Disparities in Lung Cancer: A Targeted Literature Review Examining Lung Cancer Screening, Diagnosis, Treatment, and Survival Outcomes in the United States

A Systematic Review: The Role of Artificial Intelligence in Lung Cancer Screening in Detecting Lung Nodules on Chest X-Rays

Background: Lung cancer remains one of the leading causes of cancer-related deaths worldwide. Artificial intelligence (AI) holds significant potential roles in enhancing the detection of lung nodules through chest X-ray (CXR), enabling earlier diagnosis and improved outcomes. Methods: Papers were identified through a comprehensive search of the Web of Science (WOS), Scopus, and Ovid Medline databases for publications dated between 2020 and 2024. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, 34 studies that met the inclusion criteria were selected for quality assessment and data extraction. Results: AI demonstrated sensitivity rates of 56.4-95.7% and specificities of 71.9-97.5%, with the area under the receiver operating characteristic (AUROC) values between 0.89 and 0.99, compared to radiologists' mean area under the curve (AUC) of 0.81. AI performed better with larger nodules (>2 cm) and solid nodules, showing higher AUC values for calcified (0.71) compared to non-calcified nodules (0.55). Performance was lower in hilar areas (30%) and lower lung fields (43.8%). A combined AI-radiologist approach improved overall detection rates, particularly benefiting less experienced readers; however, AI showed limitations in detecting ground-glass opacities (GGOs). Conclusions: AI shows promise as a supplementary tool for radiologists in lung nodule detection. However, the variability in AI results across studies highlights the need for standardized assessment methods and diverse datasets for model training. Future studies should focus on developing more precise and applicable algorithms while evaluating the effectiveness and cost-efficiency of AI in lung cancer screening interventions.

Ethnic disparities in lung cancer incidence and differences in diagnostic characteristics: a population-based cohort study in England

Background

Lung cancer is a leading cause of mortality, yet disparities in lung cancer across different sociodemographic groups in the UK remain unclear. This study investigates ethnicity and sociodemographic disparities and differences in lung cancer in a nationally representative English cohort, aiming to highlight inequalities and promote equitable access to diagnostic advancements.

Methods

We conducted a population-based cohort study using health care records from QResearch, a large primary care database in England. The study included adults aged 25 and over, spanning the period of 2005-2019. Lung cancer incidence rates were calculated using age-standardized methods. Multinomial logistic regression was applied to assess associations between ethnicity/sociodemographic factors and diagnostic characteristics (histological type, stage, and cancer grade), adjusting for confounders.

Findings

From a cohort of over 17.5 million people, we identified disparities in incidence rates across ethnic groups from 2005 to 2019. Analysis of 84,253 lung cancer cases revealed that younger woman and Individuals of Indian, other Asian, Black African, Caribbean and Chinese backgrounds had a significantly higher risks of adenocarcinoma compared with squamous cell carcinoma than their White counterparts (relative risk ratios [RRR] spanning from 1.52 (95% CI 1.18-1.94) to 2.69 (95% CI 1.43-5.05). Men and current smokers were more likely to be diagnosed at an advanced stage than women and never smokers (RRR: 1.72 [95% CI 1.56-1.90]-2.45 [95% CI 2.16-2.78]). Socioeconomic deprivation was associated with higher risks of moderate or poorly differentiated adenocarcinoma compared with well differentiated (RRRs between 1.35 [CI: 1.02-1.79] and 1.37 [1.05-1.80]).

Interpretation

Our study highlights significant differences in lung cancer incidence and in lung cancer diagnostic characteristics related to ethnicity, deprivation and other demographic factors. These findings have important implications for the provision of equitable screening and prevention programmes to mitigate health inequalities.

Funding

DART (The Integration and Analysis of Data using Artificial Intelligence to Improve Patient Outcomes with Thoracic Diseases) project, Innovate UK (UK Research and Innovation), QResearch® and grants from the NIHR Biomedical Research Centre (Oxford), John Fell Oxford University Press Research Fund, Cancer Research UK, and the Oxford Wellcome Institutional Strategic Support Fund.

