This week in Hematology

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Hematology. This week we're covering 10 notable papers spanning T-cell redirection and targeted therapy in B-cell lymphomas, new diagnostic and treatment thinking in bone marrow failure, and efforts to standardize risk assessment in leukemia and transplantation. Let's dive in.

We start with B-cell lymphoma, where the question is shifting from whether T-cell engagers and cellular therapies work to how far they can be pushed and at what cost. In Blood Advances, Dickinson and colleagues report a phase 1 dose-escalation study pairing glofitamab, the CD20 by CD3 bispecific, with a novel CD19-CD28 costimulatory bispecific designed to deliver the second signal that T cells need to stay active [1]. Fifty-two heavily pretreated patients received obinutuzumab pretreatment, glofitamab step-up dosing, and then the costimulatory antibody. No maximum tolerated dose was reached. Cytokine release syndrome affected just over half of patients, mostly low grade and clustered during glofitamab step-up, with grade 4 events in about four percent. There were five deaths on study, all judged unrelated to treatment. In aggressive lymphoma, close to two thirds of patients responded and about four in ten achieved complete response, while in indolent disease roughly nine in ten patients achieved complete response. Pharmacodynamic work confirmed enhanced and sustained T-cell proliferation. This is early, uncontrolled, dose-finding data, so it is a proof of concept for CD28 costimulation rather than evidence of superiority over glofitamab alone. Staying with engineered T-cell approaches, Shadman and colleagues in Blood describe CD20-directed autologous CAR T-cell therapy in heavily pretreated Waldenström macroglobulinemia refractory to Bruton tyrosine kinase inhibitors, a disease with no approved CAR T product [2]. Eighty percent of patients reached a complete response or very good partial response. The summary we have is brief and the cohort small, and the authors themselves frame this as a signal warranting larger trials rather than a new standard.

As these therapies spread, infection is becoming the dominant long-term toxicity, and a Blood Cancer Journal analysis by Michael and colleagues tackles it head to head [3]. Using the TriNetX network, they propensity-matched about eleven hundred adults treated with bispecific antibodies against the same number treated with CAR T cells; most patients had multiple myeloma. Overall one-year infection rates were essentially the same, a little over half of patients in each group. But the timing differed. Infections tended to be slightly lower early with bispecifics, though not significantly, and then ran roughly a quarter higher between one and six months and about a fifth higher in the second half of the year. Late fungal infections roughly doubled with bispecifics, and respiratory infections were more common throughout. Read alongside the combination data, this matters: continuous T-cell engagement means continuous immune pressure, and adding costimulation may not be neutral. The authors acknowledge that coded database outcomes and residual confounding from disease severity and treatment duration limit firm conclusions, but the analysis strengthens the case for extended infection surveillance with bispecifics. Rounding out this theme, Wang and colleagues in the American Journal of Hematology present the final BRUIN analysis of pirtobrutinib, the non-covalent Bruton tyrosine kinase inhibitor, in relapsed mantle cell lymphoma [4]. Among 152 patients previously exposed to a covalent inhibitor, about half responded, with a median response duration close to two years, but median progression-free survival was only about five and a half months, a reminder that benefit is concentrated in responders. In a small group of fourteen inhibitor-naive patients, roughly six in seven responded, and median progression-free survival approached four years. Grade 3 or higher bleeding and atrial arrhythmias were uncommon. The naive cohort is too small to anchor practice, but the data confirm pirtobrutinib as a meaningful option after covalent inhibitor failure.

Our second theme is bone marrow failure, where both diagnosis and salvage treatment remain difficult. In Blood, Fox and colleagues tested upfront whole genome and transcriptome sequencing on non-hematopoietic DNA in 237 patients with suspected inherited marrow failure, spanning infancy to age 76 [5]. A molecular diagnosis was made in about 37 percent of patients, covering telomere biology disorders, Diamond-Blackfan anemia, Fanconi anemia, hereditary thrombocytopenia and others, with results returned in real time to guide management and family testing. The approach also uncovered novel mechanisms such as retrotransposon-mediated gene disruption and new gene associations including THRA, and acceptability among patients and carers was high. The authors argue this yield compares favourably with panels and exome sequencing; this is a single programme without a head-to-head comparison, but the case for genome-first testing in this population is getting stronger, particularly given how often inherited disease masquerades as immune aplastic anemia. On the acquired side, Unlu and colleagues in Leukemia retrospectively examined salvage immunosuppression in 151 patients with relapsed or refractory aplastic anemia, pure red cell aplasia, large granular lymphocytic leukemia and related conditions [6]. Sirolimus, tofacitinib, and subcutaneous abatacept and alemtuzumab produced meaningful salvage responses, some lineage-restricted and some durable. The abstract gives no pooled figures, and this is retrospective, heterogeneous data, so the authors frame it as a rationale for prospective trials rather than a treatment algorithm.

Our third theme covers rare myeloid neoplasms, where targeted agents are reshaping outcomes. Hermine and colleagues in Blood report an international registry of 201 patients with systemic mastocytosis with an associated hematologic neoplasm [7]. Exposure to KIT D816V-selective tyrosine kinase inhibitors was associated with median overall survival of about 38 months versus about 18 months, roughly double, and a three-year rate of transformation to acute myeloid leukemia of about 6 percent versus 27 percent. This is observational, so selection effects cannot be excluded, but a fourfold-plus difference in leukemic transformation is a striking signal that the associated myeloid component may be modifiable. Also in Blood, Beeler and Uy review pivekimab sunirine, an anti-CD123 antibody-drug conjugate with a DNA-alkylating payload, now approved by the United States Food and Drug Administration for adults with blastic plasmacytoid dendritic cell neoplasm based on the CADENZA study [8]. The review places it alongside other CD123-directed options and discusses its potential in acute myeloid leukemia.

Our final theme is about standardizing how we measure risk. In Blood, Kotrova and colleagues present an international, cross-age consensus on measurable residual disease in acute lymphoblastic leukemia, built from a practice survey and a multi-round Delphi process involving pediatric and adult groups and reference laboratories [9]. Consensus was reached on 38 of 41 statements. Bone marrow is endorsed as the reference specimen, residual disease testing takes priority over morphology when the two disagree, a standardized assay nomenclature is adopted, and principles are set for late monitoring, confirming reappearance, and defining relapse. The unresolved areas, including isolated BCR::ABL1 reconversion and whether residual disease alone can define remission, are as informative as the agreements. In Haematologica, Lin and colleagues report a single-centre phase 2 study of geriatric assessment-guided allogeneic transplantation in 32 patients aged 60 and over, median age about 69 [10]. Robust patients received myeloablative conditioning, vulnerable patients reduced intensity, and a geriatrics team followed everyone longitudinally. One-year non-relapse mortality was about 6 percent, against a historical benchmark of 25 percent, and two-year overall survival was about 72 percent, with functional outcomes stable or improved. The comparison is historical and the sample small, so this supports the approach as promising rather than proven.

If you only have time for one paper this week, make it the international measurable residual disease consensus in acute lymphoblastic leukemia [9]. It gives pediatric and adult hematologists a shared language for remission and relapse for the first time, and it clearly marks the questions that remain open.

That's your roundup for This Week in Hematology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.