Liquid Biopsies in Lung Cancer: The Future Is Now [Video Transcript]
In this segment I'm going to be talking about liquid biopsies which maybe 5 to 10 years ago sounded like science fiction but now it's real and it's here to stay. I think what we found out over the past 5 to 10 years is that technological advances have really ushered in a new way to perform molecular interrogation of lung cancer.
Historically, the gold standard for molecular interrogation or genotyping for lung cancer has been tissue biopsies. It's so important that we identify certain mutations within lung cancer for advanced age patients. Usually that way has been either an FNA or a core biopsy of their lung mass or metastatic site.
I think all of us who've tried to coordinate these biopsies and witness patients go through them know that these biopsies are fraught with complications: you have to coordinate them, they are invasive, they can lead to complications. Just getting the tissue sent to the right place: the pathologist has to read it, the tissue is then sent out for molecular testing and the result has to be yielded back to you. All of these steps lend itself to mishaps and missteps in which the information that you're trying to get, you're not able to receive.
I think thankfully and fortunately, we now have an alternative way in which we can genotype or interrogate or find the genetic underpinnings of a tumor by a simple, minimally invasive blood test. I think we think of liquid biopsies historically as circulating tumor cells and that's not what this is. Circulating tumor cells certainly have a lot of data prognostically for prostate cancer and breast cancer. This is cell-free DNA or circulating tumor DNA. This is DNA that is shed from the tumor into the bloodstream that now can be picked up and analyzed with a simple assay.
I think in lung cancer, specifically, this has really led the way. Certainly other tumor types have started to use liquid biopsies. But in lung cancer, I think what we know is that for specific mutations like T790M, the sensitivity's around anywhere from 60% to 70% but the specificity's quite high, it's around 90%.
I think we see this with both EGFR mutations, maybe a little bit low for T790M mutations, that when you have a specificity that high, if you have a positive that's fairly certain that this inaccurate molecular proxy of disease and that patient does indeed have that mutation which you have identified in the blood. With the specificity that high, that certainly is important. Sensitivity being 60-70% means that if it's negative in the blood, it still could be in the tissue.
Also what we have seen is that for patients that are plasma-positive for T790M, the response rates are almost identical to those patients that are tissue-positive T790M, suggesting again that if you are T790M either in the blood or in the tissue, your response rates are identical to T790M-directed therapy with osimertinib or Tagrisso, suggesting that this is an accurate molecular proxy for tissue.
Currently we have an FDA approved blood-based test for EGFR mutations and I will tell you that patients who are on an EGFR TKI who are progressing and which are thinking about a therapy switch, I will start with blood first to identify T790M. There are multiple platforms out there, some of them do Next Gen sequencing, some of them do digital droplet PCR. I can tell you that if you are in the cancer world and treating patients. You certainly will find a vendor or company that will do this for you. It can be done in the office or it can even be done at an off-site lab.
So how do I see this all playing out in the future? For one, it's already playing out in the clinical scenario where a patient who is on an EGFR TKI is progressing or their cancer is growing, you're thinking about a therapy switch, this is a case where you should do a blood test. A blood test should be done. You should interrogate for T790M. If it's positive you can give a T790M-directed therapy with Tagrisso.
The second place I see this or the second scenario I see this playing out is upfront for patients in which you've done the biopsy, you confirmed that they have a lung cancer but there's not enough tissue for genetic interrogation. I think it's very reasonable to consider doing a blood-based test on them to see if you can pick it up in the blood and we've had many success stories in our own clinic in patients in which we did the biopsy, we confirmed that they had an adenocarcinoma, but there wasn't enough tissue left over to do the genotyping, we were able to do blood, identify an EGFR mutation and one patient identifying an ALK rearrangement in which we gave him TKI therapy.
The third area or clinical scenario where I think this may play a part – hasn't done it yet, we need the trials – is looking at dynamic changes in mutational load in the blood over time to see if that correlates without cause. This is not ready for primetime, but for instance, looking at EGFR copy number in the blood and seeing whether acute drop in this number when patients are getting targeted therapies means anything for that patient on down the road, or if patients don't have that acute drop does that mean they should be switched to something right away and different.
Again we're not ready for time for this scenario. There's been some nice publications on this looking at drops in EGFR copy number in the blood and correlating it with a scan at 6 weeks showing that those patient that had a fast drop, their scan showed a much greater shrinkage in tumor than those that did not have a fast drop.
Finally moving to the curative setting, and again, not ready for primetime here, but certainly some data suggesting that for patients in which you've done curative intent surgery or chemoradiation, and what you're trying to go for cure rather than monitoring them with scans, perhaps monitoring them with scans plus a blood test to pick up on minimal residual disease. I think that would be really where we need to head if we are trying to cure patients, currently when patients come to the office and they receive surgery and chemotherapy, the only thing I can offer them is a scan every 3 to 6 months to monitor for recurrence.
If I'm able to do, in addition, a blood test that may be able to be more sensitive at picking up disease where we can act on something sooner, that would be a huge win. We are certainly not there yet. I think currently where I see liquid biopsies playing a big role is in the resistance setting for EGFR TKI therapy and in the upfront settings when you don't have enough tissue left over or it's exhausted to do the tissue genotyping. But dynamic changes over time and then monitoring for patients who are cured may be where we're headed with this next and hopefully in the next 5 to 10 years we'll see that.