Download to read offline
Uploaded byRoberto Scarafia
No results: significato e implicazioni
The document discusses the reasons why cell-free DNA (cfDNA) tests may fail, exploring both technical/statistical and biological factors that contribute to low assay quality and fetal fraction issues. It highlights the importance of quality control metrics, sample handling, and specific conditions that affect the success rate of cfDNA testing, alongside the varying no-result rates influenced by technology and biological variables. Ultimately, it emphasizes the complex interplay between fetal fraction, sample quality, and the accuracy of test results in prenatal diagnostics.
More Related Content
![062016CV_Johnson_Debra [147732]](https://cdn.slidesharecdn.com/ss_thumbnails/19b81a04-95e2-42e9-865f-f4fa76591197-161129173015-thumbnail.jpg?width=640&height=640&fit=bounds)
Similar to No results: significato e implicazioni
No results: significato e implicazioni
- 1. Francesca Romana GRATI,Ph.D., ErCLG R&D Director TOMA Advanced Biomedical Assays, S.p.A. [email protected] No results: significato e implicazioni
- 2. OUTLINE Why cfDNA testsfail Role of internal quality metrics No results for low assay/cfDNA quality Technical/statistical reasons Biological reasons No results for low FF% Factors affecting FF% Fetal fraction and no results in twins No result rate @TOMA lab
- 3. Grati FR, KaganKO. Ultrasound Obstet Gynecol. 2017 Jul;50(1):134-137 CFDNA TEST WORKFLOW Submitted initial sample population Pre- analytic QC Analysis of samples cfDNA test Post- analytic QC Samples with a report Samples rejected: • Administrative issues • Handling issues (blood collection, maintenance and transportation) • Hematic plasma • Insufficient blood/plasma amount • Submitted sample are not suitable for the analysis (e.g.: twin pregnancy when cfDNA test not validated for twins; pregnancy obtained with heterologous egg donation when cfDNA test not validated for heterologous pregnancies, …) • Low quality metrics of the extracted cfDNA No results for assay quality issues (low mat and/or fetal cfDNA quality) QC metrics filter Low quality QC metrics No results for low FF% (typically 4%) Quality QC metrics OK
- 4. Grati FR, KaganKO. Ultrasound Obstet Gynecol. 2017 Jul;50(1):134-137; Jani et al. Ultrasound Obstet Gynecol. 2015;46:515-517. WHICH QC METRICS THRESHOLD? QC metrics filter Stringency too high: •High rate of no results •Complex follow-up and counseling Stringency too low: •FF not accurately measured •Suboptimal samples included and reported •Higher FPs and FNs •Unreliable results
- 5. Grati FR, KaganKO. Ultrasound Obstet Gynecol. 2017 Jul;50(1):134-137; Jani et al. Ultrasound Obstet Gynecol. 2015;46:515-517. WHICH QC METRICS THRESHOLD? Lowest rate of no results Highest performances, reliability, accuracy and quality of test results Balance threshold is not unique and easy to determine Dependent on the technology and laboratory quality metrics policy BALANCE THRESHOLD
- 6. Studies where noresults were not reported VARIABLE RATE OF NO RESULTS – DEPENDING ON TECHNOLOGY Gil et al, Ultrasound Obstet Gynecol. 2017 Apr 11. doi: 10.1002/uog.17484
- 7. Studies where overallno results were reported VARIABLE RATE OF NO RESULTS – DEPENDING ON TECHNOLOGY Gil et al, Ultrasound Obstet Gynecol. 2017 Apr 11. doi: 10.1002/uog.17484
- 8. Studies where detailsfor no results have been reported VARIABLE RATE OF NO RESULTS – DEPENDING ON TECHNOLOGY No result rate attributable to low FF% ranges from 0.1% to 6.1% No result rate for reasons related to low quality assay ranges from 0.1% to 2.8% Gil et al, Ultrasound Obstet Gynecol. 2017 Apr 11. doi: 10.1002/uog.17484
