Abstract
Objective:
Management of tracheal squamous cell carcinoma (TSCC) has been complicated by the lack of prognostic data and staging. We describe the epidemiology of TSCC and current treatment approaches.
Methods:
Five hundred thirty-two adult patients with primary TSCC from 2004 to 2012 in the National Cancer Database were identified. Demographic, clinical factors, and 5-year overall survival were analyzed. Staging was classified as localized, regional extension, and distant spread. Treatment modality was defined as “no treatment (NT),” “limited surgery (LS),” “curative surgery (CS),” “LS with any adjuvant therapy (AT) (LS+AT),” “CS with AT (CS+AT),” “radiation therapy (RT),” or “chemoradiation (CRT).”
Results:
Overall survival was 25%. Majority of cases were males, white, and occurred in sixth/seventh decades. Twenty-six percent of cases received CRT, 20% underwent LS+AT or CS+AT, 20% underwent LS or CS only, and 17% underwent RT alone. On multivariate analysis, CS (HR 0.42, 95% CI: 0.26–0.69), CS+AT (HR 0.44, 95% CI: 0.36–0.77), CRT (HR 0.48, 95% CI: 0.35–0.67), and RT (HR, 0.66 95% CI: 0.46–0.94) were associated with decreased likelihood of death compared to NT. Elderly patients and those with poor performance status had worse outcomes even on multivariate analysis.
Conclusions:
TSCC is increasingly treated with surgery and systemic therapy in addition to RT, with improved survival outcomes. CS, CS+AT, CRT, or RT provided improved survival advantage in patients with variable levels of improvement based on the extent of the disease. Prospective trials would help differentiate survival advantages between treatment modalities. Patients’ goals of care, comorbidities, and age should be considered when deciding appropriate treatment recommendations.
Keywords: Tracheal carcinoma, thoracic diseases, head and neck cancer, survival analysis, squamous cell carcinoma
INTRODUCTION
Squamous cell carcinoma (SCC) is the predominant histology of primary tracheal carcinomas.1,2 Altogether, tracheal malignancies are exceedingly rare, deadly, and debilitating diseases with a variable incidence of approximately 2.6 new cases per 1 million people per year and represent 0.1% to 0.4% of all newly diagnosed cancers.1–3 The reported 5-year overall survival (OS) rates are as low as 13% to 30%, rates that largely encompass all histolo-gies.3,4 The exact incidence and prognosis of tracheal squamous cell carcinoma (TSCC) are unknown. Much of our limited understanding about TSCC’s natural history and treatment is based on case reports and series.1–3,5–7 With the lack of patient-specific prognostication, the optimal treatment approach remains open-ended and largely anecdotal.
Given the paucity of understanding surrounding this malignancy, we described the epidemiology and analyzed various treatment approaches for TSCC using the National Cancer Database (NCDB). Our goal was to provide methods for assessing prognosis and evidence-based treatment strategies for surgeons, radiation oncologists, and medical oncologists and reduce institutional and provider-based subjectivity in making clinical judgments regarding treatment.
MATERIALS AND METHODS
Data Source and Patient Selection
The NCDB, a joint project of the American College of Surgeons Commission on Cancer and the American Cancer Society, is a hospital-based registry capturing approximately 70% of incident cancer cases in the United States. The current analysis was performed with the approval of our local institutional review board. We queried the NCDB for all primary TSCC patients aged ≥18 years diagnosed from 2004 through 2012 with 5-year follow-up data available. Tumor size and regional spread(T), and presence distant metastasis(M) were used to define the anatomic extent of the cancer, and patients without complete information were excluded. Clinical regional lymph node spread was not available and thus not used in this analysis. T-stage was defined as T1 as <20 mm, T2 as 20 to 39 mm, T3 as ≥40 mm, and T4 as advanced local disease with invasion into adjacent connective tissue or organs. Stages were defined based on anatomical knowledge and statistical discrimination of survival curves.
The primary independent variable was treatment modality defined as “no treatment (NT),” “limited surgery (LS),” “curative surgery (CS),” “limited surgery with adjuvant therapy (AT) (LS+AT),” “curative surgery with AT (CS+AT),” “radiation therapy (RT) only,” or “chemoradiation therapy (CRT) only.” Limited surgery was defined as local tumor excision or tumor debulking (surgical codes 10–29). Curative surgery was defined as procedures with curative intent such as wide local excision, simple/partial tracheal removal, total tracheal removal, or other (surgical codes 30–90). Patients were defined as receiving primary or adjuvant RT if they underwent RT within 6 months of diagnosis or surgery. Patients who had a known chemotherapy start date within 30 days before or after RT start date were defined as receiving CRT. Demographic information including gender, age, race, insurance, median zip code household income, proximity to a metropolitan area, and geographical region was compiled for analysis. Clinical variables including comorbidities based on the Charlson-Deyo score, tumor grade, the type of institution where they received treatment, and whether treatment was palliative were also evaluated.
