Subpopulation Of Patients Research Articles

Novel Preoperative Risk Stratification Using Digital Phenotyping Applying a Scalable Machine-Learning Approach.

Classification of perioperative risk is important for patient care, resource allocation, and guiding shared decision-making. Using discriminative features from the electronic health record (EHR), machine-learning algorithms can create digital phenotypes among heterogenous populations, representing distinct patient subpopulations grouped by shared characteristics, from which we can personalize care, anticipate clinical care trajectories, and explore therapies. We hypothesized that digital phenotypes in preoperative settings are associated with postoperative adverse events including in-hospital and 30-day mortality, 30-day surgical redo, intensive care unit (ICU) admission, and hospital length of stay (LOS). We identified all laminectomies, colectomies, and thoracic surgeries performed over a 9-year period from a large hospital system. Seventy-seven readily extractable preoperative features were first selected from clinical consensus, including demographics, medical history, and lab results. Three surgery-specific datasets were built and split into derivation and validation cohorts using chronological occurrence. Consensus k -means clustering was performed independently on each derivation cohort, from which phenotypes' characteristics were explored. Cluster assignments were used to train a random forest model to assign patient phenotypes in validation cohorts. We reconducted descriptive analyses on validation cohorts to confirm the similarity of patient characteristics with derivation cohorts, and quantified the association of each phenotype with postoperative adverse events by using the area under receiver operating characteristic curve (AUROC). We compared our approach to American Society of Anesthesiologists (ASA) alone and investigated a combination of our phenotypes with the ASA score. A total of 7251 patients met inclusion criteria, of which 2770 were held out in a validation dataset based on chronological occurrence. Using segmentation metrics and clinical consensus, 3 distinct phenotypes were created for each surgery. The main features used for segmentation included urgency of the procedure, preoperative LOS, age, and comorbidities. The most relevant characteristics varied for each of the 3 surgeries. Low-risk phenotype alpha was the most common (2039 of 2770, 74%), while high-risk phenotype gamma was the rarest (302 of 2770, 11%). Adverse outcomes progressively increased from phenotypes alpha to gamma, including 30-day mortality (0.3%, 2.1%, and 6.0%, respectively), in-hospital mortality (0.2%, 2.3%, and 7.3%), and prolonged hospital LOS (3.4%, 22.1%, and 25.8%). When combined with the ASA score, digital phenotypes achieved higher AUROC than the ASA score alone (hospital mortality: 0.91 vs 0.84; prolonged hospitalization: 0.80 vs 0.71). For 3 frequently performed surgeries, we identified 3 digital phenotypes. The typical profiles of each phenotype were described and could be used to anticipate adverse postoperative events.

Neuronal hyperexcitability in AD

AbstractBackgroundBecause neuronal and network deficits are the imperative immediate consequences of pathological changes of disease mechanisms, neurophysiological manifestations associated with Aβ and tau are uniquely positioned to demonstrate the abnormal homeostatic and maladaptive circuit alterations and the dynamic integrated effects of underlying genetics and neuropathological changes in Alzheimer’s disease (AD). Importantly, network dysfunction manifesting as abnormal hyper‐ as well as hypo‐active oscillatory brain rhythms is detectable even during the prodromal phase of disease. Here we leveraged the higher spatial and precise temporal resolution of Magnetoencephalography (MEG) to define frequency‐specific and region‐dependent patterns of oscillatory deficits in AD patients, and to demonstrate the earliest neurophysiological abnormalities associated with AD neuropathology and their mechanistic relationships to network hyperexcitability.MethodWe studied the resting brain oscillations in individuals with AD‐spectrum (n = 88) and age‐matched controls (n = 70) using MEG. In addition, subsets of participants were evaluated with positron emission tomography (PET) with Aβ and tau specific tracers and fluid biomarker including plasma P‐tau181 and Neuro‐Filament‐Light (NFL). We used statistical mixed models to investigate the associations between neural oscillations and pathology. We used a linear Neural‐Mass‐Model (NMM) in combination with empirical MEG spectra to derive the excitatory‐to‐inhibitory activity within neuronal subpopulations in AD patients. Based on standard neurophysiological monitoring we determined clinical evidence of hyperexcitability as AD with and without epileptiform activity (AD‐EPI+; AD‐EPI‐).ResultWhile increase and decrease of oscillatory activity are distinctly associated Aβ and tau, respectively, the characteristic frequency‐specific manifestation in AD was increased spectral signature within delta‐theta (2‐7Hz) and reduced spectral signature within alpha (8‐12Hz). Earliest spectral changes were detected in Aβ+ cognitively unimpaired and the increased spectral signature associated with Aβ was replicated in identical associations with plasma p‐tau181. The presence of APO‐E4 allele strengthened these associations. NMM demonstrated greater excitatory‐inhibitory‐activity in AD‐EPI+ vs. AD‐EPI‐ while these were specifically correlated with the reduced alpha spectrum which was more profound in AD‐EPI+ as well as uniquely associated with increased tau depositions.ConclusionOur results provide critical insights about potential mechanistic links between abnormal neural oscillations and cellular correlates of impaired excitatory and inhibitory synaptic functions associated with tau and Aβ, in patients with AD.

Metástasis pulmonares aisladas de adenocarcinoma ductal de páncreas: una población con supervivencia prolongada

The incidence of isolated pulmonary metastases in pancreatic ductal adenocarcinoma is about 13%. Resection of these metastases is uncommon; however, patients presenting only with pulmonary metastases have better survival compared to those with metastases on other locations. We report the case of a female patient who underwent lobectomy for metastases from pancreatic ductal adenocarcinoma. After resection, disease-free interval and specific diseases-free survival were 84 and 152 moths, respectively. We consider that the next step in the treatment of this subpopulation of patients is to select those patients with favorable tumor biology who would benefit from a more aggressive approach.

Abstract A134: Predicting activity of IMM-1-104 as single agent and in combination for patients with RAS or RAF mutant tumors

