How Are Prenatal Blood Tests Detecting Cancer?
20:55 Share

This episode is brought to you by Chevy Silverado. When it's time for you to ditch the blacktop and head off-road, do it in a truck that says no to nothing. The Chevy Silverado Trail Boss. Get the rugged capability of its Z71 suspension and 2-inch factory lift. Plus, impressive torque and towing capacity thanks to an available Duramax 3-liter turbo diesel engine. Where other trucks call it quits, you'll just be getting started. Visit Chevy.com to learn more.

Hi, I'm Clara Moskowitz, senior editor for space and physics at Scientific American. Like many kids, I once dreamed of becoming an astronaut. While I never made it to space, my work at Scientific American has given me the next best thing.

Exploring the cosmos through stories and sharing its wonders with science lovers like you. When I research a story, I immerse myself in the reporting to bring you an exciting and accurate account. Over the years, I've covered breathtaking rocket launches, visited one of the world's highest altitude telescopes in Chile, and even trained for suborbital spaceflight.

Space is vast, beautiful, and full of the unexpected. Taking a moment to look beyond our daily routines and reflect on its mysteries can be a powerful escape. Join me on this journey of discovery. Subscribe to Scientific American today at sciam.com slash get sciam. For Scientific American Science Quickly, I'm Rachel Feltman. ♪

For more than a decade, non-invasive prenatal blood testing, or NIPT, has been a fairly routine aspect of pregnancy care. This testing searches a pregnant person's blood for fragments of DNA that have been shed by the placenta. NIPT is designed to spot chromosomal disorders in the fetus, but in rare cases, the blood test can detect something else: cancer in the parent.

My guest today is Laura Herscher, a genetic counselor and director of student research at the Sarah Lawrence College Joan H. Marks Graduate Program in Human Genetics. She recently wrote a piece for Scientific American about the researchers working to understand how NIPT finds cancer in some pregnant people.

She's here to tell us more about the incidental detection of maternal neoplasia through non-invasive cell-free DNA analysis study, or IDENTIFY for short. Laura, thank you so much for coming on to chat today. So you recently wrote about something called the IDENTIFY study. How did you get interested in the story? Well, the first time I heard about IDENTIFY was when the principal investigator, Diana Bianchi, from National Institute of Health,

talked about her initial results, which was just about a year ago. It was like March of 2024. And she said that she was looking at something I had heard about that had crossed my radar, but I really wasn't paying that much attention, which was these weird and funky prenatal testing results, which instead of giving back information on the fetus, unexpectedly provided information on the pregnant individual

themselves. And I'd heard somewhat almost by rumor from prenatal genetic counselors that sometimes these oddball genetic testing results actually seem to see a signal for cancer in the mom. But you know, most cancer screens, even when they're set up to be cancer screens, the return on them is usually limited. However, in this case, she reported that they had looked at the first about 100 women who

that were worked up after getting these unusual results on a test that they thought was on their fetus, and almost exactly half of them

had cancer. And you have to consider that this is a group of young individuals because they're all of childbearing age, right? Right. So to see such a strong cancer signal in that population, it really blew me away. So I wanted to know more and so I investigated into it. Yeah, absolutely. So backing up a little bit, what is the prenatal testing here that we're talking about and how common is it?

How many pregnant people are getting these tests? Millions. So it's quite common. At least, I would say, about 50% of all pregnancies today use this test. It's called non-invasive prenatal testing. Let's just call it NIPT. So NIPT is a test that came along that sort of solved a problem from the point of view of expecting couples.

The problem was that before NIPT, we offered testing for Down syndrome and other unusual chromosomal presentations in the fetus to essentially every pregnant individual, right? But there were sort of two types of testing. And one was amniocentesis or CVS, where you use a needle to get a sample from either the placenta or the amniotic fluid in the uterus.