Statistical Analysis
All statistical analyses were performed with Stata 13.1 (StataCorp LP, College Station, TX). Overall survival was the primary endpoint and was defined as the time of diagnosis until the time of death or censored at 5 years. Pearson chi-square tests were used to assess associations between categorical variables and treatment modality. Kaplan-Meier (KM) survivals were generated and compared using the log-rank method. Univariate (UVA) and multivariate (MVA) analysis for OS were performed with Cox proportional hazards regression to determine hazard ratios (HR), utilizing the Breslow method for ties, with an HR >1 corresponding to worse OS. Model predictors were chosen a priori and if P values were ≤0.25. Tests were two-sided with a level of significance of less than 0.05.
RESULTS
We identified 532 patients with primary TSCC in the NCDB with a comparison of demographic, clinical, and treatment variables presented in Table I. The majority of the cases occurred in the sixth and seventh decades of life. Almost all patients were 40 years or older at time of diagnosis (98%). Men were affected two times more than women. Most patients were largely white with limited comorbidities. The highest proportion of cases were from the Southern states, and the majority of patients were treated at community medical centers. There was no apparent association to socioeconomic status, but insurance varied largely between private and Medicare. More than half of patients presented with regional or distant spread of disease at time of diagnosis. Tumors largely consisted of moderately to poorly differentiated grades.
TABLE I.
Patient Demographic and Clinical Characteristics by Treatment Modality (n = 532).
| No Treatment (n = 6) No (%) |
Limited Surgery (n = 59) No (%) |
Curative Surgery (n = 48) No (%) |
Limited Surgery + AT (n = 70) No (%) |
Curative Surgery + AT (n=38) No (%) |
RT only (n = 92) No (%) |
CRT on (n = 139) No (%) |
Total (n = 532) No (%) |
p value | |
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| Sex | <0.01 | ||||||||
| Male | 53 (61.6) | 26 (44.1) | 32 (66.7) | 40 (57.1) | 29 (76.3) | 66 (71.7) | 95 (68.3) | 341 (64.1) | |
| Female | 33 (38.4) | 33 (55.9) | 16(33.3) | 30 (42.9) | 9 (23.7) | 26 (28.3) | 44 (31.7) | 191 (35.9) | |
| Age | 0.37 | ||||||||
| <50 years | 10 (11.6) | 4 (6.8) | 7(14.6) | 5(7.1) | 7(18.4) | 7 (7.6) | 12 (8.6) | 52 (9.8) | |
| 50–59 years | 18 (20.9) | 13 (22) | 12 (25) | 21 (30) | 13(34.2) | 19(20.7) | 42 (30.2) | 138(25.9) | |
| 60–69 years | 21 (24.4) | 20 (33.9) | 12 (25) | 24 (34.3) | 8(21.1) | 28 (30.4) | 44 (31.7) | 157(29.5) | |
| ≥70 years | 37 (43) | 22 (37.3) | 17(35.4) | 20 (28.6) | 10(26.3) | 38 (41.3) | 41 (29.5) | 185(34.8) | |
| Race | 0.39 | ||||||||
| White | 64 (74.4) | 45 (76.3) | 37(77.1) | 56 (80) | 27(71.1) | 76 (82.6) | 115 (82.7) | 420 (78.9) | |
| Black | 12 (14) | 5 (8.5) | 6(12.5) | 11 (15.7) | 7(18.4) | 11 (12) | 17 (12.2) | 69 (13) | |
| Other | 10 (11.6) | 9(15.3) | 5(10.4) | 3 (4.3) | 4(10.5) | 5 (5.4) | 7 (5) | 43 (8.1) | |
