AMERICAN COLLEGE OF MEDICAL GENETICS
Standards and Guidelines for Clinical Genetics Laboratories
2006 Edition

 
       G: CLINICAL MOLECULAR GENETICS

These Standards and Guidelines specifically refer to the use of molecular techniques to examine heritable or somatic changes in the human genome.
 

G1 Specimens and Records
 
G1.1 In addition to numerical accession files and alphabetical patient listings, each family studied is assigned a unique code (preferably numeric). Note: This requirement only applies when more than one member of a family is being tested. (See G17.2 for maintaining confidentiality in reporting of results.)
 
G1.2 For required patient information, see C2.4.
 
G1.3 For specimen labeling, see C2.1.
 
G1.4 A judgment about specimen quality should be made at intake. Any problems related to specimen collection (tubes, anticoagulants, transport solutions, etc.) or quality (lysis, clotting, etc.) must be noted.
 
G1.5 It is recommended that the nomenclature guidelines set forth by the Ad Hoc Committee on Mutation Nomenclature (Hum Mutat 1996; 8: 197-202) be followed.
 
G2 General Quality Control
 
G2.1 See C4.3. In addition, for molecular testing, quality of reagents can be evaluated prior to introduction into testing or at test outcome. However, any reagent which is used at points in a protocol that would lead to complete specimen loss or destruction (e.g., DNA preparation) must be tested prior to introduction. In-house testing can be deferred or delegated to manufacturers' quality control testing, where appropriate. Critical reagents are determined at the discretion of the laboratory director and may include but are not limited to:
a) cell lysis buffer
b) Protease (proteinase K)
c) Restriction digestion buffers
d) Enzymes: restriction or polymerizing
e) Electrophoresis buffers
f) Southern blotting solutions
g) Hybridization membranes
h) Probes (in general)
i) Probe labeling buffer or reagents
j) Oligonucleotide primers
k) Hybridization solution
l) Sequencing reagents and solutions
m) PCR reagents and solutions
G3 DNA Preparation
 
G3.1 DNA preparation must be done by accepted protocols as evidenced by the published literature. Complete references should be included in standard operating procedure manuals.
 
G3.2 Southern analysis calls for DNA of higher quantity and quality than that required for PCR.
 
G3.3 Isolated material must be stored at 4° C or frozen. Excess sample material should be stored at a temperature no higher than 0-5° C to ensure long-term stability.
 
G3.4 The requirements for DNA preparations used for PCR analysis are less rigorous than those of DNA for Southern analysis. However, appropriate controls must be used in the analysis to ensure that the DNA can serve as a suitable template for DNA amplification.
 
G4 Probe/Primer/Locus Documentation

All loci used for analysis in the laboratory need to be well documented by Human Gene Mapping Workshop, Geneatlas, the Genome Data Base (GDB) or by publication in the peer-reviewed scientific literature. This documentation must be maintained in an up-to-date laboratory book and include the following: genome location, linkage data, literature references, cloning vector, cloning site, size of insert, enzyme used for the detection of the RFLP, the sizes of the alleles and any constant bands, the allele frequencies in each racial or ethnic group for which this information exists, new mutation rate (if known), how the probe was prepared as well as hybridization and wash conditions. For oligonucleotide probes or primers, documentation sheets also must include specific sequences. For primers, PCR conditions and the size of the expected positive result should be included. There must be internal documentation that the probe/primer used is consistent with the above data (i.e., a photograph indicating that the size of the insert isolated from the vector is the correct size or that the conditions used by the laboratory produce the appropriate result).
 

G5 Assay Validation
 
G5.1 Each laboratory must validate the analytical performance characteristics (sensitivity, specificity, reproducibility) of the technique chosen for analysis of each gene. Validation with well characterized samples is critical. Where available, performance characteristics should be compared with an existing "gold standard" assay. In the absence of "gold standards" for comparison of results of new assays, the splitting of samples with another laboratory with an established clinical assay may be considered. Documentation of validation results must be available for review (see section C8).
 
G5.2 The laboratory must document clinical validity through its own or other published studies.
 
G6 Southern Analysis
 
G6.1 Restriction Digestion and Electrophoresis
 
G6.1.1 Restriction endonuclease digestion of prepared DNA for Southern analysis must be done according to a standardized protocol that will be documented in the laboratory manual.
 
G6.1.2 Quality control of restriction digests must be done by one of the following:
a) Run a test gel prior to electrophoresis. If incomplete, redigest the specimen.
b) Evaluate the analytical gel by visually comparing to size markers or to the patterns of all DNAs on the gel, including controls, for consistency of satellite bands as well as high and low molecular weight bands.
G6.1.3 Each test must include human DNA control(s) with a documented genotype at the locus tested.
 
G6.1.4 All Southern gels should include size markers to assist in the reading of the alleles. External markers may be excluded if appropriate heterozygotes or "all allele" controls are used.
 
G6.2 Membrane Preparation
 
G6.2.1 Prior to transfer, the Southern gel must be photographed to provide a hard copy documentation of the gel.
 
G6.2.2 The method of transfer must be documented in the laboratory manual with appropriate references. Efficiency of transfer must be validated and documented either at time of transfer or at the end of the study.
 
G6.3 Hybridization
 
G6.3.1 Hybridizations must be carried out by accepted procedures and documented with appropriate references.
 
G6.3.2 Hybridization can be checked by scoring the known controls included on the Southern filter.
 
G6.3.3 For those markers new to the laboratory, a previously used filter, if available, on which the DNA has been cut with the appropriate enzyme (or a test DNA of known genotype), shall be used as further quality control of the hybridization.
 
G6.3.4 The laboratory must retain a representation of the primary data (gel, film, autoradiograph, etc.) demonstrating the reported hybridization pattern.
 
G7 PCR Methodologies
 
G7.1 Containment: Avoidance of contamination and cross-contamination of specimens is of utmost concern. Since amplified products are the main source of contamination, it is recommended that:
a) All handling of post-amplification products be in a defined area with dedicated pipettes.
b) Positive displacement pipettes or tips containing filters should be used.
c) Pipettes and all reagents and solutions used for diagnostic PCR procedures should be dedicated to that use.
G7.2 Primer Documentation: See Section G4.
 
G7.3 System Validation
 
G7.3.1 Amplification
 
G7.3.1.1 To assure PCR product specificity, all reaction conditions (reagents and thermocycling parameters) must be established for each test system. Reaction conditions must provide the desired degree of PCR product specificity.
 
G7.3.1.2 When amplification of a variable length sequence is assayed, the system should be tested with DNAs from individuals representing large and small amplification products to evaluate the impact of differential amplification.
 
G7.3.1.3 Amplicons developed for use in multiplex PCR reactions must be thoroughly assessed for compatibility prior to use in clinical testing.
 
