Socioeconomic factors such as income and level of education have been shown in numerous studies to be independent prognostic factors for undertreatment and poor survival in lung cancer [3, 4]. However, some experts believe that the contributions may not be equivalent—i.e., although education level may determine the onset of illness, income may be a stronger predictor of its outcome [5].

Patients from low-income families are clearly at higher risk for developing and dying from lung cancer. Studies have shown that patients from areas with the highest poverty levels (>15 %) are more likely to present with larger (>5 cm), poorly differentiated tumors with distant metastasis at diagnosis compared to patients from more affluent communities [6]. This disparity in outcomes can be explained by several factors.

Due to lack of resources, low-income patients tend to live in impoverished communities where there is a lack of access to: (1) grocery stores with fresh fruits and vegetable, (2) parks and other exercise facilities, (3) specialized cancer hospitals and physicians, and (4) adequate public transportation which can allow them to seek help elsewhere. On the flip side, these communities are often rich in primary and environmental tobacco exposure, illicit drug use, and crime [1, 7–9].

Because of the nature of their jobs, low-income workers tend to be exposed to more carcinogens—including second-hand smoke, asbestos, radon, and cooking or chemical fumes. They often have little flexibility in their work schedule or are afraid to ask for time off for medical appointments, especially when dealing with inherent delays at public hospitals. The lack of a car may also discourage compliance with serial treatments and follow-up visits.

Clearly, low-income patients cannot afford good insurance coverage or out-of-pocket costs for cancer treatment, which are among the most expensive medical therapies available in the United States today. Thus, even when diagnosed at an early stage, these patients may still have poor outcomes if they cannot take action immediately. Their differing priorities and competing economic pressures may buffer their sense of denial and lead to further procrastination.

With the passage of the Affordable Care Act in 2012 , this aspect may diminish as public health initiatives such as universal insurance coverage improve access to care. However, disparities in lung cancer outcomes may still exist if the working and living environments and high-risk behaviors of low-income patients do not change concurrently.

No one can deny the fact that income and education are directly linked—higher education leads to higher paying job opportunities. But the extent of schooling can also impact lung cancer outcomes in an additional way as well.

Access to care depends on not only ability to pay, but also patient desires and motivation to seek the most aggressive and appropriate treatment possible. Studies have shown that the best lung cancer outcomes are usually seen at facilities which follow NCCN guidelines and evidence-based algorithms, often with the guidance of a multidisciplinary lung cancer team of experts. However, this is not universally available at all hospitals and most often occurs at an academic center, comprehensive cancer institute, or integrated systems such as Kaiser Permanente or the Veterans Health Administration. Ironically, some studies show that private community hospitals have the most inconsistent, non-guideline-based care [10].

The level of education can directly impact which type of hospital the patient selects. Poorly educated patients may seek care at public teaching hospitals, where resources are often limited and care may be delayed. Patients with higher education tend to research their disease state online and seek care at the best hospitals possible, even if it is quite a distance away. They understand the need to be aggressive and are willing (and able) to sacrifice time and money to pursue the most effective treatment modalities at centers with the most experience. Their education also makes them more proactive and knowledgeable about the value of screening and early diagnosis. In addition, highly educated patients tend not to smoke, although several studies have shown that the effects of childhood smoke exposure do not dissipate and contribute to increased risk for lung cancer later in life.

The current push for more widespread public education about the dangers of smoking, lung cancer screening, and available treatment options does not address the most basic deficit of all—which is improving access to education in general. If an adequate educational foundation can impact lung cancer outcomes, then societal changes through legislation may actually alter the course of this disease more quickly and effectively than any major breakthrough in advanced cancer technology.

