Review Article
Management of T1-T2 glottic carcinomas
William M. Mendenhall, M.D. Department of Radiation Oncology, University of Florida College of Medicine, Gainesville, Florida

Local control rates after laser excision range from approximately 80% to 90% for T1 disease and from 70% to 85% for T2 disease. Local control with laryngeal preservation generally exceeds 95% for patients with T1 malignancies. It is difficult to estimate the rates of local control with laryngeal preservation for patients with T2 malignancies because of selection bias, stage migration related to pathologic staging, the combination of results for patients with early-stage/invasive tumors, and small patient samples. With these caveats, local control rates range from approximately 70% to 90%. Limited data indicate that ultimate local control rates range from 90% to 95%.

Open partial laryngectomy

The local control rates after open partial laryngectomy ranges from approximately 90% to 95% for patients with T1 disease and from 70% to 90% for those with T2 disease. Rates of local control with larynx preservation range from approximately 90% to 95%. Ultimate local control rates exceed 95%.

Radiotherapy

The 5-year local control rates after radiotherapy vary from approximately 85% to 94% for T1 tumors and from 70% to 80% for T2 malignancies. The rates of local control with laryngeal preservation are marginally better than the local control rates without laryngeal preservation, indicating that a small number of patients with local recurrence receive successful salvage treatment via a conservative procedure. Ultimate local control rates range from 90% to 95%.

Local control after radiotherapy is adversely influenced by several parameters, including increasing T stage, prolonged overall treatment time, male gender, low pretreatment hemoglobin level, and poor histologic differentiation.The dose per fraction is related to overall treatment time. Therefore, treatment with < 2 gray (Gy) once daily is associated with a lower likelihood of cure.

Complications
                       
Transoral Laser Excision

Vilaseca-González conducted a study of 102 patients who underwent transoral laser excision at the University of Barcelona (Barcelona, Spain) between 1998 and 2001 for pathologic Stage Tis-T1 (n = 87) or T2 (n = 15) glottic carcinoma. Major complications were defined as those necessitating intensive medical treatment, blood transfusion, surgery, or admission to the intensive care unit. The incidence of major complications was 0% for patients with Tis disease, 0% for patients with T1 disease, and 13% for patients with T2 disease.

Steiner[10] investigated 159 patients who were treated between 1979 and 1985 for pathologic Stage Tis-T2 glottic carcinoma; 1 patient experienced postoperative bleeding. Spector  observed no major complications in a series of 61 patients who underwent transoral laser excision for T1N0 glottic SCC at the Washington University School of Medicine between 1971 and 1990.

Open Partial Laryngectomy

Giovanni conducted a study involving 127 patients who underwent partial frontolateral laryngectomy with epiglottic reconstruction at Universitaire Timone (Marseille, France) between 1982 and 1997 for T1N0 (n = 62) or T2N0 (n = 65) glottic carcinoma. Six patients (5%) required endoscopic laser excision of granulation tissue or adhesions, 4 patients (3%) required hospitalization for aspiration, and 1 patient (1%) required a permanent gastrostomy tube.

In their study, Crampette reported that 81 patients underwent a modified subtotal laryngectomy with CHEP for T1-T2 glottic carcinoma at the Hôpital Saint Charles (Montpellier, France) between 1984 and 1993. Of these 81 patients, 6 underwent surgery for local disease recurrence after previous surgery (n = 3) or radiotherapy (n = 3). Complications included aspiration pneumonia (n = 7), pulmonary embolism (n = 1), transoral laser excision for laryngeal edema (n = 1), laryngeal stenting for subglottic stenosis (n = 1), and placement of a permanent gastrostomy tube (n = 1). Two patients (3%) died postoperatively, 1 due to a pulmonary abscess and 1 due to a mesenteric artery infarction.

Spector  reported on 404 patients who underwent open partial laryngectomy at Washington University School of Medicine for T1N0 glottic carcinoma between 1971 and 1990. Sixteen patients (4%) experienced major complications, and 2 patients died. An additional 71 patients underwent open partial laryngectomy for T2N0 glottic carcinoma.[13] Complications included fistula (n = 3), infection (n = 2), bleeding (n = 1), and superficial soft tissue sloughs (n = 3). Six of nine patients experienced complications after salvage procedures. One patient (1%) died.

Radiotherapy

Mendenhall  reported that 519 patients received radiotherapy at the University of Florida (Gainesville, FL) for T1N0-T2N0 glottic carcinoma. Six patients (1.1%) developed severe complications, including severe mucositis necessitating hospitalization and a treatment break (n = 1), total laryngectomy for a suspected local tumor recurrence with a pathologically negative specimen (n = 1), permanent tracheostomy for severe laryngeal edema (n = 3), and a pharyngocutaneous fistula after a salvage total laryngectomy (n = 1). Five of these patients had T2N0 disease. None of these patients died.

In their study, Garden  evaluated 230 patients with T2 carcinoma who received radiotherapy at MDACC. Ten patients (4%) experienced severe complications, including tracheostomy (n = 5), hospitalization (accompanied by steroidal treatment) for laryngeal edema (n = 1), fistula after salvage surgery (n = 1), death during salvage surgery (n = 2), and death following salvage surgery (n = 1).

