Lung Cancer Podcast

What Are the Risks and Benefits of Lung Cancer Screening?

Jacob Sands, MD; Shawn Regis, PhD


August 11, 2022

This transcript has been edited for clarity.

Jacob Sands, MD: Hello. I'm Dr Jacob Sands, a thoracic medical oncologist at the Dana-Farber Cancer Institute and co-lead of a committee that helps to run the lung screening program at the Brigham and Women's Hospital. Welcome to Medscape's InDiscussion series on lung cancer. Today we'll be discussing lung cancer screening. First, let me introduce my guest. Dr Shawn Regis is a PhD and the director of analytics and clinical quality assurance for Think Health and the secretary of the Rescue Lung Society. Prior to his current role, Shawn spent 8 years as an associate research scientist and patient navigator for the CT Lung Screening Program at Lahey Hospital and Medical Center in Burlington, Massachusetts, where he managed a program that performed more than 24,000 scans on more than 7200 individuals. He is a co-author of the ALA/ATS Lung Screening Implementation Guide and the Massachusetts Statewide 2017 to 2021 Cancer Plan; he is a panelist on a thoracic oncology perspective series providing practical guidance about lung cancer screening; and he was an advisory committee member for an Association of Community Cancer Centers initiative to develop an optimal care coordination model for lung cancer patients with Medicaid. Shawn is a co-author on more than 20 manuscripts pertaining to CT lung screening and has presented across the country and internationally on various CT lung screening topics. He is certainly an expert in the field. Welcome to InDiscussion.

Shawn Regis, PhD: Thanks so much. Thanks for having me.

Sands: Before we kick off on lung screening, I just want to first ask, what is it that drew you into the field? How has your journey been to this space?

Regis: I was lucky to get into the lung screening space. We had a family friend who was a radiologist at Lahey at the time that they were starting their program there and worked with Brady McKee on developing Lung-RADS. And he notified me about an opening for coordinating that program. That was the first I had ever heard about lung screening. And once you hear about it, I don't know how you can't be interested in it. So I dove right in and saw a whole bunch of opportunities for research and improvement and getting it out there to as many people as possible. I really lucked out, but I'm certainly glad that I did.

Sands: When was that? When was your first jump into the space?

Regis: I got into that program in May of 2014.

Sands: To kick us off, there have been a lot of studies in lung screening, some prior to that, but mostly after the time that you jumped into the space. Let's start off with just a review of the data and what we have seen from lung screening from the data out there so far.

Regis: The big one was the National Lung Screening Trial [NLST]. The results for that came out in 2011, and it was really the driving force behind getting clinical lung screening started in the United States. The first USPSTF recommendations, for CMS NCD, were all really just based on those results. So, there's no understating its importance but at the same time, it's not really what's done in clinical practice. And a lot has been done since then to try to expand that coverage and make it more closely resemble what happens in clinical practice as opposed to what was done in that research trial. Since then, we've had the Nederlands–Leuvens Longkanker Screenings Onderzoek [NELSON] study and the MILD trial, which have both continued to show a mortality benefit for lung screening. They actually show a greater mortality benefit for a few reasons, one of which being they compare CT lung screening to nothing, which is really what the standard of care was for lung screening, as opposed to the NLST, which compared it to a chest x-ray and still saw a mortality benefit. And the other is they had the age a little bit lower. The pack-years were also a little bit lower in those trials. And they screened for longer, and obviously you're going to see a greater benefit as you screen for a longer period of time. So that increased benefit led to the new USPSTF recommendation and CMS decision. And now we've expanded that population and hopefully we can continue to get people in and get them screened.

