Evaluation Of Long-term Efficacy Of ¹⁸F-FDG PET SUVmax For Nasopharyngeal Carcinoma Patients With Local And Regional Recurrence

Objective:Retrospectively analyzing clinical materials of patients with locally and regionally recurrent nasopharyngeal carcinoma（rNPC）,to evaluate the value of SUVmax and other clinical factors in assessing treatment efficacy and prognosis for rNPC patients,A certain reference value for optimal treatments and prognosis assessments can be provided for those patients.Methods:From Jan.2008 to Dec.2013,80 rNPC patients were treated in our center,who were histologically or radiologically diagnosed and re-staged according to the 2008 clinical staging criteria of China.All patients accepted pretreatment ¹⁸F-FDG PET-CT and treated by IMRT with or without chemotherapy.Patients in stageⅠ,stageⅡ,stageⅢand stageⅣwere8（10%）,10（12%）,31（39%）and 31（39%）,respectively.And patients categorized to rT1,rT₂,rT₃ and rT₄ were 10（12%）,10（12%）,30（38%）and30（38%）,respectively.and that of rN0,rN1,rN2 and rN3 were 57（71%）,18（22%）,2（3%）and 3（4%）,respectively.Patients with local recurrence accounted for 71%,local and regional recurrence accounted for 29%,patients with regional recurrence were excluded.10 patients only received re-radiotherapy（IMRT）,while the rest were given with IMRT and platinum-based chemotherapy,The number of patients who received concurrent radiochemotherapy,induction chemotherapy and concurrent radiochemotherapy,inductionchemotherapyandIMRT,adjuvant chemotherapy and IMRT were 7,41,19 and 3,respectively.Induction and adjuvant chemotherapy consisted of taxol at a dose of 135¹75 mg/m² for 1day（day 1）and cisplatin（DDP）at a dose of 100¹20 mg/m² for 3 days（day1³）or nedaplatin（NDP）at a dose of 70⁸0 mg/m² for 1 day（day 2）,concurrent chemotherapy consisted of DDP or NDP only with the same dose,chemotherapy was performed at 3-4weeks intervals.The median mean dose in GTV was 69Gy（range:60⁷8Gy）,with the fractional mean dose of 2².76Gy.The relationship between SUV_max and long-term clinical outcomes were analyzed.The best cutoff value for SUV_max was determined by SPSS18.0 receiver operating characteristic（ROC）curve.The Kaplan-Meier method was used to calculate OS,DFS,LRFS,RRFS,DMFS,Log-rank test and single factor prognostic analysis,Cox model multivariate prognostic analysis.Results:The 3-year overall survival,disease-free survival,local recurrence-free survival,regional recurrence-free survival,distant metastasis-free survival were 34%,31.9%,32.8%,31.9%,34.2%,respectively.The death toll was 59,in which 8 died from nasopharyngeal hemorrhage and25 patients developed distant metastases.The incidence of cranial nerves injuries,radiation encephalopathy,trismus,hearing loss and nasopharyngeal mucosa inflammation were 13%,21.3%,22.5%,45%,15%,respectively.The median SUV_max was 8.4（2.7-21.5）.The SUV_max of 7.0 was taken as a best cut-off value for all patients.For patients with SUV_max≤7.0,the 3-year OS was significantly higher than those with SUV_max>7.0（42.1%vs.29.5%,P=0.025）.Univariate analysis revealed that SUV_max and rN were significant prognostic factors to predict 3-year overall survival（P=0.025,0.004）.In a multivariate analysis,SUV_max and rN remained significantly for OS,DFS and DMFS（HR=1.79,P=0.037 and HR=2.13,P=0.007;HR=1.72,P=0.045 and HR=2.51,P=0.001;HR=1.89,P=0.021 and HR=2.44,P=0.002）.For patients with advanced rT stage and rN₀ stage,SUV_max was an adverse prognostic factor for their DMFS（P=0.036,P=0.043）.and SUV_max showed a tendency to predict DFS for rN₀ stage patients（P=0.085）.Neither increasing radiation dose nor radiochemotherapy can improve survival for patients with SUV_max>7.0.Conclusion:SUV_max is one of the useful prognostic factors for OS,DFS and DMFS in rNPC patients.SUV_max may have a predictive value in rNPC patients rT stage and rN stage categories.