**3. Benign hematological disorders**

with cytarabine-anthracycline-based induction and high-dose consolidation are considered to have relatively good prognosis compared to other leukemia subtypes, with 10-year OS, disease-free survival (DFS), and event-free survival (EFS) of 63.9, 54.8, and 49.9%, respectively [19]. Nevertheless, 40–45% of these patients eventually relapse and die of their disease. Integration of cytogenetic results with molecular genetics and epigenetic data refines the risk stratification of CBF AML. Several variables might worse prognosis of these patients, such as the level of the respective fusion transcripts *RUNX1-RUNX1T1* and *CBFB-MYH11* [20]. Some authors even suggested that *FLT3*-ITDs carriers constitute a biologically distinct group of APL

Almost half of AML is normal cytogenetically, and this subgroup shows a remarkable heterogeneity in terms of genetic mutations and response to therapy. In patients with normal karyotype, as well as in cases with chromosome abnormalities with intermediate prognosis, the intensity of therapy is driven by the prognostic subgroup. Therefore, the current standard of care combines cytogenetic results with testing for mutations in *FLT3, NPM1, CEBPA,* and *KIT* to precise the risk. The presence of *NPM1* and *CEBPA* gene mutations is associated with

Several other gene alterations (mutations in *WT1*, *RUNX1*, *ASXL1*, *TP53*, *IDH1, IDH2, DNMT3A* genes, partial tandem duplication of *MLL* gene, overexpression of *BAALC*, *MN1, EVI1, ERG*, *WT1*) have also been demonstrated to predict prognosis and probably will play a role in future risk stratification, although some of these have not been confirmed in multiple

About 30% of AML have an unfavorable karyotype, and if treated with conventional chemotherapy, a complete response rate of about 50% and a 5-year OS of 10–20% are expected. The best chance for patients with an unfavorable karyotype who achieve a complete response is

A major achievement is the incorporation of genetic and molecular data in the current classification systems. However, the major principle of the World Health Organization (WHO) Classification of tumors of hematopoietic and lymphoid tissues (2016) is to integrate these data with essential clinical features, morphology, and immunophenotyping in order to define distinct disease entities of clinical significance [25]. Morphology is the gold standard, and though it has been the classical tool for diagnosis and classification, it is routinely performed by subjective microscopic evaluation and is strongly dependent on the morphologist's expertise. To extract more accurate and detailed information from patient tissue samples, digital pathology integrated with advances in machine learning is emerging as a powerful tool to enhance morphology-based decisions. In the fifth chapter in this book, Cecilia Lantos et al. provide up to date information about the possibilities that computational histology can provide to improve leukemia diagnosis with an automated biologically meaningful pattern recognition, as well as the additional contribution of deep-learning approach for a higher accuracy. The authors claim that if mathematical pattern recognition methods that recognize cellular phenotypes from microscopic slides and define how morphological features relate to clinical genetic data and protein signatures, this could significantly speed up leukemia

a favorable prognosis, however, only in the absence of *FLT3-*ITD [22].

studies or established as the standard of care [23].

the allogeneic HSCT [24].

patients [21].

8 Hematology - Latest Research and Clinical Advances

"Benign hematology isn't so benign" if we use the words of Prof. Alice Ma in ASH Clinical News (2015). Clotting disorders, anemias, thrombocytopenias, and so on may present as serious as malignant disorders and are a field of significant progress too.
