ABSTRACT
Objective
To investigate the impact of clinical surveillance, primary radiotherapy, and primary surgery on overall survival (OS) in laryngeal carcinoma in situ (Cis).
Methods
The 2006–2020 National Cancer Database was queried for adults with a biopsy‐proven diagnosis of laryngeal Cis. Multivariable binary logistic and Cox proportional hazards regression models were implemented.
Results
Of 3567 unique patients satisfying inclusion criteria, 514 (14.4%) underwent clinical surveillance, 1074 (30.1%) underwent primary radiotherapy, and 1979 (55.5%) underwent primary surgery. Receiving treatment at an academic/research facility was associated with higher odds of undergoing primary surgery compared to primary radiotherapy. Among 646 patients undergoing primary surgery with known pT classification and margins, 570 (76.6%) had pTis and NSM and 174 (23.4%) had pT1 and/or PSM. 5‐year OS of clinical surveillance, primary radiotherapy, and primary surgery was 73%, 81%, and 86%, respectively (p < 0.001). Patients undergoing primary surgery with invasive or residual disease (i.e., pT1 and/or PSM) had similar 5‐year OS as those without (84% vs. 88%, p = 0.057). Compared with primary radiotherapy, clinical surveillance (aHR 1.29, 95% CI 1.06–1.57, p = 0.003) was associated with worse OS, and primary surgery (aHR 0.80, 95% CI 0.69–0.92, p = 0.003) was associated with higher OS.
Conclusion
Primary surgery is associated with higher OS than clinical surveillance and primary radiotherapy among patients with laryngeal Cis.
Level of Evidence
4.
Keywords: carcinoma in situ, laryngeal squamous cell carcinoma, National Cancer Database, radiotherapy, surgery
This study investigates the impact of clinical surveillance, primary radiotherapy, and primary surgery on overall survival (OS) in laryngeal carcinoma in situ (Cis). Primary surgery is associated with higher OS than clinical surveillance and primary radiotherapy among patients with laryngeal Cis.
1. Introduction
Laryngeal carcinoma in situ (Cis) represents a range of premalignant changes, histologically defined by the complete replacement of the laryngeal epithelium with malignant cells, without invasion through the basement membrane [1]. Although relatively rare with an annual incidence of 0.4 cases per 100,000 patients, laryngeal Cis has rates of progression to invasive or residual disease varying from 33% to 90% [2, 3, 4]. Compared with low‐grade dysplasia, Cis and severe dysplasia have a three times greater chance of malignant transformation [5].
Optimal management of laryngeal Cis remains controversial, with disease presentation and physician and patient preferences dictating whether clinical surveillance, primary radiotherapy, or primary surgery (e.g., laser microflap excision, cordectomy) is utilized [6]. Important considerations for management include locoregional control, laryngeal preservation, voice quality, swallowing function, treatment convenience, cost, and patient health status [7]. The benefits of radiotherapy for Cis include laryngeal preservation, potentially reduced need for repeat procedures, and the ability to better preserve voice quality in some cases [7, 8, 9, 10]. On the other hand, surgical therapies such as cordectomy and conservative endoscopic laser excision offer the advantages of potentially requiring only one treatment session and offering more precise removal of malignant tissue with pathologic evaluation [8, 9]. However, these surgeries often require multiple procedures due to residual margins or recurrence and can be associated with decreased voice quality [8, 9, 11]. Given the uncertainty in the progression of Cis to invasive or residual cancer, clinical surveillance of laryngeal Cis may also be a viable approach that minimizes treatment‐related morbidity [5].
When selecting treatment, it is important to consider the rates of occult invasiveness or residual disease in lesions initially diagnosed as Cis. Active surveillance has a risk of progression necessitating more aggressive forms of therapy to manage invasive, residual, or metastatic disease, whereas primary radiotherapy or surgery carries the risk of over‐treatment and associated complications such as xerostomia, dysphagia, dysphonia, neck rigidity, osteoradionecrosis, paresthesia, bleeding, and infection. Our study of the National Cancer Database (NCDB) investigates the rate of occult invasiveness or residual disease in patients with laryngeal Cis undergoing surgery and compares survival outcomes for patients initially diagnosed with Cis managed with clinical surveillance, primary radiotherapy, and primary surgery.
2. Methods
2.1. Data Source
The NCDB, created by the American Cancer Society (ACS) and American College of Surgeons Commission on Cancer (CoC), captures > 70% of incident cancer cases in the United States from > 1500 CoC‐accredited hospitals. Our study was deemed exempt from review by the Institutional Review Boards at Rutgers New Jersey Medical School and the Perelman School of Medicine at the University of Pennsylvania because of the de‐identified nature of patient data. The ACS and CoC have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
2.2. Inclusion Criteria
The NCDB was retrospectively queried for adults with a biopsy‐proven diagnosis of laryngeal Cis between January 2006 and December 2020 (Figure 1). The assigned American Joint Committee on Cancer (AJCC) clinical tumor (cT) category of pTis, which describes a lesion with purely noninvasive in situ carcinoma on biopsy, was utilized to identify patients. LSCC was identified using the International Classification of Diseases for Oncology, Third Edition (ICD‐O‐3) histology (“8050‐8052,” “8070‐8078,” “8083‐8084”), behavior (“2”), and topography (“C32.0” for glottis; “C32.1” for supraglottis; “C32.2” for subglottis; “C32.3”, “C32.8”, and “C32.9” for other [laryngeal cartilage, overlapping, and unspecified, respectively]) codes. Patients with a history of prior malignancy, final AJCC pathology of pT2+, clinical evidence of distant metastasis, treatment with palliative intent, atypical treatment (i.e., laryngectomy, neoadjuvant therapy, or chemotherapy), and unknown treatment and survival were excluded.