Race is a commonly identified risk factor for poorer outcomes in lung cancer [11]. On the surface, it may imply genetic differences between ethnic groups such as African Americans, Asians, and non-Hispanic Caucasians—e.g., African Americans appear to be genetically more vulnerable to the damaging effects of cigarette smoke [12] while East Asians often carry targetable EGFR mutations [1]. Or it may suggest cultural differences in the processing of information, mistrust, or acceptance of western treatment methods [1, 13–15]. Or it could refer to actual provider discrimination in the diagnosis and management of lung cancer. But in reality, race also implies socioeconomic and educational disparities, which clearly can impact access to timely and appropriate care.

Despite the election of an African American president and major landmark legislation over the past century to reduce social inequalities, racial discrimination still exists in the United States. There is no other way to explain why the poorest zip codes in the nation, defined by a poverty rate of >20 %, are communities comprised predominantly of ethnic minorities, usually black or Hispanic [9]. For example, racial analysis of the 11 Surveillance, Epidemiology, and End Results Program (SEER) areas demonstrated that African Americans, Hispanics/Latinos, and American Indians/Alaskan Natives live in the most impoverished areas [9]. In another study, black patients were 26 % more likely than whites to reside in poor communities, and much like the subgroup of low-income patients, they present with larger tumors which are often undifferentiated and have distant metastasis at diagnosis [6].

Indeed, many experts feel that when socioeconomic variables are adjusted for, no differences in lung cancer mortality can be seen as a result of race. In an analysis of almost 11,000 SEER Medicare patients, Bach et al. found that the most important factor impacting survival among different races was the receipt of surgical resection; compared to their white cohorts, blacks were 12.7 % (p < 0.001) less likely to receive this appropriate intervention [11]. Similarly, in as study of over 900 patients at the Reed Army Medical Center, Mulligan et al. found that if universal access to appropriate medical care was available, no racial disparities in lung cancer mortality could be observed [16]. Multiple other authors concur with this finding of socioeconomic factors driving the differences in racial lung cancer outcomes [6, 16–20]. Berger et al. embodies this popular opinion by stating that the “link in support of a biologic or genetic difference among different racial or ethnic groups to explain the disparity in outcomes is still weak” [21].

There are a few papers which disagree with this consensus. Farjah et al. proposed that studies to date on this issue have not measured important differences in FEV1, comorbidities, or performance status between races which may also account for discrepancies in outcomes [17]. Williams et al. feels that, even when patients of color have the same education and income level, as well as the same histology and stage of lung cancer, the outcomes may still be worse than those for their Caucasian counterparts. For example, although they may have equivalent mortality rates at the lowest socioeconomic levels, black men are more likely to die from lung cancer than white men within the highest educational brackets. Similarly, when their level of schooling is low to midlevel, black women have a lesser or equivalent risk of death than their white counterparts; however, when they are highly educated, black women are more likely to die from lung cancer than white women. Thus, although more schooling improves overall outcomes within a racial group, it paradoxically unmasks and underscores survival differences between racial groups, despite perceived educational and economic equivalence [1].

In addition, the damaging effects of cigarette smoking seem to be racially and genetically based. Evidence suggests that not only are African Americans more vulnerable to the toxic effects of tobacco use [12], but nicotine sensitivity is higher and clearance is lower, thus contributing to low sustained quit rates, despite the fact that they tend to start smoking later in life and smoke fewer cigarettes overall [22, 23]. However, subgroup analysis demonstrates that, at every socioeconomic level, black women smoke less frequently than white women, perhaps due to a higher level of religious involvement. This may also explain the same trend in adolescent African Americans [24, 25]. Certain ethnic immigrants are also less likely to smoke than their American-born counterparts [26] and have a correspondingly lower lung cancer mortality rate [27]; unfortunately, both of these rates rise with increased length of stay in the United States [28]. Based on the SEER database, the highest prevalence of smoking was found in American Indian/Alaskan Native women (38.6 %) and men (27.4 %), and the lowest prevalence in smoking was found in Asian and Hispanic/Latino women (7.9 %) [9].

