(12) INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY(PCT)
`
`(19) World Intellectual Property
`InternationalBureau
`Oreanizati
`
`(43) International Publication Date
`17 March 2005 (17.03.2005)
`
`
`
`fe
`AAIO:Sea)
`Miles
`
`
`eS
`
`PCT
`
`UIIVALNTANONAARCATRNO
`
`(10) International Publication Number
`WO 2005/023091 A2
`
`(51) International Patent Classification’:
`
`A61B
`
`(21) International Application Number:
`PCT/US2004/028857
`
`(22) International Filing Date:
`7 September 2004 (07.09.2004)
`
`R. [US/US]; 526 Stratford Court, Del Mar, CA 92014
`(US). DING, Chunming [CN/US],; 10-C Sagamore
`Way, Waltham, MA 02453 (US). LO, Yuk, Ming, Den-
`nis [GB/CN]; 4th Tloor, 7 King Tak Street, Homantin,
`Kowloon, Hong Kong (CN). CHIU, Rossa, Wai-Kwun
`[AU/CN],; Flat 1A, Block 1, Constellation Cove,
`| Hung
`Lam Drive, Tai Po, N.T., Hong Kong (CN).
`
`(25) Filing Language:
`
`(26) Publication Language:
`
`English
`
`Lnglish
`
`(30) Priority Data:
`60/500,526
`
`5 September 2003 (05.09.2003)
`
`US
`
`(71) Applicants (for all designated States except US): THE
`TRUSTEES OF BOSTON UNIVERSITY [US/US];
`One Sherborn Street, Boston, MA 02215 (US). THE
`CHINESE UNIVERSITY OF HONG KONG[CN/CN];
`Shatin N.T., Hong Kong SAR (CN).
`
`(74) Agents: EISENSTEIN,Ronald,I. et al.; Nixon Peabody
`LLP, 100 Federal Street, Boston, MA 02110-2131 (US).
`
`(81) Designated States (unless otherwise indicated, for every
`kind of national protection available): AE, AG, AL, AM,
`AT, AU, AZ, BA, BB, BG, BR, BW, BY, BZ, CA, CH, CN,
`CO, CR, CU, CZ, DE, DK, DM, DZ, EC, EE, EG, ES, FI,
`GB, GD, GE, GH, GM, HR, HU,ID, IL, IN, IS, JP, KE,
`KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MA, MD,
`MG, MK, MN, MW, MX, MZ, NA, NI, NO, NZ, OM, PG,
`PH, PL, PT, RO, RU, SC, SD, SE, SG, SK, SL, SY, TJ, TM,
`TN,TR, TT, TZ, UA, UG, US, UZ, VC, VN, YU, ZA, ZM,
`ZW.
`
`(72)
`(75)
`
`Inventors; and
`Inventors/Applicants (for US only): CANTOR,Charles,
`
`(84) Designated States (unless otherwise indicated, for every
`kind of regional protection available): ARIPO (BW, GH,
`
`[Continued on next page]
`
`(54) Title: METIIOD FOR NON-INVASIVE PRENATAL DIAGNOSIS
`
`Spearman rank correlation
`R=0.421
`P= 0.0191
`
`Positives
`
` 50
`
`30
`
`40
`
`SRY genome-equivalents/mL
`
`60
`
`70
`
`(57) Abstract: The present invention is directed to methods of detecting nucleic acids in a biological sample. The method is based on
`Uf) 2 novel combinationofa base extension reaction, which provides excellent analytical specificity, and a mass spectrometric analysis,
`& which provides excellent specificity. ‘he method can be used, for example, for diagnostic, prognostic and treatment purposes. ‘The
`20 method can be used, for example, for diagnostic, prognostic and treatment purposes. The method allows accurate detection of nucleic
`acids that are present in very small amounts in a biological sample. For example, the method of the present invention is preferably
`used to detect fetal nucleic acid in maternal blood sample; circulating tumor-specific nucleic acids in a blood, urine or stool sample;
`and donor-specific acids in transplant recipients. In another embodiment, one can detect viral, bacterial, fungal, or other foreign
`
`= nucleic acids in biological sample.