FIGURE 1.
Inclusion criteria. LVI, lymphovascular invasion; NCDB, National Cancer Database; pTN, pathologic tumor‐nodal; RT, radiotherapy.
Patients were divided into three cohorts: (1) clinical surveillance, (2) primary radiotherapy, and (3) primary surgery. Clinical surveillance included patients not undergoing primary radiotherapy or primary surgery. Primary radiotherapy was defined as the delivery of therapeutic doses of external beam radiation (40–80 Gy) to the head and neck. Primary surgery was defined as the most definitive surgical procedure to the primary site and included local tumor excision (i.e., polypectomy, excisional biopsy, photodynamic therapy, electrocautery, cryosurgery, laser ablation, laser excision, and cordectomy) (Table S1). Among the primary surgery cohort, occult invasiveness or residual disease was determined by identifying patients with a final AJCC pathology of pT1 or those having positive surgical margins (PSM). Noninvasive or residual disease was defined as having pT0 classification with negative surgical margins (NSM).
2.3. Variables
Patient data included age, sex, race, facility type, Charlson–Deyo comorbidity score (CDCS), primary site, grade, pTNM classification, neck dissection status, surgical margin status, surgical length of stay (LOS), 30‐day readmission, and 90‐day mortality following surgery, treatment, vital status, and survival time. A CDCS of 0 indicates no comorbid medical conditions. PSM was defined as having microscopic, macroscopic, or unspecified residual tumor. Five‐year overall survival (OS) was the primary outcome of our study, with survival time calculated as the time from diagnosis to either death or 5 years of follow‐up.
2.4. Statistical Analysis
Pearson chi‐squared tests were used to compare the distributions of categorical variables across cohorts. The Shapiro–Wilk test was used to assess the normality of continuous variables; median, interquartile range (IQR), and Mann–Whitney U significance were reported, as appropriate. Univariable binary logistic regression models were implemented to identify factors associated with undergoing primary surgery. Kaplan–Meier analysis and the log–rank test were used to estimate 5‐year OS. A multivariable Cox proportional hazards regression model using listwise elimination for missing data and adjusting for variables significant on univariable regression (p < 0.05) was implemented to identify associations with OS. The proportionality of hazards assumption was evaluated using time‐dependent covariables and was not violated in any regression models. The two‐sided threshold for statistical significance was set at p < 0.05 for all statistical tests. SPSS version 25 (IBM) and Microsoft Excel (Microsoft) were utilized for statistical analysis.
3. Results
3.1. Patient Demographics, Pathologic Features, and Treatment
Of the 3567 unique patients with biopsy‐proven laryngeal Cis satisfying inclusion criteria, 514 (14.4%) underwent clinical surveillance, 1074 (30.1%) underwent primary radiotherapy, and 1979 (55.5%) underwent primary surgery (Table 1). Median (IQR) patient age at diagnosis was 66 (58–74) years, and most patients were male (78.3%), treated at non‐academic facilities (63.1%), had a CDCS of 0 (78.5%), with glottic disease (88.8%). Utilization of primary surgery was highest among younger patients aged 18–39 years (Figure 2). Primary treatment by year of diagnosis is depicted in Figure 3. Race was unequally distributed across treatment cohorts (p = 0.007), with White patients more commonly undergoing primary surgery. Patients undergoing primary surgery had the highest rates of receiving treatment at an academic research facility (43.4%) and glottic disease (91.3%), respectively (p < 0.001). Among 1979 pts. undergoing primary surgery, median (IQR) LOS was 0 (0–0) days; 33 (1.7%) had 30‐day readmission to the surgical facility, and 9 (0.5%) had mortality within 90 days of surgery. The most performed surgeries were excisional biopsies (N = 700; 35.4%) and laser excision (N = 397; 20.1%).
TABLE 1.
Patient demographics, pathologic features, and treatment, n (%).
| Clinical surveillance | Primary RT | Surgery | p | Total | |
|---|---|---|---|---|---|
| No. | 514 | 1074 | 1979 | — | 3567 |
| Age at diagnosis, median years (IQR) | 67 (58–76) | 66 (58–75) | 66 (57–74) | 0.264 | 66 (58–74) |
| Sex | |||||
| Male | 381 (74.1) | 869 (80.9) | 1542 (77.9) | 0.008 | 2792 (78.3) |
| Female | 133 (25.9) | 205 (19.1) | 437 (22.1) | 775 (21.7) | |
| Race | |||||
| White | 415 (82.5) | 897 (84.3) | 1712 (87.4) | 0.007 | 3024 (85.8) |
| Black | 66 (13.1) | 140 (13.2) | 196 (10.0) | 402 (11.4) | |
| Other | 22 (4.4) | 27 (2.5) | 50 (2.6) | 99 (2.8) | |
| Facility type | |||||
| Academic/research | 151 (30.1) | 302 (2.85) | 826 (43.4) | < 0.001 | 1279 (36.9) |
| Non‐academic/research | 350 (69.9) | 758 (71.5) | 1080 (56.7) | 2188 (63.1) | |
| Charlson–Deyo comorbidity score | |||||
| 0 | 401 (78.0) | 859 (80.0) | 1541 (77.9) | 0.380 | 2801 (78.5) |
| ≥ 1 | 113 (22.0) | 215 (20.0) | 438 (22.1) | 766 (21.5) | |
| Primary site | |||||
| Glottis | 409 (79.5) | 952 (88.6) | 1806 (91.3) | < 0.001 | 3167 (88.8) |
| Supraglottis | 53 (10.3) | 67 (6.2) | 74 (3.7) | 194 (5.4) | |
| Other | 52 (10.1) | 55 (5.1) | 99 (5.0) | 206 (5.8) | |
| pT classification | |||||
| Is | 54 (100) | 89 (95.7) | 994 (91.3) | 0.027 | 1137 (92.0) |
| pT1 | 0 (0) | 4 (4.3) | 95 (8.7) | 99 (8.0) | |
| Surgical margins | |||||
| Negative | — | — | 982 (91.5) | 982 (91.5) | |
| Positive | — | — | 91 (8.5) | 91 (8.5) | |
| Adjuvant radiotherapy | |||||
| No | 514 (100) | — | 1557 (80.5) | < 0.001 | 2071 (58.8) |
| Yes | — | 1074 (100) | 376 (19.5) | 1450 (41.2) | |
Note: Bold values indicate statistical significance p < 0.05.