Abstract Introduction: Many tumors are addicted to MAPK pathway activation, including the >20% of human tumors with mutations in RAS or RAF1. IMM-1-104 is an oral once-daily treatment currently in Phase 1 in patients with RAS-mutant solid tumors [NCT05585320]. To date, drugs disrupting the MAPK pathway have done so chronically, leading to dose-limiting toxicities (DLTs) and poor response durability. In contrast, IMM-1-104 was designed to provide deep cyclic inhibition (DCI) of the MAPK pathway via a unique pharmacokinetic (PK) profile with high peak plasma drug levels and a near zero drug trough between doses. This promotes pulsatile inhibition of MEK, depriving tumors of sustained signaling of a critical oncogenic pathway while limiting toxicity and durability issues associated with chronic MEK inhibition. In Phase 1a dose escalation, no DLTs were observed, the plasma drug half-life was ~2-hours, and pharmacodynamic (PD) data were consistent with DCI. Phase 1b dose expansion is underway. Translational efforts are focused on identifying MAPK pathway addiction and sensitivity to IMM-1-104. Tumor models displaying patient-aligned genomic profiles against large patient databases such as AACR Project GENIE1 were tested in humanized 3D tumor growth assays (3D-TGA). Computational modeling based on response and in-house genomic data was used to inform identification of patient populations for IMM-1-104 monotherapy and potential combination opportunities. Experimental Procedures: Using cancer-specific, patient-aligned cell lines, IMM-1-104 activity was characterized in the 3D-TGA. Whole exome sequencing was performed to confirm alteration status, and a further subset subjected to RNA sequencing. Pharmacogenomic data were used to generate a model predictive of response to IMM-1-104 and identify biomarker-aligned patient subpopulations. Selected model predictions were then tested in subcutaneous tumor xenograft models in female BALB/c nude mice. Summary of New Data: Assessment of IMM-1-104 across >190 patient-aligned models demonstrated diverse responses across a wide range of MAPK-driven tumor types, including those with RAS or RAF mutations. In addition to RAS, these data suggested additional potential for IMM-1-104 in BRAF-mutant disease. Therefore, IMM-1-104 was tested alone and with encorafenib in the HT-29 colorectal BRAFV600E mutant xenograft model. Monotherapy with either encorafenib or IMM-1-104 displayed superior tumor growth inhibition to binimetinib. IMM-1-104 in combination with encorafenib drove deeper regressions and superior durability of response in a head-to-head in vivo comparison versus binimetinib plus encorafenib. Conclusions: We used an integrated platform of translational experiments and informatics to identify patient-aligned model systems, prioritize factors relevant for response to IMM-1-104’s unique DCI profile, and elucidate combination opportunities to potentially inform clinical development strategies. Citation Format: Praveen Nair, Sarah Kolitz, Jason Funt, Jan de Jong, Peter King, Amy Yamamura, Mai Johnson, Jenny Zhang, Kevin D Fowler, Anna Travesa, Amy Axel, Chris Walker, Scott Barrett, Benjamin J Zeskind, Brett Hall. Predicting activity of IMM-1-104 as single agent and in combination for patients with RAS or RAF mutant tumors [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr A134.

Alteration in innate immune cues assessed by analyzing peripheral blood immune system in post-covid patients

The COVID-19 pandemic, caused by the SARS-CoV-2 virus, has led to global morbidity and mortality. Some patients fully recover from COVID-19, whereas around 45% subjects suffer from various persistent symptoms (fatigue, cognitive impairment, impaired thermoregulation, skin diseases, etc.) for at least four months after SARS-CoV-2 infection regardless of disease severity. Such persistent post-infection effects are known as long-COVID, post-acute effects of COVID-19, or post-COVID state. SARS-CoV-2 infection is accompanied by damage to the innate immune system. Considering the role of natural killer cells and the activation of the complement system in COVID-19 as well as the regulatory properties related to CD46 and its potential involvement in cell virus entry, we found necessary to study immune system parameters associated with impairment of these innate immune cues on various leukocyte subpopulations in post-COVID patients. We studied 92 immune system parameters, including: pan-leukocyte markers for gated lymphocytes, phenotyping of T cells, T-helper inducers, cytotoxic T-lymphocytes, NK- and TNK-cells, T-regulatory cells/suppressors, B-lymphocytes, including B-memory cells, activated helpers, early activation of lymphocytes, activated T-lymphocytes, and late lymphocyte activation markers. Levels of total IgM, IgG, IgA, specific IgM, IgG to coronavirus COVID-19, C1-inhibitor, C3a, and C5a complement components were measured by enzyme immunoassay using Multiscan FC Thermoscientific enzyme immunoassay analyzer (China) and Vector Best reagent kits (Russia). A complete blood count was conducted to study 25 parameters: leukocyte, erythrocyte, and platelet hematopoietic lineages as well as the quantitative and qualitative composition of hematopoietic lineages. Our study results showed that in some patients, six months after suffering from COVID-19, there was a decrease in the level of NK cells (48%) and CD46+pan-leukocyte marker cells (64.5%). A decrease in NK cell levels was accompanied by increased level of total T- and B-lymphocytes, and altered platelet and erythroid hematopoietic lineages. In patients with reduced CD46 expression on T-lymphocytes, both their total count and NK cell count were significantly reduced. Our data also suggest that CD46 might be potentially involved in development of SARS-CoV-2 infection and the post-COVID state. Thus, in 50–65% of patients who have experienced SARS-CoV-2 infection, damage to the innate immune system persists after six months being accompanied by impaired erythroid and platelet hematopoietic lineages. The data obtained indicate a need for using immunocorrective therapy in such patients.

Genetic Variants in Canonical Wnt Signaling Pathway Associated with Pediatric ITP

Introduction : Pediatric immune thrombocytopenia (ITP) is thought to develop when a triggering event inappropriately activates an autoimmune anti-platelet response in the context of predisposing genetic factors. Inborn errors of immunity are increasingly being identified in cases of chronic, refractory autoimmune cytopenias. We sought to identify genetic variants associated with predisposition to 1) primary or secondary pediatric ITP of any disease duration versus healthy controls and 2) chronic ITP (cITP) versus disease that spontaneously resolves in < 1 year (sr-ITP). Methods : DNA and clinical data were derived from 8 sites from the Pediatric ITP Consortium of North America's (ICON) biorepository. Samples were genotyped using the Illumina Global Screening Array BeadChip. Genotypes were imputed using the Michigan Imputation Server HRC reference panel. Genetic variants that passed standard quality metrics and had a MAF ≥1% were retained. Control samples (n=8,372) were obtained from the National Longitudinal Study of Adolescent to Adult Health (Add Health) cohort. Logistic regression models were used to estimate adjusted odds ratios (aORs) comparing (1) ITP and control cohorts and (2) cITP and sr-ITP cohorts. Models assumed additive allelic effects and accounted for the top 3 principal components. Genome-wide significance was defined as P <5 × 10 -8 and suggestive of genome-wide significance, P <1 x 10 -5. Results : Six-hundred ten unique patient samples were genotyped. Nineteen (3.1%) did not pass quality control, leaving 591 for analysis. This cohort included 307 (52.0%) cITP patients, 209 (35.4%) sr-ITP patients, and 75 (12.7%) individuals without complete follow-up data to allow for classification of cITP or sr-ITP. The ITP cohort was 50.1% male, 50.1% non-Hispanic (NH) white, 29.4% Hispanic, 6.1% NH Black, 3.1% NH Asian, and 11.3% Other/Unknown. Compared with healthy controls, we identified 6 loci associated with ITP at genome-wide significance (Figure 1). Located within the intronic region of SLC24A4 rs75650997 was associated with an increased risk of ITP compared with controls (aOR: 2.38, P=3.9x10 -8). An intronic variant, rs2165798, in NAV2 was inversely associated with ITP (aOR: 0.52, P=3.2x10 -11). A variant in NKD1 rs4785216, was inversely associated with ITP (aOR: 0.43, P=8.86x10 -15).Another intergenic variant, rs4785426, in close proximity to NKD1 was also inversely associated with ITP (aOR: 0.48, P=1.84x10 -16) . An additional 20 variants met suggestive of genome-wide significance. When comparing sr-ITP and cITP patients, no SNVs (single nucleotide variant) reached genome-wide significance, but one variant, rs117609649, in ABCC4 met suggestive of genome-wide significance (aOR: 0.26, P=4.49x10 -7). Discussion : This is the largest, unbiased genetic study of pediatric ITP patients. The analysis comparing ITP versus healthy controls identified 20 suggestive SNVs and 6 which met genome-wide significance. Several of the top SNVs identified in this study affect Wnt/β-catenin signaling. Wnt-signaling impacts the development and regulation of immune cell populations including dendritic cells, macrophages, NK cells, B cells, and T cells. NAV2 activates Wnt/β-catenin signaling. NAV2 variants have been implicated in autoimmune conditions including Celiac disease and rheumatoid arthritis. NKD1 inhibits the canonical Wnt/β-catenin signaling pathway. The cumulative impact of these variants on Wnt activity in ITP should be evaluated in the future. No genetic variants met genome-wide significance in a comparison of sr-ITP and cITP patients. Collectively, these findings suggest there may be genetic variants implicated in ITP risk, but those genetic features are not specific to sr-ITP or cITP. A focused analysis of 632 genes that cause inborn errors of immunity (IEI) is ongoing. This includes gene burden testing and analysis of pathogenic variants within this gene subset. Variants in IEI genes will also be evaluated in a sub-population of patients with Evans syndrome. Conclusion : ITP is highly heterogenous in its presentation, disease course, and response to therapy. Mirroring its clinical behavior, the genetic contribution to ITP development is complex. This study identified potential SNVs associated with ITP, but not specific to sr-ITP or cITP. Variants in genes that affect the canonical Wnt-signaling pathway may play a role in ITP risk.

