Good morning. Thank you host. Uh It's my pleasure to get us all started this morning by talking about lung nodules and lung cancer screening. Let's just make sure this is working. Ok. So over the next 10 minutes, I'm going to talk about some pertinent issues in lung cancer screening and possibly touch on some turn key mechanisms for lung nodule management and we'll see how far we get into that. So jumping into this as of 2022 the United States preventive task force upgraded or changed the recommendations as to who is an eligible candidate for lung cancer screening. It's anybody between the ages of 50 to 80 with a 20 pack history of smoking. A big change was changing it from 30 pack years to 20 pack years. There's uh and it included anybody who's a current smoker or quit within the last 15 years. The next big change came last year with the American Cancer Society that removing or striking off that need, that they should have quit within the last 15 years. So now anybody between 50 to 80 if you want to follow those guidelines with a 20 pack history of smoking is eligible for lung cancer screening. There's a lot of other recommendations from other societies, but they kind of are around this. So we certainly try to follow this as close as close to possible. But this has been the major issue with screening the uptake has been poor and the adherence is is really abysmal. So there's about 15 million screen eligible patients in the United States, just in our area. In the Philadelphia region, there's probably 90,000 and in the greater Phil region, about 200,000 patients. And you can see the national uptake of screening is just 6%. That grafter shows each state where they are in terms of their screening uptake of eligible patients. Pennsylvania does a little better at 8%. But still there's there's a long way to go and adherence is is poor by, you know, Silver Street study. Looking at the a lung cancer registry showed that only about 22% of patients come back for their second annual scan a year later, that does not get better than 30 or 40%. And that I think is the big problem because studies only show this works. If you do this persistently and continually over a period of years, not just one per one scan, why there's, you know, a lot of barriers, of course, by screening has been an issue, you can divide them into many different levels, whether there are issues related to patient education or physician knowledge or cultural issues or so. But I think the bigger part there is, is all issues related to organizational level, access to care, care coordinations. I think those are big things. Looking, some studies looked at what are the main factors that affect screening or so. And, and, and overwhelmingly it's lack of EMR prompts. We're all in, you know, relatively fancier, er, s and all we get inundated by a lot of messages. And so, but still, I think that becomes the, becomes the lead issue. Patient refuses another big problem. But I think that can be overcome by education, easily, so easily in terms of talking. But I think spreading the message out is much harder. Uh the risk of false positives, the perception of that, that's been an issue also. So I'll try to address a couple of these issues going forward, which I think are important. So one way to address gaps or to enrich the population of screening to kind of take care of these false positive issues to see how we can enrich, to make sure we screen people who are more at risk for cancer. We do know lung cancer is more likely to happen when you know, beyond the age of 65 and not, not when you're close to 50 or so. So there's been a push to consider risk based screening, using models to identify someone's risk versus just doing a hard cut off from age or, or tobacco use. So there's different models, I just listed a few models, there are slightly different in, in some of the factors that they screen for. So for example, the PL C model looks at COPD, the B model looks at asbestos exposure. The Liverpool model, which is the basis for the UK lung prevention screening trial uses a history of pneumonia. There's been studies looking on the on the figure next to it that the sensitivity of using a risk prediction model is much better than just the hardcore analytic criteria. That graphs are little skewed. But the top orange dot over there is the is using the PC O modified criteria as a risk risk model versus using just the analytic criteria, which is the blue diamond at the bottom. It's clearly shown that this will improve efficiency of screening. It's also shown to be cost effective. And then the other big thing has been uh oh, this is just an example of the model that we'd like to show and share. I just put it out there for people who are interested. The website is called should I screen.com? It is a good pictorial representation of a patient's risk of uh of developing lung cancer over the next six years. And it gives a nice display of risks versus benefit, makes the conversation easier and helps the patient decide, you know, in an unbiased way if they want to consider this or not disparity. So I don't know how, how many of you feel confident that uh that you get a good smoking history or in other ways that you think that every time you ask, ask a history it's gonna change. Like I feel every time I ask a patient there's something