Abstract
Objective
To determine if the clinical behavior of T1 glottic squamous cell carcinoma varies by its location on the medial free edge or upper aspect of the vocal cords.
Study design
Retrospective cohort.
Setting
Single tertiary university-affiliated medical center.
Subjects and methods
Clinical, treatment, and outcome data were collected for 104 patients with T1N0M0 glottic squamous cell carcinoma who were treated and followed at our center in 1995–2013. Findings were compared between those with a tumor on the medial ( n = 60, 57.7%) or superior ( n = 44, 42.3%) aspect of the cords.
Results
Mean follow-up time was 4.15 years. No between-group differences were found in demographic or risk factors. There was a significant association of anterior commissure involvement with disease recurrence ( P = 0.0012) and of superior (vs medial) location with higher rates of anterior commissure involvement ( P < 0.001) and recurrence ( P = 0.01) and shorter time to recurrence ( P < 0.001).
Conclusions
T1 squamous cell carcinomas on the superior aspect of the vocal cords have a poorer prognosis than medial tumors and should be closely monitored for recurrence.
1
Introduction
Laryngeal carcinoma accounts for 30–50% of all head and neck cancers . Up to 95% of cases are associated with smoking, and men are at a fourfold higher risk than women to acquire the disease .
Physicians tend to estimate the progression of laryngeal tumors according to their location. Early-stage laryngeal carcinoma usually has a good prognosis and is confined by visceral compartments which are delineated by anatomic structures and barriers, including the quadrangular membrane, the conus elasticus, the vocal ligament, thyroid cartilage, and cricoid cartilage . The majority of laryngeal tumors arise from the glottic region. Unlike tumors from other areas of the larynx, glottic tumors can usually be diagnosed at an early stage because they manifest clearly as hoarseness. Early-stage glottic tumors [T1N0M0 squamous cell carcinoma (SCC)] may vary in size and appear in different locations within the glottis, including the ventricle floor, anterior and posterior commissures, and inferior parts of the vocal cords. Vocal cord lesions may also vary morphologically (polypoid, ulcerated, and plaque-like lesions). However, the classification of early glottic cancer is mainly functional and does not consider these parameters. Furthermore, it is well established that Involvement of the anterior or posterior commissure or the posterior third of the vocal cords is considered an unfavorable prognostic factor .
The importance of the surgical tumor margins in laryngeal cancer and their adequate size are controversial, although close or involved margins are considered an unfavorable prognostic factor. Achieving adequate surgical margins requires experience with the suspension laryngoscope and even laser vestibulectomy to ensure proper exposure of lesions located laterally or anteriorly.
The vocal cord histology changes from squamous epithelium medial to the arcuate line to pseudo-stratified columnar epithelium lateral to the arcuate line. Prompted by these findings, together with the different surgical approach needed for glottic tumors that extend superiorly and laterally (addition of vestibulectomy and a larger excision), we hypothesized that the prognosis of glottic SCC may differ when the tumor is located on the medial free edge of the vocal cord facing the lumen or spreads along the upper aspect of the vocal cord extending towards the sinus of Morgagni. The aim of the present study was to evaluate the clinical behavior of T1 glottic SCCs by anatomic site of occurrence and the relationship of site of occurrence to prognosis.
2
Materials and methods
A retrospective case-series design was used. The study group included adult patients with pathologically proven T1N0M0 glottic SCC who were treated and followed for at least one year at a tertiary university-affiliated medical center in 1995 to 2013. Patients younger than 18 years or with less than one year of follow-up after treatment were excluded as were patients with a laryngeal mass other than SCC or a second primary and patients for whom vocal cord photography data were missing or not sufficient.
Of the 276 patients with T1N0M0 glottic cancer treated in our center during the study period, 104 were eligible for the study. Most of them were excluded due to insufficient photography data.
For the patients eligible for this study, the following data were recorded from the medical files of the eligible patients: age, sex, lesion laterality (right or left), risk factors, presenting symptoms, symptom duration, morphologic appearance of the tumor, commissure involvement, type of treatment, margins, recurrence, complications, and pathologic differentiation status. We revised the pre-operative images of the video stroboscope and the operative rigid fiber optic laryngoscope images (0°, 70°) examined through direct laryngoscopy. Those photos were used to determine the location of the mass and gain a better view of the laryngeal structures and tumor extension to the commissures, subglottis, and sinus of Morgagni. Findings were compared between patients in whom the tumor was located on the medial aspect of the vocal cords alone or involved the superior aspect of the vocal cords as well ( Fig. 1 ), as determined by consensus of two reviewers.
The study protocol was approved by the local Institutional Review Board.
2.1
Statistical analysis
Continuous variables are recorded as mean ± standard deviation (SD). Differences in continuous variables between groups were analyzed by Student’s t -test, and differences in categorical variables, by chi-square test. The rates of disease-specific survival (DSS) and disease-free survival (DFS) were calculated from the date of diagnosis to the date of death, recurrence, or last follow-up and analyzed by the Kaplan-Meier method. The reliability of the group division in the absence of established guidelines defining medial and superior tumors was evaluated by test re-test reliability coefficient. P values < 0.05 were considered significant.