Of note, while the association between race/ethnicity and worse lung cancer outcomes has been observed for African Americans, American Indians, Alaskan Natives, Hispanic/Latinos, and Pacific Islanders, it appears that East Asian minorities may have a better prognosis overall due to a higher prevalence of more treatable, non-smoke-related cancers such as adenocarcinomas with EGFR mutations [1].

Whereas the debate continues in regards to whether racial disparities in lung cancer are due to genetic/biologic differences or due to socioeconomic factors, the impact of cultural perceptions on the delivery of care is well accepted. Wisnivesky’s group published a recent paper which found that almost 33 % of the worse outcomes among black versus white patients can be explained by their fatalism, mistrust of providers of a different racial group, and negative beliefs regarding surgery [29]. Thus, whereas socioeconomic factors may impact the incidence and timing of lung cancer diagnosis, cultural attitudes and beliefs play an important role in what unfolds afterward [14, 15].

This influence of culture on the acceptance of recommended care has been studied most extensively in African Americans. Despite having early stage resectable lung cancer and adequate pulmonary reserve, statistics show that blacks are significantly less likely to receive surgery than whites (64 % vs. 76.7 %), resulting in a much lower 5-year survival [11]. However, the striking discrepancy in operative rates is not exclusively due to socioeconomic barriers [30].

If cost and access to treatment are not issues, then an alternative explanation for continued discrepancies in care may be subconscious provider discrimination. Clearly, inherent physician biases, whether intentional or not, may lead to misconceptions regarding the patient’s willingness or likelihood of adherence with treatment and personal preferences [9]. Providers may frame the information regarding diagnosis and management options in such a way that the patient’s decision-making is skewed toward a less optimal plan [9].

However, patients may also negatively affect their own outcomes based on their cultural perceptions, regardless of how objective and well meaning the provider is. For example, McCann et al. showed 18 % of black patients refuse surgery for lung cancer compared to only 5 % of white patients [31]. Several reasons have been hypothesized as to why racial/ethnic minorities may refuse appropriate surgery for early stage disease. These include:

Depending on their education level and degree of integration into mainstream society, racial/ethnic minorities may find it difficult to navigate a complex bureaucratic medical system or may have inherent distrust of providers who are of a different race [33]. For example, a study showed that more blacks refused appropriate surgery when the diagnosis and treatment options were presented to them by a virtual physician of a different race, even if there were no perceivable differences in sense of trust, level of engagement, and communication styles [34]. In addition, based on their cultural biases, patients may be afraid of surgery compared to less invasive (but inferior) modalities or hesitate to ask for pain medications and hospice referrals when appropriate [35].

Thus, race, culture, genetics, and socioeconomic circumstances all appear to influence lung cancer outcomes simultaneously [2, 9]. As one of the best examples of the intersectionality theory, these four factors are so tightly intertwined and interdependent that race alone cannot be dissected out as an independent cause of worsened lung cancer survival.

Walter Cronkite once said that “America’s health care system is neither healthy, caring, nor a system.” According to the U.S. Census Bureau in 2009, 48.6 million people in the United States lacked health insurance [36]. This translates into approximately 45,000 excess preventable deaths per year, or equivalently, a person dying every 12 min [37]. Yet even when patients are insured, access to medical care does not always translate into better outcomes. Depending on type of insurance, hospital resources, and whether treatment decisions are guideline driven or seeped in individual preferences, the care given may not be high quality or appropriate.

Clearly, the absence of health insurance can directly block access to nonemergency care. However, even with insurance, disparities in lung cancer outcomes can still exist based on the type of coverage.

Previously, Slatore et al. published an exhaustive systematic review of available literature on the association between type of insurance and survival from lung cancer. As expected, they found that patients with no insurance or only Medicaid consistently had worse outcomes than patients with Medicare or private insurance. This disparity may be rooted in the fact that indigent and underserved patients are more likely to smoke and avoid routine health maintenance, thus presenting to medical attention at a very late stage of lung cancer. But this disparity may also result from lack of access or undertreatment of these patients due to their insurance status [38].

