Diagnosis Of Leukemia Research Articles

NGS-based molecular genetics of leukemia-a powerful and decentralized approach

The diagnosis of acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS), originally based on morphological assessment alone, has to bring together more and more disciplines. Today, modern AML/MDS diagnostics rely on cytomorphology, cytochemistry, immunophenotyping, cytogenetics, and molecular genetics. Only the integration of all these methods allows acomprehensive and complementary characterization of each case, which is aprerequisite for optimal AML/MDS diagnosis and treatment. In the following, we present why multidisciplinary and local diagnosis is essential today and will become even more important in the future, especially in the context of precision medicine. We present our idea and strategy implemented at Augsburg University Hospital, which has realized multidisciplinary diagnostics in AML/MDS in an interdisciplinary and decentralized approach. In particular, this includes the recent technical advances that molecular genetics provides with modern methods. The enormous amount of data generated by these techniques represents amajor challenge, but also aunique opportunity. We will reflect on how this increase in knowledge can be integrated into routine practice to lead the way for personalized medicine in AML/MDS to improve patient care.

Risk profiles and incidence of cardiovascular events across different cancer types

Cancer survivors are at increased risk for cardiovascular (CV) disease, although additional data are needed to better understand the incidence of CV events across different malignancies. This study sought to characterize the incidence of major adverse CV events [myocardial infarction, stroke, unstable angina (MACE), or heart failure (HF)] across multiple cancer types after cancer diagnosis. Patients were identified from a USA-based administrative claims database who had index cancer diagnoses of breast, lung, prostate, melanoma, myeloma, kidney, colorectal, leukemia, or lymphoma between 2011 and 2019, with continuous enrollment for ≥12 months before diagnosis. Baseline CV risk factors and incidence rates of CV events post-index were identified for each cancer. Multivariable Cox hazards models assessed the cumulative incidence of MACE, accounting for baseline risk factors. Among 839 934 patients across nine cancer types, CV risk factors were prevalent. The cumulative incidence of MACE was highest in lung cancer and myeloma, and lowest in breast cancer, prostate cancer, and melanoma. MACE cumulative incidence for lung cancer was 26% by 4 years (2.7-fold higher relative to breast cancer). The incidence of stroke was especially pronounced in lung cancer, while HF was highest in myeloma and lung cancer. CV events were especially increased following certain cancer diagnoses, even after accounting for baseline risk factors. Understanding the variable patient characteristics and associated CV events across different cancers can help target appropriate CV risk factor modification and develop strategies to minimize adverse CV events and improve patient outcomes.

Tooth Abnormalities and Their Age-Dependent Occurrence in Leukemia Survivors.

The multidrug nature of anticancer treatment and different treatment protocols used in the studies are likely to be a major limitation in establishing real risk factors determining the occurrence of dental abnormalities. The authors aimed to establish a relationship between the duration and the dose of chemotherapy and the number of tooth adverse effects in the group receiving the same treatment. Of the 40 anticancer therapy recipients who attended the outpatient dental clinic, 7 leukemia survivors receiving the treatment according to the ALL IC-BFM 2002 protocol were selected. The study group consisted of four females and three males aged 92 to 207 months at the time of dental examination and 29 to 91 months at leukemia diagnosis. As a result of the clinical and radiological examination, dental abnormalities such as agenesis, tooth size reduction, root abnormalities, and taurodontia were identified, and the medical records of all survivors were reviewed in terms of drugs administered, their doses, and treatment schedules. No correlation was observed between the treatment duration of an intensive therapy, the entire therapy, and the number of tooth abnormalities. No relationship was also found between the number of dental abnormalities and the cumulative dose of vincristine, L-asparaginase, methotrexate, cyclophosphamide, cytarabine, and 6-mercaptopurine. The age at the onset of antineoplastic therapy is likely to be the strongest risk factor for toxic injury during tooth development.

Leukemia Cutis-The Current View on Pathogenesis, Diagnosis, and Treatment.

Leukemia cutis (LC) is defined as the leukemic infiltration of the epidermis, the dermis, and the subcutaneous tissue. Leukemia cutis may follow or occur simultaneously with the diagnosis of systemic leukemia. However, cutaneous lesions are occasionally diagnosed as the primary manifestation of leukemia. Leukemic skin infiltrations demonstrate considerable variation regarding a number of changes, distribution, and morphology. The highest incidence of LC is observed in chronic lymphocytic leukemia, monocytic and myelomonocytic acute myeloid leukemia, and T-cell lineage leukemia. Although the pathogenic mechanism of the invasion of leukemic cells into the skin is not well understood, chemokine receptors and adhesion molecules as well as the genetic characteristics of leukemia are thought to play a role. Leukemic skin lesions may be localized or disseminated and may occur alone or in combination on any site of the skin, most frequently in the trunk and extremities. The most common clinical presentations of leukemia cutis are papules, nodules, macules, plaques, and ulcers. In most patients, the complete or partial resolution of cutaneous infiltrations occurs simultaneously with hematologic remission. However, in patients with resistant disease or recurrent skin infiltration, local radiotherapy can be used. This review presents recent data on the pathogenesis, diagnosis, and treatment of leukemic skin involvement in different types of leukemia.

Automated Deep Learning Model with Optimization Mechanism for Segmenting Leukemia from Blood Smear Images

The advancement of digital microscopic scanning has made the study of image processing as well as categorization an exciting field of diagnostic studies. The literature describes a number of methods for detecting acute lymphocytic leukemia (ALL) using blood smear pictures. The goal of this research is to create an efficient approach for segmenting and detecting leukemia. This research has created a leukemia diagnosis module predicated on deep learning (DL) using blood smear pictures. Pre-processing, segmentation, extraction of features, and classification are performed here by the identification scheme. The presented hybrid model of African Buffalo and African Vulture Optimization (AB-AVO) performs the segmentation process, in which cytoplasm and nucleus regions are segmented. The Local Directional Pattern (LDP) and color histogram characteristics have been then retrieved from the segmented pictures and given into the presented Recurrent Neural Network (RNN) for categorization. The ALL-IDB1 and ALL-IDB2 databases’ blood smear pictures are taken into account for the investigation and assessed using metrics including F1-score, sensitivity, dice coefficient, precision, specificity, recall, and accuracy. The presented AB-AVO-RNN approach exhibits 100% accuracy, according to simulation data. Modern methodologies are used to compare the effectiveness of the suggested AB-AVO-RNN methodology. The investigation demonstrates that the suggested classifier performs comparably better and can identify leukemia from blood smear pictures.