Abbreviations: IQR, interquartile range; is, in situ; pT, pathologic tumor.
FIGURE 2.
Treatment by age at diagnosis. RT, radiotherapy.
FIGURE 3.
Treatment by year of diagnosis. RT, radiotherapy.
3.2. Factors Associated With Undergoing Primary Surgery
On multivariable binary logistic regression, undergoing treatment at an academic facility (adjusted odds ratio [aOR] 1.98, 95% confidence interval [CI] 1.68–2.33, p < 0.001) was the only factor associated with increased odds of undergoing primary surgery compared with primary radiotherapy (Table 2). Black race (aOR: 0.70, 95% CI: 0.55–0.88, p = 0.003) and disease of the supraglottis (aOR: 0.57, 95% CI: 0.40–0.81, p = 0.002) were associated with decreased odds of undergoing primary surgery.
TABLE 2.
Univariable binary logistic regression models for undergoing surgery vs. primary RT.
| OR (95% CI) | p | aOR a (95% CI) | p | |
|---|---|---|---|---|
| Age at diagnosis, years | 0.99 (0.99–1.00) | 0.070 | ||
| Sex | ||||
| Male | Ref | |||
| Female | 1.20 (1.00–1.45) | 0.053 | ||
| Race | ||||
| White | Ref | Ref | ||
| Black | 0.73 (0.58–0.93) | 0.009 | 0.70 (0.55–0.88) | 0.003 |
| Other | 0.97 (0.60–1.56) | 0.901 | 0.87 (0.53–1.41) | 0.563 |
| Facility type | ||||
| Academic/research | 1.92 (1.63–2.26) | < 0.001 | 1.98 (1.68–2.33) | < 0.001 |
| Non‐academic/research | Ref | Ref | ||
| Charlson–Deyo comorbidity score | ||||
| 0 | Ref | |||
| ≥ 1 | 1.14 (0.95–1.36) | 0.174 | ||
| Primary site | ||||
| Glottis | Ref | Ref | ||
| Supraglottis | 0.58 (0.42–0.82) | 0.002 | 0.57 (0.40–0.81) | 0.002 |
| Other | 0.95 (0.68–1.33) | 0.761 | 0.92 (0.65–1.30) | 0.623 |
Note: Bold values indicate statistical significance p < 0.05.
Abbreviations: aOR, adjusted odds ratio; OR, odds ratio; Ref, reference.
N = 2940; number of uncensored surgery events: 1890.
3.3. 5‐Year OS Among All Patients
On Kaplan–Meier analysis, 5‐year OS among patients undergoing clinical surveillance, primary radiotherapy, and primary surgery was 73%, 81%, 86%, respectively (p < 0.001) (Figure 4). On multivariable Cox regression adjusting for demographics, pathologic features, and treatment, patient age at diagnosis (adjusted hazard ratio [aHR] 1.06, 95% CI: 1.06–1.07, p < 0.001), CDCS ≥ 1 (aHR: 1.35, 95% CI: 1.17–1.57, p < 0.001), disease of the supraglottis (aHR: 2.05, 95% CI: 1.58–2.66, p < 0.001), other primary site (aHR: 1.42, 95% CI: 1.11–1.80, p = 0.005), and clinical surveillance (aHR: 1.29, 95% CI: 1.06–1.57, p = 0.003) were associated with worse OS (Table 3). Female sex (aHR: 0.76, 95% CI: 0.63–0.91) and primary surgery (aHR: 0.80, 95% CI: 0.69–0.92, p = 0.003) were associated with higher OS. Patient race, facility type, and CDCS were not significantly associated with OS.
FIGURE 4.
Kaplan–Meier analysis of 5‐year overall survival among 514 patients undergoing clinical surveillance, 1074 undergoing primary RT, and 1979 undergoing surgery (p < 0.001). Significance is derived from the log–rank test.
TABLE 3.