| Insurance | 0.03 | ||||||||
| Private | 23 (26.7) | 19 (32.2) | 15(31.3) | 23 (32.9) | 16(42.1) | 17(18.5) | 54 (38.8) | 167(31.4) | |
| Other | 63 (73.3) | 40 (67.8) | 33 (68.8) | 47 (67.1) | 22 (57.9) | 75 (81.5) | 85 (61.2) | 365 (68.6) | |
| Zip code income | 0.24 | ||||||||
| <$48,000 | 39 (45.3) | 29 (49.2) | 27 (56.3) | 27 (38.6) | 15(39.5) | 45 (48.9) | 76 (54.7) | 258 (48.5) | |
| ≥$48,000 | 46 (53.5) | 30 (50.8) | 21 (43.8) | 43 (61.4) | 21 (55.3) | 46 (50) | 59 (42.4) | 266 (50) | |
| Urban | 0.21 | ||||||||
| Lives in metropolitan area | 66 (76.7) | 48 (81.4) | 34 (70.8) | 59 (84.3) | 27(71.1) | 79 (85.9) | 100 (71.9) | 413 (77.6) | |
| Urban city | 14 (16.3) | 10 (16.9) | 13(27.1) | 11 (15.7) | 7(18.4) | 8 (8.7) | 29 (20.9) | 92 (17.3) | |
| Rural | 3 (3.5) | 0 (0) | 1 (2.1) | 0 (0) | 2 (5.3) | 4 (4.3) | 4 (2.9) | 14(2.6) | |
| Geography | 0.08 | ||||||||
| East | 15 (17.4) | 14 (23.7) | 10(20.8) | 13 (18.6) | 10(26.3) | 22 (23.9) | 16 (11.5) | 100 (18.8) | |
| Midwest | 19 (22.1) | 12 (20.3) | 12 (25) | 22 (31.4) | 10(26.3) | 21 (22.8) | 37 (26.6) | 133(25) | |
| South | 33 (38.4) | 17 (28.8) | 13(27.1) | 23 (32.9) | 12 (31.6) | 31 (33.7) | 69 (49.6) | 198(37.2) | |
| West | 18 (20.9) | 16 (27.1) | 10(20.8) | 11 (15.7) | 5(13.2) | 18(19.6) | 15 (10.8) | 93 (17.5) | |
| Facility type | <0.01 | ||||||||
| Nonacademic | 52 (60.5) | 21 (35.6) | 17(35.4) | 33 (47.1) | 15(39.5) | 61 (66.3) | 83 (59.7) | 282 (53) | |
| Academic | 33 (38.4) | 38 (64.4) | 28 (58.3) | 36 (51.4) | 22 (57.9) | 31 (33.7) | 54 (38.8) | 242 (45.5) | |
| Charlson-Deyo score | 0.46 | ||||||||
| 0 | 47 (54.7) | 36 (61) | 27 (56.3) | 36 (51.4) | 27(71.1) | 47(51.1) | 76 (54.7) | 296 (55.6) | |
| ≥1 | 39 (45.3) | 23 (39) | 21 (43.8) | 34 (48.6) | 11 (28.9) | 45 (48.9) | 63 (45.3) | 236 (44.4) | |
| Tumor differentiation | 0.13 | ||||||||
| Well to moderate | 35 (40.7) | 19 (32.2) | 23 (47.9) | 27 (38.6) | 15(39.5) | 29 (31.5) | 50 (36) | 198(37.2) | |
| Poor to undifferentiated | 23 (26.7) | 18 (30.5) | 15(31.3) | 26 (37.1) | 17(44.7) | 24(26.1) | 45 (32.4) | 168(31.6) | |
| Unknown | 28 (32.6) | 22 (37.3) | 10(20.8) | 17 (24.3) | 6(15.8) | 39 (42.4) | 44 (31.7) | 166(31.2) | |
| Tumor size | <0.01 | ||||||||
| T1: 0–19 mm | 14 (16.3) | 20 (33.9) | 16(33.3) | 13 (18.6) | 9 (23.7) | 17(18.5) | 6 (4.3) | 95 (17.9) | |
| T2: 20–39 mm | 11 (12.8) | 19 (32.2) | 11 (22.9) | 19 (27.1) | 10(26.3) | 16(17.4) | 33 (23.7) | 119 (22.4) | |
| T3: >40 mm | 15 (17.4) | 4 (6.8) | 3 (6.3) | 4 (5.7) | 2 (5.3) | 7 (7.6) | 21 (15.1) | 56 (10.5) | |
| T4: Advanced local disease | 46 (53.5) | 16 (27.1) | 18(37.5) | 34 (48.6) | 17(44.7) | 52 (56.5) | 79 (56.8) | 262 (49.2) | |
| Distant metastasis | <0.01 | ||||||||
| No | 66 (76.7) | 54 (91.5) | 47 (97.9) | 63 (90) | 36 (94.7) | 80 (87) | 124 (89.2) | 470 (88.3) | |
| Yes | 20 (23.3) | 5 (8.5) | 1 (2.1) | 7(10) | 2 (5.3) | 12 (13) | 15 (10.8) | 62 (11.7) | |
AT = Adjuvant therapy; CRT = chemoradiation; RT = radiation therapy; T = tumor size.
