Citations

Background Multiple myeloma (MM) is a hematological cancer characterized by the proliferation of plasma cells (PC). Progressive disease (PD) criteria are primarily characterized by an increase in (1) serum M-protein (SPEP), (2) urinary M-protein (UPEP), (3) serum free light chains (sFLC), or (4) PC percentage in the bone marrow. An essential aspect of the PD criteria is the requirement for confirmatory results from blood or urine analyses. However, in many cases, patients may be taken off treatment in favor of a new therapy after only one assessment. This can introduce clinical trial endpoint inaccuracies when confirmatory values are missing, resulting in increased censoring and underestimation of the progression rate (if the second assessment is not done). Moreover, in subjects with high tumor burden and rapidly progressing disease, it is often urgent to confirm PD before starting a subsequent line of treatment. This study assessed the necessity of repeating biomarkers to confirm PD if two biomarkers meet the progression criteria. Methods In this retrospective study, between 2002 and 2020, we assessed all consecutive patients with relapsed MM enrolled in any therapeutic clinical trial at Mayo Clinic. We included patients with measurable disease if SPEP and sFLC were available at the time of assessment. Patients required at least two consecutive assessments (by the same method) to be included in our study. PD was based on the IMWG criteria. Relapse characterized by the development of new bone lesions or extramedullary relapse without meeting biochemical progression criteria was an exclusion criterion. Patients starting a new line of treatment after non-confirmed PD were also excluded. Furthermore, patients who did not meet simultaneous progression criteria and had unconfirmed PD (did not meet IMWG criteria at the second assessment) were excluded from the analysis (n=15). The probability of meeting the progression threshold with 1 and 2 consecutive assessments was calculated. We then examined the proportions of patients meeting PD by 2 biomarkers simultaneously at the first assessment. Results We evaluated episodes of treatment among 434 patients enrolled in clinical trials, including 601 episodes: 593 episodes of biochemical confirmed PD and 8 episodes when patients did not progress but who met simultaneous progression criteria at the same time with two different biomarkers. Per measurable disease criteria, MM isotypes were 71.0% IgG, 16.8% IgA, 0.9% IgM, 1.2% IgD, 0.3% biclonal, and 9.8% light chain. According to the standard IMWG response criteria (sequential criteria), 78% and 9% of episodes reached the progression threshold with serum and urine M-components, respectively, with two consecutive assessments. For the FLC, 21% of episodes met the progression criteria. No patients had a second bone marrow examination for confirmation of PD. At the first available assessment, 82%, 19%, and 2% of episodes met progression threshold, respectively, with serum M-protein, urine M-protein, and both (serum + urine M-protein). Additionally, 58%, and 19%, episodes met the progression threshold at the first assessment with FLC and bone marrow examination, respectively. Moreover, 67% of patients with IgA myeloma meet the progression threshold with the total quantitative IgA. With the simultaneous criteria at time of first assessment, 403 episodes meet progression threshold: 45% with SPEP+sFLC (n=269/593), 4% with SPEP+UPEP (n=21/593), 9% with sFLC+UPEP (n=55/593), 11% with BME only (n=66/593), and 0.3% with qIgA +dFLC/UPEP (n=2/593). Thus, among these 413 episodes, which meet progression criteria by 2 simultaneous values at the time of the first assessment, 98% (n=413/(413+8)) of these patients (positive predictive value of the new criteria) subsequently had confirmed progression by sequential value. Conclusions In summary, we showed that the simultaneous criteria allow to identify patients with true progressive disease without overestimating the progression rate of myeloma. Therefore, confirmatory results of progression for relapsed MM patients having two biomarkers meeting criteria of progression at the same time is useless for determining disease progression.