G7.3.2 PCR Product Detection and Analysis
 
G7.3.2.1 Detection systems (visual, restriction site, allele specific oligonucleotide (ASO) hybridization, etc.) employed in diagnostic testing are being rapidly adapted from established research and diagnostic protocols. Such systems should be well documented and published. The laboratory must demonstrate that a level of specificity characteristic of the selected detection system has been attained internally and that the level of specificity is adequate for detecting the expected products. Adequate care must be taken to guard against failure to detect PCR products.
 
G7.4 Controls and Standards
 
G7.4.1 Appropriate positive and negative controls to provide evidence of amplification in general and amplification of specific sequences must be included. A blank containing all reaction components except the DNA must be included in all PCR assays as negative control. Positive controls must include individuals of known genotype for the locus being tested. Controls for various types of assays are as below:
 
G7.4.1.1 Assays based on presence or absence of PCR products must include known control primers yielding a positive result to check for proper amplification and sizing of the PCR products and to ensure that a negative result is accurate. This should include a positive result with control primers detecting a spiked additive or a constitutive component.
 
G7.4.1.2 When specimens are analyzed for sequence variation (RFLP sites, mutation specific sites, etc.) controls must be included.
 
G7.4.1.3 Assays in which the result is based on fragment size (VNTRs, microsatellites, etc.) must include size markers (sequencing ladders, etc.) covering the range of expected results during gel electrophoresis.
 
G7.4.1.4 Assays based on changing of electrophoretic mobility (homo/heteroduplex analysis, single strand conformation analysis, etc.) must include appropriate controls to ensure correct interpretation of results. Any unexpected results require repeat of assay. Procedures for analysis of possible new mutations should be available.
 

G8

Linkage Analysis
 
G8.1 Linkage analysis should employ software in wide general use. It should be used only by individuals with a working knowledge of the specifics of each package in use.
 
G8.2 The laboratory must keep an up-to-date reference list documenting linkage relationships (i.e., location relative to locus in question, recombination fractions and/or q values at 95% confidence intervals) for each disorder analyzed by indirect linkage methods. The laboratory must have documented linkage relationships for all in-house generated probes prior to use in a clinical setting (see G4).
 
G8.3 In order for linkage analysis involving probes with significant recombination distances from the locus in question to be reported, the analysis must contain data from two informative flanking markers. If this is not possible, the reason must be stated so as to indicate that every effort was made to provide such.
 
G8.4 For linkage analyses involving probes with negligible recombination distances from the locus in question, it is only necessary to use one highly informative marker.
 
G8.5 For each disease specific system in use, the number of informative markers to be used is dependent upon the informativeness of each marker, the disease specific recombination frequency and the availability of markers.
 
G9 Denaturing Gradient Gel Electrophoresis (DGGE) Assays
 
G9.1 Overview

Strand length and conformation determine relative electrophoretic mobility of double stranded DNA in a polyacrylamide gel. Several techniques use this characteristic as a method of identifying DNA sequence abnormalities without prior knowledge of the precise location or nature of the sequence change. DGGE makes use of the conformational changes associated with DNA double strand melting as a method for detection of sequence variations. Under DGGE conditions a double stranded DNA sequence is electrophoresced through a gradient of denaturant at an elevated temperature. The mobility of the DNA is affected by the melting behavior of the sequence as it progresses through the increasing denaturant concentration. It is possible in this manner to differentiate between the mobility of two sequences which differ by as little as a single base.

DGGE uses PCR to generate copies of gene or cDNA segments of several hundred basepairs in length. Each of these is denatured and allowed to renature under conditions that promote heteroduplex formation between the normal sequence strand and the strand with a possible mutation (most patients are assumed to be heterozygous for any unknown mutation). The heteroduplexed fragments are then separated by electrophoresis in polyacrylamide gels containing denaturants that facilitate the melting of the DNA duplexes at unique positions in the gradient. Fragments containing sequence variations will generate multiple bands, while homozygous normal (or homozygous abnormal) fragments will generate only a single band.

The sensitivity of DGGE can reach 100% when sufficient knowledge and experience with the methodology and the gene of interest are available. Variations of basic DGGE such as two-dimensional DGGE have been developed and may provide increased sensitivity. In the event that a large deletion resulting in the heterozygous loss of one or more amplicons is present, an incorrect interpretation of wild type sequence may occur. This disadvantage is shared with all mutation detection techniques. Knowledge of the distribution of mutation types in the gene of interest will permit evaluation of the sensitivity of DGGE for each gene of interest.

The high detection rate of DGGE is dependent on correct design of the assay. Several factors outlined below are of importance in the design and performance of DGGE.
 

G9.2 PCR Fragment Design
 
G9.2.1 All sequences to be analyzed by DGGE should be amplified by PCR using protocols optimized for the amplicon in question. The specificity of the PCR reaction should be such that a single amplicon is seen on a stained gel.
 
G9.2.2 Each amplicon should be designed using available software or empiric analysis to produce a single melting domain throughout the region to be assessed.
 
G9.2.3 The primers used in the amplification step should be designed to include a 5 clamp sufficient to stabilize the melting domain of the test DNA sequence.
 
G9.3 Sample Preparation
 
G9.3.1 DNA samples should be prepared and stored using established protocols (see DNA preparation section, G3).
 
G9.3.2 Amplification of target sequences should be performed using all standard PCR precautions (see PCR section, G7).
 
G9.3.3 Samples should be heated and allowed to renature prior to loading to permit heteroduplex formation. Time and temperature should be standardized.
 
G9.3.3.1 If a potential homozygous mutant condition is being analyzed, it may be appropriate to mix a known normal control and test sample to force heteroduplex formation.
 
G9.4 Gel Electrophoresis
 
G9.4.1 Appropriate denaturing gradient conditions should be established based on calculated melting profile and empiric results observed with positive controls.
 
G9.4.1.1 A set of positive controls should include (whenever possible) samples containing mutations distributed throughout the region to be analyzed.
 
G9.4.2 Equipment used to form the gradients in the gels and to run the gels under temperature controlled conditions should be standardized within each laboratory. Any change in equipment will require a re-validation of the assay.
 
G9.4.3 Samples to be run on the same gel should be denatured, renatured, and loaded on the gel at the same time.
 
G9.4.4 A positive control sample should be analyzed simultaneously to provide a measure of the adequacy of the heteroduplex formation and the gel running conditions. A negative (normal) control sample can be used to aid in sizing of the observed bands.
 
G9.4.4.1 It is not necessary to run a sample of every known mutation in each gel. A single mutation control is sufficient to document the reproducibility of the system.
 
G9.5 Data Analysis
 
G9.5.1 Gels should be stained (or visualized based on labeled DNA) in a manner adequate to detect the entire banding pattern created.
 