Drivers and barriers to health-seeking behaviors and interactions: a qualitative study of Black patients with lung cancer and with peripheral artery disease

OBJECTIVE

To identify factors that may influence health-seeking behaviors and health system interactions from the perspective of Black patients with lung cancer (LC) or peripheral artery disease (PAD).

METHODS

Semi-structured interviews were conducted virtually with Black patients in the United States. Thematic analysis of interview transcripts was performed. The Sense-Think-ACT-Relate (STAR) behavioral framework was used to map emerging themes of drivers and barriers to health-seeking behaviors and health system interactions.

RESULTS

Thirty Black patients with LC (n = 15) and PAD (n = 15) participated in this study. The mean age of participants was 53.4 years, 22 were female, and half lived in an urban area or large city. Factors that shape health-seeking behaviors spanned several framework domains including Trust, Rational and Emotional Associations, Cultural, and Situational. Having a provider who was friendly, knowledgeable, and understood the patient's lived experience was a key driver to seeking care. Barriers to care included patients not recognizing disease symptoms, reservations about seeking care, having previous negative interactions with healthcare systems, and feeling stigmatized or excluded. Situational influences, such as financial cost of accessing and receiving healthcare, also acted as barriers. Similar themes emerged for the LC versus PAD cohorts.

CONCLUSION

Multiple drivers and barriers to health-seeking behaviors exist for Black patients with LC versus PAD, including patients' perceptions of previous health system interactions. Because of known inequities experienced by Black patients, these results highlight the need for interventions that address more than just medical needs but that also encourage patients to seek care when they experience early symptoms and prioritize establishing patient-provider relationships built on trust, respect, and cultural understanding.

Intra-ethnic and geographic disparities in stage at diagnosis for non-small cell lung cancer

BACKGROUND

Despite the importance of early detection for lung cancer outcomes, staging disparities among the growing US Hispanic population remain underexplored. This population-based study aimed to identify racial/ethnic disparities among non-Hispanic White, non-Hispanic Black, and Hispanic (including specific subgroups) patients in stage at diagnosis for potentially curable non-small cell lung cancer (NSCLC).

METHODS

Incident NSCLC cases (2005-2018) were extracted from the Florida cancer registry. Stage was categorized as early (localized/regional) or advanced (distant). Multivariable logistic regression assessed the association between race/ethnicity and stage at diagnosis, adjusting for socioeconomic status, smoking, and clinical factors.

RESULTS

Among 157 034 NSCLC patients, 47.8% were diagnosed at an advanced stage. Multivariable models showed higher odds of advanced-stage diagnosis for non-Hispanic Blacks (adjusted odds ratio [ORadj] = 1.22, 95% confidence interval [CI] = 1.17 to 1.26) and Hispanics (ORadj = 1.03, 95% CI = 1.00 to 1.08) compared with non-Hispanic Whites. Regional differences were stark for Hispanics compared with non-Hispanic Whites: ORadj = 0.96 (95% CI = 0.91 to 1.01) in South Florida vs 1.12 (95% CI = 1.05 to 1.19) in the rest of Florida. In South Florida, Central Americans (ORadj = 1.49, 95% CI = 1.20 to 1.85) were the only Hispanic group showing a staging disadvantage compared with non-Hispanic Whites.

CONCLUSION

Pronounced disparities in NSCLC staging among non-Hispanic Black and Hispanic populations, with notable regional variations within Florida's Hispanic communities, indicate that targeted interventions could significantly enhance early detection. The relative advantage observed in nearly all minority groups in multicultural South Florida compared with the rest of Florida underscores the need for future research exploring how specific Hispanic and multiracial sociocultural contexts can positively influence the landscape of cancer early detection across the United States.

Cancer mortality and geographic inequalities: a detailed descriptive and spatial analysis of social determinants across US counties, 2018-2021

OBJECTIVE

In the United States, cancer mortality rates continue to decline, yet geographic and racial disparities persist and are particularly evident in the Delta region, characterized by high economic distress and disease burden. We examined cancer mortality patterns by demographic groups across geographic region (Delta vs non-Delta) and investigated the influence of macro-level social determinants of health (SDoH) in cancer death.