Background The prognostic relevance of TP53 mutations (TP53MT) in acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) has been extensively studied and the information accessed for defining the new International Consensus Classification (ICC) category of “myeloid neoplasms with mutated TP53” (Blood 2022;140:1200). The latter designation requires ≥10% TP53 variant allele frequency (VAF) for AML and MDS/AML and in case of MDS, multi-hit TP53 or ≥10% VAF associated with complex karyotype (CK). We examined the prognostic relevance of TP53MT in myeloproliferative neoplasms (MPN), specified by subtype and disease stage. Methods The current study was conducted under an institutional review board approved minimum risk protocol. Additional informative cases from the University of Florence, Italy and the South Australian Myeloid Neoplasm Registry were included. Diagnostic criteria and assignment of TP53 allelic state were according to the ICC. Next-generation sequencing (NGS) and cytogenetic studies were performed according to institutional standard protocols. Conventional statistical methods were employed. Survival analyses were censored for allogeneic stem cell transplantation (ASCT) and were calculated from the time of first detection of TP53MT. Results 115 patients with MPN and documented pathogenic TP53MT were considered (median age 68.5 years). ICC-defined subtype designations, at time of TP53MT detection, were polycythemia vera (PV) or essential thrombocythemia (ET) (N=9), chronic phase myelofibrosis (MF-CP; N=63), accelerated phase MPN (MPN-AP; N=16), and blast phase MPN (MPN-BP; N=27). Driver mutation distribution was JAK2 64%, CALR 18%, MPL 8%, and triple-negative 10%. Karyotype was monosomal (MK) in 42%, CK 9%, and normal (NK) 35%. Frequent co-mutations were ASXL1 (28%), TET2 (14%), and SRSF2 (14%). TP53MT were classified as ICC-defined multihit in 67 (58%) cases; karyotype in the latter was MK (58%), CK (16%), non-MK/CK (13%), or NK (13%). TP53 VAF was ≥50% in 27 (24%) patients and <10% in 25 (22%). TP53MT included one (N=96) or ≥2 (N=19). MPN subtype- and stage-specific distributions for multihit TP53MT were 91% for MPN-BP, 88% MPN-AP, 44% MF-CP, and 0% ET/PV. In order to examine disease subtype- and stage-specific prognostic contribution of TP53MT, the study population was divided into 10 operational groups: 1) PV/ET non-multihit/non-MK/CK (N=8; VAF range 3-32%); 2) MF-CP non-multihit/non-MK/CK with TP53MT VAF <10% (N=10); 3) MF-CP non-multihit VAF ≥10% and non-MK/CK (N=11); 4) PV/ET/MF-CP non-multihit with MK/CK (N=5); 5) MF-CP multihit/non-MK/CK (N=8); 6) MF-CP multihit and MK/CK (N=20); 7) MPN-BP/AP non-multihit/non-MK/CK (N=2); 8) MPN-BP/AP non-multihit with MK/CK (N=2); 9) MPN-BP/AP multihit/non-MK/CK (N=9); and 10) MPN-BP/AP multihit with MK/CK (N=30). The corresponding median survivals were “not reached”, 55, 42, 6, 18, 10, “not reached”, 4, 4, and 6 months (p<0.01). Survival was similar between groups 1, 2, and 3 (p>0.2 for all comparisons) now referred to as “PV/ET/MF-CP non-multihit/non-MK/CK (N=39)”. Survival was similarly dismal for groups 8, 9, and 10 (p values > 0.3 for all comparisons) now referred to as “MPN-BP/AP with or without multihit TP53MT or MK/CK (N=43)”. Survival was also similar between groups 4, 5, and 6 (p values >3 for all comparisons) now referred to as “PV/ET/MF-CP with multihit TP53MT or MK/CK (N=33)”. The three consolidated groups were subsequently compared for survival that showed a significantly worse outcome for “MPN-BP/AP with or without multihit TP53MT or MK/CK” with median survival of 4 months, compared to 12 months for “PV/ET/MF-CP with multihit TP53MT or MK/CK” (HR 2.3; 95% CI; 1.3-3.9) and 55 months for “PV/ET/MF-CP non-multihit/non-MK/CK” (10.9; 95% CI 5.3-22.4); survival in the latter two groups was also significantly different, in favor of PV/ET/MF-CP non-multihit/non-MK/CK (HR 4.8; 95% CI 2.3-10.1). Conclusions: The presence of multi-hit TP53MT or MK/CK, in MF-CP, and possibly in ET/PV, warrants their inclusion in the ICC category of “myeloid neoplasms with mutated TP53”. By contrast, non-multihit TP53MT in ET/PV/MF-CP, regardless of VAF and in the absence of MK/CK, does not appear to compromise short-term survival. The observations from the current study also suggest a prognostic hierarchy with blast percentage being the most unfavorable, followed by MK/CK and multihit TP53MT.

Introduction: Multiple myeloma (MM) with t(11;14)(q13;q32), resulting in CCND1;IGH rearrangement, remains the only subset amenable to targeted therapy with the BCL2 inhibitor, Venetoclax. Nevertheless, not all patients with this translocation respond to treatment and some progress early after initial response. The mechanisms behind this clinical heterogeneity are largely unknown. Methods: To investigate mechanisms of heterogeneity in response to Venetoclax, we interrogated 44 whole genome sequencing (WGS) and 24 whole exome sequencing data from a training cohort including 33 patients with t(11;14) MM treated with Venetoclax at the Mayo Clinic. In a validation cohort, we extracted single nucleotide variants (SNV) and indels from 17 MM patients with t(11;14) with available RNAseq data treated with Venetoclax at Emory University. Results: In the training cohort the median number of lines of therapy was 4 and Venetoclax drug regimens included combinations with dexamethasone, proteasome inhibitors, anti-CD38 agents, and immunomodulatory agents. After a median follow up of 30 months, the median progression free survival (PFS) was 9 months, and 13 patients (39%) achieved a complete remission (CR) to therapy, while 9 (27%) were refractory (i.e. no response or progressed within the first 60 days). For our driver discovery analysis, we integrated SNV, indels, copy number variants (CNV) and structural variants (SV). Deletion 17p (del17p) or TP53 mutation, MCL1 gain/amp, and hyper-APOBEC did not correlate with PFS in our cohort (p=0.59, p=0.71, and p=0.05, respectively). In contrast, mutations involving the MAPK pathway (NRAS, KRAS, BRAF, FGFR3) displayed the strongest association with short PFS, which was retained after adjusting for del17p/TP53 mutations, and 1q gain/amp (HR: 4.76, CI: 1.8-12.9, p=0.002). The association between suboptimal response and resistance to Venetoclax