Uptake of noninvasive prenatal testing (NIPT) in women following positive aneuploidy screening

Authors


  • Funding sources: None
  • Conflicts of interest: Dr. Norton is a co-principal investigator on clinical trial NCT0145167, sponsored by Ariosa Diagnostics.

ABSTRACT

Objective

The aim of this study was to investigate how the introduction of noninvasive prenatal testing (NIPT) impacted women's testing choices following a positive prenatal screening (PNS) result.

Methods

Beginning in March 2012, women referred to our Prenatal Diagnosis Center following a positive PNS result were offered NIPT or invasive prenatal diagnosis. Rates of invasive testing and declining follow-up were compared with testing decisions the prior year. Differences were compared using t-test and chi-square. Multivariable logistic regression was performed to identify predictors of test choice.

Results

Between March 2012 and February 2013, 398 screen positive women were seen: 156 (39.2%) underwent invasive testing, 157 (39.4%) had NIPT and 84 (21.1%) declined further testing. In the prior year, 638 screen positive patients were seen: 301 (47.2%) had invasive testing and 337 (52.8%) declined. The rate of invasive testing declined significantly (p = 0.012). Moreover, fewer women declined follow-up testing after introduction of NIPT, 21.2% versus 52.8%, p ≤ 0.001. Race/ethnicity and timing of results (first versus second trimester) were predictors of testing choices; payer and maternal age were not.

Conclusion

The introduction of NIPT resulted in a significant decrease in invasive diagnostic testing. Additionally, fewer women declined further testing when NIPT was available. © 2013 John Wiley & Sons, Ltd.

INTRODUCTION

The recent introduction of noninvasive prenatal testing (NIPT) for trisomies 13, 18 and 21 broadens the available prenatal testing options. Traditional screening tests for fetal aneuploidy are based on indirect serum and ultrasound markers and have reported detection rates of 85–90% and false positives rates of about 5%.[1, 2] Positive screening results necessitate follow-up diagnostic testing for confirmation; diagnostic testing via chorionic villus sampling (CVS) or amniocentesis carries a loss rate of approximately 1 in 300 procedures.[3, 4]

Noninvasive prenatal testing now offers an alternative to invasive diagnostic testing in the setting of positive screening results, with high sensitivity and specificity for trisomies 18 and 21.[5, 6] The performance of testing for trisomy 13 is somewhat poorer than for trisomies 18 and 21.[7] Invasive diagnostic testing is more definitive and offers the benefit of testing for a broad array of chromosomal abnormalities. NIPT at present is focused primarily on trisomies 13, 18 and 21, which comprise approximately 70% of all aneuploidies detected at prenatal diagnosis.[8]

Survey studies by Tischler et al. and Kooij et al. have found that women are interested in NIPT, particularly older women, those of Caucasian and Asian ethnicity and those that would terminate a pregnancy.[9, 10] However, there are limited studies to demonstrate women's choices in actual practice because NIPT has become available.

In March 2012, we began to offer NIPT as an option to all women seen for genetic counseling and follow-up after a positive result on standard aneuploidy screening with serum analytes and nuchal translucency. We were interested in how the additional option of NIPT impacted women's choices regarding invasive diagnostic testing. Because the uptake of prenatal diagnostic testing has been shown to vary across different ethnic and racial groups,[11] we were also interested in how patients of different sociodemographic backgrounds utilized prenatal testing options.

Women in California are all offered participation in the California Prenatal Screening Program. The current screening offered includes second-trimester quad marker screening, serum integrated screening and sequential integrated screening, which includes nuchal translucency ultrasound and a first-trimester result. Women who are identified as high risk on the basis of screening results are eligible for follow-up services including genetic counseling, ultrasound and invasive diagnostic testing with CVS or amniocentesis at state-approved prenatal diagnosis centers; all costs of such follow-up are covered through the state program. Women who are high risk following first-trimester serum and nuchal translucency results are also offered the option of second-trimester serum screening to complete a fully integrated (sequential) approach. At the present, the state program does not cover the cost of NIPT. The cost incurred by the patient varies by the specifications of the private insurance plan and deductible, and is typically approximately $795 or less in our center. The cost of NIPT is covered fully for patients with California public insurance (MediCal).

METHODS

We designed a retrospective cohort study to compare prenatal screening and diagnostic testing choices made by women who had positive results based on traditional aneuploidy screening with first-trimester and/or second-trimester serum analytes and nuchal translucency ultrasound. We evaluated choices regarding follow-up testing in women who were seen from March 2012 through February 2013, after the introduction of NIPT as a routine option for follow-up of positive aneuploidy screening results. We compared test choices in this group to the cohort of women seen in the prior year (March 2011–February 2012), before the introduction of NIPT as a routine option in our center.