`
`
`
`
`
`023091A2TIMNIMINVTNINNANITAINATYTNA
`
`

`

`WO 2005/023091 A2
`
`—_[IIINTIIITIAINERNTITIANNIU INTACIMATA TMT
`
`Published:
`GM, KE, LS, MW, MZ, NA, SD, SL, SZ, TZ, UG, ZM,
`ZW), Eurasian (AM, AZ, BY, KG, KZ, MD, RU, TJ,TM), — without international search report and to be republished
`European (AT, BE, BG, CH, CY, CZ, DE, DK, EE, ES, FI,
`upon receipt of that report
`FR, GB, GR, HU,IE, IT, LU, MC, NL, PL, PT, RO, SE,SI,
`SK, TR), OAPI (BF, BJ, CF, CG, CI, CM, GA, GN, GQ,
`GW, ML, MR, NE, SN, TD, TG).
`
`For two-letter codes and other abbreviations, refer to the "Guid-
`ance Notes on Codes and Abbreviations” appearing at the begin-
`ning of each regular issue of the PCT Gazette.
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`METHOD FOR NON-INVASIVE PRENATAL DIAGNOSIS
`
`CROSS-REFERENCE TO RELATED APPLICATIONS
`
`The present application claims benefit under 35 U.S.C. §119(e) of U.S. provisional
`
`patent application Serial No. 60/500,526, filed on September 5, 2003, the content of whichis
`
`herein incorporated by referencein its entirety.
`
`BACKGROUND
`
`[001]
`
`Theanalysis of circulating nucleic acids has revealed applications in the
`
`noninvasive diagnosis, monitoring, and prognostication of manyclinical conditions.
`
`[002]
`
`For example, in non-invasive method fo prenatal monitoring, fetal DNA has been
`
`found to circulate in maternal plasma (Lo, Y.M.D. et al, Lancet 350, 485-487 (1997)), and
`
`development of such non-invasive prenatal diagnosis has therefore been suggested based on
`
`the analysis of a maternal blood sample. Although the non-invasive nature of such
`
`approaches represents a major advantage over conventional methods. However, the technical
`
`challenge posed by the analysis of fetal DNA. Thus, in maternal plasmalies in the need to be
`
`able to discriminate the fetal DNA from the co-existing background maternal DNA,and the
`
`diagnostic reliability of circulating DNA analysis depends on the fractional concentration of
`
`the targeted sequence, the analytical sensitivity, and the specificity of the method.
`
`[003]
`
`Fetal DNA represents a minorfraction of the total DNA in maternal plasma,
`
`contributing approximately 3% to 6% of the total maternal plasma DNAin early and late
`
`pregnancy, respectively (Lo, Y.M.D.et al. Am J Hum Genet 62, 768-775 (1998)).
`
`[004]
`
`Mostdiagnostic applications reported to date have focused on detecting of
`
`paternally-inherited fetal traits or mutations, as these are more readily distinguishable from
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-2-
`
`the backgroundmaternal DNA.Reported applications include the prenatal diagnosis of sex-
`
`linked diseases (Costa, J.M., Benachi, A. & Gautier, E. NEngl JMed 346, 1502 (2002)), fetal
`
`RhDstatus (Lo, Y.M.D, et al. NEngl JMed 339, 1734-1738 (1998)) and certain paternally-
`
`transmitted autosomal dominant conditions, including achondroplasia and myotonic
`
`dystrophy (Chiu, R.W.K. & Lo, Y.M.D. Expert Rev Mol Diagn 2, 32-40 (2002)).
`
`[005]
`
`Fetal SRY and RHD DNAdetection from maternal plasmahas reached close to
`
`100% accuracy, as corifirmed by many large scale evaluations (Sekizawa, A., Kondo,T.,
`
`Iwasaki, M., Watanabe, A., Jimbo, M., Saito, H. & Okai, T. (2001) Clin. Chem. 47, 1856-
`1858; Finning, K. M.. Martin, P. G., Soothill, P. W. & Avent, N. D. (2002) Transfusion 42,
`
`1079-1085; Costa, J. M., Benachi, A., Gautier, E., Jouannic, J. M., Ernault, P. & Dumez, Y.
`
`(2001) Prenatal Diagn. 21, 1070-1074; Rijnders, R. J., Christiaens, G. C., Bossers, B., van
`
`der Smagt,J. J., van der Schoot, C. E. & de Haas, M. (2004) Obstet. Gynecol. 103, 157-164).
