Raymond U. Osarogiagbon, MD, discusses existing disparities in lung cancer screening, and the active interventions that are needed to close the gap in access to preventive care.
A great advancement made in the realm of lung cancer has been the evidence underscoring that screening has the power to save lives. As such, it is important to be aware of, and understand, existing geographic, socioeconomic, and racial disparities that hinder access to potentially lifesaving preventive care, according to Raymond U. Osarogiagbon, MD.
“In terms of correcting the problem, we, as a society, have to say that this is a high priority for us,” Osarogiagbon said. “[We need] to bring our advocates, our politicians, our executives, at the national, state, and county levels, [together] to figure out the policies that we need to have to promote a healthier population, [and this includes] providing access to preventive care and early diagnosis. Then, we must look at our health care systems. A patient cannot walk off the street and get a low-dose CT [LDCT]screening test if the infrastructure does not exist.”
In an interview with OncLive®, Osarogiagbon, chief scientist at Baptist Memorial Health Care, director of the Multidisciplinary Thoracic Oncology Program and the Thoracic Oncology Research Group at Baptist Cancer Center, discussed existing disparities in lung cancer screening, and the active interventions that are needed to close the gap in access to preventive care.
Osarogiagbon: One of the great [areas of] excitement in lung cancer has been the evidence that screening saves lives. Data from the [phase 3] National Lung Screening Trial [NLST; NCT00047385] that was done in the United States, the Dutch-Belgian trial [NELSON; NTR636], and then the late analysis of the MILD trial [NCT02837809] done out of Italy, consistently showed [a benefit with screening]. At least a 20% reduction in lung cancer–specific mortality [was observed in] the NLST and [this percentage was] even greater in the 2 European trials.
Lung cancer screening saves lives. The challenge has then become: How does one make this happen in real-world populations, outside of tightly controlled clinical trials? That is where we have struggled.
The most recent statistics from as recently as 2018 suggest that only about 5% of eligible patients in the United States have been able to get a LDCT screening test. It is important to put that in perspective because currently, the United States is the only country on Earth where screening is a covered health care benefit. Across the rest of the world, screening is not available as a covered benefit. However, here, in the United States, there is a clear implementation challenge of making LDCT screening for lung cancer available to everyone.
[The pandemic] created the need for people to stay apart, including to stay away from their health care providers and systems, as we social distanced and quarantined ourselves. On the health care organizational side, there was the inundation with patients who were sick and dying from COVID-19, which forced us to reprioritize what we were doing. For example, in my health care system, we suspended screening. We stopped screening patients from late March 2020 to the summer of 2020.
We saw that many patients who would have been candidates for screening, and even now, a [couple of] years down the line, are coming into [the clinic] with advanced lung cancer. Some of these patients already had [received] a LDCT screening test and were found to be low risk, with a Lung–[Reporting and Data System (RADS) score] 1 or 2, and had been recommended for a follow-up interval radiologic study later on, and never showed up. A couple of years later, we are seeing that some of these patients have advanced cancer.
There is also the aspect of it that COVID-19 has been a lung-centric pandemic in many ways. The symptoms are respiratory, leading people who get sick to go to health care systems where radiologic studies get done. Amid all the noise of COVID-19, there is a reality that…cancer has had to take a backseat to the reality of rescuing someone from a lethal, viral infection [to some regard]. As a nation, [we know] that our elective procedures, especially screening preventive care, have had to take the backseat.
The huge frustration is, even as we have learned with the cycles of the number of [COVID-19 cases] going up, going down, and then coming back up, that cyclical nature of the pandemic has made it hard for us to apply the lessons we think we have learned. For example, the Omicron variant is so highly contagious, that even when we have said that some of these [efforts] are not really that elective, that we have to keep the doors open for patients to gain access to lifesaving preventive care, the sheer numbers of patients who come in with this variant overwhelm our care-delivery infrastructure. Then, we are almost back to square one, even though we remain pushy about not dropping everything, like we did in early 2020.
