DNA probe with locked nucleic acids and use for diagnosing chronic myeloid leukemia (CML)

The use of bridged nucleic acid probes in dPCR for MRD measurement in CML and ALL addresses precision and reproducibility issues, achieving a low limit of detection and eliminating false positives, enhancing diagnostic reliability.

WO2026238998A1PCT designated stage Publication Date: 2026-11-19UNIVERSITY OF HEIDELBERG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2026/064275
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-16
Filing Date
2026-05-15
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Current methods for measuring minimal residual disease (MRD) in chronic myeloid leukemia (CML) and acute lymphoblastic leukemia (ALL), such as qRT-PCR and digital PCR (dPCR), face challenges with precision, reproducibility, and comparability due to pipetting errors, signal rain, and the need for standard curves, leading to potential false-positive results and decreased sensitivity.

Method used

A method utilizing digital polymerase chain reaction (dPCR) with DNA probes comprising bridged nucleic acids (LNAs) of 12 to 18 nucleotides, specifically designed to bind to target genes like BCR::ABL1, allowing for improved signal separation and reduced false positives, achieving a limit of detection (LoD) of 0.001% or less.

Benefits of technology

The method provides enhanced accuracy and clear separation between positive and negative signals, ensuring a limit of blank (LoB) of 0.000%, thereby improving the reliability and reproducibility of MRD measurements, particularly in diagnosing CML and ALL.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2026064275_19112026_PF_FP_ABST
    Figure EP2026064275_19112026_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to a method for detecting a target gene in a biological sample, comprising carrying out a digital polymerase chain reaction (dPCR) with primers and one or more DNA probes targeting the target gene; and determining the presence and / or amount of the target gene in the sample, wherein at least one of the DNA probes is a core probe which comprises or consists of locked nucleic acids (LNAs) and has a length of 12 to 18 nucleic acids.
Need to check novelty before this filing date? Find Prior Art

Description

[0001]

[0002] May 15, 2026

[0003] DNA probe with bridged nucleic acids and use in the diagnosis of chronic myeloid leukemia (CML)

[0004] The present invention relates to a method for detecting a target gene in a biological sample. The present invention further relates to a DNA probe for use as a nuclear probe in the method. The present invention further relates to the use of the method for diagnosing chronic myeloid leukemia (CML), in particular for measuring minimal residual disease (MRD). The present invention further relates to a kit for carrying out the method.

[0005] State of the art

[0006] Chronic myeloid leukemia (CML) is a clonal myeloproliferative disorder of the hematopoietic system, characterized by the uncontrolled proliferation of myeloid cells in the bone marrow and their accumulation in the blood. The prevalence of CML worldwide is approximately 1-2 cases per 100,000 people per year, with incidence increasing with age.

[0007] Chronic myeloid leukemia (CML) is caused by the fusion of chromosomes 9 and 22 [t(9;22)] in hematological stem cells. The fused chromosome, known as the Philadelphia chromosome, produces the fusion gene BCR::ABL1, which is also found in acute lymphoblastic leukemia (ALL). BCR::ABL1 is a constitutively active tyrosine kinase that leads to uncontrolled cell proliferation and disrupts normal cell differentiation. This results in immature leukocyte precursor cells that impair normal blood function. In particular, the displacement of the physiological cell lines of the bone marrow leads to symptoms such as anemia, susceptibility to infections, and an increased tendency to bleed. The course of CML is typically divided into three diagnostic phases: In an initial chronic phase, which can last for several years, the vesicles multiply only slowly, and the symptoms are mild or absent altogether.The subsequent accelerated phase is a transitional phase in which the number of blisters in the blood and bone marrow increases, accompanied by more severe symptoms. The most advanced and severe phase of CML is the blast crisis, which resembles acute leukemia and in which the proportion of blisters in the blood and bone marrow exceeds 20%. The prognosis in this final phase is worse, and treatment options are limited.

[0008] Since the approval of tyrosine kinase inhibitors (TK1) such as imatinib, CML - provided it is diagnosed early enough - is a treatable disease with which patients can now lead a long and almost normal life.

[0009] Crucial for successful treatment of CML and ALL is achieving deep molecular remission. This is defined by low to undetectable expression of BCR::ABL1 in the blood. Patients who achieve this remission have a significantly better prognosis than those who do not. Measuring measurable residual disease (MRD) has proven indispensable for identifying these patients and tailoring therapy to their success.

[0010] In MRD measurements, the proportion of BCR::ABL1 fusion transcripts is determined in comparison to a suitable reference transcript (e.g., ABL1, GUSB, or BCR, with ABL1 being the most commonly used). The resulting ratio of BCR::ABL1 fusion transcript to reference transcript serves as a surrogate marker for the malignant cells (blisters) present in a patient sample. Different levels of deep molecular remission (also known as (M)MR, for (major) molecular response) are distinguished, which also incorporate the amount of detected reference transcripts as a marker for sensitivity.

