The PDF was saved in your download folder or opened in a new tab in your browser. You can find the download folder on your drive or in the browser bar of Firefox and Safari at the top right next to the arrow icon.
Sequence alterations have classically been divided into two categories: "mutation" and "polymorphism", with only the former considered relevant or pathogenic. However, the rapid increase in sequencing data from healthy and a wide variety of diseased tissues has illustrated that there is a broad spectrum between clearly disease-associated pathogenic alterations and nonpathogenic polymorphisms.1 This can be illustrated by the BRAF gene. The typical alteration (V600E) allows the use of targeted therapies; this is not the case if it is another atypical alteration in the same gene. Tiering in the evaluation may also be necessary when we see a change for the first time and thus have no reference data.
Various professional societies have therefore developed recommendations for oncological questions.2 We are following the most widely used 4-tier system worldwide3 and have adapted it to central questions in hematological diagnostics:
Can a diagnosis, prognosis or therapy be currently derived?
Does the change prove the clonality of hematopoiesis?
Can the change be used as a progression marker?
This involves translating our internal data from the last 15 years, up to four databases, in silico predictions, and artificial intelligence4 into a classification that can support clinical decisions.
Positive/Tier 1: Known pathogenic or, based on the type of molecular genetic aberration, very likely pathogenic change. Relevance to diagnosis, prognosis, and/or therapy depends on disease entity.
A large proportion of RUNX1 or TP53 alterations belong to this category, as they are associated with poor prognosis in many entities. Other examples in this category include alterations that allow targeted therapy (see above, BRAF), cause resistance to such therapy (e.g., alterations in the BTK gene whenIibrutinib is administered), or allow diagnosis according to WHO (e.g., NPM1 mutation to diagnose AML with NPM1 mutation). However, the significance for diagnosis, prognosis, and therapy must always be evaluated in the overall context of all findings obtained, and the respective disease-specific interpretation of the molecular genetic findings must be considered.
Positive/Tier 2: A somatic alteration is likely, so that an association with the disease is assumed. Due to the type of molecular genetic alteration and based on the data provided, no definite statement on molecular pathogenicity (functional relevance) can be made at the time of analysis.
Even though compared to tier 1 the data status is often weaker, tier 2 alterations may well give a direction in the diagnostic process and prove the clonality of hematopoiesis. We consider interpretations in the overall context with other findings to be particularly essential here. For genes in which typically only certain "hotspot" mutations occur (e.g., JAK2 V617F, BRAF V600E, and MYD88 L265P), the significance of tier 2 alterations is unclear; in particular, they cannot be seen as targets of targeted therapy without functional studies. However, tier 2 alterations can usually be used as progression markers.
Variant/Tier 3: Based on current data, it is unclear whether this is a rare congenital germline alteration or an acquired somatic alteration associated with the disease.
Congenital alterations are present in healthy and malignant cells. Matching with germline material may be helpful. A buccal swab and a piece of fingernail are suitable for this purpose. If the variant is not detectable there, then we assume that it is acquired and therefore only present in the malignant cell and its precursors. Only such acquired variants are suitable as progression markers, since they should no longer be present after successful therapy.
Polymorphism/Tier 4 (negative evaluation): These alterations occur frequently in the population, so that no clinical/pathological relevance is assumed according to the current state of knowledge.
They are by far the most frequent alterations in our daily life, are documented internally but not listed in the report.
1Baer C, Walter W, Hutter S, et al. "Somatic" and "pathogenic" - is the classification strategy applicable in times of large-scale sequencing? Haematologica 2019;104:1515-20.
2Leichsenring J, Horak P, Kreutzfeldt S, et al. Variant classification in precision oncology. Int J Cancer 2019;145:2996-3010.
3Li MM, Datto M, Duncavage EJ, et al. Standards and Guidelines for the Interpretation and Reporting of Sequence Variants in Cancer: A Joint Consensus Recommendation of the Association for Molecular Pathology, American Society of Clinical Oncology, and College of American Pathologists. J Mol Diagn 2017;19:4-23.
4Hutter S, Baer C, Walter W, Kern W, Haferlach C, Haferlach T. A Novel Machine Learning Based in silico Pathogenicity Predictor for Missense Variants in a Hematological Setting Blood 2019;134.
To conduct molecular genetic analyses DNA or RNA are isolated from cells, according to the kind of analysis required. RNA is subsequently transcribed into cDNA.