Univariable and multivariable Cox proportional hazard regression models for 3567 patients with laryngeal cTis.
| Univariable | Multivariable | |||
|---|---|---|---|---|
| HR (95% CI) | p | aHR a (95% CI) | p | |
| Age at diagnosis, years | 1.06 (1.06–1.07) | < 0.001 | 1.06 (1.06–1.07) | < 0.001 |
| Sex | ||||
| Male | Ref | Ref | ||
| Female | 0.70 (0.59–0.83) | < 0.001 | 0.76 (0.63–0.91) | 0.001 |
| Race | ||||
| White | Ref | |||
| Black | 1.09 (0.89–1.34) | 0.420 | ||
| Other | 0.65 (0.38–1.11) | 0.112 | ||
| Facility type | ||||
| Academic/research | 0.85 (0.74–0.98) | 0.022 | 0.94 (0.81–1.08) | 0.381 |
| Non‐academic/research | Ref | Ref | ||
| Charlson–Deyo comorbidity score | ||||
| 0 | Ref | Ref | ||
| ≥ 1 | 1.50 (1.29–1.74) | < 0.001 | 1.35 (1.17–1.57) | < 0.001 |
| Primary site | ||||
| Glottis | Ref | Ref | ||
| Supraglottis | 1.55 (1.20–2.00) | < 0.001 | 2.05 (1.58–2.66) | < 0.001 |
| Other | 1.37 (1.07–1.74) | 0.011 | 1.42 (1.11–1.80) | 0.005 |
| Primary treatment | ||||
| None | 1.35 (1.11–1.64) | 0.001 | 1.29 (1.06–1.57) | 0.011 |
| Radiotherapy | Ref | Ref | ||
| Surgery | 0.76 (0.66–0.89) | < 0.001 | 0.80 (0.69–0.92) | 0.003 |
Note: Bold values indicate statistical significance p < 0.05.
Abbreviations: aHR, adjusted hazard ratio; HR, hazard ratio; Ref, reference.
N = 3467; number of uncensored deaths: 882.
3.4. Patients Undergoing Primary Surgery
Among 744 patients undergoing primary surgery with known pT classification and margin status, 570 (76.6%) had pTis and NSM, and 174 (23.4%) had pT1 and/or PSM (Table 4). Adjuvant therapy was used more frequently in patients with pTis + NSM (88.3%) compared to pT1+ and/or PSM (74.3%; p < 0.001). Among 665 patients undergoing primary surgery with known pT classification, surgical margins, and aRT status, rates of aRT utilization were 11.7% (N = 66) for pTis + NSM, 12.2% (N = 6) for pT1 and NSM, 23.3% (N = 10) for pTis + PSM, and 9.1% (N = 1) for pT1 and PSM.
TABLE 4.
Patient demographics, pathologic features, and treatment by invasiveness, n (%).
| Noninvasive (i.e., pTis + NSM) | Invasive (i.e., pT1+ and/or PSM) | p | Total | |
|---|---|---|---|---|
| No. | 570 | 174 | 744 | |
| Age at diagnosis, median years (IQR) | 65 (57–73) | 66 (57–75) | 0.260 | 66 (58–75) |
| Sex | ||||
| Male | 436 (76.5) | 146 (83.9) | 0.038 | 582 (78.2) |
| Female | 134 (23.5) | 28 (16.1) | 162 (21.8) | |
| Race | ||||
| White | 483 (86.3) | 144 (83.7) | 0.101 | 627 (85.7) |
| Black | 63 (11.3) | 18 (10.5) | 81 (11.1) | |
| Other | 14 (2.5) | 10 (5.8) | 24 (3.3) | |
| Facility type | ||||
| Academic/research | 287 (52.6) | 91 (52.9) | 0.937 | 378 (52.6) |
| Non‐academic/research | 259 (47.4) | 81 (47.1) | 340 (47.4) | |
| Charlson–Deyo comorbidity score | ||||
| 0 | 444 (77.9) | 139 (79.9) | 0.577 | 583 (78.4) |
| ≥ 1 | 126 (22.1) | 35 (20.1) | 161 (21.6) | |
| Primary site | ||||
| Glottis | 520 (91.2) | 159 (91.4) | 0.882 | 679 (91.3) |
| Supraglottis | 23 (4.0) | 8 (4.6) | 31 (4.2) | |
| Other | 27 (4.7) | 7 (4.0) | 34 (4.6) | |
| pT classification | ||||
| Is | 570 (100) | 47 (33.1) | < 0.001 | 617 (86.7) |
| pT1 | 0 (0) | 95 (66.9) | 95 (13.3) | |
| Surgical margins | ||||
| Negative | 570 (100) | 50 (35.3) | < 0.001 | 620 (87.2) |
| Positive | 0 (0) | 91 (64.5) | 91 (12.8) | |
| RT | ||||
| No | 496 (88.3) | 124 (74.3) | < 0.001 | 620 (85.0) |
| Yes | 66 (11.7) | 43 (25.7) | 109 (15.0) | |
Note: Bold values indicate statistical significance p < 0.05.
Abbreviations: IQR, interquartile range; is, in situ; pT, pathologic tumor; RT, radiotherapy.
On Kaplan–Meier analysis, among 1979 patients undergoing primary surgery, patients with invasive or residual disease (i.e., pT1 and/or PSM) had similar 5‐year OS as those without invasive or residual disease (84% vs. 88%, p = 0.057) (Figure S1). On multivariable Cox regression, age at diagnosis (aHR: 1.08, 95% CI: 1.07–1.09, p < 0.001) and CDCS ≥ 1 (aHR: 1.47, 95% CI: 1.20–1.81, p < 0.001) were associated with worse OS. Race, primary site, pT classification, surgical margins, and aRT were not associated with OS (p > 0.05) (Table 5).