In our cohort, 40% of patients underwent some form of surgical treatment, including 16% who underwent a curative procedure. Thirty-day mortality for patients who underwent limited and curative surgeries was 4.7%. Ninety-day mortality after limited surgical treatments was 8.5%, whereas it was 10.5% for curative approaches. Of the patients who received RT, the total radiation dose varied, with approximately 45% receiving ≥60 grays. The most common treatment modality was CRT (26%), whereas a fifth of patients underwent LS+AT or CS+AT, and another fifth underwent LS or CS only. About 17% of patients underwent RT alone; and lastly, 16% of patients did not receive any form of treatment around the time of diagnosis. Interestingly, only 7% of all patients were coded as receiving palliative care (Table I).
Overall Survival
Median follow-up was 60 months (interquartile range: 28–98). Median OS was 20 months and 5-year OS was 25% (Fig. 1A). Patients with metastatic disease had an OS of 4%, as compared to 27% among patients without metastatic disease (Fig. 1B). The corresponding HR was 2.6, with a 95% confidence Interval (CI) of 1.9 to 3.4 (P value <0.01). Overall survival for T1 tumors was 39% versus 29% for T2 tumors (HR 1.3; 95% CI 0.87–1.83; P = 0.21). T3 disease had an OS of 26% (HR 1.7; 95% CI 1.1–2.7; P = 0.01), whereas T4 disease had an OS of 21% (HR 2.0; 95% CI: 1.5–2.8; P < 0.01) (Fig. 1C). We generated a staging system using the above categories based on disease spread and tumor behavior: localized (T1–3, M0) (OS: 33%, median OS: 32 months), regional spread (T4, M0) (OS: 21%, median OS: 16 months, HR 1.6; 95% CI 1.3–2.0), and distant spread (Tany, M1) (OS: 4%, median OS: 7 months, HR 3.2; 95% CI 2.3–4.4), as seen in Figure 1D.
Fig. 1.
Survival trends of tracheal SCC: (A) overall survival; (B) survival by distant metastasis (M-stage); (C) survival by tumor size (T-stage); (D) survival by created staging (localized vs. regionally extensive vs. distant spread). SCC = squamous cell carcinoma. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
In UVA, median OS was 42 months for the CS+AT group, 41 months for CS group, 26 months for CRT group, 17 months for LS+AT group, 16 months for RT group, 15 months for LS group, and 6 months for NT group. Improved survival was associated with any treatment modality except for RT alone, as seen in Figure 2 and Table II. Five-year OS was greatest in patients who underwent CS as compared to NT (40% vs. 13%; logrank (LR) P < 0.01). Patients who underwent CS+AT had an OS of 34% (LR P < 0.01). Patients who underwent LS+AT or CRT had similarly crude OS of 28%. Factors associated with inferior OS included age ≥65 years, ≥1 comorbidities, and palliative treatment as described in Table II. Characteristics such as gender, race, insurance type, treatment facility type, income, geography, and degree of urbanization were not associated with survival.
Fig. 2.
Survival of tracheal SCC by treatment modality. AT = Adjuvant therapy; CRT = chemoradiation; RT = radiation therapy; SCC = squamous cell carcinoma. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
TABLE II.
Five-Year Hazard Ratio of All-Cause Mortality Among Tracheal Squamous Cell Carcinoma Patients According to Treatment Modality.
| n = 532 | ||||
|---|---|---|---|---|
|
|
||||
| Characteristic | Unadjusted Cox HR (95% CI) | P Value | Adjusted Cox HR (95% CI) | P Value |
|
| ||||
| Treatment modality | ||||
| No treatment | 1.00 (Ref) | 1.00 (Ref) | ||
| Limited surgery | 0.60 (0.4–0.91) | 0.02 | 0.75 (0.49–1.14) | 0.18 |
| Curative surgery | 0.37 (0.23–0.59) | <0.01 | 0.42 (0.26–0.69) | <0.01 |
| Limited surgery + AT | 0.54 (0.37–0.78) | <0.01 | 0.53 (0.36–0.77) | <0.01 |
| Curative surgery + AT | 0.40 (0.25–0.64) | <0.01 | 0.44 (0.27–0.72) | <0.01 |
| Radiation therapy | 0.73 (0.52–1.02) | 0.07 | 0.66 (0.46–0.94) | 0.02 |
| Chemoradiation | 0.50 (0.36–0.69) | <0.01 | 0.48 (0.35–0.67) | <0.01 |
| Staging | ||||
| Localized | 1.00 (Ref) | 1.00 (Ref) | ||
| Regional extensive | 1.59 (1.27–2.01) | <0.01 | 1.64 (1.29–2.09) | <0.01 |
| Distant spread | 3.19 (2.32–4.39) | <0.01 | 2.47 (1.76–3.47) | <0.01 |
| Gender | ||||
| Male | 1.00 (Ref) | 1.00 (Ref) | ||
| Female | 0.87 (0.7–1.09) | 0.23 | 0.92 (0.73–1.16) | 0.47 |
| Age | ||||
| <65 years | 1.00 (Ref) | 1.00 (Ref) | ||
| ≥65 years | 1.43 (1.16–1.77) | <0.01 | 1.26 (1.00–1.61) | 0.05 |
