# Perioperative leukocyte–plateletcrit shift as a prognostic signature in glioblastoma

## TL;DR

Changes in blood cell counts before and after glioblastoma treatment can predict patient survival, offering a low-cost and scalable tool for prognosis.

## Contribution

The study identifies dynamic leukocyte–plateletcrit shifts as a novel independent prognostic signature in glioblastoma.

## Key findings

- The post-operative rise in leukocytes relative to plateletcrit (Δ post-op − pre-op leu/PCT) was strongly associated with worse survival.
- Concomitant and adjuvant temozolomide use showed protective effects in multivariable analysis.
- Dynamic blood cell indices provided independent prognostic value beyond standard clinical factors.

## Abstract

Circulating inflammatory indices derived from routine blood counts may offer pragmatic prognostic information in glioblastoma (GB), yet the prognostic role of plateletcrit (PCT) and of peri-treatment dynamics in leukocyte–platelet coupling remains underexplored.

We retrospectively studied 95 adults with histologically confirmed GB (48 men, 47 women; median age 64.5 years) treated adjuvantly with radiotherapy and chemotherapy with complete blood counts obtained at four windows: pre-operative, post-operative, pre-adjuvant, and post-adjuvant. From leukocytes, platelets (PLT), plateletcrit (PCT), and mean platelet volume (MPV) we derived all within-timepoint ratios and inter-timepoint differences (Δ). Overall survival (OS) in days was modeled using Cox proportional hazards (per + 1 SD), with Benjamini–Hochberg false-discovery summaries; a multivariable model adjusted for age, sex, extent of resection, radiotherapy, concomitant temozolomide (TMZ), and number of adjuvant TMZ cycles was calculated.

Median overall survival (OS) was 406 days, with 80 deaths. In univariate models, dynamic indices predominated: the post-operative rise in leukocytes relative to plateletcrit (Δ post-op − pre-op leu/PCT) showed the strongest adverse association (HR 1.60, 95% CI 1.22–2.10; p = 0.0007; BH–FDR q = 0.09), with concordant signals for Δ leu/PLT (HR 1.43, 95% CI 1.14–1.79; p = 0.002) and a protective inverse for Δ PCT/leu (HR 0.66, 95% CI 0.50–0.87; p = 0.003). Cross-sectionally, higher post-operative leu/PCT and leu/PLT and higher post-adjuvant leukocytes and leu/MPV were adverse (all p < 0.05). In the multivariable model adjusting for age, sex, extent of resection, radiotherapy, concomitant temozolomide (TMZ), and number of adjuvant TMZ cycles, Δ leu/PCT remained independently associated with worse OS (HR 1.62, 95% CI 1.04–2.52; p = 0.031), while concomitant TMZ (HR 0.18, 95% CI 0.05–0.68; p = 0.012) and greater adjuvant TMZ exposure (per + 1 SD; HR 0.48, 95% CI 0.29–0.78; p = 0.003) were protective.

Dynamic leukocyte–platelet coupling—especially Δ (post-op − pre-op) leu/PCT—provides independent prognostic information beyond standard covariates. CBC-based trajectories are low-cost and scalable and warrant prospective validation. Interpretation is limited by the absence of systematic MGMT methylation data and requires external validation and comparison with other prognostic scoring systems.

The online version contains supplementary material available at 10.1007/s11060-025-05302-8.