Predictive value of the heterogeneity index in assessing treatment response using 18F-FDG PET/CT in locally advanced and advanced lung cancers.

OBJECTIVE: F-FDG) positron emission tomography/computed tomography (PET/CT)-based parameter to estimate this variability. This retrospective single-center study evaluated the predictive value of HI for treatment response in patients with locally advanced and advanced lung cancer undergoing chemoradiotherapy and compared its performance with conventional PET-derived volumetric parameters. SUBJECTS AND METHODS: F-FDG PET/CT followed by chemoradiotherapy were retrospectively analyzed. PET-derived parameters, including maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean), MTV, total lesion glycolysis (TLG), and HI, were calculated from baseline scans. HI was derived from the slope of linear regression using MTV values measured at multiple threshold levels. Treatment response was classified as regression or progression based on follow-up imaging according to positron emission tomography response criteria in solid tumors (PERCIST). Receiver operating characteristic (ROC) analysis was used to evaluate the predictive performance of MTV and HI, and multivariate logistic regression was performed to identify independent predictors of disease progression. RESULTS: Of the 44 patients, 19 (43.2%) showed regression and 25 (56.8%) showed progression. Metabolic tumor volume and HI were significantly higher in patients with progression (P=0.047 and P=0.040, respectively). Receiver operating characteristic analysis demonstrated modest discriminative performance for MTV (area under the curve [AUC]: 0.677; 95% confidence interval [CI]: 0.519-0.810) and HI (AUC: 0.682; 95% CI: 0.525-0.814), with high sensitivity but low specificity for both parameters. Heterogeneity index showed an extremely strong correlation with MTV (r=0.997), indicating substantial overlap between the two measures. In multivariate analysis, only advanced tumor stage (T3-T4) remained an independent predictor of progression (odds ratio [OR]: 6.801; 95% CI: 1.271-36.398; P=0.025), whereas HI did not retain independent significance. CONCLUSION: These findings indicate that HI has limited but statistically significant predictive value for treatment response in locally advanced and advanced lung cancer and performs similarly to MTV. Given its extreme collinearity with MTV, HI should be interpreted primarily as a derivative parameter reflecting tumor burden rather than as an independent heterogeneity biomarker. Nevertheless, because of its simple calculation and broad availability, HI may serve as a practical adjunct metabolic parameter in clinical assessment, although larger prospective studies with external validation are needed to clarify its incremental clinical utility.