`
`However,its general applicability is limited. The high level of diagnostic accuracy in these
`
`conditions is attained by the analytical sensitivity contributed by the use of real-time
`
`quantitative PCR (Lo Y et al. Am. J. Hum. Genet. 62:768-775, 1998; Heid et al., Genome
`
`Res. 6:986-994, 1996), and the analytical specificity conferred choosing fetal DNA targets
`
`that are absolutely fetal-specific. The RHD sequence does not exist in the genome ofa
`
`rhesus D negative mother, and SRY, which is used to detect the presence of a Y
`
`chromosome, does not exist in a genome of a normal woman. Consequently, the maternal
`
`plasma SRY and RHDanalysesare relatively free from interference by the background
`
`maternal DNA. This differs from a numberof other conditions.
`
`[006]
`
`Manyfetal genetic diseases are caused by mutationsthat result in more subtle
`
`genetic differences between the maternal and fetal DNA sequences in maternal plasma. While
`
`such fetal diseases may theoretically be diagnosed non-invasively by means of the detection
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-3-
`
`orexclusion of the paternally inherited mutantallele in maternal plasma, the development of
`
`robust assays for the discrimination of less dramatic differences between fetal and maternal
`
`DNAin maternal plasma has been technically challenging (Nasis, O., Thompson,S., Hong,
`T., Sherwood, M., Radcliffe, S., Jackson, L. & Otevrel, T. (2004) Clin. Chem. 50, 694-701).
`
`Therefore, despite many potential applications reported for fetal mutation detection in
`
`maternal plasma, such as achondroplasia, Huntington’s disease, cystic fibrosis, and
`hemoglobin E (Nasis, O., Thompson, S., Hong, T., Sherwood, M., Radcliffe, S., Jackson, L.
`& Otevrel, T. (2004) Clin. Chem. 50, 694-701; Saito, H., Sekizawa, A., Morimoto,T.,
`
`Suzuki, M. & Yanaihara, T. (2000) Lancet 356, 1170; Gonzalez-Gonzalez, M. C., Trujillo,
`
`M.J., Rodriguez de Alba, M. & Ramos, C. (2003) Neurology60, 1214-1215; Gonzalez-
`
`Gonzalez, M. C., Garcia-Hoyos, M., Trujillo, M. J., Rodriguez de Alba, M., Lorda-Sanchez,
`
`I., Diaz-Recasens, J., Gallardo, E., Ayuso, C. & Ramos, C. (2002) Prenatal Diagn. 22, 946-
`
`948; Fucharoen, G., Tungwiwat, W., Ratanasiri, T., Sanchaisuriya, K. & Fucharoen,S.
`
`(2003) Prenatal Diagn. 23, 393-396), most published data only involve case reports of
`
`isolated patients. Large-scale evaluation of analytical protocols for circulating fetal DNA
`
`discrimination has been limited. Reliable discrimination between the fetal and maternal DNA
`
`sequences would depend heavily on the analytical specificity of the assay system. The degree
`
`ofanalytical specificity required for accurate analysis is inversely related to the degree of
`
`genetic difference between thealleles of interest and the background DNA (Lo, Y. M. D.
`
`(1994) J. Pathol. 174, 1-6). Thus a need exists for methodsthat can reliably analyze such
`
`subtle genetic differences.
`
`[007]
`
`The prenatal assessment of autosomal recessive diseases based on fetal DNA
`
`analysis in maternal plasma presents another challenge. The manifestation of an autosomal
`
`recessive disease results fromthe inheritance of a mutantallele from each parent. Thus, an
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-4.
`
`autosomal recessive condition could either be confirmed prenatally through the
`
`demonstration of the inheritance of two mutantalleles, or could be excluded by the
`
`demonstration ofthe inheritance of at least one non-mutant allele. The current strategies look
`
`at exclusion. For example, one such strategy is based on the haplotype assessment of
`
`polymorphisms associated with the paternally-inherited non-mutantallele (Chiu, R.W.K.et
`
`al. Clin Chem 48, 778-780 (2002)).
`[008]
`B-thalasseinia is an autosomalrecessive condition resulting from the reduced or
`absent synthesis ofthe B-globin chains ofhemoglobin.It is highly prevalent in the
`Mediterranean, the Middle East, the Indian subcontinent and Southeast Asia (Weatherall, D.J.
`& Clegg,J.B. Bull World Health Organ 79, 704-712 (2001)). More that 200 B-thalassemia.
