Disparities in Thoracic Surgical Oncology

Abstract

Disparities in access and outcomes of thoracic surgical oncology are long standing. This article examines the patient, population, and systems-level factors that contribute to these disparities and inequities. The need for research and policy to identify and solve these problems is apparent. As leaders in the field of thoracic oncology, surgeons will be instrumental in narrowing these gaps and moving the discipline forward.

Introduction

Outcomes for patients with thoracic malignancies are overall improving owing to advances in smoking cessation campaigns, screening programs, anesthetic and surgical technology, perioperative care pathways, immunotherapy, and coordinated multidisciplinary care. However, disparities in risk, incidence, presentation, treatment, and outcomes of patients with thoracic malignancies persist and deserve attention. These disparities are complex and attributable to the interplay of patient, social, and system-level determinants, creating several opportunities for improvement.

Here, we describe national and global trends in these areas with a focus on disparities for patients with lung and esophageal cancer. The impact of policy-making led by thoracic surgeons on disparities must be carefully considered. Therefore, we provide the background required to inform the modern thoracic surgeon on disparities in thoracic surgical care with an eye toward highlighting areas for continued work in our field.

Epidemiology and Risk Factors

Lung cancer

Lung cancer is the second most common malignancy worldwide (approximately 2.2 million new cases annually) and the leading cause of global cancer-related mortality (~2 million deaths) [1, 2]. The global incidence of lung cancer is steadily increasing (by 37% from 2007–2017), due to aging and growing industrialized populations [3]. In 2022, there will be > 200,000 new cases of lung and bronchial cancer in the USA with an equal distribution between men and women [4]. Smoking remains the major risk for lung cancer.Demographic and regional variations in incidence are largely driven by tobacco use trends. Thus, the populations with the highest current rates of lung cancer are also those in whom rates of cigarette smoking peaked in the latter half of the 20th century [1, 5, 6]. Accordingly, the current highest rates of lung cancer are in European countries [7]. Indoor air pollution from household smoke is the other major risk factor, particularly for women in countries with low socio-demographic index (SDI) [1].

Disparities in smoking rates between low- and high-resource countries are mediated by many of the factors which shape disparities in the developed world. Despite the introduction of the WHO’s MPOWER campaign, communities in low- and middle-income countries (LMICs) lag behind high-income countries (HICs) in their implementation of tobacco control policies [8]. Tobacco product advertising is more prevalent, cigarettes are cheaper, and overall knowledge of tobacco’s ill health effects is lower in LMICs than in HICs [9]. Analysis of data from the Global Adult Tobacco Survey (GATS) has found that the socioeconomic axes of educational attainment and income are significant predictors of tobacco use in LMICs (e.g., higher levels of education and wealth correlate with a decreased likelihood of current smoking behavior). Similarly, the interaction of education and income is observed in smoking cessation success rates in HICs [10].

An additional variable acting as a key driver of increases in LMIC smoking rates is the continued prevalence of flavored tobacco products in these markets [11]. While many developed countries have begun to curb the sale of tobacco products flavored with menthol and other additives, the use of these products is growing in many LMICs [12]. These products tend to be more attractive to younger smokers, who may find flavored tobacco smoke less aversive [13]. Because GATS data has demonstrated that quit rates are low in many LMICs, increased youth smoking rates may translate into higher proportions of adult smokers in these locales [14]. These factors can be used to explain the increasing smoking rates currently observed in much of the developing world [15]. As this trend continues, it may set the stage for a corresponding shift in global lung cancer disease burden in decades to come [16].

In the United States, smoking rates have been nearly halved in the last two decades, but there are an estimated 12.5% of American adults (30.8 million) still smoking in 2020 [17]. Further, socioeconomic disparities in smoking status have widened with women and those from disadvantaged backgrounds including low education and income levels benefitting less from common tobacco control programs [18, 19]. This widening of disparities may be due to the focus of current smoking cessation procedures on individual behavior modification (e.g. education about smoking harmful effects, pharmacological interventions for nicotine dependence) as opposed to programming with a focus on improving the social conditions which shape smoking behavior (e.g. poor educational attainment) [18, 20]. Similar to trends observed in LMICs, marketing practices that have led to the higher use of flavored tobacco products (such as menthol cigarettes) in Blacks, also should be regulated [21].