- 9. Grati FR, KaganKO. Ultrasound Obstet Gynecol. 2017 Jul;50(1):134-137 CFDNA TEST WORKFLOW Submitted initial sample population Pre- analytic QC Analysis of samples cfDNA test Post- analytic QC Samples with a report Samples rejected: • Administrative issues • Handling issues (blood collection, maintenance and transportation) • Hematic plasma • Insufficient blood/plasma amount • Submitted sample are not suitable for the analysis (e.g.: twin pregnancy when cfDNA test not validated for twins; pregnancy obtained with heterologous egg donation when cfDNA test not validated for heterologous pregnancies, …) • Low quality metrics of the extracted cfDNA No results for assay quality issues (low mat and/or fetal cfDNA quality) QC metrics filter Low quality QC metrics No results for low FF% (typically 4%) Quality QC metrics OK
- 10. NO RESULTS FORLOW ASSAY QUALITY – POSSIBLE CAUSES Technical/statistical • Assay failure – Instrument/disposable issues – Test is too complex for the lab – Technology not enough validated • Sample mix-up (air drop) – FF is a minor component that must be preserved with a specific lab set-up for NIPT and automation • Poor assay/cfDNA quality – Intrinsic contaminants (proteins, salts, …) – Sample variability Biological • Contaminant DNA – Dizygotic co-twin demise – Wrong assumption about egg donor • Maternal and/or maternal-fetal diseases • Drugs? • Borderline FF (4-6%) • Issues in reference/normalizing samples/chromosomes
- 11. NO RESULTS FORLOW ASSAY QUALITY – POSSIBLE CAUSES Biological • Contaminant DNA – Dizygotic co-twin demise – Wrong assumption about egg donor • Maternal and/or maternal-fetal diseases • Drugs? • Borderline FF (4-6%) • Issues in reference/normalizing samples/chromosomes LOW SAMPLE PURITY - cfDNA TESTS ANALYSING SNPs! Technical/statistical • Assay failure – Instrument/disposable issues – Test is too complex for the lab – Technology not enough validated • Sample mix-up (air drop) – FF is a minor component that must be preserved with a specific lab set-up for NIPT and automation • Poor assay/cfDNA quality – Intrinsic contaminants (proteins, salts, …) – Sample variability
- 12. NO RESULT FORLOW SAMPLE PURITY Detection of an unexpected additional genotype Patient 1 Patient 2 G C G SNPsA T A T C
- 13. NO RESULTS FORLOW ASSAY QUALITY – POSSIBLE CAUSES Biological • Contaminant DNA – Dizygotic co-twin demise – Wrong assumption about egg donor • Maternal and/or maternal-fetal diseases • Drugs? • Borderline FF (4-6%) • Issues in reference/normalizing samples/chromosomes LOW ASSAY CONFIDENCE VALUE Technical/statistical • Assay failure – Instrument/disposable issues – Test is too complex for the lab – Technology not enough validated • Sample mix-up (air drop) – FF is a minor component that must be preserved with a specific lab set-up for NIPT and automation • Poor assay/cfDNA quality – Intrinsic contaminants (proteins, salts, …) – Sample variability
- 14.
- 15. NO RESULTS FORLOW ASSAY QUALITY – POSSIBLE CAUSES Clinical conditions affecting both cfDNA fragmentation, amount and quality Gestational diabetes Pre-eclampsia Maternal pernicious anemia Maternal autoimmune diseases (SLE) Intrahepatic cholestasis of pregnancy Little is known about other possible concurrent metabolic processes affecting cfDNA fragmentation and physiology during pregnancy Inflammation? exercise-induced overtraining? Other immune and auto-immune diseases? Vascular diseases? Cancer?... Schuring-Blom et al, Prenat Diagn 2016:36:790–793; Chan et al, Proc Natl Acad Sci USA 2014;11:E5302–5311; Vlokova et al, Prenat Diagn 2016;36:1156–1158; Dharajiya et al, Prenat Diagn 2015;35:990–993; Pavlidis NA. Oncologist 2002;7:279–287; Chan et al, BJOG 2017; https://doi.org/10.1111/1471-0528.15006; Bianchi Genet Med. 2017 Dec 7. doi: 10.1038/gim.2017.219.