Bradley and colleagues published a series of four papers that evaluated the impact of having Medicaid on various subpopulations of lung cancer patients. At all stages of lung cancer, regardless of age or gender, it appears that all-cause and cancer specific mortality rates wer e much higher in comparison to patients with Medicare or private insurance [39–43]. Interestingly, other studies have found that patients with combined Medicare/Medicaid did worse compared to patients with Medicare alone [44], and even patients who underwent curative-intent surgical resection had a discrepancy in outcomes related to Medicaid status [45].

Medicaid is not unique in being associated with poor outcomes. Compared with either health maintenance organization (HMO) or fee for service (FFS) coverage alone, the combination of HMO and FFS has been associated with increased all-cause mortality (but not lung cancer-specific) for unclear reasons [46]. This is interesting because patients with the latter combination were more likely to receive surgery compared to patients with only FFS plans [46]. Furthermore, other studies showed that private/commercial insurance was associated with higher surgery rates compared with non-commercial insurance [47, 48].

Although there is no arguing that uninsured and Medicaid patients have the worst outcomes in lung cancer [29, 38], there is conflicting data in regards to which kind of other insurance is superior—Medicare or private/commercial. Based on a study by Potosky et al., it appears that private insurance status was associated with lower rates of guideline concordant care, and that ironically, patients with public insurance, or a mixture of public and private plans, were more likely to receive stage-appropriate care [10]. In contrast, Harlan and colleagues found that adherence to NCCN guidelines was highest among those with private insurance and lowest among those with non-private insurance, with uninsured patients falling somewhere in between [49]. Groth et al. also found that patients with private insurance were more likely to undergo lobectomy for early stage lung cancer than patients with Medicare, Medicaid, or no insurance at all [50].

Finally, even within a single academic medical center, where 29 % of patients were covered by an indigent care plan (defined as Medicare or a “county” health plan), discrepancies in outcomes were found based on insurance type. As Yorio et al. discovered, the odds of receiving “standard therapy” were dramatically reduced for patients who were covered by an indigent plan compared to private insurance. Among these socioeconomically disadvantaged patients with early stage non-small cell lung cancer (NSCLC), the hazard of death was almost twice as high and they were less likely to undergo surgery compared to privately insured patients (OR 0.13, 95 % CI 0.04–0.43). At first, the authors considered patient factors as an explanation for these findings; their indigent patients tended to be nonwhite male smokers and usually presented at a more advanced stage of disease with non-adenocarcinoma histology. They also often had smoking-related comorbidities which affected their surgical candidacy. However, even after controlling for all of these variables, socioeconomic status (and resultant insurance type) remained an independent risk factor for undertreatment and poor survival. Interestingly, although ethnic minority status also had a trend toward treatment disparities, it was not statistically significant [51].

Another factor accounting for disparities in lung cancer outcomes is hospital type. Even if patients have insurance and access to care, the characteristics of the treating institution can influence the likelihood of survival. Since surgery offers the best chance at cure for early stage disease, the frequency at which lung resection occurs at a particular hospital can help explain discrepancies in mortality rates [52].

Hospitals that traditionally have lower surgical volumes and no dedicated thoracic surgeon tend to offer lung cancer resection less frequently due to their lack of experience and expertise [52–55]. Thus, it is no surprise that seeking care at smaller hospitals may lead to a poorer prognosis.

It is also not surprising that large safety net and public hospitals tend to offer surgery less often to lung cancer patients due to a lack of resources and the fact that many of their underserved patients present with advanced stage disease [39, 52]. These patients are usually indigent or undocumented, and thus have Medicaid or no insurance at all. As a result, their insurance type can drive hospital selection, which then, in turn, impacts surgical decision-making, thus leading to further disparities in lung cancer outcomes. The exception to the rule, however, is county teaching institutions, where the chances of getting surgery are much higher [52].