in MM patients with MAPK pathway mutations (n=7) was also observed in our validation cohort from Emory University (p=0.046). This is in line with what has been reported in functional studies in acute myeloid leukemia, where RAS pathway promotes MCL1 under the selective pressure of Venetoclax (Zhang et al. Signal Transduct Target Ther 2022). While MAPK mutations explained most cases of refractory patients (86%), the mechanisms through which initially responding patients subsequently progressed remain unclear. To investigate this, we interrogated 10 patients with pre- and post-treatment samples collected at disease progression (PD). We identified genomic events acquired/selected at PD of genes involved in the anti-apoptotic MCL1/BCL2 and MAPK pathways in 6/10 cases. One patient was refractory to Venetoclax and had double minutes involving BCL2, which caused multiple focal amplifications (>30 copies), likely reflecting extrachromosomal DNA. Another patient progressed with an acquired, focal deletion on BCL2L11. In two patients, the progressing clones harbored mutations in MAPK pathways (one BRAF and the other WNK2). Two patients who relapsed after sustained CR had major branching evolution where a previously undetectable focal deletion of PMAIP1 (NOXA) was seen at PD. Importantly, one of these deletions was a biallelic event. Venetoclax sensitivity assays conducted on PMAIP1 knockout in the MM t(11;14) cell line, KMS12PE, demonstrated significantly reduced cell death across all drug concentrations (p<0.05), validating the loss of NOXA as a mechanism of resistance to Venetoclax in MM. Finally, similarly with what was previously reported in chronic lymphocyte leukemia treated with Venetoclax (e.g., Herling et al Nat Comm 2018), CDKN2A/B monoallelic and biallelic focal deletions were selected post-Venetoclax, at PD in 4 patients while TP53 SNV/deletions were selected from subclonal to clonal in 2 patients at PD. Conclusion: In this study, we demonstrate that mutations involving the MAPK pathway identify a population of MM patients with t(11;14)(CCND1;IGH) who are unlikely to respond to Venetoclax. We also show that genomic events involving the MAPK and BCL2/MCL1 pathways are primarily responsible for disease progression after initial response. These findings highlight the critical role of comprehensive genomic profiling through WGS in identifying mechanisms of resistance to targeted therapies in MM, aiding patient selection for Venetoclax, and guiding the development of novel drug combinations.

Introduction: The BMT CTN 1704 study developed and validated the CHARM that stratified risk for non-relapse mortality and overall mortality among older adults, performing better than the HCT-CI alone (Artz A et al, Blood. 2023;142:109) and equally to two machine learning models. CHARM assigns a total score for 7 health variables: increasing age, higher HCT-CI scores, lower albumin, higher C-reactive protein, higher percent of weight loss over the preceding year, lower patient-reported performance status scores, and lower cognitive score per Montreal cognitive assessment (MoCA). A CHARM calculator is available at:https://cibmtr.org/CIBMTR/OffNav/DevSandbox/CHARM-Risk-NRM-Calculator No multi-institutional prospective data exist on functional trajectories and morbidity after HCT in older patients. We now report on the association of CHARM to trajectories of secondary morbidity outcomes among allo-HCT survivors of this large, prospective study. Patients and Methods: Allo-HCT candidates, aged ≥60 years (yrs), were enrolled (n=1226) from 49 centers in the US between 2019 - 2021. The primary analysis includes 1105 patients proceeding to allo-HCT on study and secondary endpoints were assessed at day (D) 100, 180 and 365 except MoCA and organ toxicity were restricted to D100 and frailty had inadequate data for D100. A sequential multiple imputation strategy was implemented to impute endpoints for survivors at each time point with missing data. Analyses on multiply imputed datasets were conducted and the results combined using Rubin's rule. Associations between CHARM scores and secondary outcomes were analyzed using a multivariable Cox, Fine-Gray, Generalized Estimating Equations, and logistic regression model for survival, competing risks, continuous, and binary outcomes, respectively, with latter two focused on surviving patients. Models were adjusted for other variables including conditioning intensity, graft-versus-host disease (GVHD) prophylaxis regimen, disease-risk index, donor-recipient gender match, ethnicity, baseline value of the dependent variable, and visit timepoints. Results: Higher CHARM scores were associated with development of serious organ toxicities by D100 (OR: 2.05, [1.52-2.78], p<0.0001) and ≥2 worsening score on the MoCA (odds ratio (OR) 1.55 [1.16-2.1], p=0.003). Among survivors at all timepoints, higher CHARM scores were associated with greater disability by instrumental activities of daily living (IADL) (Slope -0.640 [-0.433-0.846], p<0.001) and worsening Patient-Reported Outcomes Measurement Reporting System (PROMIS) physical function (Slope -0.981 [-0.057 - -1.904], p=0.037), depression (Slope 0.763 [0.042-1.484], p=0.038) and in a lesser magnitude anxiety (Slope 0.659, p=0.076). Among survivors at D180 and D365, higher CHARM scores were associated with worse frailty (Slope 0.193 [0.081-0.305], p<0.001). CHARM scores were not associated with development of acute GVHD grades 2-4 or 3-4 but were associated with post-GVHD increased mortality (HR: 1.61, [1.25-2.08], p=0.0002). Higher CHARM scores are associated with a lower incidence of chronic GVHD (HR: 0.83, p=0.026), likely due to the effect of CHARM on the competing risk of death leaving fewer patients at risk for chronic GVHD. Conclusions: The novel primary CHARM, originally developed to predict risks of NRM, also predicts worse frailty, disability, cognitive decline, and serious organ toxicities; outcomes that are critically important to older recipients of allo-HCT. CHARM therefore informs risks of transplant morbidity, separate from risks of developing acute GVHD. Results further support adopting CHARM in practice to counsel patients, expedite HCT referrals for lower risk CHARM, and design trials for high CHARM score patients.