Integrative Analysis of the Tumor and Microenvironment to Model the Molecular Heterogeneity Underlying the Response to Cevostamab in Relapsed/Refractory Multiple Myeloma

Background: Cevostamab is an FcRH5xCD3 bispecific antibody that facilitates T-cell directed killing of myeloma cells and has shown promising activity and a favorable safety profile as monotherapy in patients with relapsed/refractory multiple myeloma (RRMM) in the ongoing Phase I GO39775 trial (NCT03275103; Trudel et al. ASH 2021). We performed an integrative analysis of peripheral blood (PB) and bone marrow aspirate (BMA) samples collected at baseline from patients in GO39775 to understand the heterogeneity of patients with RRMM and potential correlations with clinical outcome. Methods: PB and BMA samples were collected at baseline from 100 patients in GO39775. Patients who had less than 25% missingness in any data layer were selected from the single-step (3.6/90-252mg dose levels [n=49] and 3.6/90mg tocilizumab premedication group [n=14; Trudel et al. ASH 2022]) and double-step (0.3-1.2/3.6-7.2/60-160mg dose levels [n=37]) dose-escalation and dose-expansion cohorts. To characterize immune cells, samples were profiled at a central laboratory by 8-color flow cytometry. Cytokine profiling was performed on baseline plasma samples via electrochemiluminescence and digital ELISA. Serum B-cell maturation antigen (sBCMA) levels were quantified using hybrid immunoaffinity capture by liquid chromatography with tandem mass spectrometry. For the unbiased integrative analysis, a Similarity Network Fusion algorithm was applied to preprocessed flow cytometry data from PB and BMA samples, and to baseline characteristics including demographics, prior therapies, and selected clinical laboratory analytes. Data were imputed in samples with less than 25% missingness using a random forest algorithm. Results: Initial findings from unsupervised machine learning-based integrative clustering analysis of baseline samples from multiple immunological and clinical data types revealed three patient subpopulations with unique biological features, illustrating the molecular heterogeneity of patients with RRMM. Each cluster was characterized by distinct patterns of immune cell proportions and activation states. Interestingly, these immunologically defined clusters also corresponded to clinical outcomes. Cluster 1 had the strongest response to cevostamab, followed by Cluster 2 and Cluster 3. Conclusions: Our novel, unbiased, machine learning, integrative clustering approach not only revealed unique molecular subtypes that illustrate the heterogeneity of patients with RRMM, but also identified immunological features associated with potentially reduced response to cevostamab therapy, which could provide direction for future research.

An Honeur Framework for Generating Computer Understandable Cohort Definitions from Clinical Trial Protocols in Multiple Myeloma through Generative AI for Comparing with Real-World Data

Background Study protocols define the population(s) of interest through a set of inclusion and exclusion criteria -often in the form of a textual description. In haematology, these criteria can become very complex and are difficult to translate into consistent and robust computer interpretable cohort definitions. Addressing this interoperability challenge has the potential to vastly accelerate the evaluation of populations of interest from a trial to a real world setting or to replicate studies. Generative AI (GenAI) and its potential to extract valuable information from textual data such as medical records or research papers has recently received a lot of attention. Building on the clinical data harmonisation that already took place across the Haematology Outcomes Network in Europe (HONEUR), the application of the latest evolutions in GenAI technology for the generation of computer interpretable cohort definitions from descriptions in study protocols has been assessed and potential limitations have been identified. Methods HONEUR is a federated data network of 21 registries and hospitals in 9 different countries, comprising 45,000 multiple myeloma patients, where the data stays local and all participating data partners have full governance (www.honeur.org). The main aim of the network is to accelerate evidence generation to expand the knowledge and understanding of patients with hematological malignancies. The pivotal step in enabling the federated analysis is the clinical data harmonization. This is based on structural and semantic harmonization to the OMOP data model (https://www.ohdsi.org/data-standardization/). For studies involving subpopulations of multiple myeloma patients, a simplified tabular structure is generated from the OMOP data. This structure contains an extensive list of baseline and line of treatment specific variables. For the conversion of the inclusion/exclusion criteria into a standard format that could be executed by the HONEUR tools, ChatGPT (https://chat.openai.com/) version 4.0 was used. The following steps were hereby followed as a set of ‘prompts’: Define the role and system prompts: explaining in plain language what the intent is and what steps will be followed.Provide information on the data structure that contains the harmonized data: the data dictionary from the tabular structure mentioned above with the variable names, the description of these and the allowed values of the variables.Providing the target structure of the output fileProvide the inclusion / exclusion criteria as “training” promptsInstruct ChatGPT on how to behave with unknown scenarios.Repeat Step 5 until the outcome file correctly represents the cohort description as provided in Step 4 The result of the above steps is a correctly formatted file that is evaluated by Nuffield Cancer Centre London through upload in the local HONEUR real-world database and evaluated to assess the validity of the cohort definition(s) Results In total 6 patients met all criteria in Nuffield Cancer Centre London (see Table 1). The system was further evaluated on more complex inclusion of cytogenetic and subsequent treatment inclusion criteria. [Table 1] Conclusions In our setting, we've found that one-shot or few-shot learning is adequate to correctly translate complex scenarios of inclusion / exclusion rules into executable cohort definitions through the use of GenAI. The upfront semantic clinical data harmonization is however a prerequisite as the tools are not yet adequately trained to detect subtle but sometimes critical nuances in the clinical criteria. The application of GenAI is promising for conversion of textual data to structured and computer interpretable information. This offers the possibility to accelerate evidence generation in a robust and reproducible way. We showed this system can work in identifying 6 patients in one center and is therefore extrapolatable to all 21 participating centers.