different in their smoking exposure. So, so this question hits at something that's been going on now for the last few years. It's actually an older question but has picked up more debate and discussion for the last couple of years now is duration of smoking more important or is the intensity of smoking more important. So I'm just giving an example of one study here, a Southern community cohort study looking at cancer prevalence in their population. In their study, the four panel graph over there shows that all black individuals who develop lung cancer, they had significantly fewer pack years exposure. As compared to the white population. You can see that that's 2525 versus 49 pack year history. And that difference was mainly attributed to the less amount of intensity of cigarette smoking that 1212 cigarette use versus versus median of 20 cigarettes, cigarette users in the white population. So we we do know already that uh uh non Caucasian populations with the same amount of smoking exposure can develop lung cancer at a younger age or so. So this is trying to hit, hit on that disparity. And I think this was a pretty moving graph for me when I was reviewing literature over the last year or so, what if we adjusted the eligibility criteria from pack years to just years or duration of exposure? So in panels A and B that scatter plot shows all the individuals who were diagnosed with lung cancer in the cohort, the red lines mark the US PFT criteria of ages 50 to 80 with the with panels A and B being 20 pack years. But then in panel C and D, it is just a duration of 20 years. And if if you look at the first two panels, only 62% of black individuals would have been eligible for lung cancer screening by the US PFT criteria versus 82% of white individuals if you change that to just so you miss all those people in the in the shaded green area. If you try to use only a hard cut off of 20 years and forget the intensity part of this, you kind of take care of those scatter plots now kind of mirror each other. So not only does it pick up more individuals but but the disparity goes away. So it's not a final word on this, but I think this is where things will probably change and move. So certainly to address some of those issues. A quick word on screening in non-smokers, I often get that question in the clinic from family members and patients and all the bottom panel there represents, just discusses. The ILC has a working committee. Now looking at, you know, addressing this question as best as possible. They're working on constructing risk prediction models in non-smokers, looking at other risk factors involving family history, airflow obstruction BMI. So under demographics and incorporating biomarkers, that's probably where the field is going to go towards a little more. I only got one study up there, the one the talent study from Taiwan, they are actually, I think in their sixth year now of this study where they enrolled all non smokers who have a significant, either had a significant family history of lung cancer or COPD. They look at cooking exposure and there, there's a nice, nice factor there and then passive smoking. And so far four year results show that there are 2.6% of their participants were diagnosed with lung cancer. These are all non smokers again and the pertinent factors that are coming out in the individuals with the family history, uh females age greater than 60. There was, those are significant factors. Again, this is not a, there's no control arm to this. This is just a cohort of non-smokers being followed prospectively. So sometimes I will, there's a lot of patients who come in clinic and they are really worried that they've had, you know, one of their, their parents had a lung cancer. I certainly will sometimes quote this and then let them choose or decide if you should order a scan or not, it may or may not be covered by insurance, which I usually tell them upfront. Uh I'm going to switch gears and just talk about a little bit on, on, on the nodule. Follow up. So, nodules, whether they're screen detected or incidental and they're a little different in there in, in how you would follow them or so. But it really comes down to this basic question about how am I going to risk stratifying nodule? There's, there's, you know, I encourage the use of calculators, there's calculators out there which are easy to use. I think it helps put things in perspective and in objective perspective to a patient, you know, certainly incorporating a pet in there. A lot of us see patients already with pet scans and all working up. No, I think that's, that's helpful. The number to, to look there from the Brock calculator is more than 10% that is significant. In fact, the British Thoracic Society guidelines now incorporate the Brock model into the work up of, of a lung nodule to decide whether we should do an invasive test or not, do an invasive test future. Though, I think is going to go towards hes looking at A I and machine language algorithms and all to further risk stratify knowledge. There's a couple of products already out there if the approved, not ready for prime time. But I think, I think they will get to that and you know, traditionally, we've always classified nodules as a low risk as less than 5% high risk as greater than 65%. But it's that 5 to 65% piece that