Background: Inotuzumab Ozogamicin (InO) is an antibody-cytotoxin conjugate with high activity in B-cell acute lymphoblastic leukemia (B-ALL). InO targets the B-cell surface protein CD22, which is expressed in most B-ALL cases, albeit with great variability in expression levels. Emerging evidence suggest that this variation is associated with the stage of leukemia differentiation arrest [PMID: 21348573]. However, the exact relationship between CD22 expression and leukemia sensitivity to InO, particularly in the context of leukemia developmental state, remains unclear. Furthermore, our understanding of the molecular mechanisms regulating CD22 transcription in B-ALL is still incomplete. Methods: To comprehensively investigate the genomic basis of variability in leukemia CD22 expression and thus B-ALL sensitivity to InO, we performed a multi-omic characterization of 196 primary human B-ALL samples, using whole transcriptome sequencing (RNAseq), assay for transposase-accessible chromatin with sequencing (ATAC-seq), coupled with ex vivo profiling of InO sensitivity. Additionally, genomic editing studies were performed in Nalm6 and REH B-ALL cell lines, and patient derived xenograft (PDX) models were used for in vitro and in vivo validations. Results: Applying digital cytometry to leukemia bulk RNAseq data to infer the B-ALL developmental state, we found that blast cells with early Pre-pro-B-like signature had low levels of CD22 expression and were resistant to InO. As B-cell differentiation is controlled by stage-specific transcription factors (TFs), we performed a targeted CRISPR library screen of 1,639 TF genes and identified EBF1 as a key activator of CD22 expression (FDR=7.1x10-4). In primary B-ALL samples, EBF1 expression was significantly correlated with CD22 expression (R=0.69, P<0.0001) and leukemia response to InO (measured as LC50, R=-0.42, P<0.0001). In addition, TF footprint analysis of leukemia ATAC-seq data identified EBF1 as the most differentially activated TF comparing InO-sensitive and -resistant B-ALL cases (P=8x10-174). Blocking the EBF1 binding sites within the CD22 promoter by CRISPR interference resulted in a 50-fold and 49-fold decrease in cell surface CD22 expression, and a 22-fold and 25.7-fold increase in InO LC50 for Nalm6 and REH, respectively, confirming the direct impact of EBF1 on CD22 transcription and InO sensitivity. Because BCR::ABL1-positive ALL showed the greatest variability in InO sensitivity among all B-ALL subtypes, we further investigate the role of the EBF1/CD22 axis on InO sensitivity within this subtype. Unsupervised clustering of gene expression profiles revealed two distinct groups, namely BCR::ABL1 ALL with multilineage (BCR::ABL1-M) or lymphoid only (BCR::ABL1-L) involvement [PMID: 38153913]. Compared to BCR::ABL1-L, BCR::ABL1-M displayed markedly reduced CD22 expression (P=8.3×10-11) and resistance to InO (P=1.9×10-4), likely due to the lower expression and frequent somatic alteration of EBF1. Finally, we established PDX models for representative cases of BCR::ABL1-M and BCR::ABL1-L. Strikingly, InO treatment did not affect leukemia progression in the BCR::ABL1-M PDX model, whereas the BCR::ABL1-L PDX model achieved complete remission and has remained leukemia free to date. Conclusion: Collectively, these data demonstrate the direct impact of EBF1 on CD22 expression during B-cell development, and explain the heterogeneity of InO sensitivity, even within the same molecular B-ALL subtype.

Introduction: Cilta-cel is approved in the US and EU for the treatment (tx) of patients (pts) with len-refractory MM after ≥1 line based on the randomized, phase 3 CARTITUDE-4 trial (NCT04181827). At the first interim analysis (15.9-month [mo] median follow-up), cilta-cel significantly improved progression-free survival (PFS) vs SoC (hazard ratio [HR], 0.26 [protocol-specified weighted analysis]; P<0.0001). At the second interim analysis, overall survival (OS) was significantly improved with cilta-cel vs SoC (HR, 0.55; P=0.0009). MRD negativity is a prognostic marker of prolonged survival outcomes for pts with MM. We report MRD negativity, including overall and sustained MRD negativity, overall MRD-negative complete response or better (≥CR), and MRD-negative ≥CR at mo 12, from the prespecified second interim analysis of CARTITUDE-4. Methods: Eligibility criteria were previously described. Pts were assigned 1:1 to cilta-cel or SoC (pomalidomide, bortezomib, and dexamethasone [PVd]/daratumumab, pomalidomide, and dexamethasone [DPd]). Pts in the cilta-cel arm underwent apheresis, received bridging therapy (PVd/DPd), and then a single cilta-cel infusion (target dose, 0.75×106 CAR+ viable T cells/kg) 5-7 days (d) after the start of lymphodepletion. PFS was the primary endpoint; ≥CR rate, overall response rate, overall MRD-negativity rate, and OS were key secondary endpoints. MRD was assessed centrally via next-generation sequencing (clonoSEQ v2.0; Adaptive Biotechnologies). MRD was evaluated at d 56 post infusion in the cilta-cel arm; and in both arms at suspected ≥CR and at 6, 12, 18, and 24 mo post infusion (cilta-cel arm) or cycle 1 d 1 (SoC arm), as well as yearly until progression/start of subsequent therapy in pts in ≥CR. Sustained MRD negativity (10−5) was defined as confirmed MRD negativity ≥12 mo apart and without MRD positivity in between. Pts were evaluable for sustained MRD negativity if they achieved MRD negativity and had ≥1 evaluable MRD sample ≥12 mo after the first negative result or progressed/died/started subsequent tx <12 mo after the first negative result. Results: 419 pts were randomized (intent-to-treat [ITT] set; cilta-cel, n=208; SoC, n=211); 176 pts in the cilta-cel arm received cilta-cel as study tx. As of May 1, 2024, median follow-up for the study was 33.6 mo. At the 10−5 threshold, 145 pts in the cilta-cel arm and 103 in the SoC arm were evaluable for MRD. MRD-negativity rates (10−5) in the ITT set and the MRD-evaluable subset were higher with cilta-cel vs SoC (ITT, 62% vs 18%; MRD evaluable, 89% vs 38%; both P<0.0001). Across subgroups, cilta-cel vs SoC consistently increased overall MRD-negativity rates (10−5). In the ITT set, 48% of the cilta-cel arm achieved MRD negativity (10−5) by d 56, with the MRD-negative rate rising to 60% by 6 mo post cilta-cel infusion. Overall MRD-negativity rates at the 10−6 threshold in the ITT set were higher with cilta-cel vs SoC (57% vs 9%; P<0.0001). In the cilta-cel arm, 119 (57%) pts vs 26 (12%) in the SoC arm achieved overall MRD-negative (10−5) ≥CR (P<0.0001). At the 12-mo MRD assessment, 92 (44%) pts in the cilta-cel arm vs 17 (8%) in the SoC arm (P<0.0001) had MRD-negative (10−5) ≥CR. In pts with MRD-negative (10−5) ≥CR at mo 12, median PFS was not reached (NR; 95% CI, not estimable [NE]-NE) with cilta-cel and 37.8 mo (95% CI, 25.0-NE) with SoC; median OS was NR (95% CI, NE-NE) and NR (95% CI, 37.8 mo-NE), respectively. Sustained MRD-negativity rates (10−5) in the ITT set were 40% in the cilta-cel arm vs 6% in the SoC arm (P<0.0001); among 110 and 26 evaluable pts for sustained MRD, sustained MRD-negativity rates were 75% vs 50% (P=0.0159). Among the 176 pts who received cilta-cel as study tx, overall MRD negativity at 10−5 was achieved by 129 (73%) pts (89% of 145 evaluable pts). Conclusions: At 33.6-mo median follow-up in CARTITUDE-4, cilta-cel vs SoC significantly increased overall MRD-negativity rates >3-fold in the ITT set, with pts achieving MRD negativity rapidly post cilta-cel. The prognostic value of MRD-negative ≥CR at mo 12 was shown, as median PFS was >3 years in pts with MRD-negative ≥CR at mo 12, regardless of tx. These data further underscore the benefit of cilta-cel, which led to significant >3-fold increases vs SoC in rates of MRD-negative ≥CR at any time and at mo 12, and sustained MRD negativity. Our MRD data demonstrate higher rates of deep and sustained MRD negativity achieved with cilta-cel vs SoC in tx of len-refractory MM as early as first relapse.

