April 4, 2006 Presentation: Prenatal Screening for Down Syndrome: New Developments For Biochemical And Ultrasound Markers
Welcome to AACC’s Expert Access Live Online program
This month's expert is Nathalie Lepage, PhD. View the presentation and direct your questions to our online expert. AACC would like to thank Bayer HealthCare Diagnostics for making this program possible.
What is the difference between a screening test and a diagnostic test?
Nathalie Lepage, PhD: A screening test is used to identify patients at sufficient risk for having a disease to benefit from further investigations. It is usually performed before the onset of clinical manifestations. The screening performance of a test will depend on the prevalence of the disease and the chosen cut-off to define individuals at increased risk. Typically, screening tests are associated with sensitivity <100% and specificity <100%. A diagnostic test is used to confirm the presence of a disease in a patient. Diagnostic tests usually have 100% sensitivity and 100% specificity, they could require invasive specimen collection.
What is the biochemical reason for Trisomy? Is it always a female fault, such as the age of the mother (above 35 years of age)? or can the sperm also have something to do with the Trisomy? Why young women (in the early 20's have occassionally children with Trisomy and why some mom's in the 50's give birth to normal kids? My half brother Daniel was born when mom was 56.
George E. Bernett, PhD; Laguna Niguel, California
Nathalie Lepage, PhD: Down syndrome is a chromosomal abnormality. In most instances, it results from the presence of an extra chromosome 21, as a free chromosome or as part of a fusion chromosome; it can also be due to presence of triplicate of smaller length of distal part of chromosome 21. All cells in the patients would be identical. Down syndrome can also be due to mosaicism: some cells with trisomy 21, other cells being diploid. The frequency of Down syndrome (47 chromosomes, +21) is a function of maternal age: it increases with maternal age. Trisomy 21 is due to the chromosomal non-disjunction in maternal eggs. The risk of recurrence of non-disjunction is also increased in mothers of previous children with Down syndrome. It is not well understood why there is a strong effect of maternal age on incidence of Down syndrome. On the other hand, the age of the father has only a very limited effect on incidence of Down syndrome. There is no well proven impact of paternal age on the frequency of aneuploidy sperm.
Thank you for an excellent review of the MSS testing options and the main issues to be addressed with your database. One subject that you touch on only briefly is the reproducibility of the test methods. This can be a major source of unknown error in these risk assessments - lab bias relative to fixed limits would appear as natural patient variability. What procedures can be used to assess various test methods' performances over time and in different laboratories?
Traverse City, Michigan
Nathalie Lepage, PhD: Ongoing quality assurance program needs to be implemented in order to ensure accurate laboratory results. Internal quality controls (composed of materials provided by the vendor, third party materials and in-house pools of patient samples) should be utilized daily. External quality assessment programs (organized surveys, sharing of patient samples among laboratories with same or different methodologies, etc.) should also be an integral part of the laboratory. There is a need to assess the detection limits and the linearity of the various methodologies, and to evaluate the possibility of diluting specimens with high results. Patient correlation upon changing lot numbers is required, with patient samples with “extreme” levels included in the correlation. Other procedures that are necessary on a frequent basis are the monitoring of MoMs for each parameter, of initial positive rates for each screening algorithm offered in your laboratory, of detection rates (less frequent and with a delay in monitoring, since we need to wait for pregnancy outcomes). For each parameter, benchmarks need to be established and corrective actions would need to happen to maintain adequate benchmarks.
I work in a private laboratory and process two-test and three-test. I'd like to know if I rely on PRISCA 4.0 software and if INIBIN is important for this elaboration. Excuse me for my English! Sincerely Domenico Dentamare member AACC
ITALY - NAPLES
Nathalie Lepage, PhD: There are various software from several vendors available for the interpretation of Down syndrome risks in the context of prenatal screening. I cannot comment on the best software, since it will depend on your specific needs. One typical process for selection of a vendor is the following. You should evaluate your needs, such as screening protocols offered by your center, markers included (you are mentioning incorporating inhibin A as one biochemical marker), need for correction factors for specific ethnicities, monitoring of medians and MoMs, language for data entry, others. You issue a “request for proposal” with your stated needs. You establish a shortlist of vendors, and contact users for their comments on advantages and disadvantages of the software. You can proceed with evaluation, then finally select one vendor.
