This Week in Hematology — Aug 5, 2026

Welcome to This Week in Hematology. This week we're covering nine notable papers spanning myeloma therapeutics and supportive care, the myeloproliferative neoplasms, and a cluster of bleeding-and-clotting papers with immediate bedside relevance, plus new work on how we phenotype sickle cell disease. Let's dive in.

We'll start with multiple myeloma, where two papers approach the same disease from opposite ends of the treatment pathway. In Blood Advances, Bashir and colleagues report a phase one-two trial of Evomela, the propylene-glycol-free melphalan formulation, as conditioning for autologous transplant in newly diagnosed myeloma [1]. Sixty patients were randomised using a Bayesian design to either a short infusion of thirty to sixty minutes or a long infusion of eight to nine hours, at either two hundred or two hundred and twenty-five milligrams per square metre. The safety signal was reassuring — no grade four or higher non-haematologic toxicities and no non-relapse mortality at day one hundred. Forty-five percent of patients achieved a measurable-residual-disease-negative complete or stringent complete response at day ninety, and the two infusion schedules performed essentially identically on response, toxicity and progression-free survival. The more interesting finding is pharmacokinetic: a higher melphalan area under the curve was strongly associated with achieving measurable-residual-disease negativity, with a posterior probability of benefit of ninety-nine percent, and importantly without more toxicity — although that exposure advantage did not translate into a progression-free survival benefit. In propensity-matched comparison against standard melphalan two hundred, Evomela was associated with longer progression-free survival, though at a posterior probability of ninety percent this is suggestive rather than definitive. The practical message is that exposure-guided, area-under-the-curve-directed conditioning is feasible and deserves a randomised test, and that the long infusion buys you nothing over the short one. At the other end, a retrospective case-control study in the British Journal of Haematology flags something easy to miss in clinic [3]. Among newly diagnosed myeloma patients treated with daratumumab, ixazomib, lenalidomide and dexamethasone, close to half met criteria for iron deficiency with a ferritin under thirty at one year — compared with none of the age- and sex-matched patients who received bortezomib, lenalidomide and dexamethasone. Every single daratumumab-treated patient had a fall in ferritin from baseline. The numbers are small, fifty-five cases against twenty controls, and the authors are careful to say this cannot establish causality or exclude regimen-related confounding. But anaemia in a myeloma patient on an anti-CD38 antibody is almost reflexively attributed to disease or marrow suppression, and this suggests that checking iron studies is a cheap, potentially high-yield habit.

Turning to the myeloproliferative neoplasms, two papers ask whether we are diagnosing and treating these diseases well enough. In the British Journal of Haematology, the Spanish Myelofibrosis Registry reports on one thousand six hundred and forty-nine patients diagnosed from twenty ten onwards across sixty-four centres [4]. Roughly half received a JAK inhibitor, and median survival was six point six years, with no meaningful difference between primary and secondary myelofibrosis. Compared with a matched general population, relative survival was sixty-eight percent at five years and forty-four percent at ten. Among patients aged seventy or under with intermediate-two or high-risk disease by the International Prognostic Scoring System, only a quarter actually made it to transplant — a striking care-delivery gap. Acute leukaemic transformation accounted for one in five deaths, and in the high-risk group ten-year relative survival was just twelve percent. Critically, survival began diverging from expectation early even in lower-risk categories, which argues against complacency about so-called low-risk myelofibrosis. Complementing this, Leukemia publishes a prospective study of non-invasive biomarkers across one hundred and twenty-eight patients with essential thrombocythemia, prefibrotic myelofibrosis, and overt primary myelofibrosis [9]. Inflammatory mediators including interleukin-6, interleukin-1 receptor antagonist, S100A8 and A9, plus matrix regulators and immune activation markers, differed significantly across the spectrum. A regression model identified overt myelofibrosis very well, with an area under the curve of about zero point nine one, but discriminated prefibrotic myelofibrosis from essential thrombocythemia only reasonably, at about zero point seven six. The authors' own reading is that this modest separation supports a pathophysiological continuum rather than sharp categorical boundaries — so blood-based markers are not yet ready to replace the marrow biopsy for that particular distinction.

