MOSAIC: An Artificial Intelligence-Based Framework for Multimodal Analysis, Classification, and Personalized Prognostic Assessment in Rare Cancers.

PURPOSE: Rare cancers constitute over 20% of human neoplasms, often affecting patients with unmet medical needs. The development of effective classification and prognostication systems is crucial to improve the decision-making process and drive innovative treatment strategies. We have created and implemented MOSAIC, an artificial intelligence (AI)-based framework designed for multimodal analysis, classification, and personalized prognostic assessment in rare cancers. Clinical validation was performed on myelodysplastic syndrome (MDS), a rare hematologic cancer with clinical and genomic heterogeneities. METHODS: We analyzed 4,427 patients with MDS divided into training and validation cohorts. Deep learning methods were applied to integrate and impute clinical/genomic features. Clustering was performed by combining Uniform Manifold Approximation and Projection for Dimension Reduction + Hierarchical Density-Based Spatial Clustering of Applications with Noise (UMAP + HDBSCAN) methods, compared with the conventional Hierarchical Dirichlet Process (HDP). Linear and AI-based nonlinear approaches were compared for survival prediction. Explainable AI (Shapley Additive Explanations approach [SHAP]) and federated learning were used to improve the interpretation and the performance of the clinical models, integrating them into distributed infrastructure. RESULTS: UMAP + HDBSCAN clustering obtained a more granular patient stratification, achieving a higher average silhouette coefficient (0.16) with respect to HDP (0.01) and higher balanced accuracy in cluster classification by Random Forest (92.7% ± 1.3% and 85.8% ± 0.8%). AI methods for survival prediction outperform conventional statistical techniques and the reference prognostic tool for MDS. Nonlinear Gradient Boosting Survival stands in the internal (Concordance-Index [C-Index], 0.77; SD, 0.01) and external validation (C-Index, 0.74; SD, 0.02). SHAP analysis revealed that similar features drove patients' subgroups and outcomes in both training and validation cohorts. Federated implementation improved the accuracy of developed models. CONCLUSION: MOSAIC provides an explainable and robust framework to optimize classification and prognostic assessment of rare cancers. AI-based approaches demonstrated superior accuracy in capturing genomic similarities and providing individual prognostic information compared with conventional statistical methods. Its federated implementation ensures broad clinical application, guaranteeing high performance and data protection.

Classification, risk stratification and response assessment in myelodysplastic syndromes/neoplasms (MDS): A state-of-the-art report on behalf of the International Consortium for MDS (icMDS).

The guidelines for classification, prognostication, and response assessment of myelodysplastic syndromes/neoplasms (MDS) have all recently been updated. In this report on behalf of the International Consortium for MDS (icMDS) we summarize these developments. We first critically examine the updated World Health Organization (WHO) classification and the International Consensus Classification (ICC) of MDS. We then compare traditional and molecularly based risk MDS risk assessment tools. Lastly, we discuss limitations of criteria in measuring therapeutic benefit and highlight how the International Working Group (IWG) 2018 and 2023 response criteria addressed these deficiencies and are endorsed by the icMDS. We also address the importance of patient centered care by discussing the value of quality-of-life assessment. We hope that the reader of this review will have a better understanding of how to classify MDS, predict clinical outcomes and evaluate therapeutic outcomes.

Classification and Personalized Prognostic Assessment on the Basis of Clinical and Genomic Features in Myelodysplastic Syndromes.