NIMG-51. RADIOGRAPHIC IMPROVEMENT IN A PATIENT WITH OLIGODENDROGLIOMA UNDERGOING TREATMENT FOR ACUTE MYELOID LEUKEMIA

Abstract INTRODUCTION: The standard chemotherapy for treating oligodendroglioma consists of a combination of procarbazine, lomustine, and vincristine (PCV). The combination of hypomethylating agents, like azacitidine, and BCL2 inhibitors like venetoclax have not been formally studied in the treatment of glial tumors. The combination of these two drugs is commonly used to treat acute myeloid leukemia, with IDH mutant disease being a particularly sensitive subtype. CASE REPORT: A 48-year-old man was diagnosed with a left frontal, grade II oligodendroglioma following multiple episodes of speech arrest and right-hand numbness. He was subsequently treated with six cycles of PCV with a minor response. After slow progression over eight years, the tumor started displaying contrast enhancement. While planning for radiation therapy, he presented to the Emergency Department with severe fatigue and shortness of breath. Workup led to a diagnosis of pure erythroid leukemia, with molecular and cytogenetic studies suggesting a neoplasm related to prior exposure to cytotoxic chemotherapy. He was treated with three cycles of venetoclax and azacitidine, but unfortunately, the erythroleukemia progressed after a transient response to this treatment. On the other hand, however, a re-staging MRI brain showed a remarkable reduction of his tumor size of approximately 40% per RANO criteria. The patient was off steroids and remained neurologically asymptomatic during this time. Discussion: The use of azacitidine for the treatment of IDH mutant gliomas has been reported in the literature, with mixed results that might suggest at least some benefit in a subtype of patients. Other studies have also suggested that the BCL2 gene is associated with treatment resistance and tumor recurrence in gliomas. Our case presents an interesting observation that combining a hypomethylating agent with a BCL2 inhibitor might convey clinical and radiographic benefits for patients with relapsed IDH mutant gliomas.

A combined neural network mechanism for categorizing the normal and cancer cells

The dangerous form of acute lymphocytic leukemia damages the bone marrow tissue and white blood cells. Unformed white blood cells multiply and exchange healthy cells in the bone marrow. Everything spreads fast and, if not noticed, can be dangerous in some months. As a result, computer assisted diagnosis of acute lymphocytic leukemia has the possible to protect several lives, but it needs a high-precision categorization of malignant cells, that is difficult due to the graphical similarity of malignant and normal cells. Though deep convolutional neural networks and the classic machine learning algorithms have exposed excellent results in classifying blood cell images, they have not been able to make effective use of the long-term correlation between some important image attributes and image labeling. This study introduced Long short-term memory (LSTM) to solve this difficulty. This study actually combines VGG16 neural networks and LSTM, which provide the VGG16-LSTM framework, which improves the recognizing of image content and learns the structural features of images, as well as initiate big data training in clinical image processing. The transfer learning approach was used to transfer pre-trained weight parameters to the VGG16 area in the ImageNet database, and the custom loss function was used to train and integrate the network quickly and accurately with the weight parameters. Experimental findings reveal that the suggested network approach is more relevant and effective in categorizing cancer cell images than other existing methods.

Automated Deep Learning-Based Diagnosis and Molecular Characterization of Acute Myeloid Leukemia Using Flow Cytometry

The current flow cytometric analysis of blood and bone marrow samples for diagnosis of acute myeloid leukemia (AML) relies heavily on manual intervention in the processing and analysis steps, introducing significant subjectivity into resulting diagnoses and necessitating highly trained personnel. Furthermore, concurrent molecular characterization via cytogenetics and targeted sequencing can take multiple days, delaying patient diagnosis and treatment. Attention-based multi-instance learning models (ABMILMs) are deep learning models that make accurate predictions and generate interpretable insights regarding the classification of a sample from individual events/cells; nonetheless, these models have yet to be applied to flow cytometry data. In this study, we developed a computational pipeline using ABMILMs for the automated diagnosis of AML cases based exclusively on flow cytometric data. Analysis of 1820 flow cytometry samples shows that this pipeline provides accurate diagnoses of acute leukemia (area under the receiver operating characteristic curve [AUROC] 0.961) and accurately differentiates AML vs B- and T-lymphoblastic leukemia (AUROC 0.965). Models for prediction of 9 cytogenetic aberrancies and 32 pathogenic variants in AML provide accurate predictions, particularly for t(15;17)(PML::RARA) [AUROC 0.929], t(8;21)(RUNX1::RUNX1T1) (AUROC 0.814), and NPM1 variants (AUROC 0.807). Finally, we demonstrate how these models generate interpretable insights into which individual flow cytometric events and markers deliver optimal diagnostic utility, providing hematopathologists with a data visualization tool for improved data interpretation, as well as novel biological associations between flow cytometric marker expression and cytogenetic/molecular variants in AML. Our study is the first to illustrate the feasibility of using deep learning-based analysis of flow cytometric data for automated AML diagnosis and molecular characterization.

Platelet-Based Dose-Adjusted Low Molecular Weight Heparin Treatment Is Efficacious and Safe in Newly Diagnosed Acute Leukemia Patients with Catheter-Related Thrombosis