Esophageal Cancer

Esophageal cancer is a leading cause of morbidity and mortality worldwide. Globally, esophageal cancer is the 8^th most diagnosed cancer with an estimated 604,000 new cases and the 6^th most common cause of cancer death with 544,000 deaths in 2020 [22]. The majority (80%) of all cases of esophageal cancer occur in developing countries [23]. Of the two main histologic subtypes, esophageal squamous cell carcinoma (ESCC) is the most common, particularly in East Africa and East Asia [24]. Esophageal adenocarcinoma (EAC) is the most common histology in Western countries, and its prevalence has increased in recent decades [25].

Risk factors for esophageal SCC include tobacco and alcohol (which increase risk synergistically), water pollutants, and dietary factors including micronutrient deficiencies, and consumption of hot beverages [26, 27]. In addition to tobacco use, risk for EAC is most strongly linked with Barrett’s esophagus, obesity, and gastroesophageal reflux disease – which is also increasing in incidence in North American and Western European populations [26, 28]. These socially determined risk factors partially explain EAC’s geographic distribution [29, 30].

In the US, incidence of EAC is now greater than the incidence of ESCC [31]. EAC is most prevalent in White populations and is more common in men than women [32]. ESCC incidence, however, has been reported to be as much as four times greater in Black than White patients [32]. Similarly, data from SEER indicates that Black patients have worse disease-specific survival for ESCC compared to all other ethnic and racial groups, with a mortality rate almost double that of White patients with ESCC (7.79 vs. 3.96, p<0.05) [33, 34]. Analysis of the prevalence of risk factors for ESCC in the U.S. can help to explain these disparities. For example, recent data from the National Health Interview Survey indicates that smoking rates remain higher in Blacks than Whites, especially for men, although alcohol consumption is reportedly lower in Blacks than Whites [35, 36]. Disparities in access to healthy food could also play a role in the increased prevalence of ESCC in Blacks. Research has demonstrated that predominantly Black neighborhoods are more likely to be food deserts, where access to affordable, healthy food is limited and poor diet has been shown to correlate with elevated ESCC risk [37, 38].

Lung Cancer Screening

Lung cancer survival is poor unless detected and treated at an early stage. Screening with low-dose computed tomography (LDCT) for patients with high risk reduces lung cancer mortality (24% and 33% reduction at 10 years for men and women, respectively) [39–41]. Those at high risk include smokers and older adults [42] as the vast majority of lung cancer cases are attributable to cumulative smoking [43]. Therefore, current guidelines recommend screening individuals in the 50 to 80-year-old age group who have at least a 20 pack year smoking history and currently or formerly smoked within the last 15 years [42, 44, 45].

Racial disparities exist in both eligibility and access to lung cancer screening. Prior to the 2021 update, the U.S. Preventive Services Task Force (USPSTF) required at least a 30-pack year smoking history and age 55–80 years old to be eligible for lung cancer screening [44]. This guideline was generated based on data from large trials such as the National Lung Screening Trial (NLST), of which only 5% of participants were Black [39]. Black smokers have been shown to be use fewer cigarettes than White smokers yet have a higher risk of lung cancer even when controlling for socioeconomic and other risk factors [46]. Additionally, lung cancer risk is higher in younger age groups among Black men when compared to White men [47]. Risk models that include race, such as the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial model have been proposed to quantify these differences in risk [48]. While the 2021 update to the USPSTF lung cancer screening guidelines have improved the sensitivity of lung cancer screening in Black smokers, further evaluation of these screening guidelines will be needed to continue to mitigate racial disparities [49].

Further, the application of lung cancer screening has been low in certain groups, including women, patients with low socioeconomic status, and minorities. Despite a rising incidence of lung cancer in women, only 4% of women participate in screening [50]. Similarly, underrepresented racial and ethnic populations have had low uptake of lung cancer screening [51]. Further studies analyzing lung cancer screening note geospatial disparities in screening behavior, finding evidence of lower participation in screening programs across the South and reduced access to screening facilities in many rural areas [52, 53]. Insurance status remains a major barrier to screening, especially for those on government insurance. In order to qualify for reimbursement through Medicare or Medicaid for lung cancer screening, a shared-decision making visit is mandatory, and the center must submit screening data to a national registry [54]. While these criteria were designed to ensure high-quality screening practices, they also present significant logistic hurdles for the implementation of lung cancer screening programs, which could heighten disparities in areas already underserved by the healthcare system [55]. Further difficulty accessing lung cancer screening may result from the fact that coverage for Medicaid patients is determined at the state level. In states without Medicaid expansion, lung cancer screening is not required to be covered [56]. As patients with low socioeconomic status have high rates of smoking, this is a high-risk population that will lack access to screening in certain states [35]. Even when patients experiencing health disparities are able to be screened for lung cancer, studies have shown less than 35% adherence to follow-up care and subsequent screening and that Black patients are less likely to follow-up after a positive lung cancer screen [57, 58].