And as that's probably suggest to people, it's invasive, right? And so not everyone was comfortable with that. And most importantly, even though it was very small, there was a risk of losing the pregnancy associated with both of those tests. So there were some inexpensive and it's a big deal. And a lot of people would rather avoid that needle if they could. And there was another type of test called

which was simply a blood test from a pregnant person that looked at biomarkers associated with various

chromosomal issues in the fetus. And that was an easy test and cheap, but had tons of false positives. In fact, you know, when I was first in this field and would see prenatal patients, they'd come in with a positive biomarker screen. And it could be quite reassuring because like the chances were usually always greater than not that it was nothing.

Right? Right. So that's nice, but also it means that a lot of people get agitated, have follow-up care and so on, and turn out that it's nothing. It wasn't super popular, you know? Right. There was a push to look for a test

that would find fetal cells in the maternal bloodstream. So you could directly look at the fetal DNA, but you wouldn't have to get into the uterus in any fashion. Right. Really, that's never turned out to be possible. It's very complicated. It's difficult. There's been sort of tantalizing hints, but no, it's never been successful as a test.

But what a Hong Kong-based researcher realized was that, you know, as cells die, in normal course of cell death, as cells turn over, they dump little bits of DNA into the bloodstream. That's not when you're pregnant. That's for everybody all the time. And this gets very rapidly cleaned up, recycled. And that DNA isn't nice and neatly contained in chromosomes. It's chopped up into little pieces.

like a jigsaw puzzle, which is good to hold onto that image, right? And you can sequence those little pieces and then trace them back to what chromosome they came from by using the knowledge we have of the human genome like it was the picture on the cover of the box when you're doing a jigsaw puzzle, right? So you can trace it right back and figure out, oh, this little segment came from chromosome one, this little segment came from chromosome 10, and so on.

And those pieces should show up in the bloodstream in direct proportion to the size of the chromosome. So chromosome one is the biggest chromosome, so it should have the most pieces, right? Like that's just math. Like if you get a good sample, that's what the sample should look like from the biggest to the smallest chromosome.

So if you're pregnant, a portion of that cell-free DNA, those little pieces, comes from the placenta. So what Lowe figured out was that you didn't need to disentangle the fetal DNA from the maternal DNA, which is super complicated, in order to get a sense of whether the numbers were off.

because you could just assume that the maternal snippets of DNA would represent typical chromosomes because you know that person's chromosomes. And so any deviation from the expected numbers should be coming from the fetus. So it's super hard math, like it's really have to be very precise, but the idea of it's pretty straightforward, right? And it mostly works. But what it means is there's a

fewer, not zero, it's still a screen, but there's fewer false positives. But it's really quick, the adoption of this test. It goes from zero to millions very, very fast. And then right away, rarely but regularly, we start seeing this funny thing. The funny thing are results where you're seeing a signal of extra or missing chromosomes.

Not one, but multiple. And you look at this report, and it's supposed to be a report on fetal chromosomes, and you're like, this fetus should not be alive. Right. This is not compatible with life. And yet, you look at the ultrasound, and there you have a happily developing fetus. Looks fine, looks normal. It is what the doctors call discordant.

It doesn't make any sense. So we didn't quite know what to make of them. And the labs started off by just sort of saying, like, we're turning you these results. We don't understand them. They came to call them non-reportable, which is different from sort of a test fail. Let's do it over. These were like, nope, don't do it over. This just doesn't work. Something's off. Right.

And they didn't know, but then they started to have, you know, incidental findings where doctors would come back and say, look, six months ago I had this person come in and they had these non-reportable results. And now I hear that this person has cancer. And at first it was just the occasional anecdote.

which, you know, you can't really send a report based on an occasional anecdote. So over time, they started looking into this, became clearer that what we were getting was a signal, not from the fetus, but from the mother. And it was happening not often, but one in every 8,000 to 10,000 cases, which when you have millions of tests a year is really quite a number of people. Yeah. It sounds like...

