luminal, HER-2 enriched, and basal-like, which is defined in large part by human epidermal growth factor receptor 2 (HER-2), progesterone receptor (PR), and estrogen receptor (ER). However, while dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has been generally employed in the screening and therapy of breast cancer, there is a challenging problem to noninvasively predict breast cancer molecular subtypes, which have extremely low-data regimes. In this paper, a novel few-shot learning scheme, which combines lightweight contrastive convolutional neural network (LC-CNN) and multi-contrast learning strategy (MCLS), is worthwhile to be developed for predicting molecular subtype of breast cancer in DCE-MRI. Moreover, MCLS is designed to construct One-vs-Rest and One-vs-One classification tasks, which addresses inter-class similarity among normal-like, luminal, HER-2 enriched, and basal-like. Extensive experiments demonstrate the superiority of our proposed scheme over state-of-the-art methods. Furthermore, our scheme is able to achieve competitive results on few samples due to joint LC-CNN and MCLS for excavating contrastive correlations of a pair of DCE-MRI.

Application of dose-gradient function in reducing radiation induced lung injury in breast cancer radiotherapy.

OBJECTIVE: Try to create a dose gradient function (DGF) and test its effectiveness in reducing radiation induced lung injury in breast cancer radiotherapy. MATERIALS AND METHODS: Radiotherapy plans of 30 patients after breast-conserving surgery were included in the study. The dose gradient function was defined as DGH=VDVp3, then the area under the DGF curve of each plan was calculated in rectangular coordinate system, and the minimum area was used as the trigger factor, and other plans were triggered to optimize for area reduction. The dosimetric parameters of target area and organs at risk in 30 cases before and after re-optimization were compared. RESULTS: On the premise of ensuring that the target dose met the clinical requirements, the trigger factor obtained based on DGF could further reduce the V5, V10, V20, V30 and mean lung dose (MLD) of the ipsilateral lung in breast cancer radiotherapy, P < 0.01. And the D2cc and mean heart dose (MHD) of the heart were also reduced, P < 0.01. Besides, the NTCPs of the ipsilateral lung and the heart were also reduced, P < 0.01. CONCLUSION: The trigger factor obtained based on DGF is efficient in reducing radiation induced lung injury in breast cancer radiotherapy.

CGMA-Net: Cross-Level Guidance and Multi-Scale Aggregation Network for Polyp Segmentation.

Colonoscopy is considered the best prevention and control method for colorectal cancer, which suffers extremely high rates of mortality and morbidity. Automated polyp segmentation of colonoscopy images is of great importance since manual polyp segmentation requires a considerable time of experienced specialists. However, due to the high similarity between polyps and mucosa, accompanied by the complex morphological features of colonic polyps, the performance of automatic polyp segmentation is still unsatisfactory. Accordingly, we propose a network, namely Cross-level Guidance and Multi-scale Aggregation (CGMA-Net), to earn a performance promotion. Specifically, three modules, including Cross-level Feature Guidance (CFG), Multi-scale Aggregation Decoder (MAD), and Details Refinement (DR), are individually proposed and synergistically assembled. With CFG, we generate spatial attention maps from the higher-level features and then multiply them with the lower-level features, highlighting the region of interest and suppressing the background information. In MAD, we parallelly use multiple dilated convolutions of different sizes to capture long-range dependencies between features. For DR, an asynchronous convolution is used along with the attention mechanism to enhance both the local details and the global information. The proposed CGMA-Net is evaluated on two benchmark datasets, i.e., CVC-ClinicDB and Kvasir-SEG, whose results demonstrate that our method not only presents state-of-the-art performance but also holds relatively fewer parameters. Concretely, we achieve the Dice Similarity Coefficient (DSC) of 91.85% and 95.73% on Kvasir-SEG and CVC-ClinicDB, respectively. The assessment of model generalization is also conducted, resulting in DSC scores of 86.25% and 86.97% on the two datasets respectively.

Pulmonary embolism after diagnostic curettage in patient with adenomyosis and hysteromyoma: A case report and brief review of literature.