Identification of Unexpected Responders to Gilteritinib in a Cohort of FLT3-WT AML Patients Using Ex-Vivo Testing

Introduction: Despite the numerous significant advances in leveraging actionable genetic biomarkers, targeted therapies continue to exhibit limited success in AML. This may be attributed to the genetic heterogeneity and plasticity of cancer cells, which ultimately overrides the driving force of individual mutations and their resulting phenotypes. In the face of this well-known challenge, there is a growing promise of functional assays that can leverage the integrated phenotypic response of cancer patients' cells to determine treatment schemes in a personalized manner. In addition, functional testing holds the potential to identify elusive responders to targeted therapies in cohorts of patients who do not present the actionable mutations that are used as biomarkers for prescription. Aim: To perform pre-clinical correlation studies between FLT3 mutational status and the ex-vivo response to Gilteritinib using a functional assay of Patient Micro Avatars (PMAs) developed at OncoPrecision. Materials and Methods: Ex-vivo testing with PMAs was performed to evaluate the ex-vivo response to Gilteritinib in 67 bone marrow or peripheral blood samples from patients diagnosed with different AML subtypes (de novo, relapsed/refractory and secondary). Pathological subpopulations were analyzed using lineage markers and survival of blasts was assessed using flow cytometry. Samples were classified as “responders” and “non-responders” using in-house supervised and non-supervised machine learning tools. In parallel, samples were analyzed by NGS and molecular biology to identify mutations and internal tandem duplications in the FLT3 gene. Finally, FLT3 alterations were correlated with the phenotypic response obtained in the functional assay using statistical analysis to compare mutant and non-mutant frequencies in the two groups. Results: The results of the phenotypic screening revealed that ~70% of patients with FLT3 mutations exhibited a strong response to Gilteritinib, which is in line with the presence of the actionable genetic biomarker (Figure). The fact that the remaining ~30% of mutated patients presented a poor response to Gilteritinib suggests the existence of acquired mechanisms of resistance that are dominant over the mutational status of FLT3. Interestingly, we also identified a large group of FLT3wt patients (~30%) which showed a phenotypic response to Gilteritinib equivalent to the one observed in FLT3mut patients (Figure). This finding unveils the existence of a subpopulation of FLT3wt patients who may benefit from Gilteritinib treatment and cannot be identified using conventional genetic testing. Conclusions: The present study highlights the reach of functional testing as a powerful tool to predict Gilteritinib resistance in FLT3mut backgrounds and to identify exceptional responders in FLT3wt backgrounds, creating opportunities for AML patients beyond the currently available genetic biomarkers.

Soluble PD-L1 Predicts Poor Outcome and Disease Progression in Patients with De Novo Myelodysplastic Syndromes

Background: Myelodysplastic syndromes (MDS) consist of a spectrum of myeloid malignancies typified by regular genetic abnormalities, clonal hematopoiesis, dysfunctional myelopoiesis, inefficient blood cell production, peripheral-blood cytopenia, and an elevated risk for progressing to acute myeloid leukemia (AML). Mounting evidence suggests that the dysregulation of immune checkpoints (ICs) plays critical roles in immune evasion of MDS, which thus led to a series of trials using HMAs in combination with immune checkpoints inhibitors (ICIs). Of these, programmed cell death-1 (PD-1)/programmed cell death ligand-1 (PD-L1) pathway has been studied most extensively. Recently, forms of cell-free soluble PD-1/PD-L1 (sPD-1/sPD-L1) detection in plasm of patients with solid tumors has open a new paradigm for PD-1/PD-L1 investigations. However, within the context of MDS, the potential dysregulation and prognostic impact of sPD-1/sPD-L1 remains uncertain. This study was aimed to assess roles of sPD-1/sPD-L1 in newly diagnosed MDS patients. Method: Between July 2020 and March 2023, 130 MDS/sAML patients matched the inclusion criteria and 55 healthy individuals with no familial or personal history of autoimmune or cancer disorders were included. The criteria for patient inclusion included: (1) a diagnosis of MDS confirmed pathologically according to 2016/2022 World Health Organization (WHO) classification, or sAML patients with a verified prior MDS history; (2) absence of other conditions like autoimmune diseases or different types of tumors; (3) availability of complete clinical information and survival data. MDS patients were categorized by IPSS-R as: lower-risk (very low risk/ low risk; higher-risk (intermediate risk/high risk/very high risk. Follow-up was conducted until June 16th, 2023. The median follow-up was 14.32 months (range 3.2-34.63 months). Profile of sICs (sPD-L1, sPD-1, sCTLA-4, sGITR, sTIM-3, sOX40, sLAG-3, s4-1BB, sICOS) were tested by ELISA, cytokines (IFN-γ, IFN-α2, IL-2, IL-2α, IL-6, IL-7, IL-10, IL-15, IL-17, TNF-α) were detected using Luminex analysis, and the expression levels of PD-1 on T cells (CD3 +, CD4 +, CD8 +, DPT, DNT）were measured using flow cytometry (FCM). Result: In the plasma of 98 newly diagnosed MDS patients, sPD-L1 levels (median 66.80 pg/ml, range 25.00-219.5 pg/ml) were significantly elevated compared to 55 healthy blood donors (median 49.22 pg/ml, range 12.5-218.1 pg/ml). sPD-L1 level was higher in higher-risk IPSS-R groups ( P=0.0380). In addition, higher sPD-L1 levels also showed a correlation with lower hemoglobin concentration, implying a potential link between higher sPD-L1 levels and a more advanced MDS disease state. The AUC for sPD-1 was found to be 0.6333 [95% CI 0.5331-0.7334, P=0.0100] and for sPD-L1, it was 0.6785 [95% CI 0.5836-0.7733, P=0.0002]. The optimal cut-off values were established at 187.5 pg/mL for sPD-1 and 55.08 pg/mL for sPD-L1. Based on these cut-off values, patients with increased sPD-L1 experienced shorter OS ( P=0.0305). By using multivariate Cox model, high expression of sPD-L1 ( P=0.019, HR = 4.172, 95% CI: 1.265-13.755) and high IPSS-R scores ( P=0.030, HR = 2.898, 95% CI: 1.111-7.565) were found to be independent risk factors for newly diagnosed MDS patients. In addition, compared to healthy controls, the PD-1 + T cell subpopulations in MDS patients were significantly reduced, sPD-1 levels were positively correlated with proportion of CD4 + PD-1 + T cells, absolute count of CD4 + PD-1 + T cells and PD-1 + DPT cells, suggesting that CD4 + T cells and DPT cells may be potential sources of sPD-1 in plasma. Subsequently, we analysed the potential roles of cytokines and HMAs in sPD-1/sPD-L1 regulation. It was found that sPD-1 levels had a positive correlation with sPD-L1, IL-2Rα, IP-10, MIG, TNFα, GROα, and TRAIL. sPD-L1 was positively correlated with IL-2Rα, IP-10, MCP-1, MIG, MIP1α, and SCF (Figure 1 A). Notably. We found a reduced sPD-L1 levels in MDS patients who achieved remission after HMA treatment compared to treatment-naïve patients. ( P=0.0302) (Figure 1 B). Conclusions: In conclusion, we found sPD-L1 is an independent risk factors for overall survival in newly diagnosed MDS patients. The elevation of plasma sPD-L1 levels is associated with disease progression in MDS, and may be a potential target for MDS immunotherapy.