STUDY DESIGN AND METHODS

This observational study included cancer death records of individuals aged ≥20 years from 2018 to 2021 in the United States. County-level characteristics were ascertained through the linkage of multiple national administrative and community surveys. We estimated age-standardized mortality rates (ASR) and rate ratios. We calculated the adjusted relative risks by county-level SDoH (geographic region, rurality, household income, income inequality, health insurance, and education) and other factors using age-adjusted multivariate quasi-Poisson regression.

RESULTS

In 2018-2021, approximately 2.4 million cancer deaths occurred in the United States. We observed important declines in the Black-White disparities, from 16.6% in 2018 (ASR = 289.9 vs 248.6 per 100,000) to 12.1% in 2021 (281.1 vs 250.8) in the Delta region and from 15.9% (254.9 vs 219.9) to 10.7% (240.6 vs 217.3) in the non-Delta region, though Black men in the Delta region remained the highest rate (ASR2021 = 346.9 per 100,000). County-level analyses provided strong evidence of geographic inequality and the role of SDoH, particularly education and income inequality.

CONCLUSIONS

Unfavorable SDoH are associated with increased cancer death risk. Region-specific health policies and interventions in the Delta region are essential to advance cancer health equity.

Disruptions in Lung Cancer Detection During COVID-19

Objective: To quantify the extent of the disruption and recovery of lung cancer detection during the first two years of the pandemic, focusing on disparities across demographic and community factors. Methods: This retrospective cohort study used the SEER database to identify lung cancer cases from 2001 to 2021. Expected incidence rates for 2020 and 2021 were projected based on pre-pandemic trends (2000-2019) using the NCI's Joinpoint Regression program. Percent differences between expected and observed incidence rates were calculated. Multivariate and propensity score analyses were conducted to quantify changes in the odds of being diagnosed with metastatic disease during the pandemic. Results: Lung cancer incidence fell by 10% in 2020 compared to pre-pandemic projections, with rural populations, non-Hispanic Black and Asian patients, and females disproportionately affected. By 2021, detection rates partially recovered but remained 5% below expected levels. Localized disease detection improved significantly, while rural communities not adjacent to metropolitan areas faced further declines. Adjusted analyses showed that patients with small-cell lung cancer (SCLC) and those of Hispanic, non-Hispanic Black, or Asian/Pacific Islander ethnicity had persistently higher rates of distant disease presentation through 2021. Conclusions: The pandemic significantly reduced lung cancer detection, with only partial recovery by 2021. Persistent gaps, particularly in rural and minoritized populations, highlight the need for targeted interventions to reengage these communities.

Exploring PDK3 inhibition in lung cancer through drug repurposing for potential therapeutic interventions

The pyruvate dehydrogenase kinase-3 (PDK3) plays an important role in the regulation of a variety of cancers, including lung, by inhibiting the pyruvate dehydrogenase complex (PDC), shifting energy production towards glycolysis necessary for cancer metabolism. In this study, we aimed to identify potential PDK3 inhibitors using a computer-based drug design approach. Virtual screening of the FDA-approved library of 3839 compounds was carried out, from which Bagrosin and Dehydrocholic acid appeared best due to their strong binding affinity, specific interactions, and potential biological characteristics, and thus were selected for further investigations. Both compounds show strong interactions with functionally important residues of the PDK3 with a binding affinity of - 10.6 and - 10.5 kcal/mol for Bagrosin and Dehydrocholic acid, respectively. MD simulation studies for 100 ns suggest the formation of stable complexes, which is evident from RMSD, RMSF, Rg, and SASA parameters. The PCA and FEL analysis suggested admirable global energy minima for the bagrosin-PDK3 and dehydrocholic acid-PDK3 complexes. Finally, we identified FDA-approved drugs, Bagrosin and Dehydrocholic acid, that offer valuable resources and potential therapeutic molecules for targeting lung cancer. Further clinical investigations are required to validate the clinical utility of selected molecules.