Introduction: Catheter-related thrombosis (CRT) is a major complication in patients with acute leukemia. Severe thrombocytopenia related to either the concomitant disease or therapy underscores the challenge of CRT management, resulting in the use of suboptimal anticoagulation, potentially leading to high rates of short-term complications, including recurrent thromboembolism and hemorrhage. Due to the paucity of data regarding the impact of reduced-dose anticoagulation on this clinical setting, the current study aimed to assess the safety and efficacy of this treatment approach to CRT management in this challenging patient population. Methods: In this retrospective, case-control study, all adult (age ≥18 years) acute leukemia patients, who had arrived at the Rambam Health Care Campus (RHCC, Haifa, Israel) between 1.11.2004-30.6.2022 to receive anti-leukemia therapy via a peripherally inserted central catheter (PICC) line were screened. The inclusion criteria were: patients with acute myeloid, promyelocytic and lymphoid leukemia (AML, APL, and ALL, respectively) and CRT confirmed by Doppler ultrasound within 3 months of leukemia diagnosis. Patients were divided into full- (i.e., 1 mg/kg bid) vs. partial-dose low molecular weight heparin (LMWH) groups. The LMWH dose was based on the platelet count with a cutoff of 50K/L, distinguishing between the selected doses. Primary endpoints: A record of venous thromboembolism (VTE; including pulmonary embolism (PE), deep vein thrombosis (DVT), and catheter-related thrombosis (CRT) recurrence within 12 months as well as a major hemorrhage (including intracranial, retinal and gastrointestinal hemorrhage) within 6 months. Secondary endpoints: blood product requirement (unit/day) before and after the CRT diagnosis (from leukemia diagnosis till CRT identification and 90 days or death (if prior) from the event, respectively) assessed in a self-controlled case series manner. Data were collected from the hospital electronic medical records. Chi-square and paired samples Wilcoxon tests were utilized for statistical analysis. Results: Overall,195 patients [aged 54±15 years; 103 (53%) males], diagnosed with CRT were included. The most frequent leukemia diagnosis was AML (n=149; 76%), with ALL and APL established in 23 (12%) patients each. CRT was diagnosed in the left upper limb in 136 (70%) patients, with 99 (46%) patients suffering from DVT. Eighty-eight (45%) patients were treated with full-dose LMWH. Recurrent VTE was identified in 11(12.5%) patients treated with full-dose LMWH vs. 9 (8.4%) receiving partial-dose LMWH (p-value 0.484), while major hemorrhage was diagnosed in 12 (13.6%) patients from the former group vs. 14 (13.0%) patients from the latter group (table 1). Similar results were documented upon stratification to deep vs. superficial CRT, and the leukemia type. The packed red blood cell (RBC) administration rate was lower in the post-CRT phase (0.12 units/day, IQR 0.06-0.20) compared to the pre-CRT phase (0.20 units/day, IQR 0-0.36; p-value <0.0001). Platelet utilization rates were similar in both the pre-CRT and post-CRT phases (0.05 units/day, IQR 0-0.24 and 0.06 units/day, IQR 0.01-0.16, respectively; p-value 0.126). Comparable results were observed in both full and partial-dose LMWH groups. Conclusion: LMWH dose adjustment based on platelet counts, in newly diagnosed acute leukemia patients with CRT, is not associated with higher rates of VTE recurrence. Moreover, this strategy does not result in major hemorrhagic complications even in patients with severe thrombocytopenia. Notably, RBC infusion requirements were lower in the post-CRT diagnosis phase, while platelet transfusion was similar during the pre- and post-CRT phases. This analysis, including a large number of patients, demonstrates that the adjusted therapy is safe and efficient in this challenging clinical scenario.

Retrospective Review of Patients Treated with Intrathecal Rituximab for Primary and Secondary CNS Lymphoma and Leukemia

Background: Rituximab has been studied and incorporated into the treatment of primary and secondary CNS lymphoma (Grommes 2017; Bobillo 2023). Systemic rituximab, a CD20-targeting monoclonal antibody, is widely used in the treatment of various B-cell malignancies (Klein 2021); however penetration into the central nervous system (CNS) is limited (Petereit 2009; Harjunpää 2001). Although there are conflicting data regarding the efficacy of IV rituximab in primary CNS lymphoma (Schmitt 2019; Ferrari 2022; Bromberg 2019), its role in other B-cell malignancies with CNS involvement has not been adequately described. In particular, its role in treating the cerebrospinal fluid (CSF) space is unclear. The rationale for direct administration of rituximab into the CSF via lumbar puncture or Ommaya reservoir is to allow for consistent treatment of this compartment in order to effectively clear the CSF of lymphoma cells (Kadoch 2014). Limited data exist regarding the efficacy and safety of intrathecal rituximab (Bonney 2011 Aslan 2022; van de Fling 2008; Santiago 2021; Soares 2019). Current intrathecal armamentarium for leptomeningeal lymphoma is very limited and more data on the efficacy and safety of intrathecal rituximab is needed in these patients. Methods: This is a retrospective, single center study of adult patients with a diagnosis of leukemia or lymphoma who received intrathecal (IT) rituximab between 1/1/2016 and 12/31/2022 at Hackensack Meridian Health. Rituximab 25 mg IT was administered weekly either via lumbar puncture or into an Ommaya reservoir. The primary objective of this study was to evaluate the efficacy and safety of IT rituximab. Progression-free survival (PFS) and overall survival (OS) were calculated using the Kaplan Meier method. Descriptive statistics were used to summarize data for categorical and continuous variables. Results: Twelve patients were identified to have received IT rituximab at our institution between 1/1/2016 and 12/31/2022. Median age was 60 years (range 45-81), 54% were male, and 69% white. Disease types included primary CNS lymphoma (67%), diffuse large B-cell lymphoma (17%), PTLD of the CNS (8%), and chronic lymphocytic leukemia (8%). In regards to CNS disease involvement, 58% of patients presented with parenchymal and leptomeningeal disease, 33% with leptomeningeal disease only, and 8% with parenchymal disease only. There were 12 patients who received IT rituximab for the treatment of malignant CNS disease with characteristics and outcomes as shown in Table 1. The median number of IT rituximab treatments administered were 8 (range 2-9). Of the 10 patients evaluable for response, 92% of patients achieved complete response in the CSF following treatment with IT rituximab as evaluated by flow cytometry, cytology, cell count and protein levels in CSF. One patient that did not achieve CSF remission has an unusual case of relapsing/remitting CLL and leukemic meningitis. There were 2 patients unevaluable for response to IT rituximab: one patient received only 2 treatments prior to switching to hospice care; and the other patient had a severe adverse reaction (tightening of throat, flushing, toe numbness) after IT rituximab and was not continued on treatment. At a median follow-up of 19.7 months, median PFS was not reached (NR) (95% CI, 0.5-NR) ( Figure 1). Median OS was 70.1 months (95% CI, 1.2-NR). Six (50%) patients experienced adverse reactions from IT rituximab treatment, including perioral numbness, headache, nausea, and flushing ( Table 1). No treatment-related deaths occurred. One (8%) patient experienced a severe adverse reaction following administration of IT rituximab. This patient developed tightening of the throat, flushing, and toe numbness after the first IT rituximab and symptoms resolved with administration of hydrocortisone and diphenhydramine. No further IT rituximab doses were administered. Another patient experienced a mild reaction consisting of headache, fatigue, nausea, diaphoresis, hypotension, and bradycardia following the fourth dose of IT rituximab in combination with IV rituximab, but was able to complete his treatment course. Conclusions: The use of IT rituximab appears safe and showed promising activity in B cell malignancies, specifically as part of induction in patients with leptomeningeal disease and PCNSL as those patients achieved cytologic response.

