DOI: 10.1097/mnm.0000000000002263 ISSN: 0143-3636

Is there any correlation between F-18 FDG PET-CT scan metabolic parameters and programmed death-ligand 1 expression in patients with metastatic non-small cell lung carcinoma?

Lokesh Kumar, Divya Shivdasani, Natasha Singh, Melvika Pereira

Aim

The study aimed to evaluate whether there is any correlation between metabolic parameters derived from baseline fluorine-18 fluorodeoxyglucose (F-18 FDG) PET-computed tomography (CT) and programmed death-ligand 1 (PDL-1) expression in patients with metastatic non-small cell lung carcinoma (NSCLC).

Methods

33 biopsy-proven metastatic NSCLC patients were included. On immunohistochemistry, a tumor proportion score (TPS) of ≥1% was considered PDL-1 positive, while a TPS of <1% was considered negative. Metabolic parameters of the F-18 FDG PET-CT scan were obtained by drawing a region of interest around the index lung lesion. Data were analyzed using Shapiro–Wilk test of normality, unpaired t-tests/Mann–Whitney U test, Spearman’s rank correlation, Youden’s index, and receiver operator characteristic (ROC) analysis.

Results

Spearman’s correlation analysis demonstrated a moderate positive correlation ( r  = 0.466, P  = 0.0063) between maximum standardized uptake value (SUVmax) and overall PDL-1 expression. Similarly, a moderate positive correlation was noted between SUVmax and tumors with positive PDL-1 expression ( r  = 0.643, P  = 0.013), with mean SUVmax in PD-L1-positive tumors being significantly higher ( P  = 0.032). The ROC curve analysis yielded a cutoff value of SUVmax >18 to differentiate between PDL-1-positive and -negative tumors with a sensitivity of 57.1% and specificity of 89.5%. Total lesion glycolysis (TLG) and metabolic tumor volume (MTV) showed no significant association with PDL-1 expression ( P  = 0.58 and P  = 0.88, respectively).

Conclusion

The study indicates that SUVmax has the potential to be used as a noninvasive biomarker for predicting PDL-1 expression in NSCLC. No significant difference was noted in the median MTV and TLG values between the PDL-1-positive and -negative groups.

More from our Archive