| Insurance | ||||
| Private | 1.00 (Ref) | 1.00 (Ref) | ||
| Other/government | 1.22 (0.97–1.54 | 0.08 | 1.09 (0.84–1.42) | 0.51 |
| Palliative treatment | ||||
| No | 1.00 (Ref) | 1.00 (Ref) | ||
| Yes | 2.30 (1.60–3.31) | <0.01 | 1.85 (1.25–2.73) | <0.01 |
| Charlson-Deyo score | ||||
| 0 | 1.00 (Ref) | 1.00 (Ref) | ||
| ≥1 | 1.39 (1.13–1.72) | <0.01 | 1.39 (1.12–1.73) | <0.01 |
| Facility type | ||||
| Nonacademic | 1.00 (Ref) | 1.00 (Ref) | ||
| Academic | 0.98 (0.79–1.21) | 0.85 | 1.09 (0.87–1.35) | 0.45 |
| Zip code Income | ||||
| <$48,000 | 1.00 (Ref) | |||
| ≥$48,000 | 1.06 (0.85–1.31) | 0.61 | ||
| Race | ||||
| White | 1.00 (Ref) | |||
| Black | 0.98 (0.71–1.36) | 0.92 | ||
| Other | 1.41 (0.97–2.05) | 0.08 | ||
| Geography | ||||
| East | 1.00 (Ref) | |||
| Midwest | 0.94 (0.68–1.28) | 0.69 | ||
| South | 0.94 (0.7–1.26) | 0.68 | ||
| West | 0.95 (0.67–1.35) | 0.78 | ||
| Urban | ||||
| Lives in metro | 1.00 (Ref) | |||
| Urban | 1.01 (0.76–1.33) | 0.97 | ||
| Rural | 1.17 (0.64–2.14) | 0.61 | ||
AT = Adjuvant therapy; CI = confidence interval; HR = hazard ratio.
In the MVA, we incorporated stage, gender, age, insurance, facility type, comorbidities, and palliative care based on UVA with treatment modality and Cox proportional HRs. All treatment modalities except for LS demonstrated a decrease in the likelihood of death compared to the NT group. CS, CS+AT, and CRT were associated with largest reductions in HR of about 52% to 58%, whereas LS +AT exhibited a 47% reduction and RT revealed a 34% reduction (refer to Table II). Worsening disease spread, age ≥65 years, ≥1 comorbidity, and palliative care continued to be significantly associated with worse HR. Pairwise comparisons of the treatment groups were performed; however, no statistically significant protection against mortality was found between treatment groups except when comparing against LS. Comparison to LS demonstrated similar results as compared to NT.
The effect of treatment modality on survival was further investigated by stratified analysis of disease spread, that is, staging. In patients with localized disease (T1–3, M0), CS and CS+AT were associated with over 60% reduction in risk of death compared to NT, whereas CRT was associated with a 45% reduction (Table III) (Fig. 3A). Limited surgery, LS+AT, and RT groups also demonstrated an improved HR compared to the NT group in localized disease, but it was not statistically significant. In patients with advanced regional disease spread (T4, M0), CS+AT revealed the greatest reduction in HR at 68%, followed closely by the CRT and CS groups (Table III) (Fig. 3B). LS+AT was associated with a 62% reduction in the likely-hood of death, followed by RT alone at 52% reduction. Among treatment groups with statistically significant HR, pairwise comparisons were performed but did not demonstrate significant decrease in hazard between one treatment group over another. In patients with distant metastatic disease, KM survival curves and adjusted and unadjusted Cox proportional hazards showed no statistically significant difference between the treatment modality groups given the low number of patients in each subset and very poor survival. As expected, the majority of these patients underwent CRT, RT, or NT. Despite the inability to demonstrate statistical significance, the CRT cohort had the highest OS.
TABLE III.
Five-Year Hazard Ratio of All-Cause Mortality Among Tracheal Squamous Cell Carcinoma Patients According to Treatment Modality by Localized and Advanced Staging.
| Characteristic | Localized (n = 248) | Regional Extension (n = 222) | ||
|---|---|---|---|---|
|
|
|
|||
| Adjusted Cox HR* (95% CI) | P Value | Adjusted Cox HR* (95% CI) | P Value | |
|
| ||||
| Treatment modality | ||||
| No treatment | 1.00 (Ref) | 1.00 (Ref) | ||
| Limited surgery | 0.56 (0.29–1.07) | 0.08 | 0.82 (0.38–1.78) | 0.62 |
| Curative surgery | 0.38 (0.18–0.79) | 0.01 | 0.35 (0.17–0.76) | 0.01 |
| Limited surgery + AT | 0.53 (0.28–1.02) | 0.06 | 0.38 (0.21 −0.7) | <0.01 |
| Cur surgery + AT | 0.39 (0.17–0.86) | 0.02 | 0.32 (0.16–0.66) | <0.01 |
| RT | 0.72 (0.39–1.33) | 0.29 | 0.48(0.29–0.81) | 0.01 |
| CRT | 0.55 (0.3–1.00) | 0.05 | 0.34 (0.21 −0.56) | <0.01 |
Adjusted for sex, age, insurance type, comorbidities, treatment facility type, and palliative treatment.