TABLE 5.
Univariable and multivariable Cox proportional hazard regression models for 1979 patients undergoing surgery.
| Univariable | Multivariable | |||
|---|---|---|---|---|
| HR (95% CI) | p | aHR a (95% CI) | p | |
| Age at diagnosis, years | 1.08 (1.07–1.09) | < 0.001 | 1.08 (1.07–1.09) | < 0.001 |
| Sex | ||||
| Male | Ref | Ref | ||
| Female | 0.67 (0.52–0.87) | 0.002 | 0.82 (0.63–1.05) | 0.116 |
| Race | ||||
| White | Ref | |||
| Black | 0.88 (0.63–1.24) | 0.474 | ||
| Other | 0.61 (0.27–1.37) | 0.228 | ||
| Facility type | ||||
| Academic/research | Ref | Ref | ||
| Non‐academic/research | 0.78 (0.65–0.95) | 0.014 | 0.86 (0.71–1.04) | 0.119 |
| Charlson–Deyo comorbidity score | ||||
| 0 | Ref | |||
| ≥ 1 | 1.58 (1.29–1.95) | < 0.001 | 1.47 (1.20–1.81) | < 0.001 |
| Primary site | ||||
| Glottis | Ref | |||
| Supraglottis | 0.93 (0.54–1.58) | 0.777 | ||
| Other | 1.33 (0.93–1.91) | 0.121 | ||
| pT classification | ||||
| Is | Ref | |||
| pT1 | 1.02 (0.62–1.67) | 0.940 | ||
| Surgical margins | ||||
| Negative | Ref | |||
| Positive | 1.47 (0.96–2.26) | 0.079 | ||
| Adjuvant radiotherapy | ||||
| No | Ref | |||
| Yes | 1.24 (0.99–1.55) | 0.062 | ||
Note: Bold values indicate statistical significance p < 0.05.
Abbreviations: aHR, adjusted hazard ratio; aRT, adjuvant radiotherapy; HR, hazard ratio; Ref, reference.
N = 1906; number of uncensored deaths: 437.
4. Discussion
Optimal management of laryngeal Cis is controversial, with clinical surveillance, primary radiotherapy, and primary surgery, all representing reasonable strategies depending on disease presentation and patient and physician preferences. There is a paucity of literature examining long‐term survival trends in a large cohort of patients with laryngeal Cis. In our study of the NCDB, we investigate factors associated with the choice of treatment, determine the rates of invasive or residual disease identified after primary surgery, and compare 5‐year OS in patients with an initial diagnosis of laryngeal Cis. Among patients undergoing primary surgery, 15.9% were found to have invasive or residual disease (defined as pT1+ or PSM) on pathologic examination. This observed rate is comparable with rates of occult invasiveness or residual disease for Cis in other diseases, including oral cavity squamous cell carcinoma (OCSCC) (28.0%), breast ductal carcinoma (25.9%), non‐small cell lung cancer (49.3%), and cutaneous squamous cell carcinoma (31%) [12, 13, 14, 15]. Patients presenting with and without occult invasiveness or residual disease had similar demographic and clinicopathologic factors, with no significant differences noted between the two cohorts.
Our results suggest that surgery is associated with higher OS than clinical surveillance or primary radiotherapy. Most existing studies on laryngeal Cis consist of small case reports and series of patients undergoing primary surgery [11, 16, 17, 18, 19] or radiotherapy alone [9, 20, 21, 22, 23, 24, 25]. In a review of 82 cases, Le et al. [8] found that patients with glottic Cis had comparable locoregional control whether treated with surgery (e.g., vocal cord stripping, laser excision, cordectomy, hemilaryngectomy, or total laryngectomy) after diagnostic biopsy, or with radiotherapy following biopsy or a single stripping procedure. However, radiotherapy was associated with fewer local relapses [8]. Based on these findings, the authors recommended an individualized approach to initial treatment, considering factors such as patient age, reliability, tumor extent, and pre‐ and post‐treatment voice quality [8]. However, the NCDB does not capture the same level of granular detail regarding treatment sequencing and specific surgical techniques, limiting our ability to make similar group assignments. Nevertheless, our study builds upon the existing literature with a larger cohort and a subsequent analysis of patients undergoing primary surgery stratified by findings on pathological examination. Although invasiveness can only be determined after surgery, our results suggest that there may be a survival benefit in patients who received primary surgery regardless of whether pathological examination reveals occult invasion or has positive margins. Although clinical surveillance may be considered, several findings support the benefit of primary surgery. Prior studies have reported a 30.4% malignant transformation rate for severe dysplasia or Cis, and our study found occult invasiveness or residual disease in over 20% of cases [5]. Additionally, we observed survival disparities favoring surgical management. Of note, older age and CDCS were associated with worse OS among patients undergoing primary surgery. This finding aligns with the understanding that advanced age and the presence of multiple comorbidities can lead to reduced physiological reserve and poorer tolerance to surgical interventions and possible associated complications [26].
Our study found that undergoing treatment at an academic facility was associated with undergoing primary surgery. This could be attributed to the access that academic facilities have to otolaryngology subspecialists (e.g., laryngologists and head and neck oncologic surgeons), minimally invasive surgical techniques and equipment, and multidisciplinary tumor boards, which may drive differences in outcomes. On the other hand, Black race was associated with reduced odds of undergoing primary surgery. Previous studies have demonstrated that Black patients often face barriers such as limited access to specialized care, delays in treatment, and socioeconomic challenges, contributing to poorer outcomes in head and neck cancer [27, 28]. These disparities emphasize the need for targeted interventions to address healthcare inequities and ensure more equitable treatment access within this patient population. Rates of primary surgery were also highest among younger patients aged 18–39 years, possibly reflecting a preference among providers to avoid radiotherapy in this age group due to concerns about treatment‐related toxicity or the potential for secondary malignancies.