indeterminate nodule, which then needs further stratification. Biomarkers are kind of moving into that. There's a biomarker out there based on the pan optic structure, looking at five risk factors plus two proteins or rate ratio of two proteins. We use that more to kind of rule you rule out cancer. Basically the negative predictive value of that is very high. But I think this is probably the bigger challenge. So I think that is all fancy and all this has been the issue and I think continues to be the issue. It's access to the right kind of providers loss to follow up management delays. So right from the identification of a nodule where where a radiologist picks up an either in single finding or a screen nodule from them, getting to their PC P or from the PC P to the specialist. Once they're to the specialist, what kind of work up happens? They go from pulmonary to thoracic to cardiology or so at every step, there is attrition at every step or so, there's attrition. And even with screen detected nodules, we sometimes realize it's been three months before somebody acted on, you know, a lung grads for scan or so. So we we tried over the last year or two to look at internally ourselves. And we came up with two pilot projects to look at how we can bridge this gap a little bit. So we created a lung nodule pathway. The top of that just talks about dring rapid access to nodules, like which is the most concerning nodule. The nodule that bothers your patient is the most concerning nodule. We didn't come up with any size right here with anybody calling with a nodule. That's what we need to get them rapid access into our system. So doing that. So we, we, we constructed what's called a nodule clinic. But all the things in gray is what is the usual work up of any nodule that you would think of whether it's a interval city is most likely. But then all the other things that are needed up to get them management. We try to order them up front. The other big thing we changed was we tried to create that follow up results encounter in the hopes to standardize things where we have a touch point with a lung nodule coordinator whose job is not just just reporting and screen, but they can actually place orders and act on the next step or so. And then from subsequent to there get getting on to the therapeutic arms or so. So, and then we leverage a tool in the EMR that's available called episodes of care. It, it basically is a, is a tool or a reporting structure that helps link a preselect set of orders or encounters. So you could place you could, you could link it to a diagnosis. In our case, we link a diagnosis of lung nodule to a pet scan to an MR to a PFT and then to consults whether to, to pulmonary medicine consult or thoracic surgery or medical oncology consult. And we, we try to use that to create, create, create pathways. And the good thing about this system is that you can let me skip ahead. That's what the list looks like. See if I have a pointer here. Yeah. So you can look at it. It tells you, for example, when was the pet ordered? And, and where does the pet actually schedule? It tells you how many dates out this. So we've used this as a, as a working list for our, for our nodule coordinator. Look at, oh, a better scheduled way out or so let me just call and pull that sooner, you know. So this is kind of an active and this is, this is a real time list. It's very easy to filter. I can tell you within seconds. Now, how many patients I saw within the last month? And where are the work up or? So you don't have to keep Excel sheets or do everything in your EMR. So using this and looking and we were able to look at some, some, you know, each physician's profile to see if there's variability in management or so. And our, our, our pre pre intervention before we use that pilot and incorporate our lung nodule coordinator, we were surprised we didn't know we were going to be that bad where it takes 82 days from hitting our doors to getting surgery. It was about 80 or surgery or radiation or so was 80 days. We were able to drop that time down to 46 in our, in our post intervention pilot. So and similarly, for late stage cancer, late stage, there is stage three and four. So we were able to drop that time to treatment. Also from 40 to 30. We between a lot of people in this room, we have an active project going on now to just validate these, these findings more and hopefully, uh hopefully we will get better than what we are. So in summary, I'd say that, you know, this continued effort certainly needed to increase uptake of CT screening. The more we talk, the more we educate our patients about that and ourselves. Hopefully these numbers will come up, screening will probably continue to evolve and you might see some changes related to using either adding biomarkers or, or or or changing criteria in terms of uh of some things coming in the future. And of course, improved access and timely management to decrease time to treatment I think is one of the key challenges that we face in terms of not World Cup, we we're very lucky to be in a very closely integrated system. I think our, our teams are, are awesome. We, we integrate all our teams along with tobacco cessation and screening programs, I think is one of our big advantages. So I'll stop with that. And thank you very much.
Related Presenters
