3600 - Voice Outcome Following Salvage Moderately Hypofractionated Radiotherapy after Transoral Microsurgery for Early-Stage Glottic Squamous Cell Carcinoma

Purpose/Objective(s): Transoral CO2 laser microsurgery and moderately hypofractionated radiotherapy are standard organ-preserving treatments for early glottic cancer. RT may be preferred in some cases because of field cancerization and the goal of maximal voice preservation. However, salvage laryngectomy can be necessary in the relatively small cases of RT failure. In well-selected patients who initially undergo microsurgery but fails, voice outcome after salvage radiotherapy (in effect, double-modality treatment) is unclear.

Materials/Methods: A total of 53 consecutive patients without local failure following RT were included in the voice outcome analysis. At each time point (3 months, 6–12 months, and 2 years post-RT), Fisher’s exact test was used to compare voice quality between the two cohorts, with and without prior surgery at the time of RT. The relative risk of having at least moderately hoarse voice (score 2+ vs 0/1) was calculated. The 95% confidence intervals for the relative risks were calculated with the exact method using SAS software version 9.4 and R version 4.4.3. We also analyzed rates of oncologic failure following RT.

Results: The mean age of the patients was 69.9 (SD 10.6). 79.3% of patients were male, and 69.8% were former or active smokers. Distribution of stage: CIS 15 (34.8%), T1a 13 (30.2%), T1b 6 (14%), T2a 6 (14%), and T2b 3 (7%). 17 (32.1%) had prior CO2 laser surgery vs. 36 (67.9%) did not prior to RT. Of the 17 who had prior surgery, 9 (52.9%) had more than one surgery. 76.5% of prior surgery group comprised of mostly CIS or T1a disease, compared to 28% in RT only group. At the time of RT, 70.7% of entire cohort had anterior commissure involvement and 70.6% in prior surgery group. 34 (64.2%) of entire cohort received 63Gy, 19 (35.8%) received 65.25Gy, in 2.25Gy fraction. 70.6% of prior surgery group received 63Gy. At 3 months post RT, the voice qualities did not significantly differ between the two cohorts with and without prior surgery. The voice qualities significantly differed between the two groups at 6-12 months post RT (p = 0.01) and at 2-year post RT (p=0.01). The relative risk of being at least moderately hoarse (2 vs. 0/1) for no prior surgery vs prior surgery group was 0.13 (95% CI: 0.02-0.47) at 6-12 months post RT and 0.14 (0.02-0.64) at 2-year post RT. Of the larger cohort that includes 7 post RT failures, local control rates for the entire cohort were 100% and 90.5% for CIS and T1, respectively. Most of the failures occurred in T2 disease. Almost all recurrences were detected within 6 months of RT, and all required laryngectomy (2 partial, 5 total).

Conclusion: In early-stage glottic SCC treated with moderately hypofractionated RT, prior transoral CO2 laser microsurgery was associated with significantly worse long-term voice outcomes despite similarly favorable oncologic control for CIS and T1 disease. These findings suggest that, when the likelihood of requiring RT after microsurgery is nontrivial and voice preservation is a priority, upfront RT may be preferable to an initial surgical approach.