A query of the genetic counseling database was performed to identify all women referred because of positive aneuploidy screening results between 1 March 2011 and 28 February 2013. Women seen between March 2011 and February 2012 were offered invasive prenatal diagnosis with CVS or amniocentesis, or the option to decline further testing. Beginning in March 2012, women were also offered the option of NIPT. Genetic counseling sessions were performed between 11 and 24 weeks' gestational age, following abnormal first-trimester or second-trimester screening results. Invasive testing was described and risks were explained. CVS and amniocentesis were described as diagnostic procedures that would give information regarding the full karyotype; patients were quoted a 1 in 300 risk of pregnancy loss for these procedures. Accuracy for invasive testing was described as greater than 99%. NIPT was described as a screening test with an accuracy of 99% for Down syndrome and 97% for trisomy 18. Women were counseled that results would be available in approximately 10–14 days for both invasive testing and NIPT. If fluorescence in situ hybridization was performed at the time of amniocentesis, then results would be available in 2–3 days; however, an additional cost would be incurred.

We determined the rates of different testing decisions, including the number of women who underwent invasive diagnostic testing, NIPT or no further testing. We compared rates of invasive testing and no further testing before and after the introduction of NIPT. Rates were compared using t-test or chi-square tests as appropriate.

Demographic data including race and ethnicity, maternal age, gestational age and insurance status were recorded. Predictors of testing decisions were assessed with multiple logistic regression.

RESULTS

Between March 2012 and February 2013, 430 women with positive serum screening results were seen for evaluation and genetic counseling. Women with incorrect pregnancy dating as a cause of the positive screen (n = 32) were excluded. Of the 398 remaining patients, 125 were seen following a first-trimester positive result, and 214 were seen following second-trimester results. For 59 patients, the timing of screening was unknown. In the prior year, between March 2011 and February 2012, 695 women were seen for genetic counseling following a positive aneuploidy screen. Of this group, 57 were excluded because of incorrect dating, leaving 638 available for analysis. The women in the two periods were similar with respect to maternal age, race/ethnicity and insurance payer. Characteristics of the patients with positive aneuploidy screening results prior to and following the introduction of NIPT are provided in Table 1.

Table 1. Demographics of patients with positive aneuploidy screening before and after introduction of NIPT
 Pre-NIPT n = 638Post-NIPT n = 398p-value
  1. p-value: t-test for continuous variables, chi-squared for categorical variables.

  2. NIPT, noninvasive prenatal testing.

Maternal age, mean ± SD34.2 ± 6.434.9 ± 5.70.080
Maternal age ≥ 35 years, n (%)349 (54.7)234 (58.8)0.206
Race/ethnicity, n (%)   
Caucasian146 (22.9)110 (27.6)0.084
Hispanic289 (45.3)177 (44.5) 
Asian/Pacific Islander182 (28.5)106 (26.6) 
Other21 (3.3)5 (1.3) 
Primary payer, n (%)   
Medicaid258 (40.4)158 (39.7)0.813
Private380 (59.5)240 (60.3) 

Following the introduction of NIPT, 398 screen positive women were offered follow-up options of invasive diagnostic testing with CVS or amniocentesis, or NIPT. They were also told that they could decline follow-up testing. Of the total, 156 (39.2%) underwent invasive testing, 157 (39.4%) chose NIPT and 84 (21.1%) declined follow-up testing. In women 35 years and older, 39.6% chose NIPT, 40.2% chose invasive testing and 19.7% declined all testing. With regard to race/ethnicity, Caucasian and Asian patients were least likely to decline all testing (13.6% and 12.3%, respectively). In Hispanic women, 42.9% had NIPT, 31.6% declined all testing and 25.4% had invasive testing (p < 0.001). Patients with Medicaid coverage were more likely to have NIPT (46.8%) than to decline all testing (26.6%) or undergo invasive testing (26.6%; Table 2). Of the 158 patients who had NIPT, seven pursued invasive testing despite low-risk results for aneuploidy; all had normal karyotypes. Five patients underwent amniocentesis for confirmation following NIPT results, indicating a high risk for T21 or T18; all results were confirmed.

Table 2. Rates of accepting and declining invasive prenatal diagnosis by socioeconomic characteristics
 Pre-NIPT n = 638Post-NIPT n = 398p-value
  1. p-value from chi-squared test.

