Abstract
1326
Objectives: To prospectively evaluate the role of MTV and SUV based parameters evaluated from baseline and post-therapy whole body 18F-FDG PET/CT as well as their percentage changes in response assessment of NSCLC patients. Materials and Methodology: Sixty four patients (49 male, 15 female) with mean age 55.19 ± 11.99 years (range: 31-80) were enrolled in the study. All patients underwent a baseline scan and a post-therapy scan either after four cycles of platinum based chemotherapy/ three months of the start of gefitinib or crizotinib. Scans were performed on Biograph mCT PET/CT with 64 slice CT after 60 minutes of injecting 5.18-7.77 MBq/kg of 18F-FDG intravenously. Four MTV based parameters [primary tumor MTV (MTVt), primary tumor total lesion glycolysis (TLGt), whole body MTV (MTVwb) and whole body TLG (TLGwb)] and three SUV based parameters [maximum SUV (SUVmax), average SUV (SUVavg) and tumor to background ratio (TBR)] were evaluated for baseline (pre-) and post-therapy (post-) scans using an ellipsoid isocontour ROI with threshold 2.5 SUVmax. Percentage changes (∆) in these parameters were also evaluated. Treatment response was evaluated on the basis of RECIST 1.1 using the contrast enhanced CT done during PET/CT scan. Mean values of parameters evaluated from whole body 18F-FDG PET/CT were compared among the RECIST 1.1 responders and non-responders using two sample t-test or Wilcoxon rank sum test as appropriate. Significant parameters related to response evaluation, were further analysed by ROC analysis to determine best cut-off value, and sensitivity and specificity at that cut-off value to predict response. Logistic regression was also performed on each parameter to calculate odds ratio at the respective threshold.
Results: Among the post-therapy parameters, post-MTVwb, post-TLGwb, post-MTV, post-TLG, post-TBR, post-SUVmax and post-SUVavg was found significant with AUROCs 0.85 (cut-off 60 ml), 0.83 (cut-off 255 ml), 0.82 (cut-off 21 ml), 0.81 (cut-off 80 ml), 0.77 (cut-off 5.7), 0.77 (cut-off 10) and 0.71 (cut-off 4.7), respectively. Among the percentage changes, ∆MTVwb had the highest AUROC (AUROC 0.91, cut-off 55%) followed by ∆MTV (AUROC 0.90, cut-off 57%), ∆TLGwb (AUROC 0.90, cut-off 70%), ∆TLG (AUROC 0.89, cut-off 68%), ∆SUVmax (AUROC 0.72, cut-off 33%), ∆SUVavg (AUROC 0.68, cut-off 23%) and ∆TBR (AUROC 0.66, cut-off 27%). In multivariate analysis, post-TBR and ∆MTVwb were identified as independent predictors for response to therapy. If post-TBR was ≤5.7, odds ratio for response was 5.66 (95% CI, 1.19-26.87) and if ∆MTVwb was >55%, odds ratio was 31.94 (95% CI, 6.51-156.68).
Conclusions: Our study proves that a multitude of semiquantitative parameters as documented above can be used for response assessment in NSCLC. MTV based parameters performed remarkably well in treatment response assessment of these patients. Moreover, in multivariate analysis, post-TBR and ∆MTVwb in combination came out as the independent predictors of treatment response in our study. Keywords: 18F-FDG PET/CT; treatment response assessment; MTV; TLG; SUV