The identified diseases that impact the Down¡¯s screening risk analysis include diabetes status, smoking, hereditary disease, hepatitis, anemia etc.. In the pre-natal screening approach, diabetes factor can be used to correct the risk analysis result. How to deal with the rest of the mentioned diseases above?
Nathalie Lepage, PhD: As you mentioned, there are few publications that evaluated the impact of some diseases on the blood levels of biochemical markers. However, there have been no systematic prospective studies that looked at the need for correction factors for selected diseases. I am not aware that any commercial software offers the possibility to correct for presence of maternal diseases on the risk for Down syndrome. Please note that the impact of IDDM (type 1 diabetes) is mostly important for the risk assessment of open neural tube defects (open spina bifida), and not so important for the risk assessment of Down syndrome.
Is there any new screening standard set up for singlet & multiple pregnancies?
Nathalie Lepage, PhD: Standard of care is difficult to define, since it is always adjusting to new findings from research publications. In few jurisdictions, the standard of care has been defined based on screening performance, rather than on usage of specific markers. As an example, the UK National Screening committee has a very informative website. On it, they defined the expected benchmark to be achieved by UK screening laboratories, by April 2007 for singleton pregnancies: detection rate of 75% (or more), initial positive rate of 3% (or less). They recommend screening algorithms but will not make them mandatory to the testing laboratories. Most other jurisdictions do not have specific screening performance and the decision on the selection of screening algorithms and markers can depent on various stakeholders (in isolation or after agreement on protocols) such as the laboratory director(s), the Ministry of Health, the health care providers (clinical geneticists, family care providers, others), etc. For multiple pregnancies, the standard of care is not well defined, since the number of such pregnancies is lower than singleton pregnancies, and the publications related to these pregnancies did not reach consensus among the prenatal screening community and among stakeholders.
Are there any factors supposed to be considered for second-time pregnancy, e.g. miscarriage, abortion etc. occurred in the first pregnancy?
Nathalie Lepage, PhD: The risk assessment of Down syndrome is based on biochemical markers and ultrasound markers. Second-time pregnancy screening results will depend on risk assessment of first-time pregnancy. If the first pregnancy was screened as false-positive or if the first pregnancy was affected with aneuploidy, there is a greater chance of having an initial-positive result in the subsequent pregnancy. The most likely explanation for this finding is that serum marker levels are related across pregnancies. Correction factors have been proposed but such factors have not been implemented in routine screening laboratories. The counselling of initial-positive results in second-time pregnancies needs to focus on recurrent false-positive results. I am not aware of correction factors for other conditions in the first pregnancy, such as miscarriage.
For preparation of pre-natal screening, how to improve the screening detection rate besides information consulling provide to patients and well implemented the QC procedures within the lab?
Nathalie Lepage, PhD: Please see answers above for description of proposed quality assurance program.
For quad screening, how to relate the Inhibin A, run in the 1st trimester, to the triple markers run during the 2nd trimester?
Nathalie Lepage, PhD: Publications from FASTER and SURUSS studies have measured inhibin A (DIA) in the second trimester (>15 weeks). I am aware of less than 10 publications where authors have proposed measuring inhibin A in the first trimester of pregnancy. Approx. 300 Down syndrome cases and approx. 2000 controls (only) have been used for the evaluation of the effectiveness of this marker <14 weeks. The data is still insufficient to conclude on the time period when inhibin A should be measured to achieve the best screening performance. Until then, in my center, we are measuring DIA after 15 weeks gestational age (as per FASTER and SURUSS results).
Are there any standards for the screening results tracking, e.g. post-maternal?
Nathalie Lepage, PhD: I am unsure what your question is. I would refer you to above answers on quality assurance program for on-going monitoring of screening performance.
Can NT of the fetus be used as one of the screening references?
Nathalie Lepage, PhD: For singleton pregnancies, NT (an ultrasound marker) can be included in various screening algorithms for the risk assessment of Down syndrome. They are the first trimester screening, the integrated screening, the sequential screening, and the contingency screening. In fact, only the second trimester screenings (triple or quad marker protocols) do not use NT as one marker for risk calculation. For twin pregnancies, NT can also be used for risk assessment of Down syndrome, combined with maternal age only, or with Biochemical markers to provide pseudo-risk assessment. No consensus has been reached in the screening community on the best screening strategy for Down syndrome in twin and multiple pregnancies.
