AFP / FreeBeta Screening

Second Trimester Screening

Introduction
In the United States, it is standard of care to recommend all mothers 35 years of age and older have diagnostic amniocentesis for chromosome abnormalities. Amniocentesis diagnoses 99.99% of all of the hundreds of types of chromosome abnormalities, whereas prenatal mother's blood screening is designed to produce a patient-specific risk estimate for only two of these abnormalities.

Down syndrome and trisomy 18/13 maternal screening protocols are designed to determine a risk for low risk women less than 35 years of age. Each patient's age and level of biochemical markers in the patient's blood sample are then combined with other individual health factors such as weight, family history, ethnic background, etc. and compared within normal range results and increased risk results to determine an individual Down syndrome and trisomy 18/13 risk. The goal is to obtain the best balance between increasing detection efficiency and lowering the false positive rate.

First trimester screening involves biochemical analysis of freeBeta hCG and PAPP-A (pregnancy-associated plasma protein-A), and ultrasound measurement of nuchal translucency (NT). Using this combination, a 91% detection rate for Down syndrome and Trisomy 18/13 is achieved, with an initial positive rate of less than 5%.

Not all patients are able to take advantage of the 91% detection rates of first trimester screening between 11w0d and 13w6d. For those who present between 15w0d and 21w6d gestation, second trimester AFP/freeBeta screening using Bayesian and multivariate discriminate analysis should be offered as a significant improvement (31%) over the triple screen. The tables below compare the performances of freeBeta/AFP screen with the less effective triple screen.

Detection Rate, IPR, and Yield
The two tables below document the higher accuracy of AFP/free Beta screening over the less effective triple screen. For all disorders being screened, detection is significantly higher.

All things being equal, one would think that increased detection would result in an increased initial positive rate (the number of screened women under 35 years of age with a positive screening test after ultrasound confirmation of dates and fetal number). Instead of increasing, however, the initial positive rate (IPR) from second trimester AFP/freeBeta screening is less than half that of the triple screen (5.6% versus 12%). This statistic translates into fewer anxious patients, reduced medical costs, and less coordination effort by the physician's office, and fewer diagnostic amniocenteses.

A function of IPR and detection rate is yield; that is, the number of amniocenteses necessary to identify one affected fetus. A significantly lower number of amniocenteses performed after second trimester screening documents its superiority as a screening test.

Another major advantage of AFP/freeBeta screening over the triple screen is the fact that it can be offered as early as 15 weeks 0 days post LMP without loss of detection power, instead of 16 weeks recommended by the triple screen. Earlier screening means earlier detection and fewer women facing the prospect of a late second trimester termination.

Table 1: Performance of the AFP/freeBeta Screen between 13w4d and 22w3d gestation.

	IPR*	Detection Rate	Yield**
ONTD***	2.7%	98%	1 in 25
Down syndrome	2.8%	80%	1 in 25
Trisomy 18/13	0.3%	70%	1 in 6
Total =	5.6%

Table 2: Performance of triple screen (MSAFP/hCG/estriol) between 16 and 20 weeks gestation.

	IPR*	Detection Rate	Yield**
ONTD***	5.0%	85%	1 in 42
Down syndrome	6.5%	60%	1 in 80
Trisomy 18/13	0.5%	60%	1 in 16
Total =	12%

* Initial Positive Rate, after ultrasound confirmation of gestational age
** Number of amniocenteses necessary to identify one affected fetus
*** ONTD = open spine and skull defects

NOTE: Detection rates for the triple screen decrease significantly at gestations before 16 and after 20 weeks.

Performance Comparison of AFP/freeBeta versus Triple Screen

Open Neural Tube Defects Screening

The AFP detection rate for ONTDs is dramatically improved over that of the triple screen (98% versus 80-85%) with a decrease in the IPR (2.7% versus 4-5%). A look at yield shows that 42 amniocenteses must be performed in order to identify a fetus with an ONTD by triple screening, versus 25 amniocentesis by the standards of AFP/freeBeta screening. This is accomplished by:

Using true patient-specific risk cut-offs, as opposed to a Mom (multiple of the median), to identify a positive screen. Generating the patient-specific risk from Bayesian and linear multivariate discriminate analysis. Incorporating additional information, such as family history and geographical demographics, into the risk calculation to better define each subpopulation being screened. Most programs ask about family history, but the majority use this information by simply not calculating a patient risk when there is a positive family history for ONTD.