Now to bleeding and thrombosis, where three papers each change something you might do next week. The American Journal of Hematology reports an observational cohort of thirty-two patients, median age seventy-five, with acquired gastrointestinal vascular malformations — idiopathic angiodysplasia, chronic liver disease, or von Willebrand factor deficiency — treated on a predefined institutional bevacizumab pathway [2]. This is the patient who keeps coming back for endoscopic cautery, iron infusions and transfusions. Their composite haematologic support score fell from about fifteen red-cell-unit equivalents in the six months before treatment to about four during the first six months of bevacizumab, and to zero in months seven through twelve. Annualised hospital and emergency visits dropped from three per year to one, and median haemoglobin rose by three point one grams per decilitre. The trade-off is the expected anti-VEGF toxicity: proteinuria in a quarter of patients and hypertension in sixteen percent, with nine percent discontinuing. It is uncontrolled and single-institution, so regression to the mean is a real concern, but the magnitude and durability are hard to ignore for refractory cases. In the Journal of Thrombosis and Haemostasis, a single-centre retrospective study addresses a question that comes up on every consult service: can you ever give heparin again after heparin-induced thrombocytopenia [5]? Among thirty-five patients with prior HIT who were re-exposed, only one — under three percent — developed recurrent HIT, after nine days of intravenous unfractionated heparin, and that episode was not associated with thrombosis, bleeding, or death. The median interval from diagnosis to re-exposure was eight years, so this speaks to remote HIT, not to the patient you diagnosed last month. With thirty-five patients the confidence intervals are wide, but it supports the immunobiology: these antibodies are transient, and a remote history need not be an absolute lifelong contraindication when parenteral anticoagulation is genuinely needed. Alongside this, a concise review in the British Journal of Haematology covers hereditary antithrombin deficiency in pregnancy, emphasising individualised low-molecular-weight heparin dosing, selective use of antithrombin concentrate around delivery, and multidisciplinary care for what remains the highest-risk inherited thrombophilia in obstetrics [8].

Finally, two papers on disease characterisation. In the American Journal of Hematology, latent class analysis of seven thousand six hundred and thirty-six Californian patients with sickle cell disease over nearly three decades identified four reproducible sub-phenotypes in both sexes [6]: a vaso-occlusive class, a haemolytic class marked by chronic kidney disease, pulmonary hypertension and stroke, an overlap class, and — accounting for around sixty-two percent of patients — a low-complication class. The first three classes carried mortality hazard ratios of roughly one point three to two point four, and in patients twenty-five and under the haemolytic class had dramatically worse survival, with a hazard ratio near eight. That youth finding is the one to remember: early renal and pulmonary vascular complications identify a group needing aggressive attention. And in JAMA, Patel and Moskowitz provide a clinical review of classic Hodgkin lymphoma [7], reinforcing that anti-PD-1 immunotherapy or brentuximab vedotin now belongs in frontline therapy for unfavourable early-stage and all advanced-stage disease, with cure rates approaching ninety percent — which makes long-term surveillance for second malignancies, cardiopulmonary and thyroid dysfunction a core part of the job.

If you only have time for one paper this week, make it the bevacizumab cohort in the American Journal of Hematology [2]. Transfusion-dependent angiodysplasia in older adults is a problem most haematologists manage badly and repeatedly, and this offers a mechanistically rational option with a defined monitoring plan.

Here are the key takeaways from this week in Hematology. Evomela conditioning is safe at up to two hundred and twenty-five milligrams per square metre, higher drug exposure tracks with deeper responses, and a short infusion is as good as a long one. Check iron studies in myeloma patients on daratumumab-based regimens before blaming anaemia on the disease. Real-world myelofibrosis survival remains poor even with JAK inhibitors, and only a quarter of transplant-eligible higher-risk patients under seventy actually get transplanted — audit your own referral practice. Blood-based inflammatory markers identify overt myelofibrosis well but cannot reliably separate prefibrotic myelofibrosis from essential thrombocythemia. Consider bevacizumab for refractory acquired gastrointestinal vascular malformations, monitoring blood pressure and urine protein. And a remote history of heparin-induced thrombocytopenia, years in the past, may not preclude careful heparin re-exposure when it is truly needed.

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.