Machine Learning-Facilitated Cross-Panel Automatic Classification at Both Sample and Cell Level in Acute Myeloid Leukemia

Introduction Flow cytometry (FC) immunophenotyping is essential for accurate and prompt diagnosis of acute myeloid leukemia (AML), with up to 95% of AML patients showing detectable aberrant immunophenotype (IP) by FC. Current manual and subjective analysis of clinical FC data, however, lacks reproducibility and leads to interpretation variation. In previous studies, we showed that supervised machine learning (ML) approaches can detect IP abnormalities and identify the acute leukemia subclassification accurately at the specimen level. We also found that comparable classification performance can be achieved using only a subset of markers of the reagent panel (Ko et al., 2018; Monaghan et al., 2022). To interpret the ML classification results, we have developed data clustering and discriminative learning methods to identify and visualize the diagnostic cell populations in a 2D space for hematopathology review (Ji et al., 2020; Lee et al., 2018). Built upon these approaches, this study reports our findings on developing an interpretable cross-panel ML classification model to support timely and accurate diagnosis and subclassification of AML. Materials and Method The AML diagnosis model at the sample level utilized FC data obtained from 53 bone marrow (BM) aspirate samples from Roswell Park Comprehensive Cancer Center (RPCCC) and 50 BM aspirate samples from UPMC. Data from RPCCC were acquired using a Beckman Coulter Navios EX and measured with the ClearLLab10C panel; while data from UPMC were acquired using a BD FACSCantoII and measured with a UPMC-developed diagnostic panel. We used only the common parameters collected within these 103 FC data sets and applied a previously published Gaussian Mixture Model-Support Vector Machine (GMM-SVM) based approach to train a cross panel sample-level classification model for AML diagnosis (Ko et al., 2018; Monaghan et al., 2022). Three-fold cross-validation was conducted to assess the model performance, using sensitivity, specificity, classification accuracy, and area under the receiver operating characteristic curve (AUC). For the cell-level visualization and analysis, all samples in the training set were pooled together, before a uniform manifold approximation and projection (UMAP) process was applied to transform and visualize the cell populations within the samples. Specifically, 26 non-neoplastic and 27 AML samples from the M2 panel of the ClearLLab10C from the RPCCC cohort were combined for the UMAP visualization. Then the FlowSOM unsupervised clustering method and the DAFI semi-supervised clustering method (Lee et al., 2018) were applied to identify all of the cell populations that have significant differences between the AML samples and the non-neoplastic controls. Results The cross-panel AML versus non-neoplastic classifier, combining FC data from UPMC and RPCCC achieved exceptional performance. The AML detection model using all parameters on the ClearLLab10C panel achieved an AUC of 100%. When using only the overlapping parameters between the ClearLLab10C and the UPMC panels, the classification model achieved an AUC of 99.5%, while the cross-panel classifier achieved an AUC of 99.1% (Table 1). Specimen-level visualization confirmed the separation of AML versus non-neoplastic controls. The UMAP plots, created by pooling non-neoplastic and AML samples, exhibited distinct patterns for different AML subtypes (M1 to M5) and a genetic mutation (e.g., AML IDH1), which were distinct from control samples. Using the automated gating analysis, cell populations identified as different between AML and control samples were characterized and visualized in 2D-by-2D scatter plots for straightforward interpretation and assessment based on their immunophenotype. Conclusion Our study demonstrated the feasibility to develop and assess a ML classification model for AML detection and diagnosis using overlapping markers in FC data across reagent panels and sites. Advanced visualization and automated gating analysis enabled the interpretation of the ML classification by identifying and characterizing the diagnostic cell populations and their phenotypic heterogeneity, providing insights into AML heterogeneity at both the specimen and the single cell level. These findings highlight the potential of ML capability for improving patient care by timely and accurate diagnosis and subclassification of AML.

Toward a Measurable Residual Disease-Based Algorithm for Selection of Allogeneic Hematopoietic Stem Cell Transplant in Acute Myeloid Leukemia

Background: The diagnosis and management of acute myeloid leukemia (AML) have transformed in recent years with an improved understanding of leukemogenesis and improved methods of measurable residual disease (MRD) detection. Current treatment algorithms include consideration of allogeneic hematopoietic stem cell transplant (HCT) for intermediate and adverse-risk AML. MRD negativity is widely accepted as a predictive and prognostic marker for disease-free survival and overall survival. However, the clinical use of MRD status in consolidative HCT planning remains unclear. Methods: Two hundred thirty-three consecutive patients achieving complete remission (CR) as defined in ELN 2022 after receiving therapy at Massey Cancer Center were analyzed. MRD status was determined either by multiparameter flow cytometry, PCR-based sequencing, and/or next-generation sequencing. Patients were analyzed via retrospective chart review to obtain baseline characteristics, molecular and cytogenetic profiles, and survival data. Median overall survival (mOS) was calculated and compared using the Kaplan-Meier method. Continuous variables were compared using one-way ANOVA. Categorical variables were analyzed using Fisher's exact test. Results: In total,109 patients achieved MRD-negative CR and 124 patients achieved an MRD-positive CR. Of the MRD-negative patients, 42 went received HCT (39%). In the MRD-positive cohort, 44 underwent HCT (35%). Among the MRD negative group, those receiving HCT tended to be younger (52.1 y. [range: 24.8-71.7] vs 61. 8 y. (range: 27.1-82.9) p = 0.026) with fewer co-morbidities (Charlson comorbidity index [CCI] score: 3 [range: 2-9] vs 4 (range: 0-14), p = 0.322). We observed a similar trend in those who had MRD positivity at best response or at the time of transplant (age 55 y. [range: 24.5-72.2] vs 62 y. [range: 25.1-88], p = 0.142; CCI 4 [range: 2-7] vs 5 (range 2-11], p = 0.014). In the MRD-negative/no HCT group, 49.3% had ELN favorable-risk disease at initial AML diagnosis, 28.4% had intermediate risk, and 22.4% had adverse risk. In the MRD-negative/HCT group, 31% were favorable risk, 33.3% intermediate risk, and 35.7% were adverse risk. In the MRD-positive/HCT group, 46.3% favorable risk, 23.8% intermediate risk, and 29.5% were adverse risk. There were no significant differences in ELN risk between any group. MRD-positive patients who underwent HCT had a mOS of 25.7 m. vs 20.7 m. for no HCT (HR=0.68 [95% CI, 0.41-1.11], p = 0.125). The median OS was not reached for MRD-negative patients, regardless of receipt of HCT (HR=0.61 [95% CI, 0.30-1.23], p = 0.168). Patients undergoing HCT in MRD-negative CR had improved OS compared to those undergoing HCT with persistent MRD positivity (p = 0.006). Discussion: MRD positivity is a known predictor of relapse and decreased survival despite HCT. However, the use of MRD as a decision point predicting which patients will benefit from transplant a priori needs to be better understood. Our data suggest that MRD negativity has a more significant impact on overall prognosis than HCT status.Likewise, the negative impact of MRD positivity was not significantly abrogated by subsequent HCT. This highlights the need for refinement in HCT selection beyond ELN 2022 intermediate/adverse-risk patients, and new validated algorithms incorporating MRD status are warranted.

