Rosai-Dorfman disease (RDD) is a rare non-Langerhans cell histiocytosis with limited data guiding management. This study aimed to characterize the clinical, treatment, and outcome features of patients with RDD. We retrospectively reviewed 65 individuals with histopathologically confirmed RDD from 1994-2025 at a single institution. Demographic, clinical, treatment, and outcome data were analyzed. Overall response (OR) was defined as complete or partial response. Median age at diagnosis was 46 years; 82% had extranodal disease, 63% had localized involvement, and 29% presented with cutaneous disease. Initial treatments included surgery (n=32), observation (n=13), corticosteroids (n=10), systemic therapy (n=9), and radiotherapy (RT; n=1). Thirty-eight percent of observed patients experienced spontaneous complete or partial regression. Initial OR rates were 80% for corticosteroids and 22% for systemic therapy. Initial surgery resulted in complete resection in 72% of patients. The single individual treated with initial RT achieved a partial response. Twenty-six individuals (40%) progressed after initial treatment. Nine patients received RT (20-38 Gy) during their disease course. Among 11 evaluable lesions, best responses included one complete response, six partial responses, three stable disease, and one progression. Four had no post-RT progression. No RT late effects were observed. At median follow-up of 56.2 months, 17% required hospitalization, mostly due to mass effect. Six individuals (9%) died of unrelated causes. In the largest single-center retrospective cohort to date, RDD demonstrated heterogeneous behavior and treatment responses but generally indolent courses. RT, though used in a small patient subset, showed promising local activity with minimal toxicity.
CD19-targeted CAR T-cell therapy (CAR-T) has transformed treatment for elderly patients with relapsed or refractory large B-cell lymphoma (rrLBCL), yet data on its safety and efficacy in octogenarians (aged ≥80 years) often with more comorbidities and frailty remain limited. In this single-institution retrospective study, 24 octogenarians were identified and matched with 48 patients aged 65-79 years receiving standard-of-care CAR-T using 2:1 propensity score matching. Baseline characteristics were similar, except for more use of lisocabtagene maraleucel and tisagenlecleucel (P<0.001) and bridging therapy (P=0.024) in the octogenarian cohort. Median follow-up was 15.8 and 36.7 months for the octogenarian and comparator cohorts, respectively. The octogenarian cohort had a higher rate of grade ≥3 immune effector cell-associated neurotoxicity syndrome (ICANS) (25% vs 21%; age treatment effect (ATE) P=0.027). Among those experiencing grade ≥3 cytokine release syndrome (CRS) or ICANS, hospital stays were significantly longer in the octogenarian cohort (30 days vs. 16 days, ATE P<0.001). CAR T-cell expansion was reduced in octogenarians, with lower peak counts (ATE P=0.038) and expansion over 28 days post-infusion (ATE P<0.01). Overall and complete response rates were comparable between cohorts. Six- and 12-month progression-free survival rates in octogenarians were 57% and 45%, and overall survival rates were 88% and 81%, respectively, with no significant differences from those of the comparator cohort. These findings suggest that selected octogenarians tolerate and benefit from CAR T-cell therapy and underscore the need for strategies to mitigate severe toxicities through close monitoring and supportive care.