Defining the population being studied with greater precision and using sophisticated statistical analysis to arrive at a refined patient-specific risk has been advocated by national regulatory bodies for years:

American Academy of Pediatrics, Alpha-fetoprotein Screening, Pediatrics 80: 444-445, 1987. "Laboratory measurements are obtained in quantitative units (nanograms or international units per millileter) and are frequently reported as multiples of the median. Such expressions do not address the degree of risk to the individual patient. As information is collected that permits an estimate of risk in subpopulations comparable to the woman being screened, laboratories should include in the laboratory report the risk to the woman on the basis of her test result."

American Society of Human Genetics, Policy Statement for Maternal Serum Alpha-fetoprotein Screening Programs and Quality Control for Laboratories Performing Maternal Serum and Amniotic Fluid Alpha-fetoprotein Assays, Am J Hum Genet 40: 75-82, 1987. "As data are collected to permit an estimate of risk in subpopulations whose members are comparable to the woman being screened, laboratories should include in the laboratory report information about the risk to the woman on the basis of her test result."

Despite the clear directive by regulatory agencies, most laboratories still make only a few adjustments to the data (weight, race, diabetic status) and rely on the multiple of the median to generate a general risk figure. GeneCare also considers: geographic location, anti-seizure medications, incremental high and low weight adjustments, multiple fetuses, and family history of ONTDs and chromosome abnormalities. Once the patient is classified into the appropriate subpopulation, her laboratory data is then compared to that of women with similar backgrounds who have had normal and affected offspring. Comparison data is derived from over 1,000,000 prospective samples with known outcomes. Higher order statistical analysis reduces the overlap between the distributions of normal and affected pregnancies. It is this process, in conjunction with better analytes, that produces AFP/freeBeta improved detection rates and lower IPR.

Down Syndrome Screening

Because of the superior screening capability of freeBeta compared to intact hCG, AFP/freeBeta detects 80% of Down syndrome fetuses compared to 55-60% by the triple screen and it does so at less than half the initial positive rate (2.8% versus 6.5%). This is a remarkable improvement over the triple screen and it is accomplished with only two markers, instead of three, thereby reducing the cost of the test.

***Adjusted Down Syndrome Screening in Hispanics

A recent paper in Prenatal Diagnosis documents a higher than expected frequency of Down syndrome among Hispanic liveborn children. Overall, the paper confirmed a 20% increase of Down syndrome in Hispanics compared to non-Hispanics. This finding has implications for prenatal screening and diagnosis. The adjustment for maternal serum screening programs is straightforward: before screening, the maternal-age rate for Down syndrome should be increased by 20% in Hispanic women. This adjustment has been made in GeneCare's screening program and will ensure the most accurate risk assessment across all populations.

The increased rate of Down syndrome in Hispanics also would justify a recommendation for amniocentesis/CVS when the mother is 34 years old rather than 35. Though totally appropriate, this may be difficult to implement because medical insurance guidelines have fixed on a certain age, rather than the patient's risk.

Hook, E.B., Carothers, A.D., and Hecht, C.A. 1999. Elevated maternal age-specific rates of Down syndrome liveborn offspring of women of Mexican and Central American origin in California. Prenat. Diag.; 19: 245-251.

Trisomy 18/13 Screening

AFP/freeBeta identifies more cases of Trisomy 18/13 than the triple screen (70% versus 60%) with a decrease in the initial positive rate (0.3% versus 0.5%). The more effective AFP/freeBeta screening requires that only six amniocenteses be performed to identify one case of Trisomy 18/13, compared to 16 amniocenteses required by the triple screen.