A similar study using the NCDB compared OS following surgical and non‐surgical management of cTis OCSCC, which identified a 28% rate of occult invasiveness or residual disease, defined as having a final AJCC classification of pT1+, pN1+ classification, or PSM [14]. Factors associated with invasiveness included female sex, Black race, and alveolar ridge, vestibule, and retromolar primary sites [14]. Our study did not associate occult invasiveness or residual disease with a specific larynx subsite, possibly related to similarities in surgical visualization between the glottis and supraglottis and the rarity of subglottic tumors which limits any predisposition for impaired detection. Although our study did not identify an association between sociodemographic factors and rates of occult invasiveness or residual disease, it remains important to acknowledge disparities in head and neck cancer management and outcomes [29, 30].
The NCDB does not provide specific details regarding biopsy; the NCDB categorizes excisional biopsy as surgery, and these patients were considered to have undergone primary surgery. Patients undergoing clinical surveillance, therefore, presumably underwent a simple representative biopsy, but the NCDB does not capture whether the entirety of visible disease was removed. Vocal fold biopsies are often plagued with sampling error when excision biopsy is not performed, and this error is magnified in large databases that fail to capture specific operative details (i.e., operative reports). In fact, the NCDB does not differentiate severe dysplasia, which may be grouped with laryngeal Cis in some classifications. Occult invasiveness and residual disease, therefore, may be an artifact of the initial biopsy that insufficiently sampled the specimen; the initial disease having pT1+ staging rather than occult invasiveness or progression of Cis on final pathology remains a possibility.
The NCDB lacks data on important clinicopathologic factors, including depth of invasion, extent and location of disease as determined by laryngoscopy, and diagnostic imaging—all of which may help identify tumors at high risk for occult malignancy. Although patients in our cohort had high CDCS scores, and CDCS was associated with OS, the NCDB does not capture data on tumor progression, re‐resection, locoregional recurrence, disease‐free survival, or treatment‐related complications and toxicities, limiting our analysis to OS [15]. Although our study controlled for various demographic and clinicopathologic factors on multivariable regression, there are likely uncontrolled confounders such as specific medical comorbidities and history of alcohol or tobacco use that have prognostic significance. Regarding the determination of treatment cohorts, the NCDB does not reliably indicate the timing of treatment relative to diagnosis, particularly among patients undergoing clinical surveillance; for example, patients deciding to undergo clinical surveillance, but then electing to undergo surgery based on evolving preferences and physician recommendations, were categorized as undergoing primary surgery. In addition, the NCDB only reports the most definitive surgical procedure to the primary site, and patients undergoing multiple surgical procedures are therefore not differentiated; “primary surgery” includes a broad range of procedures that are highly variable and differ in terms of aggressiveness and outcomes (e.g., voice preservation and survival). Future research should investigate patterns in management and outcomes among these distinct clinical entities. Additionally, the NCDB does not account for nuances of tumor presentation, such as anterior commissure involvement, bilateral vocal cord involvement, or decreased vocal cord mobility, which may influence treatment decisions and limit the ability to determine the optimal treatment approach for individual patients [8, 31, 32]. Given the high rates of surgical re‐resection in laryngeal Cis and the potential need for salvage surgery, our study may not capture the full spectrum of treatment combinations among patients undergoing surgery. The inclusion of many medical centers in the NCDB with varying practice patterns and lack of uniformity in reporting makes it difficult to distinguish the clinical surveillance and primary surgery cohorts, introducing potential geographic bias and sampling variability. At last, our retrospective study design limits any interpretation of data to correlation, not causation, and limits the applicability of our data in determining which initial treatment would be best for a patient with a diagnosis of laryngeal Cis. Despite these limitations, our study associated primary surgery with higher OS than primary radiotherapy and clinical surveillance and identified a 15.9% rate of invasive or residual disease among patients undergoing primary surgery.
5. Conclusion
Our study identifies a high rate of occult invasiveness or residual disease of 15.9% among patients with biopsy‐proven laryngeal Cis and associates that primary surgery with higher OS than clinical surveillance or primary radiotherapy. Undergoing treatment at an academic facility was associated with undergoing primary surgery. Among patients undergoing primary surgery, OS was not impacted by pT classification or margin status. These findings suggest that primary surgery should be given strong consideration in appropriately selected patients.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Figure S1: Kaplan–Meier analysis of 5‐year overall survival among 570 patients undergoing surgery and having noninvasive disease, and 174 undergoing surgery and having invasive disease (p = 0.057). Significance is derived from the log–rank test.
Table S1: Types and frequencies of primary surgeries.
Kaki P. C., Patel A. M., Huang L., et al., “Treatment of Biopsy‐Proven Laryngeal Squamous Cell Carcinoma In Situ,” Laryngoscope Investigative Otolaryngology 10, no. 5 (2025): e70267, 10.1002/lio2.70267.
