(S037) Phenotypic Diversity Measured in PET/CT Scans Predicts Overall Survival in Early-Stage Lung Adenocarcinoma

April 30, 2015

These results suggest that within-tumor phenotypic diversity, as quantified in PET-CT scans, can predict OS in patients with early-stage lung adenocarcinoma. Quantification of within-tumor heterogeneity in this routine clinical imaging may provide a noninvasive method for identifying a high-risk subset of patients with early-stage non–small-cell lung cancer.

Jennifer S. Chang, Viola Walther, Natalie Lui, Aleah F. Caulin, David Jablons, Carlo Maley, Trevor Graham, Sue S. Yom; University of California, San Francisco; Barts Cancer Institute, Queen Mary University of London

BACKGROUND: Although most patients with early-stage lung adenocarcinoma have favorable outcomes, stratification of a high-risk subset may facilitate clinical decision making on therapy and surveillance. The degree of within-tumor heterogeneity may be a prognostic biomarker, as more diverse tumors are more likely to contain a subpopulation that is adapted to a new selective pressure (eg, radiotherapy), whereas homogeneous tumors are less likely to harbor a resistant subclone. We recently found that within-tumor genetic heterogeneity predicts overall survival (OS) in patients with early-stage lung adenocarcinoma. Here, we investigated whether quantification of within-tumor phenotypic diversity, measured by 18-fluorodeoxyglucose-positron emission tomography (FDG-PET) and computed tomography (CT) imaging, was also a predictor of OS.

METHODS: Genetic analysis had previously been performed on 58 patients initially diagnosed with stage I/II lung adenocarcinoma, treated with surgery and no neoadjuvant therapy. We obtained the presurgery PET and CT scans. The primary lung lesion was contoured on PET and CT scans using the software program MIM 5.6. Phenotypic diversity was then quantified using the Moran I statistic, which describes spatial autocorrelation. The prognostic value of the Moran I coefficient was assessed using Cox proportional hazards (CPH) models and Kaplan-Meier curves. Statistical analyses were performed in R.

RESULTS: Tumors were contoured on 46 CT and 32 PET scans. Differences in phenotypic diversity were identified across the cohort for both PET and CT images. Patients in the upper quartile of Moran I values (phenotypic diversity) had significantly shorter survival using PET scans (log-ranked: P = .02; CPH: P = .08; hazard ratio [HR] = 2.72; 95% confidence interval [CI], 0.88–8.39), and the upper quartile was borderline significant for CT scans (log-ranked: P = .051; CPH: P = .061; HR = 2.83; 95% CI, 0.95–8.45). There was a weak rank correlation between Moran’s I statistic and genetic diversity for both PET (P = .85) and CT (P = .21). Moran I and genetic diversity were significant predictors for overall survival using PET scans (Moran I: P = .045; genetic diversity: P = .032).

CONCLUSIONS: These results suggest that within-tumor phenotypic diversity, as quantified in PET-CT scans, can predict OS in patients with early-stage lung adenocarcinoma. Quantification of within-tumor heterogeneity in this routine clinical imaging may provide a noninvasive method for identifying a high-risk subset of patients with early-stage non–small-cell lung cancer.

Proceedings of the 97th Annual Meeting of the American Radium Society - americanradiumsociety.org