DOI: 10.1097/rct.0000000000001902 ISSN: 1532-3145

Can Spectral CT Serve as a Surrogate for Perfusion CT to Evaluate Lymphovascular and Perineural Invasion in Gastric Cancer? A Prospective Study Using a One-Stop CT Protocol

Tiezhu Ren, Long Ma, Yue Peng, Qianqian Chen, Yunhang Jing, Wei Ren, Yuan Xu, Xinan Su, Hongman Chen, Jiahui Liang, Wenjuan Zhang, Junlin Zhou
Show PDF Cite

Objective:

To evaluate the clinical utility of a “One-stop” spectral CT combined with perfusion CT scan for preoperative assessment of lymphovascular invasion (LVI) and perineural invasion (PNI) in gastric cancer (GC) and to investigate whether spectral CT parameters can serve as reliable surrogates for perfusion CT biomarkers in risk stratification.

Methods:

In this prospective study, 87 patients with histologically confirmed gastric adenocarcinoma underwent preoperative contrast-enhanced “One-stop” spectral CT combined with perfusion CT scan at our institution. Patients were stratified into LVI/PNI-positive (n=65) and LVI/PNI-negative (n=22) groups based on postoperative pathology. Spectral parameters—including iodine concentration (IC), effective atomic number (Zeff), and water concentration (WC)—were measured in arterial (AP) and venous (VP) phases. Perfusion parameters included blood volume (BV), blood flow (BF), mean transit time (MTT), permeability-surface area product (PS), and time to peak (Tmax). Intergroup comparisons, diagnostic performance evaluation, and correlation analyses between spectral and perfusion parameters were performed.

Results:

Significant differences were observed between LVI/PNI-positive and -negative groups for Zeff (in both AP and VP), IC (in VP), BV, and MTT (all P <0.05). In contrast, no significant differences were found for WC, BF, PS, or Tmax. A combined diagnostic model integrating spectral and perfusion parameters (AP+VP+P) achieved the highest performance, with an area under the curve (AUC) of 0.862 and an F 1 score of 0.892. Notably, VP-IC demonstrated a strong positive correlation with BV (Spearman ρ =0.751, P <0.001), underscoring the potential of spectral CT to reflect tumor perfusion noninvasively.

Conclusion:

Spectral CT parameters, particularly VP-IC, show a strong correlation with the perfusion biomarker BV and offer comparable diagnostic accuracy for detecting LVI/PNI in GC. This supports the role of spectral CT as a viable surrogate for perfusion CT. The integrated “One-stop” protocol, enabling simultaneous acquisition of both data sets in a single scan, provides a streamlined and clinically practical approach to preoperative risk stratification.

More from our Archive