CI = confidence interval; HR = hazard ratio.
Fig. 3.
Survival of tracheal SCC stratified by staging: (A) among localized disease only; (B) among non-metastatic advanced localized disease. CRT = chemoradiation; RT = radiation therapy; AT = Adjuvant therapy; SCC = squamous cell carcinoma. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
DISCUSSION
Our analysis of tracheal squamous cell carcinoma provides insight into prognosis and treatment approach for a rare and deadly disease. Multiple staging systems for primary tracheal carcinoma have been proposed, but none to date have been incorporated into standards of care. Previously described staging systems consisted of small cohorts as well as incomplete staging information with poor internal and external validity.5,8,9 Evidence to guide treatment is also plagued by data spanning multiple decades and confounding by inclusion of multiple his-tologies.2,8,10 Whereas multiple case series have advocated for definitive resection for localized tumors and multimodality therapy for tumors not amenable to complete surgical resection, epidemiologic evidence from national registries has shown that tracheal carcinomas are rarely considered for surgical resection despite these findings.1,7,9,11–13 Fewer than half of all patients with potentially resectable tumors underwent RT instead of surgery in the Netherlands Registry.14 This may be due to the lack of awareness among physicians about available treatment approaches and their effectiveness.8,14
In our analysis, we found that almost half of patients with TSCC underwent surgery, with nearly half of those receiving a curative procedure. This is increased from earlier studies that reported surgical rates as high as 34%.3,15 Surgical expertise of tracheal carcinoma resections has likely increased over time, leading to lower rates of postoperative morbidity and mortality.6 Advanced techniques in tracheal surgery today have made it possible to safely resect approximately half of the trachea.8,16–18 Unquestionably, this depends on the patient’s age, weight, neck mobility, and comorbidity. Elderly patients with kyphosis may not tolerate the resection of more than 2 centimeters of trachea, whereas young patients with long, mobile necks may undergo successful reconstruction after resection of 6 or more centimeters. Contrary to prior studies, our analysis demonstrated that a near majority of patients received multimodality therapy with either adjuvant RT/CRT or definitive CRT, whereas 17% underwent definitive RT solely. Most patients in prior epidemiological studies were reported as receiving RT alone because systemic therapy was not often analyzed or part of treatment.15 Interestingly, 26% of patients in our cohort underwent definitive CRT, making it the most common form of treatment. The variation in treatment rates across studies may be due to changing trends over time, with many earlier studies consisting of data from several decades ago and due to systemic therapy not being previously recorded in national databases.2,3,15
Whereas fewer than one-fifth of patients underwent upfront curative surgical treatment, our analysis demonstrated CS was associated with a significant decrease in risk of death compared to no treatment, even when controlling for stage and other demographic and clinical factors. CS+AT and CRT also demonstrated comparable risk reductions. These associations of improved survival were stronger among patients with localized TSCC. Although the risks of acute surgical morbidity need to be accounted for, the results suggest that localized diseases may be effectively treated with CS alone. Unfortunately, we cannot assess why certain patients had adjuvant RT/CRT because assessment of pathologic risk factors, such as margin status or extracapsular spread or number of nodes in the NCDB, is highly inaccurate. We speculate that differences in outcomes by modality may represent such postoperative pathological findings concerning for aggressive spread or incomplete resection. The improved OS rates with surgical resection has been corroborated in other studies looking at all primary tracheal malignancies.3,6 Gaissert et al. study from 2004 showed statistically improved 5-year survival rates in surgically treated tracheal carcinomas versus unresected disease in both adenoid cystic carcinoma and SCC.6 Several European studies reported a significant improved 5-year survival rate in patients who underwent surgical resection versus other treatment (51% vs. 15%, respectively).15,19 Analysis of an English national registry also demonstrated improved survival for those who underwent curative resection compared to survival overall (60.8% vs. 19.5%).7 Lastly, an analysis of the Surveillance, Epidemiology, and End Results (SEER) database in 2017 by Agrawal et al. demonstrated improved OS and cancer-specific survival among patients who underwent surgery alone or surgery plus RT compared to no treatment or RT alone.3 Arguments against these findings state that differences observed between surgical candidates and nonsurgical candidates are most likely because patients with resectable tumors present with more localized malignancy, with smaller tumors, without distant metastasis, and with younger age and fewer health comorbidities. We attempted to control for these imbalances to the extent possible, but undetected confounders could also underlie discrepant results by treatment modality. In patients with regionally invasive TSCC, we see a transition in treatment effectiveness with RT and CRT modalities having improved survival rates. Not only were CS, CS+AT, and definitive CRT associated with a decreased risk of death, but RT and LS+AT were as well. This may be attributed to the adjuvant RT or CRT therapy patients receive, highlighting the need for at least bimodality therapy in patients with advanced disease.