- 16. NO RESULTS FORLOW ASSAY QUALITY – POSSIBLE CAUSES Biological • Contaminant DNA – Dizygotic co-twin demise – Wrong assumption about egg donor • Maternal and/or maternal-fetal diseases • Drugs? • Borderline FF (4-6%) • Issues in reference/normalizing samples/chromosomes Technical/statistical • Assay failure – Instrument/disposable issues – Test is too complex for the lab – Technology not enough validated • Sample mix-up (air drop) – FF is a minor component that must be preserved with a specific lab set-up for NIPT and automation • Poor assay quality – Intrinsic contaminants (proteins, salts, …) – Sample variability
- 17. Aneuploidy of off-targetnormalizing chromosomes DANSR likely fails: reference chromosomes at denominator are not euploid not accurate risk estimation counts (21) counts (1/T2/3/…/12) MPSS GW reports rare trisomy Complex management, counseling burden, unecessary invasive procedure, unecessary TOP SNP-counting likely not affected: this technology does not use reference chromosomes NO RESULT DUE TO ISSUES IN REFERENCE CHROMOSOMES
- 18. NO RESULTS FORLOW ASSAY QUALITY – POSSIBLE CAUSES Aneuploidies of normalizing/reference chromosomes Dizygotic co-twin demise with an aneuploidy of normalizing chr CPM for normalizing chr Maternal malignancies involving normalizing chr - also benign Uterine leiomyomas http://atlasgeneticsoncology.org//Tumors/leiomyomID5031.html Approximately 40% show abnormal karyotypes Usually with single or few changes; rarely, they may show complex karyotypes t(12;14), trisomy 12, deletions of portions of the long arms of chromosomes 3 or 7 and the short arm of chromosome 1, rearrangements of the short arm of chromosome 6 and rearrangements of chromosomes 1, 3, 10, 13 and X. Schuring-Blom et al, Prenat Diagn 2016:36:790–793; Chan et al, Proc Natl Acad Sci USA 2014;11:E5302–5311; Vlokova et al, Prenat Diagn 2016;36:1156–1158; Dharajiya et al, Prenat Diagn 2015;35:990–993; Pavlidis NA. Oncologist 2002;7:279–287; Chan et al, BJOG 2017; https://doi.org/10.1111/1471-0528.15006; Bianchi Genet Med. 2017 Dec 7. doi: 10.1038/gim.2017.219.
- 19. Grati FR, KaganKO. Ultrasound Obstet Gynecol. 2017 Jul;50(1):134-137 CFDNA TEST WORKFLOW Samples rejected: • Administrative issues • Handling issues (blood collection, maintenance and transportation) • Hematic plasma • Insufficient blood/plasma amount • Submitted sample are not suitable for the analysis (e.g.: twin pregnancy when cfDNA test not validated for twins; pregnancy obtained with heterologous egg donation when cfDNA test not validated for heterologous pregnancies, …) • Low quality metrics of the extracted cfDNA No results for assay quality issues (low mat and/or fetal cfDNA quality) QC metrics filter Low quality QC metrics No results for low FF% (typically 4%) Submitted initial sample population Pre- analytic QC Analysis of samples cfDNA test Post- analytic QC Samples with a report Quality QC metrics OK
- 20. Palomaki et al,Genet Med 2011; 13: 913–920; Canick, et al. Prenat Diagn 2013, 33, 1–8 cfDNA results by MPSS are expressed in ‘z-score’ values Distribution of Z-scores: euploid fetuses: distribution is Gaussian, with a mean of 0 and a SD of 1 trisomic fetuses: the mean of the Z-score > 0 and increases with increasing fetal fraction NO RESULTS FOR LOW FETAL FRACTION – WHY FF MATTERS
- 21. Wang E etal, Prenat Diagn. 2013 Jul;33(7):662-6; Ashoor et al, Fetal Diagn Ther 2012;31:237–243, 2012 Fetal fraction of cell-free DNA in maternal plasma At 10GA FF is in average 10%
- 22. Courtesy of ThomasMusci; Wang E et al, Prenat Diagn. 2013 Jul;33(7):662-6; Ashoor et al, Fetal Diagn Ther 2012;31:237–243, 2012 Fetal fraction of cell-free DNA in maternal plasma 22,384 samples 2% samples < 4% fetal cfDNA