RATIONALE: Pulmonary embolism (PE) is a common cause of cardiovascular death whose major acquired risk factors include postoperative states, pregnancy, malignancy, and age. We report a case of PE that occurred after diagnostic curettage for abnormal uterine bleeding, with a medical history of adenomyosis and hysteromyoma. PATIENT CONCERNS AND DIAGNOSES: A 31-year-old Han Chinese female was referred to our hospital with menstrual disorders, increased menstrual flow, and severe anemia. After admission, the patient was treated with a blood transfusion, iron supplementation, and erythropoietin, and diagnostic curettage was performed the following day. On the first postoperative day, the patient developed pulmonary embolism with dyspnea and fever diagnosed by CT pulmonary angiography and significantly elevated D-dimer. INTERVENTIONS AND OUTCOMES: Molecular weight heparin was administered for PE for 2 weeks, dyspnea was relieved significantly after 2 days of treatment and the uterine bleeding did not increase; and gonadotropin-releasing hormone agonists were administered for adenomyosis after 1 week of anticoagulant therapy to reduce bleeding. We followed up for 6 months, and the patient had no recurrence of thrombosis and uterine bleeding had improved. CONCLUSION: We speculate that the occurrence of pulmonary embolism was closely related to adenomyosis, hysteromyoma, and curettage in this patient. Treating the presence of both menstrual bleeding and thromboembolism is challenging, and careful management is necessary to avoid therapeutic contradictions.

Application research on reducing radiation-induced lung injury with a trigger operator based on overlap volume histogram (OVH) in breast cancer postoperative radiotherapy.

This study aims to develop a trigger operator based on the Overlap Volume Histogram (OVH) and examined its effectiveness in enhancing plan quality to minimize radiation-induced lung injury in postoperative radiotherapy for breast cancer. This trigger operator was applied for plan re-optimization to the previous Volumetric Modulated Arc Therapy (VMAT) plans of 16 left breast conserving surgery cases. These cases were categorized into a Contiguous Group (CG) and a Separated Group (SG) based on the relative position between the target and the Left-Lung (L-Lung). We investigated the changes in Vx, mean dose, and Normal Tissue Complication Probability (NTCP) values of organs-at-risk (OARs) before and after using the trigger operator. The Pairwise Sample T test was employed to evaluate the differences in indices between the two groups before and after optimizations. The trigger operator effectively initiated plan re-optimization. The values of V5, V10, V20, V30, and V40 of the L-Lung, as well as the mean dose of the heart, all decreased after re-optimization. The Pairwise Sample T test results showed statistically significant differences in the V20, V30, and V40 of the L-Lung in the CG (P < 0.01), and in the V5, V10, V20, V30, and V40 of the L-Lung in the SG (P < 0.01). Our findings suggest that the proposed trigger operator can improve plan quality, thereby reducing radiation-induced lung injury in postoperative radiotherapy for breast cancer.

The heterogeneity of tumour immune microenvironment revealing the CRABP2/CD69 signature discriminates distinct clinical outcomes in breast cancer.

BACKGROUND: It has been acknowledged that the tumour immune microenvironment (TIME) plays a critical role in determining therapeutic responses and clinical outcomes in breast cancer (BrCa). Thus, the identification of the TIME features is essential for guiding therapy and prognostic assessment for BrCa. METHODS: The heterogeneous cellular composition of the TIME in BrCa by single-cell RNA sequencing (scRNA-seq). Two subtype-special genes upregulated in the tumour-rich subtype and the immune-infiltrating subtype were extracted, respectively. The CRABP2/CD69 signature was established based on CRABP2 and CD69 expression, and its predictive values for the clinical outcome and the neoadjuvant chemotherapy (NAT) responses were validated in multiple cohorts. Moreover, the oncogenic role of CRABP2 was explored in BrCa cells. RESULTS: Based on the heterogeneous cellular composition of the TIME in BrCa, the BrCa samples could be divided into the tumour-rich subtype and the immune-infiltrating subtype, which exhibited distinct prognosis and chemotherapeutic responses. Next, we extracted CRABP2 as the biomarker for the tumour-rich subtype and CD69 as the biomarker for the immune-infiltrating subtype. Based on the CRABP2/CD69 signature, BrCa samples were re-divided into three subtypes, and the CRABP2highCD69low subtype exhibited the worst prognosis and the lowest chemotherapeutic response, while the CRABP2lowCD69high subtype showed the opposite results. Furthermore, CARBP2 functioned as a novel oncogene in BrCa, which promoted tumour cell proliferation, migration, and invasion, and CRABP2 inhibition triggered the activation of cytotoxic T lymphocytes (CTLs). CONCLUSION: The CRABP2/CD69 signature is significantly associated with the TIME features and could effectively predict the clinical outcome. Also, CRABP2 is determined to be a novel oncogene, which could be a therapeutic target in BrCa.

