Acute anemia is a medical emergency that impairs oxygen delivery and is usually treated with red blood cell transfusion. However, some patients cannot receive transfusions due to religious beliefs, creating a clinical challenge. This study aimed to evaluate the effectiveness of erythropoietin-based (EPO) therapy, with the majority of patients receiving concomitant intravenous (IV) iron, in restoring hemoglobin in patients with emergent anemia who decline blood products. We conducted a retrospective observational study of 66 patients at Montefiore Medical Center from 2016-2025 who received EPO at 20,000-40,000 units daily for at least three consecutive days. The primary outcome was hemoglobin recovery, defined as hemoglobin >8 g/dL or a greater than 2 g/dL increase from treatment initiation. Secondary outcomes included time to recovery, peak hemoglobin, mortality, and hospital length of stay. Hemoglobin recovery was achieved in 49 patients (74%), with a median increase of 4.4 g/dL. Recovery rates did not differ by anemia type, EPO route, or EPO dose, while higher baseline hemoglobin demonstrated a trend toward increased likelihood recovery. Median time to recovery was 13 days overall. Of the 17 patients who did not recover, 6 died during hospitalization. Overall mortality was 14%. These findings demonstrate that EPO-based therapy can effectively increase hemoglobin in patients unable to receive transfusions. This approach may be life-saving when transfusion is not an option, but one in four patients may not respond, highlighting the need for careful patient selection and monitoring.

Venetoclax (Ven) in combination with hypomethylating agents (HMA) (azacitidine or decitabine) is the standard of care for elderly or unfit patients with acute myeloid leukemia (AML) and is being explored in high-risk myelodysplastic syndrome (HR-MDS). However, currently approved dosing of HMA/Ven is associated with prolonged cytopenias, without a clear improvement in survival for TP53-mutated myeloid malignancies. In order to reduce hospitalizations during COVID, a once-weekly, metronomic schedule of decitabine (0.2 mg/Kg) and venetoclax (400 mg) was developed for patients with MDS and AML. Based on encouraging results, a phase 2 trial was performed. In the current study, we analyzed response rates and survival for all patients with TP53-mutated disease treated on the metronomic schedule. In total, 40 patients with TP53-mutated MDS and AML (26 in a prospective trial and 14 in the retrospective cohort) were included; 26 had HR-MDS and 14 had AML. The median age was 76.5 years, 70% had complex cytogenetics and 82% had bi-allelic TP53 mutations. The ORR for AML (CR+Cri) was 70% and 57% (CR+mCR) for MDS. With a median follow-up of 12.9 months, the median OS for the entire cohort was 11.3 months (11.6 months for AML, 9.9 months for MDS), and median OS in the 31 patients with bi-allelic mutated TP53 was 10.4 months. Transfusion independence was achieved in 58%. Neutropenic fever occurred in 15%, there were no therapy-related fatalities, and the 100-day mortality was 7.5%. A non-cytotoxic metronomic dosing schedule of decitabine/Ven has a low toxicity profile in TP53-mutated myeloid malignancies

e15160 Background: BAL0891 is a first-in-class dual threonine tyrosine kinase (TTK) and polo-like kinase 1 (PLK1) inhibitor designed to disrupt the spindle assembly checkpoint (SAC), resulting in aberrant mitosis and tumor cell death. We report interim safety, PK/PD, and preliminary antitumor activity from the ongoing first-in-human (FIH) phase I TTK-CS-101 study evaluating BAL0891 as monotherapy (Substudy 1) and in combination with paclitaxel (Substudy 3) in patients with advanced solid tumors. Methods: Patients (pts) with advanced/metastatic solid tumors refractory to or intolerant of standard therapy were enrolled. In this ongoing, multicenter, open-label FIH phase I study, pts received escalating BAL0891 as monotherapy in Substudy 1 Regimen A (RA; D1/D8 Q3W) or Regimen C (RC; D1/D15 Q4W), or in Substudy 3 (BAL0891 D1/D15 plus paclitaxel 60 mg/m² D1/D8/D15 Q4W). The primary objectives were to assess safety/tolerability and to determine MTD/RP2D. Results: As of December 31, 2025, 61 pts received ≥1 dose across 9 dose levels (5–280 mg): Substudy 1 RA n = 42 (24 escalation, 18 backfill), Substudy 1 RC n = 9, and Substudy 3 n = 10. DLTs (neutropenia, febrile neutropenia, anemia) occurred at 160 mg (RA), 240 mg (RA/RC), and 280 mg (RC); no DLTs were reported in Substudy 3 to date. The Substudy 1 RA MTD was 240 mg, and RP2D was 160 mg. Most pts (88%) had ≥2 prior metastatic treatment lines. Grade ≥3 treatment-related adverse events (TRAEs) occurred in 38 pts (62.3%); the most common TRAEs were neutropenia (n = 41), fatigue (G1/2; n = 17), and leukopenia (n = 15). PK in Substudy 1 RA showed predictable exposure with minimal inter-occasion variability. Thirty-nine pts were efficacy-evaluable (≥1 post-baseline assessment). Dose escalation is ongoing, and updated results will be presented. Conclusions: BAL0891 demonstrated a tolerable safety profile and consistent systemic exposure in heavily pretreated pts. These interim data support continued clinical evaluation of BAL0891 as monotherapy and in combination with paclitaxel. Clinical trial information: NCT05768932 . Best overall response and clinical outcomes. Substudy 1 (D1/D8 Q3W) Substudy 1 (D1/D15 Q4W) Substudy 3 (Paclitaxel Combination) N 23 8 8 Median prior treatment lines (range) 3 (1–5) 3 (1–6) 2 (1–5) Best overall response CR 0 1 0 PR 0 1 3 SD 16 4 4 PD 7 3 1 DCR, % (95% CI) 70 (47–87) 63 (25–92) 88 (47–100) CBR, % (95% CI) 52 (31–73) 25 (3–65) 63 (25–92) ORR, % (95% CI) 0.00 (0.00–15) 25 (7–53) 38 (9–76)