Sands: Among those, there's the NLST and the NELSON trials which, of course, are the two biggest studies, the NLST being more than 50,000 patients. And they both have a really similar table, which I always show the residents and fellows around lung screening. So NLST showed a 20% lung cancer–specific mortality benefit and that led to early stopping of the study. Of course, it also showed an all-cause mortality benefit. But Table 5 in the NLST, and there's a similar table in NELSON, essentially showed that a screen-detected lung cancer ends up being around 70% early stage vs the group that got randomized to the CT screening group but ends up getting a lung cancer diagnosis in the follow-up years within the NLST or between years even within NELSON. So, randomized to get a CAT scan and got a lung cancer but it wasn't on a screening year, and that ends up being 70% late stage. So even within these trials, we see a really significant shift in cancer diagnoses, which I think speaks to the importance of yearly scans. But you've had experience managing a huge program. Can you talk a bit about the difference between that baseline year, the first year of getting the scan, vs the later years and what you see as far as stage?

Regis: That's another reason that the NELSON study, and the MILD [study] as well, showed such an increased benefit over the NLST — they screened for longer. It's the annual scans, like you mentioned, every single year, where you really see the most benefit from lung screening. At baseline, like you mentioned, we at Lahey saw about 70%-80% that are early stage. But even in those baseline patients, you have some that are really close to presenting with symptoms. You're still finding some bulky disease in there, whereas if they're in annual screening, they've already had that baseline scan. So that bulkier disease has been moved out mostly. And so, in the annual rounds of screening, we see closer to 80%-90% that are early stage. Getting patients in is a big deal but keeping them in is also a big deal because that's really where the benefit comes in — those annual rounds of screening.

Sands: And in those follow-up years, what is the stage breakdown? What's your experience been in terms of staging in those following years?

Regis: We're talking between 70% and 80% are stage I, and then around 10% or so are stage II cases, and then later stage is another 10 or maybe 15%. But for the most part, we're talking early stage all the time in those annual screenings. So it's big.

Sands: And that's such an important point. Basically the point that makes is that the screening data we have that led to lung screening implementation … that the real world of being in yearly scanning is actually even better than the data that we're seeing. Can you speak to that? To what degree do you think that's true? And what do you think the real-world data look like compared to what we've described in NLST and NELSON?

Regis: It's almost hard to put it into words if you don't have a clinical program or at least have an idea of what the data look like clinically. Because even when people are really trying to promote lung screening, we hear a 20% mortality benefit from the NLST, which is big. That's a big deal. But it's also kind of hard to picture exactly what that means. And it's an undershot because we're doing things differently now than we did in the NLST. So, as you mentioned, even in between rounds, but really in the absence of screening, we're talking 70% late-stage disease, which is why it's such a problem. The NLST completely flipped that to 70% early stage, so already that's a massive difference. And then we've improved on that by doing annual screening with a better positive predictive value. And the benefits are unreal. If you look at some of the numbers that the ALA [American Lung Association] has given, they estimate that about 48,000 lives could be saved every year if all the eligible people for lung screening got screened. It saves more lives than curing breast cancer.

Sands: I agree. I think the true benefit of lung screening is actually understated from these studies. So, we're saying that the benefit is really substantial. What are the risks of lung screening?

Regis: Well, if the benefits are understated and the risks are overstated a bit, the one that always kind of jumps out at me because it's repeated over and over, is that it has a high false-positive rate. And there are a couple issues with that. One is if you're hearing that it has a 96% false-positive rate, that's not the false-positive rate. That's the number of positive scans that end up not being cancer, which is different. In the NLST, the false-positive rate was really closer to 23%-25%, somewhere in that range, which is elevated and not terrific. But again, that's not what we do clinically. The NLST had a lower threshold for what they were calling positive. In clinical practice, with a 6-mm threshold on nodule size, the false-positive rate is in the 8%-12% range. And if you do things the way the NELSON trial did it, where you need a follow-up scan before you actually determine if something is positive or not, it's down to 2% for a false-positive rate. So it's very comparable to what you see in mammography. And it's not a high false-positive rate at all.

Sands: I think that's been actually one of the most confusing things, particularly for primary care, where they're seeing different headlines and such. But I want to pause for a moment and really focus on something you said there. The NELSON trial suggests that the false-positive rate is much better when using their strategy. Of course, they essentially don't call something positive but say "indeterminate." And then you need to do another scan at whatever the recommended time point is, whether that be 3 months, 6 months, essentially doing that scan. And then if the nodule is growing, labeling it as positive vs, in the ACR Lung-RADS®, calling something positive even if it's something that you're then going to do a scan in a few months to determine. It's a different format, which you outlined a bit, but you also have published data from within the Lahey program. What does that look like as far as your false-positive rates and your detection rates?