You know, it was a long process to sort of turn the ANIC data into a signal that was worth pursuing. How did the researchers in charge of this study get really interested and figure out what was going on? Well, Diana Bianchi, who's the principal investigator, is somebody who's been working with this test, MIPT, really from its very beginning.

And I think what Dr. Bianchi found was that there were some studies showing a pretty intriguing signal of cancer, but some of them were from the labs themselves. In the United States, you have multiple labs. You have like 12 different labs that offer NIBT. Each one is slightly different. They're not exactly the same. And so it was hard to get like one simple answer. And without a simple answer, you can send a report back saying we're concerned about maternal malignancy, but...

Is the insurance company going to pay for that? It's not validated. And also, were the doctors going to take it seriously? They hadn't seen it before. There was a lot of ways for this not to be used properly, this information to get lost. So this group at NIH said, "Right, okay, we're going to do an objective study. Anybody that fits those criteria, discordant results, that person, they would pay all of their expenses to come to NIH. They would give them a full and total workup.

And that would allow them to see how many of them actually, if any, had cancer, what was the effective way of finding it, what type of cancers were these, so on, all these questions, and provide guidance for the labs. And they thought they were going to find something. But what Dr. Bianchi said to me, I said, were you surprised by how many? She said, hell yeah, we were surprised. I mean, almost half. And if you actually looked at it closely, said in the paper,

but they could further refine the signal. So when the women showed up and they actually had more than two full chromosomes missing or added, I think the numbers were 49 of them had this pattern and 47 of those people had cancer. Wow. So that was...

really, really strong signal, but even in the bigger group. And there were other things. Sometimes people had fibroids. Sometimes the test was wrong. And sometimes there were other things. Sometimes there's people that they think like, well, we don't know. We got to follow those people out. Maybe we just can't find it yet. But really, a lot of them already had cancer and

And essentially, they either had no symptoms or they had symptoms that were easily mistaken for pregnancy. Right. Something that really struck me in your piece was, and I'm definitely oversimplifying this, but you got into sort of the idea that in medicine, people are very uncomfortable about the idea of pregnant people with cancer. And it's a very fraught topic and how

maybe that contributed somewhat to the disconnect between the many pieces here as this data was coming together. Could you talk a little bit more about that? Sure. I think there's different pieces to it. One is simply, as Dr. Bianchi said, there's sort of a lot of history around the idea of like, don't touch pregnant people. It's not comfortable. And, you know, medicine's quite siloed. So the OBGYNs, not super comfortable.

having patients who have cancer or might have cancer, how does he work that up? It really requires them to go out and find

oncologists who they would be able to explain this to, right? Because this isn't some way that some, you know, show up and they're like, well, they have no symptoms. Right. There's no particular reason except I have this funky test result and we have no idea what cancer this might be. I mean, it's a very odd presentation, right? And the oncologists are obviously not used to working with pregnant people.

We talked to one person who had this experience. I talked to one woman who showed up for her prenatal results session and the geneticist she was talking to said, well, actually, you know, we think this signal is coming from you. And she almost didn't follow up on it because she said, I felt great. I actually, you know, had never been in better shape in my life. I was like, that's crazy. I'm fine. Like,

I thought you were going to say there's something about the fetus. And the person kind of said, look, go get the workup. So she did. And to her shock and horror, they found a fairly large and aggressive lymphoma. And she had to be treated during her pregnancy. She had her last chemotherapy treatment two weeks before she gave birth. And she said that she was really lucky because in Washington, she was able to find a center where they could do coordinated care. And for her, that meant that

OB would send somebody over every time she had an infusion to just check on the baby's heartbeat, make sure it was okay. There was constant communication back and forth. And obviously not everybody's gonna have that available and it can be scary and uncomfortable. Beyond that, there's a whole other layer of this story

which is that people who discover they have cancer during a pregnancy, there are times when appropriate care means discussing terminating the pregnancy because that may result in a better outcome. That doesn't mean that they have to do it or that's the option they're going to choose. But in the situation where best outcomes are associated with treatment that can only be undertaken if a person is not pregnant...