Lost and Found: A Mixed Methods Study on Treatment Abandonment in Acute Leukemia

Introduction: Acute leukemias represent an important public health problem due to their high morbidity and mortality. Treatment failure due to abandonment or loss of follow-up (LOF) is a contributing factor to poor outcomes. Objective: to determine the causes and outcomes of LOF in a cohort of adults with a diagnosis of acute leukemia and its impact on survival in a limited-resource setting. Materials and Methods: We conducted a prospective, single-center cohort study from 2016 to 2022 in the university hospital; this institution is subsidized with funds from private and public sources. We included patients undergoing induction, consolidation, and maintenance phases of treatment for acute lymphoblastic (ALL) and acute myeloid leukemia (AML). LOF events were defined as discontinuation of follow-up beyond the last documented visit in the electronic medical records, with a timeframe of 2 to 4 weeks during the induction and 4 to 8 weeks in the consolidation or maintenance phases. Patients with LOF were contacted via telephone. Those who responded underwent a structured interview that assessed factors potentially contributing to the discontinuation of treatment. Treatment response, event-free survival (EFS), and overall survival (OS) in patients who lost follow-up were also assessed. Results: From January 1, 2016, to October 31, 2022, a total of 391 patients were treated n=99 patients were found to have lost follow-up during first-line treatment, and 73 patients, (%) answered the phone call, 75.3% had been diagnosed with ALL, and 24.7% with AML. Of the 55 patients diagnosed with ALL, 52.7% were women, the median age was 26 years (range 18-86), and 96.4% were classified as a high-risk disease according to age, white blood cell count at diagnosis, CNS involvement, or steroid response. Most patients lived in an urban area (90.9%), and 74.5% didn't have healthcare coverage. Among the patients diagnosed with AML, 50% were male, the median age of diagnosis was 33 years (range 23-67), 94.4% resided in urban areas, and 83.3% of the patients did not have healthcare coverage, thus and molecular risk could not be determined due to the high cost of comprehensive genetic analysis. All patients were classified as low socioeconomic status, with a median income of $130 US dollars per month, (range, 81 -378 USD) for the ALL cohort and $180 USD for the AML group. Based on the findings of the telephone survey, it was evident that economic constraints were the primary driver of treatment abandonment in both ALL and AML patients, accounting for 92.8% and 94.4% of cases, respectively. Specifically, 92.7% of the ALL group expressed difficulties in covering the costs of treatment, while 83.3% in the AML group faced similar challenges. For ALL patients LOF was more frequent during the maintenance phase, (61.8%) compared to consolidation in AML patients (44%). Despite telephone contact, 76.4% of those with ALL and 77.8% of those with AML, could not continue follow-up. Concerning disease status, 11.1% of patients diagnosed with ALL and 34.5% of patients with AML that resumed follow-up after the phone call were diagnosed with relapsed disease. OS at 2 years from the time LOF was identified 49.1% for ALL and 61.1% for AML. Additionally, the 2-year EFS rates were 43.6% for ALL and 44.4% for AML. Conclusions: Lack of access to financial resources and catastrophic healthcare expenditures were the main reason for abandoning treatment in patients with acute leukemia. Despite the loss of follow-up, the overall survival in both groups of patients showed favorable long-term outcomes; however, risk assessment strategies should be employed, and support programs implemented to improve outcomes in vulnerable populations.

Intensive Chemotherapy (IC) Followed By Oral Azacitidine (AZA) Maintenance Versus Venetoclax (VEN) Plus AZA for Patients (pts) with Acute Myeloid Leukemia (AML): Retrospective Analysis of an Electronic Medical Record (EMR) Database in the United States

Introduction:Frontline therapy with IC is usually recommended for eligible pts with newly diagnosed (ND) AML. Pts who achieve remission with induction and are transplant-ineligible can receive oral AZA (Oral-AZA) maintenance. For pts ineligible for IC, VEN with injectable AZA (VEN-AZA) is the recommended treatment. Selection of frontline therapy can be challenging for pts considered reasonable candidates for either IC or VEN-AZA. A prior analysis suggested that survival outcomes may be comparable in matched pts with ND AML treated with either IC or VEN-AZA (Zeidan AM, et al, Annal Hematol 2023;102:749-754). However, the impact of Oral-AZA maintenance was not accounted for. In this analysis, we compare the survival outcomes of pts who received IC followed by Oral-AZA maintenance (IC-Oral-AZA) versus those who received frontline VEN-AZA and achieved remission, before and after matching for baseline and clinical characteristics Methods:In this retrospective, observational, study using Flatiron Health's database, de-identified data from adult pts with ND AML who received either frontline IC-Oral-AZA or frontline VEN-AZA and achieved remission were analyzed. Exclusion criteria included diagnosis of acute promyelocytic leukemia, treatment with arsenic trioxide/tretinoin, clinical trial participation, or prior hematopoietic stem cell transplant (HSCT). Treatment with Oral-AZA for IC-Oral-AZA pts had to commence prior to a 2nd line of therapy or disease relapse, and after Sept 1, 2020. Cohorts were matched 1:1 on baseline demographic and clinical characteristics using propensity scores to minimize differences. Matched characteristics included sex, age at diagnosis (18-64, 65-74, ≥ 75 years), race/ethnicity, region, ECOG performance status (PS), BMI, previous diagnosis of myelodysplastic syndromes or myeloproliferative neoplasm, cytogenetic risk category, whether AML is treatment-related or was diagnosed as mixed phenotype. Baseline molecular data were insufficient. Relapse-free survival (RFS) and overall survival (OS) were estimated for unmatched and matched cohorts using Kaplan-Meier analysis from date of remission, censoring pts at the time of HSCT/end of follow-up. Cox regression analyses were conducted for RFS and OS, both on an unadjusted basis and after adjusting for any remaining differences in age between the matched cohorts. Confirmatory landmark analysis is being conducted. Results:In total, 50 IC-Oral-AZA and 403 VEN-AZA pts met the selection criteria prior to matching. The matching process yielded 33 pts in each cohort, with median follow-up from start of therapy of 13.1 and 11.2 months for IC-Oral-AZA and VEN-AZA cohorts, respectively. The cohorts matched well on median age at diagnosis (63 years for both), sex (54.5%, 51.5% males), ECOG PS 2+ (18.2%, 21.2%), cytogenetic risk (46.7% poor, 26.7% intermediate, 26.7% favorable risk for both groups, among those with known risk status), whether AML was treatment-related (12.1%, 21.2%), race/ethnicity, region, and BMI. Three pts (9.1%) in the IC-Oral-AZA cohort and 8 (24.2%) in the VEN-AZA cohort received HSCT during follow-up. Median time from remission to start of Oral-AZA for the IC-Oral-AZA cohort was 128 days. For the unmatched cohorts, median RFS was significantly longer for IC-Oral-AZA (20.0 months [95% CI, 13.4-not reached (NR)]) than for VEN-AZA (9.3 months [8.3-10.7]; P<0.001) as was median OS (IC-Oral-AZA, NR [15.5-NR]) versus VEN-AZA (13.4 months [10.7-15.5]; P < 0.001). For the matched cohorts, median RFS was significantly longer for IC-Oral-AZA (20.0 months [95% CI, 9.9-NR]) than for VEN-AZA (8.7 months [4.1-15.5]; P = 0.002) as was median OS (IC-Oral-AZA, NR [14.6-NR] versus VEN-AZA, 15.2 months [7.4-19.7]; P = 0.002) (Figures A and B). The Cox analyses confirmed the superior outcomes of the IC-Oral-AZA cohort for both RFS (unadjusted HR, 0.33 [95% CI, 0.17-0.65]; P < 0.01, adjusted HR, 0.34 [0.17-0.70]; P < 0.01) and OS (unadjusted HR, 0.28 [0.11-0.68]; P < 0.01, adjusted HR, 0.29 [0.11-0.74]; P < 0.01). Conclusion:Pts treated with frontline IC followed by Oral-AZA as maintenance had significantly better survival outcomes than pts who were treated and achieved remission with frontline VEN-AZA. These results suggest that pts with ND AML eligible for frontline IC should receive this regimen followed by Oral-AZA maintenance, if they are not candidates for transplant at remission.