DNA or RNA are extracted from the aliquots of cells previously isolated from the patient’s submitted sample material (e.g. bone marrow or peripheral blood).
In an initial step, cells are mixed with lysis buffer in order to isolate nucleic acids (DNA or RNA). The buffer disrupts the plasma membrane of the cells, releasing and concomitantly stabilizing nucleic acids. Then, magnetic beads that bind DNA or RNA molecules on their surface are added. Using a magnet the nucleic acid-coated magnetic beads can be immobilized on the vessel wall, facilitating separation from the surrounding liquid. Subsequently, remaining proteins and cellular debris are entirely removed by a number of washing steps and the purified nucleic acids are eluted from the magnetic beads. The entire process is performed in a fully-automated manner with up to 96 samples in parallel.
The DNA can directly be subjected to molecular genetic analysis, e.g. via NGS. In contrast, isolated RNA first requires transformation into significantly more stable cDNA (complementary DNA) in an additional step termed reverse transcription. The viral enzyme reverse transcriptase – an RNA-dependent DNA polymerase – transcribes RNA into DNA. Similar to every other polymerase, reverse transcriptase requires a primer as point of synthesis initiation. For this purpose, either gene-specific primers or a mix of random primers (oligonucleotides composed of six randomly assembled nucleobases) are used. In this manner, a DNA-RNA hybrid strand is produced in an initial reaction. Subsequently, the RNA strand is enzymatically degraded and the resulting single-stranded cDNA serves as template for a DNA-dependent DNA polymerase to be transformed into a double-stranded molecule and further amplified according to the principles of classic PCR (see section PCR). Compared to genomic DNA cDNA misses introns and is thus best suited for detection of fusion transcripts, since break points are often spaced far apart on genomic level hindering analysis. Further, usage of cDNA allows for expression analysis of individual genes.
Hematological diagnostics faces significant challenges due to the rising diversity of clinically relevant molecular markers. Extremely sensitive, parallel analysis of hundreds of thousands of genome regions became possible with the introduction of modern, high-throughput sequencing methods, a.k.a. next-generation sequencing (NGS). This opens up new opportunities in mutation analysis and for progress control of hematopoietic neoplasias. For instance, NGS is used for BCR-AB1 mutation analysis among CML patients or to prepare comprehensive genetic profiles during diagnosis and prognosis of AML or MDS.
NGS testing can be conducted using any material. Between 5 and 10 ml of bone marrow with heparin, EDTA or citrate anticoagulant, i.e. peripheral blood, are needed for the test. Peripheral blood is sufficient as a test material, provided that malignant cells have passed into the blood. Therefore, bone marrow should be tested if passage has not occurred.
Next Generation Sequencing permits a staggering parallelization of the sequencing process. Methods based on amplicon deep sequencing or targeted enrichment are particularly suitable for routine diagnostics. Depending on the entity, the methods can sequence relevant gene segments with a high degree of sensitivity (1–3% mutational load). Panel testing permits the analysis of several hundred genes or genetic hotspots in one test cycle in less than a week. The findings are assessed using complex data processing and sequencing variants are then compared with SNP and mutation databases.
Several modern NGS platforms are available from routine diagnostics, for instance at Illumina or Thermo Fisher.
Detection of fusion transcripts via PCR requires cDNA, which is synthesized from RNA isolated from a patient’s sample material. Five to ten ml of bone marrow or peripheral blood anticoagulated by addition of either EDTA, heparin or citrate are required for molecular genetic analyses. In most cases, peripheral blood is sufficient as sample material, provided it contains neoplastic cells at detectable levels.
Aside from the target sequence containing the region to be amplified, a set of two so called oligonucleotide primers designed to flank the sequence of interest are required. The primers serve as points of initiation for a polymerase – a thermostable enzyme capable of producing a reverse complement copy of a single-stranded molecule upon addition of nucleotides (ribose triphosphates of adenine, guanine, cytosine and thymine).