Background:Patients with cancer often face significant financial challenges, known as financial toxicity (FT), which is associated with reduced quality of life. Patients with hematologic malignancies (HMs) are especially vulnerable due to intensive and prolonged treatments, frequent hospital visits, and a high risk of complications. While FT affects many in the general population, it is particularly severe among racial and ethnic minorities, especially those below the poverty line. To our knowledge, no studies have specifically examined FT in this vulnerable group in the United States. Objective:This study aimed to evaluate the severity of FT in patients receiving treatment for HMs in a socioeconomically underserved population, explore sociodemographic factors that may predict the severity of FT, and evaluate the subjective experiences of these patients as they relate to FT. Methods:We conducted a prospective, observational, longitudinal study at the Montefiore Cancer Center's outpatient department in the Bronx, New York, from October 1, 2022, to October 30, 2023. Participants included either adult patients newly diagnosed (ND) with HMs or those already diagnosed, undergoing cellular therapy (CT). The severity of FT was assessed using the validated Comprehensive Score for Financial Toxicity-Functional Assessment of Chronic Illness Therapy (COST-FACIT) questionnaire. Additionally, an investigator-designed questionnaire was developed to gather sociodemographic data and evaluate the subjective effects of financial burden on patient care. Patients in both the ND and CT groups were followed for 90 days. Data collection occurred at their initial presentation, as well as on days 30 and 90. Results:Ninety patients participated in the study (ND=52 and CT=38). The median age was 59 (IQR 44-66) years, with 27% (n=24) African American and 55% (n=48) Hispanic. Overall, 75% (n=67) of participants experienced some degree of FT, most with mild FT at baseline (day 0, median COST-FACIT score=19.4). In the CT group, FT worsened significantly over time, with a decline in median COST-FACIT scores from 19.9 at day 0 to 15.5 on day 90 (P=.02). In a multivariable linear regression model, race and ethnicity were a significant predictor of FT burden: identifying as African American or Hispanic was associated with a significantly lower COST-FACIT score (ie, higher FT) compared to non-Hispanic White participants (B=-3.08, P=.04, 95% CI -6.05 to -0.12). Additionally, over half of ND and CT participants reported difficulty affording basic necessities (ND: 28/52, 54%; CT: 23/38, 61%) and concerns regarding transportation access and costs (ND: 26/50, 52%; CT: n=18/38, 47%). Conclusions:FT is prevalent among patients with HMs receiving care in underserved populations, and the burden is significantly higher among African American and Hispanic populations.

6547 Background: TP53 mutations drive poor outcomes in Myelodysplastic Syndrome (MDS) and Acute Myeloid Leukemia (AML), aggressive hematologic malignancies characterized by clonal abnormalities in myeloid hematopoiesis. Prognostic scoring systems incorporate cytogenetic and molecular abnormalities based on data from majority White populations and have limited applicability to minorities. Racial disparities in survival are well-known, with Black patients experiencing worse outcomes. While TP53 mutations are associated with poor outcomes, there is limited data about their implications in minority populations. Objective: To study TP53 mutation status, variant allele frequency (VAF), and co-mutations on outcomes in MDS and AML between White and minority populations in the Bronx, a racially diverse region. Methods: This retrospective cohort study analyzed 84 patients diagnosed with TP53 mutated AML and MDS between 2014 and 2024 at Montefiore Medical Center. Data included race/ethnicity (Black, White, Hispanic), age, age at diagnosis, gender, diagnosis, first-line chemotherapy, and bone marrow blast percentage. Molecular data included TP53 mutation status (bi- vs. mono-allelic), variant allele frequency (VAF), and co-mutations. Co-mutation patterns will be presented at the meeting. Comparisons were made using descriptive statistics, and survival outcomes were analyzed using Cox proportional hazards models adjusted for age and gender. Results: Black patients were diagnosed at a younger age than White or Hispanic patients (mean: 63.7 vs. 72.6 vs. 65.9 years, p = 0.029) and had shorter median overall survival (3.9 vs. 16.4 vs. 10 months, p = 0.014). Black patients had higher mean VAF (48.1 vs. 31.9 vs. 38.2, p = 0.047) and were more likely to have co-mutations than isolated TP53 mutations, though this was not significant (OR 4.21, p = 0.06). Patients with VAF above the median had a threefold increased risk of death (aOR 3.14, p = 0.019), independent of age or gender. Conclusions: Black patients with TP53 mutant MDS and AML present younger and have worse outcomes than White patients, associated with a higher TP53 VAF and more frequent co-mutations. This is to our knowledge the largest dataset of TP53 in minorities with MDS and AML; highlighting the need for personalized prognostic models that incorporate minorities to overcome racial disparities in myeloid malignancies.

