Citations

Abstract Chemotherapy regimens for triple-negative breast cancer (TNBC) combine agents without knowing which agents drive response. Consequently, predictors derived from multi-agent regimens cannot be assumed to generalize to individual drugs or other regimens, motivating development of treatment-matched predictors. Here, we used TNBC patient-derived xenografts (PDXs) treated with carboplatin, docetaxel, or the combination to deconvolute drug-specific responses and identify associated molecular features. Combination treatment rarely improved upon the best single agent, with enhanced responses in only 13% of PDXs and antagonism in a comparable fraction. Proteogenomic analyses identified high cytokeratin-5 (KRT5) as a general marker of chemotherapy responsiveness and KRT5 immunohistochemistry discriminated responsive PDXs (AUROC, 0.83). To train treatment-specific predictors, we integrated these data with independent PDX and clinical cohorts with responses assessed after anthracycline-free platinum, taxane, or platinum-taxane therapy, ensuring response corresponded to the modeled treatment. Four feature selection strategies yielded largely nonoverlapping biomarker panels converging on treatment-relevant pathways. On independent test data, RNA-based predictors of complete response (CR) to platinum-based (carboplatin or cisplatin) and taxane-based (docetaxel or paclitaxel) chemotherapy achieved AUROCs of 0.80 and 0.86, respectively. For platinum-taxane regimens (carboplatin plus docetaxel or paclitaxel), proteomic-guided feature selection generated a 10-biomarker, protein- informed RNA predictor of pathologic complete response (pCR) that outperformed RNA-only feature selection and achieved an AUROC of 0.85 in an independent clinical cohort, while retaining practicality as an RNA-based assay. Treatment-matched integration of multi-omic PDX and clinical datasets provides a framework for chemotherapy-specific predictors with clinically relevant performance, supporting biomarker-guided precision selection and treatment optimization for patients with TNBC. Statement of significance Integration of multi-omic data from patient-derived xenografts with treatment-matched clinical cohorts yielded three retrospectively validated predictors of platinum, taxane, and platinum+taxane response, enabling biomarker-guided chemotherapy selection for patients with triple-negative breast cancer.

BACKGROUND Previous epidemiologic studies of autoimmune diseases in the US have included a limited number of diseases or used metaanalyses that rely on different data collection methods and analyses for each disease. METHODS To estimate the prevalence of autoimmune diseases in the US, we used electronic health record data from 6 large medical systems in the US. We developed a software program using common methodology to compute the estimated prevalence of autoimmune diseases alone and in aggregate that can be readily used by other investigators to replicate or modify the analysis over time. RESULTS Our findings indicate that over 15 million people, or 4.6% of the US population, have been diagnosed with at least 1 autoimmune disease from January 1, 2011, to June 1, 2022, and 34% of those are diagnosed with more than 1 autoimmune disease. As expected, females (63% of those with autoimmune disease) were almost twice as likely as males to be diagnosed with an autoimmune disease. We identified the top 20 autoimmune diseases based on prevalence and according to sex and age. CONCLUSION Here, we provide, for what we believe to be the first time, a large-scale prevalence estimate of autoimmune disease in the US by sex and age. FUNDING Autoimmune Registry Inc., the National Heart Lung and Blood Institute, the National Center for Advancing Translational Sciences, the Intramural Research Program of the National Institute of Environmental Health Sciences.

In this case series, we present several examples from participants (2 patients and 1 healthy control) of a 12-week pilot feasibility study to create a digital phenotype of depression (unipolar or bipolar) through active and passive data collection from a smartphone and a wearable device combined with routine clinical care for mood disorders. The selected cases represent real clinical examples that highlight the intrinsic challenges that should be expected when conducting similar studies, including appropriate health data privacy protection, clinical standardization, and interindividual differences in levels of engagement and acceptability of active and passive data collection (ie, self-reported, behavioral, cognitive, and physiological data), particularly with patient-generated data in mobile apps, digital proficiency habituation, and consistent use of wearable devices. In the context of the rapidly growing use of digital technologies in psychiatry, anticipating challenges for the integration of personal mobile devices and smartphone mental health apps as aides to track specific aspects of depressive disorders is critical for a clinically meaningful digital transformation of mood disorders care.