G9.5.1.1 Heteroduplexes are often present in smaller amounts than the homoduplex forms and may produce a lighter signal.
 
G9.5.2 Samples on the gels should be identified by an unambiguous method clearly identifying positive and negative controls.
 
G9.5.3 Documentation of gel results by photography or other image storage system is necessary.
 
G9.5.3.1 Computerized image analysis may be helpful in identification of recurring mutations.
 
G9.5.4 The presence of putative mutations identified by DGGE must be confirmed by sequencing.
 
G9.6 Validation
 
G9.6.1 Each laboratory must validate the technique for each sequence to be analyzed. Validation with known mutations as well as normal samples is required. Results of validation studies for each gene analyzed must be available for review.
 
G10 Protein Truncation Tests for Mutation Detection
 
G10.1 Overview

The protein truncation assay uses RNA (or DNA in the case of large exons) to produce a PCR amplified modified cDNA. The cDNA is then placed in a linked transcription/translation system to produce a protein product that can be analyzed by gel electrophoresis to identify abnormally sized products. Protein truncation analysis can be used to search for possible mutations in a gene of interest. Knowledge of the proportion of previously identified mutations known to result in a truncated protein product must be available before use of this methodology in a clinical setting can be considered.

This assay system is very complex and each gene analyzed will present a unique set of challenges. Therefore, extensive experience with each gene is required before application of the assay to clinical use.
 

G10.2 Source of Samples
 
G10.2.1 The starting material for this assay is DNA or RNA obtained from any tissue by standard methodology (see DNA/RNA preparation section). RNA is the material of choice unless one or more large exons provide a useful target for analysis.
 
G10.2.2 The source tissue must be of sufficient quality to provide high molecular weight DNA or RNA. In addition, the tissue type must express the mRNA of interest in sufficient quantity for accurate and sensitive analysis.
 
G10.3 PCR Amplification of DNA or cDNA (see G10.4 for cDNA synthesis)
 
G10.3.1 The usual safeguards against contamination by PCR products should be used (see G7.1).
 
G10.3.2 The 5 primer is designed to introduce a bacteriophage promoter sequence and a mammalian translation initiation sequence (Kozak sequence) into the PCR product. It is not necessary to include a stop codon in the 3 primer since absence of a stop codon does not appear to influence the translation efficiency of PCR products failing to reach the natural stop codon.
 
G10.3.3 Although PCR products of at least 5 kb can be translated, it is recommended that multiple overlapping segments be amplified, each less than 2 kb with a minimum overlap of 200-300 bases. This minimizes the risk of missing mutations that are close to the primer sequences.
 
G10.3.4 Each PCR reaction should be run in duplicate or triplicate to avoid false identification of artifactual mutations arising through amplification of chance polymerase errors leading to production of truncated polypeptides.
 
G10.3.5 A normal control for the specific region of the gene to be analyzed must be included in each assay.
 
G10.4 RT-PCR Amplification from RNA (Reverse-transcription PCR)
 
G10.4.1 When RNA is the starting material, cDNA is first synthesized from the RNA using oligo (dT), random hexamer primers, or mRNA-specific primers.
 
G10.4.2 A second round of PCR using a nested primer pair may be necessary to amplify low abundance mRNA transcripts. PCR controls including a water blank must also be reamplified to permit detection of low-level contamination.
 
G10.4.3 Amplification by RT-PCR followed by electrophoresis may reveal gross rearrangements such as gene deletions (complete or partial), duplications, insertions or splice mutations without need for the protein truncation assay.
 
G10.4.4 Differences between the two alleles in terms of transcription efficiency or RNA stability can influence results. A genomic DNA control segment with a previously identified heterozygous sequence in the gene must be PCR-amplified in parallel to confirm that both alleles have been amplified in each patient sample.
 
G10.4.5 The quality of RNA should be documented by either gel analysis or by amplification of a housekeeping gene to ensure that it is an appropriate starting template.
 
G10.5 Coupled Transcription and Translation
 
G10.5.1 After amplification, the unpurified PCR product is added to the mixed components of a reticulocyte lysate system which enable transcription and translation to be accomplished.
 
G10.5.2 It may be necessary to optimize potassium salt concentration to overcome inappropriate translation termination.
 
G10.6 SDS-PAGE Electrophoresis
 
G10.6.1 Translation products are separated by discontinuous SDS-PAGE. Commercially available protein markers are usually used as molecular size standards. If the protein product of interest is very large, special standards may be required.
 
G10.6.2 A normal control must be run with each batch of test samples. Previously prepared (known product size) controls may be used as an external size indicator, but a simultaneously transcribed/translated control is also required.
 
G10.7 Interpretation
 
G10.7.1 A mutation is indicated by the presence of a novel band of lower-than-normal molecular weight representing a truncated peptide. If the band representing the full-length polypeptide is present in the same sample, it can serve as an internal control.
 
G10.7.2 "Background" bands are often observed. Some of these are artifactual, resulting from translation from internal AUG codons downstream from the authentic start codon or erroneous translation termination due to a non-optimized "in vitro" system (see G10.5.2). Other background bands present may represent proteins in the reticulocyte lysate or alternatively-spliced products from the gene of interest. Again, comparison of bands with those from a known normal control assayed simultaneously is essential.
 
G10.7.3 The presence of a truncated polypeptide is suggestive of an underlying genomic mutation. In most cases, the length of the truncated polypeptide (determined by using the protein markers as standards) can be used to localize the putative mutation. If the polypeptide is truncated due to a large deletion, the deletion site can be determined by restriction endonuclease mapping.
 
G10.7.4 The analytical specificity and sensitivity of the protein truncation assay is not known. It is essential to verify the presence of each mutation by either sequencing genomic DNA or sequencing cDNA followed by analysis of genomic DNA using RFLP or ASO methodologies.
 
G10.8 Validation
 
G10.8.1 Each laboratory must validate the technique for each gene to be analyzed. Validation with known mutations as well as normal samples is required. Results of validation studies for each gene analyzed must be available for review.
 
G11 Single-strand Conformation Polymorphism (SSCP) Assays
 
G11.1 Overview

Single-strand conformation polymorphism (SSCP) analysis is a method for detecting mutations and sequence polymorphisms in genes. SSCP is generally performed by denaturing PCR products and electrophorescing under nondenaturing conditions. The technique relies on the fact that single-strand DNA under certain conditions has defined secondary structure. The electrophoretic mobility of folded single-strand DNA molecules depends on both length and conformation. Mutations can alter the mobility of one or both single strands. Direct sequencing is performed after SSCP analysis to ascertain the nature of the sequence changes.

