Prediction of local control in early glottic carcinoma using the maximum standardised uptake value.

PURPOSE: This retrospective study aimed to determine whether the maximum standardised uptake value (SUVmax ) can predict local tumour control in early glottic cancer (Tis, T1, and T2).

PATIENTS AND METHODS: Fifty-nine patients treated with definitive radiotherapy for early glottic cancer between 2003 and 2011 were enrolled. We evaluated the SUVmax in the region of interest around the original tumour site. Local tumour control and survival were estimated using Kaplan-Meier curves. Receiver operating characteristic curves were used to assess the optimal SUVmax cut-off for predicting local control.

RESULTS: As determined by laryngoscopy, all patients achieved a complete response. Eleven patients experienced local recurrence, while no distant metastasis occurred. One patient died due to local recurrence, while five lost their larynxes. The median follow-up was 61.5 (range: 6.2-123.4) months. The five-year local progression-free survival was 84.7%, and larynx preservation was possible in 89.6% of cases. The median SUVmax was 2.2. The optimal SUVmax for predicting local tumour control was identified as 3.4. Patients with glottic cancers with an SUVmax >3.4 showed a significantly lower local progression-free survival rate than those with tumours with an SUVmax <3.4 (five-year local progression-free survival rate: 53.4% vs. 95.4%, P<0.01). Multivariate analysis confirmed that a high SUVmax was an independent predictive factor for local progression-free survival (P=0.006).

CONCLUSION: The use of (18 F)-fluorodeoxyglucose-positron emission tomography for evaluation of the SUVmax is useful to predict local progression-free survival in patients with early glottic cancer treated by radiation. Early glottic cancer with a high SUVmax may require aggressive local treatment and careful surveillance.