What is interesting, however, is that the racial composition of the hospital patients also plays a role. Lathan et al. compared hospitals with a higher percentage of Medicare-insured black patients with those who served primarily white patients; they found that if the racial composition was ≥30 % African American, surgery was less likely to occur for early stage disease. This was in addition to, rather than because of, racial disparities. All patients in those hospitals, regardless of race, received less surgery [52].

Therefore, hospital type is a key determinant of surv ival rates for lung cancer. The Affordable Care Act, which theoretically will improve access to care by providing universal insurance coverage, may not solve this disparity if the quality of care at all hospitals is not standardized. Ideally, hospitals should follow national guidelines in lung cancer treatment and refer their patients to the closest major cancer center if they are unable to provide guideline-based care because of lack of experience or resource limitations.

The popular paradigm put forth in the best-selling book, “Men are From Mars, Women are From Venus” may hold true in more than just the psychology of relationships. From the perspective of lung cancer outcomes, the disparities between men and women are so striking that many experts believe they may actually represent different disease states [56–58] that require alternate approaches to treatment. In the era of personalized molecular medicine, where one glove no longer fits all, this not only translates into the need for individually tailored therapies, but we also need to rethink the design and interpretation of clinical trials with such differences in mind [56, 57, 59–61].

In light of this new appreciation of biological differences, it is ironic that the epidemic of lung cancer among women may have resulted in part from a desire to be more like men. Since the nineteenth century, women have faced inequality in several arenas, and in their quest to break free from traditional stereotyped roles, more and more women adopted what had been a predominantly male habit—smoking. Since the 1940s, tobacco companies aggressively portrayed smoking as a sign of independence among women, and after a latency period of several decades, we are now seeing the unfortunate consequences in recent years [60]. Although smoking cessation campaigns have been successful in reducing overall female tobacco use since the 1960s [57, 61], the habit continues to increase in prevalence among adolescent girls. That this demographic group is at risk should not be surprising given the fact that men and women may smoke for different reasons, with the latter more frequently using cigarettes to boost their self-image; thus, teenage girls with depression and weight issues are particularly vulnerable [57, 59, 60, 62, 63].

As a result, some experts estimate that the incidence of female lung cancer in the U.S. has jumped by 600 % in the past 50 years [57]. Since 1987, lung cancer has exceeded breast cancer as the number one cause of death among women [64]. Annually, 30,000 more women die of lung cancer than breast cancer, and the gap continues to widen [58]. Part of this discrepancy may be due to effective public education, vigilant screening, and improved treatments for breast cancer. However, a more concerning possibility is the fact that the biology of lung cancer in women is different from that in men, and up until now, we have been treating them the same.

Multiple studies have shown that female lungs are more vulnerable to the effects of carcinogens [58, 63–65]. This may be partially attributed to the fact that they have reduced pulmonary reserve. Although the lungs appear to develop at a similar pace for both genders throughout childhood, at puberty boys have a major growth spurt which results in larger airway calibers with less bronchoreactivity [58]. Given the same age and dose of tobacco exposure, women suffer more damage and decline in FEV1, with a resultant higher susceptibility to COPD and lung cancer compared to men [58, 62]. In fact, females are at a 1.5–2.7-fold increased risk of developing lung cancer compared to males [65, 66].

Early second-hand smoke (SHS) exposure also seems to plays a role. Compared to children raised in a nonsmoking household, those who grew up surrounded by SHS have 30 % increased odds of developing lung cancer as an adult if exposed during the first 25 years of life, yet their risk of lung cancer is reduced dramatically if exposed after the age of 25 years [67]. Yet upon stratification by gender, there is a disparity in this window of vulnerability. Whereas the damaging effects of SHS appear to plateau at age 20 for males, this risk threshold extends up to age 25 for females [58, 63]. Thus, females are more prone to developing lung cancer than men, despite similar or lesser exposures.