`
`mutations have been described, many of which are. point mutations (Weatherall, D. J. (1997)
`
`BMJ 314, 1675-1678). $-thalassemia major is an otherwise lethal condition where survivalis
`
`dependent on life-long blood transfusions and iron chelation therapy. Curative therapies are
`
`not readily available and therefore, much focus has been devoted to disease prevention
`through prenatal diagnosis.
`|
`
`[009]
`
`The alpha and beta loci determine the structure of the 2 types of polypeptide
`
`chains in adult hemoglobin, Hb A. Mutantbeta globin that sickles causes sickle cell anemia -
`
`(http://www.ncbi.nlm.nih.gov/entrez/dispomim). Absenceof the beta chain causes beta-zero-
`
`thalassemia. Reduced amounts of detectable beta globin causes beta-plus-thalassemia, which
`
`is one of the most commonsingle gene disorders in the world.
`
`[010]
`
`For clinical purposes, beta-thalassemia is divided into thalassemia major
`
`(transfusion dependent), thalassemia intermedia (of intermediate severity), and thalassemia
`
`minor (asymptomatic). Patients with thalassemia major present in the first year of life have
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-5-
`
`severe anemia; they are unable to maintain a hemoglobin level above 5 gm/di. Clinical details
`
`of this disorder have been detailed extensively in numerous monographs and are summarized
`
`by Weatherall, et al. (The hemoglobinopathies. In: Scriver, C.; Beaudet, A. L.; Sly, W.S.;
`
`Valle, D. (eds.) : The Metabolic and Molecular Bases of Inherited Disease. (7th ed.) New
`
`York: McGraw-Hill 1995. Pp. 3417-3484). The prognosis for individuals with beta-
`
`thalassemia is very poor. For example, in 2000 it was reported that about 50% of U.K.
`patients with beta-thalassemia majordie before the age of35 years, mainly because
`conventional iron-chelation therapy is too burdensome for full adherence (Modeletal.
`
`Survival in beta-thalassaemia major in the UK: data from the UK Thalassaemia Register.
`
`Lancet 355: 2051-2052, 2000).
`
`[O11]
`
`The molecular pathology of disorders resulting from mutations in the nonalpha-
`
`globin gene region is the best known,this elucidation having started with sickle cell anemia
`
`in the late 1940s. Steinberg and Adams reviewed the molecular defects identified in
`
`thalassemias: (1) gene deletion, e.g., of the terminal portion of the beta gene (2) chain
`
`termination (nonsense) mutations; (3) point mutation in an intervening sequence; (4) point
`
`mutation at an intervening sequence splice junction; (5) frameshift deletion; (6) fusion genes,
`e.g., the hemoglobins Lepore; and (7) single amino acid mutation leading to very unstable
`
`globin, e.g., Hb Vicksburg (beta 75 leu-to-0) Steinberg, M. H.; Adams,J. G., IIL:
`
`Thalassemia: recent insights into molecular mechanisms. Am. J. Hemat. 12: 81-92, 1982.
`
`{012]
`
`Because ofthe frequency of the mutations in the populations and the devastating
`
`clinical symptoms including the markedly reduced life span, prenatal diagnosis is important.
`
`For example, it can provide a means for disease prevention. However, the conventional
`methods ofprenatal diagnosis such as, amniocentesis, chorionic villus sampling and
`
`cordocentesis, are all associated with a small but finite risk of fetal loss. Therefore, it would
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-6-
`
`be important to develop a non-invasive method for prenatal diagnosis of thalassemias.
`
`Attempts have been made in the past to develop other means of non-invasive prenatal
`diagnosis of6-thalassemia, including the analysis offetal cells in maternal blood (Cheung,
`
`M.C., Goldberg, J.D. & Kan, Y.W. Nat Genet 14, 264-268 (1996)). However, these methods
`
`are labor-intensive and time-consuming. Consequently, the need exists to develop tools that
`
`accurately permit highly specific and sensitive detection of nucleic acids in biological
`
`samples, particularly parentally inherited alleles.
`
`SUMMARY
`
`[013]j
`
`Accordingly, the present invention is directed to methodsof detecting nucleic
`
`acids in a biological sample.
`1014]
`Weshowthe feasibility of the use ofmass spectrometric analysis for the
`
`discrimination of fetal point mutations in maternal plasma and developed an approach for the
`
`reliable exclusion of mutations in maternal plasma. We further show the feasibility of the
`approachfor the minimally invasive prenatal diagnosis in a situation where a mother and
`
`father share an identical disease causing mutation, a concurrence previously perceived as a
`
`challenge for maternal plasma-based prenatal diagnosis for autosomal recessive diseases.