The Iceland Screens, Treats, or Prevents Multiple Myeloma (iStopMM) risk stratification model, developed to predict ≥10

7514 Background: MM remains an incurable disease and resistance mechanisms are emerging. ISB 2001, a first-in-class trispecific T cell engager, redirects cytotoxic T cells to BCMA and/or CD38-expressing myeloma cells. By simultaneously targeting two TAA, ISB 2001 enhances avidity binding to tumor cells in vitro, hence potency, while the distal positioning of the CD38 vs CD3 binders minimizes CD38-related off-tumor adverse events. Methods: We report data from the DE portion of a Phase 1 study of ISB 2001, assessing safety, tolerability, efficacy, pharmacokinetics (PK), and pharmacodynamics (PD) in RRMM patients (pts) exposed to immunomodulatory drugs, proteasome inhibitors, and anti-CD38 therapies and refractory or intolerant to established therapies. Prior BCMA-targeted and/or T-cell directed therapies were allowed. ISB 2001 was administered weekly subcutaneously (SC) in 28-day cycles, with initial step-up doses on Days 1 (15µg/kg) and 4 (variable). DE utilized an accelerated titration design (initial 3 cohorts with single-patient dosing) followed by a standard 3+3 design. DLTs were evaluated in the first 28 days. After DE, the study will proceed with Part 2 (dose expansion) to confirm safety and select the recommended Ph2 dose under FDA Project Optimus. Results: As of January 13, 2025, 24 pts were treated with ISB 2001 across 8 dose levels (5–1800 μg/kg) with a median follow-up of 6 months (range: 2–12). DL9 (2700 µg/kg) is last dose level and currently enrolling. Among 24 pts, median age was 66 years; 58% male, 83% white with a median of 6 prior lines of therapy (range: 3-11). All pts were triple-exposed, 17/24 (71%) penta-exposed, and 3/17 penta-refractory (18 %). No DLT, adverse events (AE) leading to treatment discontinuation or deaths occurred. Serious AEs were reported in 8 (33%) patients. Drug related Grade (Gr) 3-4 AEs were seen in 13 (54%) pts. CRS was reported in 17 (70.8%) patients, primarily Grade 1–2, with a median onset of 3 days and a median duration of 2 days. No neurologic AEs or ICANs. The Overall Response Rate (ORR) was 75% across all 8 doses, including stringent CR (sCR) 13%, CR 13%, VGPR 38%, PR 13%. Responses were observed from dose level as low as 50 µg/kg (MRD-neg, sCR) with an ORR of 82% at doses ≥50 µg/kg. Median time to response was 36 days. ISB 2001 showed near dose-proportional PK, a half-life >10 days, and consistent T-cell activation, supporting its mechanism of action. Conclusions: ISB 2001 was well tolerated with manageable CRS, no ICANS and demonstrated robust anti-myeloma activity in heavily pretreated RRMM pts (NCT05862012). Full clinical data, including PK and PD results from the dose-escalation portion of the study, will be presented at the conference. Clinical trial information: NCT05862012 .