Lung cancer is the public health challenge of our age, here, in the United States, and essentially worldwide. However, the patterns of incidence and mortality are not uniform. Across the globe, there are countries with much higher incidence and mortality rates. Even within nations, there are geographic differences. For example, in the United States, if you look at the state-stratified incidence and mortality from lung cancer, there is a leaderboard. Kentucky is number 1, Mississippi is number 2, Arkansas is number 3, West Virginia is number 4, and Tennessee is number 5.
We know that the lung cancer incidence and mortality map is always a carbon copy of the tobacco use map. We know smoking is, by far, still the most powerful risk factor for lung cancer. As such, part of what we are speaking about is disparities in the use and cultural acceptance of tobacco. The tobacco belt of the United States is also the cancer mortality belt of the country—especially [with regard to] lung cancer. As such, that is 1 kind of disparity: geographic.
We also know that, at least in the United States, tobacco use, and therefore, [resultant] lung cancer, is a phenomenon of socioeconomic deprivation. People with a college degree are much less likely to be active cigarette or tobacco users than those with a high school diploma or less. People who live in wealthier parts of a city are much less likely to have lung cancer than those who live in less wealthy, maybe more industrial or red-lined areas [within] cities.
There is also a very clear racial disparity. The people with the highest incidence and mortality from lung cancer are Black men and Native American men. On the female side, it is a little bit different: Native American women and White women have a higher risk than Black women. Smoking is a primary risk factor, but smoking as a risk factor for lung cancer also has a racial distribution. The smoking intensity that leads to lung cancer in African Americans and Native Americans is much lower than that in those who are White.
Therefore, when we calibrate our risk on the basis of those who typically have access [to screening], which is always in the United States, White people, we miscalibrate risk for everyone else—including Black people, Native Americans and Asians, for whom, for reasons that we still do not fully understand, the risk of lung cancer is high, and much less dependent on tobacco exposure.
These are just some examples of the disparate nature of lung cancer as a public health burden: geographic, socioeconomic, and racial.
This is a very interesting phenomenon. When we are all in the dark, when there is no knowledge, there oftentimes is hardly any difference between person and person. Where there are differences, they tend to be relatively narrow differences. From the point of discovery, when something is known to be of benefit, and therefore, creates the opportunity to change the natural history of health care or any other challenge we are dealing with, we quickly begin to see disparities emerge. The people who have ready access turn out to be essentially the same people. They belly up to the bar first and get the full benefit of that discovery. The usual people then get left behind, almost without exception.
For example, if we talk about screening, the moment the NLST clearly demonstrated that a LDCT scan once a year for people above a certain age with a certain level of cigarette/tobacco exposure can identify enough lung cancer in enough people to provide access to early diagnosis and curative treatment, and therefore, lead to lower mortality, we went from baseline in which there was a slight plurality of those kinds of tests being done in African Americans, to where there is already a clearly emergent racial disparity. A significantly higher proportion of people who have had a screening test are White people—especially White men. Even White women get a raw deal. Of course, everyone else, the other races, have lagged [with regard to screening].
These things are predictable. They are sociologic constructs that emerge out of our complex history of socioeconomic and political decisions that were oftentimes, made centuries ago. We must not act as if they come as a surprise. They are predictable, and therefore, as we work to discover and implement, we must bear that reality in mind. In bearing that in mind, we must then act on that predictable knowledge that these disparities will emerge, and to the degree that we, as a society, believe that disparities are undesirable, we must invest resources in preventing or minimizing them.
The NLST had over 60,000 people enrolled; 4.5% of its population were African American, and this was after some effort to shore up those numbers. Naturally comes the suspicion of, ‘Black people do not like to participate in clinical trials, and that is why the number [is so low].’ No, the reality is that the places where the trial was conducted were not the places that those people went to. As such, one of the lessons we must learn is that when we have these important projects, we need to make them available in the places where the full spectrum of the population chooses to seek care. That is oftentimes not our high temples of clinical science; it is not the wonderful network of National Cancer Institute [NCI]–designated comprehensive cancer centers. It is oftentimes the community health care systems who carry the bulk of the burden of clinical care responsibility and who, unfortunately, lack clinical trials infrastructure.