[0011] WF / WF 241383WO 15 May 2026 MMR: Fusion transcript / reference transcript ratio < 0.1% in a sample containing at least 20,000 ABL 1 reference transcripts. Achieving MMR (a 3-log reduction) within the first 12 months of treatment is a positive indicator of long-term disease control and a lower risk of disease progression.

[0012] MR4: Fusion transcript / reference transcript ratio < 0.01% in a sample containing at least 20,000 ABL1 reference transcripts. MR4 is considered a deep molecular remission (4 log reduction) and indicates that the disease is well controlled. It is an important marker for long-term prognosis.

[0013] MR4.5: Fusion transcript / reference transcript ratio < 0.0032% in a sample containing at least 32,000 ABL1 reference transcripts. MR4.5 indicates an even deeper molecular remission (4.5 log reduction) and is often considered a target for patients considering a treatment break. MR5: Fusion transcript / reference transcript ratio < 0.001% in a sample containing at least 100,000 ABL1 reference transcripts. MR5 is the deepest molecular remission currently diagnosable (5 log reduction) and indicates a very low number of remaining malignant cells. It is considered an optimal marker for potential cure.

[0014] The requirements for a BCR::ABL1 MRD measurement (also known as an MRD assay) are high. Precision, reproducibility, comparability between different centers, and the exclusion of false-positive results are just as important as outstanding sensitivity. To detect MR5 (BCR::ABL1 / ABL1 ratio < 0.001%), a single BCR::ABL1 transcript must be detectable in a sample containing 100,000 ABL1 reference transcripts.

[0015] The most frequently used method for MRD measurement in CML and ALL is quantitative real-time polymerase chain reaction (qRT-PCR). qRT-PCR is a widely used technique in molecular diagnostics and is established in many laboratories. In this method, the sequence

[0016] WF / WF 241383WO 15 May 2026 des BCR::ABL1 Transcripts are amplified using specific oligonucleotides, called primers, from RNA isolated from a patient sample. This RNA has been previously reverse-transcribed into cDNA. The amplified RNA is then amplified through repeated cycles of denaturation, annealing, and elongation. The reaction mixture also contains fluorescently labeled probes or intercalating dyes that generate a fluorescence signal. This signal is measured in real time, with the intensity of the fluorescence signal being proportional to the amount of amplified target molecules (which can be DNA, cDNA, or RNA). The fluorescence is measured after each cycle, and the data are used to calculate the initial amount of target molecules in the sample.

[0017] The advantages of qRT-PCR include its relatively simple operation and the wide selection of reagents available that are certified for in vitro diagnostics. However, qRT-PCR has the significant disadvantage that it cannot quantify the number of BCR::ABL1 and ABL1 transcripts in a sample absolutely, but only relatively. To calculate the absolute number of transcripts in the sample from the measurement, a so-called standard curve must be used. For this purpose, reference samples containing defined amounts of transcripts are measured. The number of transcripts in the patient sample can then be calculated from the results of the sample measurements and the standard curve.However, even the smallest variations, which cannot be completely avoided in daily laboratory operations despite the utmost care and quality standards, as well as the use of different reagent batches or different measuring instruments, can influence the result. Furthermore, in qRT-PCR, the measurements of the fusion transcript BCR::ABL1 and the reference transcript (e.g., ABL1) are often performed separately, since the amplification efficiencies of the PCR reaction play a significant role in this method, and thus usually only one gene sequence can be measured per reaction (so-called monoplex reaction). This carries the risk that pipetting errors can also lead to erroneous results. Additionally, qRT-PCR has the disadvantage that the precision of the results decreases with increasing BCR::ABL1 levels (i.e., with disease progression).

[0018] WF / WF 241383WO 15 May 2026 noticeably decreases and thus alternative reference genes, e.g. GUSB or BCR, must be used at higher BCR::ABL1 / ABL1 ratios.

[0019] These partly unavoidable disadvantages of qRT-PCR impair the reliability of MRD assays with regard to precision, reproducibility, and, above all, comparability between different laboratories. As a standalone method, qRT-PCR cannot fully meet all the performance requirements for a BCR::ABL1 MRD measurement. To improve comparability, the International Scale (IS) was introduced by the European LeukemiaNet (ELN) in 2013. For this purpose, samples with defined BCR::ABL1 / ABL1 ratios were measured at various centers, and an individual conversion factor was calculated for each laboratory based on the results, intended to harmonize the findings. However, the widespread implementation of the IS is proving difficult, as there are currently no commercially available and sufficiently certified reference samples to determine the conversion factor.

[0020] Although qRT-PCR remains the most widely used method for MRD measurement in CML, digital PCR (dPCR) is gaining increasing importance. Unlike conventional qRT-PCR, where amplification takes place in a single reaction vessel, the dPCR reaction is divided into a large number (> 20,000) of partitions. Each of these partitions, which can be, for example, individual droplets (so-called digital droplet PCR, ddPCR) or microchambers, undergoes an independent reaction. After amplification is complete, each chamber is checked for the presence of a fluorescence signal, indicating whether the target transcript was present in that individual reaction. By counting the positive and negative partitions, the absolute number of target transcripts in the original sample can be calculated.

