Summary
A medRxiv preprint analysing preoperative imaging from 999 people with newly diagnosed glioblastoma found that a tractography-based measure of tumour involvement across white-matter pathways was more consistently associated with overall survival than tumour volume. The finding is based on a multi-institution imaging and survival analysis, not a treatment trial.
A preprint study suggests that the way glioblastoma involves the brain’s connected white-matter pathways may provide more consistent prognostic information than the tumour’s measured volume alone. The researchers analysed preoperative imaging from 999 people with newly diagnosed glioblastoma across multiple institutions and compared conventional volume measurements with a tractography-derived measure called the lesion-tract density index, or L-TDI.
The study was posted on medRxiv on September 15, 2026. Its results describe associations with overall survival and are based on analysis of existing imaging and clinical data rather than a prospective treatment trial.
From local tumour size to network involvement
Tumour volume measures the amount of anatomical space occupied by visible tumour tissue. It is a useful representation of local extent, but glioblastoma can also affect or interact with structures distributed across the brain. White-matter pathways connect brain regions, and tractography uses imaging data to model the course of those pathways.
The L-TDI is described by the researchers as a measure of tumour involvement of white-matter pathways. In practical terms, it adds a brain-wide dimension to the assessment: rather than considering only how large the lesion is, it estimates how extensively the lesion is associated with the brain’s structural connections.
The researchers evaluated both types of measurement across tumour compartments and different stratification thresholds. A tumour compartment is a region defined by its imaging characteristics; the study specifically reports that the prognostic value of volume was largely concentrated in the contrast-enhancing compartment.
L-TDI showed more consistent survival associations
The analysis found that the relationship between tumour volume and overall survival depended substantially on how the data were analysed. Volume provided prognostic information primarily in the contrast-enhancing compartment, while its performance varied across the tested thresholds and compartments.
L-TDI showed more robust associations with overall survival across those thresholds and tumour compartments. The researchers report that it outperformed volumetric measures in their analyses and retained prognostic value after adjustment for established clinical variables.
This pattern supports a broader interpretation of glioblastoma burden. A lesion’s location within the brain’s connected architecture may capture information about disease involvement that a single measurement of local size misses. The result could help guide future prognostic models that combine conventional imaging with measures of structural connectivity.
The study does not describe a new treatment or a clinical decision tool. Its evidence comes from a multi-institution analysis of preoperative scans and survival outcomes, and the report is currently a medRxiv preprint. The abstract summarises the direction and comparative robustness of the findings but does not provide numerical effect estimates, confidence intervals or the specific threshold values used. Further work would be needed to determine how the measure performs in clinical workflows and whether using it changes patient management.