CD123 As a Therapeutic Target in Accelerated and Blast Phase Myeloproliferative Neoplasms (MPNs)

Background: CD123 expression in MPNs has been reported in case series and CD123 has been a therapeutic target in one phase I study of patients with chronic myelomonocytic leukemia or myelofibrosis in chronic phase. CD123 has also been described as a marker of leukemic stem cells and incorporated in chimeric antigen receptor T-cell products for treatment of acute myeloid leukemia in clinical trials. Herein, we describe the clinical phenotype of patients with accelerated (AP) or blast phase (BP) MPNs paired with CD123 co-expression on CD45(dim) myeloid blasts. Methods: This is a retrospective, single-institution study of patients with AP and BP MPNs. Subjects were identified by search for MPN diagnoses in the Ohio State University Comprehensive Cancer Center Leukemia Tissue Bank (OSUCCC-LTB). Relevant clinical data was collected by retrospective chart review with approval of Institutional Review Board. CD123 expression was measured by immunophenotyping of cryopreserved peripheral blood (PB) or marrow (BM) samples. Blasts were identified by CD45(dim) gating of lymphocytes. Lymphocytes were also gated on CD7, CD13, CD33, CD34, and CD117. CD123 expression was stratified by quartile and correlation with disease phase, symptoms, splenomegaly, transfusion-dependence, and molecular mutations measured by Chi-square test. Results: Forty-one PB or BM samples from 24 subjects were identified from the OSUCCC-LTB. Nine subjects had paired PB and BM specimens and 6 subjects had samples obtained at multiple timepoints in the course of their disease. Median age at time of sample collection was 65.8 years (range 35.2-79.7) and median time from initial diagnosis to sampling was 13.1 months (range 0-96.5). Constitutional symptoms were present at the time of collection in 24 specimens (58.5%) and splenomegaly present in 20 specimens (51.3%). At sample collection, 19.5% had never received treatment, 19.5% had been previously treated, and 61% were currently on treatment. Most patients on treatment (n=16, 64%) were receiving JAK inhibitor at sample collection. Nineteen samples were obtained in chronic phase (46.3%), nine in AP (22.0%), and thirteen in BP (31.7%). Median WBC count was 4.37K/uL (range 0.3-174.6), median peripheral blasts were 3.2% (range 0-76%), median hemoglobin was 8.8 g/dL (range 5.6-13.2), and median platelets were 42K/uL (range 5-564). Sixteen samples (39.0%) were from transfusion-dependent patients, of which, 68.8% were both red blood cell (RBC) and platelet transfusion-dependent, 25.0% were RBC transfusion-dependent only, and 6.2% were platelet transfusion-dependent only. Median lactate dehydrogenase at time of sampling was 300 U/L (range 116-2,142). Median bone marrow blasts at time of sampling were 3% (range 0-82%). Molecular sequencing and karyotype +/- 30 days of sample collection were available for 18 (43.9%) and 27 (65.9%) of specimens, respectively. The most common mutations identified were JAK2 (50%), ASXL1 (27.8%), SRSF2 (22.2%), CALR (16.7%), and TP53 (16.7%). Normal karyotype was reported in 37% of the samples (one was consistent with donor chimerism post-allogeneic SCT), 33.3% had complex karyotype, 11.1% had del 5q or 7q/-7, and 18.5% had del 13q or 20q. CD123 expression on the CD45(dim) gated blasts ranged from 0.3-95.6% with a median value of 11.02%. CD123 expression is presented in figure 1 box plot. Figure 2a shows correlation between CD123 expression and AP/BP compared to chronic phase disease (p < 0.01). Figure 2b shows CD123 expression correlation with ASXL1 compared to unmutated ASXL1 (p =0.023). There were no statistically significant associations between CD123 expression and any other tested variables. Conclusions: These single-institution data suggest that high CD123 expression on blasts is associated with leukemic transformation and high-risk disease. This is demonstrated by correlation of higher expression in AP and BP MPN as well as correlation with ASXL1 mutation. CD123 expression in MPNs has been previously reported and explored as a therapeutic target in a single phase I trial, though these data suggest that specifically targeting high risk disease, defined by blasts >10% or the presence of ASXL1 mutation, may be more effective with greater target availability. This is an especially important subpopulation of patients with MPNs in need of more effective and less toxic therapy than current standards of care that typically include cytotoxic chemotherapy.

Lintuzumab-Ac225, a CD33-Directed Antibody Radiotherapy, Targets AML in a Mutation Agnostic Manner