- 23. Courtesy of ThomasMusci (Ariosa); Wang E et al, Prenat Diagn. 2013 Jul;33(7):662-6. Fetal fraction increases with gestational age Post 21 wks GA: fetal fraction increases by 1.1% per week Prior to 21 wks GA: fetal fraction increases by 0.11% per week 10 15 20 25 30 35 40 0.00.10.20.30.4 GestationalAgeFractionalWeeks FractionFetal
- 24. Courtesy of ThomasMusci (Ariosa); Wang E et al, Prenat Diagn. 2013 Jul;33(7):662-6, Ashoor et al, Fetal Diagn Ther 2012;31:237–243, 2012 Fetal fraction decreases with increasing maternal weight Maternal Weight Bin (kg) Pregnancies with ≥ 4% fetal cell- free DNA (%) <50 >99 ≥50 & <60 >99 ≥60 & <70 >99 ≥70 & <80 99 ≥80 & <90 98 ≥90 & <100 96 ≥100 & <110 95 ≥110 & <120 90 ≥120 & <130 88 ≥130 & <140 81 ≥140 71 Increased apoptosis of cfmatDNA for a fixed amount of cffDNA determined by the size of the placenta dilution effect
- 25. Sarno et alUltrasound Obstet Gynecol 2016; 47: 705–711 Global no result rate by cfDNA testing: risk factors Increases with increasing maternal BMI
- 26. Wang E etal, Prenat Diagn. 2013 Jul;33(7):662-6; Canick, et al. Prenat Diagn 2013, 33, 1–8; Brar H et al., J Matern Fetal Neonatal Med. 2013 Jan;26(2):143-5; Poon et al, Fetal Diagn Ther 2013;33:215–223; Musci, Schmid et al, Oral presentation ISPD 2015; Ashoor et al, Fetal Diagn Ther 2012;31:237–243, 2012; Ma et al, Ultrasound Obstet Gynecol; Ultrasound Obstet Gynecol 2018; 51: 274–277; Rolnik et al, Ultrasound Obstet Gynecol. 2018 Jan 10. doi: 10.1002/uog.18993. Other factors affecting FF% ↑ with maternal serum level of PAPP-A, ß-HCG* ↑ with CRL ↑ in women of Afro-Caribbean and East-Asian racial origin than in Caucasians*** ↓ in smokers ** ↓ in IVF pregnancies (probably related to PAPP-A, ß-HCG) ↓ in Twins ↓ in AMA ↓ chronic hypertension, ↓ type 1 or type 2 diabetes mellitus ↓ South Asian ethnicity Use of low-molecular-weight heparin: heparin exerts a direct effect on the trophoblast by reducing apoptosis *** increased placental mass and/or activity in these racial groups reflected in the serum levels of free β-hCG and PAPP-A. * the number of apoptotic cells would be proportional to the placental mass, reflected in the serum concentration of free β-hCG and PAPP-A, which are also produced by the placenta ** impaired placental development and decreased placental mass in such patients
- 27. Supplement to JANUARY2015, American Journal of Obstetrics & Gynecology, S45-Abstract 65; Pergament et al, OBSTETRICS & GYNECOLOGY, 2014; Palomaki et al, Prenatal Diagnosis 2015; Norton et al, NEJM 2015; Musci, Schmid et al, Oral presentation ISPD 2015; Revello et al, UOG 2016 Fetal fraction decreases in presence of a fetal trisomy Samples with FF<4%: increased incidence, statistically significant, of chr abnormalities compared with the overall cohort with FF≥4% (13.8% vs 2.4% - 2.7% vs 0.4%) ↓ %FF trisomy 18-13-triploids-MX (Not in T21) Median fetal fraction: 11.0% (IQR 8.3-14.4%) in the unaffected group 10.7% (IQR 7.8-14.3%) in trisomy 21 8.6% (IQR 5.0-10.2%) in trisomy 18 7.0% (IQR 6.0-9.4%) in trisomy 13 Via placental restrictionreduced cytotrophoblastic layerreduced ‘cfpDNA’ shedding: Likely to involve any chr abn affecting placental growth
- 28. Rolnik et al,Ultrasound Obstet Gynecol. 2018 Jan 10. doi: 10.1002/uog.18993. Fetal fraction as marker of placental dysfunction Mean arterial pressure (MAP) and the mean uterine arteries pulsatility indices (UtAPI) are negatively associated PAPP-A and PlGF are positively associated with fetal fraction Women at increased risk for PE and FGR and for placental disorders tend to have lower fetal fraction and would be more likely to have reduced fetal fraction on cfDNA testing or to receive a failed result report Lower fetal fraction may be a consequence of smaller placental mass and even an early sign of placental dysfunction.