Background: Risk stratification for suspected pulmonary emboli (PE) remains an area of active research. Current risk stratification algorithms have moderate discriminatory capabilities and are guideline-recommended, but they are poorly utilized in practice, with most clinicians relying predominantly on their clinical training and personal “gestalt.” Our group recently derived an Artificial Intelligence (AI) algorithm using electrocardiograms (ECG) to predict the presence of PE (AUC 0.69) on computed tomography scans of the pulmonary arteries (CTPA) in a cohort of 79K patients. Aim: Validate the AI-ECG algorithm in a contemporary dataset of patients with suspected PE. Methods: Emergency Department (ED) visits between 9/1/2022 and 12/31/2023 across the Mayo Clinic Enterprise were searched using a unified data platform. ED visits in patients over 18 years of age with a discretely documented Wells' PE or PERC Rule, a D-dimer result, or a CTPA were included. Only CTPA imaging and D-dimers performed within 48 hours of ED admission were eligible to be included. Patients were included in the final cohort if an ECG was performed within 6 hours of ED admission, D-dimer testing, or CTPA. Patients were considered negative for PE if CTPA was negative, and when this wasn't performed, a negative D-dimer (high sensitivity) or PERC rule of 0 was considered negative. A previously derived Artificial Neural Network algorithm to analyze the ECG to predict PE was applied to the dataset for validation. Results: A total of 27,028 ED visits in 24,288 unique patients meet inclusion criteria. Among this group, the first ED visit with a CTPA and an ECG within 6 hours was included, leaving 18,936 patients for the final validation cohort. The mean age of the cohort was 54.7 (SD 18.7), and 57% were male. Among the entire cohort, 13,573 underwent D-dimer testing for which 4,895 (36%) were positive at standard thresholds. A total of 10,486 patients underwent CTPA (55% of patients evaluated), of which 455 were positive for PE (4.3%). Using the AI-ECG model from the derivation cohort applied to this validation cohort resulted in an AUC of 0.69, identical to the derivation cohort. Using ECG model estimates alone, low (0.74%, n=2447, NPV 99.3%), elevated (2.27%, n=15404, NPV, 97.72%), and high-risk (8.02%, n=1085, NPV 91.98%) categories for PE can be created. A negative D-dimer in the elevated risk group increased the negative predictive value (NPV) to >99.9% and could be used to reduce imaging in this group further. Using the proposed ECG to D-dimer cascade risk stratification could further reduce the need for PE imaging for an overall NPV of 99.9% and sensitivity of 95.2%. We then used additional clinical parameters to train a new algorithm with additional clinical data and the AI-ECG algorithm estimates. Multiple machine learning algorithms were then evaluated in a dataset divided into training (2/3) and testing (1/3) datasets, after downsampling (to address class imbalance). The best results were obtained with AI-ECG results and D-dimer: Ada Boost (AUC 0.93, F1 score 0.895) compared to standard logistic regression (AUC 0.79, F1 score 0.746). The addition of age, sex, and oxygen saturation did not improve model performance. Conclusion: We have now validated the previously derived AI-ANN ECG algorithm for PE in a contemporary and independent cohort and demonstrate potential for integration into clinical PE assessment algorithms. Unlike prior risk stratification processes, ECGs are invariably performed in patients with cardiopulmonary symptoms and have the potential to prevent missed or delayed PE diagnosis by proactively identifying high-risk patients. Additionally, we demonstrate that the results of this algorithm can be combined with D-dimer values, resulting in a simple, powerful, predictive algorithm (AUC 0.93) that pairs cardiovascular electrophysiology with laboratory biomarkers and could lead to a new paradigm in clinical risk stratification for PE.

Background. Study of myeloid neoplasms (MN) has been rapidly transformed by genome characterization. Gene mutations have been reported to influence disease phenotype and risk of disease progression, and the evaluation of mutational status can provide valuable insights to improve the decision-making process. In this international study, we evaluated the clinical impact of mutational screening in patients with CMML, which is characterized by significant clinical and genomic heterogeneity, including a substantial proportion of patients with poor outcomes and unmet needs. Currently, only limited molecular information has been incorporated into CMML classification and prognostication. Methods. We retrospectively collected 3182 CMML patients. We used MOSAIC, an AI-based framework for multimodal analysis in rare cancers (PMID: 38875514) to develop innovative, molecular-based tools for classification and prognostication. An independent validation of the results on 516 prospectively collected patients was performed. Results. Bayesian Networs and Hierarchical Dirichlet Processes were used to identify genomic associations and define a CMML molecular classification. We identified 11 clusters with distinct clinical and genomic features, including splicing machinery, signal transduction and tyrosine kinase pathways aberrations, and high-risk molecular signatures (i.e., AML-like and TP53 mutations). Up to 15% of patients showed clear genomic overlap with other MN. Importantly, extensive multi-color flow cytometry on T lymphocytes, NK and myeloid cells (BD FACSymphony A5 Cell Analyzer) revealed specific immunologic and inflammation signatures associated with molecular subgroups. We then integrated molecular and clinical information to build an international CMML Prognostic Scoring System (iCPSS). We assessed the performance of different statistical and AI-based methods; L1-penalized Cox Model resulted as the best performing method. Selected features included hematological parameters (WBC, Hb, PLT and marrow blasts), cytogenetic abnormalities, and mutations in 10 genes (ASXL1, DNMT3A, EZH2, NRAS, RUNX1, SETBP1, STAG2, TET2, TP53, U2AF1). By Bayesian thresholds optimization, we identified 5 risk classes (i.e., very-low, low, intermediate, high and very-high) with median overall survival (OS) ranging from 99 to 9 months (P<.001), and median leukemia-free survival (LFS) ranging from 210 to 18 months (P<.001). iCPSS provided better patient discrimination across all clinical endpoints compared to currently available prognostic tools, with a Concordance Index (CI) for OS of 0.75 versus 0.62-0.64, respectively. Notably, compared to available scoring systems, up to 40% of patients were reassigned to higher or lower risk classes by the iCPSS. External validation on an independent prospective cohort confirmed that the performance of iCPSS is superior to existing prognostic models (CI for OS 0.71 vs 0.54-0.61, respectively) To demonstrate the clinical utility of iCPSS, we focused on 753 patients who underwent allogeneic stem cell transplantation (HSCT). In multivariable analysis, iCPSS stratified the probability of OS post-HSCT (P<.001) and identified groups of patients with different probabilities of disease relapse, ranging from 9% to 62% (P<.001). We employed a clinical- and genomic-based decision support system (PMID: 38723212) to determine the optimal timing of HSCT. Preliminary analyses indicated that higher-risk patients according to iCPSS (i.e., intermediate, high, and very high) benefit from an immediate HSCT strategy. Lower-risk patients (i.e., very low and low) had life expectancy maximized with delayed HSCT. Modeling decision analysis using iCPSS compared to conventional scores, resulted in a change in transplantation policy for a significant proportion of patients (up to 25%). Conclusion. Molecular information significantly improves the classification of CMML patients, providing a basis for refining diagnostic boundaries with other MN and for a more rational inclusion of patients in clinical trials. The iCPSS demonstrated superior performance over currently available scoring systems and improved HSCT decision making process.