Clime: CLL Immune Micro-Environment and Risk of Infections Pre- and Post-Diagnosis: A Population-Based Study

Background: Immune dysfunction precedes a diagnosis of chronic lymphocytic leukemia (CLL) in some patients. More information is needed about which patients experience immune dysfunction and how it impacts disease progression, as well as the development of post-diagnosis infections and second malignancies. The objectives of this study were to 1) examine the rate of pre-diagnosis infection for individuals diagnosed with CLL compared to population controls, 2) evaluate the impact of pre-diagnosis infection on time to first treatment, 3) compare the rate of immune dysfunction before the start of treatment by stage, and 4) compare the rate of immune dysfunction before the start of treatment for each year post-treatment. Methods: We used a retrospective matched cohort study design that included all individuals in Manitoba diagnosed with CLL from 2006 to 2019. Individuals were identified from the Manitoba CLL database and the Manitoba Cancer Registry (MCR). Individuals with CLL were matched by age, sex, and area of residence (urban versus rural) to population controls identified from the Manitoba Health Population Registry. Provincial pharmacy, hospital admissions, medical claims, and MCR data were used to determine infection, second malignancies, and IgR therapy. Immune dysfunction was defined as two or more infections, a diagnosis with a secondary malignancy, or receipt of IgR therapy. The pre-diagnosis time period was defined as three to 15 months prior to diagnosis. The pre-treatment time period was defined as 12 months prior to treatment. The following analyses were used: conditional logistic regression for objective 1, competing risk for objective 2, logistic regression for objective 3, and logistic GEE models for objective 4. Results: A total of 1,776 CLL patients were matched to 17,760 non-CLL controls. Sixty-two percent were male, the mean age at diagnosis was 71 years, and 62% lived in an urban area. Fifty-one percent of CLL patients had a pre-diagnosis infection compared to 46% of controls. Compared to individuals with no prior infection, those who had a prior infection were more likely to be diagnosed with CLL (Odds Ratio (OR) 1.17, 95% confidence Interval (CI) 1.03, 1.33 for one infection, OR 1.29, 95% CI 1.15, 1.45 for two or more infections). After adjusting for the number of outpatient visits and comorbidity, individuals with two or more infections were 14% more likely to be diagnosed with CLL (OR 1.14, 95% CI 1.00-1.30). Conclusions: We found that infections that occurred in the three to 15 months before diagnosis increased the odds of being diagnosed with CLL. Work is ongoing examining time to treatment, immune dysfunction by stage, as well comparing immune dysfunction before and after treatment.

RNA Fusions and Their Association with DNA Alterations in Myeloid Neoplasia Patients Identified By a Single Tube Multimodal Comprehensive Genomic Profiling Test