The sensitivity of SSCP is not 100%. Sensitivity depends on the size and sequence of the segment as well as the gel matrix utilized, the temperature, and the concentration of glycerol in the loading buffer. At present, there is no reliable way to predict the sensitivity of novel mutation detection, which typically varies from 50-90%. For segments of a given size under a given set of conditions, the sensitivity depends on the mobility of the wild type sequence relative to the distribution of mobilities of all the possible single base changes.

Each laboratory must determine its own sensitivity and specificity for each gene analyzed.

In order to increase the expected sensitivity of SSCP, two to four different conditions are sometimes employed. However, use of multiple conditions defeats the major advantage of the technique, speed.

Hybrids of SSCP and other methods have been developed in order to increase sensitivity. Three of these methods have the advantage of detecting virtually all mutations, as judged by blinded analysis. Dideoxyfingerprinting (ddF) is best for segments of 300 bp or less, bi-directional dideoxyfingerprinting (Bi-ddF) is best for segments of 300-600 bp, and restriction endonuclease fingerprinting is best for segments of 800-2000 bp.

When performing SSCP, attention to safeguards for PCR-based assays as described in Section G7 is required. Particular attention should be given to Section G7.3.1.2 (amplification of variable length sequences) to ensure amplification of the range of sizes possible at the locus, and Section G7.4.1.4 (changing of electrophoretic mobility) for correct interpretation of results. Additional considerations include:
 

G11.2 Assay Design
 
G11.2.1 When screening for unknown mutations, DNA fragments between 150 and 300 bp are typically used. Larger fragments can be used if it is known that the specific mutation/polymorphism of interest produces an abnormal SSCP pattern in that DNA segment.
 
G11.3 Polyacrylamide Gel Electrophoresis
 
G11.3.1 Gels should be run for a sufficient length of time (dependent on fragment length) to detect possible mobility shifts. In order to reduce the risk of missing mutations, samples may be run under two electrophoretic conditions that may differ in length of time, temperature, buffer concentration, crosslinking ratio, crosslinking reagents, and presence or absence of glycerol.
 
G11.3.2 It is preferable to standardize electrophoretic conditions for as many different mutations as possible. This can be done by using more than one control mutation (see below).
 
G11.3.3 SSCP requires a stable, uniform temperature throughout the gel. Unstable cooling (as occurs with cooling fans) may produce unreliable results.
 
G11.4 Controls
 
G11.4.1 A double-stranded DNA control should be run alongside single-stranded fragments to allow identification of both fragments.
 
G11.4.2 Some mobility shifts are observed only with double-stranded fragments.
 
G11.4.3 Optimal denaturation of double-stranded fragments should involve a dilution of the PCR product. This will necessitate use of a sensitive detection method (fluorescence, radioactivity, or silver staining).
 
G11.4.4 The PCR product from at least one normal control should be included on every SSCP gel.
 
G11.4.5 The PCR product from at least one control sample containing a mutation should be included on each SSCP gel in order to ensure that the electrophoresis conditions are optimal for detection of at least one mutation. Inclusion of more than one control mutation is advisable to improve the accuracy and standardization of the assay. If screening for several known mutations in a DNA fragment, use of control samples for each is desirable to ensure that the sequence alteration produces an abnormal SSCP band under the conditions used.
 
G11.5 Visualization of Results
 
G11.5.1 For manual approaches to SSCP using 32P-labeled or 33P-labeled deoxynucleotides, multiple X-ray film exposures are recommended to visualize all signals. Some abnormal SSCP bands may be faint, requiring longer exposures than normal bands.
 
G11.5.2 For SSCP by automated fluorescent analysis, internal size markers help prevent artifactual lane shifting from influencing mobility shift data. It may be necessary to adjust the volume of sample loaded to achieve detection.
 
G11.6 Interpretation of Results
 
G11.6.1 All samples showing a mobility shift should be sequenced to determine the nature of the sequence change. It is possible for different sequence variations to produce similar SSCP results.
 
G11.7 Validation
 
G11.7.1 Each laboratory must validate the technique for each gene to be analyzed. Validation with known mutations as well as normal samples is required. Results of validation studies for each gene analyzed must be available for review.
 
G12 Heteroduplex Assays
 
G12.1 Heteroduplexed double-stranded DNA molecules result from the annealing of complementary DNA strands containing base mismatch(es) due to a mutation or polymorphism in one of the strands. Regions of interest can be amplified, denatured, and allowed to reanneal to facilitate heteroduplex formation. Mutations or polymorphisms can be detected by differential migration of heteroduplexes vs. homoduplexes on acrylamide gels, presumably due to sequence-dependent conformational changes in double-stranded DNA. Sequence changes as little as a single-base substitution may be detected by heteroduplex analysis, depending on factors such as the type of base mismatch, the size of the PCR product, and the distance of the mismatch from the ends of the fragment. Gel matrices developed for heteroduplex analysis are available commercially (MDE) or have been described in the literature (CSGE), and isotopic or non-isotopic detection systems can be used. Heteroduplex analysis is a relatively simple technique to perform and has been applied successfully for numerous genetic disorders. Detection rates of approximately 80-90% have been reported for small DNA fragments (<300 bp), which is comparable to that of SSCP.
 
G12.2 PCR Fragment Design
 
G12.2.1 PCR product sizes of approximately 150-300 bp are ideal for screening unknown mutations by heteroduplex analysis. Larger fragments can be used to detect specific mutations or polymorphisms once it has been established that a heteroduplex band can be consistently detected under standardized conditions.
 
G12.2.2 The location of the mutation/polymorphism of interest should be at least 40-50 bases from the ends of the DNA fragments. Thus, PCR primers in flanking intron sequences should be at 40-50 bases from the intron-exon junctions.
 
G12.3 Sample Preparation
 
G12.3.1 The preparation and storage of DNA samples should be performed according to standard protocols (see DNA preparation section G3).
 
G12.3.2 PCR amplification of the regions of interest should be carried out according to all standard precautions (see PCR section G7). It is critical that each amplicon produce a clean, single band for use in heteroduplex analysis.
 
G12.3.3 Samples should be heat denatured and allowed to reanneal to facilitate heteroduplex formation. The time and temperature for denaturation and annealing should be standardized.
 
G12.3.4 In case of potential homozygous mutations, PCR products from wild type controls should be mixed, denatured and reannealed with the test samples to force the formation of heteroduplexes.
 
G12.4 Gel Electrophoresis
 
G12.4.1 The composition of the gel matrix to be used for heteroduplex analysis, the thickness of the gel, the length and time of the run, and the electrophoresis equipment should be standardized within each laboratory.
 
G12.4.2 Samples to be analyzed on the same gel should be denatured, reannealed and loaded on the gel run to validate the results for each gel.
 
