Citations

Abstract Machine learning (ML) offers opportunities to advance pathological diagnosis, especially with increasing trends in digitalizing microscopic images. Diagnosing leukemia is time‐consuming and challenging in many areas globally and there is a growing trend in utilizing ML techniques for its diagnosis. In this review, we aimed to describe the literature of ML utilization in the diagnosis of the four common types of leukemia: acute lymphocytic leukemia (ALL), chronic lymphocytic leukemia (CLL), acute myeloid leukemia (AML), and chronic myelogenous leukemia (CML). Using a strict selection criterion, utilizing MeSH terminology and Boolean logic, an electronic search of MEDLINE and IEEE Xplore Digital Library was performed. The electronic search was complemented by handsearching of references of related studies and the top results of Google Scholar. The full texts of 58 articles were reviewed, out of which, 22 studies were included. The number of studies discussing ALL, AML, CLL, and CML was 12, 8, 3, and 1, respectively. No studies were prospectively applying algorithms in real‐world scenarios. Majority of studies had small and homogenous samples and used supervised learning for classification tasks. 91% of the studies were performed after 2010, and 74% of the included studies applied ML algorithms to microscopic diagnosis of leukemia. The included studies illustrated the need to develop the field of ML research, including the transformation from solely designing algorithms to practically applying them clinically.

In this work related to familial aggregation of familial venous thromboembolism, the investigators report genomic and transcriptomic association of 16 novel susceptibility loci for venous thromboembolism.

Background Clinical characteristics and outcomes of takotsubo syndrome ( TTS ) patients with malignancy have not been fully elucidated. This study sought to explore differences in clinical characteristics and to investigate short‐ and long‐term outcomes in TTS patients with or without malignancy. Methods and Results TTS patients were enrolled from the International Takotsubo Registry. The TTS cohort was divided into patients with and without malignancy to investigate differences in clinical characteristics and to assess short‐ and long‐term mortality. A subanalysis was performed comparing long‐term mortality between a subset of TTS patients with or without malignancy and acute coronary syndrome (ACS) patients with or without malignancy. Malignancy was observed in 16.6% of 1604 TTS patients. Patients with malignancy were older and more likely to have physical triggers, but less likely to have emotional triggers compared with those without malignancy. Long‐term mortality was higher in patients with malignancy ( P <0.001), while short‐term outcome was comparable ( P =0.17). In a subanalysis, long‐term mortality was comparable between TTS patients with malignancies and ACS patients with malignancies ( P =0.13). Malignancy emerged as an independent predictor of long‐term mortality. Conclusions A substantial number of TTS patients show an association with malignancy. History of malignancy might increase the risk for TTS , and therefore, appropriate screening for malignancy should be considered in these patients. Clinical Trial Registration URL : http://www.clinicaltrial.gov . Unique identifier: NCT 01947621.

e15029 Background: ICIs induce durable responses in dMMR CRC pts with overall response rates (ORR) of 30-50%. Even though the loss of expression of any MMR gene predicts ICIs response, it is unknown if ORRs are similar across all MMR genes (MLH1, PMS2, MSH2, and MSH6). In this study, we analyzed the impact of each specific MMR gene loss and clinical characteristics of pts with best response to ICIs. Methods: Pts were eligible if they had confirmed dMMR CRC by IHC or microsatellite instability-high (MSI-H) by PCR, and received ICIs between 01/01/2012 and 10/01/2018 at Winship Cancer Institute of Emory University, Mayo Clinic or Vanderbilt University Medical Center. Due to the pattern of frequent concurrent loss and functional dependency, the groups were categorized as MLH1 ±PMS2 vs. MSH2 ±MSH6. Cox proportional hazard model and Fisher’s exact test were used for the best response and the distribution of variable among the subgroups. Results: A total of 45 pts with dMMR CRC were identified. ORRs in MLH1 ±PMS2 and MSH2 ±MSH6 groups were 68% and 57.1% respectively without statistical difference (Table). Pts with age < 50 and 50-65 years old had better ORRs compared to pts with age >65 (58.3%, 85.7% and 42.1% respectively, P=0.036). Left-sided tumors had a trend toward higher ORRs compared to right-sided tumors (83.3% vs 51.5% P=0.086). Gender and BRAF status were not predictors of response. BRAF mutations were more common in right-sided tumors (29.6% vs 11.1% respectively) and in older patients. Conclusions: Our data suggest that MSI-H CRC pts aged 50-65 treated with ICIs, have improved ORR compared to pts > 65; pts with left-sided tumors have a trend toward improved ORR compared to those with right sided tumors. [Table: see text]