Introduction: Light-chain amyloidosis (ALA) is a rare plasma cell disorder characterized by multiorgan involvement. Patients often present with non-specific symptoms, which can lead to a delay in diagnosis. This is associated with progressive organ involvement and shorter survival. Reducing the time from symptom onset to diagnosis is a critical step in improving outcomes. This study was designed to evaluate the clinical prodrome and challenges associated with the diagnosis of ALA from the patient perspective. Methods: This retrospective study surveyed patients with ALA through the online HealthTree® Cure Hub platform, which helps patients navigate their disease in a holistic way. The survey consisted of 32 questions, including 23 multiple-choice, 7 date-entry, and two free-response questions. Data on diagnosis and demographics were extracted from the electronic health record and self-reported data linked to HealthTree® Cure Hub. Results were reported as numbers and percentages, and comparisons were made by age (≥70 vs. <70 years) and sex. The following statistical methods were used: Fisher's exact test for categorical data, two-sample t-test and F-test for normally distributed continuous data, Wilcoxon rank-sum test for skewed data and Spearman correlation coefficient. Results: We surveyed 248 patients. Of those, a total of 87 patients with a self-reported diagnosis of ALA completed the questionnaire between March 12, 2023, and May 8, 2024. The median age was 66 years; 55% were female, and 91% were white. The most common initial symptoms prompting patients to seek medical evaluation were fatigue/weakness (47%), swelling (38%), shortness of breath (28%), chest pain (23%), and foamy urine (22%). Pain in hands/feet, carpal tunnel syndrome, loss of appetite, hoarseness, macroglossia, diarrhea, periorbital puffiness, nail changes, and erectile dysfunction each accounted for <10% of cases. Signs and symptoms were first detected by patients themselves in 58% of cases, by a physician in 33% and by a family member or friend for the rest. The median number of symptoms per individual was 2; 31% reported only one symptom, 37% 2-3, 18% 4-6, and 14% >6; with chest pain being the most common (39%), followed by swelling (22%), shortness of breath (13%), and foamy urine (13%); one patient reported erectile dysfunction and another reported arrhythmias as their only symptom. There was a correlation between hoarseness and nail changes (ρ=0.71), and carpal tunnel syndrome and orthostatic hypotension (ρ=0.63). Only 16% were aware of ALA as a condition prior to symptom onset. Thirty six percent presented to a general practitioner for initial evaluation; about 50% sought a specialist, including a myeloma/amyloid specialist (21%), and 14% were first evaluated in the emergency department. While most patients (45%) were diagnosed by a hematologist/oncologist, patients reported seeing a median of 3 physicians (range: 1-10) before the diagnosis was established. Among responders, 61% reported having myeloma at the time of testing. After diagnosis, about 50% were referred to a center specialized in ALA, and 72% met with a stem cell transplant specialist. Most patients reported receiving multidisciplinary care, with a median of 3 additional specialists participating in ALA care (range: 0-16). More than half of patients (59%) felt the diagnosis was delayed; among those, 42%, 24%, 24%, and 11% reported a delay of 1-6, 7-12, 13-24, and >24 months, respectively. The proportion of patients reporting >12-month delay was higher among older patients (54% vs. 27%, P=0.02). A quarter of responders felt their treatment was delayed. Half of patients felt the diagnostic delay contributed to a decrease in treatment efficacy and 48% did not feel that all medical professionals were knowledgeable about ALA. Only 38% felt the diagnostic process was simple and easy within their healthcare system. Conclusion: Most patients with AL amyloidosis initially present with non-specific symptoms, with fatigue/weakness being the most common. Over a third were evaluated first by a general practitioner, and about a third reported a diagnostic delay of >12 months. The diagnostic process is associated with high healthcare utilization, even among patients with concurrent multiple myeloma. Increasing disease awareness among patients and medical providers may shorten the time to diagnosis and contribute to improved outcomes.