Funding: The authors received no specific funding for this work.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Hellquist H., Lundgren J., and Olofsson J., “Hyperplasia, Keratosis, Dysplasia and Carcinoma In Situ of the Vocal Cords—A Follow‐Up Study,” Clinical Otolaryngology and Allied Sciences 7, no. 1 (1982): 11–27, 10.1111/j.1365-2273.1982.tb01557.x. [DOI] [PubMed] [Google Scholar]
- 2. Hintz B. L., Kagan A. R., Nussbaum H., Rao A. R., Chan P. Y., and Miles J., “A ‘Watchful Waiting’ Policy for In Situ Carcinoma of the Vocal Cords,” Archives of Otolaryngology 107, no. 12 (1981): 746–751, 10.1001/archotol.1981.00790480022006. [DOI] [PubMed] [Google Scholar]
- 3. Nguyen C., Naghibzadeh B., Black M. J., Rochon L., and Shenouda G., “Carcinoma In Situ of the Glottic Larynx: Excision or Irradiation?,” Head & Neck 18, no. 3 (1996): 225–228, . [DOI] [PubMed] [Google Scholar]
- 4. Stenersen T. C., Hoel P. S., and Boysen M., “Carcinoma In Situ of the Larynx: An Evaluation of Its Natural Clinical Course,” Clinical Otolaryngology and Allied Sciences 16, no. 4 (1991): 358–363, 10.1111/j.1365-2273.1991.tb00948.x. [DOI] [PubMed] [Google Scholar]
- 5. Weller M. D., Nankivell P. C., McConkey C., Paleri V., and Mehanna H. M., “The Risk and Interval to Malignancy of Patients With Laryngeal Dysplasia; a Systematic Review of Case Series and Meta‐Analysis,” Clinical Otolaryngology 35, no. 5 (2010): 364–372, 10.1111/j.1749-4486.2010.02181.x. [DOI] [PubMed] [Google Scholar]
- 6. Sadri M., McMahon J., and Parker A., “Management of Laryngeal Dysplasia: A Review,” European Archives of Oto‐Rhino‐Laryngology 263, no. 9 (2006): 843–852, 10.1007/s00405-006-0078-y. [DOI] [PubMed] [Google Scholar]
- 7. Smitt M. C. and Goffinet D. R., “Radiotherapy for Carcinoma‐In‐Situ of the Glottic Larynx,” International Journal of Radiation Oncology, Biology, Physics 28, no. 1 (1994): 251–255, 10.1016/0360-3016(94)90164-3. [DOI] [PubMed] [Google Scholar]
- 8. Le Q. T., Takamiya R., Shu H. K., et al., “Treatment Results of Carcinoma In Situ of the Glottis: An Analysis of 82 Cases,” Archives of Otolaryngology – Head & Neck Surgery 126, no. 11 (2000): 1305–1312, 10.1001/archotol.126.11.1305. [DOI] [PubMed] [Google Scholar]
- 9. W. Small, Jr. , Mittal B. B., Brand W. N., et al., “Role of Radiation Therapy in the Management of Carcinoma In Situ of the Larynx,” Laryngoscope 103, no. 6 (1993): 663–667, 10.1288/00005537-199306000-00015. [DOI] [PubMed] [Google Scholar]
- 10. Spayne J. A., Warde P., O'Sullivan B., et al., “Carcinoma‐In‐Situ of the Glottic Larynx: Results of Treatment With Radiation Therapy,” International Journal of Radiation Oncology, Biology, Physics 49, no. 5 (2001): 1235–1238, 10.1016/s0360-3016(00)01517-0. [DOI] [PubMed] [Google Scholar]
- 11. Wetmore S. J., Key J. M., and Suen J. Y., “Laser Therapy for T1 Glottic Carcinoma of the Larynx,” Archives of Otolaryngology—Head & Neck Surgery 112, no. 8 (1986): 853–855, 10.1001/archotol.1986.03780080053011. [DOI] [PubMed] [Google Scholar]
- 12. Brennan M. E., Turner R. M., Ciatto S., et al., “Ductal Carcinoma In Situ at Core‐Needle Biopsy: Meta‐Analysis of Underestimation and Predictors of Invasive Breast Cancer,” Radiology 260, no. 1 (2011): 119–128, 10.1148/radiol.11102368. [DOI] [PubMed] [Google Scholar]
- 13. Chuang G. S., Lu L. K., Cummins D. L., et al., “Incidence of Invasive Squamous Cell Carcinomas in Biopsy‐Proven Squamous Cell Carcinomas In Situ Sent for Mohs Micrographic Surgery,” Dermatologic Surgery 38, no. 9 (2012): 1456–1460, 10.1111/j.1524-4725.2012.02507.x. [DOI] [PubMed] [Google Scholar]
- 14. Cooper D. J., Ziemba Y., Pereira L., et al., “Rates of Occult Invasive Disease in Patients With Biopsy‐Proven Oral Cavity Squamous Cell Carcinoma In Situ,” JAMA Otolaryngology. Head & Neck Surgery 150, no. 2 (2024): 151–156, 10.1001/jamaoto.2023.3754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Talcott W. J., Miccio J. A., Park H. S., et al., “Rates of Invasive Disease and Outcomes in NSCLC Patients With Biopsy Suggestive of Carcinoma In Situ,” Lung Cancer 157 (2021): 17–20, 10.1016/j.lungcan.2021.05.028. [DOI] [PubMed] [Google Scholar]
- 16. Annyas A. A., Van Overbeek J. J., Escajadillo J. R., and Hoeksema P. E., “CO2 Laser in Malignant Lesions of the Larynx,” Laryngoscope 94, no. 6 (1984): 836–838, 10.1288/00005537-198406000-00022. [DOI] [PubMed] [Google Scholar]