  2. NIPT, noninvasive prenatal testing.

Total accepting invasive prenatal diagnosis, n (%)301/638 (47.2)156/398 (39.2)0.012
Maternal age ≥ 35 years185/349 (53.0)94/234(40.2)0.002
Race/ethnicity   
Caucasian97/146 (66.4)53/110 (48.2)0.003
Hispanic80/289 (27.7)45/177 (25.4)0.593
Asian/Pacific Islander114/182 (62.6)56/106 (52.8)0.103
Other10/21 (47.6)2/5 (40.0)0.759
Insurance   
Medicaid71/258 (27.5)42/158 (26.6)0.835
Private230/380 (60.5)114/240 (47.5) 
Total declining further testing337/638 (52.8)84/398 (21.2)<0.001
Maternal age ≥ 35 years164/349 (47.0)46/234 (19.7)<0.001
Race/ethnicity   
Caucasian49/146 (33.6)15/110 (13.6)<0.001
Hispanic209/289 (72.3)56/177 (31.6)<0.001
Asian/Pacific Islander68/182 (37.4)13/106 (12.3)<0.001
Other11/21 (52.4)0/5 (0.0)0.033
Insurance   
Medicaid187/258 (72.5)42/158 (26.6)<0.001
Private150/380 (39.5)41/240 (17) 

These results were compared with those of the prior year, before availability of NIPT. Between March 2011 and February 2012, 638 women were counseled and offered follow-up for positive screening results. Of this group, 301 (47.2%) had invasive testing, and 337 (52.8%) declined follow-up testing. Just over half of the women who were older than 35 years chose invasive testing (53%), whereas the remainder declined (47%). In Hispanic women, 72.3% declined invasive testing, whereas 27.7% underwent invasive testing. Similar to the pattern observed in the group following the availability of NIPT, 66.4% of Caucasian women and 62.6% of Asian women chose to undergo invasive testing, whereas the remainder declined follow-up testing (33.6% and 37.4%, respectively; Table 2).

Introduction of NIPT was associated with a decreased acceptance of invasive prenatal diagnosis (39.2% when NIPT was available vs 47.2% in the prior year, p = 0.012, odds ratio (OR) 0.72, 95% confidence interval (CI) 0.56–0.93; Table 2). This association persisted after controlling for race/ethnicity, maternal age ≥ 35 years and payer (adjusted OR 0.67, 95% CI 0.51–0.88). Furthermore, the introduction of NIPT significantly decreased the likelihood that a patient would decline further testing (21.2% when NIPT was available vs 52.8% prior to its introduction, p ≤ 0.001, OR 0.23, 95% CI 0.18–0.32). This association also persisted after controlling for race/ethnicity, maternal age ≥ 35 years and payer (adjusted OR 0.21, 95% CI 0.15–0.28).

We also examined sociodemographic factors predictive of testing decisions. Women who were 35 years or older were more likely to choose invasive testing in the period prior to the introduction of NIPT (OR 1.67, 95% CI 1.21–2.29). Hispanic women (OR 0.31, 95% CI 0.19–0.51) and women with public insurance (OR 0.25, 95% CI 0.18–0.35) were less likely to choose invasive testing. In multivariable analysis concurrently adjusting for race/ethnicity, advanced maternal age and insurance payer, following the introduction of NIPT, only Hispanic ethnicity was a significant predictor of declining follow-up testing (OR 3.63, 95% CI 1.69–7.80; Table 3).

Table 3. Characteristics of women accepting NIPT (only after introduction of NIPT), adjusted odds ratios
 Accepted NIPTAccepted iPNDDeclined further testing
  1. NIPT, noninvasive prenatal testing; iPND, invasive prenatal diagnosis (i.e. chorionic villus sampling or amniocentesis); aOR, odds ratio adjusted for other characteristics in table; n/c, not calculable (too few numbers).

Race/ethnicityaOR (95% CI)aOR (95% CI)aOR (95% CI)
CaucasianReferenceReferenceReference
Hispanic1.09 (0.58–2.03)0.36 (0.19–0.68)3.63 (1.69–7.80)
Asian/Pacific Islander0.77 (0.42–1.39)1.25 (0.71–2.20)0.94 (0.42–2.12)
Other2.01 (0.30–13.41)0.68 (0.10–4.47)n/c
Maternal age ≥ 35 years vs age < 35 years0.97 (0.62–1.51)1.18 (0.75–1.86)0.93 (0.56–1.55)
Medicaid vs private1.63 (0.92–2.87)0.66 (0.37–1.18)0.80 (0.41–1.51)

Of the patients seen following abnormal first-trimester screening, 20 (16%) declined further testing, 70 (56%) had NIPT, 24 (19%) had invasive testing and 11 (8%) had both NIPT and invasive testing. In the patients seen following abnormal second-trimester screening, 45 (21.5%) declined further testing, 78 (37%) had NIPT, 85 (40.6%) had invasive testing and 1 patient had both NIPT and invasive testing. Abnormal first-trimester screen (OR 7.59, 95% CI 3.84–15) was more strongly associated with acceptance of NIPT than abnormal second-trimester screen (OR 2.03, 95% CI 1.07–3.84) after controlling for ethnicity, maternal age ≥ 35 years and insurance payer. However, both were associated strongly with acceptance of NIPT, but not invasive testing. Interestingly, a positive first-trimester screen was associated in this dataset with a decreased likelihood of acceptance of invasive testing (OR 0.24, 95% CI 0.13–0.46). This may be because patients were given the option of undergoing second-trimester screening to complete a sequential screening protocol prior to deciding on invasive testing versus NIPT.