Introduction: In recent years, multiple myeloma (MM) research has increasingly focused on unraveling the complex interactions between malignant plasma cells and their surrounding bone marrow microenvironment (BMME). To further elucidate these interactions, we developed a comprehensive single-cell atlas of the MM BMME, integrating both coding and noncoding (ncRNA) transcriptional profiles. By correlating these profiles with specific cytogenetic abnormalities in myeloma cells, we aim to uncover how ncRNAs regulate the phenotypic and functional states of myeloma cells and, consequently, shape the cellular landscape of the BMME. A detailed understanding of how the noncoding transcriptome affects the cellular components and immune landscape of the BMME is essential for advancing our knowledge of MM pathogenesis and identifying novel biomarkers and therapeutic targets. Methods: Our atlas was constructed using single-cell RNA sequencing data from 481 samples of CD138neg cells sorted from bone marrow aspirates of MM patients enrolled in the MMRF CoMMpass study. This dataset includes accompanying whole-genome sequencing and survival data, enabling correlative analysis between transcriptional profiles, cytogenetic abnormalities, and patient outcomes. To generate a combined coding and non-coding transcriptome, we developed an expanded human reference genome by systematically merging mRNA and ncRNA transcripts from the LncBook2.0 and GENECODEv42 reference genomes. The sequencing data was aligned to this expanded genome, followed by quality control processing, batch correction, clustering, and supervised analysis. Results: The resulting coding and non-coding atlas comprises over 1.9 million cells, spanning immune, plasma, and stromal compartments. Remarkably, ncRNAs constituted 46% of the associated genes, including 22,102 long ncRNAs (lncRNAs) and 3,112 small ncRNAs. Subclustering analysis of the major immune compartments (B lymphoid, myeloid, and NK and T lymphoid) yielded 70 subclusters, with ncRNAs representing 24-38% of the top 100 differentially expressed genes within each compartment. This underscores the significant role of ncRNAs in distinguishing immune subpopulations. Moreover, examining the association between specific cytogenetic abnormalities and the immune composition of the BMME, 30 subclusters were found to be differentially abundant (P < 0.05) between cytogenetic abnormalities, supporting our hypothesis that myeloma cell genotypes significantly shape the immune microenvironment. For instance, patients with CCND1 amplification displayed an increase CD4+ central memory T cells with a decrease in CD8+ cytotoxic T cells. Similarly, patients with 1q21 gain exhibited an increase in TGFβ-stimulated monocytes, while a decrease in cytotoxic CD8+ T and CD56dim NK cells was observed in patients with 17p13 deletion. Next, examining which ncRNAs are associated with specific cytogenetic abnormalities, we identified 14 ncRNAs significantly differentially expressed across cytogenetic abnormalities. Increased expression of six of these ncRNAs were associated with poor overall survival both independently and in combination (P < 0.05, HR = 3.7). Notably, three of these ncRNAs have documented roles in cancer progression but have not been previously studied in the context of MM. Subsequently stratifying patients based on their enrichment of these outcome-associated ncRNAs revealed even stronger associations with immune composition, including an increase in CD4+ regulatory T cells (P < 0.05). This further supports our hypothesis that myeloma cell genotypes shape the BMME. Lastly, by constructing a gene regulatory network to predict the interactions between coding and noncoding RNAs, we illustrate the regulatory networks of ncRNAs modulating the phenotypes of myeloma and immune cell subpopulations. Conclusion: In summary, we present the first high-resolution transcriptomic atlas of the MM BMME, integrating both coding and noncoding RNAs. This study reveals distinct immune subpopulations associated with specific myeloma cell genotypes and highlights the central role of ncRNAs in modulating the phenotypic and functional states of both malignant and non-malignant cells in the MM BMME. Our findings emphasize the potential of ncRNAs as therapeutic targets and prognostic markers for MM, offering new avenues for research and clinical intervention.

