Salama, Mohamed E.

Yacoub et al report excellent responses to pegylated interferon alfa-2a in patients with hydroxyurea-resistant/intolerant polycythemia vera or essential thrombocythemia.

Abstract Introduction Our objective is to improve the assistive role of a deep, densely connected convolutional neural network (CNN) to hematopathologists in differentiating histologic images of Burkitt lymphoma (BL) from diffuse large B-cell lymphoma (DLBCL). We hypothesized that for the majority of cases, a CNN would accurately differentiate BL from DLBCL and attempted to identify design and input variables to improve performance. Methods and Materials In total, 3608 images of BL were collected from 18 cases and 2,071 images of DLBCL were collected from 20 cases using Aperio Image Scope (Leica Biosystems, Buffalo Grove, IL). Cases were randomized into training and unknowns, and three separate CNNs were trained and applied to unknown images. Networks differed by either the size in pixels of the images (network 1 used 32 × 32 while networks 2 and 3 used 224 × 224) or absence of color (in network 2). Decreased numbers of training images were used to evaluate network 3. Results Network 3 performed the best with 17 of 19 (89%) cases classified correctly, 10 with 100% of images correct. Overall, network 3 had 88% accuracy, 80% precision, 81% recall, and an F1 score of 0.81. Receiver operating characteristic (ROC) curve analysis showed an area under the curve (AUC) of 0.89 for DLBCL and 0.88 for BL. Networks 1 and 2 performed more poorly with F1 scores of 0.63 and 0.39, respectively. Decreasing training images by one-half and one-fourth resulted in decreased F1 scores of 0.63 and 0.60, respectively. Conclusions CNNs are promising augmented human intelligence tools for differentiating DLBCL from BL cases and potentially could contribute to improved workflow in the laboratory. More training images, larger image size, and color significantly improved performance of the CNN because of the added information available to the network. More research is under way to optimize the performance of CNNs’ lymphoma application.

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.

Abstract Myeloproliferative neoplasms (MPN), including polycythemia vera, essential thrombocythemia, and primary myelofibrosis, have a propensity to evolve into accelerated and blast-phase disease (MPN-AP/BP), carrying a dismal prognosis. Conventional antileukemia therapy has limited efficacy in this setting. Thus, MPN-AP/BP is an urgent unmet clinical need. Modest responses to hypomethylating agents and single-agent ruxolitinib have been reported. More recently, combination of ruxolitinib and decitabine has demonstrated synergistic in vitro activity in human and murine systems. These observations led us to conduct a phase 1 study to explore the safety of combined decitabine and dose-escalated ruxolitinib in patients with MPN-AP/BP. A total of 21 patients were accrued to this multicenter study. Ruxolitinib was administered at doses of 10, 15, 25, or 50 mg twice daily in combination with decitabine (20 mg/m2 per day for 5 days) in 28-day cycles. The maximum tolerated dose was not reached. The most common reasons for study discontinuation were toxicity/adverse events (37%) and disease progression (21%). Fourteen patients died during study treatment period or follow-up. The median overall survival for patients on study was 7.9 months (95% confidence interval, 4.1-not reached). Among evaluable patients, the overall response rate by protocol-defined criteria (complete remission with incomplete count recovery + partial remission) was 9/17 (53%) and by intention-to-treat analysis was 9/21 (42.9%). The combination of decitabine and ruxolitinib was generally well tolerated by patients with MPN-AP/BP and demonstrates potentially promising clinical activity. A phase 2 trial evaluating the efficacy of this combination regimen is ongoing within the Myeloproliferative Disorder Research Consortium.

Key PointsSVECs support the development of hematopoiesis in MF. LCs and SVECs each have distinct properties and functions that differ between normal and MF spleens.

Abstract Given the frequent and largely incurable occurrence of multiple myeloma, identification of germline genetic mutations that predispose cells to multiple myeloma may provide insight into disease etiology and the developmental mechanisms of its cell of origin, the plasma cell (PC). Here, we identified familial and early-onset multiple myeloma kindreds with truncating mutations in lysine-specific demethylase 1 (LSD1/KDM1A), an epigenetic transcriptional repressor that primarily demethylates histone H3 on lysine 4 and regulates hematopoietic stem cell self-renewal. In addition, we found higher rates of germline truncating and predicted deleterious missense KDM1A mutations in patients with multiple myeloma unselected for family history compared with controls. Both monoclonal gammopathy of undetermined significance (MGUS) and multiple myeloma cells have significantly lower KDM1A transcript levels compared with normal PCs. Transcriptome analysis of multiple myeloma cells from KDM1A mutation carriers shows enrichment of pathways and MYC target genes previously associated with myeloma pathogenesis. In mice, antigen challenge followed by pharmacologic inhibition of KDM1A promoted PC expansion, enhanced secondary immune response, elicited appearance of serum paraprotein, and mediated upregulation of MYC transcriptional targets. These changes are consistent with the development of MGUS. Collectively, our findings show that KDM1A is the first autosomal-dominant multiple myeloma germline predisposition gene providing new insights into its mechanistic roles as a tumor suppressor during post-germinal center B-cell differentiation. Significance: KDM1A is the first germline autosomal dominant predisposition gene identified in multiple myeloma and provides new insights into multiple myeloma etiology and the mechanistic role of KDM1A as a tumor suppressor during post-germinal center B-cell differentiation. Cancer Res; 78(10); 2747–59. ©2018 AACR.