Although our results correlate with prior studies that recommend curative surgical resection when feasible, these analyses are controversial in comparing RT and CRT versus no treatment. We found improved survival rates and decreased risk of death with primary RT in advanced TSCC, albeit not as strong as other modalities. Furthermore, risk of death was comparable between CRT and CS/CS+AT versus NT. The inclusion of adjuvant RT/CRT appeared to provide similar risk reduction compared to the surgical approach alone in localized stages of TSCC. Whereas adjuvant therapy did not provide a statistical survival benefit in comparison to other treatments, we believe that those who were selected to undergo adjuvant therapy had risk factors that prompted additional therapy such as positive margins, extracapsular extension, and perineural or lymphovascular invasion. In a retrospective matched-pair analysis of the SEER database, Xie et al. advocated for RT as an adjuvant therapy after surgery for patients with suspected positive margins or lymph node involvement in order to improve survival.10
To the best of our knowledge, our study provides the first insight into CRT as an effective treatment for TSCC in the primary or adjuvant setting. Based on prior studies, chemotherapy has not been well studied for tracheal malignancies and has been largely used with palliative intent.8 Few case studies have suggested CRT as an option for inoperable patients with unresectable localized disease. In a case report, triplet consisting of cisplatin, 5-fluorouracil, and etoposide was delivered concurrently with radiation in a patient with TSCC.20 Nouraei et al. found a significant survival advantage with chemotherapy but not with RT in tracheal carcinomas, but the discrepancy may be due to tumor characteristics related to histology.7 In randomized trials of patients who underwent resection of advanced head and neck squamous cell carcinoma (HNSCC), patients who received postoperative CRT experienced improved survival and decreased local recurrence but higher adverse effects than patients who received adjuvant RT alone.21,22 Given similarities in SCC histology, our findings are similar to these large prospective studies in HNSCC. It would be logical to consider using chemotherapy in addition to RT for unresectable or metastatic TSCC to improve survival.
Encouragingly, we find that OS has greatly improved for TSCC in the US compared to other populations and prior decades. Cross-sectional and retrospective studies from the 1970s to early 2000s have reported 5-year survival rates as low as 3%.4,20,21 Our data, which includes 2004 through 2012, demonstrates a 5-year OS of 25% that is nearly double that from a SEER study that showed 5-year survival in patients with TSCC from 1973 to 2004 to be 12.6%.2 This remarkably improved 5-year prognosis for SCC may reflect improvement in rates of cure over the past few decades stemming from the increasing utilization of multi-modality therapy that we identified.
Our retrospective study is not without limitations. The first is inherent to the NCDB data set because information on tumor recurrence was not available, meaning that disease-free survival could not be calculated. A limitation of this type of calculation is that it would overestimate TSCC-associated mortality if TSCC patients have coexisting comorbidities that can contribute to mortality. The NCDB does not contain information about specific patient comorbidities, smoking status, chemotherapy type and administration, and other patient-specific demographics that may also contribute to survival—or information about which treatments patients are offered or undergo. Treatment selection bias is likely present, and modality selection is correlated with the extent of tumor spread and metastasis, which may affect survival despite adjusting for tumor stage.
CONCLUSION
TSCC remains a rare malignancy with a poor 5-year overall survival of 25%. We are seeing increasing rates of surgical and systemic therapy with improved survival outcomes. Our staging classification provides excellent prognostication for patients and clinicians. We find that CS, CS+AT, definitive CRT, or RT have variable survival advantage in patients with this disease. Not surprisingly, increasing comorbidities and older age represent adverse prognostic factors. Prospective clinical trials would give provide insight into treatment-specific guidelines. Patients’ goals of care, comorbidities, and age should be considered when deciding appropriate treatment recommendations.
Acknowledgments
The authors have no funding, financial relationships, or conflicts of interest to disclose.
BIBLIOGRAPHY
- 1.Licht PB, Friis S, Pettersson G. Tracheal cancer in Denmark: a nationwide study. Eur J Cardiothorac Surg 2001;19:339–345. [DOI] [PubMed] [Google Scholar]
- 2.Urdaneta AI, Yu JB, Wilson LD. Population based cancer registry analysis of primary tracheal carcinoma. Am J CUn Oncol 2011;34:32–37. [DOI] [PubMed] [Google Scholar]
- 3.Agrawal S, Jackson C, Celie KB, et al. Survival trends in patients with tra cheal carcinoma from 1973 to 2011. Am J Otolaryngol 2017;38:673–677. [DOI] [PubMed] [Google Scholar]
- 4.Ampil FL. Primary malignant tracheal neoplasms: case reports and litera ture radiotherapy review. J Surg Oncol 1986;33:20–23. [DOI] [PubMed] [Google Scholar]
- 5.Bhattacharyya N Contemporary staging and prognosis for primary tracheal malignancies: a population-based analysis. Otolaryngol Head Neck Surg 2004;131:639–642. [DOI] [PubMed] [Google Scholar]
- 6.Gaissert HA, Grillo HC, Shadmehr MB, et al. Long-term survival after re section of primary adenoid cystic and squamous cell carcinoma of the trachea and carina. Ann Thorac Surg 2004;78:1889–1896; discussion 1896–1887. [DOI] [PubMed] [Google Scholar]
- 7.Nouraei SM, Middleton SE, Nouraei SA, et al. Management and prognosis of primary tracheal cancer: a national analysis. Laryngoscope 2014;124:145–150. [DOI] [PubMed] [Google Scholar]
- 8.Madariaga MLL, Gaissert HA. Overview of malignant tracheal tumors. Ann Cardiothorac Surg 2018;7:244–254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Macchiarini P Primary tracheal tumours. Lancet Oncol 2006;7:83–91. [DOI] [PubMed] [Google Scholar]
- 10.Xie L, Fan M, Sheets NC, Chen RC, Jiang GL, Marks LB. The use of radia tion therapy appears to improve outcome in patients with malignant primary tracheal tumors: a SEER-based analysis. Int J Radiat Oncol Biol Phys. 2012;84:464–470. [DOI] [PubMed] [Google Scholar]
- 11.Honings J, Gaissert HA, van der Heijden HF, Verhagen AF, Kaanders JH, Marres HA. Clinical aspects and treatment of primary tracheal malignancies. Acta Otolaryngol 2010;130:763–772. [DOI] [PubMed] [Google Scholar]
- 12.Behringer D, Konemann S, Hecker E. Treatment approaches to primary tra cheal cancer. Thorac Surg Clin 2014;24:73–76. [DOI] [PubMed] [Google Scholar]
- 13.Hetnal M, Kielaszek-Cmiel A, Wolanin M, et al. Tracheal cancer: role of radiation therapy. Rep Pract Oncol Radiother 2010;15:113–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Honings J, Gaissert HA, Verhagen AF, et al. Undertreatment of tracheal carcinoma: multidisciplinary audit of epidemiologic data. Ann Surg Oncol 2009;16:246–253. [DOI] [PubMed] [Google Scholar]
- 15.Honings J, van Dijck JA, Verhagen AF, van der Heijden HF, Marres HA. Incidence and treatment of tracheal cancer: a nationwide study in the Netherlands. Ann Surg Oncol 2007;14:968–976. [DOI] [PubMed] [Google Scholar]
- 16.Grillo HC. Development of tracheal surgery: a historical review. Part 1: techniques of tracheal surgery. Ann Thorac Surg 2003;75:610–619. [DOI] [PubMed] [Google Scholar]
- 17.Grillo HC. Reconstruction of the trachea. Experience in 100 consecutive cases. Thorax 1973;28:667–679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Pearson FG, Cooper JD, Nelems JM, Van Nostrand AW. Primary tracheal anastomosis after resection of the cricoid cartilage with preservation of recurrent laryngeal nerves. J Thorac Cardiovasc Surg 1975;70:806–816. [PubMed] [Google Scholar]
- 19.Regnard JF, Fourquier P, Levasseur P. Results and prognostic factors in resections of primary tracheal tumors: a multicenter retrospective study. The French Society of Cardiovascular Surgery. J Thorac Cardiovasc Surg 1996;111:808–813; discussion 813–804. [DOI] [PubMed] [Google Scholar]
- 20.Videtic GM, Campbell C, Vincent MD. Primary chemoradiation as definitive treatment for unresectable cancer of the trachea. Can Respir J 2003;10: 143–144. [DOI] [PubMed] [Google Scholar]
- 21.Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med 2004;350:1945–1952. [DOI] [PubMed] [Google Scholar]
- 22.Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radio therapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 2004;350:1937–1944. [DOI] [PubMed] [Google Scholar]