Introduction: Mantle cell lymphoma (MCL) is an uncommon B-cell non-Hodgkin lymphoma (NHL) Prognosis is highly variable due to underlying biological and molecular heterogeneity. The first-line (1L) treatment paradigm for MCL continues to evolve with increasing use of bruton tyrosine kinase inhibitors (BTKi) and decreasing use of autologous stem cell transplant (ASCT). TP53 alterations are associated with inferior outcomes in MCL. The aim of this retrospective study is to determine treatment strategies and clinical outcomes in MCL patients based on TP53 gene mutation status. Methods: Retrospective data were collected from 19 institutions in the United States from 1/1993 to 1/2024. We examined treatment patterns and outcomes among MCL patients >18 years of age with known TP53 mutation status by sequencing. Patients who did not receive treatment were excluded. The analysis was stratified based on presence of TP53 mutations. Systemic therapies were categorized as chemo-immunotherapy (CIT), non-chemotherapy (including monoclonal antibodies, lenalidomide, venetoclax, and novel agents on clinical trial), and regimens containing BTKi. The primary outcome was overall survival (OS) and secondary outcomes were progression free survival (PFS) and time to next treatment (TTNT). OS and PFS were estimated with the Kaplan-Meier method and log-rank tests were used to compare groups. Results: 645 patients were included in the study, with 468 men (72.6%) and median age was 64 years (range 30-89). 230/645 patients (35.7%) had a TP53 mutation, of which 182 (79%) were identified prior to 1L. While the male/female distribution was the same between the mutated and unmutated groups, median age at diagnosis and MIPI risk score was higher in the TP53 mutated group (age 65.9 vs 63.1 yrs, p= 0.015; MIPI 6.4 vs 6.1, p<0.001). TP53 unmutated patients had a median of 1 line of therapy vs. 2 for the mutated group (p<0.001, range 1-8). Blastoid histology and higher ki67 were also associated with the TP53 mutated group (p=0.045, 0.01 respectively) The median OS was 14.2 years for TP53 unmutated and 8.3 years for TP53 mutated patients (p<0.0001). PFS in 1L was 0.95 years in TP53 mutated compared to 2.02 years in TP53 unmutated group (p<0.0001). Selection of 1L therapeutic strategy was not associated with presence of TP53 mutation (Χ2 p= 0.21) although more patients in the TP53 unmutated group underwent ASCT (119/415 vs 38/230; Χ2 p=0.0006) TP53 mutation was associated with inferior survival compared to TP53 unmutated patients for CIT in 1L but the difference did not reach statistical significance when non-chemotherapy regimens and regimens containing BTKi were used. For CIT in 1L, 5 year OS was 78% [71-84%] for unmutated compared to 58% [48-67%] for the mutated group (p=0.0001 ). For non-chemotherapy regimens, 5 year OS for unmutated patients was 73% [52- 86%] compared to 52% [25-73%] for the mutated group (p=0.09). For regimens containing BTKi, 5 year OS for unmutated was 93% [81-97%] compared to 57% [28-78%] for mutated (p=0.07). Within the TP53 mutated group, there were no differences in any clinical outcomes between those receiving CIT, non-chemotherapy, or BTKi in 1L. While median OS appeared numerically longer in the BTKi group (14.4 yrs vs 8.8 yrs for CIT and 6.8 yrs for non-chemotherapy), the difference did not reach statistical significance (p=0.58). In contrast,TP53 unmutated patients who received CIT in 1L had better TTNT (p=0.0039) compared to those in non-chemotherapy and BTKi groups. There was no improvement in PFS (p=0.79) or OS (p=0.47). Conclusions To our knowledge this is the largest retrospective real world dataset of MCL treatment outcomes stratified by TP53 mutation. As in prior studies, TP53 mutation portends a worse prognosis in MCL compared to TP53 unmutated patients, but outcomes are better than historically reported. CIT in 1L was not associated with improved OS in either group. Irrespective of TP53 mutation status, incorporation of BTKi in first line was not associated with improved clinical outcomes. The gap in OS between TP53 mutated and unmutated MCL may be narrowing as treatment paradigms shift away from intensive induction chemotherapy and ASCT, and due to an increasing number of effective therapies for relapsed patients. Clinical trials of novel agents are needed to further improve the OS of TP53 mutated MCL patients.