Surgical Treatment Access

Multiple sociodemographic disparities exist in the access to treatment of thoracic cancers. For patients with early-stage non-small cell lung cancer (NSCLC), surgical resection has been established as the standard of care [59]. Nonetheless, in the U.S., it has been consistently demonstrated that Black and elderly patients, patients with low socioeconomic and education status, and patients on government insurance have significantly lower odds of receiving guideline concordant care for early-stage NSCLC [60, 61]. Surgical access disparities for Black patients are among the most well documented. Since the last decade of the 20^th century, researchers began to quantify the surgical disparities confronted by Black patients with NSCLC, finding evidence that they were significantly less likely to undergo surgical resection, a potentially curative procedure, when presenting with early-stage disease [62, 63].

Despite this knowledge of disparities in offering lung cancer care, for more than two decades, studies have also found that Black patients are still less likely than White patients to receive guideline concordant care for their NSCLC, regardless of disease stage [64]. A recent study noted that Black patients are still almost 25% less likely to be offered surgical resection for stage I NSCLC than White patients and more likely to receive radiation therapy or chemotherapy alone than surgery [65]. In patients with locally advanced lung cancer in the U.S., minoritized patients and those residing in a region with lower education or low median incomes are also associated with lower rates of guideline concordant care [61]. This is important, as advanced lung cancer patients managed in accordance with current guidelines have been shown to have superior overall 5-year survival (51.6%) compared to those treated with non-concordant therapies (36%) [66].

Even for patients offered surgery, disparities persist. Race has been reported to be a risk factor for surgical deferral, with one recent study finding that Black patients were over 2 times more likely to decline surgical management of their early-stage NSCLC than White patients [67]. A related study used the SEER database to demonstrate that Black patients referred to surgical providers within 6 months of NSCLC diagnosis were half as likely as a White patient to have their cancer resected; this analysis also found a significant difference between surgeons in the likelihood of recommending surgery to Black versus White patients, suggesting provider-level bias may play an important role in perpetuating this disparity [68]. Although the surgical rate gap remains to be definitively closed, at least one recent study has reported evidence that it may be narrowing nationwide, down from over 10% in 2004 to just 3.8% in 2015 [69]. Subsequent analysis has demonstrated, however, that the trend toward parity in surgical resection rates at a national level obscures lingering and—in certain cases—widening disparities apparent after the data was broken down by region or facility type [70].

The surgical rate gap is not the only indication of the Black-White disparity in the treatment of lung cancer. Another such indication is the difference in “time-to-treatment”—the time elapsed from radiologic diagnosis of NSCLC to the initiation of definitive treatment, surgical or otherwise, in Black and White patients. Using the NCDB, one group found that Black race was a significant risk factor for delayed time-to-treatment, accounting for a 22% increase in time between diagnosis and initiation of therapy relative to a White patient [71].

Esophageal cancer is also subject to multiple treatment-related disparities. In the U.S., race has been shown to influence esophageal cancer mortality: Blacks are more likely to be diagnosed with esophageal squamous cell carcinoma, which is associated with a poorer prognosis. Black patients are also more frequently diagnosed at a later stage than Whites [72]. Black patients of all stages in the U.S. are less likely to undergo surgery than White patients [73]. In stage I-III esophageal cancer patients, Blacks, women, those with government insurance, and those in regions with lower education were more likely to be offered no treatment at all. Ultimately, programs to minimize exposure to risk factors and detect disease earlier will be needed to improve the outcomes in esophageal cancer as the 5-year overall survival of patients with Stage I disease was higher in those that underwent surgery (54.2%) than those who received no treatment at all (34.1%), with no difference in survival differences observed in resected versus unresected Stage II and III disease [74]. The issue of late presenting stage impacts survivability across the global “hot spots” of this disease. In East Africa, as many as 90% of patients present with inoperable disease [75].

