The Laryngoscope C 2015 The American Laryngological, V
Rhinological and Otological Society, Inc.
Laryngeal Adenoid Cystic Carcinoma: A Population-Based Perspective Pariket M. Dubal, BA; Peter F. Svider, MD; Adam J. Folbe, MD; Ho-Sheng Lin, MD; Richard C. Park, MD; Soly Baredes, MD, FACS; Jean Anderson Eloy, MD, FACS Objectives/Hypothesis: Adenoid cystic carcinoma (ACC) occurs infrequently in the larynx. Consequently, no large samples describing its clinical behavior are available in the literature. Our objective was to use a nationally representative population-based resource to evaluate clinical behavior, patient demographics, and outcomes among patients diagnosed with laryngeal ACC (LACC). Study Design: Retrospective database analysis. Methods: The National Cancer Institute’s Surveillance, Epidemiology, and End Results database was analyzed for patients diagnosed with LACC between 1973 and 2011. Patient demographics, incidence, treatment, and survival between LACC and other laryngeal malignancies were compared. Results: Of 69 LACC patients, 63.8% were female, 78.2% Caucasian, and the median age was 54 years. LACC patients were much more likely to have subglottic lesions (44.9%) than individuals with other malignancies (1.6%). The incidence of LACC was 0.005/100,000 individuals. The majority of patients with LACC harbored T4 lesions at initial diagnosis, although 87.9% had N0 disease, and only 6.1% had distant metastasis at diagnosis. Disease-specific survival (DSS) was greater at 1 year for LACC compared to other laryngeal malignancies, but not at 5 or 10 years. Five-year DSS was greater for LACC patients who underwent surgery versus those who did not undergo surgery. Conclusions: This analysis notes that LACC has a low incidence with no significant change in incidence over the study period. Compared to other laryngeal malignancies, LACC has a female preponderance, is much more common in the subglottis, presents at a younger age, and more often presents with T4 disease. Surgery was noted to confer a survival advantage in LACC. Key Words: Adenoid cystic carcinoma, laryngeal cancer, Surveillance, Epidemiology, and End Results, population-based, laryngeal adenoid cystic carcinoma, survival, disease-specific survival, incidence, laryngeal cancer. Level of Evidence: 4. Laryngoscope, 125:2485–2490, 2015
INTRODUCTION Adenoid cystic carcinoma (ACC) of the head and neck is most frequently found in the minor salivary glands. Although the histological architecture appears relatively benign, these slow- growing lesions can be aggressive, often demonstrating perineural spread, and are notable for late recurrence and distant metastasis.1 From the Department of Otolaryngology–Head and Neck Surgery (P.M.D., R.C.P., S.B., J.A.E.), Rutgers New Jersey Medical School, Newark, New Jersey; Department of Otolaryngology–Head and Neck Surgery (P.F.S., A.J.F., H.-S.L.), Wayne State University School of Medicine, Detroit, Michigan; Section of Otolaryngology, Department of Surgery (H.-S.L.), John D. Dingell VA Medical Center, Detroit, Michigan; Center for Skull Base and Pituitary Surgery (S.B., J.A.E.), Neurological Institute of New Jersey, Rutgers New Jersey Medical School, Newark, New Jersey; Department of Neurological Surgery J.A.E., Rutgers New Jersey Medical School, Newark, New Jersey; Department of Ophthalmology and Visual Science J.A.E., Rutgers New Jersey Medical School, Newark, New Jersey, U.S.A. Editor’s Note: This Manuscript was accepted for publication May 24, 2015. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Jean Anderson Eloy, MD, Department of Otolaryngology–Head and Neck Surgery, Neurological Institute of New Jersey, Rutgers New Jersey Medical School, 90 Bergen Street, Suite 8100, Newark, NJ 07103. E-mail: [email protected] DOI: 10.1002/lary.25448
Laryngoscope 125: November 2015
In contrast to many other malignancies, long-term survival rates do not stabilize at 5 years.2–11 Although common in the minor salivary glands, ACC of the larynx is a rare entity. Laryngeal ACC (LACC) arises from minor salivary glands and glandular elements in the larynx,12 and the characterization of its histology and behavior in the literature is comprised primarily of case reports and small case series. Although much is known about the clinical behavior of extralaryngeal ACC, current analyses available in the literature on LACC have limited usefulness due to small sample size and statistical power. For example, in one review of 1,342 laryngeal neoplasms, only five cases were identified as LACC.13,14 In contrast to squamous cell carcinoma (SCC), the most common laryngeal malignancy, LACC has not been demonstrated to be associated with tobacco abuse.14,15 Despite this difference in etiology, histologic examination of ACC demonstrates a close resemblance to basaloid SCC, a recognized variant of traditional SCC.16,17 Under the microscope, ACC appears as bundles of small cells arranged in groups of varying sizes, with a tendency to form glandular spaces filled with mucoid or hyaline material.18 Similarly, basaloid SCC appears as densely packed groups of cells alongside cystic spaces19 and, depending on the size of the biopsy, may demonstrate Dubal et al.: Laryngeal Adenoid Cystic Carcinoma