Background: Patients with acute lymphoblastic leukemia (ALL) receiving L-asparaginase (ASP) based induction therapy have a high risk of venous thromboembolism (VTE). Studies investigating universal thromboprophylaxis strategies have yielded sub-optimal results. Risk prediction models are needed to identify high-risk patients who would be candidates for intensification of thromboprophylaxis, and lower risk patients who are unlikely to benefit. Aims: Derive and externally validate a risk prediction model for VTE in patients with ALL receiving ASP therapy. Methods: We conducted a multicenter, international, cohort study of patients with newly-diagnosed ALL receiving ASP-based induction therapy (≥18yrs). The derivation and external validation cohorts included 306 and 97 patients without VTE prior to ASP therapy, respectively. Patients on therapeutic anticoagulation at index were excluded. Patients were followed from date of first ASP dose for 100 days post ALL diagnosis or until VTE, therapeutic anticoagulation or death. Candidate predictors were documented at ALL diagnosis and included demographics, disease and treatment characteristics, VTE history, comorbidities, thromboprophylaxis, complete blood count, fibrinogen and d-dimer. D-dimer was only analyzed within the 7 derivation cohort centers who routinely tested this at ALL diagnosis with <20% missing values. Primary outcome was VTE at any site (centrally confirmed). Secondary outcomes included major bleeding by ISTH criteria. Cause-specific Cox proportional hazard regression was performed to identify VTE risk factors in the derivation cohort while treating death as a competing event. Variables with p-value <0.1 on univariable analysis were included in a multivariable model and subsequently used in the risk model if p-value <0.05. Lastly, the derived VTE risk prediction model was tested in the external validation cohort for association with time to VTE using cause-specific Cox proportional hazard model. All analyses were stratified by center and use of low molecular weight heparin prophylaxis (at any dose on the date of first ASP dose). Results: The cumulative incidence of VTE in the derivation cohort was 17% at 30 days (95% confidence interval [CI] 13% - 21%) and 22% at 90 days (95% CI 17% - 27%). The multivariable model included hyperlipidemia (hazard ratio [HR] 1.35; 95% CI 0.56 - 3.27), body mass index ≥ 25 kg/m² (HR 1.49; 95% CI 0.75 - 2.95), platelet count ≥ 25 x 10⁹/L (HR 1.42; 95% CI 0.67 - 3.00), increasing hemoglobin per 1 g/dL (HR 1.17; 95% CI 1.03 - 1.32) and d-dimer ≥ 1 µg FEU/mL (HR 2.68; 95% CI 1.08 - 6.66). Hemoglobin as a continuous variable and d-dimer ≥ 1 µg FEU/mL were included in the final model and used to create a risk score which was weighted by estimated log-HR sum of these risk factors. The lowest risk quartile (risk score below 2.1) successfully differentiated between a lower 30-day cumulative VTE incidence (4%, 95% CI 0.72% - 12%) compared with the other quartiles (20%, 95% CI 14% - 27%; HR 4.40, 95% CI 1.54 - 12.5) which were defined as high risk (score ≥2.1). The risk score's negative predictive value (NPV) for VTE at 30 days was 96% and positive predictive value (PPV) was 20%. The validation cohort (n=97 from 3 external cohorts with routine d-dimer measurement) included 16 (16.5%) patients in the low-risk group and 81 (83.5%) in the high-risk group, with a 30-day VTE incidence of 6.3% (95% CI 0% - 17%) and 21% (95% CI 12% - 30%), respectively (HR 6.22; 95% CI 0.85 - 45.8; p = 0.073)). The NPV (93.7%) and PPV (21%) were comparable to the derivation cohort. The 30-day incidence of major bleeding was 0% (95% CI NR - NR) in low-risk patients (HR 1.36; 95% CI 0.27 - 6.77) and 2.0% (95% CI 0.55% - 5.4%) in the high-risk group. Conclusion: Patients with ALL receiving ASP induction therapy have a 17% VTE risk at 30 days. We derived and externally validated a VTE risk prediction model based upon d-dimer and hemoglobin. This model successfully identifies at least 16.5% of patients with ALL who have a low VTE risk after ASP treatment. Using this model, novel prevention strategies which may be associated with cost and risk could now be tailored to patients with higher risk. Prospective studies are needed to validate the model's performance and role in VTE risk management.

