Matched single-cell analyses may reveal glioblastoma recurrence changes obscured by pooled comparisons. We reanalyzed tumor-classified cells from 9 Wang primary–recurrent pairs and 59 paired CARE patients using recurrent-minus-primary program vectors. Five of nine Wang pairs were assigned an astrocytic-like (AC-like) route, but its exact 95% proportion interval was 21–86%, and an exact paired sign-flip test did not establish an excess of this route ( p = 0.227). Continuous positive-program contributions indicated mixed trajectories. A 10-gene candidate expression module selected from 5 Wang AC-like pairs increased in all 11 CARE AC-like patients in the original sample-level analysis (median T2–T1 score +0.277), versus 13 of 48 other patients (median −0.125). In the CARE authors’ copy-number-informed malignant-cell table, 242,702 cells from paired T1/T2 samples were scored; the mean paired score change was +0.412 in AC-like patients and −0.158 in others (between-group difference +0.570, 95% confidence interval 0.321–0.820). Enrichment implicated glutamate transport but does not establish a single pathway or EGFR / NFIA causality. These findings support a recurrence-associated candidate module in an independent cohort while leaving discovery-route frequency, mechanistic function, and spatial localization unresolved.
No takes yet. Share an insight, caveat, or question.
Alexopoulos et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: