Racial Disparities in the Use of Voice Preservation Therapy for Locally Advanced Laryngeal Cancer

Objective To identify potential racial disparities in the use of larynx preservation.

Design Retrospective database review.

Setting Academic medical center.

Patients The Surveillance Epidemiology and End Results (SEER) database was used to identify white, black, Hispanic, and Asian patients with stage III and IV laryngeal cancers that were diagnosed during 1991 through 2008. Patients with T4 disease or distant metastasis were intentionally excluded.

Main Outcome Measure Univariate and multivariate logistic regression analysis, with odds ratios and 95% confidence intervals, was used to investigate the relationship between race/ethnicity and the use of larynx preservation with radiation therapy as initial therapy.

Results Among the 5385 cases of laryngeal cancers that met the selection criteria, the racial distribution was white (72.7%), black (16.8%), Hispanic (7.4%), and Asian (3.1%). On univariate analysis, blacks (odds ratio [OR], 0.72; 95% CI, 0.59-0.88) were significantly less likely to undergo larynx preservation. This racial disparity persisted on multivariate analysis for blacks (OR, 0.78; 95% CI, 0.63-0.96) and was still observed among patients treated more recently between 2001 and 2008 (OR, 0.74; 95% CI, 0.56-0.96).

Conclusions Pronounced racial disparities exist in the use of larynx preservation therapy for locally advanced laryngeal cancer. While acknowledging the potential biases of socioeconomic factors, further research to better elucidate the underlying reasons for these findings may be warranted.

Approximately 12 000 cases of laryngeal cancer are diagnosed in the United States annually.¹ Historically, the standard of care for patients with locally advanced laryngeal cancer consisted of total laryngectomy followed by adjuvant radiation therapy. However, the landmark Veteran Affairs (VA) Laryngeal Cancer Study in 1991 demonstrated that when compared with total laryngectomy followed by adjuvant radiation, the use of radiation therapy in patients who responded to induction chemotherapy conferred a high rate of larynx preservation without compromising survival.² Subsequently, the Radiation Therapy Oncology Group (RTOG) trial 91-11 showed that concurrent chemoradiation is superior to induction chemotherapy, followed by radiation or radiation alone for larynx preservation.³ These studies, along with others,^4,5 have resulted in the widespread acceptance of larynx preservation using radiation therapy with concurrent chemotherapy as the initial treatment for locally advanced laryngeal cancer, with total laryngectomy reserved for salvage therapy.⁶

While nonsurgical larynx preservation appears to be gaining popularity among medical practitioners,⁷ it is unclear if racial disparities exist in its use.⁸ Since racial minorities often have worse access to care, particularly with respect to radiation therapy facilities and receive a lower quality of care,⁹ we hypothesized that black and Hispanic patients with locally advanced laryngeal cancers are less likely to undergo larynx preservation treatment than their white counterparts. To address this question, we identified patients with American Joint Committee on Cancer (AJCC) stage III and stage IV nonmetastatic laryngeal cancer from the Surveillance Epidemiology and End Results (SEER) database and analyzed trends in the use of larynx preservation.

The SEER database of the National Cancer Institute was used to identify all white, black, Hispanic, and Asian patients with AJCC stage III and IV cancer of the larynx diagnosed during 1991 through 2008. SEER collects cancer incidence and survival data from 17 populations-based cancer registry representing 26% of the US population. Current SEER registry consist of the states of Connecticut, Hawaii, Iowa, Kentucky, Louisiana, New Jersey, New Mexico, and Utah; the metropolitan areas of Atlanta (Georgia), Detroit (Michigan), San Francisco–Oakland (California), Seattle-Puget Sound (Washington), and San Jose–Monterey (California); and the Alaska Native Tumor Registry, rural Georgia, greater California, and Los Angeles County (California). SEER registries routinely collect data on patient demographics, primary tumor site, tumor morphology, stage at diagnosis, and first course of treatment.

To select suitable candidates for larynx-preservation, patients with T4 disease or distant metastasis were intentionally excluded from this analysis. Further exclusions were made if race was other than white, black, Hispanic, or Asian; if cancer site was other than supraglottis, glottis, or subglottis; or if initial treatment was anything other than radiation therapy or total laryngectomy. Patients treated for recurrent disease were not included. “Larynx preservation” identified the group of patients who underwent radiation therapy as initial treatment.

Univariate analysis was used to evaluate the relationship between race/ethnicity and the use of definitive larynx preservation using white patients as the referent population. Odds ratios (ORs) were reported with associated 95% confidence intervals. Differences were considered significant at P ≤ .05. With larynx preservation therapy as the outcome variable, multivariate logistic regression analysis was adjusted for race, age, sex, year of diagnosis, stage, subsite, and histologic features. Additional multivariate logistic regression models were constructed to stratify for patients diagnosed during 1991 through 2000 and 2001 through 2008, and for patients with stage III and IV disease. Statistical analyses were conducted using Minitab version 16 (Minitab Inc).

