Investigating Incidental Pulmonary Nodules in Underserved Communities

Montefiore Medicine Academic Health System

Status: Withdrawn

Conditions: Lung; Node

Study type: Observational

Funder types: Other

Identifiers

NCT05738031

2021-13475

Details and patient eligibility

About

The study team hypothesizes that incidentally discovered pulmonary nodules are often under captured and/or not surveilled in accordance with published guidelines in the Montefiore Health System, which cares for a large proportion of Black and Hispanic patients. Incidental Pulmonary Nodules (IPNs) require a pragmatic approach to follow-up and management, especially in racially disparate populations who have greater potential for lung cancer morbidity and mortality.

Full description

An estimated 235,760 people will be newly diagnosed with lung cancer in 2021 and over 130,000 people will die as a result. Computed tomography (CT) screening has demonstrated reduced lung cancer mortality, and is now recommended for adults aged 50 to 80 years with a 20-pack-year smoking history who are current smoker or have quit within 15 years. Along with the early identification of at-risk pulmonary nodules, screening programs also have the ability to direct patients into pathways for protocolized follow-up. Despite defined eligibility requirements, lung cancer screening (LCS) is underutilized, with less than 10% of eligible patients undergoing screening. Screening is particularly challenging in minority and socioeconomically disadvantaged populations, who face an increased risk of death due to lung cancer. Of over 50,000 patients enrolled in the National Lung Screening Trial, for example, only 4.5% of participants identified as Black and 1.7% identified as Hispanic.

Outside of screening programs, lung nodules are commonly detected incidentally on imaging done for other reasons. Each year, more than 1.5 million patients are diagnosed with an incidental pulmonary nodule. Of these, 5-9% are estimated to represent cancers, higher than the malignancy rate noted in lung cancer screening programs. Guidelines exist for the follow-up of IPNs, however compliance is often poor. IPNs may be overlooked in the context of the other illnesses for which imaging is obtained. Tests may also be ordered by providers without continuity of care, as occurs in the emergency department (ED). Patients may thus be unaware of incidental findings or receive inadequate direction for follow-up when there is no clear chain of responsibility. Racially disparate populations are specifically at risk and often face barriers to accessing primary care providers (PCPs), leading to increased use of the ED. In one study, a higher proportion of Black and Hispanic patients (38.3% and 28.1%, respectively) had initial imaging identifying an IPN performed in the ED compared to White patients (10.7%), who were more likely to have outpatient scans.

Previous studies indicate that only 38% of patients receive guideline concordant care once diagnosed with an IPN. Lung cancer was diagnosed in 8% of these patients undergoing such care compared to only 1% in those who received less intensive evaluation. Similarly, the median time to diagnosis of a lung cancer was 1.3 months in the guideline concordant care group versus 12 months in the less intensive evaluation patients, again underscoring the importance of appropriate mechanisms of follow-up. The consequences of a missed nodule are clear.

Data suggest that racial and ethnic disparities exist in the follow-up of IPNs. In a study of 1,562 patients with an IPN requiring follow-up at a tertiary care center, only 49.1% of Hispanic patients and 55.1% of Black patients were notified of IPNs compared to 79.5% of White patients. Similarly, non-White patients had significantly lower rates of ordering and adherence to follow-up imaging and had an increased odds of delaying follow-up. While this discrepancy in care has been identified, few solutions exist to bridge the gap and underrepresented patients remain at higher risk of delayed diagnosis until advanced stages of disease.

Further compounding the difficulty in managing underserved patients with IPNs is the lack of programs for formalized follow-up, specifically in urban areas. In an advisory board meeting of major medical centers within New York City, only one formal nodule evaluation program associated with a center's ED was identified. New York Presbyterian Hospital Weill Cornell identified 539 patients with IPNs over a two-year period. After radiologic review, chest radiologists referred 289 patients for further consultation and of these 142 (26.3% of original population) were referred for evaluation by a pulmonologist or oncologist. While the results of this investigation and rates of cancer diagnoses are currently being tabulated, the large proportion of patients referred for concerning findings is quite notable.