Lung Cancer Screening in the US, 2022

Importance

The US Preventive Services Task Force (USPSTF) recommends annual lung cancer screening (LCS) with low-dose computed tomography in high-risk individuals (age 50-80 years, ≥20 pack-years currently smoking or formerly smoked, and quit <15 years ago) for early detection of LC. However, representative state-level LCS data are unavailable nationwide.

Objective

To estimate the contemporary prevalence of up-to-date (UTD) LCS in the US nationwide and across the 50 states and the District of Columbia.

Design, Setting, and Participants

This cross-sectional study used data from the 2022 Behavioral Risk Factor Surveillance System (BRFSS) population-based, nationwide, state-representative survey for respondents aged 50 to 79 years who were eligible for LCS according to the 2021 USPSTF eligibility criteria. Data analysis was performed from October 1, 2023, to March 20, 2024.

Main Outcomes and Measures

The main outcome was self-reported UTD-LCS (defined as past-year) prevalence according to the 2021 USPSTF eligibility criteria in respondents aged 50 to 79 years. Adjusted prevalence ratios (APRs) and 95% CIs compared differences.

Results

Among 25 958 sample respondents eligible for LCS (median [IQR] age, 62 [11] years), 61.5% reported currently smoking, 54.4% were male, 64.4% were aged 60 years or older, and 53.0% had a high school education or less. The UTD-LCS prevalence was 18.1% overall, but varied across states (range, 9.7%-31.0%), with relatively lower levels in southern states characterized by high LC mortality burden. The UTD-LCS prevalence increased with age (50-54 years: 6.7%; 70-79 years: 27.1%) and number of comorbidities (≥3: 24.6%; none: 8.7%). A total of 3.7% of those without insurance and 5.1% of those without a usual source of care were UTD with LCS, but state-level Medicaid expansions (APR, 2.68; 95% CI, 1.30-5.53) and higher screening capacity levels (high vs low: APR, 1.93; 95% CI, 1.36-2.75) were associated with higher UTD-LCS prevalence.

Conclusions and Relevance

This study of data from the 2022 BRFSS found that the overall prevalence of UTD-LCS was low. Disparities were largest according to health care access and geographically across US states, with low prevalence in southern states with high LC burden. The findings suggest that state-based initiatives to expand access to health care and screening facilities may be associated with improved LCS rates and reduced disparities.

Three-year follow-up study reveals improved survival rate in NSCLC patients underwent guideline-concordant diagnosis and treatment

Background

No studies in China have assessed the guideline-concordance level of the first-course of non-small cell lung cancer (NSCLC) diagnosis and treatment and its relationship with survival. This study comprehensively assesses the current status of guideline-concordant diagnosis (GCD) and guideline-concordant treatment (GCT) of NSCLC in China and explores its impact on survival.

Methods

First course diagnosis and treatment data for NSCLC patients in Liaoning, China in 2017 and 2018 (n=1828) were used and classified by whether they underwent GCD and GCT according to Chinese Society of Clinical Oncology (CSCO) guidelines. Pearson's chi-squared test was used to determine unadjusted associations between categorical variables of interest. Logistic models were constructed to identify variables associated with GCD and GCT. Kaplan-Meier analysis and log-rank tests were used to estimate and compare 3-year survival rates. Multivariate Cox proportional risk models were constructed to assess the risk of cancer mortality associated with guideline-concordant diagnosis and treatment.

Results

Of the 1828 patients we studied, 48.1% underwent GCD, and 70.1% underwent GCT. The proportions of patients who underwent both GCD and GCT, GCD alone, GCT alone and neither GCD nor GCT were 36.7%, 11.4%, 33.5% and 18.4%, respectively. Patients in advanced stage and non-oncology hospitals were significantly less likely to undergo GCD and GCT. Compared with those who underwent neither GCD nor GCT, patients who underwent both GCD and GCT, GCD alone and GCT alone had 35.2%, 26.7% and 35.7% higher 3-year survival rates; the adjusted lung cancer mortality risk significantly decreased by 29% (adjusted hazard ratio[aHR], 0.71; 95% CI, 0.53-0.95), 29% (aHR, 0.71; 95% CI, 0.50-1.00) and 32% (aHR, 0.68; 95% CI, 0.51-0.90).

