PURPOSE: Approximately 30%-40% of patients with relapsed/refractory (R/R) large B-cell lymphoma (LBCL) infused with CD19-targeted chimeric antigen receptor (CAR) T cells achieve durable responses. Consensus guidelines suggest avoiding bendamustine before apheresis, but specific data in this setting are lacking. We report distinct outcomes after CAR T-cell therapy according to previous bendamustine exposure. METHODS: The study included CAR T-cell recipients from seven European sites. Safety, efficacy, and CAR T-cell expansion kinetics were analyzed according to preapheresis bendamustine exposure. Additional studies on the impact of the washout period and bendamustine dose were performed. Inverse probability treatment weighting (IPTW) and propensity score matching (PSM) analyses were carried out for all efficacy comparisons between bendamustine-exposed and bendamustine-naïve patients. RESULTS: The study included 439 patients with R/R LBCL infused with CD19-targeted commercial CAR T cells, of whom 80 had received bendamustine before apheresis. Exposed patients had significantly lower CD3⁺ cells and platelets at apheresis. These patients had a lower overall response rate (ORR, 53% v 72%; P < .01), a shorter progression-free survival (PFS, 3.1 v 6.2 months; P = .04), and overall survival (OS, 10.3 v 23.5 months; P = .01) in comparison with the bendamustine-naïve group. Following adjustment methods for baseline variables, these differences were mitigated. Focusing on the impact of bendamustine washout before apheresis, those with recent (<9 months) exposure (N = 42) displayed a lower ORR (40% v 72%; P < .01), shorter PFS (1.3 v 6.2 months; P < .01), and OS (4.6 v 23.5 months; P < .01) in comparison with bendamustine-naïve patients. These differences remained significant after IPTW and PSM analysis. Conversely, the cumulative dose of bendamustine before apheresis did not affect CAR-T efficacy outcomes. CONCLUSION: Recent bendamustine exposure before apheresis was associated with negative treatment outcomes after CD19-targeted CAR T-cell therapy and should be therefore avoided in CAR T-cell candidates.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To identify precise 3D radiomics features in CT images that enable computation of stable and biologically meaningful habitats with machine learning for cancer heterogeneity assessment. Materials and Methods This retrospective study included 2436 liver or lung lesions from 605 CT scans (November 2010-December 2021) in 318 patients with cancer (mean age, 64.5 years ± 10.1 [SD]; 185 male patients). 3D radiomics were computed from original and perturbed (simulated retest) images with different combinations of feature computation kernel radius (R) and bin size (B). The lower 95% confidence limit (LCL) of the intraclass correlation coefficient was used to measure repeatability and reproducibility. Precise features were identified by combining repeatability and reproducibility results (LCL ≥ 0.50). Habitats were obtained with Gaussian Mixture Models in original and perturbed data using precise radiomics features and compared with habitats obtained using all features. The Dice similarity coefficient (DSC) was used to assess habitat stability. Biologic correlates of CT habitats were explored in a case study, with a cohort of 13 patients with CT, multiparametric MRI (mpMRI) and tumor biopsies. Results 3D radiomics showed poor repeatability (median LCL[IQR] of 0.442 [0.312-0.516]) and poor reproducibility against R (0.44[0.33-0.526]) but excellent reproducibility against B (0.929[0.853-0.988]). Twentysix radiomics features were precise, differing in lung and liver lesions. Habitats obtained with precise features (DSC, 0.601 (0.494-0.712) and 0.651 (0.52-0.784) for lung and liver lesions, respectively) were more stable than those obtained with all features (DSC, 0.532 (0.424-0.637) and 0.587 (0.465-0.703), respectively; P < .001). In the case study, CT habitats correlated quantitatively and qualitatively, with heterogeneity observed in mpMRI habitats and histology. Conclusion Precise 3D radiomics features were identified on CT that enabled tumor heterogeneity assessment through stable tumor habitat computation. ©RSNA, 2024.

Over 60% of relapsed/refractory (R/R) large B‐cell lymphoma (LBCL) patients who receive chimeric antigen receptor (CAR) T cells will experience disease progression. There is no standard next line of therapy and information in this setting is scarce and heterogeneous. We analyzed 387 R/R LBCL patients who progressed after CAR T cells from July 2018 until March 2022 in Spain and the United Kingdom. Median overall survival (OS) was 5.3 months, with significant differences according to the interval between infusion and progression (6 months [not reached]). After progression, 237 (61%) patients received treatment. Focusing on the first subsequent therapy, overall (complete) response rates were 67% (38%) for polatuzumab–bendamustine–rituximab (POLA), 51% (36%) for bispecific antibodies (BsAb), 45% (35%) for radiotherapy (RT), 33% (26%) for immune checkpoint inhibitors (ICIs), 25% (0%) for lenalidomide (LENA), and 25% (14%) for chemotherapy (CT). In terms of survival, 12‐month progression‐free survival and OS was 36.2% and 51.0% for POLA, 32.0% and 50.1% for BsAb, 30.8% and 37.5% for RT, 29.9% and 27.8% for ICI, 7.3% and 20.8% for LENA, and 6.1% and 18.3% for CT. Thirty‐two (14%) patients received an allogeneic hematopoietic cell transplant with median OS not reached after a median follow‐up of 15.1 months. In conclusion, patients with R/R LBCL who progress within the first 2 months after CAR T‐cell therapy have dismal outcomes. Novel targeted agents, such as polatuzumab and BsAbs, can achieve prolonged survival after CAR T‐cell therapy failure.