G12.5 Data Analysis
 
G12.5.1 Heteroduplex gels should be visualized by staining or by autoradiography, depending on the detection system employed, to detect the entire banding pattern required for mutation detection. The detection system used to detect the heteroduplex bands (e.g., the specific staining protocol) should be standardized in each laboratory.
 
G12.5.2 Heteroduplex bands are usually seen at a lighter intensity because they comprise a stoichiometrically smaller amount of the total DNA sample.
 
G12.5.3 Results should be scored unambiguously by comparison with the positive and negative controls. All putative positive results detected by heteroduplex analysis should be confirmed by sequencing to identify the mutation or polymorphism involved.
 
G12.6 Validation
 
G12.6.1 The heteroduplex analysis technique should be validated by each laboratory where this assay is to be performed. Validation should be carried out using sequence variations (which should exhibit detectable and in many cases characteristic heteroduplex banding patterns for specific sequence variations), as well as normal control samples. For each gene analyzed by heteroduplex analysis, validation test results should be available for review.
 
G13 DNA Sequencing Analysis
 
G13.1 Overview

DNA sequencing is the "gold standard" for the analytic validation of all new DNA-based mutation testing. In addition, since it is capable of the exact determination of every base within a gene including the promoter and splice sites, it can be used to unambiguously determine the genotype of an individual. It is the method of choice for genes with a large number of rare mutations with each mutation seen in only one or a few families. Other methods such as SSCP, ASO or protein truncation assays are far more cost effective and should be applied prior to end-to-end sequencing. Clinically, DNA sequencing technology should be applied only when the gene in question is well characterized as follows:

a) The full and complete sequence is available in Genbank.
b) An initial and ongoing catalog (i.e., database) is widely available for the identification and location of "definitive" mutations (amino acid or stops or frameshifts) as well as polymorphisms.
c) The presence or absence of any pseudogene sequences complicating interpretation has been established.
G13.2 Sequence Standards
 
G13.2.1 Although the sequence assay shares elements in common with all other DNA diagnostic assays, there are unique concerns and areas that require separate address. Unique issues that arise in DNA sequence assays result from the large number of analytical points measured in each particular assay (i.e., the number of bases analyzed) and the relatively small signal strengths that are obtained from any base at any position. The technology for the generation of the sequence information is also generally complicated. Therefore, the sequence information must be verified and controlled at multiple points in the generation and interpretation of the sequencing data.
 
G13.2.2 One very positive aspect of the emerging use of sequencing for molecular diagnostics is that the likely errors will be biased very strongly towards the generation of false positives, rather than false negatives. This is a consequence of the fact that it is much easier to produce a sequence that looks as if it contains the wrong base(s) than a clear profile showing only the correct base. As each positive can and should be tested by an independent determination, this direction of bias is desirable. Potential for missing a heterozygous base substitution is a concern. To increase the sensitivity of heterozygote detection, both the sequencing chemistry and polymerase used should be optimized to produce uniform peak intensities in the case of fluorescent sequencing, since variations can result in false negatives. Both of these scenarios underscore the need to sequence both strands of the DNA region analyzed to optimize sensitivity and specificity of the assay.
 
G13.3 Methodologies
 
G13.3.1 Presently the most widely used method is Sanger dideoxy chain termination, which can be applied in several forms.
 
G13.3.2 Manual sequencing requires a radioactive label (32P, 33P or 35S) in one of the four dNTPs or at the 5 end of a sequencing primer. The disadvantages are low throughput, requirement for radioactivity, and lack of automation in base-calling, data collection, analysis, and records. Both manual and computer-assisted reading formats can be used, but computerized systems provide more accurate transfer of data.
 
G13.3.3 Fluorescent sequencing reactions can be performed using dye primers or dye-labeled primers or dye terminator chemistries and one of several polymerases. Data collection uses an imaging system and appropriate software.
 
G13.3.4 Automated fluorescent sequencing can be performed using automated sequencer formats providing automated gel running and data collection.
 
G13.3.5 Capillary gel electrophoresis for sequencing is in widespread use in genome centers and has advantages over gel-based systems with potential in diagnostics.
 
G13.4 DNA Preparation
 
G13.4.1 All normal concerns for collection and preparation of the DNA sample apply. The use of a commercially available DNA preparation kit is recommended to provide consistency in sample concentrations. However, well standardized in-house methods are also acceptable.
 
G13.5 PCR Amplification
 
G13.5.1 The length of the region to be sequenced in a single run must be limited. An upper limit of accurately readable bases exists for each methodology and gel apparatus type.
 
G13.5.2 The quantity of the DNA must be sufficient to generate adequate PCR product. This can be determined by meeting an expectation of PCR efficiency (e.g., an agarose or acrylamide gel separation of an aliquot of the PCR can be compared to a standard).
 
G13.5.3 The PCR product should be analyzed by gel and purified prior to the sequence reaction to ensure the highest quality of results.
 
G13.6 Sanger Sequencing
 
G13.6.1 Primers directed towards the end of the fragments are used.
 
G13.6.2 There are several chemistries available, but each should be aimed at providing the best possible sequence coverage of the fragment.
 
G13.7 Gel Electrophoresis
 
G13.7.1 Following the Sanger reaction, materials must be pooled (dye primer reactions) or purified from unincorporated materials. Normal care is needed to prevent sample mix-up.
 
G13.7.2 The tracking of individual samples on gels is a difficult and potentially error-prone step. Standard loading formats should be used to ensure this part of the process is accurate.
 
G13.7.3 Gel preparation using commercially available premixed solutions may provide additional quality control. If the supplier of the solutions changes, separation characteristics must be reevaluated.
 
G13.7.4 The characteristics of each gel apparatus/power supply combination are unique. Therefore timing, voltage requirements and separation characteristics must be standardized for each individual set-up.
 
G13.8 Primary Base Calling
 
G13.8.1 The overall quality of the sequence reactions must be monitored. The concern is that poor sequence reactions containing artifacts such as "stops," compressions, or "Ns" will be difficult to interpret and will result in the classification of normal bases as mutant or vice versa. Every effort should be made to resolve any such regions. Routine analysis of the opposite strand sequence will be useful for that purpose. The use of a different sequencing chemistry or polymerase may resolve specific regions, since artifacts may not occur in identical spots under alternate conditions. Currently available criteria include the number of positions at which computer base calling is not possible. A comparison of each test with a known standard (e.g., Genbank) is required, including judgment of peak height. (Caution should be exercised, since not all sequences in Genbank are correct.) Future options will include objective measurement of the base determination by statistically generated "quality factors."
 
G13.8.2 Manual re-reading of areas where the software has had difficulty should be performed with caution. The chromatograms of both the forward and reverse strands should be evaluated and the consensus compared to the standard sequence.
 