Background: Recent updates into NCCN professional guidelines have included adding several genomic biomarkers for myeloid disorders. Detecting SNVs, indels, select CNVs and genomic rearrangements in a single comprehensive genomic profiling test is invaluable for clinical care. Moreover, WHO recognizes 23 genomic rearrangements or fusions which define subclasses of AML, MDS/MPN and related neoplasms, and their detection is essential for patient management. Here we present joint prevalence data of gene fusions and other genomic alterations in myeloid disorders in a cohort of 312 patients analyzed by a CLIA grade single-tube NGS assay capable of concurrent analysis of DNA and RNA alterations. Methods: Total nucleic acid (TNA) from bone marrow or peripheral blood was analyzed by a CLIA grade custom amplicon-based multimodal NGS test reporting DNA mutations in 126 genes and gains/losses in 17 genes by DNA-seq, and RNA fusions from 40 genes by RNA-seq. Libraries were sequenced on a NovaSeq6000 instrument, and fusions were called with in-house developed BI pipeline using the distribution of AI-assisted fusion confidence scores to improve the signal to noise discrimination for fusion calls before validation. Deidentified patient data was used according to an IRB approved protocol. Results: Analytical validation of RNA fusion calling against FISH and Sanger-seq in 74 hematologic disorder samples demonstrated 96.7% sensitivity and 98.2% specificity with 100% reproducibility. This improved fusion detection module was added to our CLIA validated NGS assay, which at tumor purity of >20%, already detects SNVs (VAF >3%), Indels (<81 bp, VAF >3%) and copy number gains/losses with a sensitivity of 98.3%, 98.9 % and 96 % for SNV, Indels and copy number gains/losses, respectively and a specificity of >99.9% for SNV/Indels and 98.4% for gene gains/losses. Next, data from a total of 789 patients serially tested with this assay was used to study the distribution of myeloid fusion events in community cases which included 312 adult patients with confirmed/suspected myeloid disorders including AML, CML, MDS/MPN and other disorders while 477 patients had a diagnosis of lymphoid leukemias. Fifty five (55) % of the Myeloid leukemia patients were male and 45% female, with a median age of 67.5 (22-87) and 71 (22-89) years, respectively. 27% (84/312) of them presented a gene fusion, of which in 21% of cases it involved a gene from WHO/NCCN fusion gene recommendations. Prevalence for common fusions were 6.7% for BCR::ABL1; 5.1% for PML::RARA; 2.6% (8/312) for KMT2A; 1.9% for RUNX1::RUNX1T1; 1.6% CBFB::MYH11; 1.3% for PICALM::MLLT10 and 1 % for NUP98. Other fusions were detected in <1% patients. Fusions of PDGFRA, ETV6, ZNF384, FGFR1 and other genes were also observed. BCR::ABL1 fusions were seen not only in CML patients but also in a patient with AML. For KMT2A, 25% (2/8) of fusions detected by NGS in all patients were confirmed by Sanger-seq but missed by FISH, which correlates with higher sensitivity of the NGS assay. Interestingly, the prevalence of fusion positive cases on the lymphoid leukemia patients (17%, 80/477) tested on the same period of time. Novel fusions were called in ~8% of all patients with high confidence. We validated by Sanger sequencing the CCND2::MGP fusion, which was identified on a patient with AML. This fusion protein is predicted to be oncogenic as it lacks the region from CCND2 including Thr280, a residue required for ccnd2 degradation. Finally, we analyzed the relationship between fusions and recurrently mutated genes. We found that DNA alterations in CALR, EZH2, FLT3, KIT, and ZRSR2 were enriched in fusion positive cases. Conversely alterations on a subset of genes were absent on fusion positive cases and enriched on fusion negative cases. Those included alterations on CEBPA, IDH1/2, KMT2A, MPL, NPM1, BCOR, IKZF1, FBXW7 STAG2, CSF3R, PDGFRA, PHF6, PTPN11. Conclusions: A robust low-noise RNA fusion detection assay coupled us DNA alterations on myeloid disorders in a single assay enables to fully molecularly characterize acute myeloid leukemias and other myeloid disorders. Frequencies of well-known fusions in a small community-based cohort were similar to studies performed in academic settings with subsets of gene alterations being mutually exclusive from fusions. Larger studies are needed to confirm those associations.

Label-Free Cell Detection of Acute Leukemia Using High-Content Morphological Profiling in Flow

Background: Early diagnosis and prompt initiation of appropriate treatment are critical for improving the prognosis of acute leukemia. Acute leukemia is typically diagnosed through microscopic morphological examination of bone marrow smears and flow cytometric immunophenotyping of bone marrow cells stained with fluorophore-conjugated antibodies. However, these diagnostic processes require trained professionals and are time and resource-intensive. Here, we present a novel diagnostic approach using ghost cytometry (GC), a recently developed high-content flow cytometric approach that leverages machine vision-based, stain-free, high-speed morphological characterization and analysis of cells. Methods: Label-free morphological information of cells was acquired as compressive imaging temporal waveforms. These waveforms were directly analyzed using the support vector machine (SVM) algorithm without computational image production, enabling high-speed and accurate information processing. Conventional flow cytometric data modalities, such as forward-scattering (FSC), backscattering (BSC), and fluorescence signals from biological labeling, were used as ground truth labels to identify and annotate blast and normal cells in bone marrow (BM) based on flow cytometric gating schemes. Subsequently, we trained a leukemia cell classifier using the annotated GC (compressive imaging) waveforms and evaluated its classification ability by comparing the predicted results with the ground truth labels. Finally, we applied the trained classifier to unidentified GC waveforms (a test dataset not used for classifier development) to demonstrate the detection of blast cells in BM based on their morphology, without relying on biological labels. Following this protocol, we first trained and tested a machine classifier using a human leukemia NB-4 cell line spiked into BM from a healthy donor. Next, we trained and tested two classifiers using BM aspirates from patients diagnosed with acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML), where CD45 was used as the ground truth marker of leukemic cells. Results: The leukemia cell classifier, developed using a human NB-4 cell line spiked into BM cells from a healthy donor, demonstrated excellent classification accuracy with an ROC-AUC score of 0.9998. The high classification accuracy was reproducible across different spike-in concentrations ranging from 1% to 95%. The classifiers trained using BM cells from patients diagnosed with ALL and AML showed high ROC-AUC scores of 0.986 and 0.941, respectively. When we used the classifiers to evaluate blast cell proportions in test datasets that were not used to develop the classifiers, they were also highly accurate in identifying the percentage of blast cells: 56.8% for ALL with a decision boundary of SVM scores at 0, and 30.4% for AML with a decision boundary of SVM scores at 0.37, when compared to May-Grunwald-Giemsa staining of the same cases by medical technologists at diagnosis (60.0% for ALL and 31.2% for AML). Conclusions: Here, we present a novel diagnostic approach that leverages machine vision-based, stain-free, high-speed morphological characterization and analysis of cells. We demonstrate that the approach can detect leukemic cells from the bone marrow cells of patients diagnosed with ALL and AML, without requiring biological staining. The method presented here holds promise as a rapid, precise, simple, and cost-effective method for the diagnosis of acute leukemia in clinical settings.

Disseminated aspergillosis due to Aspergillus flavus in a pediatric patient with a recent diagnosis of acute lymphoblastic leukemia

Disseminated aspergillosis due to Aspergillus flavus in a pediatric patient with a recent diagnosis of acute lymphoblastic leukemia

Risk of Therapy-Related Acute Myeloid Leukemia (t-AML) in Adolescent and Young Adults Cancer Patients Treated with Chemotherapy with or without Radiotherapy: A National Population-Based Cohort Study Using SEER Registry