A total of 3862 patients with locally advanced laryngeal cancer diagnosed from 1991 through 2008 met our selection criteria. Patient factors, tumor characteristics, and receipt of therapies according to racial/ethnic groups are given in Table 1. The mean age of the entire patient population was 63 years (range 12-100 years). The racial distribution was white (72.7%), black (16.8%), Hispanic (7.4%), and Asian (3.1%). Male patients composed 73.9% of the entire study population. The most common subsite of disease was supraglottis (66.9%), followed by glottis (31.7%) and subglottis (1.4%). Patients predominantly had T3 disease (64.4%), followed by T2 (24.2%) and T1 diseases (11.4%). The majority of patients received larynx preservation as the initial treatment strategy (79.0%). More than 90% of patients received radiation as a component of their cancer treatment with no significant differences among racial/ethnic groups (P = .69).

Blacks (OR, 0.72; 95% CI, 0.59-0.88; P = .001) were less likely to undergo larynx preservation than whites (Table 2). The ORs for Hispanics (OR, 0.87; 95% CI, 0.65-1.17; P = .370) and Asians (OR, 0.76; 95% CI, 0.50-1.17; P = .22) to receive larynx preservation were not significantly different from whites. The observed racial disparity in the use of larynx preservation among all analyzed ethnicities over time from 1991 through 2008 is illustrated in Figure 1.

In the all-inclusive multivariate logistic regression model given in Table 3, use of larynx preservation occurred more commonly in cases diagnosed during 2001 through 2008 (OR, 3.25; 95% CI, 2.75-3.85; P < .001). Conversely, use of larynx preservation was less likely among patients with stage IV disease (OR, 0.69; 95% CI, 0.57-0.83; P < .001) and in blacks (OR, 0.78; 95% CI, 0.63-0.96; P = .02). Use of larynx preservation was also less common in patients with T2 (OR, 0.61; 95% CI, 0.43-0.87; P = .006) or T3 disease (OR, 0.34; 95% CI, 0.25-0.47; P < .001).

Table 4 summarizes the results of multivariate logistic regression analyses stratified by period (1991-2000 vs 2001-2008). Between 1991 and 2000, the use of larynx preservation was less likely among patients with stage IV disease (OR, 0.66; 95% CI, 0.48-0.90; P = .008). In addition, there was a trend toward a lack of use of larynx preservation among blacks (OR, 0.76; 95% CI, 0.55-1.07; P = .12). For the era spanning 2001 through 2008, the use of larynx preservation was more likely in patients with supraglottic disease (OR, 1.41; 95% CI, 1.12-1.79; P = .004) and less likely in blacks (OR, 0.74; 95% CI, 0.56-0.96; P = .02).

Table 5 summarizes the likelihood of larynx preservation according to race stratified by AJCC stage. For stage III disease, use of larynx preservation was more likely in patients diagnosed between 2001 and 2008 (OR, 2.60; 95% CI, 2.07-3.26; P < .001). No significant difference was observed between racial/ethnic groups. For stage IV disease, use of larynx preservation was again more likely in patients diagnosed between 2001 and 2008 (OR, 4.12; 95% CI, 3.21-5.28; P < .001). Blacks were less likely to undergo larynx preservation for stage IV disease (OR, 0.57; 95% CI, 0.42-0.78; P < .001).

While the use of nonsurgical larynx preservation incorporating primary radiation therapy for locally advanced laryngeal cancer increased significantly among the general population from 1991 through 2008 (Figure 2), it is notable that pronounced racial disparities continue to exist. After controlling for potentially confounding variables including age, sex, year of diagnosis, stage, and subsite, we demonstrated that blacks were less likely to undergo larynx preservation than whites. Following the trend within the general population, the use of larynx preservation has also increased among black patients over time, and our data suggest that the observed racial disparity may in fact be narrowing.

Our results are noteworthy because larynx preservation with chemoradiation therapy confers a high probability of retaining a functional larynx without a negative impact on survival.⁸ Retrospective studies have demonstrated that surgically treated patients experience more respiratory complications, pain, depression, and social dysfunction than patients with an intact larynx.^10,11 Due to a lack of nasal air conditioning, patients without a larynx also develop cough, dyspnea, and sleep disturbances.¹² Nerve irritation and tissue damage following neck dissection often results in chronic shoulder pain.¹³ In addition, patients treated with total laryngectomy experience worse social functioning and isolation from speech impairment and body disfigurement.¹⁰ As such, choice of initial therapy (total laryngectomy vs radiation therapy) has a dramatic impact on posttherapy quality of life, and the use of larynx preservation is an accurate barometer for assessing quality of care.

