Genetic flukes

Prenatal test accidentally picks up cancer in 50% of those with wonky results

The prenatal test results flagged cancers when other screens and tests missed them.

Beth Mole – Dec 5, 2024 7:18 am | 65

In 2013, researchers reported an eye-opening case of a healthy pregnant woman with a puzzling prenatal test result. A routine genetic screen using cell-free DNA—a highly accurate blood test—suggested her fetus had an extra copy of chromosome 13 (Patau syndrome) and only one copy of chromosome 18. These results are devastating; both conditions can cause severe abnormalities. Those with Patau syndrome often only survive a few days or weeks after birth. But, when doctors looked at scans and did additional pregnancy testing, all they found was a healthy fetus developing normally. The woman carried on with her uncomplicated pregnancy and gave birth to a healthy baby.

The alarming genetic results may have been written off as a freak testing flub. But soon after giving birth, the otherwise healthy 37-year-old mother of two reported severe pelvic pain. Imaging revealed what looked like multiple bone tumors, and she was subsequently diagnosed with metastatic small cell carcinoma of vaginal origin. Tragically, she has since died.

Testing of one of her tumors found that the cancerous cells had an increased number of chromosome 13 relative to chromosome 18. Her prenatal test had picked up her deadly cancer.

Prenatal testing

Prenatal cell-free DNA (cfDNA) screening during pregnancy is a non-invasive blood test used to look for genetic variations early in pregnancy, particularly for chromosomal abnormalities like Patau syndrome. The test works because, normally, about 10 percent of the DNA circulating freely in a pregnant person's blood comes from the placenta. The other 90 percent or so comes from the pregnant person. Algorithms and reference samples are used to establish ratios that can estimate the chances of an abnormality in the pregnancy. But those ratios can be thrown off if there's a third source of DNA in the blood: cancer.

Tumors and other cancers can shed DNA into the bloodstream, distorting the ratios. Sometimes this can lead to alarming prenatal test results, like extra or missing chromosomes. Sometimes it can just lead to unreportable gobbledygook.

In the decade since, doctors have become more aware of such scenarios, and doctors have reported more cases of prenatal screens foreshadowing cancer diagnoses. However, there is still a dearth of data on the phenomenon and no standardized guidance on what clinicians should do when their patient gets a strange cfDNA result. Researchers don't even have solid data on how good cfDNA tests are at detecting cancer.

To address the lack of data, researchers at the National Institute of Health set up a study to look at outcomes after abnormal cfDNA tests—and specifically look for cancer in patients. Women were enrolled if they were pregnant or recently postpartum and had no known cancers. Another big requirement was that they had received some perplexing cfDNA result, such as an unreportable result or an abnormal result that didn't match a healthy fetus or secondary genetic test results, like the woman's case in 2013.

Fresh data

For the study, published Wednesday in the New England Journal of Medicine, NIH researchers set out to repeat the cfDNA tests and then screen the women for cancer using various methods. Those included family and clinical histories, blood tests, measurements of serum tumor markers, fecal occult blood tests, pap smears, physical exams, and—the big one—whole-body magnetic resonance imaging (MRI).

In all, 107 women were enrolled. Of those, 52 (48.6 percent) were found to have hidden cancers. Thirty-two participants had blood cancers (31 were lymphomas), while 20 had solid tumors, including breast, bile duct, colorectal, pancreatic, lung, kidney, bone, and adrenal gland cancers.

Of the 52 with cancer, 29 (55.8 percent) had no symptoms of their disease. Thirteen had cancer symptoms that were ascribed to pregnancy-related causes. For example, stomach pain from pancreatic cancer was diagnosed as reflux. Ten had symptoms that were either overlooked by the woman or were evaluated but deemed not concerning.

Of the 20 cases with solid tumors, most were in later stages (five were stage 2 or 3, and 13 were stage 4). But of the 20 cases overall, 13 were eligible for potentially curative treatments.

Then there were the other 55 women who had no detectable cancer. Of those, researchers concluded that 15 had simply gotten a rotten cfDNA result, which they based on repeat cfDNA testing and the cancer screens. Thirty were found to have noncancerous biological reasons for their weird cfDNA results, including fibroids, placental mosaicism, a stem cell abnormality, or a fetal finding. But, the remaining 10 had no clear explanation for their odd cfDNA results, stumping the researchers. These participants are now being followed for five years to assess their outcomes.

Takeaways

While the study is small and the findings should be considered preliminary, the results suggest that peculiar cfDNA tests are effective at signaling hidden cancers, and clinicians should follow up on them. The findings also point toward a pattern that could signal a higher risk that a weird cfDNA result is indicating cancer: Most of the women who were diagnosed with cancer had a cfDNA result that indicated both chromosomal gains and losses—just like the 2013 case.

The NIH researchers were able to successfully repeat cfDNA testing for 105 of the women. Possible results included normal findings, both gains and losses of chromosomes, only losses, only gains, maternal variants, or borderline results. Of the 105 repeated tests, 49 had both gains and losses of chromosomes, and 47 were in women found to also have cancer. That is, 90 percent of the participants with cancer had both gains and losses in their cfDNA results. The other two who had both gains and losses on their cfDNA test but no cancer are in the unexplained group that's being followed for five years. "There is ongoing concern that an existing cancer was not detected or that cancer might develop in the other two participants," the NIH researchers write.

Another takeaway from the study is that, by far, the most effective way to detect the cancers flagged by the cfDNA tests was the whole-body MRI. The blood tests and other screens were "of limited use," the researchers found. The MRIs, on the other hand, only missed one cancer case and had a low rate of false positives (only six out of 52). But, for now, obstetricians don't often order whole-body MRIs and insurance companies don't often cover them.

In an accompanying editorial, Neeta Vora, one of the authors of the 2013 case report, called for more research, clinician awareness, and clinical guidance.

"Although we have come a long way since 2013, when we published the case report of our patient with a neuroendocrine tumor and abnormal cfDNA results, we still have a long way to go to improve provider education regarding the possibility of identifying maternal cancer through cfDNA screening, as well as to improve laboratory standardization, reporting of suspicious findings, and access to imaging methods such as whole-body MRI," Vora wrote.

This post has been updated to correct an error in reporting the false positive rate.

Beth Mole Senior Health Reporter

Beth is Ars Technica’s Senior Health Reporter. Beth has a Ph.D. in microbiology from the University of North Carolina at Chapel Hill and attended the Science Communication program at the University of California, Santa Cruz. She specializes in covering infectious diseases, public health, and microbes.

65 Comments