INTRODUCTION The risk of treatment-related acute myeloid leukemia (t-AML) in adolescents and young adults (AYA) with first cancers treated with chemotherapy with or without radiation is not well characterized. We analyzed Surveillance, Epidemiology and End Result (SEER) data for the past 45-years (1975-2020) to identify the overall risk of t-AML in treated AYA cancer patients (pts) and by index cancer sites as well as latency periods. METHOD We analyzed NCI's SEER-8 registries. ICD-O-3/WHO-2008 codes were used to identify pts with index cancer types between the ages of 15-39 years (AYA), treated with either chemotherapy or radiotherapy, and with a subsequent diagnosis of acute myeloid leukemia (AML) at least one year after index cancer, between January 1975 and December 2020. The risk of t-AML was estimated by standardized incidence ratio (SIR), defined as the ratio of observed t-AML-cases in the AYA-with-cancer, to the expected number of AML-cases in the matched general US population. A reference rate file for AML incidence in age, gender, race and calendar-year matched US population is available in SEERStat software. A Poisson distribution of observed t-AML was assumed for the calculation of 95% confidence intervals (CIs). A P-value <.05 was considered statistically significant. We stratified the risk of t-AML for index cancer types with ten or more pts, by attained-calendar-years, and latency at multiple time intervals (Table-1) from index cancer diagnosis. RESULT Among 102,879 AYA cancer pts eligible for analysis, 236 pts developed t-AML with SIR of 6.19 (95%CI: 5.43-7.04, P <.05) with excess risk of 1.46 per-10000 person-years (pyrs). (Table-1). We stratified risk for index cancer types with at least ten patients developing t-AML, these included breast, Hodgkin lymphoma (HL), non-Hodgkin lymphoma (NHL), testicular, thyroid, bone and joint, and soft tissue including heart. As a general trend, risk of t-AML was highest during 1-3 years after index cancer diagnosis compared to subsequent latency periods. However, risk for t-AML was significantly greater than the matched-general US population even after 20-40 years of breast cancer, HL, and NHL (Table-1). When stratified by attained calendar-year periods (1975-1985, 1986-1995, 1996-2005, 2006-2015, 2016-2020), risk for t-AML was higher than general population in all time periods. However, overall SIR was comparatively lower for recent time periods, particularly lowest during 2016-2020 [SIR: 3.49 (2.51-4.72, P <.05)]. Notably, the highest risk for t-AML in 2016-2020 was observed with index cancer of soft tissue including heart [SIR: 10.12 (1.23-36.57), P <.05]. CONCLUSION In this study spanning between 1975-2020, we observed the risk of t-AML trending down among AYA in recent years, albeit higher than the age matched general population. The risk of t-AML remained significantly elevated even after 20-40 from diagnosis of breast cancer, HL and NHL. Since these pts likely received radiation and older (higher intensity) chemotherapy based-regimens, it is essential to analyze the long-term impact of newer therapies for these cancers and assess the risk of t-AML in future studies.

Durable Transfusion Independence in Lower-Risk Myelodysplastic Syndrome (LR-MDS) Is Associated with Better Survival: A Population Level Analysis Based on a Large US Health Insurance Claims Database

CONCLUSIONS Red blood cell (RBC) transfusions are commonly used to treat patients with LR-MDS and anemia. RBC transfusion dependence (TD) is an impairment on patients' quality of life and associated with worsened survival (Lemos et al. Sci Rep. 2021). The few approved therapeutic options for LR-MDS have limited efficacy and durability. In this study, the baseline RBC-TD before first-line (1L) and second-line (2L) of therapy, durability of transfusion independence (TI), and associated survival among patients with LR-MDS treated with current standard-of-care therapies were assessed in a large US health insurance claims database. Optum Clinformatics ®, a large single-payer database of administrative health claims for members of large commercial and Medicare Advantage health plans, includes approximately 17-19 million annual covered lives, for a total of >76 million unique lives over a 9-year period. Patients with LR-MDS were identified through 5 relevant International Classification of Diseases, Tenth Revision (ICD-10) diagnosis codes (D46.0, D46.1, D46.9, D46.A, and D46.B) between Oct 2015 and Jun 2022 and were required to have no MDS/acute myeloid leukemia (AML) diagnosis and no use of higher-risk MDS or AML medication before their respective index diagnosis dates. As revised International Prognostic Scoring System or other risk score classification information was not available in the database, ICD-10 diagnosis codes were used as a proxy for the identification of LR-MDS; these codes have been used previously in published studies (Zeidan et al. Leuk Lymphoma. 2019). Lines of treatment were determined based on claims for MDS treatments contained in the database: erythropoietin stimulating agents (ESAs; darbepoetin, epoetin alfa), hypomethylating agents (HMAs; azacitidine, decitabine, decitabine-cedazuridine), lenalidomide, luspatercept, eltrombopag, and cyclosporine. Outcomes of interest included transfusion burden (RBC units [U] in 8 weeks), proportion of patients who were transfusion independent before and after different lines of treatment, time to 8- and 16-week continuous TI, Kaplan-Meier analysis of real-world progression-free survival (starting a next line of treatment, progression to higher-risk MDS or AML, and death were considered events), and overall survival. The analysis included 5662 patients with LR-MDS (median age, 79 years; 57% male; 76% non-Hispanic White, primarily covered by Medicare Advantage) who received ≥1 line of treatment. Overall, 66% and 17% of patients received frontline monotherapy with ESAs and HMAs, respectively. In the 16 weeks before 1L treatment initiation, 35% of patients received ≥1 RBC transfusion. During 1L treatment, 45% of patients received ≥1 RBC transfusion, of whom 49% received >3 U and 24% received >6 U during any 8-week period. 2L treatment consisted of ESA monotherapy (40%), HMA monotherapy (21%), luspatercept (9%), or a combination thereof (19%). In the 16 weeks before 2L treatment initiation, 49% of patients received ≥1 RBC transfusion, and 55% received transfusions during 2L treatment (Table). Among patients receiving ≥1 transfusion during 2L treatment, 61% and 31% had >3 and >6 U/8 weeks, respectively. Transfusion burden was greater for patients with ring sideroblast-positive disease and increased with subsequent lines of therapy. Among 612 patients who received ≥1 transfusion in the 16-week period before 2L, 33% achieved 16-week TI with subsequent therapies. Median real-world progression-free survival from start of 2L was 20.0 months in patients who achieved 16-week TI and 4.1 months in patients who did not ( P < .0001). The median overall survival from start of 2L was 23.4 months overall, and 37.9 months vs 9.3 months among responders becoming RBC-TI vs nonresponders, respectively ( P < .0001; Figure). Among patients with LR-MDS, 35% and 49% were RBC-TD before 1L and 2L of therapy, respectively. Achievement of TI was associated with improved survival, suggesting that RBC-TD is a modifiable predictor of clinical outcomes in LR-MDS. Despite the currently available therapies, RBC-TD after any line of therapy is associated with poorer outcomes. Our study results show that novel therapies that provide durable TI are expected to yield decreased progression and improved survival.