DISCUSSION

In our practice, the introduction of NIPT resulted in a significant decrease in the number of women who opted for invasive testing following a positive aneuploidy screening result. We also found a significant decrease in the number of women who declined all follow-up testing when the option of NIPT was presented.

A study by Hill et al. compared the preferences of women and health professionals for noninvasive prenatal diagnosis for Down syndrome relative to current invasive tests. Results of surveys showed that women and their health care providers preferred tests conducted early in pregnancy, with high accuracy, and an ability to give information about conditions beyond Down syndrome alone. In contrast to their providers, women were willing to wait longer for test results with lower accuracy if the test had no risk of miscarriage.[12] Interestingly, we found that although the rate of invasive prenatal diagnostic testing decreased, it remained equivalent to the rate of NIPT when both were offered. Although we do not know all the reasons that women had for making specific choices, this may be due to costs of NIPT as well as the increased information and more definitive results provided by invasive testing.

We observed an increased uptake of NIPT following abnormal first-trimester screening compared with abnormal second-trimester screening (56% vs 37%). As part of the California program, women with abnormal first-trimester screening results are offered the option of invasive diagnostic testing or a second-trimester serum screen to complete a fully integrated (sequential) screening approach. For many of these patients, NIPT was likely seen as an alternative to integrated screening, still noninvasive when compared with invasive testing, but with a more definitive result. In addition, the uptake of invasive testing was higher in the second trimester than in the first (41% vs 19%); this may have been due to the greater comfort of some patients with amniocentesis than CVS. In addition, amniocentesis was more readily available in our satellite clinics, as CVS more frequently requires travel to another center.

Our study demonstrates that NIPT influences the decisions that women with a positive aneuploidy screening result make for follow-up testing. It appeared to have had a significant impact on both women who wanted more definitive information, as well as on women who would previously have declined follow-up. Similar to prior studies of patient preferences regarding diagnostic testing, when offered an alternative option with no risk but potential to gain valuable information, some women opt for alternatives that are less invasive, whereas others prefer the certainty of invasive testing and full karyotype.[9, 13]

Although prenatal testing is at times considered as valuable primarily so that pregnancy termination can be considered when results are abnormal, for many, there is other value in the information. When women are presented with the option of invasive testing, they may think more critically about the information to be gained by the procedure as there is some risk involved. However, when the procedure-associated risk is eliminated, women may be less likely to decline testing as they do see value in the information.

Limitations of our study include the retrospective study design, which limits our ability to determine the precise reason that a given testing decision was made. Women's decisions to proceed with invasive testing may be influenced by various factors, including perceptions of risk, prior experience with invasive testing, prior knowledge of NIPT and reasons for obtaining further information. Our study was performed at one center with significant ethnic diversity, although the number of women older than 35 years may be overrepresented as compared with centers in other parts of the country. The California State Screening program may also influence the decision-making process of these patients compared with women where prenatal screening is not performed in the same standardized manner. Additionally, the relatively short period examined in our study may also influence the observed pattern, as awareness of NIPT is rapidly increasing over time.

Noninvasive prenatal testing is clearly having a great impact on the field of prenatal aneuploidy testing. Our study demonstrates that even in the first year of introduction of NIPT as a prenatal testing option following abnormal aneuploidy screening, there was a significant decrease in invasive testing. A significant proportion of women who were identified as screen positive by first-trimester and second-trimester screening were identified as low risk by NIPT, significantly decreasing the risk of the common aneuploidies and the incurred risk of invasive procedures. NIPT also offered reassurance to those women for whom invasive diagnostic testing was not an acceptable option and who would previously have declined follow-up and likely continued with some degree of anxiety throughout their pregnancy. As providers to these women, the additional information gained by NIPT in women with high risk for aneuploidy based on serum screening allows for improved use and access of pediatric resources if there is a significant concern for any of the trisomies. Although it is unknown how the narrow scope of abnormalities detected by NIPT influenced testing decisions, it is likely that as the scope of NIPT broadens beyond detection of trisomies, additional women may opt for NIPT rather than invasive testing.

WHAT'S ALREADY KNOWN ABOUT THIS TOPIC?

  • NIPT now offers an alternative to invasive diagnostic testing in the setting of positive screening results, with high sensitivity and specificity for trisomies 18 and 21.

WHAT DOES THIS STUDY ADD?

  • The introduction of NIPT in our population has influenced the prenatal diagnostic choices made by women with abnormal aneuploidy screening. We found that rates of invasive testing decreased significantly following the introduction of NIPT. Additionally, fewer women declined follow-up testing when NIPT was an option.

Ancillary