then that's a discussion the oncologist needs to have. But that's one of the areas where we've seen since 2022 that exceptions are really challenging in stringent anti-abortion laws. Right. Because very often the doctor is caught in a bind where the standard of care would be to present that as one of the options, but the law may say you're not allowed to present that as an option because

unless the person is literally dying. And in oncology, that's not the way that's gonna look. What's it gonna look like is what are your chances of being alive in five years? What are your chances of being alive to raise this child? And so the laws weren't written by the doctors. They don't really have a lot of flexibility or nuance in them. And in those cases, oncologists can be really in a bind. And so this test,

which doesn't identify, you know, there's not more people getting cancer because of the test, but what it's really doing is the early identification is moving people that would maybe be diagnosed with cancer a month, two months, six months, two years later. That's happening earlier on and while they're pregnant. Right. So it just creates more of these sort of difficult and conflictual situations.

Yeah. Well, and, you know, with the average age of pregnancy being higher and, of course, seeing these upticks in cancer in younger cohorts, you know, is there concern that this is an issue we need to get better at handling in general, the idea of people having cancers diagnosed during pregnancy? Yeah, no, absolutely. I mean, all over the world, the average age at which people have children is increasing.

And, you know, cancer is a direct in a linear fashion associated with older age. So we do have more. Our numbers of what percentage of women who are pregnant will present with cancer are out of date. So it's definitely a rare thing, but it's an increasing rare thing. And the other thing which Dr. Bianchi really stressed is we

We are also getting better at being able to treat people while they're pregnant. We can do more of it than we thought. She told me, I thought this was really striking. She told me that particularly there was a group in Belgium that's been very active about looking at this and like being creative about the ways to treat. And one of the problems they have is the dye that you use when you do this test to look for tumors, the contrast dye.

was toxic for people who were pregnant. And they found that by getting them to drink a whole lot of pineapple juice, that worked as well. So, you know, they're finding that things that maybe they thought they couldn't do, they really can do. And it really, really emphasizes the need to not lose these people to make sure that they are found when they can be found. I mean, a number that really stuck with me from this, of those 47 individuals

cancer, by the time they went to press with this article, six of those, remember, relatively young people were dead. Wow. Yeah. So these were not a series of trivial or meaningless findings. The most common finding was lymphoma. The second most common finding was

colon cancer, which separately but relatedly is on the rise in younger adults in America today. So there are a lot of trends kind of like weaving their way in and out of this story. It was one of the ones that I found it so interesting to write about. It's an important finding on its own, but it's also something you have to situate within the context of what's happening in the United States, right? And it asks us to work outside of silos and be really smart and really intervene and

It's yet another thing where I'm worried about the fractures in our healthcare. Because right now, they had an answer. They have resources for people in this position. They're like, you should go to the IDENTIFY study. They will fly you to NIH, cover all of your costs. But when the study goes away...

how many of these people are going to have to really fight with their insurance companies to get this covered? And if not, is it just going to be, you know, like if you have money to pay for it out of pocket, are you going to be able to get this care, but otherwise not? You know, so I really want to shine some light on this. I hope that people see the importance of A, responding to these findings with an appropriate workup and B, that they should be covered. Yeah, absolutely. Absolutely.

Thank you so much for taking the time to come on and tell us a little bit about your article. I definitely encourage our listeners to check it out in full on SIAM.com. Thanks. It's a pleasure to be here. That's all for today's episode. We'll be back with our usual news roundup on Monday.

Science Quickly is produced by me, Rachel Feltman, along with Fondam Wangi, Kelso Harper, Naima Marci, and Jeff Dalvisio. This episode was edited by Alex Zaghiara. Shana Posis and Aaron Shattuck fact-check our show. Our theme music was composed by Dominic Smith. Subscribe to Scientific American for more up-to-date and in-depth science news. For Scientific American, this is Rachel Feltman. Have a great weekend. ♪