Voice Quality
                       
Rosier  reported on 18 patients treated at St.-Luc University Hospital (Brussels, Belgium) between 1979 and 1995 for T1N0 glottic carcinoma with transoral laser resection (n = 6), open partial laryngectomy (n = 5), or radiotherapy (n = 7). Perceptual voice ratings were provided by the patients, three non-speech therapists, and two speech therapists. There was a trend toward poorer patient satisfaction after open partial laryngectomy compared with laser excision or radiotherapy. There was also a nonsignificant trend toward less hoarseness and breathiness after radiotherapy compared with transoral laser excision.

Rydell  evaluated 36 patients at the University Hospital of Lund (Lund, Sweden) who received either transoral laser excision (n = 18) or radiotherapy (n = 18) for T1a glottic carcinoma. Outcomes were determined via computer assessment as well as subjective assessment by four experienced listeners and four naive listeners. Voice quality at 3 months and at 2 years after treatment (as assessed by the acoustic variables breathiness, jitter, fundamental frequency average, running speech voice sample, reading time, and number of breaths) was significantly better for patients who received radiotherapy compared with patients who received transoral laser excision. Voice quality as assessed by perceptual variables also was significantly superior for patients who received radiotherapy.

McGuirt evaluated voice outcomes 6 months after radiotherapy (n = 13) or laser cordectomy (n = 11) at the Bowman Gray School of Medicine (Winston-Salem, NC). All patients were free of disease after treatment, and the laser treatment group experienced resection of less than one-half of the cordal depth. Physician- and patient-rated voice outcomes were better after radiotherapy compared with transoral laser excision. Speech pathologists detected no differences in results.

Voice quality after transoral laser excision is closely associated with extent of resection, as was demonstrated in a recent study conducted by Delsupehe These investigators reported on a perceptual voice analysis comparing 12 patients treated with radiotherapy with 30 patients who underwent narrow-margin laser cordectomy. Although voice quality deteriorated initially after surgery compared with radiotherapy, no significant differences were detected at 6 months or 2 years follow-up.

Crampette e reported on 81 patients who underwent modified subtotal laryngectomy with CHEP at Hôpital Saint Charles for T1-T2 glottic carcinoma. At 18 months after surgery, 86% of patients were able to communicate effectively in all situations, and 42% were satisfied with their voice quality. The authors concluded that the surgery resulted in a significant level of voice disability, but this represented only a minor issue.

Verdonck-de Leeuw  compared the voice quality, voice function, and vocal performance of 60 patients treated with radiotherapy at the University of Amsterdam (Amsterdam, The Netherlands) with 20 matched control speakers. Voice characteristics improved after radiotherapy, and 50% of patients developed voice quality that was comparable to that of the control speakers. Vocal cord stripping at diagnosis and continued smoking had an adverse impact on vocal performance after radiotherapy.

The data pertaining to voice quality are based on relatively small patient samples. With this caveat in mind, we believe that it is likely that voice quality is similar after radiotherapy compared with transoral laser excision for patients with limited T1a tumors. As the extent of laser excision increases for larger tumors, voice quality results are likely to deteriorate relative to the results associated with radiotherapy. Open partial laryngectomy results in relatively poor voice quality compared with transoral laser excision or irradiation.

Conclusions
                       
To summarize, the rates of local control, laryngeal voice preservation, ultimate local control, and survival are comparable for patients treated with transoral laser excision, open partial laryngectomy, and radiotherapy. Patients with anterior commissure involvement may have a higher risk of local disease recurrence after partial laryngectomy, particularly if transoral laser excision is performed. Anterior commissure invasion does not significantly influence the likelihood of local control after radiotherapy. In addition, a subset of patients with unfavorable T2 tumors may have a better local control rate after open partial laryngectomy. The risk of major complications is comparable for patients treated with either laser resection or radiotherapy. Patients who undergo an open partial laryngectomy are likely to experience greater morbidity. Voice quality outcomes associated with laser excision and radiotherapy probably are similar for patients with limited T1a tumors. However, laser resection of larger tumors will probably result in worse outcomes.[2] Open partial laryngectomy results in poorer voice quality compared with either laser resection or radiotherapy. It is likely that the cost of transoral laser excision is similar to the cost of radiotherapy, although it must be noted that the missed work and inconvenience associated with a 5-6-week course of radiotherapy are not reflected in cost analyses. However, many patients are elderly and retired, and missed work therefore may not be an issue for a significant subset of patients. Open partial laryngectomy usually requires hospitalization and is significantly more expensive than either transoral laser resection or radiotherapy.

Patients with limited, well defined T1aN0 carcinoma evaluated at the University of Florida are treated with either radiotherapy or transoral laser excision. In practice, our preference has been to use radiotherapy for the majority of patients. The remaining patients with T1N0 tumors and all patients with T2N0 lesions are treated with definitive radiotherapy. Patients are treated at 2.25 Gy per fraction to 63 Gy (T1 and T2a) or 65.25 Gy (T2b) in a continuous course involving fields limited to the glottis. The risk of subclinical disease in the neck is very low, and elective neck irradiation is not performed.[1] Open partial laryngectomy is reserved for those who experience local disease recurrence. In addition, partial laryngectomy may be the optimal treatment option for patients who are unable to be present for a 5.5-6-week radiotherapy course.