AIM to evaluate the predictive role of pre-therapy 18F-FDG uptake indices (metabolic tumor volume -MTV, SUVmean and total lesion glycolysis -TLG) in head and neck cancer. MATERIALS AND METHODS Nineteen pts (15 men, 4 women, mean age 59.2 yrs, range 23-81) were retrospectively included in the study. Twelve (63.1%) pts had stage IV, 3 (15.8%) had stage III, 3 (15.8%) had stage II and 1 (5.3%) had stage I disease. Tumour site was oropharynx in 12 (63.1%), rynopharynx in 5 (26.3%) and larynx in 2 (10.5%) pts. All pts underwent 18F-FDG PET/CT prior to treatment with Simultaneous Integrated Boost Intensity-Modulated Radiotherapy (SIB-IMRT) guided by Biological Target Volume (BTV) defined on PET/CT images. Based on disease extension, 10 pts received concomitant chemotherapy. The median follow-up was 19.3 months (range 4 - 24). The prognostic role of primary tumor MTV, SUVmean and TLG, was assessed, using ROC analysis, considering 2-year disease free survival (DFS), local relapse free survival (LRFS) and distant metastasis free survival (DMFS). SUVmean and TLG were used with and without correction for PartialVolume Effect (PVC). RESULTS A MTV cut-off value could divide the whole population into two groups, with MTV lower and higher than 32.4cc, characterized by different survival: DFS was 73.3 vs 25% (p<0.025), LRFS was 80 vs 25% (p<0.01) and DMFS was 93.3 vs 37.5% (p<0.01), respectively. A SUVmean cut-off value could separate the whole population into two groups, with SUVmean lower and higher than 10.8, characterized by DFS of 90 vs 33.3% (p<0.025), LRFS of 90 vs 44.4% (p<0.05) and DMFS of 90 vs 77.7% (p>0.05), respectively. When PVC was applied, DFS was 91 vs 25% (p<0.01), LRFS was 91 vs 37.5% (p<0.025) and DMFS was 91 vs 75%(p>0.05), with SUVmean lower and higher than 13.3, respectively. A TLG cut-off value of 469.8 g had the highest degree of significance in dividing the population in groups, with TLG below and higher than 469.8g, characterized by DFS of 70.6 vs <1% (p<0.001), LRFS of 76.5 vs <1% (p<0.001) and DMFS of 94.1 vs <1% (p<0.001), respectively. When PVC was applied, 547.3 g TLG cut-off value could equally divide the population and the figures. CONCLUSIONS In our series MTV and TLG demonstrated a prognostic value in predicting DFS, LRFS and DMFS, while SUVmean could only predict DFS and LRFS. When PVC was applied the uptake indices clustered the patients with higher degree of statistical significance.
Prognostic Value of pre-therapy 18FDG PET/CT for the Outcome of 18FDG PET-guided SIB-IMRT in Patients with Head and Neck Cancer
Isabella Castiglioni;Francesca Gallivanone;Maria Carla Gilardi;Cristina Messa;
2012
Abstract
AIM to evaluate the predictive role of pre-therapy 18F-FDG uptake indices (metabolic tumor volume -MTV, SUVmean and total lesion glycolysis -TLG) in head and neck cancer. MATERIALS AND METHODS Nineteen pts (15 men, 4 women, mean age 59.2 yrs, range 23-81) were retrospectively included in the study. Twelve (63.1%) pts had stage IV, 3 (15.8%) had stage III, 3 (15.8%) had stage II and 1 (5.3%) had stage I disease. Tumour site was oropharynx in 12 (63.1%), rynopharynx in 5 (26.3%) and larynx in 2 (10.5%) pts. All pts underwent 18F-FDG PET/CT prior to treatment with Simultaneous Integrated Boost Intensity-Modulated Radiotherapy (SIB-IMRT) guided by Biological Target Volume (BTV) defined on PET/CT images. Based on disease extension, 10 pts received concomitant chemotherapy. The median follow-up was 19.3 months (range 4 - 24). The prognostic role of primary tumor MTV, SUVmean and TLG, was assessed, using ROC analysis, considering 2-year disease free survival (DFS), local relapse free survival (LRFS) and distant metastasis free survival (DMFS). SUVmean and TLG were used with and without correction for PartialVolume Effect (PVC). RESULTS A MTV cut-off value could divide the whole population into two groups, with MTV lower and higher than 32.4cc, characterized by different survival: DFS was 73.3 vs 25% (p<0.025), LRFS was 80 vs 25% (p<0.01) and DMFS was 93.3 vs 37.5% (p<0.01), respectively. A SUVmean cut-off value could separate the whole population into two groups, with SUVmean lower and higher than 10.8, characterized by DFS of 90 vs 33.3% (p<0.025), LRFS of 90 vs 44.4% (p<0.05) and DMFS of 90 vs 77.7% (p>0.05), respectively. When PVC was applied, DFS was 91 vs 25% (p<0.01), LRFS was 91 vs 37.5% (p<0.025) and DMFS was 91 vs 75%(p>0.05), with SUVmean lower and higher than 13.3, respectively. A TLG cut-off value of 469.8 g had the highest degree of significance in dividing the population in groups, with TLG below and higher than 469.8g, characterized by DFS of 70.6 vs <1% (p<0.001), LRFS of 76.5 vs <1% (p<0.001) and DMFS of 94.1 vs <1% (p<0.001), respectively. When PVC was applied, 547.3 g TLG cut-off value could equally divide the population and the figures. CONCLUSIONS In our series MTV and TLG demonstrated a prognostic value in predicting DFS, LRFS and DMFS, while SUVmean could only predict DFS and LRFS. When PVC was applied the uptake indices clustered the patients with higher degree of statistical significance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