PURPOSE: Recurrently mutated genes and chromosomal abnormalities have been identified in myelodysplastic syndromes (MDS). We aim to integrate these genomic features into disease classification and prognostication. METHODS: We retrospectively enrolled 2,043 patients. Using Bayesian networks and Dirichlet processes, we combined mutations in 47 genes with cytogenetic abnormalities to identify genetic associations and subgroups. Random-effects Cox proportional hazards multistate modeling was used for developing prognostic models. An independent validation on 318 cases was performed. RESULTS: We identify eight MDS groups (clusters) according to specific genomic features. In five groups, dominant genomic features include splicing gene mutations (SF3B1, SRSF2, and U2AF1) that occur early in disease history, determine specific phenotypes, and drive disease evolution. These groups display different prognosis (groups with SF3B1 mutations being associated with better survival). Specific co-mutation patterns account for clinical heterogeneity within SF3B1- and SRSF2-related MDS. MDS with complex karyotype and/or TP53 gene abnormalities and MDS with acute leukemia-like mutations show poorest prognosis. MDS with 5q deletion are clustered into two distinct groups according to the number of mutated genes and/or presence of TP53 mutations. By integrating 63 clinical and genomic variables, we define a novel prognostic model that generates personally tailored predictions of survival. The predicted and observed outcomes correlate well in internal cross-validation and in an independent external cohort. This model substantially improves predictive accuracy of currently available prognostic tools. We have created a Web portal that allows outcome predictions to be generated for user-defined constellations of genomic and clinical features. CONCLUSION: Genomic landscape in MDS reveals distinct subgroups associated with specific clinical features and discrete patterns of evolution, providing a proof of concept for next-generation disease classification and prognosis.

Special considerations in the management of adult patients with acute leukaemias and myeloid neoplasms in the COVID-19 era: recommendations from a panel of international experts.

The ongoing COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 is a global public health crisis. Multiple observations indicate poorer post-infection outcomes for patients with cancer than for the general population. Herein, we highlight the challenges in caring for patients with acute leukaemias and myeloid neoplasms amid the COVID-19 pandemic. We summarise key changes related to service allocation, clinical and supportive care, clinical trial participation, and ethical considerations regarding the use of lifesaving measures for these patients. We recognise that these recommendations might be more applicable to high-income countries and might not be generalisable because of regional differences in health-care infrastructure, individual circumstances, and a complex and highly fluid health-care environment. Despite these limitations, we aim to provide a general framework for the care of patients with acute leukaemias and myeloid neoplasms during the COVID-19 pandemic on the basis of recommendations from international experts.

Assessment of ASC specks as a putative biomarker of pyroptosis in myelodysplastic syndromes: an observational cohort study.

BACKGROUND: NLRP3 inflammasome-directed pyroptotic cell death drives ineffective haemopoiesis in myelodysplastic syndromes. During inflammasome assembly, the apoptosis-associated speck-like protein containing a CARD (PYCARD, commonly known as ASC) adaptor protein polymerises into large, filamentous clusters termed ASC specks that are released upon cytolysis. Specks are resistant to proteolytic degradation because of their prion-like structure, and therefore might serve as a biomarker for pyroptotic cell death in myelodysplastic syndromes. METHODS: This observational cohort study was done at the H Lee Moffitt Cancer Center (Tampa, FL, USA). Patients with myelodysplastic syndromes, healthy controls, and patients with non-myelodysplastic syndrome haematological cancers or type 2 diabetes were recruited. We used confocal and electron microscopy to visualise, and flow cytometry to quantify, ASC specks in peripheral blood and bone marrow plasma samples. Speck percentages were compared by t test or ANOVA, correlations were assessed by Spearman's rank correlation coefficient, and biomarker efficiency was assessed by receiver operating characteristics and area under the curve (AUC) analysis. FINDINGS: Between Jan 1, 2005, and Jan 12, 2017, we obtained samples from 177 patients with myelodysplastic syndromes and 29 healthy controls for the discovery cohort, and 113 patients with myelodysplastic syndromes and 31 healthy controls for the validation cohort. We also obtained samples from 22 patients with del(5q) myelodysplastic syndromes, 230 patients with non-myelodysplastic syndrome haematological cancers and 23 patients with type 2 diabetes. After adjustment for glucose concentration, the log10-transformed mean percentage of peripheral blood plasma-derived ASC specks was significantly higher in the 177 patients with myelodysplastic syndromes versus the 29 age-matched, healthy donors (-0·41 [SD 0·49] vs -0·67 [0·59], p=0·034). The percentages of ASC specks in samples from patients with myelodysplastic syndromes were significantly greater than those in samples from individuals with every other haematological cancer studied (all p<0·05) except myelofibrosis (p=0·19). The findings were confirmed in the independent validation cohort (p<0·0001). Peripheral blood plasma danger-associated molecular pattern protein S100-A8 and protein S100-A9 concentrations from 144 patients with myelodysplastic syndromes from the discovery cohort directly correlated with ASC speck percentage (r=0·4, p<0·0001 for S100-A8 and r=0·2, p=0·017 for S100-A9). Patients with at least two somatic gene mutations had a significantly greater mean percentage of peripheral blood plasma ASC specks than patients with one or no mutation (-0·22 [SD 0·63] vs -0·53 [0·44], p=0·008). The percentage of plasma ASC specks was a robust marker for pyroptosis in myelodysplastic syndromes (AUC=0·888), in which a cutoff of 0·80 maximised sensitivity at 0·84 (95% CI 0·65-0·91) and specificity at 0·87 (0·58-0·97). INTERPRETATION: Our results underscore the pathobiological relevance of ASC specks and suggest that ASC specks are a sensitive and specific candidate plasma biomarker that provides an index of medullary pyroptotic cell death and ineffective haemopoiesis in patients with myelodysplastic syndromes. FUNDING: T32 Training Grant (NIH/NCI 5T32 CA115308-08), Edward P Evans Foundation, The Taub Foundation Grants Program, the Flow Cytometry, Analytic Microscopy, and Tissue Core Facilities at the H Lee Moffitt Cancer Center and Research Institute, a National Cancer Institute-designated Comprehensive Cancer Center (P30-CA076292).