2485
obvious squamous differentiation with peripheral palisading and thick basement membranes.20 However, in small sections, the areas indicating a squamous component may be absent, making differentiation between ACC and basaloid SCC difficult or impossible. The distinction, however, can be made by immunohistochemistry.19–21 The differentiation is paramount to optimal management of these entities, as prognosis is much more favorable for ACC in comparison to basaloid SCC.17 Because of the rarity of LACC, a population-based resource is valuable in learning more about its clinical behavior. In prior analyses, the Surveillance, Epidemiology, and End Results (SEER) database has had remarkable utility in characterizing uncommon clinical entities,22–27 as this nationally representative resource allows for evaluating trends from a population-based perspective. Our objective was to use the SEER database, which represents approximately a quarter of the national population, to evaluate trends in LACC, including incidence, management, outcomes, and survival. Information from this analysis can be useful in determining prognosis and therapeutic strategies, thus facilitating comprehensive discussion with patients and among practitioners.
MATERIALS AND METHODS The SEER database was examined for survival and incidence data for LACC diagnosed between 1973 and 2011. Given that no patient identifiers were used, institutional review board (IRB) approval was not necessary per the guidelines of the IRB of Rutgers New Jersey Medical School (Newark, NJ). We searched the dataset for the International Statistical Classification of Diseases for Oncology, 3rd Edition (ICD-O-3) histology code corresponding to ACC (8200/3), and then limited the results to identify cases arising in subsites of the larynx (ICDO-3 site codes: 32.0–32.3, 32.8, and 32.9). Data were further organized by patient demographics, including age, gender, and race. The data were also organized by laryngeal subsites and tumor staging (American Joint Committee on Cancer staging system). Incidence data were obtained by searching SEER for LACC and other laryngeal malignancies rates per 100,000 individuals from 2000 to 2010. All incidence data were age adjusted to the standard 2000 United States population per Census P25–1130. Disease-specific survival (DSS) and relative survival (RS) were used for description of survival trends. The former takes into account cases in which the malignancy is considered to have been the cause of death, whereas RS is a ratio between observed survival and expected age-adjusted survival for a comparable segment of the population. To determine DSS, we exported individual case listings with comprehensive data regarding dates of diagnosis, death, and loss to follow-up into Microsoft Excel 2013 (Microsoft Corp., Redmond, WA). Data were reorganized and assigned binary survival codes (with 0 corresponding to disease-specific deaths and 1 corresponding to survival or deaths due to other causes). The resultant data were imported into JMP Statistical Discovery (SAS Institute, Cary, NC) for Kaplan-Meier analysis to yield DSS rates. RS rates, on the other hand, were available in the SEER database. For comparison with LACC patients, we also collected data on all other malignant laryngeal histologies. The laryngeal
Laryngoscope 125: November 2015
2486
subsite codes detailed above were used to search for all cases with malignant behavior found in the SEER database, and LACC cases were removed from this control cohort. These are referred to as “other laryngeal malignancies” in this analysis and represent an average of non-ACC malignancies of the larynx.
Statistical Analysis The SEER*Stat 8.1.5 (National Cancer Institute, Bethesda, MD) program was used to extract data from the SEER 18 database. Two-proportion Z tests with Power Analysis and Sample Software 13 (NCSS Statistical Software, Silver Spring, MD) was used to compare LACC and other laryngeal malignancies’ demographic, treatment modalities, primary site, and staging characteristics. The annual percent change (APC) in incidence was determined using weighted least squares and 1-year endpoints with the SEER*Stat 8.1.5 software. JMP Statistical Discovery was used to carry out Kaplan-Meier analysis to yield DSS rates as outlined above. Log-rank analysis was used to determine statistical significance for DSS. Threshold for significance was set at probability values (P values) below .05.
RESULTS Tumor Characteristics Utilizing the ICD-O-3 site codes, laryngeal subsites were evaluated for LACC and other laryngeal malignancies. When compared to other laryngeal malignancies, a significantly greater proportion of LACC originated from the subglottis (44.9% vs. 1.6%, P
Rhinological and Otological Society, Inc.