Introduction Acute myeloid leukemia (AML) is a highly heterogeneous hematologic malignancy with a poor survival prognosis. Recurrent gene mutations are the major obstacle that impact treatment efficacy of AML patients. Actimab-A, an anti-CD33 antibody conjugated with Actinium-225 (Ac-225) alpha particle-emitting radionuclide (lintuzumab-Ac225), has demonstrated promising therapeutic responses in AML patients, including in combination trials with chemotherapy regimen CLAG-M and targeted therapy venetoclax. Because lintuzumab-Ac225 broadly targets myeloid cells via CD33, we hypothesized that anti-leukemic activity is agnostic of genetic abnormalities commonly found in relapsed/refractory AML, including FMS-like tyrosine kinase-3 ( FLT3) mutations and mixed lineage leukemia ( MLL) rearrangements. While small molecule FLT3 and menin inhibitors that target these patient subpopulations show anti-leukemic benefit, most patients eventually relapse. This study uses preclinical models to evaluate the anti-leukemic activity of lintuzumab-Ac225 in FLT3 and MLL mutation harboring AML cells as either a single agent or in combination with FLT3 or menin inhibitors. Methods Lintuzumab-Ac225 was generated by conjugating lintuzumab with p-SCN-Bn-DOTA and subsequently radiolabeled with Ac-225. A viability assay was performed using flow cytometry to assess the response of the FLT3 mutant and MLL rearranged AML cell line MV-4-11 to lintuzumab-Ac225, FLT3 inhibitor (Gilteritinib) or MLL inhibitor (Revumenib) single agents and the combinations. Furthermore, the effects of lintuzumab-Ac225 on AML growth in vivo were investigated using the subcutaneously injected MV-4-11 leukemia xenograft model in nude mice. Results Treatment of MV-4-11 AML cells with lintuzumab-Ac225 revealed a dose-dependent reduction of leukemia cell viability which was significantly more potent relative to cold lintuzumab. Single agent cytotoxicity of Gilteritinib or Revumenib was also observed, but the combination of lintuzumab-Ac225 with FLT3 or MLL inhibitors was additive and synergistic at all dose levels studied. In a preclinical AML model in vivo, each inhibitor delayed tumor growth as a monotherapy. However, the combination of each of these inhibitors with lintuzumab-Ac225 demonstrated statistically significant enhanced tumor control in the MV-4-11 AML model. Conclusions Lintuzumab-Ac225 has potent anti-leukemic activity in AML cells harboring FLT3 or MLL genetic abnormalities. Combination of CD33-targeted radiation with either FLT3 or menin inhibitors significantly improves AML control, suggesting how FLT3 and/or menin efficacy may be augmented by radiation-induced DNA damage. The combination of lintuzumab-Ac225 to existing mutation-targeted standard-of-care may enhance AML targeting and treatment durability versus single-agent approaches.

Distribution characteristics of circulating B cell subpopulations in patients with chronic kidney disease

This study compared the levels of circulating B cell subpopulations in patients with different stages of chronic kidney disease (CKD), investigated the effects of haemodialysis (HD) on the B cell-related immune spectrum in patients with end-stage renal disease, and evaluated the link between renal function and immune homeostasis. Overall, 197 patients with CKD (158 non-dialysis patients with CKD stages I–V and 39 end-stage patients undergoing maintenance HD) and 77 healthy controls were included. Compared to healthy controls, patients with CKD stages I–II showed no significant differences except for the proportion of transitional B cells; patients with CKD stage V showed a significant decrease in the proportions of transitional B cells and CD5+ B cells and a significant increase in double-negative (DN) B cells. Compared with early-stage patients with CKD, the absolute count of various B cell subpopulations in advanced-stage patients with CKD showed a significant decrease. The distribution of circulating B cell subpopulations in patients with CKD was significantly altered and was associated with CKD progression. Furthermore, the proportion of DN B cells and CD5+ B cells was inconsistent pre- and post-HD. This in-depth study of the immune status of patients with CKD may have important clinical value.

PATH-42. UNCOVERING AN ASSOCIATION BETWEEN OLIGODENDROGLIOMA AND THE HUMAN PAPILLOMAVIRUS (HPV) THROUGH MOLECULAR SUBPOPULATION ANALYSIS USING NETRAAI

Abstract Oligodendrogliomas (OGs) are a rare subset of primary brain tumors, accounting for approximately 5% of all brain tumors. While the etiology of OG has yet to be fully characterized, genomic heterogeneity is linked to clinical heterogeneity with variability in progression and survival. Currently, OGs are stratified into three molecular subtypes based on IDH mutation and 1p/19q co-deletion with additional molecular heterogeneity related to EGFR, PTEN, 10q deletion, and several other markers observed among 1p/19q co-deleted OGs. Using a public OG dataset, we assembled a set consisting of 156 primary OG tumors, 14 primary glioma samples, and 9 normal samples profiled using mRNA expression arrays. Using NetraAI, a novel machine learning platform, we identified three OG patient subpopulations based on survival and differentially expressed genes. Examining low- versus high-grade OG, we identified two subpopulations: (1) 26 low-grade and 21 high-grade OGs characterized by higher RPS6KV1 expression, and (2) 6 low-grade and 61 high-grade OGs characterized by higher LFNG expression. A third subpopulation emerged examining 1p/19p co-deletion, consisting of 60 no co-deletion and 3 co-deletion OGs characterized by higher HDAC1 expression. These driving genes have a common link to human papillomavirus (HPV) infection based on functional genomic analysis. HDAC1 is a histone deacetylase involved in gene expression regulation, LFNG is implicated in the Notch signaling pathway, and RPS6KB1 is a kinase regulating protein synthesis and cell growth. We hypothesize that concomitant HPV infection may negatively impact OG prognosis, warranting further investigation into the precise molecular interactions linking HPV infection to clinical outcomes. Furthermore, exploring the link between HPV infection and vaccination to OG outcomes on a population basis may provide valuable insights. The identification of these subpopulations, not previously described by other groups, demonstrates the utility of the NetraAI approach in uncovering subpopulations with potential implications in developing targeted therapies for OG.

Height, weight, and body mass index in patients with familial dysautonomia.

Children with familial dysautonomia (FD) are smaller and grow more slowly than the general population. It is unknown whether this abnormal growth is due to comorbidities that patients with FD live with, or if it is a direct effect of the disease-causing homozygous ELP-1 mutations. Here, we created growth curves for weight, height, and body mass index (BMI) in male and female children with FD to monitor the nutritional status of patients with FD. We used the New York University (NYU) FD Registry which includes data from 680 individuals with FD who were followed longitudinally since birth. We generated sex-specific FD growth charts for three age ranges (birth to 36 months, 2 to 20 years, and 2 to 40 years) and compared them to the general population. We generated Kaplan-Meier curves to test the hypothesis that FD patients with low BMI had shorter survival than the rest of the cohort. Growth charts generated from 591 individuals with FD show that these patients grow more slowly, reach less height, and gain less weight than the general population. The impact of FD on height was more pronounced in girls than in boys. However, both groups showed markedly low weights, which resulted in low BMI. Low weight, but not height, is already evident at birth. In a subpopulation of FD patients, we found that treatment with growth hormone or spinal fusion surgery helped patients achieve the expected growth characteristic of FD patients, but these treatments did not lead FD patients to achieve the growth pattern of the general population. Contrary to our hypothesis, low BMI had no impact on patient survival. Pediatric patients with FD have lower height, weight, and BMI compared to the general pediatric population, but this does not appear to affect survival. Growth curves specific to the FD population are an important tool to monitor growth and nutritional status in pediatric patients with FD when the general population growth curves are of limited use.