A PCR is divided into three steps, which are characterized by different temperatures. Initially, a denaturation step is performed at about 95°C to separate the double strand into two single strands. During the following annealing step oligonucleotide primers bind the single-strand. The temperature is adjusted according to the base composition of both primers. During the final step, the so-called elongation, the polymerase synthesizes a new strand based on the single-stranded template. This step is conducted at 72°C, which is the thermal optimum of the enzyme. Subsequently, the denaturation step follows and the cycle is repeated 35 - 45 times, depending on the respective assay. Under optimum conditions the amount of pre-existing PCR products is doubled during each cycle. Following amplification, the PCR products are analyzed via capillary electrophoresis, during which the amplicons pass through a thin, polymer-containing capillary at a speed determined by their length. Application of a concomitantly separated internal size standard allows estimation of amplicon length. The presence of a PCR product of expected size serves as proof of the existence of a distinct fusion transcript.
Quantitative real-time PCR is another method for the amplification and detection of PCR products. Unlike the other detection methods, it is not a qualitative endpoint analysis, and is instead measured in real time during propagation of the PCR. Analysis of the logarithmic phase of this amplification permits precise quantification of the target sequences in the test material.
Real-time PCR testing can be conducted using any material. Between 5 and 10 ml of bone marrow with heparin, EDTA or citrate anticoagulant, i.e. peripheral blood, are needed for the analysis. Peripheral blood is sufficient as a test material, provided that malignant cells have passed into the blood. Therefore, bone marrow should be tested if passage has not occurred.
The method is based on introducing fluorescence-marked probes to the reaction mixture in addition to the specific primers required for PCR. During PCR, they hybridize with the continuously propagating amplification products and therefore emit fluorescence signals that are detected by the optical unit. Therefore, provided the specific target sequence is present, a rise in fluorescence intensity will occur directly during PCR. The point (PCR cycle) at which fluorescence above background levels is detectable for the first time correlates with the number of molecules that can be identified in the input material. The number of identified molecules of the target gene are benchmarked against a constant gene or transcript (normalized), which permits calculation of the number of malignant cells that the input material contains. Special PCR equipment fitted with optical units is necessary in order to conduct this detection method. While this method can also be used for the normal identification of a PCR product, real-time PCR reveals its true strength in particular during progress testing, e.g. during treatment, as it provides an opportunity for precise quantification and hence is used as a tool to measure treatment success. Real-time PCR also achieves extremely high sensitivities, so that – depending on the type of mutation and the input material – one malignant cell can be detected in 104 to 105 the number of healthy cells.
Neben der quantitativen real-time PCR ist die digitale PCR (dPCR) eine alternative Methode zum Nachweis und zur Quantifizierung einer spezifischen Zielsequenz. Sie eignet sich besonders gut für Verlaufskontrollen von Patienten unter Therapie, um den Erfolg bzw. das Versagen sensitiv bestimmen zu können, ohne eine Standardisierung zu benötigen.
Die digitale PCR kann aus jedem Material durchgeführt werden. Für die Analyse werden 5 bis 10 ml antikoaguliertes (Heparin, EDTA, Citrat) Knochenmark bzw. peripheres Blut benötigt. Peripheres Blut ist als Untersuchungsmaterial ausreichend, sofern eine Ausschwemmung maligner Zellen vorliegt. Entsprechend sollte bei fehlender Ausschwemmung Knochenmark untersucht werden.
Die digitale PCR beruht auf der Partitionierung des PCR-Ansatzes. Dies kann entweder in Tröpfchen (droplets) oder array-basiert erfolgen. Die eingesetzte DNA wird zufällig in die einzelnen Reaktionsräume verteilt, sodass einige keine und andere eine oder mehrere Kopien der Zielsequenz enthalten. Anschließend erfolgt die Amplifikation der Zielsequenz mittels Endpunkt-PCR. Die Fluoreszenzintensität der einzelnen Reaktionsräume wird erfasst, um den Anteil der positiven Partitionen zu bestimmen. Hiermit wird die Ausgangskonzentration der Zielsequenz berechnet. Die Anzahl an Kopien mit der spezifischen Mutation wird zu derer mit entsprechender Wildtyp-Sequenz ins Verhältnis gesetzt. Da die Quantifizierung mittels des ja/nein Prinzips geschieht, ist keine Kalibrierung und Standardkurve notwendig und es kann die absolute Kopienanzahl des amplifizierten Genabschnitts im Ausgangsmaterial bestimmt werden.
Bei dem in unserem Labor verwendeten droplet dPCR-System (ddPCR) erfolgt die Kompartimentierung des PCR-Ansatzes in bis zu 20 000 Einzelreaktionen. In Wasser-Öl-Tröpfchen findet die Amplifikation der Zielsequenz statt. Die Sensitivität beträgt in der Regel zwischen 0,005 - 0,05% je nach Zielsequenz.