Regis: What we did at Lahey was to consider a Lung-RADS 3 or higher to be positive. Now, a Lung-RADS 3 is considered probably benign but it does require something other than the patient just coming back in a year. So that's what we call positive. And what we found is that really the cancer detection rate in that group is only between 1% and 2%, probably. So, if we waited to call those positive until after that 6-month scan, you would see our false-positive rate drop way off as well. But even including that group as positive, it's about baseline, I think it's around a 12% false-positive rate. And then annually after that, it's more like 8% because again, now you have a comparison, and you can see whether there is a growth or not and that bumps off a whole bunch of those false positives that you were seeing before.

Sands: And what about the cancer detection rate?

Regis: On baseline, it's about 2% at Lahey. And then annually, every single year thereafter, it's about 1%. So that goes to your point about doing this every year. There's been talk about should we do it biannually or, if they're a Lung-RADS 1, should they be pushed out? It's 1% every single year that we see consistently in that program. And I think it's a big deal to make sure that they're coming back every single year because that's where you find these cancers and get that benefit.

Sands: In the US, it's approved to be done yearly and that is the recommendation, that's in the USPSTF [United States Preventive Services Task Force] guidelines. In other countries around the globe, they're deciding how to run their screening programs. And there's been some debate about that. You know, one of the things that did not come out in the NELSON publication, but I do remember it being presented at World Lung, at the IASLC World Conference on Lung Cancer, was that when there was a two-and-a-half-year time frame between scans, they actually had an increased number of late-stage diagnoses. Now, I don't remember whether they actually did a statistical analysis of that, so I can't say it was significant, but it was substantial. It was certainly a numerically clear difference within those. And I do worry about the time frame being more than a year between scans based upon that and that really substantial stage shift that we discussed in the very beginning that we see in both the NLST and the NELSON study. So I agree, you know, the confusion around false positives certainly is out there. Are there any other risks that you think are important to highlight?

Regis: One of the other big ones is the radiation dose, which I think is what contributes a lot to that conversation about should we not do this annually? Because that's a lot of CT scans. And it's just important to point out that these are low-dose CTs and, again, even lower than what was done in the NLST. The amount of radiation that was being used at Lahey for these was almost the same as what you see in a mammogram. So, if there's no problem with doing that [mammography] annually, there's really no problem with doing this annually. There are concerns about additional workup for those positive findings that don't end up being lung cancer. But there are several publications out there that show that the rates of intervention on benign disease for lung screening are really low, especially if you have multidisciplinary involvement and you're discussing these cases and what the next steps should be. It's always a risk that you're going to intervene somewhere that's not cancer. But if you're not intervening enough, then you're missing some. So, you know, you've got to weigh that out and make sure that you're doing everything you can to find and take care of these cancers. And the last is overdiagnosis, where people are concerned that, you know, you're diagnosing lung cancer that isn't really going to affect anybody in their lifetime, but that hasn't really panned out either. The NLST, in a follow-up study, showed that that's about 3% for lung cancer. So again, it seems the benefit significantly outweighs the risk. And it's important to have all those things in mind when we're talking about these with eligible patients.

