ELCC 2021 | Lung cancer screening: NELSON and NLST

I have a privilege of talking about NELSON, NLST, are the results similar? And actually, what I start off with is actually, we probably should have had a different title. Now that we have two large international trials, why are we waiting to implement lung cancer screening?

However, I think that we actually can say that we have two excellent trials. Both of them demonstrating a mortality advantage of 20-24% in the CT arm. And it’s really what it comes out of the NELSON trial that I’m going to go through them without sort of going through slide by slide. I think the thing is that one has to appreciate that there are a couple of areas I’m going to focus on.

And I think something that probably people picked up at the initial presentation that Harry de Koning gave at the world lung cancer meeting, is that in fact there may be actually a greater benefit for women. The NELSON trial didn’t actually have a sufficient number of women to have that as a statistical outcome. However, the German trial demonstrated their actual benefit was actually in women. So, I think it’s something that we need to actually look out into a much greater detail, and it’s quite possible that there are a number of reasons behind this. And it may be that there’s higher, let’s say, disease in men, let’s say cardiac disease and so on.

So, I think thing is that it’s something that we need to consider. And also, actually does that mean, do we change the recruitment way for women? Do we take them on earlier? Do we identify the disease earlier? So I think it opens up a completely, let’s say, new area that hadn’t really been considered until Harry de Koning presented his data.

The other is actually probably over-diagnosis. Now, again, over-diagnosis is a very large question and the reasons for it. And I think the thing is that the NELSON reported 8.9% where the NLST actually produced a figure of 18% initially, then they suggested it was possibly went down much lower to 3% in a recent publication. But I think what we have to consider is that there’s no room for complacency in this regard. And there is a balance of what we would call over-diagnosis or mortality reduction which will be acceptable. And that is an ongoing question I think that we need to consider.

The other area is further investigations. The NLST actually had approximately 20% of our participants that had to undergo at least one additional scan for a tumor growth or aggression. Now, the NELSON introduced a, if I can call it, a very rigorous radiology protocol and they utilized volume and volume doubling time. And therefore, they actually probably managed to bring this figure down dramatically. So, they have percentages between 1.9 and 6.7% in subsequent screenings, but really overall, the average was probably less than 10%. And therefore, we have to consider, again, going forward, what is an acceptable, further investigation rate. And I think 10% at this moment in time is probably something that we would consider acceptable.

If I can just sort of move slightly to one side. I’ve been involved recently in a Nature Reviews and Clinical Oncology review. And this was with my colleague Matthijs Oudkerk. And one of the areas is actually still outstanding is cost-effectiveness. And we provided data on all of the cost-effectiveness, and it really depends on what each nation considers cost-effective. So, it’s about a $100,000 in the States per [inaudible], whereas in the UK were working around £20,000. But actually, if you look at the data, and there’s about eight or nine publications out, we have actually demonstrated and a lot of it is modelling, we undertook it also for the UKLS, that in fact it would appear that lung cancer will be cost-effective. And I think this is a very important question that we answer in the years that are coming forward.

And also, looking forward, I have already said we need to just get on with implementing lung cancer screening, but there are outstanding questions. I ran a workshop at the meeting in Geneva back in 2019 from which there was a ESMO publication. And that was when we came out with series of recommendations. But actually, we need to think of implementation research. And probably what’s at the forefront of this is a new Horizon program called 4-IN LUNG RUN. The PI for this is Harry de Koning, and there are a number of outstanding questions that can be answered in ongoing trials. It doesn’t mean we stop; we carry on. But in fact, we can learn by let’s say, moving forward and asking these questions.

But I think I also need to mention, I feel that the UK has really led implementation of lung cancer screening within the European environment. And after the UKLS trial was published, first of all the Liverpool Healthy Lung Project was initiated and it was covered by the Liverpool CCG, followed by the Manchester Lung Health Check, followed by the Lung Screen Uptake Trial, followed by the West London Lung Cancer Screening Trial, and then there’s a Yorkshire Lung Screening Trial. And all of these have sort of built up a lot of impact within the country that they’d actually want to get on and undertake screening. And that, I feel, really created the springboard for the NHS England targeted screening.

Now we were all about to start this and the area I’m in is Liverpool, Holton and Knowsley, but there first of all were 10 sites. Now there’s many more that have been identified. Unfortunately, COVID has brought that to a, let’s say, it’s delayed its implementation, but hopefully we can start that again if I sort of say this summer going into the autumn. And when you actually think of the numbers of the rollout has the potential, and his was really at the first stage, of reaching about 600,000 people over four years and detecting probably about 3,400 cancers. And if we look at all the new sites that are now being involved, that figure will be increased.

And in fact, I just wanted to leave the participants watching my presentation with what we call SPIRAL, Screening, Planning, and Implementation Rationale for Lung cancer. And this is an area that we feel we have identified all of the major questions that need to be addressed in the coming years. How to identify high-risk individuals. If I can just put in, we’ve done validated the LLP, that’s now LLP version three, and we’ve published in Thorax this year, and NHS England are using the LLP and the PLCO to identify high-risk individuals. It’s how we recruit individuals, how we actually bring people on board. Probably after COVID it’s going to be through telephone calls, but actually do we have one-to-one with nurses. Then the whole area of nodule management and British Thoracic Society have produced a guidelines. And that has continuously been upgraded, let’s say revised. And one really looks at how one can improve that.

The screening frequency. Do we actually screen every year, or can we actually use the results from the first year of screening, excuse me, that will help us actually say, well, these people are at lower risk and possibly only screen every two years? These are suggestions and these are our areas that people are working on. And in fact, from a financial point of view, yes, there will be benefits, but how many cancers would we miss? And that is something that is continuously considered.

I’ve already mentioned the sex differences. Is there something within the results that came from NELSON that we should be looking into a greater detail. I’ve already mentioned cost-effectiveness. And I think that will be extremely important looking across Europe to ensure that lung cancer screening is cost-effective. And that is actually building the framework for the research implementation projects. So, my view is yes, we have sufficient information, we can learn a great deal by taking it forward, but the sooner we actually implement lung cancer screening within Europe the better.