Introduction: T-PLL is a rare, aggressive malignancy with poor prognosis and limited therapies. Outside of prospective trials utilizing alemtuzumab, data is derived primarily from small retrospective series, and limited data on survival outcomes and therapeutic efficacy, particularly beyond frontline therapy are available. In this large multi-center series, with comprehensive clinical data, we evaluated survival and therapeutic outcomes in T-PLL patients, with a focus on identification of prognostic subgroups of T-PLL, that will inform future investigations into the biology and treatment of this disease. Methods: We retrospectively evaluated all patients diagnosed with T-PLL at 15 academic cancer centers throughout the USA between 2000-2023 using the recent TPLL-ISG staging and response criteria with overall response rate (ORR) defined as CR and PR. Overall survival (OS) and progression-free survival (PFS) were assessed using Kaplan-Meier methods and cox proportional hazard models. Single-cell RNA sequencing (scRNAseq) was performed using the standard 10X Genomics platform. Results: 377 patients, with a median follow-up time of 29 months (range: 0-196) were evaluated. The median age at diagnosis was 64 and 51% were male. In evaluated patients (n=223), TCL1A was positive by cytogenetics or FISH in 82%. 61% of patients had CD4+CD8- T-PLL, 11% of patients had CD4-CD8+ T-PLL, and 24% of patients had CD4+CD8+ T-PLL, with 4% unknown. 175 patients (46.4%) had known nodal involvement at diagnosis identified by PET or CT scan. Among treated patients (n=342, 91%), using TPLL-ISG criteria, the ORR to any frontline treatment was 67% (49% CR, 18% PR). Median OS and PFS was 19 months and 10 months in responding patients from the start of treatment initiation. For those treated with frontline alemtuzumab (n=250), the ORR was 74% (58% CR), with a median OS and PFS of 18 and 12 months. 20 patients (5%) were treated with a combination of alemtuzumab/pentostatin frontline with an ORR of 85% (70% CR), which was superior to alemtuzumab alone (ORR 74%, 58% CR). Patients that had frontline chemotherapy (excluding pentostatin, n=48) had a worse ORR of 35% (15% CR) (p<0.001), and a worse PFS of 4 months (p=0.002), when compared to those who received alemtuzumab. Among patients in 1st CR/PR (n=226), 44% (n=99) received an allograft and had an improved 3-year OS of 50% vs 17%, (HR=0.67, 95%CI 0.48-0.93), and 50% improvement in PFS (HR 0.48, 0.33-0.68) compared to those who did not receive an allograft. For patients that received second line treatment (n=176, 52%), re-treatment with alemtuzumab (n=33) produced a 46% ORR (21% CR). Other active second-line treatments included: pentostatin (n=23; 44% ORR, 17% CR), ruxolitinib-based regimens (n=4; 25% ORR, 0% CR), venetoclax-based regimens (n=13; 39% ORR, 8% CR), bendamustine (n=10; 30% ORR, 0% CR), and nelarabine (n=4; 75% ORR, 50% CR). In subgroup survival analysis, TCL1A+ patients had a worse OS and PFS when compared to TCL1A- patients (23 months vs 43 months; p=0.02), and (14 months vs 24 months; p=0.02) respectively, with HR 1.58 (1.08, 2.31) (p=0.02) for OS and HR 1.62 (1.09-2.40) (p=0.02) for PFS. Furthermore, CD4- T-PLL had worse OS (16 months vs 25 months) when compared to CD4+ patients [HR 1.57 (1.08, 2.27), p=0.02]. Single cell RNA seq in 13 patients, including 3 TCL1A- patients, with a total of 66,269 T-PLL cells, demonstrated a unique RNA signature between the CD4+/- and TCL1A+/- patients, suggesting these represent unique molecular subtypes of T-PLL. Conclusions: In this large, multi-center study of T-PLL, frontline treatment with a combination of alemtuzumab/pentostatin improved response rates and OS. Further, among patients who attained a CR, allograft improved OS & PFS vs no allograft. In relapsed T-PLL, retreatment with alemtuzumab had a high ORR, and we report significant activity of multiple novel/alternative agents with the largest reported numbers to date. Intriguingly, TCL1A+, and CD4- T- PLL had worse OS/PFS, and scRNAseq confirmed unique molecular signatures in these populations, suggesting these represent novel molecular subtypes of T-PLL with prognostic significance. These studies form the foundation for future, targeted, therapeutic studies in this rare, aggressive disease with few treatment options.