Background: Severe anemia is a common medical emergency causing impaired oxygen delivery to vital tissues and organs. Prompt recognition, diagnosis, and management of anemia is vital to restore adequate tissue perfusion and prevent downstream complications. Severe anemia, defined by hemoglobin (Hb) <8.0 g/dL, and life-threatening anemia by Hb <6.5 g/dL, is treated with RBC transfusion. However, some patients cannot receive transfusions due to religious beliefs, most notably Jehovah's Witnesses. This creates a unique clinical challenge and highlights the importance of utilizing other methods to rapidly correct anemia. Previous studies have focused on preoperative optimization in this patient population, but data on emergent management remain limited. Objective: Evaluate clinical outcomes following administration of epoetin alfa (EPO) and intravenous (IV) iron in patients with critical anemia who decline blood transfusion for religious reasons. Methods: We performed a retrospective single-arm observational study using data from Montefiore Medical Center's electronic health records from 2016 to 2025. Patients with severe anemia who both declined blood transfusion for religious reasons and were treated with EPO at a dose of 20,000 to 40,000 for at least three consecutive days were included for analysis. Patients who concurrently received EPO for chronic anemia related to CKD or malignancy or who had other transfusion contraindications, including alloimmunity to blood products or hemolytic anemia, were excluded. The primary outcome of analysis was time to hemoglobin recovery defined as Hb >8 g/dL or an absolute Hb increase >2 g/dL from treatment initiation. Secondary outcomes included peak Hb, time to peak Hb, and in-hospital mortality. For patients who did not recover during the index hospitalization but did show recovery at a follow-up visit, the date of first follow-up appointment showing recovery was assigned as the time of hemoglobin recovery. Results: Of 135 patients screened, 66 met inclusion criteria. Median age was 70 years and the population skewed 64% female. Acute anemia was present in 54.5% of patients, while 45.5% had acute on chronic anemia. Acute blood loss was the cause of anemia in 74% of cases with the remainder attributed to subacute bleeding or chronic inflammatory conditions. Hemoglobin recovery occurred in 74% of the examined population, of which 31 (63%) recovered during the index admission and 18 (37%) were seen to have met recovery on repeat laboratory work performed at a follow-up appointment. Median time to recovery was 13 days overall. The median increase in Hb was 4.4 g/dL and the median peak Hb, defined as within 90 days of treatment, was 9.7 g/dL. For the 31 patients who recovered during admission, the median time to peak hemoglobin was 31 days. Most patients (88%) received concomitant IV iron; there was no statistically significant difference in Hb recovery in the patients for which iron was deferred (p=0.664). The most common reason for iron deferral was hyperferritinemia. Among the 17 non-recovered patients, 35% died during hospitalization. Overall, in-hospital mortality was 14%. One thrombotic event occurred post EPO. Conclusions: This study represents the first evaluation of EPO and IV iron treatment outcomes in an emergent setting for patients who decline blood transfusions. Our data demonstrate that this approach is a feasible alternative to transfusion, as evidenced by a 74% hemoglobin recovery rate with a median increase in Hb of 4.4 g/dL. However, approximately 1 in 4 patients did not recover, and overall mortality remained significant at 14%, underscoring EPO and IV iron therapy as a second-line option after transfusion. Overall, further prospective studies are warranted to clarify safety, optimize dosing and route of administration, and better define the role of IV iron therapy across clinical settings.

e19011 Background: Chimeric antigen receptor (CAR) T-cell therapy for B-cell lymphomas is increasing in use and predictive tools, like CAR-HEMATOTOX, help identify patients at risk for prolonged neutropenia (NP). This tool, validated in two cohorts, lacks confounding by race, but applicability to minority populations is unknown. Here we describe hematologic toxicities and evaluate possible predictors of toxicity in patients who received axicabtagene ciloleucel (AC) in a majority minority population. Methods: We conducted a single-center retrospective analysis of adults ≥ 18 years who received AC at Montefiore Einstein Comprehensive Cancer Center in the Bronx, NY. Descriptive statistics were used to define patient characteristics. Univariate linear regression models were used to evaluate possible predictors of hematologic toxicity. Statistical analyses were performed using R programming. Results: Between June 2018 and December 2024, 81 patients received AC; 23 (28%) identified as Hispanic, 19 (23%) non-Hispanic Black, 27 (33%) non-Hispanic White, four (5%) Asian/Pacific Islander, and eight (10%) were self-identified as other race/ethnicity. 35 patients (43.2%) experienced profound NP (absolute neutrophil count [ANC] < 100 cells per µL) and 73 patients (90.1%) experienced severe NP (ANC < 500 cells per µL). The median duration of severe NP was 8 days (range 0 - 49 days). On univariate analysis lower baseline white blood cell count (WBC) and ANC measured 30 - 40 days prior to AC significantly correlated with duration and severity of NP (p < 0.01). WBC and ANC were highly correlated; WBC was a surrogate for ANC. Lower hemoglobin level on the day of infusion (D0) also significantly correlated with duration of NP (p < 0.05), not with severity. Baseline c-reactive protein (n = 64) and ferritin (n = 52) did not correlate with duration or severity of NP. Age at the time of CART infusion, sex (male or female) and race did not correlate with either outcome, but number of previous lines of treatment correlated with both duration (p < 0.01) and severity (p < 0.05). 32 out of 81 patients (39.5%) were classified as low-risk and 49 (60.5%) as high-risk based on the CAR-HEMATOTOX model. Neither CAR-HEMATOTOX score nor risk class correlated with duration or severity of NP. Each factor was a poor predictor of duration or severity of NP based on the models adjusted r 2 value. The best predictor of severity of NP was the WBC on the D0 (r 2 = 0.30), and of duration of NP was the number of lines of previous treatment (r 2 = 0.11). Conclusions: CAR-HEMATOTOX is a tool to stratify patients likely to have cytopenias, and in this majority minority population it did not correlate with severe or prolonged NP after AC. We plan to expand these results to include patients who received other CAR-Ts, and to describe the prevalence of non-hematologic toxicities. Further studies are needed to predict cytopenias after CAR-T, especially as outpatient CAR-T is explored.