Outcomes in Thoracic Surgical Oncology

For patients diagnosed with lung cancer, the prognosis is historically poor. Global estimates of 5-year overall survival (OS) across all stages ranged from 10–20% in the years 2010–14, with recent U.S. data demonstrating a 5-year OS of 19% [76, 77]. For patients in the U.S. with NSCLC, those diagnosed with stage I disease have significantly improved 5-year survival at 68% compared to 6% in those with stage IV lung cancer [78]. Unfortunately, only 29% of lung cancers are identified at early stages, compared to the 44% of patients diagnosed with stage IV disease at presentation [78]. While the USPSTF guidelines do attempt to diagnose more patients with early-stage disease. screening has failed thus far to rectify existing disparities in early detection of lung cancer. Hence, Black individuals, are diagnosed with advanced stage disease more frequently than their Caucasian counterparts (49% of Black patients compared to 45% of White patients) [79]. In addition to racial disparities, low socioeconomic status has been shown to correlate with advanced stage at diagnosis [80].

Despite the persistence of disparities associated with sociodemographic and socioeconomic variables in access to thoracic cancer treatment, recent studies on the correlation between these same factors and thoracic surgical outcomes themselves provide some indications for optimism. In patients receiving similar treatment strategies, it has been shown that the 5-year OS for Black and White patients with early-stage lung cancer does not significantly differ [65, 67]. This result suggests that the disparity seen in unadjusted 5-year OS reported in the SEER database is not the result of biological differences in the lung cancer itself [81]. These data further suggest that it is primarily the disparities in access to equitable treatment which drive the disparities in lung-cancer mortality between patients, underlining the continued importance of efforts to reduce treatment gaps between groups.

Survival is not the only measure of thoracic surgical outcomes where disparities persist. Other areas include access to immune- and targeted therapy, adequacy of oncologic procedures, and short-term surgical outcomes. In the IMpower010 trial, for example, White participants outnumbered Black participants 123:1 [82]. This imbalance is in keeping with a longstanding precedent in randomized clinical trials, where enrollment has historically been predominantly White [83–86]. As others have argued, however, this disparity deserves even greater attention in the era of targeted and immunotherapies, as bias towards White patients in research populations may lead to selection algorithms for these drugs which differentially exclude racial and ethnic minority patients, thus perpetuating disparities in oncological outcomes which have been observed for decades [87]. Beyond the trial setting, ensuring equitable access to these drugs is vital, as data from other surgical and oncological specialties demonstrates that the prohibitive cost of novel therapies is already widening the existing gaps in care and outcomes between socioeconomic and sociodemographic groups [88, 89].

One recent retrospective analysis found that even when they underwent surgical resection for early-stage lung cancer, Black patients underwent a less extensive lymph node dissection during pulmonary lobectomy, with the number of lymph node stations sampled and the total number of lymph nodes taken being significantly less in Black patients than in White patients [90]. Low nodal sampling rates can contribute to stage underestimation, which can in turn impact prognosis, recommendations for adjuvant therapy and overall survival. Similar disparities exist in the perioperative space as well, with evidence demonstrating that Black patients experience significantly longer hospital stays and intubation periods after pulmonary lobectomy [91].

The factors influencing these disparities are not limited to race alone, as low socioeconomic status has also been shown to correlate significantly with perioperative complications, including in-hospital death, following oncological lung resections [92]. Another important axis of outcome disparity in patients with surgically resected NSCLC is insurance status. Reports in the literature suggest that nonprivate insurance status (i.e., insurance via Medicare or Medicaid) is associated with reduced overall survival and a less favorable disposition at discharge [93, 94].

Finally, the equality amongst outcomes in global settings remains an elusive goal. For example, esophageal cancer, which portends a particularly poor prognosis, with an overall 5-year survival of less than 20% worldwide, has both the highest incidence and highest mortality rates in East Asia and Southern and Eastern Africa [95]. Notably, a region’s sociodemographic index has been negatively correlated with esophageal cancer morbidity and mortality rates [95].

Conclusions

The disparities in thoracic surgical oncology discussed in this review are multifaceted and complex. There are emerging data regarding the intersection of race, insurance and financial status, and education levels which must be examined given their impact on the social determinants of health and the well-being of our society as a whole. As heath care evolves, access to emerging technologies including digital health innovations, hybrid surgery, and new periadjuvant regimens are areas to monitor for future inequal distribution in access [96]. Increased awareness of these issues, in combination with systems level interventions to target the source of disparities, will lead to improved delivery of care by surgeons and our colleagues for thoracic oncology patients worldwide [97].