`
`[015]
`
`In one embodiment, the invention is directed to a method for the detection of
`
`paternally-inherited fetal-specific B-thalassemia mutations in maternal plasma based on
`
`methods for looking at nucleic acid segments using methods such as the primer-extension of
`
`polymerase chain reaction (PCR) products, at about single molecule dilution. This is
`
`preferably followed by mass spectrometric detection. The technique allows the non-invasive
`
`prenatal exclusion of B-thalassemia with high throughput capacity and is applicable to any
`
`disease caused by mutations in a single gene including, but not limited recessive single gene
`
`diseases such as thalassemias, such as beta thalassemias, cystic fibrosis, and congenital
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-7-
`
`adrenal hyperplasia. The invention is also useful in detection oftumor-derived DNA
`
`mutations isolated from cells in the plasma of a cancer patient, and detecting donor-derived
`
`DNA in the plasmaofa transplant recipient.
`
`[016]
`
`The invention is based upona discovery that a highly sensitive and specific
`
`mutation-specific analysis of the paternally-inherited mutation in maternal plasma can be
`
`used to exclude the fetal inheritance of the paternal mutation based onits negative detection.
`
`For example, using a real-time quantitative allele-specific polymerase chain reaction (PCR)
`
`approach to exclude the inheritance of the B-thalassemia mutation codons (CD) 41/42(-
`
`CTTT), involving the deletion of four nucleotides (CTTT) between codons 41 and 42 ofthe
`
`6-globin gene, HBB (Chiu, R.W.K.et al. Lancet 360, 998-1000 (2002)), showsthat the
`
`negative exclusion proposedherein can readily be used.
`[017]
`To achieve single nucleotide discrimination at low fractional concentrations, an
`
`analytical system that combines the use of an approach with better allele-specificity and high
`
`detection sensitivity is required. One exampleis the use of primer extension analysis in a-
`
`system such as the MassARRAY system (SEQUENOM),that allows a high throughput
`
`approach for the detection and exclusion of paternally-inherited fetal mutations in maternal
`
`plasma with the capability of single base discrimination. The MassARRAYsystem is based
`
`on matrix-assisted laser desorption ionization/time-of-flight (MALDI-TOF) mass
`
`spectrometric (MS) analysis of primer-extension products (Tang, K.et al. Proc Natl Acad Sci
`
`US A 96, 10016-10020 (1999)),
`
`[018]
`
`In one embodiment, the invention is directed to a method of detecting a genetic
`
`disorderin a fetus from a blood, serum or plasma sample of a pregnant woman, the method
`
`comprising: a) analyzing both isolated maternal and paternal DNAfora disease-causing
`
`mutation for the single gene disorder; b) if both maternal and paternal nucleic acid, e.g., DNA
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-8-
`
`carry a disease causing mutation for the same disease then isolating the nucleic acid,e.g.,
`
`DNAfrom plasma,blood, or serum of the pregnant mother; c) determining a fetal genotype
`
`from the isolated maternal plasma DNA using primerscorrespondingto the paternally
`identified mutation and performing a mutation-specificprimer-extension assay in at least two,
`
`preferably several replicates, for example 3, 5, or about 10, 12, 15, 20, 25-100 replicates and
`
`even up to about 1000 replicates. Most preferably about 15-25 replicates are used, wherein a
`
`detection ofthe paternal mutation in any ofthe replicate sample is indicative ofthe presence
`
`of the single-gene disorderin the fetus.
`(01 9]
`In one preferred embodiment, the single gene disease is an autosomal recessive
`
`disease. In the most preferred embodiment, the autosomal recessive disease is selected from
`
`beta thalassemias, cystic fibrosis and congenital adrenal hyperplasia. In the most preferred
`embodiment, the disease is beta thalassemia caused by mutations selected from the group
`
`consisting of CD 41/42 -CTTT; IVS2 654 (C->T); nucleotide -28 (A->G); and CD 17 (A-
`
`>T).
`
`[020] Inthe preferred embodiment the numberofreplicatesis at least two, preferably at
`
`least about 3, 5, 10-25, or 25-100, up to at least about 1000 replicates. Most preferably the
`
`numberof replicates is about 10-25.
`[021]
`In one embodiment, the primer-extension analysis is performed using the
`
`MassARRAYsystem (SEQUENOM).