Laryngeal Adenoid Cystic Carcinoma: A Population-Based Perspective Pariket M. Dubal, BA; Peter F. Svider, MD; Adam J. Folbe, MD; Ho-Sheng Lin, MD; Richard C. Park, MD; Soly Baredes, MD, FACS; Jean Anderson Eloy, MD, FACS Objectives/Hypothesis: Adenoid cystic carcinoma (ACC) occurs infrequently in the larynx. Consequently, no large samples describing its clinical behavior are available in the literature. Our objective was to use a nationally representative population-based resource to evaluate clinical behavior, patient demographics, and outcomes among patients diagnosed with laryngeal ACC (LACC). Study Design: Retrospective database analysis. Methods: The National Cancer Institute’s Surveillance, Epidemiology, and End Results database was analyzed for patients diagnosed with LACC between 1973 and 2011. Patient demographics, incidence, treatment, and survival between LACC and other laryngeal malignancies were compared. Results: Of 69 LACC patients, 63.8% were female, 78.2% Caucasian, and the median age was 54 years. LACC patients were much more likely to have subglottic lesions (44.9%) than individuals with other malignancies (1.6%). The incidence of LACC was 0.005/100,000 individuals. The majority of patients with LACC harbored T4 lesions at initial diagnosis, although 87.9% had N0 disease, and only 6.1% had distant metastasis at diagnosis. Disease-specific survival (DSS) was greater at 1 year for LACC compared to other laryngeal malignancies, but not at 5 or 10 years. Five-year DSS was greater for LACC patients who underwent surgery versus those who did not undergo surgery. Conclusions: This analysis notes that LACC has a low incidence with no significant change in incidence over the study period. Compared to other laryngeal malignancies, LACC has a female preponderance, is much more common in the subglottis, presents at a younger age, and more often presents with T4 disease. Surgery was noted to confer a survival advantage in LACC. Key Words: Adenoid cystic carcinoma, laryngeal cancer, Surveillance, Epidemiology, and End Results, population-based, laryngeal adenoid cystic carcinoma, survival, disease-specific survival, incidence, laryngeal cancer. Level of Evidence: 4. Laryngoscope, 125:2485–2490, 2015
INTRODUCTION Adenoid cystic carcinoma (ACC) of the head and neck is most frequently found in the minor salivary glands. Although the histological architecture appears relatively benign, these slow- growing lesions can be aggressive, often demonstrating perineural spread, and are notable for late recurrence and distant metastasis.1 From the Department of Otolaryngology–Head and Neck Surgery (P.M.D., R.C.P., S.B., J.A.E.), Rutgers New Jersey Medical School, Newark, New Jersey; Department of Otolaryngology–Head and Neck Surgery (P.F.S., A.J.F., H.-S.L.), Wayne State University School of Medicine, Detroit, Michigan; Section of Otolaryngology, Department of Surgery (H.-S.L.), John D. Dingell VA Medical Center, Detroit, Michigan; Center for Skull Base and Pituitary Surgery (S.B., J.A.E.), Neurological Institute of New Jersey, Rutgers New Jersey Medical School, Newark, New Jersey; Department of Neurological Surgery J.A.E., Rutgers New Jersey Medical School, Newark, New Jersey; Department of Ophthalmology and Visual Science J.A.E., Rutgers New Jersey Medical School, Newark, New Jersey, U.S.A. Editor’s Note: This Manuscript was accepted for publication May 24, 2015. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Jean Anderson Eloy, MD, Department of Otolaryngology–Head and Neck Surgery, Neurological Institute of New Jersey, Rutgers New Jersey Medical School, 90 Bergen Street, Suite 8100, Newark, NJ 07103. E-mail: [email protected] DOI: 10.1002/lary.25448
Laryngoscope 125: November 2015
In contrast to many other malignancies, long-term survival rates do not stabilize at 5 years.2–11 Although common in the minor salivary glands, ACC of the larynx is a rare entity. Laryngeal ACC (LACC) arises from minor salivary glands and glandular elements in the larynx,12 and the characterization of its histology and behavior in the literature is comprised primarily of case reports and small case series. Although much is known about the clinical behavior of extralaryngeal ACC, current analyses available in the literature on LACC have limited usefulness due to small sample size and statistical power. For example, in one review of 1,342 laryngeal neoplasms, only five cases were identified as LACC.13,14 In contrast to squamous cell carcinoma (SCC), the most common laryngeal malignancy, LACC has not been demonstrated to be associated with tobacco abuse.14,15 Despite this difference in etiology, histologic examination of ACC demonstrates a close resemblance to basaloid SCC, a recognized variant of traditional SCC.16,17 Under the microscope, ACC appears as bundles of small cells arranged in groups of varying sizes, with a tendency to form glandular spaces filled with mucoid or hyaline material.18 Similarly, basaloid SCC appears as densely packed groups of cells alongside cystic spaces19 and, depending on the size of the biopsy, may demonstrate Dubal et al.: Laryngeal Adenoid Cystic Carcinoma