6514 Background: Mutations in the tumor suppressor TP53 gene are common in elderly patients with myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) and confer resistance to conventional chemotherapeutic DNA damaging agents. Venetoclax (Ven) added to the hypomethylating agents (HMA) of Decitabine or Azacitidine is the current standard of care for elderly patients with AML and is frequently used in high-risk MDS (HR-MDS). Currently approved dosing schedules of HMA/Ven rely on cytotoxicity and have not improved outcomes in the TP53 mutant population. The efficacy and tolerability of metronomic weekly dosing of Decitabine and Ven in HR-MDS and AML were previously described (Goldfinger et al, Blood 2024). Mechanistically, metronomic dosing relies on terminal differentiation, rather than cytotoxicity making it an attractive regimen for TP53 mutant MDS/AML. Methods: Patients with histologically confirmed AML or MDS and a TP53 mutation received a once-weekly dose of decitabine 0.2 mg/kg subcutaneously and one dose of Ven 400 mg on days 1, 8, 15 and 22 of a 28-day cycle. Results: Between April 2020 and January 2025, 40 patients with TP53 mutated myeloid malignancies were treated with metronomic weekly low-dose Decitabine/Ven (14 AML, 26 MDS). Twenty-four patients were followed prospectively as part of a clinical trial (NCT05184842), and 16 were treated off-trial and had data collected retrospectively. Median age at diagnosis was 76.5 years, 13 (32%) were from minority backgrounds, 28 (70%) had complex cytogenetics and 31 (82%) had biallelic TP53 mutations (median VAF 36%). All AML patients were ELN-poor risk, 21 MDS patients (82%) were R-IPSS high or very high risk. The median time on therapy was 5.8 months, with 10 (25%) patients still on therapy at time of data cut-off. Four patients in the AML and five in the MDS cohorts were not evaluable (2 withdrew consent, 1 lost to follow-up and 6 did not have a BM biopsy for evaluation). Of the evaluable AML patients, 7 (70%) achieved a complete remission (CR), 3 (30%) did not respond. In the evaluable MDS patients, 9 (43%) achieved a CR and 3 (15%) a marrow CR, 4 (19%) with stable disease, 5 (24%) with no response. Of the 26 patients who were transfusion-dependent at the start of therapy, 15 (58%) became transfusion-independent. For the entire cohort (n=40), the median overall survival (OS) was 11.3 months. For the AML and MDS cohorts, the OS was 11.6 and 9.9 months, respectively. In patients who underwent allogeneic stem cell transplant (n=6), OS was 16 months. Non-heme therapy-related adverse events of ≥ grade 3 was seen in 13 (54%) of patients. Conclusions: In this cohort, of elderly patients with poor risk TP53 mutated MDS and AML the use of a non-cytotoxic dosing schedule of Decitabine and Ven resulted in over half the patients achieving a CR and transfusion independence. The median OS of 11.3 months compares favorably to currently approved cytotoxic dosing of HMA/Ven. Clinical trial information: NCT05184842 .

TPS652 Background: Claudin-6 (CLDN6) is overexpressed in solid tumors including germ cell tumors (GCTs) with minimal expression in healthy tissue. This differential expression profile makes CLDN6 a suitable tumor-associated antigen for a T-cell engaging bispecific antibody (bsAb). XmAb541 is a first-in-class humanized, anti-CLDN6 x anti-CD3 bsAb that directs T-cell cytotoxicity. The XmAb 2+1 format provides avid tumor targeting and selectivity. Importantly, XmAb541 selectively binds CLDN6, with reduced binding to other claudin family proteins (e.g., CLDN9) and non-CLDN6-expressing healthy tissue [1]. This study focuses on subjects with GCT and other CLDN6+ solid tumors who have failed standard therapy and have a high unmet medical need. The primary objective of the trial is to determine an optimal and tolerated dose(s) for further evaluation. Methods: This is a first-in-human, Phase 1, open-label study to evaluate the safety and tolerability of XmAb541 in advanced solid tumors. The study will be conducted in 2 Parts. Part 1, Dose Escalation, will establish a dosing regimen inclusive of a priming dose, step-up dose(s), and the target dose. Part 2 Dose Expansion will further evaluate safety and tolerability and provide an initial evaluation of preliminary antitumor activity for the dose regimen(s) established in Part 1. XmAb541 will be administered intravenously. Key inclusion criteria are age ≥ 18 years (for subjects with GCTs, age ≥15 years). Have histological evidence of locally advanced, recurrent, or metastatic GCT, ovarian, fallopian tube, peritoneal cancer, or adenocarcinoma of the endometrium. Have documented progressive disease on standard-of-care therapies; exhausted therapies with a survival benefit, or the standard therapy has no survival benefit or proven to be ineffective, intolerable, or subject is not a candidate for such available therapy. Subjects must have an Eastern Cooperative Oncology Group performance status of 0-2 and a life expectancy ≥ 3 months with adequate liver, kidney, and bone marrow function. Key exclusion criteria include the following: known active central nervous system metastases and/or carcinomatous meningitis. Subjects with treated brain metastases may participate, provided they are radiologically stable. Ethics approval: This study was approved by WCG IRB. References: 1. Faber, M. Bispecific claudin-6 x CD3 antibodies in a 2+1 format demonstrate selectivity and activity on human ovarian cancer cells. (AACR 2021, Abstract No. 1860). Clinical trial information: NCT06276491 .

