Validation of the Artificial Intelligence Prognostic Scoring System for Myelodysplastic Syndromes in chronic myelomonocytic leukaemia: A novel approach for improved risk stratification.

Chronic myelomonocytic leukaemia (CMML) is a rare haematological disorder characterized by monocytosis and dysplastic changes in myeloid cell lineages. Accurate risk stratification is essential for guiding treatment decisions and assessing prognosis. This study aimed to validate the Artificial Intelligence Prognostic Scoring System for Myelodysplastic Syndromes (AIPSS-MDS) in CMML and to assess its performance compared with traditional scores using data from a Spanish registry (n = 1343) and a Taiwanese hospital (n = 75). In the Spanish cohort, the AIPSS-MDS accurately predicted overall survival (OS) and leukaemia-free survival (LFS), outperforming the Revised-IPSS score. Similarly, in the Taiwanese cohort, the AIPSS-MDS demonstrated accurate predictions for OS and LFS, showing superiority over the IPSS score and performing better than the CPSS and molecular CPSS scores in differentiating patient outcomes. The consistent performance of the AIPSS-MDS across both cohorts highlights its generalizability. Its adoption as a valuable tool for personalized treatment decision-making in CMML enables clinicians to identify high-risk patients who may benefit from different therapeutic interventions. Future studies should explore the integration of genetic information into the AIPSS-MDS to further refine risk stratification in CMML and improve patient outcomes.

Machine Learning Improves Risk Stratification in Myelodysplastic Neoplasms: An Analysis of the Spanish Group of Myelodysplastic Syndromes.

Myelodysplastic neoplasms (MDS) are a heterogeneous group of hematological stem cell disorders characterized by dysplasia, cytopenias, and increased risk of acute leukemia. As prognosis differs widely between patients, and treatment options vary from observation to allogeneic stem cell transplantation, accurate and precise disease risk prognostication is critical for decision making. With this aim, we retrieved registry data from MDS patients from 90 Spanish institutions. A total of 7202 patients were included, which were divided into a training (80%) and a test (20%) set. A machine learning technique (random survival forests) was used to model overall survival (OS) and leukemia-free survival (LFS). The optimal model was based on 8 variables (age, gender, hemoglobin, leukocyte count, platelet count, neutrophil percentage, bone marrow blast, and cytogenetic risk group). This model achieved high accuracy in predicting OS (c-indexes; 0.759 and 0.776) and LFS (c-indexes; 0.812 and 0.845). Importantly, the model was superior to the revised International Prognostic Scoring System (IPSS-R) and the age-adjusted IPSS-R. This difference persisted in different age ranges and in all evaluated disease subgroups. Finally, we validated our results in an external cohort, confirming the superiority of the Artificial Intelligence Prognostic Scoring System for MDS (AIPSS-MDS) over the IPSS-R, and achieving a similar performance as the molecular IPSS. In conclusion, the AIPSS-MDS score is a new prognostic model based exclusively on traditional clinical, hematological, and cytogenetic variables. AIPSS-MDS has a high prognostic accuracy in predicting survival in MDS patients, outperforming other well-established risk-scoring systems.

Response to erythropoietic-stimulating agents in patients with chronic myelomonocytic leukemia.

BACKGROUND: The efficacy of erythropoietic-stimulating agents (ESA) in chronic myelomonocytic leukemia (CMML) is unknown. Our objective was to analyze erythroid response (ER) and overall survival (OS) in a series of 94 patients with CMML treated with ESA. METHODS: We analyzed a series of 94 patients with CMML treated with ESA included in the Spanish and Düsseldorf-MDS registries. FINDINGS: ER was observed in 64% of patients and red blood cell (RBC) transfusion independence in 31%. The median duration of ER was 7 months (range, 0-88). CPSS and EPO level were significantly associated with ER in multivariate analysis (P = 0.003). Considering only patients with CPSS low- or intermediate-1-risk group, the absence of RBC transfusion dependence and erythropoietin (EPO) level predicted ER (P = 0.003 and P = 0.008, respectively). In multivariate analysis, only the EPO level retained its prognostic value (P = 0.029). Achievement of ER correlated with a better survival since ER evaluation (P = 0.016). INTERPRETATION: The CPSS and EPO levels are adequate tools to select CMML patients with symptomatic anemia who may benefit from treatment with ESA. A significant ER to ESA is expected in anemic patients with low/intermediate-1 CMML risk by the CPSS and a low endogenous serum EPO level.