In patients with myelodysplastic syndromes (MDS) treated with hypomethylating agents (HMA) and proceeding to allogeneic stem-cell transplantation (alloHSCT), pretransplant risk assessment remains dominated by morphologic remission–based frameworks despite uncertain prognostic value in the transplant setting. We retrospectively analyzed 280 HMA-treated MDS patients undergoing alloHSCT at DFCI and Moffitt between 2013-2021 to evaluate the predictive performance of IWG 2023 response criteria and to develop an alternative framework integrating functional marrow reserve (FMR) with baseline molecular risk. Progression-free survival (PFS) was the primary outcome. Median overall survival (OS) was 52 months, with 5-year PFS and OS of 43% and 48%, respectively. IWG 2023 response categories showed substantial overlap in outcomes, and assessment by composite complete remission status did not discriminate PFS (5-year PFS 46% vs 40%; p=0.16). In contrast, preconditioning hemoglobin and platelet counts were strongly associated with outcomes, whereas marrow blasts and absolute neutrophil count were not. Recursive partitioning identified clinically relevant platelet and hemoglobin thresholds, enabling construction of FMR strata with markedly distinct 5-year PFS (70%, 40%, and 24% for favorable, intermediate and poor risk, respectively; p<0.0001). Baseline IPSS-M independently predicted outcomes (5-year PFS 58%, 45%, and 29%; p=0.003) and further refined prognostication when integrated with FMR. In multivariable analysis, the integrated IPSS-M/FMR model remained independently associated with PFS (HR 3.00–8.31 across risk tiers; p<0.004) and showed superior discrimination compared with IWG 2023 and classical transplant variables (PFS c-index 0.676). These findings support an integrated framework that outperforms remission-based criteria for post-alloHSCT risk prediction in HMA-treated MDS.
Immunomodulatory drugs (IMiDs) are a cornerstone of multiple myeloma treatment and newer cereblon E3 ligase modulating drugs (CELMoDs) are in clinical trials. However, a major barrier to improving patient outcomes is the inevitable development of resistance to these agents. To explore the mechanisms underlying IMiD/CELMoD resistance, and identify novel targets for treatment, human cell line models of acquired IMiD/CELMoD resistance were generated. Quantitative proteome analysis of these models identified common changes in lipid synthesis proteins and glucose labelling experiments confirmed altered lipid flux. Proteomic analysis of paired patient samples, from diagnosis and relapse on IMiD, suggested similar changes in lipid pathways. A genome wide CRISPR screen performed in a human multiple myeloma cell line with acquired CELMoD resistance identified dependencies in the lipid pathway genes Stearoyl-CoA Desaturase (SCD) and Membrane Bound Transcription Factor Peptidase Site 1 (MBTPS1). This work has led to novel insights into lipid pathway changes in the IMiD/CELMoD-resistant state which may represent targetable cancer cell vulnerabilities.
Limited-stage mantle cell lymphoma (MCL) is rare, and optimal management remains undefined due to limited representation in clinical studies. We conducted a large population-based analysis using the National Cancer Database (NCDB) to evaluate overall survival (OS) in patients with Ann Arbor stage I–II MCL diagnosed between 2004 and 2015. Initial treatment was categorized to chemotherapy alone, radiotherapy (RT) alone, or combined chemotherapy plus RT (Chemo+RT), and outcomes were further stratified by stage and era of diagnosis (2004–2009 vs 2010–2015) to account for advances in systemic therapy, such as the adoption of bendamustine-rituximab for frontline therapy and the availability of Bruton tyrosine kinase inhibitors (BTKis) for relapsed disease. Among 2,428 patients included, median OS for the entire cohort was 8.1 years. Chemo+RT was associated with superior OS compared with chemotherapy alone or RT alone (median OS 11.2 vs 7.4 vs 8.0 years, respectively; P<0.001). This survival benefit was observed in both stage I and stage II disease and in both eras. On multivariable analysis, treatment with chemotherapy alone (hazard ratio [HR] 1.50) or RT alone (HR 1.33) were independently associated with inferior OS compared with Chemo+RT. In this largest real-world analysis of limited-stage MCL to date, combined chemoimmunotherapy and RT was associated with a clinically meaningful survival benefit, including in the modern treatment era. These findings suggest that combined-modality therapy may be preferred for appropriately selected patients with limited-stage MCL.