Introduction: Peripheral T-cell lymphomas (PTCL) are a group of heterogeneous lymphomas that arise from mature T-cells and comprise about 10% of all non-Hodgkin lymphomas. While diverse clinically and pathologically, these lymphomas often present at advanced stages and have a poor prognosis. Accurate prognostication for these patients is critical to inform clinical decision making and treatment strategies. Traditional tools to estimate overall survival (OS) can fall short when it comes to handling complex clinical data. Established indices such as the International Prognostic Index (IPI), Prognostic Index of PTCL-NOS (PIT), modified PIT, and others typically include factors such as age, disease stage, performance status, and lactate dehydrogenase (LDH), but may not capture the full spectrum of clinical and laboratory data in individual patient cases. In this study we aimed to develop and validate a machine learning (ML) model to predict OS using comprehensive demographic, clinical, and laboratory data from diagnosis. Methods: We identified 97 patients with PTCL diagnosed at Montefiore Medical Center between January 1st, 2010 and December 31st, 2022. Data was collected via manual chart review. Inclusion criteria included a diagnosis of PTCL, age >18, and complete data at diagnosis. Patients with incomplete data were excluded. Data processing and modeling was performed in Python using libraries such as scikit-learn, scikit-survival, and pandas. A random forest survival model was used for model development. Model performance was evaluated using concordance index (C-index) to assess discriminatory ability and Brier scores were calculated at different time points (30, 365, 730, and 1095 days) to assess calibration. Permutation feature importance (PFI) was used to evaluate individual variable contribution, and partial dependence plots were used to interpret model behavior. Results: Among the 97 patients with PTCL, 42% were diagnosed with Adult T-cell Leukemia/Lymphoma (ATLL), 28% with Peripheral T-cell Lymphoma not otherwise specified (PTCL-NOS), 8.2% with Extranodal NK/T-cell Lymphoma, Nasal Type (ENKTCL), 7.2% with Angioimmunoblastic T-cell Lymphoma (AITL), and 13.4% with other PTCL subtypes. The cohort was 54.6% male and 45.4% female. 49.6% were Black, 36.1% were categorized as Other, 9.2% were White, 4.1% were Unknown, and 1% was Asian. 38% were of Hispanic ethnicity. The average age at diagnosis was 60 years. During the study period, 57.7% of patients died and 25.8% were lost to follow-up. The average time from diagnosis to death or censorship was 760 days. Our ML model demonstrated predictive performance in the full dataset with a C-index of 0.86, suggesting strong overall discriminative performance, though on the validation set the C-index decreased to 0.68. Brier scores at 30, 365, 730, and 1095 days were 0.07, 0.21, 0.21, and 0.19, respectively, indicating good calibration over time. The most predictive features for the full data set included baseline LDH, ECOG at diagnosis, Charlson Comorbidity Index score, ejection fraction (EF), absolute neutrophil count (ANC), hemoglobin, and if the patient received a transplant after first- or second-line therapy (PFI of 0.03, 0.02, 0.02, 0.02, 0.02, 0.01, and 0.01 respectively). Creatinine (PFI= 0.009), number of extranodal sites (VI=0.009), Medicaid coverage (PFI= 0.007), and GFR (PFI= 0.007) had the next highest feature importance scores. Baseline LFTs and bilirubin did not significantly contribute, and neither did age >60. Partial dependence plots were suggestive of shorter predicted survival among patients of Hispanic ethnicity and those with Medicaid. Conclusions: Our machine learning model demonstrated predictive capacity for OS in our full cohort. The discrepancy in the C-index in the full cohort and validation sets is likely reflective of the impact of limited sample size and indicates potential overfitting in our model. This emphasizes the need for further model refinement and validation on independent data. Overall, these findings do support the feasibility of machine learning in generating survival prediction models in PTCL but again highlight the importance of external validation and a larger sample size to improve model robustness and possibility for clinical use. Future work will focus on refining the model, incorporating further molecular and genetic data, and validation in multi-institutional cohorts.