Introduction CD19-directed CAR T-cell therapies are an important option for the management of relapsed/refractory (R/R) DLBCL. Studies suggest that ~50% of pts receiving CAR T-cell therapies will relapse within 6 months, with an estimated overall survival (OS) of only 5 months. Additionally, 25-35% of pts do not receive subsequent treatment, indicating a high unmet need. CD19 loss is a potential mechanism of resistance to CAR T-cell therapies; thus, treatments that target alternate B-cell antigens are an attractive option. Here, we report for the first time the primary analysis of the ELM-1 expansion cohort investigating the efficacy and safety of odronextamab, a CD20×CD3 bispecific antibody, in pts with R/R DLBCL progressing after CAR T-cell therapy (NCT02290951). Methods Pts aged ≥18 with DLBCL and disease progression after CAR T-cell therapy received intravenous odronextamab weekly in 21-day (D) cycles during Cycles (C) 1-4, with steroid prophylaxis and step-up dosing in C1 (described previously), followed by 160 mg on D 1, 8, and 15 of C2-4. After C4, odronextamab maintenance treatment continued at 320 mg Q2W until disease progression or unacceptable toxicity. Pts with complete response (CR) for ≥9 months transitioned to Q4W dosing. All pts received infection prophylaxis. The primary analysis was conducted when all pts had completed ≥9 months of follow-up. Primary endpoint was objective response rate (ORR), as assessed by independent central review (ICR) according to the Lugano classification. Key secondary endpoints included duration of response (DOR), progression-free survival (PFS), and OS. Exploratory endpoints included immune biomarker assessment. Results As of January 22, 2024, 60 pts were treated and evaluable for efficacy and safety, with median follow-up of 16.2 months. Prior CAR T-cell therapies were axi-cel (50%), tisa-cel (12%), liso-cel (10%), and not specified (28%). Median time since prior CAR T-cell therapy was 6.5 months (range 1.2-46.2). Median age was 63 years (range 27-82; aged ≥75 years: 10%), and 65% were male. Pts had received a median of 3 prior lines of therapies (range 2-9); 77% of pts were refractory to their last therapy, 77% were double refractory, and 72% were refractory to prior CAR T-cell therapy. 48% of pts relapsed within 90 D of CAR T-cell therapy, and 67% relapsed within 180 D. Of the 20 pts with available tumor biopsies at baseline, 10 were CD19(-) and only 2 were CD20(-). 92% of pts completed C1 and 52% completed ≥4 C of treatment. The median duration of treatment was 12 weeks (range 0.7-154.1). The ORR and CR rate by ICR were 48% and 32%, respectively, and were consistent across pts with high-risk features and across prior CAR T products. ORR in pts relapsing in ≤90 D was 21% (6/29) and in pts relapsing in ≤180 D was 33% (13/40). ORR was observed even in pts with low CD20. Median DOR was 14.8 months (95% CI 2.8-not estimable [NE]) and median duration of CR was not reached (NR) (95% CI 3.3-NE); the probability of maintaining CR for 24 months was 68%. Median PFS was 4.8 months (95% CI 2.6-5.4) and median OS was 10.2 months (95% CI 4.6-15.8). Median PFS and OS in complete responders was NR. T-cell counts increased at Week 14 and demonstrated a functional phenotype (CD69 expression and TEMRA count), which was confirmed by ex vivo stimulation. T-cell expansion was observed regardless of CAR T product or time from CAR T administration. The most common TEAE was CRS (any grade [G]: 48%; G1: 25%; G2: 23%; G≥3: 0%). G≥3 TEAEs occurred in 77% of pts (most common: neutropenia 17%, anemia 13%, and neutrophil and white blood cell count decrease 10% each). Five pts (8%) permanently discontinued odronextamab due to AEs (COVID-19, pneumonia, encephalopathy, epilepsy, and gait disturbance [n=1 each]). There was one treatment-related death (COVID-19 pneumonia). No cases of ICANS were reported. G≥3 infections occurred in 12 pts (20%); 2 were due to COVID-19. No cases of tumor flare or TLS were reported. Conclusions ELM-1 is the only study to prospectively evaluate efficacy and safety of a CD20×CD3 bispecific antibody in pts with R/R DLBCL progressing after CAR T-cell therapy. Odronextamab monotherapy was associated with encouraging efficacy and generally manageable safety in a setting with limited treatment options and historically dismal outcomes. Efficacy was observed even in pts relapsing in ≤90 D, supporting the potential of odronextamab as an accessible, off-the-shelf option in the post-CAR T setting.