The initial response might be that the [facilities] do not have the ability to do clinical trials; that is why we cannot do clinical trials there. However, my response to that is if this is important to us, as a society, we must invest our resources to build that infrastructure. This is also why one can be heartened by the [effort being made by the] NCI’s Community Oncology Research Program to create minority, underserved, and core programs that are designed to do that, to set their stall in the places where such patients routinely gain access [to care].
The more you need, the less you get, unfortunately. The rich get richer, and the poor get poorer. Unfortunately, that is the way we are set up. Those who have the greatest burden of pathology also, paradoxically, have least well-put-together care delivery infrastructure.
If we said LDCT scans save lives, and the per capita incidence of lung cancer is heavily clustered at the high end in these places, the natural question is, if we as a society decide, yes, we have a solution, and we want to apply it, where are we going to create that infrastructure? You would think you would see a high density of LDCT facilities in the heart of the lung cancer mortality battlefield. What you see is almost a [reverse] image. The predominance of screening facilities have been set up in the Northeast and the West Coast, where there are large populations of lung cancer victims. There are large populations of people who stand to benefit from screening. I do not begrudge them one bit.
However, the per capita mortality in those places is significantly lower than the southern parts of the United States, parts of Appalachia, and the Upper Midwest, where the highest density of lung cancer exists. They also have, with the exception of the state of Kentucky, which has figured out some way to create that infrastructure, what you see is a [reverse] image of geographic dispersal of screening facilities vs the lung cancer mortality map.
It is a complex set of etiologic factors that intertwine into a tightly wound burden. To untangle it, one does have to come up with a strategic plan. A way to look at it is, where do we have our biggest potential influence? With this, comes different levels of intervention. At the base, there is person level, and then, the provider level, the people who serve as gatekeepers between service and patient. Then, there are the health care systems, the organizations within which those providers work, and then the social policies and laws that guide the institutional functions.
The way these things are layered, there are more patients who need a service than there are providers, and there are more providers than the number of organizations and healthcare systems within which they work, and there are more healthcare systems and organizations than the social policies and laws within which we function. In terms of the efficiency of intervention, if you then see the reality that your leverage is always going to be greater at the social policy level, and then subsequently, at the organizational level, and then, at the provider level, then the least efficient is to go directly [to the people], and expecting them to know what they are supposed to do and get it done.
The key questions become: Why have the organizations, the health care systems, not created the infrastructure, and why have the providers within those organizations not created awareness among their populations to provide access to [this kind of care]? You must flow from there: the social policy that makes it conducive for organizations to provide the services, and then task their clinicians to provide access to their patients. In terms of getting the right ‘bang for our buck,’ this is the way we must do it. It is not nagging all those patients out there who are at risk and telling them that it is their problem that they have not taken advantage of the service.
LDCT screening saves lives, but we cannot seem to get enough people through it fast enough. We are discovering that people do get radiologic studies done for many different reasons. If [a person goes] into an emergency room from a car accident, chances are [that they will] get an X-ray or a CT scan. Not infrequently, these images reveal the presence of a lesion, and a small but significant minority of such lesions turn out, over time, to become lung cancer. You have this experience where you see sequential images taken of someone with a lesion that keeps growing. However, because we are all distracted by several other [medical issues, signs of cancer are not always fully acknowledged].
This idea of incidentally detected lung nodules has turned out to be a potential avenue to create access to early detection in those who otherwise would not have gained access to a LDCT scan. The advantage of that is that we are not talking about encouraging someone to go get an X-ray with no understanding of any real personal risk, other than the theoretical risk from age and past smoking history. Rather, someone has already had a test, and it already showed an abnormality, and guidelines already exist for clinicians to follow, to know how to safely identify those who may turn out to have lung cancer and provide care for them.
The challenge here is, those guidelines are often not followed. These incidental lung nodule programs, where they bring, for example, the Fleischner Society guidelines, or the American College of Chest Physicians guidelines, they seem to be able to find many people who have early lung cancer, and the demographics are a little bit different. Suddenly, you find higher proportions of the kinds of people who stumble into emergency rooms for other reasons, who otherwise would never have had a preventive test.