Could you please kindly provide any related references and/or pre-natal screening software that can be used as methods or criteria in the pre-natal screening programs?
Nathalie Lepage, PhD: I will not provide a list of possible software vendors. The selection should be based on your local needs. However, you could search internet and you will find potential vendors. I will refer you to answer above for proposed steps for the selection and evaluation of software.
Thank you for the informative presentation on Down´s syndrome screening. There is one question concerning slide #16: here you mention the repeated measures screening with the biochemical markers PAPP-A and uE3 in first and second trimester. Is there any data published, which shows the performance of the particular markers in said semesters. To my knowledge PAPP-A is not discriminatory in second trimester... therefore I was wondering whether this works. Thanks in advance, kind regards Verena Hofmann
Nathalie Lepage, PhD: The reference is Wright DE and Bradbury I. BJOG: 112: 80-83, 2005. The concept of repeated measures is new, as you see by the year of publication (2005). This new concept has opened a field of research for prenatal screening. More publications are expected in the near future to confirm these preliminary findings.
Do you think the first trimester screening for Down syndrome will replace the current second trimester?
Nathalie Lepage, PhD: This answer is partly based on personal experience. In my center, we proceeded to implementation of integrated screening (first and second trimester screening) with massive number of education sessions to health care providers. The uptake of prenatal screening increased dramatically following these sessions. Also, providers appreciated the fact that screening performance was improved compared to “traditional” triple marker second trimester screening. However, second trimester screening is still requested for approx. 30% of pregnancies, most likely reflecting that there are many patients that still wait to the second trimester of pregnancy to confirm pregnancy and to seek prenatal care. The answer to your question is no. First trimester screening is a new option, but it will not completely replace second trimester screening. The impact of multiple screening options and strategies has not been fully evaluated. What will be the screening performances (sensitivity, specificity, positive predictive values) of second trimester screening with the implementation and availability of earlier screening options? These issues will need to be addressed in order for adequate counselling (pre-test and post-test) be offered to pregnant patients.
Dr. Lepage, Previously, diagnostic testing (karyotype) was offered to women who would be >35 years at the time of birth yet now it seems that all women, regardless of age, are offered biochemical screening. This seems illogical because regardless of what risk is used as the cutoff for a positive screening result, it becomes increasingly more difficult to modify an age-based risk to be less than the cutoff risk as maternal age increases. Would you please comment on this?
Nathalie Lepage, PhD: Based on experience with prenatal screening, the patients that have a result as “screen positive for Down syndrome” have patterns of biochemical and ultrasound markers consistent with Down syndrome pregnancies (I refer to one of my slide for MoM levels). The ages of these screen positive pregnant patients vary from 16 to 40 and more years. Therefore, the screening algorithms not only take into account the a priori risk based on maternal age, but also the MoM pattern of the tested markers. The final risk is almost never identical to the a priori risk (based on maternal age only) and could be increased or decreased Vs a priori risk. I see great benefit in offering prenatal screening to all pregnant ladies.
Dr. Lepage, What effect does assay imprecision have on DS detection rates? Is there a threshold CV beyond which an assay is unsuitable for use in maternal serum screening progams?
Chapel Hill, NC
Nathalie Lepage, PhD: Since detection rates are difficult to monitor on a regular basis, the parameter that is usually monitored in the testing laboratory, on a monthly basis is the initial positive rate (IPR). Assay imprecision can have great impact on the IPR. It is difficult to define a specific threshold CV, since the marker assays (PAPP-A, free beta hCG, total hCG, unconjugated estriol, AFP, inhibin A) do not perform identically. However, during the evaluation phase of a methodology, an assay or an instrument, the target CV from the manufacturer should be checked and confirmed in your laboratory setting. Imprecision is also monitored with the peer-user CV calculated in the external proficiency programs for the various markers. Patient correlation between reagent lots is an important factor and each laboratory should establish benchmarks above which new lots could be rejected for routine implementation, and alternate lots should be requested for evaluation. Therefore, several analytical conditions need to be met in order for an assay to be implemented and maintained as part of the prenatal screening program. It is always possible that methodologies that are not performing as per pre-determined laboratory expectations will be or become unsuitable and will need to be converted to other methodologies or vendors.