Within New York City, the Montefiore Health System is uniquely positioned to conduct clinical research and bridge health care disparities by engaging underserved and underrepresented communities. The health system is comprised of 11 hospitals in the Bronx, Westchester, and the Hudson Valley in New York. The main campuses include two high volume EDs, including one of the five busiest in the country, and serve a diverse population of nearly 1.5 million residents in the Bronx. Previous data from Montefiore have demonstrated that of 855 primary lung cancers diagnosed between 2013 and 2016, only 417 (55%) were found in patients with an in-network PCP, illustrating the need for a better support system and for a systematic approach to identify and guide these patients. Furthermore, of the 175 of these patients who were eligible for LCS, only 33 had completed screening. Among screened patients, 64% were diagnosed with stage I/II non-small cell lung cancer, compared to only 29% of the lung cancers found outside of screening. In this latter group, 46% were diagnosed with metastatic disease. This demonstrates not only the value of screening in this at-risk population, but also emphasizes the need for prompt follow-up of incidentally detected lung abnormalities. A large proportion of this population in whom lung cancers were identified outside of screening was comprised of Black (46%) and Hispanic (34%) patients with a median per capita income of only approximately $20,000.12. The evaluation and implementation of a nodule detection program may thus extend care and improve the potential for survival in patients with reduced access to health care, where the ED may function as the primary care hub.

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

Part A of the study seeks to capture IPNs and standardize mechanism for management will employ the following inclusion criteria:

• Above the age of 18
• Underwent an imaging study for any purpose other than lung cancer screening at a Montefiore Medical Center facility that identifies a pulmonary nodule
• Imaging results captured into the Epic electronic medical record
• Referred to the lung nodule clinic
• Underwent an in-person or telephone/telehealth encounter at a Montefiore Medical Center facility during the study period
• Lung cancers diagnosed at a Montefiore Medical Center facility between April 1, 2021 and March 31 2022 will also be included for retrospective review and comparison to the cohort of lung cancers diagnosed following the instillation of the IPN program herein described

Part B of the study will seek to develop an algorithm for incidental lung nodule risk stratification utilizing data from the electronic medical record. Inclusion criteria are:

• Above the age of 18
• Underwent an imaging study for any purpose other than lung cancer screening at a Montefiore Medical Center facility that identifies a pulmonary nodule
• Imaging results captured into the Epic electronic medical record
• Referred to the lung nodule clinic
• Underwent an in-person or telephone/telehealth encounter at a Montefiore Medical Center facility during the study period

The third part (Part C) of the study is a substudy examining plasma-based expression of markers associated with lung cancer in patients with lung nodules, and will include the following patients:

• Above the age of 18
• Underwent an imaging study, including imaging performed for lung cancer screening, at a Montefiore Medical Center facility that identifies a pulmonary nodule
• Imaging results captured into the Epic electronic medical record
• Referred to the lung nodule clinic
• Underwent an in-person or telephone/telehealth encounter at a Montefiore Medical Center facility during the study period
• Recommended follow-up by means of either continued imaging surveillance of nodule or procedure for diagnosis, including percutaneous transthoracic needle biopsy, endobronchial transbronchial needle aspiration, surgical biopsy or resection of pulmonary nodule

Exclusion criteria

• <18 years of age
• No evidence of a lung nodule through imaging

Trial design

0 participants in 3 patient groups

Group A

Description:

Aim 1 (Part A). To utilize natural language processing (NLP) to identify all ED patients with incidentally detected lung nodules found on chest radiographs or chest or abdominal CT scans, and to develop a standardized referral and notification process Aim 1.1) Utilization of NLP to screen radiologic reports and identify nodules meeting criteria for follow-up per Fleischner Society Guidelines Aim 1.2) Creation of an electronic medical record-based notification system to alert patients and providers of the identification of an IPN that requires follow-up, tracked by a dedicated patient navigator Aim 1.3) Establishment of a multidisciplinary lung nodule management team, hereafter referred to as the lung nodule clinic, to ensure guideline-directed management of nodules with emphasis on high risk nodules as identified in subsequent aims

Group B

Description:

Aim 2 (Part B). To clinically risk stratify patients with IPNs utilizing artificial intelligence (AI) processing of known clinical risks factors for pulmonary malignancy, such as age, smoking history, and history of malignancy, along with radiographic risk classifiers including nodule location, size, and imaging features. Aim 2.1) Development of an integrated classifier based on automated scanning and data retrieval from the electronic medical record (EMR) to stratify patients with IPNs as low, intermediate or high risk for malignancy, with factor analysis to assess contributions of individual factors to the model Aim 2.2) Prospective evaluation of the integrated classifier and comparison of automated integrated classifier to established manual risk calculators

Group C

Description:

Aim 3 (Part C). To investigate biologic risk classifiers that may aid in the risk stratification of pulmonary nodules Aim 3.1) Evaluation of a blood-based gene expression assay for risk stratification of pulmonary nodules using biobanked specimens Aim 3.2) Prospective collection of plasma from patients enrolled in lung nodule clinic and evaluation of gene expression to assess malignancy risk