`
`[022}
`
`Alternatively, the invention provides a methodofdetecting a genetic disease in a
`
`fetus using maternally isolated DNA from plasma, serum, or blood, the method comprising:
`
`a) selecting one or more single nucleotide polymorphisms (SNP) which are not disease-
`
`causing polymorphisms and which are associated either with a paternal disease-causing allele
`
`or with a paternal healthy allele and which SNP differs between the maternal and the paternal
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-9-
`
`genotype; b) determining the fetal genotype from a sample DNAisolated from the plasma,
`
`serum or whole blood of the pregnant mother, wherein the determination is performed using
`
`primers corresponding to both the selected SNP andthe disease-causing mutation and
`
`performing an SNP and disease mutation-specific primer-extension assay in several replicates
`
`using said primers; c) wherein detection of the SNP associated with the paternal allele in any
`
`of the replicate samples is indicative of the presence of the paternal allele inherited by the
`
`fetus and the detection of the paternal disease-causing mutation in any of the replicate
`
`samples indicates detection of the genetic disease inherited bythe fetus, wherein detection of
`
`the SNP associated with the healthy paternalallele in any ofthe replicate samplesis
`
`indicative of the presence of the healthy allele inherited by the fetus and excludes the
`
`inheritance of the genetic disease by the fetus.
`
`BRIEF DESCRIPTION OF DRAWINGS
`
`[023]
`
`Figure 1 demonstrates the relationship between the numberofpositives detected
`
`in the 15-replicate PCR experiments for fetal gender determination (y-axis) and the fetal
`
`DNAconcentration measured by real-time quantitative PCR targeting the Y-chromosome
`
`gene, SRY (x-axis). The theoretical basis of using the 15-replicate format is based on the
`
`Poisson distribution of fetal DNA molecules at single molecule concentration. The equation
`
`for the Poisson distribution is:
`
`[024]
`
`P(n)=
`
`
`m Hn e”
`al
`
`, where, n= numberoffetal DNA
`
`molecules per PCR, P(n)=probability of n fetal DNA molecules in a particular PCR; m=
`
`mean numberof fetal DNA molecules in a particular plasma DNA sample.
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-10-
`
`[025]
`
`Using our standard plasma DNA extraction protocol (see text), each PCR using I
`
`LL of maternal plasma DNA contains 0.4 genome-equivalent of the paternally-inherited fetal
`
`DNA. Thus, the probability that a particular PCR will be negative due to having nofetal
`
`DNAmoleculeis:
`
`[026]
`
`0-04
`
`P(0) = wie = 0.670
`
`[027]
`
`To reducethe false-negative rate, the probability that all replicates are negative for
`
`a 15-replicate PCR experimentis: 0.670'° = 0.0025.
`
`[028]
`
`Figures 2A and 2B show the Mass Spectrometric analyses of the SRY DNA and
`
`the paternally inherited thalassemia TVS2 654 mutation. Mass spectra for the other three
`
`thalassemia mutations are similar. For all mass spectra, Mass (x-axis) represents the
`
`molecular weight of the marked peaks. The molecular weights ofall relevant peaks are
`
`calculated before the analysis and the Mass values measured by mass spectrometry are .
`
`generally only 0-5 Daltonoff. Intensity (y-axis) is of an arbitrary unit. P and PP represent
`unextended primer and pausing product(i.e., premature termination ofthe base extension
`
`reaction), respectively. For SRY DNA analysis, the SRY peak is present (thus a positive
`
`result, marked as POS at the left side of the figure) in some of the 15 replicates (see Fig. 1).
`
`None of the 15 replicates has a SRY peak (thus a negative result, marked as NEG in the
`figure), ifa women was pregnant with a female fetus. For thalassemia IVS2 654 mutation
`
`analysis, the pT peak is from the paternally inherited thalassemia IVS2 654 mutation and is
`
`present in some ofthe 15 replicates for fetuses carrying a paternally inherited thalassemia
`
`IVS2 654 mutation (see Table 1). The nP peakis from all other S-globinalleles except the
`
`paternally inherited thalassemia IVS2 654 allele.