Chronic myelomonocytic leukemia (CMML) is a heterogeneous myeloid neoplasm with a highly variable prognosis. TET2 is mutated in ∼60% of cases, with several studies linking TET2 mutations and mutation burden to favorable outcomes. However, no prior study has quantitatively characterised TET2 variant allele frequency (VAF) as a continuous, non-linear prognostic variable across large multi-institutional cohorts, nor leveraged it to refine contemporary CMML risk-stratification systems. We examined the prognostic role of TET2 VAF in 1,206 CMML patients across a UK discovery cohort (DC, TET2-mutated: 433/645, 67%), an international validation cohort (VC1, TET2-mutated: 141/291, 48%), and a French CD14+ monocyte-sequenced cohort (VC2, TET2-mutated: 176/270, 65%). Segmented regression of TET2 VAF distributions reproducibly identified two inflection points (DC 32%/55%; VC1 33%/54%; VC2 36%/56%), implying consistent subdivision of TET2 mutation burden across CMML of epidemiological (and potentially biological) relevance. Restricted cubic spline Cox regression revealed a non-linear relationship between TET2 VAF and survival, with risk lowest near VAF 39% (the empirical first quartile), whilst TET2 wild-type patients had hazards comparable to high-VAF cases. Capitalising on this continuous relationship, we constructed TET2-VAF-adjusted CPSS-Mol, GFM, and MMM scores by adding a per-integer-VAF adjustment to each patient's original score. Across cohorts and prognostic tools, the adjusted scores improved discrimination (Harrell's C-index gain in 16/18 comparisons, IDI significant in 18/18, p<0.01), and group-level stratification for overall survival and acute myeloid leukemia-free survival. These findings suggest that TET2 VAF represents a continuous prognostic variable in CMML and provide a robust, reproducible framework for refining standard risk-stratification systems.
MEIS2 was identified biochemically as a substrate of cereblon (CRBN), a receptor of the CRL4CRBN E3 ubiquitin ligase required for the anti-myeloma activity of immunomodulatory drugs (IMiD)s and the second generation cereblon E3 ligase modulatory drug (CELMoD). However, the function and clinical relevance of MEIS2 in myeloma is unknown. We discovered that MEIS2 is aberrantly expressed in bone marrow myeloma cells (BMMC)s, and that high MEIS2, CDK4 or CDK6 and low CRBN expression predisposes patients to inferior progression-free survival in IMiD therapy. Inhibition of CDK4/6 (CDK4/6i) enhances the IMiD vulnerability in BMMCs ex vivo that mimics patient’s response to IMiDs. Mechanistically, CDK4/6i both rapidly accelerates the dissociation of MEIS2 from CRBN by IMiD and destabilizes the MEIS2 protein while increasing the CRBN protein in cooperation with IMiD and CELMoD. This enhances ubiquitination-degradation of IKZF3 and IKZF1 by CRL4CRBN to exacerbate the loss of IRF4 and relieve IRF7 for induction of the interferon response, culminating in TRAIL-mediated apoptosis. Moreover, MEIS2 promotes survival of myeloma cells by sustaining BCMA (TNFRSF17) expression and antagonizes repression of BCMA by IMiD and CELMoD. Thus, inhibition of CDK4/6 reverses MEIS2 inhibition of CRL4CRBN in cooperation with IMiD and CELMoD, and mitigates MEIS2-mediated BCMA signaling for survival, suggesting targeting MEIS2 or CELMoD and CDK4/6 as a new strategy to advance immunomodulatory drug therapy for multiple myeloma.
Chronic lymphocytic leukemia (CLL) comprises immunogenetically defined stereotyped subsets of patients with distinct B-cell receptor immunoglobulin (BcR IG) features and clinical trajectories, yet the molecular pathways underlying subset-specific differences remain incompletely characterized. To resolve disease-relevant heterogeneity obscured in bulk analyses, we performed integrated single-cell transcriptomic and immunogenetic profiling of 48,557 malignant and bystander immune cells from 13 treatment-naïve primary patient samples representing poor-prognostic subsets #1 and #2 and the indolent subset #4. Despite inter-patient variability, leukemic cells exhibited pronounced subset-specific transcriptional features, with enrichment of hypoxia-related genes in subset #1, oxidative phosphorylation (OXPHOS), MYC/E2F targets, and mTORC1 signaling in subset #2, and negative enrichment of hypoxia, apoptosis, and reactive oxygen species-related pathways in subset #4. Notably, compared with indolent subset #4, aggressive subsets #1 and #2 harbored increased proportions of metabolically active and recently emigrated/proliferative leukemic cells, characterized by a CXCR4dim CD5bright transcriptional phenotype and concerted enrichment of MYC target genes, mTORC1 signaling, oxidative phosphorylation, and BcR signaling. Immunogenetic analyses revealed dominant malignant clones with evidence of intraclonal immunogenetic diversification in 11/13 cases. T cells were dominated by effector-memory phenotypes exhibiting progressive differentiation toward terminal cytotoxic states, accompanied by exhaustion-associated programs, and expanded T-cell clones largely restricted to terminal states. Ligand–receptor analysis further indicated co-stimulatory and inhibitory signaling between leukemic and immune cells. Together, this study provides an integrated view of the transcriptional and immunogenetic landscape of three major stereotyped CLL subsets, linking BcR IG configuration with proliferative capacity and microenvironmental crosstalk, thereby shaping clonal behavior.