Despite the surge of tobacco use in the past half century, a significant fraction of women who develop lung cancer are never smokers [57, 61, 62, 64]. In the United States, approximately 15–20 % of lung cancers occur in nonsmokers, and of these, 70–80 % are women [62]. This phenomenon of lung cancer in female nonsmokers may be partially explained by unintentional exposure to environmental carcinogens, such as indoor fumes from cooking, residential radon, and second-hand tobacco exposure, which can increase the risk of lung cancer by 24–30 % [56, 63, 66].

Of historical interest, the segregation of women to the home and men to the workplace in past decades resulted in distinctly different patterns of exposures accounting for lung cancer in nonsmokers. Among nonsmoking men who developed lung cancer, the chief carcinogens they were exposed to were found in the workplace: asbestos, radon, and chemical toxins. The majority of nonsmoking women, however, developed lung cancer as a result of second-hand smoke from their husbands [68]. These patterns of predominant carcinogen exposure are changing as a result of changing roles of men and women in the workplace and home. As smoking cessation efforts continue and laws are passed banning cigarettes in work and public places, the impact of tobacco on the development of lung cancer may diminish, and other exposures and factors (such as genetics) will become more important contributors to lung cancer.

Despite a higher incidence and greater vulnerability to lung cancer, women have one major advantage over men: the type of lung cancers they develop tends to be less advanced and more responsive to treatment. Multiple studies have shown that regardless of the histology and stage, women appear to have “superior survival” statistics compared with men [58, 59, 61, 62, 64, 69, 70].

Compared with men, lung cancer in wome n tends to develop at a younger age, is detected at an earlier stage, and is predominantly adenocarcinoma that is frequently associated with EGFR mutations [56, 58, 61, 62, 71]. On average, these features result in better survival and fewer treatment side effects, if not absolute cure. In the presence of EGFR mutations, treatment with TKI have shown a response rate of up to 75 % in advanced NSCLC [72]. Even when their tumors lack actionable mutations, however, women tend to have better responses than men to standard platinum-based chemotherapy regimens [60]. And if they are fortunate enough to be diagnosed with early stage disease, women historically have had lower operative mortality rates than their male counterparts [57, 63].

In contrast, lung cancer presents very differently in men. Although adenocarcinoma is still the most prevalent subtype, smoking-related forms such as squamous cell and small cell are seen more frequently than in women [62]. Men with lung cancer tend to be diagnosed at a more advanced stage and frequently have tumors that lack EGFR mutations and are less responsive to surgery or chemotherapy. Thus, while women are at higher risk of lung cancer, they tend to develop forms which are milder and easier to treat, with better overall prognosis than men [70].

These distinctions in lung cancer presentation and outcomes by sex support the idea that biological and behavioral differences between men and women may influence lung cancer disparities. For example, cervical infection with HPV can lead to airway infection by HPV—either due to hematologic dissemination or risky sexual practices—which can then result in viral-induced squamous cell airway tumors, particularly in Asian women [56, 64, 69, 73]. Another example is HER2/neu or BRCA mutations, which are commonly associated with breast cancer, and are now being found in lung cancer tissue as well. Although studies have not shown benefit in giving lung cancer patients trastuzumab (a monoclonal antibody to HER2/neu) in addition to standard chemotherapy [74], this remains a potential target for therapy in the future. BRCA1 overexpression appears to increase the sensitivity of tumors to taxol therapy and therefore may help tailor treatment [64].

A high estrogen milieu (due to endogenous estrogen production or use of hormone replacement therapy (HRT) after menopause) is also being implicated in the pathogenesis of lung cancer and is a potential therapeutic target. No longer considered just a “sex steroid” involved in breast, ovarian, and endometrial cancers, estrogen is now being investigated as a controller molecule for abnormal proliferation in the lungs, acting as a direct carcinogen in forming DNA adducts [60, 62]. In vitro studies have shown a 17-fold increase in proliferation of lung cancer cell lines when incubated with B-estradiol. Early menopause, which ironically increases the risk of cardiovascular disease in women, seems to decrease the risk of lung adenocarcinoma in women [57, 69].

