This Week in Hematology — Aug 12, 2026

Welcome to This Week in Hematology. This week we're covering 10 notable papers spanning thrombosis risk stratification and antibody-mediated clotting, long-term outcomes in lymphoma and myeloma immunotherapy, and the diagnosis and management of under-recognised benign hematology conditions. Let's dive in.

We start with a paper that challenges one of the most entrenched rules in thrombosis practice — that provoked venous thromboembolism carries a low recurrence risk and needs only short-term anticoagulation. In Blood, a prospective cohort study followed 242 patients with symptomatic deep vein thrombosis or pulmonary embolism who had completed at least three months of anticoagulation and then stopped, classifying their provoking factors using the International Society on Thrombosis and Haemostasis definitions of major versus minor transient risk factors [1]. Patients with cancer, major thrombophilia, pregnancy-associated events, or an indication for extended treatment were excluded. The trial was actually stopped early because recurrence in the major-provoking-factor group was unexpectedly high — at two years, roughly one in five of those patients had a recurrent event, compared with about one in twenty after a minor provoking factor. And that minor-risk-factor figure conceals something important. Among the 81 women whose event was hormone-associated, there were no recurrences at all; strip those women out and the remaining minor-risk-factor patients had a two-year recurrence risk of about nine percent — no longer reassuringly low. The practical message is that "provoked" is not a single category. Surgery-provoked events, in particular, may not deserve the automatic short-course reflex, while hormone-associated venous thromboembolism in women, once the hormone is stopped, looks genuinely low risk.

Staying with recurrence prediction, Thrombosis and Haemostasis reports a post hoc analysis of the double-blind randomised PADIS-PE trial, restricted to the 180 patients whose first unprovoked pulmonary embolism was diagnosed on computed tomography pulmonary angiography [9]. The question was whether the initial clot burden — the pulmonary vascular obstruction index — predicts recurrence after anticoagulation stops, and whether you need the laborious Qanadli score to measure it. Over a median 36 months, 42 patients had recurrent pulmonary embolism, and all four scoring approaches, including two simplified semiquantitative methods, performed comparably with areas under the curve around 0.70 to 0.74. What is clinically usable is the anatomical gradient: compared with segmental-artery-only involvement, where 36-month cumulative recurrence was about ten percent, lobar involvement carried roughly a three-fold higher hazard with recurrence around a quarter, and main pulmonary artery involvement a four- to five-fold higher hazard with recurrence approaching half. In other words, the anatomical level of the most proximal clot on the original scan — something already in your radiology report — carries prognostic information equal to a formal index. The authors are appropriately cautious that this is post hoc and needs prospective validation, but read alongside the Blood cohort, both papers point the same direction: recurrence risk after stopping anticoagulation is far more granular than our current binary categories allow.

A third thrombosis paper reframes anti-platelet factor 4 antibodies. In the Journal of Thrombosis and Haemostasis, a Greifswald laboratory cohort studied 474 consecutive sera with a strongly positive anti-platelet factor 4 heparin immunoassay, testing each with both heparin-dependent and heparin-independent functional platelet activation assays [10]. Nearly half — about 48 percent — had platelet-activating antibodies, and importantly, about seven percent showed isolated heparin-independent platelet activation, the pattern seen in vaccine-induced immune thrombocytopenia and thrombosis and its non-vaccine equivalents. Thrombosis prevalence rose stepwise from about a third in patients negative on both functional assays, to 40 percent in those positive only on the heparin-dependent assay, to nearly 48 percent when both were positive. Prevalence also trended upward with higher optical density, though that trend did not reach statistical significance. The take-home for the clinician: if you see thrombosis with thrombocytopenia that precedes any heparin exposure, standard heparin-induced thrombocytopenia assays may miss it, and heparin-independent platelet factor 4 testing should be requested explicitly.

And rounding out the coagulation theme, the American Journal of Hematology proposes a new diagnostic category — monoclonal gammopathy of thrombotic significance — for thrombotic disorders where an M-protein is mechanistically proven to drive clotting rather than merely coexisting with it [7]. By their strict criteria, only monoclonal protein-induced immune thrombocytopenia and thrombosis currently qualifies; thrombotic microangiopathy and antiphospholipid syndrome occurring alongside a paraprotein are provisionally labelled thrombotic syndromes associated with M-proteins, pending causal evidence. It is a framework paper, but a useful one when you next encounter unexplained thrombosis in a patient with a small clone.

Turning to long-term outcomes with immunotherapy and response-adapted chemotherapy. Nature Medicine reports persistence of 4-1BB-costimulated anti-CD19 chimeric antigen receptor T cells up to ten years after infusion in 38 patients with non-Hodgkin lymphoma [5]. Beyond year five, the transgene was still detectable in five of eight long-term responders, and three maintained B cell aplasia, implying ongoing functional activity. In one patient progression-free at ten years, chimeric antigen receptor T cells still made up over one percent of circulating T cells more than nine years out. These persisting cells shifted over time toward a double-negative, effector-memory-like phenotype with heightened aerobic metabolism, and T cell receptor sequencing showed oligoclonal persistence — with the dominant clone, seventy percent of the product at nine years, already detectable at vanishingly low frequency by day fourteen. That last observation is the interesting one biologically: the clones that will still be there a decade later may be determined in the first two weeks. Integration site analysis found no known drivers of malignant expansion, which is reassuring for the secondary malignancy conversation you are having in clinic.