Sands: These are such important points. Just to add to the NLST follow-up you're describing, the initial publication suggested that there may be a big overdiagnosis just because of the incidence of cancer detection between the two groups. But as you highlight in further follow-up, the incidence of cancer between the two groups really became … I think it was a 20-person difference in the follow-up time frame, essentially showing that those that were in the chest x-ray arm (so, not getting a CT scan) had decreased numbers of lung cancer in the initial publication, but in follow-up years essentially they had more people diagnosed with a lung cancer during those follow-up years, that essentially brought those numbers pretty approximate. Just to put some numbers on what you said about the radiation dose, I've heard Andrea McKee, radiation oncologist at the Lahey program, talk about the fact that, in radiation oncology and radiology, the box they wear can get up to 50 mSV per year of radiation exposure and it's still considered occupationally safe. A lung screening scan is less than 1 mSV, so less than 1/50 of what our colleagues can get and still considered safe. But the other topic you brought up about people getting a procedure and not having a lung cancer diagnosis, I mean, this is a tough one, because obviously we want that to be 0%. But you did highlight the challenge of this and the balance of the number of lives that you're saving by doing the scanning, by intervening vs potentially leaving a cancer there. That seems to be a challenging one. Can you speak to, within the Lahey system or even beyond, what the best way of addressing that is?

Regis: The main difference between lung screening and mammography in this case is what we're doing with those positive findings. In mammography, it's pretty straightforward, where if you see something, you're going to get some tissue because it's really hard to do a ton of damage with a breast biopsy. But you can't be doing that with lungs, you know, just sticking a needle in everything that looks like it might be a problem. One of the bigger things is growth, for lung cancer screening, of whatever you're seeing in there. And so the vast majority of follow-up is just imaging. It's just imaging that's happening less than a year later as opposed to those who are in the annual screening. If it looks stable or if it goes away in that time, then you don't have to worry about anything. If it's growing or it still seems concerning for whatever reason … now we're talking about advanced imaging, like a PET scan or tissue sampling. But again, the way that we managed that at Lahey was basically everybody was going through a pulmonary specialist. So, you have somebody making those decisions who has not only the imaging to look at but also clinical expertise and an idea of the patient's individual situation in terms of what are the best next steps. That's really all you can ask for. We never did anything based on “okay, this lung screening exam shows this, so you need a biopsy.” It was always that somebody had to see a specialist first who was going to take everything into account before those kinds of decisions were made. And you see that the numbers are really low when it comes to benign disease interventions just by involving those specialists in making sure that they have a say in what happens with the patient moving forward.

Sands: And your group has published the very low numbers of intervention for non-lung cancer. So, we've covered the very substantial benefits and pretty limited risk. That being said, the American Lung Association releases a "state of lung cancer" report each year. They highlighted that the state that you've done your work in — Massachusetts — leads the country, with 18% of people who qualify getting screened. So really every state is failing, but Massachusetts leads at just 18%. Why do you think the numbers are so low?

Regis: There are a few reasons. And the first one, which often gets overlooked but is worth pointing out, is that this is still new. I mean, a lot of the other screening modalities were low for a while, too, and they didn't have one of the second problems, which is the stigma that's associated with lung cancer and smoking that has existed for quite a while, and it makes it easy for people to be not so supportive of lung screening or trying to find lung cancer in these patients. A lot of it is that this is largely something that your PCP would be recommending, like mammography, only less well known. And until this most recent USPSTF recommendation, the society for physicians — the AAFP [American Academy of Family Physicians] — wasn't supporting it. And so that makes it difficult to convince your primary care providers — when they have all of this other stuff that they need to talk about — that lung screening should be at the top of the list. And really it should. I'm hoping that we've turned that corner a bit. The conversation about risk-to-benefit ratio should be done. It's got more randomized controlled trials than any other screening modality. Now we're at the point of, all right, we've got to spread the word. We've got to get more facilities doing this and doing this correctly. And we've just got to screen patients. Let's get them in and get them screened because we know what this can do when we do that.

Sands: Now, all that being said, the program that you were running had a screening rate that was much higher. So how did you get there? What were the important cornerstones to building a really optimized screening program?