2485
obvious squamous differentiation with peripheral palisading and thick basement membranes.20 However, in small sections, the areas indicating a squamous component may be absent, making differentiation between ACC and basaloid SCC difficult or impossible. The distinction, however, can be made by immunohistochemistry.19–21 The differentiation is paramount to optimal management of these entities, as prognosis is much more favorable for ACC in comparison to basaloid SCC.17 Because of the rarity of LACC, a population-based resource is valuable in learning more about its clinical behavior. In prior analyses, the Surveillance, Epidemiology, and End Results (SEER) database has had remarkable utility in characterizing uncommon clinical entities,22–27 as this nationally representative resource allows for evaluating trends from a population-based perspective. Our objective was to use the SEER database, which represents approximately a quarter of the national population, to evaluate trends in LACC, including incidence, management, outcomes, and survival. Information from this analysis can be useful in determining prognosis and therapeutic strategies, thus facilitating comprehensive discussion with patients and among practitioners.
MATERIALS AND METHODS The SEER database was examined for survival and incidence data for LACC diagnosed between 1973 and 2011. Given that no patient identifiers were used, institutional review board (IRB) approval was not necessary per the guidelines of the IRB of Rutgers New Jersey Medical School (Newark, NJ). We searched the dataset for the International Statistical Classification of Diseases for Oncology, 3rd Edition (ICD-O-3) histology code corresponding to ACC (8200/3), and then limited the results to identify cases arising in subsites of the larynx (ICDO-3 site codes: 32.0–32.3, 32.8, and 32.9). Data were further organized by patient demographics, including age, gender, and race. The data were also organized by laryngeal subsites and tumor staging (American Joint Committee on Cancer staging system). Incidence data were obtained by searching SEER for LACC and other laryngeal malignancies rates per 100,000 individuals from 2000 to 2010. All incidence data were age adjusted to the standard 2000 United States population per Census P25–1130. Disease-specific survival (DSS) and relative survival (RS) were used for description of survival trends. The former takes into account cases in which the malignancy is considered to have been the cause of death, whereas RS is a ratio between observed survival and expected age-adjusted survival for a comparable segment of the population. To determine DSS, we exported individual case listings with comprehensive data regarding dates of diagnosis, death, and loss to follow-up into Microsoft Excel 2013 (Microsoft Corp., Redmond, WA). Data were reorganized and assigned binary survival codes (with 0 corresponding to disease-specific deaths and 1 corresponding to survival or deaths due to other causes). The resultant data were imported into JMP Statistical Discovery (SAS Institute, Cary, NC) for Kaplan-Meier analysis to yield DSS rates. RS rates, on the other hand, were available in the SEER database. For comparison with LACC patients, we also collected data on all other malignant laryngeal histologies. The laryngeal
Laryngoscope 125: November 2015
2486
subsite codes detailed above were used to search for all cases with malignant behavior found in the SEER database, and LACC cases were removed from this control cohort. These are referred to as “other laryngeal malignancies” in this analysis and represent an average of non-ACC malignancies of the larynx.
Statistical Analysis The SEER*Stat 8.1.5 (National Cancer Institute, Bethesda, MD) program was used to extract data from the SEER 18 database. Two-proportion Z tests with Power Analysis and Sample Software 13 (NCSS Statistical Software, Silver Spring, MD) was used to compare LACC and other laryngeal malignancies’ demographic, treatment modalities, primary site, and staging characteristics. The annual percent change (APC) in incidence was determined using weighted least squares and 1-year endpoints with the SEER*Stat 8.1.5 software. JMP Statistical Discovery was used to carry out Kaplan-Meier analysis to yield DSS rates as outlined above. Log-rank analysis was used to determine statistical significance for DSS. Threshold for significance was set at probability values (P values) below .05.
RESULTS Tumor Characteristics Utilizing the ICD-O-3 site codes, laryngeal subsites were evaluated for LACC and other laryngeal malignancies. When compared to other laryngeal malignancies, a significantly greater proportion of LACC originated from the subglottis (44.9% vs. 1.6%, P
