Histogram analysis of multiple diffusion models for predicting advanced non-small cell lung cancer response to chemoimmunotherapy

Yu Zheng ^1,2, Liang Zhou ^1,2, Wenjing Huang ^1,2

⁰Department of Magnetic Resonance, The Second Hospital & Clinical Medical School, Lanzhou University, Lanzhou, 730030, China.

Cancer Imaging : the Official Publication of the International Cancer Imaging Society +

|

June 11, 2024

View abstract on PubMed

Summary

Intravoxel incoherent motion (IVIM) and diffusion kurtosis imaging (DKI) histogram analysis can predict treatment response in advanced non-small cell lung cancer (NSCLC) patients undergoing chemoimmunotherapy. Whole tumor analysis showed superior diagnostic performance compared to single slice analysis.

Area Of Science

Radiology and Imaging
Oncology
Medical Physics

Background

Chemoimmunotherapy efficacy in advanced non-small cell lung cancer (NSCLC) requires reliable imaging biomarkers.
Intravoxel incoherent motion (IVIM) and diffusion kurtosis imaging (DKI) offer potential for non-invasive assessment.

Purpose Of The Study

To evaluate the predictive capability of IVIM and DKI histogram analysis for treatment response in advanced NSCLC.
To compare different region of interest (ROI) selection methods (whole tumor vs. single slice) for histogram analysis.

Main Methods

Seventy-two advanced NSCLC patients receiving chemoimmunotherapy underwent pre-treatment IVIM and DKI.
Histogram parameters (ADC, Dslow, f, Dfast, Dk, K) were analyzed using whole tumor and single slice ROIs.
Logistic regression and ROC curve analysis assessed prediction performance.

Main Results

Responders showed significantly lower ADC, Dslow, Dk and higher f values than non-responders (P < 0.05).
Mean f value from whole tumor ROI yielded the highest AUC (0.886) among single parameters.
A combined model achieved AUCs of 0.968 (whole tumor) and 0.893 (single slice).

Conclusions

Whole tumor histogram analysis of IVIM and DKI is a promising tool for predicting chemoimmunotherapy response in advanced NSCLC.
Whole tumor ROI selection demonstrated superior diagnostic ability compared to single slice ROI analysis.
These imaging techniques can aid in early assessment of treatment efficacy.

Keywords:

Chemoimmunotherapy Diffusion kurtosis imaging Histogram analysis Intravoxel incoherent motion Non-small cell lung cancer