MEAI: an artificial intelligence platform for predicting distant and lymph node metastases directly from primary breast cancer.

PURPOSE: Breast cancer patients typically have decent prognoses, with a 5-year survival rate of more than 90%, but when the disease metastases to lymph node or distant, the prognosis drastically declines. Therefore, it is essential for future treatment and patient survival to quickly and accurately identify tumor metastasis in patients. An artificial intelligence system was developed to recognize lymph node and distant tumor metastases on whole-slide images (WSIs) of primary breast cancer. METHODS: In this study, a total of 832 WSIs from 520 patients without tumor metastases and 312 patients with breast cancer metastases (including lymph node, bone, lung, liver, and other) were gathered. Based on the WSIs were randomly divided into the training and testing cohorts, a brand-new artificial intelligence system called MEAI was built to identify lymph node and distant metastases in primary breast cancer. RESULTS: The final AI system attained an area under the receiver operating characteristic curve of 0.934 in a test set of 187 patients. In addition, the potential for AI system to increase the precision, consistency, and effectiveness of tumor metastasis detection in patients with breast cancer was highlighted by the AI's achievement of an AUROC higher than the average of six board-certified pathologists (AUROC 0.811) in a retrospective pathologist evaluation. CONCLUSION: The proposed MEAI system can provide a non-invasive approach to assess the metastatic probability of patients with primary breast cancer.

Colon cancer with colovesical fistula: A report of four cases and a literature review.

Colon cancer with colovesical fistula (CVF) is a rare complication of colon cancer that possesses an extremely poor prognosis. Surgical treatment can improve the prognosis. The current study presents four cases of CVF, in which the first two cases were treated conservatively and the other two were treated surgically. The first case presented with intestinal obstruction for 3 days, and computed tomography (CT) was performed. The patient refused surgery and still exhibited lower abdominal pain 11 months later. The second case presented with urinary frequency and urgency that lasted for 2 days, and CT was performed. The patient refused surgery and died 2 months later. The third case presented with fecaluria that lasted for 1 month, and CT, endoscopy and one-stage palliative surgery were performed. The patient was lost to follow-up 5 months later. The fourth case presented with acute urinary tract symptoms for 4 months, and CT, endoscopy and one-stage radical surgery were performed. The patient remained disease-free 10 months later. The four cases reported in the present study not only represent excellent examples of the disease spectrum, but also act as a reminder of the possibility of detecting CVF at an early stage of the disease. The present study discusses the epidemiology of CVF, and presents the pattern of CVF in terms of signs, symptoms and imaging examinations, including CT, cystoscopy and colonoscopy, as well as treatment in the early stage of the disease.

Deep learning-extracted CT imaging phenotypes predict response to total resection in colorectal cancer.