Health-related quality of life (HRQoL) measured by patient-reported outcomes (PROs) is understudied in patients who receive chimeric antigen receptor T-cell (CAR-T) therapies. We hypothesized that CAR-T therapies generate long-term HRQoL benefits, and that PRO trajectories in the immediate post-infusion period may serve as prognostic indicators for treatment response. We measured longitudinal HRQoL among patients with hematologic malignancies receiving CAR-T using the PROMIS Global Health-10 and PROMIS-29 Profile instruments at 13 timepoints over a 2-year period after CAR-T. PROs were analyzed by linear mixed-effect modeling. We enrolled 83 patients, mostly with aggressive lymphomas (49%) and myeloma (43%). Long-term improvements in physical and social role functioning as well as sleep were observed in the entire study cohort at 2 years after CAR-T, regardless of age ≤65 or >65. Women reported significant improvements in fatigue (p = 0.016) and global physical health (p = 0.041) at 2 years despite having worse pre-CAR-T HRQoL than men. Patients who later achieved a response to CAR-T at 3 months had significantly worse fatigue at day 15 following CAR-T compared with non-responders, with no significant differences in other HRQoL domains. At day 30 following CAR-T, fatigue was similar for responders and non-responders. At 2 years after CAR-T, responders had significant improvements in social role functioning (p = 0.041) compared to non-responders. CAR-T confers long-term HRQoL benefits in patients with hematologic malignancies. Patient-reported fatigue in the acute post-CAR-T period may serve as a prognostic “PRO-marker” for treatment response, which warrants further prospective study.
Mosunetuzumab (mosun) is a CD20xCD3 bispecific antibody approved for follicular lymphoma. Despite promising results in clinical trials, there are limited real-world data to assess outcomes with mosun, particularly in patients with poor performance status or significant comorbidities. To address this need, we performed a multicenter retrospective study at nine centers in the Collaborative US Bispecifics Consortium (CUBIC) and identified 119 patients who received mosun for non-Hodgkin lymphoma. Ninety patients (75.6%) had indolent lymphomas, 26% of whom would not have been eligible for the mosun registrational trial. Of these patients, 97% received single-agent mosun. Overall response rate (ORR) was 87% and complete response rate (CRR) was 71%. With median follow-up 17.0 months, median progression-free survival (PFS) and overall survival (OS) were not reached (NR). 18-month PFS and OS estimates were 56.7% and 92.9% respectively. Most cases of CRS or ICANS were low grade, with no differences in survival between trial-eligible and ineligible patients. There was a higher rate of infections in trial-ineligible patients than in trial eligible-patients (45% vs. 21%, p = 0.043). These findings support the use of mosun in the treatment of relapsed/refractory indolent lymphomas, even in trial-ineligible patients.