The results of the present study showing that racial disparities exist with respect to the selection of therapy for laryngeal cancer are consistent with previously published data analyzing patterns of care involving radiation therapy. Others have demonstrated that black patients with head and neck cancer present more frequently with extensive disease, are less likely to receive cancer-directed therapies, and have higher cancer-specific mortality when compared with their white counterparts.¹⁴ With respect to the use of radiation, it has also been shown that black patients with breast cancer are less likely to receive adjuvant radiation therapy following lumpectomy or mastectomy.^15,16 In addition, black patients with rectal cancer have been shown to less frequently undergo neoadjuvant radiation therapy and sphincter-preserving surgery.¹⁷

There are several possible explanations for our findings. First, larynx preservation therapy involves an intensive and prolonged course of radiation therapy with chemotherapy; therefore, a physician may deem a patient unreliable to complete radiation therapy and recommend surgical options. Second, a lack of health literacy may prevent a patient from full comprehension of management options. Third, patients may decline radiation therapy if they are unable to complete treatment owing to lack of social or familial support. Fourth, racial disparities in referral to oncology specialists can also contribute to the observed difference. Lastly, other social factors including financial barriers, health insurance, and transportation may potentially influence treatment modalities available to patients.⁹

Unexpectedly, a strong trend was noted toward an increased use of postoperative radiation therapy following total laryngectomy among black, Hispanic, and Asian patients (whites, 61.2%; blacks, 67.9%; Hispanics, 66.7%; and Asians, 79.3% [P = .11]). Postlaryngectomy radiation therapy is indicated for adverse features, including pT3-4, pN2-3, close and/or positive margin, extranodal extension, subglottic extension of 1 cm or greater, cartilage involvement, perineural invasion, or lymphovascular space invasion.¹⁸ We speculate that among patients who underwent total laryngectomy, minority patients were more frequently found to harbor high-risk disease pathologically, prompting postoperative radiation therapy. Unfortunately, the SEER database does not provide sufficient detail regarding adverse pathologic features to evaluate this hypothesis. However, if the increased use of postoperative radiation therapy in minority patients was indeed a result of more unfavorable pathologic conditions, this may reflect a combination of racial disparity in the evaluation and recommendation for surgery by physicians as well as a lack of thorough understandings of risks, benefits, and alternatives of procedures by patients.

Prior studies have consistently identified patients of low socioeconomic status as at risk for receiving inferior medical care and experiencing poorer outcomes.¹⁹ Patients of low socioeconomic status tend to have below-average education levels, with lower health literacy and increased financial barriers to health care.⁹ Unfortunately, we were unable to control for socioeconomic status in our analysis. Because of the potentially confounding effect of this variable, our conclusions must therefore be tempered by the recognition that the observed racial disparity may simply be a socioeconomic disparity, where the less-advantaged socioeconomic group included a much higher percentage of racial minorities. It also must be noted that geographic regions with much higher percentages of racial minorities may also be somewhat less likely to use larynx preservation approaches for reasons unrelated to race/ethnicity. Another shortcoming in our study was the possible inclusion of patients receiving palliative radiation therapy in the “larynx preservation” group because SEER database entries do not distinguish palliative radiation from radiation therapy of curative intent. The number of patients receiving palliative radiation therapy was presumably small, given that our study cohort excluded heavy disease burden (metastatic disease or T4 disease). In addition, SEER database entries do not provide other medical information that may potentially prevent patients from receiving radiation therapy (poor performance status, uncontrolled psychiatric illness, inability to lie flat, prior irradiation, and collagen vascular disease). The primary limitations of this study thus relate to those inherent in any secondary analysis, and caution must be exercised in drawing definitive conclusions. Despite these shortcomings, our study has identified an important area for improvement in the management of laryngeal cancers in black patients.

Lastly, information on treatment-related complications was unavailable for review. Although patients undergoing nonsurgical therapy with chemoradiation are likely to retain their larynx, laryngeal function is not always preserved.²⁰ Following radiation therapy, patients can develop laryngeal and esophageal dysfunction including dyspnea, dysphagia, and chronic aspiration with severe symptoms requiring tracheotomy, laryngectomy, or feeding tube placement.²⁰ Patients also experience varying degrees of alteration in voice quality following radiation therapy. Future studies that assess survival end points and long-term laryngoesophageal function will help further elucidate disparities in functional outcomes and quality of life for these patients.²⁰

In conclusion, although the use of nonsurgical larynx preservation therapy appears to be increasing among the general population, racial disparities continue to exist, most notably among black patients with stage IV disease. Acknowledging that socioeconomic and nonethnicity related variables have the potential to confound our observed findings, we believe that future research should focus on identifying and eliminating barriers to the use of larynx preservation for all medically suitable patients, with a particular focus on black patients with stage IV disease.

Correspondence: Allen M. Chen, MD, Department of Radiation Oncology, University of California, Davis Comprehensive Cancer Center, 4501 X St, Ste G140, Sacramento, CA 95817 (allen.chen@ucdmc.ucdavis.edu).

Submitted for Publication: January 30, 2012; final revision received April 18, 2012; accepted April 21, 2012.

Author Contributions: Drs Hou and Chen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Hou, Lee, Luu, and Chen. Acquisition of data: Hou, Lee, and Chen. Analysis and interpretation of data: Hou, Daly, Lee, Farwell, and Chen. Drafting of the manuscript: Hou, Lee, and Chen. Critical revision of the manuscript for important intellectual content: Hou, Daly, Lee, Farwell, Luu, and Chen. Statistical analysis: Hou, Lee, and Chen. Administrative, technical, and material support: Luu. Study supervision: Daly.

Previous Presentation: This work was presented at the 94th annual meeting of the American Radium Society; May 1, 2012; Las Vegas, Nevada.