2
Materials and methods
A retrospective case-series design was used. The study group included adult patients with pathologically proven T1N0M0 glottic SCC who were treated and followed for at least one year at a tertiary university-affiliated medical center in 1995 to 2013. Patients younger than 18 years or with less than one year of follow-up after treatment were excluded as were patients with a laryngeal mass other than SCC or a second primary and patients for whom vocal cord photography data were missing or not sufficient.
Of the 276 patients with T1N0M0 glottic cancer treated in our center during the study period, 104 were eligible for the study. Most of them were excluded due to insufficient photography data.
For the patients eligible for this study, the following data were recorded from the medical files of the eligible patients: age, sex, lesion laterality (right or left), risk factors, presenting symptoms, symptom duration, morphologic appearance of the tumor, commissure involvement, type of treatment, margins, recurrence, complications, and pathologic differentiation status. We revised the pre-operative images of the video stroboscope and the operative rigid fiber optic laryngoscope images (0°, 70°) examined through direct laryngoscopy. Those photos were used to determine the location of the mass and gain a better view of the laryngeal structures and tumor extension to the commissures, subglottis, and sinus of Morgagni. Findings were compared between patients in whom the tumor was located on the medial aspect of the vocal cords alone or involved the superior aspect of the vocal cords as well ( Fig. 1 ), as determined by consensus of two reviewers.
The study protocol was approved by the local Institutional Review Board.
2.1
Statistical analysis
Continuous variables are recorded as mean ± standard deviation (SD). Differences in continuous variables between groups were analyzed by Student’s t -test, and differences in categorical variables, by chi-square test. The rates of disease-specific survival (DSS) and disease-free survival (DFS) were calculated from the date of diagnosis to the date of death, recurrence, or last follow-up and analyzed by the Kaplan-Meier method. The reliability of the group division in the absence of established guidelines defining medial and superior tumors was evaluated by test re-test reliability coefficient. P values < 0.05 were considered significant.
3
Results
3.1
Patient characteristics
The study cohort included 89 men (85.6%) and 15 women (14.4%) aged 34 to 95 years (mean 66.6 ± 12.5 years). Eighty-two (78.8%) were smokers and 8 (7.7%) were chronic consumers of alcohol. Twenty-five patients (24.0%) had a history of reflux. There was no difference between patients with superior or medial tumors in age, sex, tumor laterality, or risk factors of alcohol consumption, reflux, or smoking.
Mass location was evaluated twice, at an interval of several months, and was found to be reliable for group allocation (kappa = 0.8474). In 60 patients (57.7%), the mass was located only on the medial aspect, involving only the free edge of the vocal cord, and in 44 patients (42.3%), the mass involved the superior aspect of the vocal cord, spreading to the floor of the ventricle. Of them 31 (70.4%) patients had involvement of both medial and superior aspects.
Of all the patients 15 showed involvement of the anterior commissure without involvement of the contralateral vocal cord, other 17 had T1b tumor, involving not only the commissure but the contralateral vocal cord itself. Of them 9 (15%) from the medial tumor group and 8 (18.2%) from the superior tumor group ( P = NS).
Mean duration of hospitalization was 2.2 ± 1.4 days (range 1–10 days), with no difference between the groups. In accordance with our institutional policy, the operated patients were discharged one day after surgery. There were no delayed discharges due to surgical complications. The comparison of the two groups by demographic and clinical variables is presented in Table 1 .
| Superior tumor | Medial tumor | P value | |
|---|---|---|---|
| Number of patients, n (%) | 44 (42.3%) | 60 (57.7%) | |
| Average age (yr), mean ± SD | 65.6 ± 14.0 | 67.3 ± 11.3 | NS |
| Male sex, n (%) | 39 (88.6%) | 50 (83.3%) | NS |
| Left-side lesion, n (%) | 23 (52.3%) | 31 (51.7%) | NS |
| Smoking a , n (%) | 35 (79.5%) | 47 (78.3%) | NS |
| High alcohol consumption b , n (%) | 5 (11.4%) | 3 (5%) | NS |
| Reflux, n (%) | 11 (25%) | 14 (23.3%) | NS |
| Hospitalization period (days), mean ± SD | 2.1 ± 1.7 | 2.3 ± 1.1 | NS |
| TLM, n (%) | 13 (29.5%) | 27 (45%) | NS |
| Radiation therapy, n (%) | 31 (70.5%) | 33 (55%) | NS |
| Vestibulectomy, n (%) | 5 (11.4%) | 0 | 0.025 |
| Close or positive margins, n (%) | 5 (11.4%) | 4 (6.7%) | NS |
| Vocal cord involvement, n (%) | |||
| Anterior third only | 0 (0%) | 4 (6.7%) | 0.049 |
| Middle third only | 23 (52.3%) | 41 (68.3%) | NS |
| Anterior half | 16 (36.4%) | 12 (20.0%) | NS |
| Entire length | 5 (11.4%) | 3 (5.0%) | NS |
| Anterior commissure involvement, n (%) | 23 (53.5%) | 9 (15%) | < 0.001 |
| Tumor grade, n (%) | |||
| Well differentiated | 35 (79.6%) | 44 (73.3%) | NS |
| Moderately differentiated | 9 (20.45%) | 12 (20.0%) | NS |
| Poorly differentiated | 0 (0.0%) | 4 (6.7%) | NS |
| Recurrence, n (%) | 11 (25%) | 3 (5%) | 0.01 |
| Time to recurrence (yr), median ± SD | 2.7 ± 3.3 | 5.5 ± 4.2 | < 0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