BACKGROUND: Deep learning surpasses many traditional methods for many vision tasks, allowing the transformation of hierarchical features into more abstract, high-level features. PURPOSE: To evaluate the prognostic value of preoperative computed tomography (CT) image texture features and deep learning self-learning high-throughput features (SHF) on postoperative overall survival in the treatment of patients with colorectal cancer (CRC). MATERIAL AND METHODS: The dataset consisted of 810 enrolled patients with CRC confirmed from 10 November 2011 to 10 February 2018. In contrast, SHF extracted by deep learning with multi-task training mechanism and texture features were extracted from the CT with tumor volume region of interest, respectively, and combined with the Cox proportional hazard (CoxPH) model for initial validation to obtain a RAD score to classify patients into high- and low-risk groups. The SHF stability was further validated in combination with Neural Multi-Task Logistic Regression (N-MTLR) model. The overall recognition ability and accuracy of CoxPH and N-MTLR model were evaluated by C-index and Integrated Brier Score (IBS). RESULTS: SHF had a more significant degree of differentiation than texture features. The result is (SHF vs. texture features: C-index: 0.884 vs. 0.611; IBS: 0.025 vs. 0.073) in the CoxPH model, and (SHF vs. texture features: C-index: 0.861 vs. 0.630; IBS: 0.024 vs. 0.065) in N-MTLR. CONCLUSION: SHF is superior to texture features and has potential application for the preoperative prediction of the individualized treatment of CRC.

Anthracite Releases Aromatic Carbons and Reacts with Chlorine to Form Disinfection Byproducts in Drinking Water Production.

Anthracite is globally used as a filter material for water purification. Herein, it was found that up to 15 disinfection byproducts (DBPs) were formed in the chlorination of anthracite-filtered pure water, while the levels of DBPs were below the detection limit in the chlorination of zeolite-, quartz sand-, and porcelain sandstone-filtered pure water. In new-anthracite-filtered water, the levels of dissolved organic carbon (DOC), dissolved organic nitrogen (DON), and ammonia nitrogen (NH3-N) ranged from 266.3 to 305.4 µg/L, 37 to 61 µg/L, and 8.6 to 17.1 µg/L, respectively. In aged anthracite (collected from a filter at a DWTP after one year of operation) filtered water, the levels of the above substances ranged from 475.1 to 597.5 µg/L, 62.1 to 125.6 µg/L, and 14 to 28.9 µg/L, respectively. Anthracite would release dissolved substances into filtered water, and aged anthracite releases more substances than new anthracite. The released organics were partly (around 5%) composed by the µg/L level of toxic and carcinogenic aromatic carbons including pyridine, paraxylene, benzene, naphthalene, and phenanthrene, while over 95% of the released organics could not be identified. Organic carbon may be torn off from the carbon skeleton structure of anthracite due to hydrodynamic force in the water filtration process.

Comparative analysis of dosimetry and predictive somatotype parameters of prone and supine whole-breast irradiation among Chinese women after breast-conserving surgery.

Purpose: Finding a better treatment position (prone or supine) for whole-breast irradiation for Chinese female patients diagnosed with breast cancer by identify the associations between predictive somatotype parameters and dosimetric gains. Materials and methods: Two volumetric-modulated arc therapy (VMAT) plans were deployed for whole-breast irradiation in supine and prone position with a total dose of 50 Gy in 25 fractions. Dose-volume parameters were compared and analysed both in the target volume and organs at risk, and equivalent uniform dose-based figure-of-merit (fEUD) models were further used to quantitatively evaluate the overall merits of the two plans. Body shape parameters, including body mass index (BMI), body surface area (BSA), breast shape, cup size, bust size and chest size, were collected. Anatomic features such as the central heart distance (CHD) were measured on supine CT. Spearman's correlation analysis, receiver operating characteristic (ROC) curve analysis, and the linear regression models were conducted. Results: Doses to the heart and left anterior descending coronary artery (LADCA) are greater in left-sided breast cancer (BC) patients in the prone position than in the supine position, and the opposite was true for right-sided BC patients (p<0.001). 19 of 63 patients (5 left-sided and 14 right-sided BC) achieved greater benefit from the prone position according to the fEUD score. Right-sided BC patients with a bust size ≥92.25 cm, drop-type breasts and cup size ≥B are very likely to benefit from prone-position radiotherapy. The CHD is significantly positively associated with △fEUD among right-sided BC patients (rho=0.506, p=0.004). Using a cut-off point of 2.215, the CHD had 71.4% sensitivity and 81.2% specificity in predicting a successful prone plan. Conclusions: Right-sided BC patients had better dosimetric gain in the prone position than left-sided BC patients. The CHD is an especially good and novel predictor that could help to select prone-benefitting right-sided BC patients.

Transcriptomic Features in a Single Extracellular Vesicle via Single-Cell RNA Sequencing.