Introduction Acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) in elderly patients unfit for intensive chemotherapy have substantial clinical and economic challenges. While hypomethylating agents (HMA) plus venetoclax (Ven) improve outcomes over HMAs alone, the combination at standard-dose (SD) frequently causes severe myelosuppression, leading to toxicities and hospitalizations. Recent studies suggest that a metronomic low-dose (MLD) regimen of decitabine plus Ven yields comparable efficacy with fewer toxicities. This study aims to evaluate economic outcomes and days spent in the hospital of MLD versus SD HMA/Ven using real-world cohorts at Montefiore Einstein Comprehensive Cancer Center (MECCC). Methods At MECCC, we analyzed both prospective and retrospective cohorts of myeloid malignancies (MDS, AML, chronic myelomonocytic leukemia) ineligible for intensive chemotherapy, grouping them into SD HMA/Ven (azacitidine 75 mg/m² for 5–7 days or decitabine 20 mg/m² for 5 days plus Ven 400 mg daily for 14–28 days per 28-day cycle) or MLD HMA/Ven (subcutaneous decitabine 0.2 mg/kg weekly plus Ven 400 mg weekly). Using patient-level data, we constructed a Markov model, with three exclusive health states “on treatment”, “off treatment” and “death”. Transition probabilities between different health states were calculated using the TreeAge Pro software. Overall survival was assessed by Kaplan–Meier analysis and parametric fits for overall survival and event-free survival were modeled with log-normal distributions chosen by Akaike Information Criterion. Model costs (calculated in 2024 USD) were obtained from the Centers for Medicare and Medicaid Services, Physician Fee Schedule and published literature. Utilities for progression-free and progressive disease states were referenced from prior studies. All analyses were calculated from a US third-party payer perspective. To determine cost effectiveness, total costs and quality-adjusted life years (QALY) were estimated and the incremental cost-effectiveness ratio (ICER) was calculated as the difference in cost divided by the difference in QALYs between the MLD and SD arms, with results compared against a willingness-to-pay (WTP) threshold of $150,000 per QALY. Probabilistic (10,000 Monte Carlo simulations) and one-way sensitivity analyses were used to assess robustness of the model. Additionally, longitudinal absolute neutrophil count (ANC) data and hospitalization events were extracted via a HIPAA-compliant tool. Hospitalization rates were modeled with negative binomial regression adjusting for treatment duration, age, Charlson Comorbidity Index and ECOG Performance Status. Results Between November 2018 and January 2024, 68 patients with myeloid malignancies were treated at MECCC with HMA/Ven therapies. In the MLD cohort (n = 31; median age 73 years, range 52–87), there were 21 AML, 8 MDS, and 2 chronic myelomonocytic leukemia cases. The SD cohort (n = 37; median age 69 years, range 30–94) comprised 33 AML and 4 MDS patients. Overall survival was comparable between the two cohorts (p = 0.066). Monthly treatment costs were substantially lower for MLD ($5,269 vs. $19,875). Lifetime healthcare costs were $244,223 for MLD versus $299,038 for SD, with QALYs of 13.17 versus 4.86, yielding an ICER of –$6,598 per QALY, indicating dominance of MLD regimen. All 10,000 Monte Carlo simulations in probabilistic sensitivity analysis remained below the $150,000/QALY WTP threshold. One-way sensitivity analyses identified terminal care cost, supportive care cost, and SD discontinuation probability as the main influential parameters. Longitudinal ANC analysis showed longer follow-up in MLD (median 147.5 vs. 75.5 days; p<0.001), lower neutropenia burden (time with normal ANC: 35.5% vs. 19.4%, p=0.036; time in grade 4 neutropenia: 22.4% vs. 46.5%, p<0.001), and faster ANC recovery (over 50% achieving ANC >1.5×10⁹/L by day 76 vs. never reached in SD). Hospitalization data revealed fewer days hospitalized per 100 treatment days with MLD (4.5 days vs. 8.5 days) and a 39% lower adjusted hospitalization risk (incidence rate ratio 0.61; 95% CI 0.38–0.98; p=0.042). Conclusion MLD HMA/Ven combination appears more cost-effective compared to SD and is associated with less neutropenia, fewer hospitalizations and lower overall costs. These findings support broader evaluation of MLD regimens in prospective, randomized trials and highlight the potential to optimize patient care and reduce healthcare burden.