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-11-
`
`[0299
`
`Figure 3 shows a schematic illustration of the single allele base extension reaction
`
`(SABER) and standard MassARRAYassays. Maternal plasma detection of the paternally
`
`inherited fetal-specific B-thalassemia mutation, IVS2 654 C OT,is presented as an
`
`ilfustrative example. Maternal plasma is first amplified by PCR. The PCR products are
`
`subjected to base extension by the standard and SABERprotocols. The standard protocol
`
`involves the base extension ofboth the mutantfetal allele (7 allele) and the backgroundallele
`
`(C allele), whereas the SABER method only extends the fetal-specific mutant allele. The base
`
`extension reactions are terminated by dideoxynucleotides, indicated in boxes. The extension
`
`products of the standard protocol include a predominance of the nonmutantallele (open
`
`arrows) with a small fraction of the fetal-specific mutant allele (filled arrows). The low
`
`abundanceofthe fetal allele (filled peak) is overshadowed by the nonmutantallele (open
`
`peak) on the mass spectrum. Because SABERonly involves the extension of the mutant
`
`allele, the latter’s presence (filled peak) can be robustly identified from the mass spectrum.
`
`The striped peaks represent the unextended primer.
`
`{030]
`
`Figures 4A-4D show the MS analyses of the paternally inherited B-thalassemia.
`
`IVS2 654 mutation in maternal plasma. For all mass spectra, mass (x axis) represents the
`
`molecular weight of the marked peaks. The expected molecular weights ofall relevant peaks
`
`werecalculated before the analysis. Intensity (y axis) is in arbitrary units. P and PP,
`
`unextended primer and pausing product(i.c., premature termination of the base extension
`
`reaction or incorporation of an undigested dGTP from shrimp alkaline phosphatase treatment
`
`for the wild-type DNA template), respectively. Figs 4A and 4B illustrate the mass spectra
`
`obtained by the standard MassARRAYprotocol for a fetus negative and positive for the
`
`mutation, respectively. T, expected mass of the mutantallele; C, position ofthe alleles
`
`without the IVS2 654 mutation. Figs 4C and 4D illustrate the mass spectra obtained by the
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-12-
`
`SABER MassARRAYprotocol for a fetus negative and positive for the mutation,
`
`respectively. IVS2 654, expected mass ofthe mutantallele.
`[031]
`Figure 5 shows Table 1, showing prenatal exclusion of B-thalassemia major by
`
`maternal plasma analysis. AJI of the parents are carriers for B-thalassemia and have one HBB
`
`mutation. The “mutations of the father and mother are marked by “F’ and “M”, respectively.
`
`The maternal mutation is not indicated for cases where the maternal mutation is not one of
`
`the four HBB mutations studied. The fetal genotypeis indicated by the inheritance ofthe
`paternal mutation “F’, the maternal mutation Me", or the normalallele ‘“*’’. Results of the
`MassARRAY maternal plasmaanalysis is indicated by the “numberofreplicates among the
`15 repeats where the paternally-inherited fetal allele was positively detected. The fetusis
`
`deemed to have inherited the paternal mutation if any of the 15 repeats showed a positive
`
`result.
`
`[032]
`
`Figure 6 shows Table 2 including PCR and extension primer sequences. *CCT
`
`mix is ddCTP/ddGTP/ddTTP/dATPin which dd indicates the 2’,3’-dideoxynucleoside.
`
`Similarly AC mis is ddATP/ddCTP/dGTP/dTTP.
`
`[033]
`
`Figure 7 shows Table 3 including data from detection of paternally inherited HBB
`
`mutations in maternal plasma. All the patients are carriers of B-thalassemia and have one
`HBB mutation. The maternal mutation is not indicated for cases where the maternal mutation
`
`is not one of the four HBB mutations studies. F and M, mutations of the father and mother,
`
`respectively; -, no mutation; neg, negative; pos, positive; N.A. not applicable. The fetal
`
`genotype determined by conventional methodsis indicated by the inheritance of the paternal
`
`mutation F, the maternal mutation M,or the normalallele, *.
`
`[034]
`
`Figure 8 shows Table 4 including data from haplotype analysis of paternally
`
`inherited alleles in maternal plasma. Neg, negative; Pos, positive; N.A., not applicable. ¢G
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-13-
`
`and C denote the rs2187610 allele linked to the mutant or wild-type paternal HBBalleles,
`
`respectively. { The fetal genotype determined by conventional methodsis indicated by the
`
`inheritance of the paternal mutation F, the maternal mutation M,or the normalallele, *.