Although many studies have investigated functional molecules in extracellular vesicles (EVs), the exact number of ribonucleic acid molecules in a single-EV is unknown. Therefore, it is critical to explore the transcriptomic features and heterogeneity at the level of a single-EV. Here, using the 10x Genomics platform, the RNA cargos are profiled in single EVs derived from human K562 and mesenchymal stem cells. The key steps are labeling intact EVs using calcein-AM, detecting the EV concentration via flow cytometry, and using the CB2 algorithm with adaptive thresholds to effectively distinguish real EVs from background. The gene number in a single-EV varied from 6 to 148, with a mean of 52. Ribosomal genes, mitochondrial genes, and eukaryotic translation elongation factor 1 alpha has a high EV percentage in all EV samples. Hemoglobin genes are uniquely highly expressed in K562-EVs, and cytoskeleton genes are enriched in MSC-EVs. Ten or more clusters with different marker genes in each single-EV dataset demonstrated EV heterogeneity. Moreover, integrating EVs and their parental cells reveal both EVs and cells in each cluster, indicating different cell origins of various EVs. To the best of the author's knowledge, this study provides the first high-throughput transcriptome at the single-EV level and improves the understanding of EVs.

A hybrid hemodynamic knowledge-powered and feature reconstruction-guided scheme for breast cancer segmentation based on DCE-MRI.

Automatically and accurately annotating tumor in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), which provides a noninvasive in vivo method to evaluate tumor vasculature architectures based on contrast accumulation and washout, is a crucial step in computer-aided breast cancer diagnosis and treatment. However, it remains challenging due to the varying sizes, shapes, appearances and densities of tumors caused by the high heterogeneity of breast cancer, and the high dimensionality and ill-posed artifacts of DCE-MRI. In this paper, we propose a hybrid hemodynamic knowledge-powered and feature reconstruction-guided scheme that integrates pharmacokinetics prior and feature refinement to generate sufficiently adequate features in DCE-MRI for breast cancer segmentation. The pharmacokinetics prior expressed by time intensity curve (TIC) is incorporated into the scheme through objective function called dynamic contrast-enhanced prior (DCP) loss. It contains contrast agent kinetic heterogeneity prior knowledge, which is important to optimize our model parameters. Besides, we design a spatial fusion module (SFM) embedded in the scheme to exploit intra-slices spatial structural correlations, and deploy a spatial-kinetic fusion module (SKFM) to effectively leverage the complementary information extracted from spatial-kinetic space. Furthermore, considering that low spatial resolution often leads to poor image quality in DCE-MRI, we integrate a reconstruction autoencoder into the scheme to refine feature maps in an unsupervised manner. We conduct extensive experiments to validate the proposed method and show that our approach can outperform recent state-of-the-art segmentation methods on breast cancer DCE-MRI dataset. Moreover, to explore the generalization for other segmentation tasks on dynamic imaging, we also extend the proposed method to brain segmentation in DSC-MRI sequence. Our source code will be released on https://github.com/AI-medical-diagnosis-team-of-JNU/DCEDuDoFNet.

Discovery of a novel water-soluble, rapid-release triptolide prodrug with improved drug-like properties and high efficacy in human acute myeloid leukemia.

In this work, a series of water-soluble triptolide prodrugs were synthesized, and their triptolide release rate, pharmacokinetic characteristics and anti-tumor effect were measured. We found that inserting glycolic acid as a linker between triptolide and the cyclic amino acid accelerated the release of triptolide from prodrugs into the plasma while preserving its safety. Among them, prodrug TP-P1 was significantly better than Minnelide (the only water-soluble triptolide prodrug in clinical trials) in terms of release rate in plasma and synthetic yield. In mouse models of human acute myeloid leukemia (AML), TP-P1 was effective in reducing xenograft tumors at dose levels as low as 25 µg/kg, and eliminating tumors at dose 100 µg/kg. Furthermore, TP-P1 could significantly enhance the efficacy of FLT3 inhibitors in the treatment of AML. These experimental results showed the potential of TP-P1 as water-soluble prodrugs of triptolide.