Introduction: Scant data exist on prognostic factors, survival and treatment (Tx)-related outcomes in SCNSL. We conducted a multicenter international analysis to identify prognostic factors and outcomes in SCNSL at diagnosis (de novo), or after frontline Tx (1L) either with (concomitant) or without (isolated) systemic (syst) relapse. Methods: This study collected data from patients (pts) treated between 2001-2023 at 35 centers. Overall (OS) and progression-free survival (PFS) were calculated from SCNSL diagnosis to first event (death for OS, relapse/death for PFS) or last follow-up. Uni- (UVA) and multivariable (MVA) Cox analyses were performed to assess prognostic factors; Kaplan-Meier method was used to estimate survival. Results: 1173 (US: 863, UK: 273, Canada: 37) pts with SCNSL were included. Baseline characteristics were median age: 62 years (interquartile range 52.5-69), male sex: 58.7%, elevated LDH: 71.1%, stage III/IV: 88.6%, HIV: 3%, transformed LBCL (tLBCL): 13.8%, non-germinal center B-cell like (non-GCB) subtype: 52.3%, double expressor (DE): 40%, MYC rearrangement (MYC-R): 28.3%, MYC/BCL2 double hit: 16.6%, MYC/BCL2/BCL6 triple hit (TH): 6.1%. Among 522 pts with de novo SCNSL, 84.2% had involvement of a single CNS compartment (44.7% brain parenchyma; 40.4% leptomeningeal [lepto]). 1L regimens included 36.2% R-CHOP with high-dose methotrexate (HD-MTX), 12.5% MARIETTA-type, and 9.8% R-CODOX-M/IVAC; 54.6% received intrathecal chemotherapy. Consolidation Thiotepa-based autologous stem cell transplant (TT-ASCT) occurred in 15.3%. The majority of pts who underwent TT-ASCT had previously achieved CR or PR in syst (92.5%) and CNS (72.5%) compartments. At median follow-up 25.7 months (mos), subsequent relapse was observed in 27.5% TT-ASCT recipients: 36.3% had isolated CNS, 27.4% syst, and 36.3% concomitant. 588 pts presented with SCNS after 1L (66.2% isolated, 33.8% concomitant). Brain parenchyma involvement was observed in 56.8%, lepto in 22.6%. Tx included 31.6% HD-MTX combination (eg, R-MPV, MTR); 18.4% HD-MTX +/- rituximab; 16.2% MARIETTA-type; and 9.4% platinum-based regimen. Consolidation TT-ASCT occurred in 22.4% (n=132), and 13.8% (n=81) received CAR-T cell therapy (excluding subsequent relapses in de novo group). Syst (65.1%) & CNS (69.1%) responses (CR/PR) were higher before TT-ASCT compared to CAR-T cell (49.5% & 54.5%, respectively). After TT-ASCT, 52 pts (39.3%) relapsed, mainly in the CNS (n=30, 57.7%), with only 4 (7.7%) concomitant. After CAR-T, 48 pts (59.3%) relapsed: 15 (31.2%) isolated in CNS and 15 (31.2%) concomitant. 12-month cumulative incidence of relapse (with death as competing risk) was 21.1% after TT-ASCT and 55.8% after CAR-T. In 63 pts (5.4%), the first CNS event occurred >2 lines of Tx. With median follow-up of the entire cohort of 24.6 mos, median PFS and OS were 11 and 21 mos, respectively. mPFS and mOS were significantly better in pts with de novo SCNSL (13.4 & 47.4 mos) compared to SCNSL after 1L (10.4 & 14.2 mos) and >2 lines of Tx (6.5 & 9.7 mos); P<0.0001. Factors associated with significantly shorter OS in UVA across all presentations were age >60, male sex, GCB subtype, MYC-R, tLBCL, DE, TH, elevated LDH, ECOG PS ≥2, SCNSL after 1L (vs. de novo), and failure to achieve syst/CNS response to 1L. Of these, age >60, male sex, MYC-R, ECOG PS ≥2, and failure to achieve response to 1L remained significant in MVA. In de novo cases, age >60, DE, ECOG PS≥2, and failure to achieve syst/CNS response to 1L were associated with poorer OS; for pts with SCNSL after 1L, age >60, GCB subtype, and ECOG PS ≥2 were significant. Pts treated with HD-MTX-based regimens after 1L experienced longer OS than those treated with platinum regimens or high-dose cytarabine. In de novo cases, TT-ASCT was significantly associated with longer OS (HR=0.58; P=0.025) in MVA after adjusting for significant factors, 1L type, and response. For SCNSL at relapse, TT-ASCT (HR=0.34; P<.0001) and isolated SCNSL presentation (vs. concomitant HR=0.70, P=0.007) were associated with longer OS, adjusting for significant factors, type of second-line Tx, & site of CNS relapse. Conclusions: This study represents the largest analysis of SCNSL inclusive of all modes of presentation. Concomitant SCNSL at relapse was associated with poor outcomes, while TT-ASCT consolidation was associated with improved OS in de novo & relapsed disease. Despite advances in LBCL SCNSL remains an unmet therapeutic need.