Regis: The two big things that I always preach, which are usually also the hardest to get, are a dedicated patient navigator, not somebody who is doing something else but now you're going to pull in to manage the lung screening program as well, because it's a massive undertaking and it can't be just something that you're doing on the side. And then some sort of program management system to help you track all of these patients. Because as much as I love Excel, it's really hard to manage a lung screening program doing that. But a lot of places are reluctant to dish out the money for those resources, and a lot of that goes back to what we've talked about with the misinformation that's out there and the stigma surrounding the disease, so it's not at the top of their list for them to support and it needs to be. Lahey was lucky to have Brady and Andrea McKee, who are very vocal and very persuasive. They talked to the primary care groups and got them to refer patients in. There were community events where we talked to the patients to explain to them what lung screening was and get them directly involved. So, I think the biggest thing is appropriate education. The shared decision-making aid that's in the JTO [Journal of Thoracic Oncology] has a lot of appropriate information in it. All the other stuff used to be based on the NLST study. Like we've pointed out a few times, as important as that study was, it isn't really relevant in terms of numbers anymore because it's not the way things are done clinically; so, it's important to have the right information. And that, I think, is what is going to get people to finally come to the right side.

Sands: Yes, that shared decision aid in the JTO from 2021, as you mentioned, that is really made to help primary care doctors have that discussion with patients, to be able to have a very rapid discussion with patients as to the risk and benefit of lung screening that hopefully assists in those discussions. Before we wrap up, where are some places that people can go who are trying to build their programs or optimize their programs? What kind of resources are out there?

Regis: Yeah, so there are a lot of good places to go. There is the ALA/ATS Implementation Guide that is out there, which has input from a bunch of different screening programs on how they do things and how they did things to get where they are. The Rescue Lung Society website has a lot of information about how to build a program and, again, a lot of the answers to the questions that you may have as you're going through this process. The ALA is a good resource. The Go2 foundation has a lot of good resources. There are a lot of places where you can go to get good information on how to help build your program, get patients in, and really run an effective program.

Sands: Well, to wrap things up, we've covered quite a bit today. We've discussed the risks and benefits of lung screening, essentially highlighting the NLST and NELSON studies, which really showed the benefits of lung screening that led to the ultimate USPSTF recommendation and then updates on increasing the population that qualifies, while also highlighting the fact that the true benefit of lung screening is probably more than what we see even in those studies. The risks of lung screening, as Dr. Regis said, are likely overstated. We essentially highlighted that radiation itself is pretty minimal and not really considered a substantial risk, and that the false-positive rate has really been unfortunately misstated and that the false-positive rate is quite a bit lower than what has been out there. I think you had said within the Lahey program, it's around 12% in that first scan, but only about 5% in follow-up scans. And in most cases, it's just a follow-up CT scan and not intervention that highlights the fact that ultimately ends up being negative and that the risk of intervention is actually quite low, but it is an ongoing area of effort and something that we want to get to 0% in people who do not have a lung cancer diagnosis. We've highlighted the pillars, essentially, of a lung screening program: having a navigator, having a multidisciplinary group, and working with primary care to really optimize their ability to order scans and counsel patients on that, as well as the JTO publication that has a shared decision aid that primary care practitioners can utilize in their discussions. And then the American Lung Association and ATS lung screening implementation guide is an excellent resource for a lot more information. Websites like also have some information on those as well. And with all of that, Dr. Regis, thank you so much for joining me today for this important discussion.

Regis: Thanks so much for having me. I appreciate it.

Sands: And thank you to all of you tuning in. This is Dr Jacob Sands for InDiscussion.


National Lung Screening Trial

Lung Cancer: Screening, 1996

NELSON lung cancer screening study

Early Lung Cancer Detection in High Risk Individuals (MILD)

Screening for Lung Cancer With Low-Dose Computed Tomography: An Evidence Review for the U.S. Preventive Services Task Force

Appendix A Table 5. Recommendations for Lung Cancer Screening With LDCT

ALA Press Release

Lung CT Screening Reporting & Data System (Lung-RADS)

Outcomes of Positive and Suspicious Findings in Clinical CT Lung Cancer Screening and the Road Ahead

Lung Cancer Incidence and Mortality with Extended Follow-up in the National Lung Screening Trial

Lung Screening Benefits and Challenges: A Review of The Data and Outline for Implementation

American Thoracic Society and American Lung Association Implementation Guide for Lung Cancer Screening

Rescue Lung Society

Lung Cancer Screening Resources

Go2 Foundation for Lung Cancer

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.