Individualized PPV and why it matters
5 Questions with Dr. James Goldberg, FACOG, FACMG
1. What is Positive Predictive Value (PPV)?
Dr. Goldberg: Positive predictive value (PPV) is the likelihood that a positive result is associated with an affected pregnancy. PPV is based on test performance as well as prevalence of disorders.
For conditions like Down syndrome, for example, both maternal and gestational age impact the prevalence of the condition, and should be factored into the PPV calculation, thereby generating an individualized or patient-specific PPV.
Individualized PPV is not a new concept for OB/GYNs — it is equivalent to the patient-specific risk estimates obtained from traditional maternal serum aneuploidy screening.
Just like in serum screening, we take the sensitivity and specificity of the test (test accuracy), the patient’s prior risk based on maternal and gestational age, and test results to compute an adjusted post-test risk.
The clinical community has had some trouble transitioning this same concept to cell-free DNA. I believe this is because we were first focused on the high sensitivity and specificity of non-invasive prenatal screening (NIPS).
2. How is PPV different than the sensitivity and specificity of a test?
Dr. Goldberg: There are a lot of numbers associated with a screening test that let a provider know how well the test performs at a population level, including sensitivity, specificity, false positive rate, and test failure rate.
Some may confuse PPV with sensitivity and specificity. Sensitivity is the chance an affected pregnancy would have a positive result, and specificity is the chance an unaffected pregnancy would have a negative result — for most non-invasive prenatal aneuploidy screens these numbers are about 99%.
While this speaks to the performance of the test, those values play just one part in determining a patient’s actual risk. The other crucial piece is the patient’s prior risk. What was her chance of having an affected pregnancy before she walked in the door? For disorders like Down syndrome, this depends on her maternal and gestational age, which will be different for every patient.
For example, let’s compare two patients with the same positive NIPS result for Down syndrome:
A 25 y/o patient at 16 weeks has a PPV of 27.7%, giving her a 27.7% chance of having an affected pregnancy.
A 42 y/o patient at 12 weeks with the same result has a PPV of 91.6%, giving her a 91.6% chance of having an affected pregnancy.
3. Why is PPV important?
Dr. Goldberg: NIPS is a highly accurate test and has decreased the need for invasive testing like amniocentesis by more than 60%¹ in the last several years.
But, NIPS is not a diagnostic test — it is a screening test.
When there is a positive result, providers and their patients then have the choice to pursue diagnostic testing such as amniocentesis or CVS to confirm the result.
This is a critical point because some patients may hear “99% accurate” and believe there is a 99% chance their baby is affected. Then, they may opt to take action without confirmation when their risk is actually much lower. As you can see from the examples above there can be significant differences.
An individualized or patient specific PPV is the only number that reports the likelihood that a test result is correct for that particular patient. It specifies the individual patient’s risk after a positive test result — in the same way that traditional serum screening results have been reported for decades.
4. What do the medical societies say?
Dr. Goldberg: Most of the initial recommendations related to NIPS were commentaries on the accuracy and use of the test. Recently, both ACOG and SMFM² ³ have released recommendations citing individualized PPV as an important aspect of NIPS reporting that should be included for all patient results.
“Given the importance of these data in providing accurate and understandable information to patients regarding screening test results, ACOG and SMFM encourage all laboratories to report results with PPV and residual risk values.” ²
It is also cited in this commentary that no-calls should be treated as screen positives, and that genetic counseling should be offered to all patients, even those with negative results.
5. What is Counsyl doing to address this?
Dr. Goldberg: Last year at SMFM, when I had just joined Counsyl as the Chief Medical Officer, we began discussing the importance of PPV. I felt this was absolutely crucial; and I’m proud to say we now include individualized PPV and negative residual risk on all our Informed Pregnancy Screen reports.
Counsyl is the first lab to include individualized PPV on reports so there is no need to calculate it separately. The low test failure rate of the Informed Pregnancy Screen also means fewer redraws and less anxiety for patients. We want to ensure that clinicians have what they need to responsibly offer NIPS and guide next steps for a patient’s care.
There are a lot of NIPS options available but not all tests are created equal.
While the accuracy of most NIPS options are about the same, the experience for the patient and provider can be very different.
As screening becomes more broadly available, Counsyl is working to address the challenges integrating broader screening into an already busy practice with our customizable platform of additional services.
We partner with clinics to serve as an extension of the office with results delivery, on-demand genetic counseling for both positive and negative results, and transparent billing policies. This decreases patient anxiety and enables seamless integration into clinical practice so clinicians can stay focused on patient care — that makes all the difference.
Learn more about the Informed Pregnancy Screen here.
Recommended articles of interest:
- ACOG/SMFM Committee Opinion on Cell-free DNA screening (2015)
- SMFM statement: Cell-free DNA screening is not a simple blood test (2015)
- Editorial in Scientific American (2016)
- Disrupting Prenatal Testing — A Revolution is Underway (2015)