Results of treatment with azacitidine in patients aged ≥ 75 years included in the Spanish Registry of Myelodysplastic Syndromes.

The tolerability of azacitidine (AZA) allows its administration in elderly patients. The objective of this study was to analyze the clinical and biological characteristics, transfusion independence (TI), overall survival (OS) and toxicity in a series of 107 patients ≥ 75 years of age from the Spanish Registry of Myelodysplastic Syndromes (MDS) treated with AZA. The median age (range) was 78 (75-90) years. According to the World Health Organization (WHO) classification, 86/102 (84%) had MDS, 10/102 (10%) had mixed myeloproferative/myelodysplastic disorder and 6/102 (6%) had acute myeloblastic leukemia. Regarding MDS by the International Prognostic Scoring System on initiation of AZA, 38/84 (45%) were low-intermediate-1 risk and 46/84 (55%) were intermediate-2-high risk. Ninety-five patients (89%) were red blood cell or platelet transfusion dependent. The AZA schedule was 5-0-0 in 39/106 (37%) patients, 5-2-2 in 36/106 (34%) patients and 7 consecutive days in 31/106 (29%) patients. The median number of cycles administered was 8 (range, 1-30). Thirty-eight out of 94 (40%) patients achieved TI. Median OS (95% confidence interval [CI]) was significantly better in patients achieving TI (n = 38) compared to patients who did not (n = 56) (22 [20.1-23.9] months vs. 11.1 [4.8-17.5] months, p = 0.001). No significant differences were observed in TI rate and OS among the three different schedules. With a median follow-up of 14 (min-max, 1-50) months, the median OS (95% CI) of the 107 patients was 18 (12-23) months and the probability of OS (95% CI) at 2 years was 34% (22-46%). Cycles were delayed in 31/106 (29%) patients and 47/101 patients (47%) were hospitalized for infection. These results show that treatment with AZA was feasible and effective in this elderly population, with 40% achieving TI, having a better OS than patients not achieving it. The schedule of AZA administration did not affect efficacy and toxicity.

Second mobilization and collection of peripheral blood progenitor cells in healthy donors is associated with lower CD34(+) cell yields.

We have retrospectively evaluated the results of two cycles of mobilization and collection of peripheral blood progenitor cells (PBPC) from 46 healthy donors included in the Spanish National Donor Registry. Mobilization involved the administration of granulocyte colony-stimulating factor (G-CSF) at a median dose of 10 microg/kg per day, and apheresis was begun after the fourth dose of G-CSF in both cycles. The median interval between both mobilizations was 187 days (range, 7-1428 days). The incidence and types of side-effects were similar after both donations, with 25 and 26 donors developing some toxicity after the first and second donations, respectively. The median number of CD34(+) cells collected was higher after the first mobilization than after the second (5.15 versus 3.16 x 10(6)/kg, respectively; p = 0.05), and 29 donors yielded fewer CD34(+) cells after the second mobilization (p = 0.018). A lower proportion of donors yielded CD34(+) cell counts >4 x 10(6)/kg after the second cycle than after the first (52% versus 76%, respectively; p = 0.057). Our study shows that second rounds of PBPC collection from normal donors are well tolerated but are associated with a significantly reduced number of CD34(+) cells collected when the same mobilization scheme is used.

FLAG-IDA regimen (fludarabine, cytarabine, idarubicin and G-CSF) in the treatment of patients with high-risk myeloid malignancies.