TPS2656 Background: ESCC remains the predominant histological subtype of, and accounts for most deaths from, esophageal cancer. PD-1 inhibition has demonstrated antitumor activity and was generally well tolerated in patients (pts) with advanced unresectable or metastatic ESCC. Tislelizumab, an investigational anti-PD-1 antibody, was engineered to minimize binding to FcγR on macrophages to abrogate antibody-dependent phagocytosis, a mechanism of T-cell clearance and potential resistance to anti-PD-1 therapy. Results from early phase clinical studies suggest single-agent tislelizumab was generally well tolerated and had antitumor activity in pts with solid tumors, including ESCC. Methods: This phase 3, randomized, placebo-controlled, double-blind study (NCT03783442) is designed to evaluate the efficacy and safety of tislelizumab plus chemotherapy as first-line treatment of unresectable, locally advanced recurrent or metastatic ESCC. Adult pts with histologically confirmed ESCC that had metastatic disease either at first diagnosis or with a ≥6 month treatment-free interval will be eligible. Additional eligibility criteria include measurable/evaluable disease, ECOG performance score ≤1, and no prior anti-PD-(L)-1, PD-L2, or other first-line therapy or palliative radiation treatment ≤4 weeks before treatment. Approximately 480 pts will be randomized 1:1 to receive investigator-chosen chemotherapy (ICC) plus tislelizumab 200 mg IV Q3W or ICC plus placebo. Chemotherapy options include: platinum (cisplatin 60–80 mg/m2 or oxaliplatin 130 mg/m2 IV Q3W) plus 5-FU 750–800 mg/m2 IV daily for 5 days Q3W; or platinum plus capecitabine 1000 mg/m2 orally BID for 14 days Q3W; or platinum + paclitaxel 175 mg/m2 IV Q3W. Progression-free survival and overall survival are coprimary endpoints; secondary endpoints include objective response rate, duration of response, and health-related quality-of-life. Safety will be assessed by monitoring adverse events, physical examinations, vital signs, and electrocardiograms. This study is actively enrolling. Clinical trial information: NCT03783442.

The aim of this study is to review all the published clinical trials on autologous bone marrow mesenchymal stem cells (BM-MSCs) in the repair of cartilage lesions of the knee. We performed a comprehensive search in three electronic databases: PubMed, Medline via Ovid, and Web of Science. A systematic review was conducted according to the guidelines of PRISMA protocol and the Cochrane Handbook for Systematic Reviews of Interventions. The modified Coleman methodology score was used to assess the quality of the included studies. Meta-analysis was conducted to estimate the effect size for Pain and function change after receiving BM-MSCs. Thirty-three studies—including 724 patients of mean age 44.2 years—were eligible. 50.7% of the included patients received cultured BM-MSCs for knee cartilage repair. There was improvement in the MINORS quality score over time with a positive correlation with the publication year. Meta-analysis indicated better improvement and statistical significance in the Visual Analog Scale for Pain, IKDC Function, Tegner Activity Scale, and Lysholm Knee Score after administration of noncultured BM-MSCs when compared to evaluation before the treatment. Meanwhile, there was a clear methodological defect in most studies with an average modified Coleman methodology score (MCMS) of 55. BM-MSCs revealed a clinically relevant improvement in pain, function, and histological regeneration.

Abstract Aims We aimed to evaluate the frequency, clinical features, and prognostic implications of cardiac arrest (CA) in takotsubo syndrome (TTS). Methods and results We reviewed the records of patients with CA and known heart rhythm from the International Takotsubo Registry. The main outcomes were 60-day and 5-year mortality. In addition, predictors of mortality and predictors of CA during the acute TTS phase were assessed. Of 2098 patients, 103 patients with CA and known heart rhythm during CA were included. Compared with patients without CA, CA patients were more likely to be younger, male, and have apical TTS, atrial fibrillation (AF), neurologic comorbidities, physical triggers, and longer corrected QT-interval and lower left ventricular ejection fraction on admission. In all, 57.1% of patients with CA at admission had ventricular fibrillation/tachycardia, while 73.7% of patients with CA in the acute phase had asystole/pulseless electrical activity. Patients with CA showed higher 60-day (40.3% vs. 4.0%, P &lt; 0.001) and 5-year mortality (68.9% vs. 16.7%, P &lt; 0.001) than patients without CA. T-wave inversion and intracranial haemorrhage were independently associated with higher 60-day mortality after CA, whereas female gender was associated with lower 60-day mortality. In the acute phase, CA occurred less frequently in females and more frequently in patients with AF, ST-segment elevation, and higher C-reactive protein on admission. Conclusions Cardiac arrest is relatively frequent in TTS and is associated with higher short- and long-term mortality. Clinical and electrocardiographic parameters independently predicted mortality after CA.

PURPOSE Evidence-based treatments for metastatic, human epidermal growth factor receptor 2 (HER2)–positive breast cancer to the CNS are limited. We previously reported modest activity of neratinib monotherapy for HER2-positive breast cancer brain metastases. Here we report the results from additional study cohorts. PATIENTS AND METHODS Patients with measurable, progressive, HER2-positive brain metastases (92% after receiving CNS surgery and/or radiotherapy) received neratinib 240 mg orally once per day plus capecitabine 750 mg/m 2 twice per day for 14 days, then 7 days off. Lapatinib-naïve (cohort 3A) and lapatinib-treated (cohort 3B) patients were enrolled. If nine or more of 35 (cohort 3A) or three or more of 25 (cohort 3B) had CNS objective response rates (ORR), the drug combination would be deemed promising. The primary end point was composite CNS ORR in each cohort separately, requiring a reduction of 50% or more in the sum of target CNS lesion volumes without progression of nontarget lesions, new lesions, escalating steroids, progressive neurologic signs or symptoms, or non-CNS progression. RESULTS Forty-nine patients enrolled in cohorts 3A (n = 37) and 3B (n = 12; cohort closed for slow accrual). In cohort 3A, the composite CNS ORR = 49% (95% CI, 32% to 66%), and the CNS ORR in cohort 3B = 33% (95% CI, 10% to 65%). Median progression-free survival was 5.5 and 3.1 months in cohorts 3A and 3B, respectively; median survival was 13.3 and 15.1 months. Diarrhea was the most common grade 3 toxicity (29% in cohorts 3A and 3B). Neratinib plus capecitabine is active against refractory, HER2-positive breast cancer brain metastases, adding additional evidence that the efficacy of HER2-directed therapy in the brain is enhanced by chemotherapy. For optimal tolerance, efforts to minimize diarrhea are warranted.