The role of hormone replacement therapy (HRT) in the pathogenesis of lung cancer is controversial. Initially, the Women’s Health Initiative study suggested a higher rate of lung cancer in female smokers on HRT. However, a post-hoc analysis found that this was only true when patients were supplemented with both estrogen and progesterone, not estrogen alone. Paradoxically, other studies have found a protective effect of HRT against the development of lung cancer in current smokers, so the jury is still out in regards to whether estrogen exerts a positive or negative impact on lung cancer [58, 61, 69, 75, 76].

Estrogen receptors (ER) may provide another therapeutic target in the battle against lung cancer in women. Alpha and beta ER are normally found in pulmonary tissue; the presence of ER-beta, however, has been demonstrated in 45–70 % of resected NSCLC tumors in both genders [56, 62, 77]. Soy is known to competitively bind to ER, and therefore high soy intake may exert a protective effect against lung cancer [69]. Studies have also shown decreased lung cancer mortality (although no reduction in incidence) in women who are on tamoxifen for a history of breast cancer [78]. Genetically, ER-beta seems to track with EGFR mutations in many patients and may provide another surrogate marker for selection of TKI candidates [79]. In fact, concurrent administration of fulvestrant (ER blocker) with gefitinib has been shown to sensitize the tumor to TKI and thus may synergistically increase response rates [80, 81].

A final potential therapeutic target may be aromatase. This enzyme catalyzes the conversion of androgens into estradiol and is especially important in post-menopausal women. Even when the estrogen milieu is low, lung cancer can produce its own local estrogen via high levels of aromatase, and this has been found to be a poor prognostic marker [60, 62, 77]. Aromatase inhibitors may therefore be useful in reducing triggers for proliferation [77].

In addition to the obvious hormonal differences between men and women, women may carry other types of genetic alterations which may explain both their increased vulnerability to the development of lung cancer and their superior survival compared with men. Because of the presence of polymorphisms in CYP detoxification enzymes, women tend to express more DNA adducts compared to men, have a higher level of CYP1A1 which activates certain carcinogenic polycyclic aromatic hydrocarbons (PAH), and have a reduced level of GSTM1 which prevents the clearance of toxic metabolites [57, 58, 60, 62, 69]. This accumulation of reactive intermediates in women may not only increase the risk of lung cancer [60, 64] but may also help explain the higher incidence of COPD and FEV1 decline in women with minimal tobacco exposure, both of which are also independent risk factors for the development of lung cancer [58].

Female patients have also been found to have a lower DNA repair capacity (DRC), which is necessary for fixing nucleotide mismatch or mutations [57, 58, 60, 62]. This may paradoxically increase the risk of lung cancer and yet simultaneously improve the response to platinum-based chemotherapy. Platinum-based agents work by forming DNA adducts which halt the cell cycle. An inability to repair these DNA adducts encourages apoptosis within cancerous cells [60]. Studies have shown improved survival in Stage 4 NSCLC when patients who have lower DRC are given standard chemotherapy [82].

Not all genetic defects in women confer an unfavorable outcome, however. EGFR mutations are the main driving mutations behind 8–10 % of NSCLC, especially adenocarcinomas, and are typically found in women, nonsmokers, and Asians. Three major studies have documented a statistically higher rate of EGFR mutations in women compared to men—ranging from 19 to 38 % in females compared to 9–14 % in males [64]. This lends women a distinct survival advantage because of the exquisite sensitivity of these tumors to TKIs, which have demonstrated superior outcomes in clinical trials relative to traditional chemotherapy [83, 84]. In addition, EGFR mutations are often found in adenocarcinoma in situ (AIS, formerly known as BAC), which due to its indolent growth pattern already has a better prognosis than other forms of lung cancer. In addition to their predilection for EGFR mutations, women are also two- to fourfold more likely than men to have AIS [56, 62, 69].

