Deauville 5a held up in two German Hodgkin trials: without new lesions it still marked a small, high-risk group at interim PET

The edition in brief

Today's radiology edition closes on a validation of the refined Deauville score in German Hodgkin Study Group trials HD18 and HD21 (Blood Advances, 22 September). At interim PET after two cycles in advanced classic Hodgkin lymphoma, DS5a — uptake over twice liver without new lesions — occurred in only 4–6% of patients but carried worse progression-free survival than DS1–4 across every cohort (HR range 2.57 to 5.47). Reports should now say whether a DS5 is 5a or 5b. The lead is a post hoc BRIDGE-TNK analysis (Neurology, 20 August) of 550 patients with large-vessel occlusion: tenecteplase before thrombectomy was associated with more functional independence when ASPECTS was below 8 (aRR 1.67) but not at 8–10, where mortality was numerically higher. It is exploratory and needs confirmation, but it makes the ASPECTS in the report more consequential. An AJR reader study (9 September) found LLM help using radiologists' own text descriptions raised residents' accuracy by 16 points and consultants' by 9, but residents switched to a wrong answer after misleading output 61% of the time. A Radiology 'How I Do It' (1 September) sets out the MRI approach to leiomyosarcoma versus fibroids, which reaches 88–95% accuracy with DWI and ADC. The pearl is on putting ASPECTS in the first line of a stroke CT report.