- 17. Gillis T. M., Incze J., Strong M. S., Vaughan C. W., and Simpson G. T., “Natural History and Management of Keratosis, Atypia, Carcinoma‐In Situ, and Microinvasive Cancer of the Larynx,” American Journal of Surgery 146, no. 4 (1983): 512–516, 10.1016/0002-9610(83)90243-x. [DOI] [PubMed] [Google Scholar]
- 18. Wolfensberger M. and Dort J. C., “Endoscopic Laser Surgery for Early Glottic Carcinoma: A Clinical and Experimental Study,” Laryngoscope 100, no. 10 Pt 1 (1990): 1100–1105, 10.1288/00005537-199010000-00014. [DOI] [PubMed] [Google Scholar]
- 19. Zeitels S. M., “Phonomicrosurgical Treatment of Early Glottic Cancer and Carcinoma In Situ,” American Journal of Surgery 172, no. 6 (1996): 704–709, 10.1016/s0002-9610(96)00295-4. [DOI] [PubMed] [Google Scholar]
- 20. Fein D. A., Mendenhall W. M., Parsons J. T., Stringer S. P., Cassisi N. J., and Million R. R., “Carcinoma In Situ of the Glottic Larynx: The Role of Radiotherapy,” International Journal of Radiation Oncology, Biology, Physics 27, no. 2 (1993): 379–384, 10.1016/0360-3016(93)90250-y. [DOI] [PubMed] [Google Scholar]
- 21. Harwood A. R., Hawkins N. V., Keane T., et al., “Radiotherapy of Early Glottic Cancer,” Laryngoscope 90, no. 3 (1980): 465–470, 10.1002/lary.5540900313. [DOI] [PubMed] [Google Scholar]
- 22. MacLeod P. M. and Daniel F., “The Role of Radiotherapy in In‐Situ Carcinoma of the Larynx,” International Journal of Radiation Oncology, Biology, Physics 18, no. 1 (1990): 113–117, 10.1016/0360-3016(90)90274-n. [DOI] [PubMed] [Google Scholar]
- 23. Medini E., Medini I., Lee C. K., Gapany M., and Levitt S. H., “The Role of Radiotherapy in the Management of Carcinoma In Situ of the Glottic Larynx,” American Journal of Clinical Oncology 21, no. 3 (1998): 298–301, 10.1097/00000421-199806000-00020. [DOI] [PubMed] [Google Scholar]
- 24. Pene F. and Fletcher G. H., “Results in Irradiation of the In Situ Carcinomas of the Vocal Cords,” Cancer 37, no. 6 (1976): 2586–2590, . [DOI] [PubMed] [Google Scholar]
- 25. Sung D. I., Chang C. H., Harisiadis L., and Rosenstein L. M., “Primary Radiotherapy for Carcinoma In Situ and Early Invasive Carcinoma of the Glottic Larynx,” International Journal of Radiation Oncology, Biology, Physics 5, no. 4 (1979): 467–472, 10.1016/0360-3016(79)90807-1. [DOI] [PubMed] [Google Scholar]
- 26. Partridge J. S., Harari D., and Dhesi J. K., “Frailty in the Older Surgical Patient: A Review,” Age and Ageing 41, no. 2 (2012): 142–147, 10.1093/ageing/afr182. [DOI] [PubMed] [Google Scholar]
- 27. Patel A. M., Haleem A., Choudhry H. S., Povolotskiy R., and Roden D. F., “Patterns and Trends in Adjuvant Therapy for Major Salivary Gland Cancer,” Otolaryngology and Head and Neck Surgery 171, no. 1 (2024): 155–171, 10.1002/ohn.715. [DOI] [PubMed] [Google Scholar]
- 28. Ragin C. C., Langevin S. M., Marzouk M., Grandis J., and Taioli E., “Determinants of Head and Neck Cancer Survival by Race,” Head & Neck 33, no. 8 (2011): 1092–1098, 10.1002/hed.21584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Jassal J. S. and Cramer J. D., “Explaining Racial Disparities in Surgically Treated Head and Neck Cancer,” Laryngoscope 131, no. 5 (2021): 1053–1059, 10.1002/lary.29197. [DOI] [PubMed] [Google Scholar]
- 30. Rotsides J. M., Oliver J. R., Moses L. E., et al., “Socioeconomic and Racial Disparities and Survival of Human Papillomavirus‐Associated Oropharyngeal Squamous Cell Carcinoma,” Otolaryngology and Head and Neck Surgery 164, no. 1 (2021): 131–138, 10.1177/0194599820935853. [DOI] [PubMed] [Google Scholar]
- 31. Hendriksma M. and Sjogren E. V., “Involvement of the Anterior Commissure in Early Glottic Cancer (Tis‐T2): A Review of the Literature,” Cancers (Basel) 11, no. 9 (2019): 1234, 10.3390/cancers11091234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Myssiorek D., Vambutas A., and Abramson A. L., “Carcinoma In Situ of the Glottic Larynx,” Laryngoscope 104, no. 4 (1994): 463–467, 10.1288/00005537-199404000-00011. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Kaplan–Meier analysis of 5‐year overall survival among 570 patients undergoing surgery and having noninvasive disease, and 174 undergoing surgery and having invasive disease (p = 0.057). Significance is derived from the log–rank test.
Table S1: Types and frequencies of primary surgeries.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.