`
`DETAILED DESCRIPTION
`
`[035]
`
`The present invention is directed to methods of detecting nucleic acids in a
`
`biological sample. The method is based on a novel combination of a base extension reaction,
`
`which provides excellent analytical specificity, and a mass spectrometric analysis, which
`
`provides excellent specificity. The method can be used, for example, for diagnostic,
`
`prognostic and treatment purposes. The method allows accurate detection ofnucleic acids
`
`that are present in very small amounts in a biological sample. For example, the method of the
`
`present invention is preferably used to detect fetal nucleic acid in a maternal blood sample;
`
`circulating tumor-specific nucleic acids in a blood, urine or stool] sample; and donor-specific
`
`nucleic acids in transplant recipients. In another embodiment, one can detect viral, bacterial,
`
`fungal, or other foreign nucleic acids in a biological sample.
`
`[036]
`
`The methods provided are minimally invasive, requiring generally, a small
`
`amount of a biological sample, for example, a blood, plasma, serum, urine, bucchal or nasal
`
`swap,saliva, skin scratch, hair or stool sample from an individual.
`
`[037]
`
`In the case of determining a fetal genotype or quantitating the fetal nucleic acids
`
`or alleles using the methodsofthe present invention, the sample can be any maternal tissue
`
`sample available without posing a risk to the fetus. Such biological materials include
`
`maternal blood, plasma, serum, saliva, cerebrospinal fluid, urine or stool samples.
`
`[038]
`
`In the present study, we evaluated, and showthe feasibility of, the use of mass
`
`spectrometry (MS)for the discrimination of fetal point mutations in maternal plasma and
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-14-
`
`developed an approachfor the reliable exclusion of mutations in maternal plasma. We further
`
`evaluated, and show the'feasibility of, the approach for the noninvasive prenatal diagnosis of
`
`a motherand father sharing an identical disease causing mutation, anoccurrence previously
`
`perceived as a challenge for maternal plasma~-based prenatal diagnosis for autosomal
`
`recessive diseases.
`
`[039]
`
`The methodsof the present invention are automatable. For example, use of mass
`
`spectrometry, such as MassARRAY system (Sequenom Inc, CA), in combination with the
`present invention allowsanalysis of fetal DNA with the capacity of over 2000 samples per
`day in triplicate samples thus making the method a practical system for routine use.
`
`[040]
`
`In one preferred embodiment, the invention provides an accurate method for
`
`determining differences between fetal and maternal nucleic acids in a maternal blood sample
`
`allowing for a minimally invasive and reliable method for prenatal diagnosis. The method is
`
`based on a combination of a base extension reaction and a mass spectrometric analysis.
`
`Thus, prenatal diagnosis can be performed without the potential complications to the fetus
`
`and the motherthat are associated with traditional methodsfor prenatal diagnosis including
`
`amniotic fluid and/or chorionic villus sampling.
`
`[041]
`
`Dueto the specificity of the base extension reaction, allelic differences can be
`
`accurately amplified for analysis including changed varying from single nucleotide variations
`
`to small and large deletions, insertions, invertions and other types of nucleic acid changes that
`
`occur in even a small percentage of the pool of nucleic acids present in a sample.
`
`[042]
`
`The base extension reaction according to the present invention can be performed
`
`using any standard base extension method. In general, a nucleic acid primer is designed to
`
`anneal to the target nucleic acid next to orclose to a site that differs between the different
`
`alleles in the locus.
`
`In the standard base extension methods,all the alleles present in the
`
`

`

`WO 2005/023091
`
`PCT/US2004/028857
`
`-15-
`
`biological sample are amplified, when the base extension is performed using a polymerase
`
`and a mixture of deoxy- and dideoxcynucleosides correspondingtoall relevant alleles. Thus,
`for example; ifthe allelic variation is AIC, and the primeris designed to anneal immediately
`
`before the variation site, a mixture of ddATP/ddCTP/dTTP/dGTPwill allow amplification of
`
`both of the alleles in.the sample, if both alleles are present. Table 2 in Figure 6 shows
`
`exemplary mixtures for the standard base extension reactions for detecting several different
`
`nucleic acid variations in the HBB locus.
`
`[043]
`
`The After the base extension reaction, the extension products including nucleic
`
`acids with A and C in their 3’ ends, can be separated based on their different masses.
`
`Alternatively, if the ddNTPsare labeled with different labels, such as radioactive or
`
`fluorescentlabels, the alleles can be differentiated based on the label. In a preferred
`
`embodiment, the base ex