ACMG Posters: studies do not support professional recommendations

I attended and presented a poster at this year’s American College of Medical Genetics & Genomics conference, held in Nashville, Tennessee. Here’s what some of the other posters reported on the research their authors had conducted regarding prenatal testing.

The earlier the testing, the more likely to be accepted

Poster 79: reviewed the uptake of prenatal genetic testing after non-directive genetic counseling. Results: 1/3 accepted further testing after a 2d trimester screen result; 2/3 accepted first trimester screening; and, almost 3/5 accepted NIPS. Results show the earlier and less invasive the testing, the more likely patients are to accept it.

In urban setting, NIPS has 75% false positive rate

Poster 183: NIPS in an urban setting. Check out the conclusion:

In our small series, only one abnormal NIPS result was a true positive. Three out of the four abnormal NIPS cases were false-positives [] which was higher than anticipated. … NIPS is a powerful tool in our new genetic screening armamentarium. However, incorporating it into our practice has been challenging.

So, in this practice’s experience, if you received a “positive” NIPS (which there is no such thing), you actually had a 75% chance for a false positive.

The earlier the testing, the less reliable it is

Poster 189: Sequenom examines fetal fractions and its test. Without sufficient fetal fractions, Sequenom’s test is either inconclusive or incorrect. Sequenom concluded:

Overall, these results indicate that the likelihood of a sample containing a fetal fraction sufficient to accurately classify chromosomal aneuploidies increases with gestational age. However, the gestational age range during which the vast majority of our samples are collected shows a relatively low rate of increase in fetal fraction.

When Sequenom launched, they highlighted how their test is trimester independent, meaning that it can be done at any time after the 10th week, all the way up to the 39th week. But it specifically markets its test highlighting it being offered earlier than all other screening tests. Yet, by its own study, its test would be more reliable if done later in a pregnancy, when fetal fractions are higher.

Medical residents rate their training on genomics: “inadequate.”

Poster 393: medical residents rate their training on genomics:

Ninety-three percent of students responded correctly to at least 80% of GKS items indicating a basic understanding of genomics. The majority (89%) rated genomics as an important part of training and over 75% agreed that this testing will become increasingly important and will benefit their patients. While acknowledging the importance of genomics, most students judged their medical school training as inadequate in preparing them to use it in patient care. Slightly more than one-third rated their pre-clinical training as “adequate” and fewer than 1 in 7 agreed with this rating for their clinical years.

I expect that if you spoke with them individually they would rate their training as “what training?”

More posters to follow in subsequent posts. But connecting the dots of these studies: NIPS is more accurate later in pregnancy, when it is less likely to be accepted than if offered earlier in the pregnancy. NIPS can have a high false positive rate in certain populations. And, practitioners rate their own training on genetic testing as inadequate.

But what do the guidelines say?

Because NIPS has a high accuracy in select clinical studies, let’s offer it to high risk women, regardless of its true accuracy in the commercial setting, in certain populations, and regardless of whether medical professionals are adequately trained to counsel and offer the testing to their patients.