G13.8.3 Sequence analysis software is needed to compare data of the wild type and patient sample in both forward and reverse directions.
 
G13.9 Comparison of Sequence Data with a "Within Run" Standard
 
G13.9.1 The comparison with a standard of a high quality sequence from the same run is also needed to identify base differences.
 
G13.9.2 Verification of readings using second strand and/or second aliquot sequencing is required. Some mutations may be missed if sequencing is performed in only one direction. Any positives should be confirmed by sequencing a second aliquot. For direct sequencing, a second PCR amplification product should be used for repeat sequence analysis.
 
G13.10 Interpretation and Data Reporting
 
G13.10.1 Base differences are correlated with the known gene structure and other relevant data, and the likely effect of the base change on the gene is predicted. The laboratory must follow the ACMG Recommendations for Standards for Interpretation of Sequence Variations (Genet in Med 2000; 2(5):302-303).
 
G13.10.2 The report should note the exact base change and location by nucleotide position as referenced in Genbank and the corresponding position change in the protein using standard nomenclature.
 
G13.10.3 For small deletions and insertion or nonsense mutations resulting in a predicted protein truncation, the term "mutation" is appropriate.
 
G13.10.4 For missense alterations, one must consider whether these represent mutations, polymorphisms, or rare variants. For each genetic disease, the laboratory should first refer to a polymorphism and mutation database. If the base alteration has not been previously described, the nature and significance of the change may be unclear and should be stated as such in the report. For resolution, family studies and population based studies are appropriate.
 
G13.10.5 Reports in which no mutations are detected by sequence analysis should include multiple disclaimers, primarily that the sensitivity of the test is <100%. If sequencing was confined to the coding region of the gene, the possibility of mutations in the promoter or intragenic regions not covered by the test should be clearly stated. Sequencing will not detect large gene deletions or duplications. In addition, a mutation in a different gene that contributes to the disease, as well as misdiagnosis of the proband, constitute other possibilities.
 
G13.11 Considerations for Alternate Technology
 
G13.11.1 Alternate methods may be identical to the above, except that Sanger sequencing is not used. Note that methods that simply verify the normal sequence and show a difference if a mutation is present should not be considered as sequence assays if they do not enable the precise identification of the mutations. Examples of these are SSCP or hybridization of a target fragment to a selected array of oligonucleotides.
 
G13.11.2 For each technology, the requirement for the robust identification of normal and mutant sequence must be demonstrated. Controls and standards should be in place to ensure that a standard of quality is met for each determination.
 
G13.12 Validation

Each laboratory must validate the technique for each gene to be analyzed. Validation with known mutations as well as normal samples is required. Results of validation studies for each gene analyzed must be available for review.
 

G14

Microsatellite Based Analysis

DNA microsatellite markers have general utility in a variety of molecular genetic analyses (e.g., genotyping, linkage analysis, parent of origin/uniparental disomy studies, characterization of chromosome rearrangements).

Attention to safeguards for PCR based assays as described in Section G7 is required. Particular attention must be given to Sections G7.3.1.2 (amplification of variable length sequences) to ensure amplification of the range of sizes possible at the locus, and G7.4.1.3 (fragment size markers) for appropriate sizing. In addition:
 

G14.1 For manual approaches to microsatellite analysis using polyacrylamide sequencing gels and radioisotope detection (32phosphorous-labeled deoxynucleotides), multiple X-ray film exposures are recommended to obtain all possible autoradiographic signals.
 
G14.2 For manual approaches using 35sulfur or 33phosphorous-labeled deoxynucleotides:
 
G14.2.1 Gel drying may be necessary before autoradiography and should be standardized to avoid underdrying or overdrying, both of which may affect interpretation, e.g., through blurry bands or by gel cracking.
 
G14.2.2 Individual autoradiographic exposures are necessary.
 
G14.3 For manual approaches using blotting of polyacrylamide sequencing gels followed by chemiluminescent detection, blotting should be standardized to establish a minimal blotting time as well as times for optimal autoradiographic exposure.
 
G14.4 Microsatellite analysis by automated sequencing requires unique assay controls.
 
G14.4.1 Each sample lane on the polyacrylamide sequencing gel must have internal markers detectable by the automated detection system.
 
G14.4.2 All markers must be tested to determine optimal PCR sample amounts to be loaded (i.e., amplicon intensity must be within the sensitivity parameters of the detection system).
 
G14.5 Microsatellite data interpretation is similar for each use. However, care should be taken in interpretation due to the appearance of shadow-bands and variability in gel migration.
 
G15 Interpretation of Data
 
G15.1 All results must be read by two individuals (identified in records) independently, one of whom must be the director, laboratory technical supervisor or other qualified individual.
 
G15.2 All file materials relating to individual and/or family studies should be cross-referenced for accessibility.
 
G15.3 All questionable or inconsistent data must be resolved by additional analysis (reprobing the filter or digestion of another sample followed by reprobing on a filter with appropriate controls, etc.). The use of family members of known genotype as controls can assist in checking for nonpaternity or other problems.
 
G15.4 Dosage analysis is inherently less reliable than +/- assays. For carrier or mutation detection using dosage analysis, samples must be run on a filter containing appropriate intralane and interlane controls. Sufficient controls to ensure the accuracy of dosage estimation must be used. Densitometric scanning must be done to confirm dosage analysis that is initially based upon visual interpretation.
 
G15.4.1 The report should include caveats as to the inherent reliability of these tests.
 
G15.5 For PCR assays, care must be taken to assess the possibility of differential amplification.
 
G16 Records of Molecular Testing
 
G16.1 Scoring sheets must contain the following information (if applicable):
a) specimen numbers
b) locus names tested (probe name and locus identification)
c) test system used (Southern, PCR, etc.)
d) mutation detection system (RFLP, ASO, etc.)
e) enzymes used for RFLP analysis
f) alleles detected
g) results
G16.2 All results must be recorded on sheets which are retained and kept in the patient file, the family file and/or with the photographs or autoradiographs.
 

G17

Molecular Genetics Reports
 
G17.1 A report must be issued only to the referring physician or genetic professional. In general, the laboratory must not give results directly to the patient. See C3 for issues regarding record dissemination. Additionally, a report must be issued only to the referring physician, genetic professional, or referring center. In general, the report must include the following:
a) collection date
b) time/date of receipt in the laboratory
c) name of individual
d) date of birth
e) ethnicity where appropriate
f) laboratory identification number
g) date of report
h) reason for testing or disease locus tested
i) test performed, methodology, mutations tested
j) the genotyping and/or haplotyping established for the individual
k) a statement interpreting the data (interpretation should be understandable to a non-geneticist professional), including clinical implications, follow-up test recommendations, genetic counseling indications
l) documentation if a preliminary report has been issued
m) notation of any deviation from the laboratory's standard practice
n) signature of the laboratory director or technical supervisor or other authorized individual above his/her printed name.
o) a means to contact the laboratory director or designee
p) recommendations
It is recommended that along with this information, the following be included: the family/kindred number if one is assigned, and a pedigree with the genotype information indicating if this was a linkage study.
 