Efficiency Comparison

In light of the above data, it is clear that AFP/freeBeta is a far more effective second trimester screening protocol than the triple screen. It detects substantially more affected fetuses, while requiring half as many amniocenteses. Patient anxiety and medical costs, both to the patient and to the health industry as a whole, are dramatically reduced.

In Europe, the number of labs offering triple screen has decreased to a small minority because the world literature shows AFP/freeBeta is significantly more effective, less expensive, and is not limitied to 16-20 weeks gestation (Toward the optimal screening protocol for Down syndrome screening. AM J OBSTET GYNECOL. 1996; 174(5): 1668-1669. Macri JN, Spencer K). Why do labs in the United States still offer the less effective Triple Screen? The reasons for this are several:

Lack of a commercially available kit for freeBeta in the U.S.
Additional software requirements and outcome data for data analysis
General inertia in changing protocols
Reduced profit margins for participating laboratories

Since the triple screen does not provide equally effective screening, there are no documented reasons for physicians not to switch to AFP/freeBeta screening. Physicians throughout the U.S. and around the world provide AFP/freeBeta screening, so this option is available to all patients.

Second Trimester Markers

Alpha-fetoprotein

Alpha-fetoprotein (AFP) is the best marker known for detection of open neural tube defects (ONTDs). It tends to be elevated in the blood of women carrying fetuses with ONTDs. Conversely, AFP tends to be lower in the blood of women carrying fetuses with Down syndrome and Trisomy 18/13.

FreeBeta hCG

FreeBeta is the only biochemical marker that can be used in both the first and second trimesters. In both trimesters, freeBeta hCG is elevated in the blood of women carrying fetuses with Down syndrome (second trimester median MoM = 2.69) and reduced in the blood of women carrying fetuses with Trisomy 18/13 (second trimester median Mom = 0.20). In the second trimester, the comparable median Moms for intact hCG are 2.03 and 0.36, causing greater overlap between affected and unaffected fetuses.

Estriol

Estriol was originally added to AFP and hCG to enhance detection of Down syndrome. Estriol concentrations do not influence risk calculations for ONTDs. Initial studies using AFP concentrations generated from older less refined assays suggested that estriol improved Down syndrome detection by about 1-2%, raising detection from 55% to 57%. However, studies using AFP concentrations from today's more precise assays and analysis have not been able to demonstrate a benefit from estriol. Most researchers today agree that estriol increases the initial positive rate of a screening program without improving the detection rate of Down syndrome or Trisomy 18/13. Most laboratories in the U.S. have deleted estriol to increase detection, reduce unnecessary positives and cost.

An article in The American Journal of Obstetrics and Gynecology (Evans, et. al. 172: 1138-49, 1995) addresses the effectiveness of double marker screening versus triple marker screening for Down syndrome. The article employing a large data set concludes that "...double screening is equally as effective as triple screening so that the expense of estriol is unnecessary""...eliminating estriol should achieve higher specificity for the same sensitivity and save, conservatively, about $900,000 in this series (25,504 patients)". This finding confirms data, published by other investigators, which has demonstrated that unconjugated estriol has no useful role in Down syndrome or trisomy 18/13 screening.

There is a misconception that the use of three markers provides better detection than two markers. On the contrary, the combination of freeBeta and AFP achieves a 21-24% higher Down syndrome detection rate at a lower false positive rate than the "triple test". In summary, the scientific literature has documented the AFP/freeBeta protocol as the most sensitive and specific Down syndrome, Trisomy 18/13, and ONTD screening protocol.

Who Should be Offered 2nd Trimester Screening?

Maternal blood screening is designed for pregnant women who are at low risk for the disorders being screened. Therefore, the screening programs described here are for women who are under 35 years of age at the time of delivery and who do not have a positive history of an ONTD or chromosome abnormality. High-risk patients should be referred for genetic counseling, ultrasound, and amniocentesis. High-risk patients are those who are 35 years or older at delivery, have a close relative with an ONTD (self, spouse, child, parent or sibling) or have a previous child with a chromosome abnormality.