Conclusion

The 3-year risk of death is expected to be reduced by 29% if patients with NSCLC undergo both GCD and GCT. There is a need to establish an oncology diagnosis and treatment data management platform in China to monitor, evaluate, and promote the use of clinical practice guidelines in healthcare settings.

Multilevel approaches to address disparities in lung cancer screening: a study protocol

BACKGROUND

Low-dose computed tomography (lung cancer screening) can reduce lung cancer-specific mortality by 20-24%. Based on this evidence, the United States Preventive Services Task Force recommends annual lung cancer screening for asymptomatic high-risk individuals. Despite this recommendation, utilization is low (3-20%). Lung cancer screening may be particularly beneficial for African American patients because they are more likely to have advanced disease, lower survival, and lower screening rates compared to White individuals. Evidence points to multilevel approaches that simultaneously address multiple determinants to increase screening rates and decrease lung cancer burden in minoritized populations. This study will test the effects of provider- and patient-level strategies for promoting equitable lung cancer screening utilization.

METHODS

Guided by the Health Disparities Research Framework and the Practical, Robust Implementation and Sustainability Model, we will conduct a quasi-experimental study with four primary care clinics within a large health system (MedStar Health). Individuals eligible for lung cancer screening, defined as 50-80 years old, ≥ 20 pack-years, currently smoking, or quit < 15 years, no history of lung cancer, who have an appointment scheduled with their provider, and who are non-adherent to screening will be identified via the EHR, contacted, and enrolled (N = 184 for implementation clinics, N = 184 for comparison clinics; total N = 368). Provider participants will include those practicing at the partner clinics (N = 26). To increase provider-prompted discussions about lung screening, an electronic health record (EHR) clinician reminder will be sent to providers prior to scheduled visits with the screening-eligible participants. To increase patient-level knowledge and patient activation about screening, an inreach specialist will conduct a pre-visit phone-based educational session with participants. Patient participants will be assessed at baseline and 1-week post-visit to measure provider-patient discussion, screening intentions, and knowledge. Screening referrals and screening completion rates will be assessed via the EHR at 6 months. We will use mixed methods and multilevel assessments of patients and providers to evaluate the implementation outcomes (adoption, feasibility, acceptability, and fidelity).

DISCUSSION

The study will inform future work designed to measure the independent and overlapping contributions of the multilevel implementation strategies to advance equity in lung screening rates.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT04675476. Registered December 19, 2020.

Structural Racism as a Contributor to Lung Cancer Incidence and Mortality Rates Among Black Populations in the United States

BACKGROUND

Although racial disparities in lung cancer incidence and mortality have diminished in recent years, lung cancer remains the second most diagnosed cancer among US Black populations. Many factors contributing to disparities in lung cancer are rooted in structural racism. To quantify this relationship, we examined associations between a multidimensional measure of county-level structural racism and county lung cancer incidence and mortality rates among Black populations, while accounting for county levels of environmental quality.

METHODS

We merged 2016-2020 data from the United States Cancer Statistics Data Visualization Tool, a pre-existing county-level structural racism index, the Environmental Protection Agency's 2006-2010 Environmental Quality Index (EQI), 2023 County Health Rankings, and the 2021 United States Census American Community Survey. We conducted multivariable linear regressions to examine associations between county-level structural racism and county-level lung cancer incidence and mortality rates.

RESULTS

Among Black males and females, each standard deviation increase in county-level structural racism score was associated with an increase in county-level lung cancer incidence of 6.4 (95% CI: 4.4, 8.5) cases per 100,000 and an increase of 3.3 (95% CI: 2.0, 4.6) lung cancer deaths per 100,000. When examining these associations stratified by sex, larger associations between structural racism and lung cancer rates were observed among Black male populations than among Black females.

CONCLUSION

Structural racism contributes to both the number of new lung cancer cases and the number of deaths caused by lung cancer among Black populations. Those aiming to reduce lung cancer cases and deaths should consider addressing racism as a root-cause.