G17.2 Any report must ensure the confidentiality of the other family members whose studies were used to provide information to the proband. The format can be such that one copy is detailed and for the referring genetic expert, while a cover summary sheet is provided for the proband as long as no other family members' results are revealed.
 
G17.3 Investigative studies: A written report must be issued to the referring source and must contain the same information stated in Section G17.1. However, there must be a qualifying statement clearly indicating that the results are based on an investigational study and may, therefore, not be as accurate as a test recognized by the genetic community as an accepted or proven clinical service test.
 
G18 Denaturing High Performance Liquid Chromatography (dHPLC) (Section Added November 2003)
 
G18.1 Overview

Denaturing high performance liquid chromatography (dHPLC) can be used for rapid, automated, and high-throughput mutation detection based on principles similar to those for heteroduplex analysis. Recent advances in the development of this technology have led to the introduction of automated instruments. The software is useful in both predictions of the optimum run conditions based on the DNA sequence and analysis of the results in distinguishing homoduplexes and heteroduplexes. This technology is particularly suited for detection of point mutations, small deletions and insertions. It has also been applied for analysis of fragment size differences and for sensitive detection of sequence differences in minor cell populations such as tumors. The basic principle is that DNA is negatively charged, the column cartridge is neutral, and a positively charged binding ion--triethylammonium acetate (TEAA)--links the two. Heterozygous mutations are detected through differential binding of homo- and heteroduplexes to the column. Analysis is performed at a temperature sufficient to partially denature heteroduplexes. The melted heteroduplexes are resolved from the corresponding homoduplexes by HPLC. Denaturation leads to a reduced double-stranded PCR fragment. Single-stranded fragments elute earlier than double-stranded fragments due to the reduced negative charge. Thus, heteroduplexes elute prior to homoduplexes.

Sensitivity depends upon the size and sequence of the PCR fragment, in particular the melting profile, as well as the conditions of analysis, including temperature and buffer concentration. At present, there is no reliable way to predict the sensitivity of detection for novel mutations, which have been reported in various genes to exceed well over 90%. Nevertheless, diagnostic laboratories must validate the sensitivity of this detection method for each gene test developed. For each PCR fragment under a given set of assay conditions, the sensitivity depends on the elution profile of the wild-type homoduplex sequence relative to the heteroduplex with the sequence alteration. In order to increase the sensitivity of dHPLC, two or three different temperatures may be employed.
 

G18.2 PCR Fragment Design

PCR fragment design is critical to the success of dHPLC analysis. dHPLC can be used for fragments up to 600 bp; however, generally optimum separation is achieved with fragments of 200 to 400 bp. For PCR fragment design of regions of large size, it is recommended that overlapping sets of primers be used. It is suggested that the overlap region be a minimum of 50 bp. Prior to ordering oligonucleotide primers, the melting profile of the PCR fragment should be analyzed using the software of the instrument. If there are more than two melting temperatures of the sequence, it may be useful to break the fragment into smaller fragments in order to achieve a more accurate analysis. In some cases it may be necessary to use GC clamps, and in other cases it may not be possible to achieve optimum design based on problematic sequences. It is possible to design well defined small multiplex PCR reactions to analyze by dHPLC, but care must be taken in resolving the different PCR fragments, based on size variation, yet having consistent melting profiles, allowing the same optimized analysis conditions.
 

G18.3 Sample Preparation

DNA preparation is critical to the success of this assay. Some methods, such as certain column preparations, interfere with the binding to the cartridge and cannot be used. It is critical that the laboratory use a DNA preparation protocol that does not damage the cartridge. Therefore it is strongly recommended that each laboratory consult with the manufacturer for recommended DNA preparation kits, of which many exist.

PCR products are pipetted in 96-well plates and loaded on the instrument. Sample mixing is critical to resolve homozygous mutation carriers and for analysis of males for X-linked conditions. For individuals who are heterozygous for a sequence alteration, heating to 95°C and slowly cooling produces a mixture of heteroduplexes and homoduplexes. However, for detection of homozyotes, the PCR product from the patient is mixed with a comparable amount of wild-type PCR product in order to obtain heteroduplexes.
 

G18.4 Chromatography

dHPLC should be performed under optimized conditions to detect possible heteroduplexes. In order to reduce the risk of missing mutations, samples should be analyzed under optimized melting temperatures, which may be multiple, and may also require adjustment in buffer concentrations. The use of dHPLC-grade water or an equal grade is critical for this analysis system to operate efficiently. Any change in water source will require re-standardization of the column. In addition, it is important that the column be standardized at routine intervals (at least weekly) in order to assess reproducibility and quality of performance. The column should be monitored closely for number of analyses and replaced appropriately as recommended by the manufacturer. The software keeps track of column usage, which is a valuable quality control measure for diagnostic laboratories. It is important to recognize that the reproducibility of profiles is highly dependent upon the column and the number of runs. When columns are changed and when the number of runs on a column is high (>2000), profiles may also change. Therefore it is important to run mutation standards at regular intervals in order to determine test reproducibility. It is important that diagnostic laboratories monitor columns for reproducibility of results, and change columns when mutation-detection is compromised. This should be done at the discretion of the technical director.
 

G18.5 Controls

Both wild-type and positive mutation controls, including heterozygous samples (and homozygous samples when applicable, depending upon the test) must be analyzed along side test specimens. In particular, it is critical that the wild-type fragment is used for the basis of all comparisons. However, it is impossible when scanning large genes for unknown mutations to be able to validate each sequence variation prior to introduction of this method of analysis. Therefore one mutation in each fragment of interest is sufficient. In cases where the laboratory is unable to obtain mutations for all fragments to be analyzed either because they do not exist or are not available, the laboratory must develop the conditions for analysis of that fragment using the same high standards as all other fragments analyzed. When a positive control for a particular DNA segment cannot be obtained, it is critical that the laboratory use multiple analysis conditions in order to optimize detection of an unknown mutation. It is noteworthy here that each mutation in a given PCR fragment will have a characteristic elution profile of its heteroduplex. If a pattern variation is identified, the laboratory should confirm the variant by sequence analysis.
 