Effectiveness and safety of catheter-directed thrombolysis for pulmonary embolism in key patient subgroups: results of an individual patient level analysis

Abstract Background The effectiveness of catheter-directed thrombolysis (CDT) for acute pulmonary embolism (PE) has been rarely investigated in subgroups of patients at a higher risk of complications. Furthermore, this group of patients has been excluded from prior and ongoing phase 3 trials with CDT. Purpose To assess the efficacy and safety of catheter-directed thrombolysis for pulmonary embolism among specific patient subpopulations. Material and methods We conducted an individual patient-level analysis of patient data from 16 cohort studies and randomized trials of CDT for intermediate- and high-risk acute PE. Studies were identified on the basis of a systematic review of the literature. We focused on early clinical outcomes, including haemodynamic decompensation, intracranial bleeding, major bleeding, and death, as well as on the changes in the mean pulmonary artery pressure (mPAP) before and after CDT. We predefined three subgroups of at-risk patients: (i) older age (i.e. ≥75 years), (ii) active cancer, and (iii) surgery in the prior 4 weeks. Results We studied 1244 patients, of whom 586 (47.1%) were women. The median age was 61 (Q1-Q3; 50-71) years. There were 202 (16.3% of total available) patients aged 75 years or older, 101 (10%) with active cancer, and 108 (14.8%) with prior surgery. A total of 946 (89.3%) patients had intermediate-risk PE, and 113 (10.7%) were classified in the high-risk group. Patients aged ≥75 years more frequently suffered haemodynamic decompensation (7.5% vs 1.6% aged <75 years; p< 0.001), major bleeding (12.7% vs 4.5%; p<0.001), and death (9.2% vs 2.2%; p<0.001); Table 1. In a multivariable logistic regression model, age remained associated with death, major bleeding, and haemodynamic decompensation after adjustment for a key prognostic factor (high-risk PE). The same, except for decompensation, was observed after conditioning for sex, high-risk PE, and cancer status. Patients with active cancer showed a tendency to a higher 30-day case fatality when compared with patients with no active cancer, without reaching statistical significance; Table 1. Recent surgery did not appear to influence the outcome rate in this cohort. No differences were found among the three subgroups regarding the reduction of the mPAP following CDT ; Table 2. Conclusions Our results point to age as a strong predictor of early mortality and decompensation in patients undergoing CDT, even after adjusting for the higher rate of high-risk PE in the elderly. The clinical net benefit of CDT in the elderly when accounting for bleeding and the risk of haemodynamic decompensation remains unclear.Efficacy and safety outcomes of CDTmPAP values before and after CDT

Abstract 15253: Phenotyping Sub-Populations of Heart Failure Patients Based on Clinical and Social Determinants of Health Using Unsupervised Machine Learning Models

Background: Social determinants of health affect health care utilization, access, and outcomes. When designing population health management strategies, phenotyping sub-populations of patients using both clinical and social risk factors is essential. Aims: Define sub-populations of heart failure (HF) patients using a novel machine learning method based on demographics, comorbidities, and social vulnerability. Compare days alive and out of hospital (DAOH) within 1-year post-discharge. Methods: Retrospective cohort study of patients admitted to 48 hospitals in Providence St. Joseph Health with a primary HF diagnosis from 02/01/2016 - 12/31/2022. CDC social vulnerability index (SVI) was linked to patients’ home address. Thirty-four demographic, clinical, and SVI variables were included in the model. Clusters were defined using hierarchical clustering of principal components (HCPC). Top drivers of cluster formation were assessed by v-test . Due to large sample size, 10 random samples were selected and clusters were assigned by vote. Cluster enriched subgroups (CES) were defined as patients having ≥6 cluster assignments and >80% cluster vote. Results: HCPC analysis grouped 61,126 hospitalizations into 3 clusters (Figure). CES-1 patients (n=25,873) were older, wealthier, White, with multiple comorbidities living in more educated communities. CES-2 patients (n=13,450) were younger with combined systolic/diastolic HF and fewer comorbidities living in crowded, urban communities. CES-3 patients (n=3,787) were non-White, Hispanic with higher cardiometabolic risk factors living in urban, crowded, low-income communities. CES-1 patients had fewer DAOH (230) and higher 1-year (22%) mortality. CES-2 patients had similar DAOH (274) to CES-3 (269), despite younger age and fewer comorbidities. Conclusions: In a large heart failure cohort, unique sub-populations based on clinical and social factors have distinct care needs, utilization, and outcomes.

Abstract 18918: Outcomes of Patients With Transcatheter Aortic Valve Replacement (TAVR) and Cardiac Resynchronization Therapy (CRT): A Retrospective Analysis

Introduction: Many patients with TAVR have a pre-existing reduced LVEF <35% and left bundle branch block (LBBB) requiring CRT-D for potential improvement in LV function and secondary prevention of sudden death. Few studies have examined patient outcomes in this specific subpopulation. The aim of our study was to see if this patient population had worse outcomes compared to patients with TAVR without CRT-D. Hypothesis: We hypothesized that patients undergoing CRT would have better outcomes. Methods: Patients aged >18 years with TAVR and CRT versus TAVR alone were identified from the United States National Inpatient Sample (2015-2020). ICD 9 and 10 codes were utilized and multivariate regression analysis was used to estimate the odds ratios of in-hospital mortality, average length of stay (LOS), and hospital charges (TOTHC), after adjusting for age, gender, race, insurance, and hospital demographics (type, size, region, and teaching status). Propensity matching was performed using the Kernel method. Weighted analysis utilizing Stata 17 MP was performed. Results: This study identified approximately 324,563 patients with TAVR, of which 2,613 patients had CRT. The analysis revealed mortality (OR: 2.59, p<0.025, CI: 1.13-5.94), length of stay (LOS) (+5.69 days, p<0.0001, CI: 4.44-6.95), and TOTHC ($165,520, p<0.0001, CI: 134,551-196,489) were increased for patients with TAVR and CRT compared to patients with only TAVR. Secondary analysis is described in Table 1. Conclusions: Our study revealed that patients with TAVR and CRT had worse outcomes including increased mortality, LOS, and TOTHC. Despite CRT, this patient subpopulation had worse outcomes compared to those with only a TAVR. We suspect that these patients either have a poor functional status at baseline or CRT may not have as much benefit for patients with TAVR. Further studies are required to further evaluate this patient population.