Die Fragmentanalyse (auch Genescan genannt) ist ein fluoreszenzbasiertes, molekulardiagnostisches Analyseverfahren zur Längenbestimmung von Nukleinsäuresequenzen. Sie ermöglicht neben der hochauflösende Detektion von Größenunterschieden bei PCR Amplifikaten (z.B. Insertionen, Deletionen, Duplikationen, Fusionsgenen) die quantitative Bestimmung einer Mutation im Verhältnis zu ihrem gesunden Allel. Im Rahmen der molekulargenetischen Diagnostik werden Fragmentanalysen zur Identifizierung von prognostischen und krankheitsauslösenden Mutationen bei verschiedenen hämatologischen Neoplasien verwendet. Darüber hinaus finden Fragmentanalysen bei der Chimärismusanalyse sowie zur Klonalitätsanalyse Anwendung.
Die Fragmentanalyse kann aus jedem Material durchgeführt werden. Für die Analyse werden 5 bis 10 ml mit antikoaguliertes (Heparin, EDTA, Citrat) peripheres Blut bzw. Knochenmark benötigt. Peripheres Blut ist als Untersuchungsmaterial ausreichend, sofern eine Ausschwemmung maligner Zellen vorliegt. Entsprechend sollte bei fehlender Ausschwemmung Knochenmark untersucht werden.
Die Fragmentanalyse beruht auf dem Funktionsprinzip der Polymerase Kettenreaktion (PCR). Im Gegensatz zur klassischen PCR werden für die Fragmentanalyse sequenzspezifische Primerpaare eingesetzt, von denen einer der beiden Primer mit einer Fluoreszenzmarkierung versehen ist. Die anschließende Auftrennung der PCR-Amplifikate erfolgt in einem Kapillarsequenzer. Durch das Hinzufügen eines fluoreszenzmarkierten Längenstandards ist im Gegensatz zur Gelelektrophorese eine Größenunterscheidung von PCR-Produkten selbst dann möglich, wenn sich diese nur um ein Basenpaar voneinander unterscheiden. Darüber hinaus ermöglicht die gleichzeitige Verwendung von mehreren mit unterschiedlichen Fluoreszenzfarbstoffen markierten Primerpaaren in einem Multiplex PCR Ansatz die parallele Größenbestimmung von verschiedenen Zielsequenzen.
Clonality analysis is a procedure in molecular diagnostics applied for detection of lymphoproliferative disorders. During B- and T-cell development a genomic rearrangement (termed V-(D)-J-joining) of gene sections encoding the antigen receptors on B- and T-cells occurs. Physiological polyclonal lymphatic hematopoiesis comprises a variety of different lymphatic cell clones, which differ regarding their antigen specificity and the genomic rearrangement of the V-(D)-J region. In patients suffering from lymphoproliferative disorders, physiological polyclonal hematopoiesis is suppressed and dominated by uncontrolled proliferation of one (monoclonal), two (biclonal) or several (oligoclonal) malignant lymphatic cell clones. Via PCR fragment analysis size and quantity of genomic rearrangements in the V-(D)-J region can be investigated and the presence of a dominant cell clone within the lymphatic cell population can be determined. However, analysis solely based on molecular genetics does not allow any statement about malignancy of the cell clone and rarely is capable unambiguous discrimination between mono-, bi- or oligoclonal cell populations.
Apart from establishing diagnosis of lymphoproliferative disorders, conducting clonality analyses with a sensitivity of approximately 5% allows monitoring of therapy success over the course of disease. Further, based on the results of clonality analysis a patient-specific quantitation of minimal residual disease (MRD) can be established to achieve a sensitivity of approximately 0.01-0.001%. This procedure exploits the fact that the base pair sequence in the genomic region of the B- and T-cell-receptor is highly specific for the respective cell clone.
Analysis can be conducted on any material. Five to ten ml of bone marrow or peripheral blood anticoagulated by addition of either EDTA, heparin or citrate are required for molecular genetic analyses. In most cases, peripheral blood is sufficient as sample material, provided it contains neoplastic cells at detectable levels.
