WCLC Sept 2021 | Strategies for early cancer detection with liquid biopsy

During the World Lung Cancer Conference, I had the opportunity to discuss about liquid biopsy in a special session regarding to this technology and the data that we are collecting in different scenarios. One of them is early detection, and we discussed there the different techniques and different approach that are used nowadays for detection.

One of them is the ctDNA and there, there are two technologies that are very important. One was recently approved by FDA and is the whole-genome methylation. The second one is the fragmentomics using the DELFI approach, and these two technologies bring interesting data. The first one, the methylation, is very important because we have there a high specificity in the validation, so important to remember that this study was coming from a discovery, a training, and this is the validation finally published in 2021 in Annals of Oncology, where we have a high specificity, 99.5%, in multicancer approach, so this including not only solid tumor, but also hematological disease. And finally, important also the number of patients, more than 2,800 patients including also in the validation, stage four.

If we are looking specifically in the data of lung cancer, there are differences in the sensitivity per stage. Obviously, stage III and IV are performing very high in the sensitivity, where stage I and stage II are still with a low sensitivity, specifically for stage I in all the cohorts, of all the tumors, it was around 16%. That is something that we also observed within lung cancer specifically, very low for stage I. Several concerns, obviously, about this technology that we don’t know specifically the different subtypes sensitivity, that is something that were presented in 2019, and we saw in this preliminary data the difference between squamous and adenocarcinoma. And also important that, in this trial, we didn’t have any data for screening. So we didn’t have a validation, obviously, because it’s in a multi-target, multi-cancer study. So it was not able to go indeed in high-risk populations in the case of lung cancer in the patients part of the screening, the normal screening for lung cancer.

The second study was the fragmentomics, the DELFI approach, and this is using whole-genome sequencing. And from there, we go for the fragments of DNA. And using this technology, there are data confirming the difference in histology as well and we saw that the performance is also different between adenocarcinoma and squamous. The sensitivity was analyzed and then a specificity of 80%, so lower than the methylation study. And here also, we saw that, in the early stage, some patients are missing. What is interesting in this study, in the [inaudible] cohort and in the validation cohort of this DELFI approach, is that we have data for high-risk populations. So this could be an approach that could be used in the screening. Obviously, fragmentomics, we need to recognize that it is a very complex process and different factors could interfere in the analysis and in the determination. So it’s important also to see that if we want to go for a large scale of application in the future.

Other technologies, like CTC, at the moment where negative and that is because I presume that was for the use of the technology in the isolation of the CTCs. But there are several trials going on in CTC as well that we would see an advance in the technology and a future application. Other approaches, metabolomics, and in the metabolomics, some companies with investigators were able to identify some metabolites that are able to predict early-stage and even benign versus cancer nodules, so that is important also for the screening in the future. The approach of microRNAs that we knew from the World Lung Cancer Conference as well, that this approach is interesting, obviously, and microRNAs are a very old friends in liquid biopsy, we would say, and to reproduce this data will be a little bit difficult in the future for a large scale of patients.

So, in the future, what I see that we are maybe going from a multi-omics approach, maybe that is putting together the different technologies in order that we are able to have a good sensitivity with a high specificity. But the important data is that we have already started this road for early detection and having an opportunity to save life including a higher sensitivity that is in addition to the low-dose CT scans in the screening products.