RARA overexpression defines a molecularly distinct subset of acute myeloid leukemia (AML). Tamibarotene, an oral selective RARα agonist, synergize with azacitidine (AZA) but its addition to Venetoclax (VEN) with AZA has not been assessed. SY-1425-202 was a multicenter, open-label, randomized study evaluating tamibarotene combined to VEN/AZA (TAMI/VEN/AZA) versus VEN/AZA (VEN/AZA) alone in newly diagnosed RARA-positive, unfit AML. Part 1 assessed safety of TAMI/VEN/AZA, Part 2 randomized participants to TAMI/VEN/AZA vs VEN/AZA, and Part 3 explored salvage therapy with TAMI/VEN/AZA after VEN/AZA failure in Part 2. Tamibarotene 6 mg twice daily was administered. Randomization occurred after confirmation of RARA positivity by Cycle1 Day8. The primary endpoint was complete remission (CR) + CR with incomplete hematologic recovery (CRi) in Part 2 and Part 3, and safety in Part 1. In Part 1 (n=10), overall response rate (ORR) was 77.8% (5 CR, 2 CRi). In Part 2 (n=51), CR/CRi was 60.0% for TAMI/VEN/AZA and 69.2% for VEN/AZA. Median CR/CRi duration was 293 vs 253 days. ORR was 80.0% vs 73.1%, respectively. In Part 3, 2/5 patients responded (CR, MLFS). Grade ≥3 treatment-emergent adverse event occurred in 76% of patients. Deaths were attributed mainly to disease progression or known complications of therapy; none were attributed to TAMI. The study met pre-specified futility criteria and was discontinued early. The addition of TAMI to VEN/AZA was tolerable but did not improve efficacy over VEN/AZA. These results did not demonstrate clinical benefit of TAMI in the frontline unfit RARA-positive AML setting receiving VEN/AZA. (NCT04905407)
The MK-4280-003 study evaluated favezelimab plus pembrolizumab in hematologic malignancies. We report results for anti-programmed cell death protein 1 (PD-1)-refractory classic Hodgkin lymphoma (cHL; cohort 2). Participants in the safety lead-in had relapsed or refractory (R/R) cHL, diffuse large B-cell lymphoma, or indolent B-cell lymphoma. Participants in cohort 2 had R/R cHL, had undergone or were ineligible for autologous stem cell transplantation, and had disease progression after ≥2 doses of anti-PD-1 therapy and within 12 weeks of last dose. Primary end points were safety and the recommended phase 2 dose (RP2D) of favezelimab plus pembrolizumab. Objective response rate (ORR) was secondary. Duration of response (DOR), progression-free survival (PFS), and overall survival (OS) were exploratory. In the safety lead-in, 1 of 21 participants experienced a dose-limiting toxicity (grade 4 autoimmune hepatitis). The RP2D was favezelimab 800 mg plus pembrolizumab 200 mg IV every 3 weeks. Cohort 2 included 34 participants. Treatment-related adverse events (AEs) occurred in 28 participants (82%; grade 3 or 4 in 6 participants [18%]; no grade 5 AEs). Immune-mediated AEs and infusion reactions occurred in 17 participants (50%; grade 3 or 4 in 3 participants [9%]; no grade 5 AEs). ORR was 29% (95% confidence interval [CI], 15-48). Median DOR was 21.9 months (range, 0.0+ to 26.1+). Median PFS was 9.7 months (95% CI, 5.1-14.7); 24-month PFS was 21%. Median OS was not reached (NR; 95% CI, 27.9 to NR); 24-month OS was 76%. Favezelimab plus pembrolizumab showed manageable safety and antitumor activity in heavily pretreated anti-PD-1-refractory cHL. This trial was registered at www.clinicaltrials.gov as NCT03598608.
