The Impact Of FDG PET/CT Lymph Node Staging On 3D-CRT Planning In Patients With NSCLC

Objective: ¹⁸F-fluoro-2-deoxy-D-glucose positron emission tomography (¹⁸FDG-PET) combined with computed tomography (CT) is obviously superior to CT alone in mediastinal lymph nodes(LN) staging of non-small cell lung cancer (NSCLC). We studied the potential impact of PET/CT on the radiation treatment plan of NSCLC patients. And we expected to evaluate the potential gain from PET/CT in the three-dimensional conformal treatment (3D-CRT) planning in patients with NSCLC.Methods and materials: From April 2001 to April 2005, thirty-two patients with pathologically proven NSCLC accepted ¹⁸F-FDG PET/CT examination and 3D-CRT plans. All the patients have mediastinal LN metastasis on CT or PET/CT. For each patient, two 3D-CRT plans were made: one with a CT-based planning target volume (PTV) and one with a PET/CT-based PTV. Because our focus was only mediastinal LN metastases, the primary tumor contours of the two plans were identical, the only difference between their targets volume was the LN targets. Both volumes were delivered 60 Gy in 30 fractions. The beam arrangement of both plans was as same as possible. For the tumor and metastatic LN, we evaluated the volume of GTV and PTV. From the dose-volume histograms and dose distributions on each plan, the dosimetric factors predicting esophagus, spinal cord and lung toxicity were assessed and compared. For the lung, We analysed the mean dose (MLD) and V₂₀(the volume of lung receiving 20 Gy). For the esophagus, we analysed the maximal dose(Dmax-E), the mean dose(MED), V₄₅(the volume of esophagus receiving 45Gy) and V₅₅(the volume of esophagus receiving 55Gy). We also analysed the maximaldose of spinal cord (Dmax-SC). In order to describe the relative role of information proved by PET/CT in detail, we propose a classification based on GTV and PTV. Comparison of differences between PET/CT and CT alone were performed using the paired t test and Wilcoxon signed rank test. Probability values of less than 0.05 were considered to be statistically significant.Results: In all 32 patients, LN staging in 13 patients (41%) was increased by PET vs. CT, because new metastasis LNs were found on PET/CT scan that were not enlarged on CT scan alone. There was only one case that LN staging was decreased, because mediastinal LN that enlarged on CT were excluded metastasis by PET without FDG uptake. LN staging in 18 patients (56%) was same on both PET scan and CT scan. In 23 patients (72%) the information obtained from PET would have led to a change of the treatment volumes. The PET-based gross tumor volume (PET-GTV) was 29.19(14.06, 52.93)cm3, and the CT-based gross tumor volume (CT-GTV) was 27.71(12.25, 56.80)cm3 (p>0.05). The volume of PET-PTV was 194.02(120.97, 297.15)cm3, and the CT-PTV was 172.46(88.78, 311.42)cm3 (p<0.05). In RTPpet and RTPCT, the maxcimal dose of spinal cord was 38.79(29.12, 45.88)Gy and 37.24(22.46, 43.98)Gy (p<0.05). The Dmax-E was 66.14(61.04, 67.49)Gy in RTPpet, and was 64.56(38.20, 67.28)Gy in RTPCT(p<0.05). The V45 of RTPPET and RTPct was 21.23%(3.44%, 28.70%), 15.75%(0.04%, 25.13) (p<0.05);and V55 was 14.38%(0.72%, 21.03%), 7.63%(0.00%, 20.91%) respectively(p>0.05). The MED increased from 16.50±11.84 Gy on the CT scan to 19.45 ± 11.04Gy on the PET-CT scan (p>0.05). For lung V20 was 28.56%(23.44%, 39.44%) on PET, and was 24.30%(16.26%, 35.98%) on CT(p<0.05). The MLD increased from 14.58±7.18Gy on the CT scan to 16.35±6.82Gy on the PET-CT scan (p<0.05).Conclusions: In our study, mediastinal LN treatment volume of seventy-two percent patients with NSCLC were changed by using 18FDG-PET/CT information in radiotherapy planning. The main contribution of PET was to decrease the risk ofgeographic misses. In other words, new metastatic LN that not enlarged on CT alone was found avid FDG uptake by PET/CT. FDG-PET/CT increases the radiation exposure of the esophagus, spinal cord and lung. The precision of target volume contour was improved by avoiding geographic misses. The therapeutic ratio improvement does not result from target volume decrease, but the precision of target volume contour.