Background: ctDNA clearance after frontline DLBCL tx has shown strong prognostic value for favorable outcomes, with the potential to improve MRD-driven therapeutic strategies and establish MRD as a surrogate endpoint in future studies (Roschewski M, et al. Blood 2022; Goldstein J, et al. Blood 2023). However, further study is needed to evaluate the association of this noninvasive approach with efficacy outcomes after 2L LBCL tx. We previously reported ctDNA analysis after liso-cel infusion in TRANSFORM (NCT03575351; Stepan L, et al. Blood 2023;142[suppl 1]). Here, we describe the predictive value of pre-tx and on-tx ctDNA levels for durable clinical benefit (CR and event-free survival [EFS]) in both liso-cel and SOC arms from TRANSFORM (Kamdar M, et al. J Clin Oncol 2024), evaluating ctDNA as an earlier surrogate for conventional clinical outcomes. Methods: Baseline (prerandomization) and longitudinal ctDNA levels were assessed in randomized pts treated with liso-cel or SOC in TRANSFORM using the phased variant enrichment and detection sequencing assay (PhasED-Seq; Foresight Diagnostics). Phased variants were identified from baseline plasma samples for longitudinal ctDNA monitoring at multiple on-tx time points, as previously described (Stepan L, et al. Blood 2023;142[suppl 1]). Randomization was defined as study Day 1. Association of ctDNA levels with response per independent review committee using Lugano 2014 criteria and EFS was investigated at various predefined time points, including Day 43 (after 2 cycles of salvage immunochemotherapy for SOC; Day 15 after liso-cel infusion), Day 64 (after 3 cycles of salvage immunochemotherapy for SOC; 1 month after liso-cel infusion), and Day 126 (within 2 months after ASCT; 3 months after liso-cel infusion). To assess the surrogacy value of ctDNA, ctDNA-MRD dynamics and ctDNA clearance were investigated within both arms. Results: In pooled analyses combining the liso-cel and SOC arms, higher baseline ctDNA levels (above the median) were associated with shorter EFS (P = 0.05). Pts achieving Day 126 CR had substantial reductions in ctDNA burden vs baseline, while pts with PD or stable disease had persistently high ctDNA levels vs baseline. Achieving undetectable ctDNA at different time points was associated with significantly longer EFS in pooled analyses, observed as early as study visit Day 43, with the strongest association at Day 126. HR (95% CI) for inferior EFS was 3.0 (1.6-5.7) at Day 43, 3.8 (2.1-7.0) at Day 64, and 4.2 (2.3-7.7) at Day 126 in pts with detectable vs undetectable ctDNA. In the analysis by tx arm, consistent with the pooled analyses, ctDNA clearance was associated with EFS benefit at all measured time points with either liso-cel or SOC. Significantly more pts achieved undetectable ctDNA with liso-cel vs SOC during the study (39/63 [62%] vs 25/65 [38%], respectively; P = 0.013), consistent with superior EFS (primary endpoint) with liso-cel in TRANSFORM. All pts in the SOC arm with evaluable ctDNA at Day 126 (n = 28) had received high-dose chemotherapy (HDCT)/ASCT. However, pts with undetectable ctDNA in the liso-cel vs SOC arm had longer EFS at all predefined time points and had statistically longer EFS at Day 126 (SOC vs liso-cel: HR, 3.9 [95% CI, 1.4-10.6]), indicating a deeper and more durable response with liso-cel vs SOC. In pts with CR, ctDNA showed significant predictive value beyond the response assessment at all time points in the liso-cel arm with a HR (95% CI) of 3.8 (1.3-11.0) at Day 64 and 7.0 (2.0-24.3) at Day 126 for EFS in pts with detectable ctDNA vs undetectable ctDNA. Among pts in the SOC arm who experienced subsequent PD despite achieving CR (all received HDCT/ASCT) and undetectable ctDNA at Day 126, serial ctDNA assessment showed consistent reversion to detectable ctDNA at later time points. Conclusions: ctDNA data from TRANSFORM confirm the value of ctDNA as a biomarker for disease burden monitoring and early prediction of durable clinical benefit after 2L LBCL treatment. Furthermore, longitudinal ctDNA molecular response results agree with previously reported clinical efficacy data and confirm the superiority of liso-cel with deeper responses over the historical SOC in 2L LBCL. ctDNA may have a key role as a surrogate endpoint in future LBCL studies, including after CAR T cell therapies.