1st and 2nd Trimester blood screening should not be provided to high-risk patients (women over 34 years of age at EDD or with a positive medical history) as a substitute for amniocentesis chromosomal analysis because:

Screening provides a risk for two out of the hundreds of types of chromosome abnormalities, not a diagnosis, whereas amniocentesis can diagnose 99.99% of the hundreds of types of fetal chromosome abnormalities.
Screening will miss about 1 out of 10 (10%) Down syndrome pregnancies for women under 34 years of age. Screening will miss the majority of other chromosomal abnormalities.
A within normal screening result only suggests a lower risk. Only a diagnostic amniocentesis can determine whether a fetus has a chromosome abnormality.
Because the two screening risks are limited to a few chromosome abnormalities, mother's age and medical history risks more accurately determine each high-risk patient's risk of having a baby with a chromosome abnormality.
Screening will falsely reassure many women who are carrying an affected fetus.

GeneCare recognizes that some high-risk women will elect to have maternal blood screening instead of diagnostic amniocentesis. Although screening will provide a risk assessment for two disorders, these patients should be counseled that screening will not adequately address the total chromosomal risk they face and may give them a false sense of security. The fetal chromosome risk of a 35 year old woman includes 1. Down syndrome (only 40% of the total screening risk) and 2. hundreds of other chromosome abnormalities (60% of the total risk). A within normal range Down syndrome report will only reduce a percentage of the Down syndrome risk, which leaves most of the fetal chromosome abnormality risk not addressed. Genetic counseling is strongly suggested for all high risk patients to help them understand why screening should not be considered an equal substitute for diagnostic amniocentesis chromosome analysis.

Second trimester biochemical screening may be offered to low risk women who are carrying twins or triplets, although screening is not as sensitive as in singleton pregnancies.

Screening Methods
Patients who elect to have the screening test after nondirective genetic counseling should be scheduled for blood draw between 13w4d and 22w3d. AFP/Free beta testing may ideally be scheduled at 14w with known dates and 15 weeks with uncertain dates (based on our published data).

In order to achieve the detection rates discussed in this text, blood samples must be fixed onto dried blood paper which is attached to the requisition as soon as possible after leaving the patient's body. Blood can be obtained using a sterile lancet. Alternatively, a sample of blood may be drawn into a red top tube (no additives) and then immediately spotted onto the paper after the draw. GeneCare uses a simple device (DIFF-SAFE) which is inserted into the blood tube to dispense 5 drops of blood onto the paper.

Screening and Diagnostic Procedures

Screening versus Diagnosis

In order to distinguish between affected and unaffected fetuses, one or more tests are needed to tell them apart. If, after applying a test, no overlap in the distribution of results of the affected and unaffected populations is seen, the method is diagnostic. This is the case with a chromosome test for Down syndrome: discrimination between affected and unaffected should be greater than 99%.

Screening tests differ from diagnostic tests in that the distributions of results (as from a biochemical assay) from affected and unaffected fetuses overlap. The more the distributions overlap, the poorer the discrimination between the two groups. Screening tests are limited to detecting certain birth defects and do not guarantee a perfect baby. A result within the normative range does not necessarily mean there are no abnormalities present, and an increased risk result does not guarantee that abnormalities are present. An increased risk screen places a woman at increased risk for abnormalities and warrants diagnostic testing by amniocentesis.

Screening tests should be reported as "within the normative range" or "increased risk" because screening provides a patient specific risk, not a diagnosis. The terms "normal" or "negative" and "abnormal" or "positive" suggest the screening result is a diagnosis. These terms may over reassure some patients or cause undue concern to others.