Multicenter independent assessment of outcomes in chronic myeloid leukemia patients treated with imatinib.

BACKGROUND: Imatinib slows development of chronic myeloid leukemia (CML). However, available information on morbidity and mortality is largely based on sponsored trials, whereas independent long-term field studies are lacking. PATIENTS AND METHODS: Consecutive CML patients who started imatinib treatment before 2005 and who were in complete cytogenetic remission (CCyR) after 2 years (± 3 months) were eligible for enrollment in the independent multicenter Imatinib Long-Term (Side) Effects (ILTE) study. Incidence of the first serious and nonserious adverse events and loss of CCyR were estimated according to the Kaplan-Meier method and compared with the standard log-rank test. Attainment of negative Philadelphia chromosome hematopoiesis was assessed with cytogenetics and quantitative polymerase chain reaction. Cumulative incidence of death related or unrelated to CML progression was estimated, accounting for competing risks, according to the Kalbleisch-Prentice method. Standardized incidence ratios were calculated based on population rates specific for sex and age classes. Confidence intervals were calculated by the exact method based on the χ(2) distribution. All statistical tests were two-sided. RESULTS: A total of 832 patients who were treated for a median of 5.8 years were enrolled. There were 139 recorded serious adverse events, of which 19.4% were imatinib-related. A total of 830 nonserious adverse events were observed in 53% of patients; 560 (68%) were imatinib-related. The most frequent were muscle cramps, asthenia, edema, skin fragility, diarrhea, tendon, or ligament lesions. Nineteen patients (2.3%) discontinued imatinib because of drug-related toxic effects. Forty-five patients lost CCyR, at a rate of 1.4 per 100 person-years. Durable (>1 year) negative Philadelphia chromosome hematopoiesis was attained by 179 patients. Twenty deaths were observed, with a 4.8% mortality incidence rate (standardized incidence ratio = 0.7; 95% confidence interval = 0.40 to 1.10, P = .08), with only six (30%) associated with CML progression. CONCLUSIONS: In this study, CML-related deaths were uncommon in CML patients who were in CCyR 2 years after starting imatinib, and survival was not statistically significantly different from that of the general population.

Evaluation of BCRP and MDR-1 co-expression by quantitative molecular assessment in AML patients.

Expression of two MDR genes, BCRP and MDR1, was evaluated by real-time PCR technique in 51 AML patients. Fifty-six percent expressed the BCRP gene, with the 48.2% showing intermediate levels. Eighty-eight percent expressed the MDR1, with 23.8% of cases at high expression. A significant correlation between BCRP and MDR1 values was found by regression analysis. Either levels of BCRP or MDR1 did not correlate with clinical characteristics of patients at diagnosis.