Of note, women have also been found to carry Kras mutations more frequently than men (26.2 % vs. 17.4 %) [85]. Kras mutations are present in approximately 10–30 % of adenocarcinomas, tend to be mutually exclusive to EGFR mutations, and unfortunately have no specific targeted therapy [62]. Indeed, studies have found increased mortality when patients with Kras mutations are given TKIs [72]. Women with Kras mutations have a worse prognosis.

Finally, the discussion of gender and lung cancer outcomes would not be complete without addressing lung cancer in gay, lesbian, or bisexual patients [86]. This highly marginalized group may be the most neglected of all minorities because most surveillance and research databases have no options for alternate sexual preferences, including the U.S. Census and SEER [86, 87], and therefore, there is less data on disparities. As a result, health disparities for this minority group have received little attention until recently.

Currently, the few epidemiologic studies that exist suggest that up to 6 % of all patients seen by physicians in the United States self-identify as members of a sexual minority group. However, many physicians are untrained and report feeling uncomfortable in addressing special issues related to treating this patient population. Studies show that almost 40 % of clinicians are “sometimes or often” uncomfortable providing medical care to gay patients, and 67 % of health providers report seeing “substandard care” being given to members of sexual minority groups [88]. Indeed, it is estimated that only 3 h and 26 min are devoted to the teaching about special health issues related to homosexuality in 4 years of medical school [89]. But compared to the general public, members of this minority group exhibit a much higher rate of smoking, alcoholism, and drug use, and thus patients from sexual minority groups represent an under-appreciated, under-treated, high-risk demographic for lung cancer and other diseases [86–88, 90, 91].

Interestingly, despite an equal understanding of the dangers of smoking and support of banning smoking in public places [87], members of sexual minority groups smoke heavier and longer than their heterosexual counterparts—approximately twice the volume, starting earlier in their youths, and with much lower quit rates [86, 87, 90]. Smoking rates among gays, lesbians, and bisexuals ranges from 38– 59 % in youths and up to 50 % in adults, compared to an average of 28–35 % and 28 % in heterosexual adolescents and adults, respectively [90].

Given these facts, it should be no surprise that the incidence and mortality of lung cancer is much higher among gay men and bisexual patients, with the same pattern of racial disparities (blacks with worse outcomes than whites) as that seen in the general population [86]. In contrast, the impact of smoking may be less pronounced in lesbian women than their heterosexual counterparts. Despite greater tobacco exposure, lesbian women appear to have statistically better lung cancer survival. The only explanation proposed for this perceived resilience against the damaging effects of tobacco thus far is lagtime bias; some experts feel that lesbian women, as a group, only started to smoke heavily recently. Thus, their lung cancer incidence may increase over time and actually peak a few decades from now. Based on these findings, it seems the disparities in lung cancer outcomes among sexual minorities may be a predominantly sociobehavioral phenomenon related to heavy smoking and lack of healthcare maintenance, rather than based on biologic and genetic differences.

The most promising solution to reducing lung cancer-related disparities in sexual minorities may involve improving existing communication and support. Part of the reason why sexual minorities are medically underserved is because they tend to avoid routine healthcare visits (including preventive services) because of fear of “insensitive” providers and perceived poor communication [88]. In addition, smoking cessation programs need to be geared toward their special needs, as their triggers for smoking relapse are usually different and specifically revolve around anxiety, depression, and the stress of living in an often homophobic society [87].

Given the complicated milieu of genetic, hormonal, behavioral, and environmental factors that may be differentially involved in the development of lung cancer among persons of different genders, future treatments must be highly personalized with the consideration of gender-specific needs to avoid disparities in care. This includes not only strategic use of targeted therapies against a variety of hormonal receptors and genetic mutations that are distributed differently among men and women, but also more preventive education and aggressive smoking cessation campaigns directed specifically at adolescent females, gays, lesbians, and bisexuals.