Maternal Serum Screening Improvements

Anything that can be done to reduce the overlap between affected and unaffected fetuses will enhance detection and improve the quality of the screening test. GeneCare uses several innovative approaches to reduce the overlap between distributions and thus enhance detection of maternal serum screening:

Certified filter paper (903) specimen collection fixes the patient's chemicals, eliminating sample degradation and guaranteeing that the patient's sample is representative of her true biological state.
Patented assays with the highest documented specificity and sensitivity accurately measure the freeBeta and AFP concentrations on dried filter paper.
Unlike large general laboratories, no default data is ever used. If patient information (such as weight, family history, etc.) is not provided, it is obtained from the referring physician prior to analysis.
Unlike some other laboratories, positive family history is considered in assessing the patient's prior risk for Down syndrome, Trisomy 18/13, and ONTDs.
Once the patient's prior risk is determined by multivariate discriminate analysis of her clinical information and her conditional risk is determined by discriminate analysis of her biochemical and sonographic information, Bayesian analysis integrates these two risks to generate a true patient-specific risk for Down syndrome and Trisomy 18/13.
The detection rate and initial positive rate are increased by obtaining pregnancy outcomes. Physicians and patients should be aware that only labs with outcome data can quote detection and initial positive rates.

Summary of Second Trimester Screening Parameters

freeBeta detects 80% of Down syndrome, 70% of Trisomy 18/13, some other chromosomal abnormalities and some multiple pregnancies.
AFP detects 98% of Open Neural Tube Defects.
No screening protocol identifies all disorders being screened, nor does it detect other physical or mental defects not being screened.
Multiple gestation screening is not as sensitive as singleton screening with biochemical analysis.
Report results rely on the accuracy of the clinical information submitted.
Patient-specific risks are determined by sophisticated statistical distributions (not simply MoMs) and can be accurately recalculated as needed only by GeneCare.

Targeted Fetal Ultrasound

A targeted ultrasound exam is recommended for patients with an increased risk screen to evaluate gestational age and to assess multiple pregnancy and fetal anomalies. GeneCare offers the highest resolution fetal ultrasounds by collaborating with other experienced specialists. Ultrasound can detect most but not all ONTDs, some (but not most) cases of Down syndrome and few other chromosome abnormalities. Many journal articles have documented that the fetal chromosome abnormality risk for patients with a significantly elevated ONTD risk after a normal targeted fetal ultrasound (0.8% or greater) will still exceed the possible risk of miscarriage associated with amniocentesis (less than 1 in 300 chances). Therefore, patients with an increased risk for Down syndrome, Trisomy 18/13, or ONTDs should be offered amniocentesis.

Amniocentesis in the Second Trimester

Women with increased risk for a chromosomal abnormality or an ONTD because of second trimester screening or ultrasonographic findings are considered candidates for amniocentesis. Amniocentesis has been shown to be very safe when performed by experienced physicians with continuous ultrasound guidance.

In addition to identifying all detectable chromosomal abnormalities, amniocentesis is diagnostic for 99% of open neural tube defects (ONTDs). Amniotic fluid can be assayed for amniotic fluid AFP (AFAFP) and acetylcholinesterase (AChE), tests which identify all detectable open spine and skull defects between 8w4d and 22w3d.

Optimal Screening Protocol
Maximum screening efficiency for Down syndrome, Trisomy 18/13 and ONTDs is achieved through first trimester biochemical (PAPP-A/freeBeta) and sonographic (nuchal translucency [NT]) screening, followed by AFP/freeBeta screening in the second trimester. This approach will achieve 91% detection of the two chromosomal abnormalities and 98% detection of ONTDs. Alternative screening protocols (first trimester biochemical screening alone, NT measurement alone or second trimester biochemical screening) cannot achieve these detection levels. The first trimester blood sample and NT measurement do not have to be obtained concurrently. However, GeneCare must know upon receipt of the biochemical specimen that an NT measurement is pending or it will generate a report based on biochemistry alone.

Timing of Tests

As with any screening test, there is a time range to offer the first and second trimester screening tests. Gestational age is best estimated by ultrasound when available, followed by the date of onset of the last menstrual period and finally by physical exam. The times to perform the various tests are indicated below.

FreeBeta/PAPP-A biochemical screen			9w - 13w 6d
Nuchal translucency measurement			11w - 13w 6d
MSAFP/freeBeta biochemical screen			13w 4d - 22w 3d

Second Trimester Screening References - Available upon request.

Information provided on this site is for information purposes ONLY. This should not be used as a substitute for professional medical advice, treatment or diagnosis.