Also in Blood, three-year follow-up of the phase one/two MonumenTAL-1 study of talquetamab, the GPRC5D-targeting bispecific antibody, in relapsed or refractory myeloma [2]. Across cohorts naive to T cell redirection and one with prior T cell redirection, overall response rates were 67 to 74 percent with complete response or better in a third to over forty percent. Median progression-free survival was between roughly seven and a half and eleven months, and median overall survival was 34 months in the weekly cohort, not yet reached with every-other-week dosing, and 28 months after prior T cell redirection. Cytokine release syndrome was near-universal but almost entirely low grade, taste changes affected around three-quarters of patients, and grade three or four infections occurred in about a quarter — which the authors note remains lower than with approved BCMA-directed bispecifics. Ataxia or balance disorders occurred in about five percent, none grade four or five. No deaths were attributed to the drug. The differentiating feature for practice is that infection profile, relevant when sequencing bispecifics in a heavily pre-treated, immunosuppressed population — balanced against the very real quality-of-life burden of taste and weight change.

On the chemotherapy side, the British Journal of Haematology reports long-term follow-up of the Argentine GATLA LH-05 trial, in which 563 patients with stage one through four classic Hodgkin lymphoma received three cycles of ABVD followed by interim positron emission tomography [4]. Those with Deauville one or two stopped treatment entirely; Deauville three or four completed six cycles with involved-field radiotherapy; Deauville five went to salvage. At a median follow-up of nearly eight years, ten-year progression-free survival was 76 percent overall and 84 percent among the interim-negative patients, with ten-year overall survival close to 94 percent. Notably, the interim scan was the strongest predictor of progression-free survival, and disease stage carried no independent prognostic weight once the scan was accounted for. That supports stopping at three cycles in interim-negative patients across all stages — a genuinely chemotherapy- and radiation-sparing strategy in a young, curable population, albeit from a non-randomised trial.

Finally, three papers on conditions that are easy to under-diagnose. Blood's How I Treat series tackles bleeding in women and girls with hemophilia and hemophilia carriers, reminding us that since the 2021 International Society on Thrombosis and Haemostasis reclassification, females with factor eight or nine levels below 0.40 international units per millilitre are women and girls with hemophilia, and that carriers with normal factor levels who bleed excessively are symptomatic carriers deserving care [3]. The article walks through reproductive tract and joint bleeding across the lifespan and the harm caused by diagnostic delay — a reminder to check factor levels in the sisters, mothers, and daughters in your hemophilia families. A companion How I Treat covers adult autoimmune neutropenia, which unlike the paediatric form is usually chronic, occurs predominantly in women and those with other autoimmune disease, and has detectable anti-neutrophil antibodies in only about half of cases [6]. The central message is therapeutic restraint: most patients remain asymptomatic and infection-free even with absolute neutrophil counts persistently under five hundred, so observation and supportive care are generally preferred, with response to T cell-directed second-line therapy hinting at cell-mediated destruction in some patients. And in Blood, the Myeloid Malignancy Variant Curation Expert Panel assembled 339 people with clinically diagnosed germline GATA2 deficiency across sixteen centres in seven countries and compared 73 phenotypes against UK Biobank controls, defining the condition by nearly three thousand two- or three-phenotype combinations conferring odds ratios above one thousand [8]. That gives clinical laboratories a quantitative, standardised phenotypic anchor for classifying GATA2 variants — which matters directly for donor selection and transplant timing.

If you only have time for one paper this week, make it the Blood prospective cohort on recurrence after provoked venous thromboembolism [1]. It directly undercuts a decision you make almost weekly at the three-month anticoagulation review, and it argues that surgery-provoked events and hormone-associated events should no longer be managed as though they were the same thing.

Here are the key takeaways from this week in Hematology. First, "provoked" venous thromboembolism is not one entity — recurrence approached twenty percent at two years after major transient risk factors, while hormone-associated events in women had no recurrences, so treat duration decisions as individualised rather than protocolised. Second, the anatomical level of the original pulmonary embolism on computed tomography predicts recurrence about as well as formal obstruction indices, with proximal clots carrying several-fold higher risk. Third, consider heparin-independent platelet factor 4 antibody testing when thrombocytopenia and thrombosis precede heparin exposure, because standard assays will miss it. Fourth, anti-CD19 chimeric antigen receptor T cells can persist and remain functional a decade after infusion, with no evidence of malignant integration drivers. Fifth, interim positron emission tomography after three cycles of ABVD identifies Hodgkin lymphoma patients across all stages who can safely stop therapy, with ten-year progression-free survival above eighty percent. And finally, check factor levels in female relatives of your hemophilia patients — carriers bleed, and diagnostic delay causes harm.

That's your roundup for This Week in Hematology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

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