Dr. Lepage, What is your opinion regarding the fate of 2nd trimester screening tests? As 1st trimester screening becomes more widespread, it's logical to assume that the prevalence of DS in the 2nd trimester will decrease. That decrease in prevalence will impact the predictive value of 2nd trimester tests and could lead to lower detection rates even for the integrated test.
St. Louis, MO
Nathalie Lepage, PhD: Please see answer above.
If the ultrasound suggests a normal pregnancy, what is the significance of the triple marker result? What is the "normal level" of AFP? Any study on the individual variation of AFP? and the cause of the variation? Thanks,
Nathalie Lepage, PhD: It seems that you are referring to the 18-20 weeks anatomical ultrasound. The time period for prenatal screening is before this ultrasound. The 18-20 weeks ultrasound is not limited to looking at soft markers of Down syndrome, but is looking at all organs. Prenatal screening for Down syndrome and the anatomical ultrasound have different purposes and they should both be offered to all pregnant women. "Normal" levels for each marker will depend on the assay in use in your laboratory. Medians for each week of gestation need to be established in each laboratory, by using serum from non-affected pregnancies. Monitoring of the laboratory medians needs to be performed on a regular basis. MoMs need to be compared to benchmarks and adjustments to medians could be necessary.
Would you please give your recommendations on how to go about setting up an integrated screen program from a laboratory point of view? Thank you.
Los Angeles, CA
Nathalie Lepage, PhD: Extensive planning of implementation is the key to success!! Many aspects need to be addressed: selection of analytical platform(s) for the tests and training of staff/technologists and planning of quality assurance program, implementation of laboratory processes and procedures for all the situations that are different from the integrated screening program (e.g. procedures for issuing of laboratory reports for miscarriage pregnancies, for women who declined second trimester blood collection, etc.), communication with collecting blood centers for education on needs for two serum specimens (instead of one like previously), communication with radiology/ultrasound clinics on the need for implementation of NT measurement (with the associated implementation of proper quality assurance program for NT), selection of interpretative software for risk assessment of Down syndrome, communication with genetics clinics on the impact of implementation of new screening protocols, communication with health care providers on the implementation of new prenatal screening protocol. Our planning team was active for 6 months before the implementation of the integrated screening program. Some of the positive aspects of the change are an increased uptake of the prenatal screening program, better screening performance for the patients with lesser number of amniocentesis. Good luck!!
We are currently performing the quad screen for our patients. What would be the advantage of adding first trimester screening other than determining problems earlier in the pregnancy?
Nathalie Lepage, PhD: In my slide presentation, I included the advantages and disadvantages of screening in the first and/or in the second trimester of pregnancy. I will refer you to those slides. Another aspect of the answer on the selection of one or several screening protocols is based on screening performance. What percentage of sensitivity and specificity is expected by the health care providers for whom you offer the service? Is this percentage met with the available screening protocol? Communication with health care providers is important in decision-making process.
Dr. Lepage, thank you for your excellent presentation. You indicated on your slides that you achieved a 87.5% detection rate for a 5.2% false positive rate with integrated screening. The positive rate seems much higher than the predicted 1%. These numbers appear to be similar to that achieved first trimester alone so it does not appear to be an improvement. Can you please comment?
New York, New York
Nathalie Lepage, PhD: Thank you for bringing the local statistics! There is a delay in monitoring the detection rate. The statistics that I included were only based on the first year of implementation of integrated prenatal screening. However, I monitor the initial positive rate on a monthly basis and the statistics have greatly improved over the years (with continuous monitoring and adjustment in the median levels of all markers). The initial positive rate for the past two years in my institution has been approx. 2.3%. This is slightly higher than the statistics from SURUSS and FASTER, however I expect that the average age of the screening population in my center is older than these two other studies.
Is there a training course available in the near future similar to the old FBR's Prenatal Screening? Thank you very much!
Nathalie Lepage, PhD: Part of this question, I will need to defer to our colleagues previously from FBR- Maine. AACC, at its annual meeting is regularly offering Edutrak/symposium/workshop on prenatal screening, however, the length of time dedicated to these activities is not equivalent to the FBR course. I am also aware of a prenatal screening course in the UK.