Data are scarce on cytomegalovirus (CMV) replication in patients receiving CD19-directed chimeric antigen receptor (CAR) T cell treatment. Here we describe the incidence, severity, and management of CMV infection in patients with aggressive B cell lymphoma treated with CAR T cell therapy. In this retrospective observational study, we analyzed CMV viral load and its clinical impact in patients with aggressive B cell lymphoma receiving CAR T cell therapy between July 2018 and December 2021 at a single center. Patients with a negative baseline CMV IgG or a previous allogeneic stem cell transplantation were excluded. CMV replication was determined in whole blood. Overall, 105 patients met the study's inclusion criteria. Ten patients presented with CMV replication before CAR T cell infusion and were analyzed separately. Forty-two of the remaining 95 patients (44%) had at least 1 positive CMV determination, with a viral load ≥1000 IU/mL in 21 patients (22%). Four patients in the main cohort (N = 95) and 4 patients in the preinfusion replication group (N = 10) achieved a viral load >10,000 IU/mL. Only 7 patients received preemptive antiviral treatment. No CMV end-organ disease was reported. The sole independent risk factor associated with CMV viremia ≥1000 IU/mL was dexamethasone treatment (odds ratio, 8.4; 95% confidence interval, 2.4 to 36.6; P = .002). Based on our findings, we designed an algorithm for CMV management in this setting. CMV replication is relatively frequent in patients with aggressive B cell lymphoma receiving CAR T cell therapy. It is usually self-limited and not associated with end-organ disease. Patients receiving dexamethasone or harboring CMV replication before infusion might benefit from active surveillance and preemptive treatment strategies.

The search for understanding immunotherapy response has sparked interest in diverse areas of oncology, with artificial intelligence (AI) and radiomics emerging as promising tools, capable of gathering large amounts of information to identify suitable patients for treatment. The application of AI in radiology has grown, driven by the hypothesis that radiology images capture tumor phenotypes and thus could provide valuable insights into immunotherapy response likelihood. However, despite the rapid growth of studies, no algorithms in the field have reached clinical implementation, mainly due to the lack of standardized methods, hampering study comparisons and reproducibility across different datasets. In this review, we performed a comprehensive assessment of published data to identify sources of variability in radiomics study design that hinder the comparison of the different model performance and, therefore, clinical implementation. Subsequently, we conducted a use-case meta-analysis using homogenous studies to assess the overall performance of radiomics in estimating programmed death-ligand 1 (PD-L1) expression. Our findings indicate that, despite numerous attempts to predict immunotherapy response, only a limited number of studies share comparable methodologies and report sufficient data about cohorts and methods to be suitable for meta-analysis. Nevertheless, although only a few studies meet these criteria, their promising results underscore the importance of ongoing standardization and benchmarking efforts. This review highlights the importance of uniformity in study design and reporting. Such standardization is crucial to enable meaningful comparisons and demonstrate the validity of biomarkers across diverse populations, facilitating their implementation into the immunotherapy patient selection process.

Mutations in the TP53 gene, particularly multihit alterations, have been associated with unfavorable clinical features and prognosis in patients diagnosed with myelodysplastic syndrome (MDS). Despite this, the role of TP53 gene aberrations in MDS with isolated deletion of chromosome 5 [MDS-del(5q)] remains unclear. This study aimed to assess the impact of TP53 gene mutations and their allelic state in patients with MDS-del(5q). To that end, a comprehensive analysis of TP53 abnormalities, examining both TP53 mutations and allelic imbalances, in 682 patients diagnosed with MDS-del(5q) was conducted. Twenty-four percent of TP53-mutated patients exhibited multihit alterations, whereas the remaining patients displayed monoallelic mutations. TP53-multihit alterations were predictive of an increased risk of leukemic transformation. The impact of monoallelic alterations was dependent on the variant allele frequency (VAF); patients with TP53-monoallelic mutations and VAF <20% exhibited behavior similar to TP53 wild type, and those with TP53-monoallelic mutations and VAF ≥20% presented outcomes equivalent to TP53-multihit patients. This study underscores the importance of considering TP53 allelic state and VAF in the risk stratification and treatment decision-making process for patients with MDS-del(5q).