G18.6 Visualization of Results

The observation of heteroduplex peaks in a chromatogram indicates the presence of a sequence variant, while samples without base mismatches resolve as homoduplexes. Heterodupex peaks elute earlier than homodupexes, and can be observed as separate peaks or as shoulders on the leading edge of homoduplex peaks. The manner in which a heteroduplex peak resolves is influenced by the specific nucleotide mismatch present and the melting characteristics of the surrounding bases. Elution profiles that differ from the wild-type or reference DNA indicate the presence of sequence alterations in the form of base substitutions, deletions, or insertions. One cannot predict the type of mutation (i.e., deletion, insertion, nonsense, etc.) from the heteroduplex pattern. The software of the instrument allows real-time visualization of results. Software allows overlay of the patient specimen and the wild-type fragment for aided visual comparison. The software also automatically scores the profile for the presence of a heteroduplex. This automatic scoring must be confirmed by visual observation. Similarly, it is recommended that all "negative" profiles also be confirmed visually. The homoduplex wild-type pattern is typically one peak, but may be two peaks, depending upon the melting profile. It is desirable to optimize the fragment design to have a single peak in order to more readily distinguish wild-type patterns from heterozygous mutant. In addition, it is recommended that each patient specimen that shows a positive result be documented as a hard-copy printout and inserted in the laboratory record. Currently, mutation profiles are not recorded by the instrument's software in order to enable future comparisons via "pattern recognition." Therefore these mutation heteroduplex profiles always require manual observation. Future development trends may resolve this issue.

The instrument can utilize an ultraviolet detection system or a fluorescent detection system. However, at present only one fluorescent dye can be detected during a single analysis. The rationale for using fluorescence is to achieve more sensitive detection for minor populations or use in single cell PCR. Future trends will be to include a four-dye system in order to allow multiplex analysis of heteroduplexes.

Instrument maintenance is required at routine specified intervals and must be performed and documented.
 

G18.7 Interpretation of Results

All samples identified as heteroduplexes by dHPLC analysis must be sequenced in both directions to confirm and determine the nature of the sequence change. Each sequence change within a DNA fragment is predicted to have a unique heteroduplex pattern. It is recommended that a pattern file be established for quick identification of specific sequence changes. However, pattern recognition alone is not considered sufficient for diagnostic purposes, particularly when scanning genes for unknown mutations. In the case of a recurring mutation within a well characterized DNA fragment such as a targeted mutation test, pattern recognition alone may be sufficient for mutation identification. However, sufficient validation is required by the laboratory prior to introduction of such tests.

For samples in which no heteroduplex is identified in any PCR fragment tested, the report must state the sensitivity of this technique. The laboratory must then determine whether another method should be employed to supplement detection rate, such as sequence analysis, or whether to stop testing.
 

G18.8 Validation

Each laboratory must validate this technique for each sequence to be analyzed. Validation with known mutations as well as wild-type controls is required. Results of validation studies must be documented and available for review.
 

G19 Prenatal Testing (Section Added November 2003)
 
G19.1 Samples

Many genetic analyses are amenable to prenatal diagnosis using both direct and cultured cells from amniotic fluid (AF) and chorionic villi (CVS). However, in some cases one of these particular specimen types may be more appropriate. For each prenatal genetic test, the laboratory should determine the appropriate prenatal specimen and specify the amount of material required for testing. The laboratory should provide these requirements and appropriate instructions to referring centers and professionals. When the prenatal specimen is cultured cells, it is important that the laboratory culturing the sample maintain backup flasks until the molecular analysis is completed and reported. It is recommended that the mutation status of one or both parents, as appropriate, be tested prior to testing of fetal specimens, preferably within the same laboratory. To the extent possible, laboratories should have a follow-up program in place to monitor the accuracy of their prenatal testing.
 

G19.2 Sample Processing

As with other genetic tests, prenatal testing must be performed with the utmost level of caution to ensure accuracy of the predicted result. Laboratories should have procedures in place to assure accurate sample handling. If there is sufficient material and whenever possible, it is recommended that prenatal testing be performed in duplicate using DNA extracted from two separate specimens.
 

G19.3 Maternal Cell Contamination

The contamination of both direct and cultured cells from AF and CVS with maternal cells is well documented and therefore represents a potential source of error in prenatal diagnosis. Prenatal samples should be examined in parallel with a maternal sample to rule out error due to maternal cell contamination (MCC). Laboratories should understand how their testing methods are affected by the presence and the amount of MCC. For example, prenatal detection of a deletion using PCR, as is the case in testing for DMD and SMA, is expected to be more sensitive to maternal contamination, since a normal maternal allele could mask the deletion. A prenatal test using an allele-specific PCR reaction to detect a paternal RhD gene in the fetus of a RhD-negative mother is much less sensitive to maternal contamination.

For example, Chamberlain et al. (Nucleic Acid Res 1988;16:11141-11156) explored potential problems with maternal contamination in a multiplex PCR test for deletions in the dystrophin gene by mixing DNA from a partially deleted sample and a non-deleted sample. This study demonstrated that 3-5% contamination could be tolerated if the amplification cycles were limited to 25. In contrast, Hessner et al. (Am J Obstet Gynecol 1997;Feb;176(2):327-33) used similar mixing experiments to determine the impact of maternal contamination on prenatal testing for paternally inherited alloalleles using allele-specific PCR. In this situation, where the fetus is being tested for an allele that the mother does not have, the paternal alloallele could still be detected with more than 90% contamination. These two examples illustrate how the effects of MCC depend on the specific test and the method being used.

Laboratories should perform similar studies, when possible, and in the absence of this information should seek to confirm the test results from contaminated samples. The results may be confirmed from an alternate sample, if it is available. This may include a cultured sample prepared from original direct sample or an independent culture. If necessary, the obstetrician should be contacted about the possibility of an additional amniotic fluid sample.

The laboratory should have procedures in place to assess the presence and level of maternal cell contamination. These methods should detect, at a minimum, the level of contamination that would affect the test results. A combination of several polymorphic STR or VNTR loci is recommended for ruling out MCC. Batanian et al. (Genet Testing 1998;2:347-350) showed that two VNTR loci could be used to rule out MCC in 30/30 cases. However, some of these cases required a paternal sample to complete the testing for MCC. As a paternal sample may not be available, the laboratory should be able to complete the testing for MCC without the paternal sample. Therefore it is likely the laboratory will need at least 3 loci to resolve all cases. If a paternal sample is used, the laboratory should be aware that the MCC results might identify nonpaternity.

There are a number of marker systems suitable for MCC analysis. Many multiplex kits are commercially available that enable a number markers to be analyzed in one PCR reaction. These markers systems are also used to detect chimerism in hematopoetic stem cell transplant patients. A list of the marker systems being used in engraftment testing laboratories can be found in the Monitoring Engraftment Survey distributed by the CAP proficiency testing program. The validation of MCC assays should include sensitivity studies to determine if the appropriate levels of MCC can be detected.