The diagnosis of myelodysplastic neoplasms (MDS) requires examination of the bone marrow for morphologic evidence of dysplasia. We sought to determine whether a self-supervised learning (SSL) artificial intelligence-based image analysis approach can be used to reliably distinguish MDS from its clinically relevant mimics using bone marrow biopsies (BMBx). The whole-slide images (WSIs) of hematoxylin and eosin (H&E)- and reticulin-stained BMBx sections from 243 unique patients (89 MDS, 55 non-MDS cytopenic controls [NMCCs], and 99 negative control [NC] cases) were partitioned into image tiles for analysis. These image tiles were then processed using the Barlow Twins SSL model to identify histomorphologic phenotype clusters (HPCs). Review of the HPCs revealed the clusters enriched in MDS cases captured known histopathologic features of the disease, including hypercellularity (characterized by enrichment in hypercellular image tiles), dysplastic and loosely clustered megakaryocytes, increased immature hematopoietic cells, increased vascularity, fibrosis, and cell streaming patterns. For external validation, 95 MDS BMBx WSIs from an independent institution were analyzed using the trained model. The model demonstrated consistent HPC enrichment patterns, supporting its robustness and generalizability. The trained ensemble model using H&E- and reticulin-stained slides distinguished MDS from NCs with an area under the curve (AUC) of 0.82, and from age-matched NMCCs with an AUC of 0.80. These findings demonstrate the potential of SSL approaches to capture diagnostically relevant morphologic patterns and to improve the reproducibility of MDS diagnosis.
This post hoc pooled analysis of ADMIRAL and COMMODORE trials aimed to determine if response kinetics differ for gilteritinib monotherapy in relapsed/refractory (R/R) FMS-like tyrosine kinase 3 (FLT3)-mutated acute myeloid leukemia (AML). In total, 205 patients achieved composite complete remission (CRc). This analysis only included the 166 patients (81.0%) who achieved CRc before on-study hematopoietic stem cell transplantation (HSCT). Of these, 148 patients (89.2%) achieved CRc within 6 gilteritinib cycles, 60 patients (36.1%) after cycle 2, and 119 patients (71.7%) after cycle 4. We defined early responders (ERs; n = 60) and later responders (LRs; n = 106) as those who achieved CRc within 2 cycles and after cycle 3, respectively. Multivariate logistic regression analysis indicated that female patients and patients with nucleophosmin 1 comutations were more likely to be ERs than LRs. Median CRc duration was 2.7 months (range, 0-66) for ERs and 3.5 months (range, 0-61) for LRs. More LRs (41 of 106 patients [38.7%]) had best response of CR vs ERs (13 of 60 patients [21.7%]). Median overall survival indexed at CRc date was 9 months for ERs and 10 months for LRs. In ERs and LRs, 14 of 60 patients (23.3%) and 20 of 106 patients (18.9%) underwent HSCT respectively. The rate of drug-related grade ≥3 adverse events was 11.4 events per patient-years in ERs and 7.8 events per patient-years in LRs. Patients with R/R FLT3-mutated AML should continue gilteritinib monotherapy, if tolerated, for a minimum of 6 cycles because later responses can occur. These trials were registered at www.clinicaltrials.gov as NCT02421939 (ADMIRAL) and NCT03182244 (COMMODORE).
Abstract: Relapse after allogeneic hematopoietic cell transplantation (alloHCT) remains a challenge for patients with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS), and salvage therapy in this setting is not standardized. Given the efficacy of hypomethylating agents and venetoclax (HMA-VEN) in the frontline and relapsed settings, we evaluated the use of this therapy for post-alloHCT relapse. Eighty-three patients with AML (n = 69) or MDS (n = 14) who relapsed after alloHCT were included in this retrospective analysis. The median time to relapse after alloHCT was 7.0 months. Most patients (55%) had prior VEN exposure, and 28% of these patients had previously experienced treatment failure with a VEN-based regimen. HMA-VEN treatment-emergent toxicities were predominantly neutropenia and thrombocytopenia; however, the 30-day mortality was only 3.6%. There were 3 deaths within 30 days because of infectious complications. HMA-VEN led to a complete response (CR) or CR with incomplete hematological recovery in 48% of the patients, with 58% of these patients achieving minimal residual disease negativity. In multivariable analysis, European LeukemiaNet 2024 genetic risk stratification was a predictor of survival outcome. Additionally, prior VEN failure was not a predictor of overall survival. In conclusion, HMA-VEN provides an efficacious and well-tolerated option for post-alloHCT AML or MDS relapse regardless of their prior therapy and may allow responders to undergo a second alloHCT with curative intent.