Methodologically clonality analyses are based on PCR fragment analysis, which is a high-resolution molecular diagnostic procedure for selective amplification and size-based separation of nucleic acids. In clonality analysis genomic V-(D)-J gene rearrangements are amplified in a sequence-specific manner in a multiplex PCR, employing primer pairs labelled with fluorescent tags of different color. The resulting amplicons are separated on a capillary sequencer. A fluorescently tagged size standard allows identification of clonal V-(D)-J gene rearrangements by their color and amplicon size, as well as discrimination between clonal rearrangements and polyclonal background.
In hematology chimerism analysis is performed to detect and quantify donor and recipient hematopoiesis following allogeneic stem cell transplantation. The term “chimera” describes an organism harboring DNA of two different organisms. This is the case, when a patient’s hematopoietic system originates from a different person. Monitoring chimerism allows early recognition of graft failure or relapse.
Whenever a person’s entire hematopoietic cells originate from a donor, their status is considered complete chimerism. Should hematopoietic cells of donor as well as of recipient be detectable, mixed chimerism is diagnosed. Upon analysis of whole bone marrow or peripheral blood a sensitivity of 2-5% is achieved with this method.
Five to ten ml of bone marrow or peripheral blood anticoagulated by addition of either EDTA, heparin or citrate are required for the analysis. Five to ten ml of bone marrow or peripheral blood anticoagulated by addition of either EDTA, heparin or citrate are required for molecular genetic analyses. In most cases, peripheral blood is sufficient as sample material, provided it contains neoplastic cells at detectable levels.
Additionally, at the minimum one of the following materials is required once for a patient as reference:
- The patient’s bone marrow or peripheral blood prior to grafting
- The patient’s buccal swab or nail material
- A sample of the donor (e.g. peripheral blood)
Chimerism analysis is performed by determining short tandem repeat (STR)-polymorphisms. Short tandem repeats are short sequence repeats of only a few base pairs (STR-motif), which are mainly located in non-coding DNA regions. The number of repeats at a defined position (locus) of DNA can vary among different individuals. The length of both STR-alleles of an individual’s investigated loci are determined via PCR-based amplification and subsequent fragment analysis. The result is indicated as the number of repeats of the STR-motif. Applying a multiplex PCR setup, several of these STR-loci can be amplified simultaneously. The combination of the number of sequence repeats at distinct loci provides an individual genetic profile. An STR-locus, i.e. a marker, is considered informative if the recipient harbors at least one allele that does not display the same length as the donor allele.
Array technology is a useful addition to classic cytogenetics. Array-based copy number analyses have become increasingly important in recent years. Application of this methodology permits whole genome, high-resolution detection of gains and losses in chromosomal material, including cytogenetically cryptic microdeletions and microduplications. It can also identify and characterize breakpoints and unknown fusion partner genes for unbalanced translocations. Screening for copy-neutral loss of heterozygosity (CN-LOH) is also possible. Frequently encountered synonyms for genomic array analyses include array CGH (array-based comparative genomic hybridization) and SNP array (single nucleotide polymorphism-based array).
Given that the detection of balanced translocations or intrachromosonal rearrangements is not possible without copy number changes, array technology is a useful addition to chromosome analysis and fluorescence in situ hybridization (FISH), although it cannot replace the other two techniques. Array technology is used as a complementary method, above all for patients with cytogenetically normal findings or extremely complex karyotype mutations.
Table 1: Advantages and limitations of genomic array analyses
Insight into gains and losses in the whole genome
Balanced rearrangements cannot be detected
Independent of malignant cell proliferation
Small clones (<15%) cannot be detected safely
Very high resolution
Copy-neutral LOH are detectable
Genomic array analysis involves the hybridization of patient DNA on an array. Additional hybridization of reference DNA may be necessary, irrespective of the technology used. Resolution of the genomic array varies depending on the particular array platform; the array we currently use provides resolution of approximately 100 kb. After around 48 hours of hybridization, the arrays are scanned and then characterized for mutations in the patient genome using a variety of bioinformatic algorithms.
In addition, the genomic array platform used in MLL research projects permits the identification of regions featuring copy neutral loss of heterozygosity (CN-LOH) for the validation of genomic gains and losses. Early studies have demonstrated that CN-LOH regions of tumor cells are frequently encountered in AML, myeloproliferative neoplasias and myelodysplastic syndromes and that they are associated with loss-of-function mutation for genes in the affected CN-LOH region. Neither chromosome analysis nor FISH have the capability to detect CN-LOH.