Supplementary table on representativeness

Introduction: Overall survival (OS) has improved significantly for patients with chronic myeloid leukemia (CML) since the development of BCR/ABL tyrosine kinase inhibitors (TKI). Health care disparities resulting from social determinants of health are key factors in clinical outcomes for patients with cancer. In medically underserved populations, reduced access to primary care results in delays in diagnoses and unmanaged co-morbidities. In addition, lack of medical insurance and financial distress has been shown to be associated with worse outcomes as well as high rates of non-adherence to therapy, a critical factor in the success of TKIs in CML. Located in the Bronx, New York, the Montefiore-Einstein Comprehensive Cancer Center (MECCC) serves the poorest congressional district in the United States with a high prevalence of patients with low socioeconomic status (SES). As part of an ongoing outcomes-based analysis of patients with hematologic malignancies at MECCC, we evaluated patient demographics, initial TKI choice, and clinical results including achievement of National Comprehensive Cancer Network (NCCN) management guidelines for CML and OS in this vulnerable and medically underserved population. Methods: We conducted a retrospective analysis on all patients seen at MECCC for the initiation of treatment for CML in chronic phase. Data obtained by searching the electronic medical record system for the diagnosis code of CML. Results: From 2004-2024, data was available for 104 patients with CML in chronic phase. Sixty-two (60%) were male and 42 (40%) were female with a median age of 53 years (range 15-89). Race/Ethnicity data showed 73% with a minority background, including Hispanic (41%) and non-Hispanic Black (28%). Of the 74 patients with documentation of clinical presentation, 33 (44%) were symptomatic. Seventy-two (69%) patients were treated with imatinib as their initial TKI. Twenty-six (25%) and 6 (6%) received dasatinib and nilotinib, respectively. Three month molecular real-time polymerase chain reaction milestone data using International Scale (IS) was available for 62 patients: 29 (47%) had an IS>10%, 15 (24%) had 1-log reduction (IS 1-10%), 9 (15%) had 2-log reduction (IS 0.1-1%), 6 (10%) had 3-log reduction (0.01-0.1%), 0 had 4-log reduction (0.0032-0.01%), and 3 (5%) had 4.5-log reduction (<0.00325%). Six-month data was available for 62 patients: 16 (26%) had IS>10%, 19 (31%) had 1-log reduction, 14 (23%) had 2-log reduction, 9 (15%) had 3-log reduction, 0 had 4-log reduction, and 3 (5%) had 4.5-log reduction. Twelve-month molecular data for 61 patients showed 9 (15%) had an IS>10%, 12 (20%) had 1-log reduction, 15 (25%) had 2-log reduction, 11 (18%) had 3-log reduction, 3 (5%) had 4-log reduction, and 11 (18%) had 4.5-log reduction. Overall at 6-months, 45 patients (74%) met the NCCN criteria for TKI-sensitive disease with at least a 1-log reduction. At the 12-months, 25 patients (41%) had at least a 3-log reduction, or a major molecular response (MMR), again meeting criteria for TKI-sensitive disease. As per current NCCN guidelines, a total of 20 patients (19%) were eligible for treatment free remission (TFR). Of these, 5 discontinued therapy and 15 elected to continue their TKI. Five-year OS was 91.3%. Of the 9 patients who died, only 1 CML-related death was observed. Conclusions: We present data for a diverse cohort of CML patients located in one of the poorest areas in the United States. A high number of patients were symptomatic at presentation, which may represent the lack of access to primary care in this economically and social marginalized population. Imatinib was the most common TKI prescribed. Less than half of the patients met MMR at 12 months, a milestone that has been shown to predict future loss of response and ability to achieve TFR. This was confirmed by the low number of patients that met criteria for TFR and ultimately the even lower number of patients that successfully attempted TKI discontinuation. Despite all these unfavorable factors, patients showed an OS similar to that of other clinical data reported in non-minority majority populations. These observations suggests that highly effective drugs, such as TKIs in CML, can help overcome the well-known barriers and poor health outcomes of patients in lower SES. Of particular interest is the observation of the overall lack of CML-related mortality despite failure to achieve strict milestones for TKI management.

Background: Recent advancements in multiple myeloma (MM) diagnosis and treatment significantly improved patients' survival and quality of life. Severe Renal Impairment (RI) remains a significant challenge in newly diagnosed MM (NDMM), complicating treatment and affecting outcomes. Prior studies reported improved overall renal recovery in the era of novel therapies. However, there is a lack of evidence documenting the outcomes of patients presenting with severe RI. We present an updated analysis of the characteristics and outcomes of patients with NDMM presenting with severe RI in an ethnically diverse, low socio-economic population. Methods: A single-center retrospective review of patients with NDMM presenting to Montefiore Medical Center (Bronx, New York) was conducted from 2011-2023. Severe RI was defined as an estimated glomerular filtration rate (eGFR) of less than 30 mL/min/1.73 m². MM risk was assessed using the mSMART 3.0 tool. Results: Out of 284 patients with NDMM, 58 (20%) patients presented with severe RI. Black patients (p = 0.018) and older adults (64 years vs 60 years, p value 0.032) were more likely to present with severe RI. Severe RI at diagnosis correlated with high-risk myeloma (p = 0.002), high R-ISS stage (p < 0.001), and a lower likelihood of receiving stem cell transplants (p < 0.001). Among those with severe RI, 23/58 required dialysis of which 6/23 were taken off dialysis at last follow-up after renal recovery. Median eGFR at diagnosis, 6 months and 1 year after diagnosis was 11, 23 and 23.5 ml/min/1.73 m2, respectively. 14 out of 58 patients (39%) achieved renal recovery to eGFR > 30 one year post-diagnosis, with no significant difference in race, sex, or ethnicity. The most common first-line treatment for patients with severe RI was CyBorD (62%). The median time from diagnosis to first treatment (including plasma exchange) was 10 days. By the last follow-up, 35/58 patients (60%) had a stem cell transplant, with a median time from diagnosis to transplant of 275 days. The median overall survival (OS) for patients with severe RI at diagnosis was 49 months, while it was not reached for those without severe RI (p < 0.001). Early death (≤ 2 months after diagnosis) occurred in 2 patients, both with severe RI. Among patients with severe RI, multivariate Cox regression survival analysis showed that achieving free light chain (FLC) below 500 mg/L (HR 0.212, p < 0.001) and receiving a stem cell transplant (HR 0.209, p 0.002) improved outcome. Conclusions: Despite improvements in the life expectancy of patients with MM, severe renal failure remains a challenge, particularly among minorities. Socioeconomic factors need to be explored as a contributor to outcomes in patients with severe RI. Achieving FLC clearance and undergoing stem cell transplant may play a role in survival in patients with NDMM and severe RI.