- 29. Gil et al,Fetal Diagn Ther. 2014;35(3):204-11; Struble et al, Fetal Diagn Ther. 2014;35(3):199-203; Bevilacqua, Ultrasound Obstet Gynecol. 2014 Oct 9. doi: 10.1002/uog.14690 Twins: risk assessment with cfDNA testing Monozygotic: • with rare exception concordant and much like singleton • Generally increased fetal cfDNA % (fetal fraction) Dizygotic: • Different contributions of cfDNA in maternal plasma: may vary by 2-fold • DANSR technology proposes analysis based on lower of the two fetal fraction values • Reduces possibility of ‘false negative’
- 30. Sarno et alUltrasound Obstet Gynecol 2016; 47: 705–711; Lee et al, Hum Reprod. 2018 Feb 15. doi: 10.1093/humrep/dey033 Global no result rate by cfDNA testing: risk factors IVF Spontaneous Singleton ---- Twins Higher in twins than singleton (3-folds higher at first blood draw) Higher in IVF than spontaneous A repeated sampling provided a result in 51.3% of twin pregnancies and 63% of singletons
- 31. No results forlow FF% in twins The rationale for this choice is to avoid a false negative result in a dizygotic twin pregnancy discordant for aneuploidy where the total fetal fraction is satisfactory but the contribution of the affected fetus could be less than 4%. An unavoidable consequence of such policy is that no result rate is higher with those technologies measuring individual contribution from each zygote MPSS cannot quantify the contribution from each twin as SNP genotyping is not performed a lower no result rate is expected at the cost of a higher FNR Sarno et al Ultrasound Obstet Gynecol 2016; 47: 705–711; Grömminger et al, J. Clin. Med. 2014, 3, 679-692; doi:10.3390/jcm3030679
- 32. NO RESULT RATE(NRR) OF TOMA LAB Overall NRR after 2nd redraw for T21,18,13: 0.65% NRR for FF<4% at 2nd redraw: 0.1% NRR for low cfDNA quality: after 2nd redraw: 0.55% Partial NRR for SCA (all for quality metrics) after 1st draw : 0.85% Unpublished data
- 33. NO RESULT FORLOW QUALITY OF TOMA LAB Unpublished data • No results for quality issues increase at borderline FF% (4-6%)
- 34. No result ratedoes not always mean ‘lab failure’ rather ‘fail to meet quality standards’ for a reliable result If NRR is too high Technology is too complex for the lab or not enough validated or high CQ metrics stringency If NRR is too low or absent No/inaccurate FF measurement and/or low CQ metrics stringency There are not unique reasons for no results: different reasons for different technologies Strictly dependent on the laboratory performing the test Medical providers Should be aware of the reasons for no results of the chosen technology and laboratory Ensure appropriate technology-based pre/post test counselling is provided CONCLUSION