Forty-five patients with high-risk myeloid malignancies (32 acute myeloid leukemia and 13 high-risk myelodysplastic syndromes) were treated with fludarabine, cytarabine, idarubicin, and G-CSF (FLAG-IDA). Twenty-four (53%) patients achieved complete remission (CR), and five (11%) partial remission. Infection predominantly with pulmonary involvement was the most common regimen-related toxicity. Mucositis (15 patients) and pulmonary toxicity (19 patients) were the most frequently observed non-hematologic side effects. There were four early deaths and 12 patients presented with resistant disease. Overall survival (OS) at 12 months was 40%. The FLAG-IDA regimen shows evident antileukemic activity in patients with high-risk myeloid malignancies with acceptable toxicity.

Cefepime plus amikacin versus piperacillin-tazobactam plus amikacin for initial antibiotic therapy in haematology patients with febrile neutropenia: results of an open, randomized, multicentre trial.

BACKGROUND: Standard therapy for suspected infections in patients with profound neutropenia is the combination of a beta-lactam antibiotic plus an aminoglycoside. Cefepime's broad-spectrum activity makes it an option for initial empirical therapy in neutropenic patients. The aim of this study is to evaluate the efficacy and safety of cefepime plus amikacin compared with piperacillin-tazobactam plus amikacin for initial empirical treatment of fever in adult haematology patients with severe neutropenia. METHODS: In this prospective multicentre trial, 969 patients with 984 febrile neutropenic episodes were randomized to receive iv amikacin (20 mg/kg every 24 h) combined with either cefepime (2 g every 8 h) or piperacillin-tazobactam (4 g/500 mg every 6 h). Clinical response was determined at 72 h and at completion of therapy. RESULTS: Eight hundred and sixty-seven episodes were assessable for efficacy (432 cefepime, 435 piperacillin-tazobactam). The frequency of success without modification of the empirical therapy was nearly identical for cefepime plus amikacin (49%) compared with piperacillin-tazobactam plus amikacin (51%). Similar rates of success were found for microbiologically documented infection: 40% versus 39%, respectively. Antibiotic modification was necessary in 49% of cefepime and 44% of piperacillin-tazobactam patients. The overall response rate, with or without modification of the assigned treatment, was 94% in both groups. Drug-related adverse events were reported in 10% of cefepime plus amikacin versus 11% of piperacillin-tazobactam plus amikacin patients. Mortality due to infection occurred in a total of 10 patients (two cefepime, eight piperacillin-tazobactam). CONCLUSION: The empirical regimen of cefepime plus amikacin is equivalent to piperacillin-tazobactam plus amikacin in febrile adult haematology patients with severe neutropenia. KEYWORDS: cefepime, piperacillin-tazobactam, amikacin, empirical antibiotic therapy, febrile neutropenia, haematological malignancy

Analysis of factors associated with low peripheral blood progenitor cell collection in normal donors.

BACKGROUND: Predictive factors of the response to rHuG-CSF in normal donors have not been extensively studied. STUDY DESIGN AND METHODS: We analyzed factors influencing CD34+ cell yield in the 1st day of collection in 261 healthy donors from the Spanish National Donor Registry. The median age was 38 years (range, 2-72). The median dose of rHuG-CSF was 10 microg per kg per day (range, 5-20) over 4 days. In 103 donors (40%), <4 x 10(6) per kg CD34+ cells were collected. The variables that were analyzed included age, sex, weight, basal complete blood cell count, dose, type of rHuGCSF and schedule of administration, and maximum WBC count before apheresis. RESULTS: By univariate analysis, the maximum WBC count (<50 vs. >or=50 x 10(9)/L, p = 0.004), advanced age (p = 0.008), and number of daily rHuG-CSF doses (one vs. two; p = 0.01) correlated with the number of CD34+ cells collected. By multivariate analysis, donors age (<38 vs. >or=38 years; p = 0.014) and a single daily dose of rHuG-CSF (p = 0.005) were the two variables that significantly predicted a low CD34+ cell yield. CONCLUSION: Donors' age, with a threshold of 38 years or more, and the rHuG-CSF schedule are the factors that significantly affected CD34+ cell mobilization and collection in healthy donors.