Multiple Myeloma B Cells and Pre-Plasma Cells Are Important Reservoirs for Myeloma Relapse Following Plasma Cell-Directed Therapy and Prevent Cure with Standard Therapies

Multiple Myeloma (MM) is a B cell neoplasm characterized by the accumulation of mature plasma cells (PCs) within the bone marrow (BM). Despite treatment advances, MM remains incurable in the vast majority of patients. Disease relapse typically occurs regardless of treatment with proteasome inhibitors (such as bortezomib and carfilzomib), immunomodulatory drugs (such as lenalidomide and pomalidomide), anti-CD38 monoclonal antibodies (such as daratumumab and isatuximab) or myeloablative melphalan and autologous stem cell transplantation, and also occurs following treatment with immunotherapeutics such as bispecific antibodies and/or CAR-T cells (targeting BCMA, FcRH5 or GPRC5D). This failure to achieve reliable cure in MM, despite the routine attainment of deep treatment responses against PCs, indicates the existence of important intra-tumor heterogeneity and the presence of rare drug-resistant MM cells with full tumorigenic capacity. To identify intra-clonal MM cell subpopulations, we examined BM samples from MM patients using a combination of FACS, single cell resolved immunoflourescence-FISH (IF-FISH), whole exome sequencing (WES), custom-capture targeted deep sequencing (CC-Seq) and single cell RNA sequencing (scRNAseq). We sought to characterize the genomic landscape and to track the dynamic interconnectedness of these intra-clonal subpopulations in patients over time. Using sequential FACS-IF-FISH studies of MM BM samples (n=140) we identified MM clone cellular subpopulations within the BM of MM patients that recapitulate the normal maturation stages between non-malignant post germinal centre B cells and mature PCs. These subpopulations include rare MM cells that resemble CD20 +CD38 -CD138 -Bcma -Irf4 - Xbp1s - B cells, CD20 -CD38 -CD138 - Bcma -Irf4 -Xbp1s - pre-plasmablasts and CD38 +CD138 -Bcma -/lowIrf4 -/+Xbp1s + pre-plasma cells, as well as predominating tumor-bulk CD38 +CD138 +Irf4 +Bcma +Xbp1s + plasma cells. Importantly, MM progenitor cells were typically found to possess all of the chromosomal translocations and copy number variations (CNV) present in MM PCs including whole chromosome ploidy changes, translocations such as t(11;14), t(4;14) and t(14;16), and secondary aberrations such as gain(1q), del(1p) and del(17p). To examine if MM progenitor cells also possess the single nucleotide variations (SNV) present within MM PCs, and are thus fully malignant, and to track MM progenitor subpopulations and their inter-relatedness within patients over time, we performed WES (n=60) plus targeted deep CC-Seq (n=210) on purified cell subpopulations from 17 patients, including 5 patients whom we sampled serially over >3 years. A median of 5 cellular subpopulations were isolated from each BM sample, enriched by factors of up to 1000-fold or more, and each subpopulation was then sequenced to a depth of 4,000-20,000x. Significantly, SNVs that were present in tumor-bulk MM PCs were also regularly detected in MM progenitor cells, although the low frequency of the clonal progenitor cells often prevented capture of complete SNV profiles. At the same time, analyses of serial BM samples from repeat donor patients taken pre- and post-treatment demonstrated that relapsing MM PCs typically did not appear to derive linearly from pre-treatment PCs, as the relapsing PCs commonly lacked a number of SNVs that had been present in pre-treatment PCs. Importantly, we instead observed that relapsed MM plasma cells appeared to consistently derive from MM progenitor cell populations, frequently B cells, as emergent relapse-specific SNVs that were exclusively detected in relapsing post-treatment PCs (but in not pre-treatment PCs) could often also be detected within pre-treatment CD138- CD38- progenitor cells isolated from BM samples that had been collected up to 3 years prior to relapse. We conclude that MM B cells and pre-plasma cell progenitors possess all of the clonal genomic aberrations required for full malignant potential. Tracking of these cells in vivo in patients over time by their SNV profile implicates them as the cellular origin of PC relapse and recurrent MM disease. We propose that cure of MM requires eradication of MM progenitor cells alongside plasma cells. We are currently conducting scRNA-seq and CITE-seq studies of primary MM samples to further characterize MM progenitor cells for gene and protein expression in order to define their optimal therapeutic targets.

Post-Hoc Analysis of the Final, Three-Year Follow-up Results of a Phase I/II Study on Tirabrutinib (ONO-4059) in Patients with Relapsed or Refractory Primary Central Nervous System Lymphoma

[Background] Tirabrutinib (TIR; ONO-4059) is a second-generation inhibitor of Bruton's tyrosine kinase with high kinase selectivity. Based on the results of study ONO-4059-02 in patients with relapsed or refractory (r/r) primary central nervous system lymphoma (PCNSL) (Phase I/II study, jRCT: 2080223590), TIR was approved in Japan, Korea, and Taiwan. We have reported the results of 3-year follow-up (median follow-up: 37.1 months) of study ONO-4059-02, in which the overall response rate (ORR), the primary endpoint, was 63.6% and median duration of response was 9.2 months. The 3-year PFS and OS rates were 13.9% and 56.7%, respectively (Asai et al. ASCO 2023, Abstract 7548). Several clinical questions still remained to be addressed, including the proportion of patients who received subsequent treatment after discontinuation of TIR, and their treatment types; the characteristics of patients with higher efficacy of TIR; and the impact of long-term TIR treatment on Karnofsky Performance Status (KPS) and the Quality of Life (QoL). Here, we report the results of post-hoc analyses of the 37-month follow-up data of study ONO-4059-02. [Methods] In study ONO-4059-02, 44 patients with r/r PCNSL, age ≥20, and KPS≥70 were treated with TIR once daily at doses of 320 mg (n=20), 480 mg (n=7), or 480 mg (n=17) under a fasted condition. TIR was administered in 28-day cycles, and treatment was continued until disease progression or clinically unacceptable toxicity was observed. We assessed the number of patients who received subsequent treatment after discontinuation of TIR and their treatment types. ORR was assessed in patient subgroups according to their characteristics. In addition, patient characteristics associated with 12-month PFS rates were identified using univariate/multivariate logistic regression analysis. The time course of KPS and QoL during the TIR treatment was also assessed. QoL was evaluated using the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30, EORTC QLQ-BN20, and EuroQol 5 dimensions. [Results] Of the 44 patients, 6 patients were treated for 3 years. At the time of last visit, 2 patients (4.5%) continued the TIR treatment, 10 (22.7%) discontinued the TIR treatment but had not received subsequent treatment, and 32 (72.7%) received subsequent treatment after discontinuation of TIR. The subsequent treatments included systemic therapy, radiotherapy, hematopoietic stem cell transplantation, and surgery. Major subsequent treatments were MTX-based therapy (n=18) and radiotherapy (n=10) with the 3-year OS rate of 61.1% and 40.0%, respectively. The subgroup analyses of ORR by patient characteristics showed that response to TIR was observed in all patient subgroups; ORR was low in those refractory to the treatment just before TIR (33.3%), those receiving ≥ 4 previous treatments (33.3%), those receiving hematopoietic stem cell transplantation (28.6%), those with high tumor burden (42.9%), and those having mutations in CD79B and MYD88 (40.0%). The univariate/multivariate analyses for the 12-month PFS rate identified that baseline KPS 90-100 was significantly associated with long-term response (odds ratio 5.99, P=0.0289). The median baseline KPS was 80.0 (range: 70~100) and was maintained during the TIR treatment (max: 49 cycles). QoL scores were generally maintained. [Conclusion] The majority of patients were able to receive subsequent treatment, including MTX, after TIR. TIR responses were observed in all patient subpopulations, and the baseline KPS was suggested to be associated with long-term response. KPS and QoL were generally maintained during the long-term observation period.