[0021] Compared to qRT-PCR, dPCR has several advantages. The number of transcripts used in the reaction can be calculated from the signal using Poisson statistics, which is why a standard curve is not necessary for dPCR. Furthermore, the

[0022] WF / WF 241383WO 15 May 2026 dPCR, unlike qRT-PCR, is very well suited for the simultaneous measurement of multiple target gene sequences, so that, for example, BCR::ABL1 (fusion transcript) and ABL1 (reference transcript) can easily be measured in the same reaction (multiplexing). The absolute quantification without the need for a standard curve and the multiplexing capabilities make dPCR more precise and reproducible than qRT-PCR and allow for greater comparability between different analytical laboratories. Although dPCR is not inherently more sensitive than qRT-PCR, dPCR can measure more replicates of a patient sample, thus achieving an overall higher sensitivity.

[0023] However, the increased sensitivity of dPCR also carries risks: the more sensitive a method is, the greater the danger of obtaining false-positive results, especially since even a single false-positive signal can lead to a change in the diagnostic category. Furthermore, a clear distinction between positive and negative signals is crucial for the success of the MRD assay: commercially available dPCR kits for MRD measurement in CML, for example, exhibit a so-called signal rain (see Fig. 1A), which significantly complicates the reliable determination of truly positive signals.

[0024] Object of the invention

[0025] Against this background, the present invention aims to provide an improved method for detecting a target gene in a biological sample, suitable for measuring MRD in CML and / or ALL, and which at least partially overcomes the disadvantages of known methods described above. In particular, the method should achieve a limit of detection (LoD) of 10 5 or exhibit fewer and allow for improved separation between positive and negative signals.

[0026] Description of the invention

[0027] The aforementioned problem is solved according to the invention by a method for detecting a target gene in a biological sample, comprising:

[0028] WF / WF 241383WO 15 May 2026a. Performing a digital polymerase chain reaction (dPCR) with primers and one or more DNA probes targeting the target gene; and

[0029] b. Determining the presence and / or quantity of the target gene in the sample, characterized in that at least one of the DNA probes is a nuclear probe comprising or consisting of bridged nucleic acids (LNAs) and having a length of 12 to 18 nucleic acids.

[0030] The aforementioned problem is further solved according to the invention by a DNA probe for use as a nuclear probe in a method according to the invention, which comprises or consists of bridged nucleic acids (LNAs) and has a length of 12 to 18 nucleic acids. The DNA probe is optionally configured to bind to the antisense strand of a target gene or gene segment, and / or is preferably configured to bind to a variant-specific sequence, preferably a fusion site, of a fusion gene.

[0031] Unless explicitly defined herein, all technical and scientific terms used have the same meaning as they are commonly understood by a professional in the field of gene therapy, biochemistry, genetics and molecular biology.

[0032] Unless otherwise specified, the present invention employs conventional techniques of cell biology, cell culture, molecular biology, biotechnology, and immunology, which are part of the prior art. Such techniques are described in detail in the literature.

[0033] The examples, features, and particular embodiments described herein are not to be interpreted as limiting the present invention only to the expressly described embodiments or combinations of features. Rather, the present description should be understood as disclosing and encompassing embodiments that combine the expressly described embodiments with any number of the

[0034] WF / WF 241383WO 15 May 2026 combine the elements described herein and / or preferred. Furthermore, all permutations and combinations of the elements described herein should be considered disclosed by this disclosure unless otherwise indicated in the context.

[0035] For the purposes of this disclosure, the term “target gene” (also referred to as “target transcript” or “target sequence”) refers to a transcript, gene, or gene segment that is specifically investigated or manipulated in molecular biology experiments or diagnostic tests. In particular, the target gene comprises the nucleotide sequence that is to be amplified and quantified by the PCR underlying the MRD measurement. The target gene is recognized by specific primers and probes that bind to the complementary sequences of the target gene.

[0036] Within the scope of this disclosure, the term "digital PCR" describes a configuration of the polymerase chain reaction (PCR) in which the sample to be analyzed is divided into a large number (10,000 to 100,000) of small reaction volumes or reaction chambers (so-called "partitions") in which a PCR procedure is carried out in parallel.

[0037] In the context of this disclosure, the term "probe" describes a fluorescently labeled oligonucleotide probe used in digital PCR for the specific detection and quantification of target sequences. The probe consists of a short, single-stranded DNA sequence that is complementary to the target sequence. Typically, the core probe is labeled with a fluorescent dye (reporter) at the 5' end and a quencher at the 3' end. During PCR amplification, the core probe binds specifically to the target sequence. The DNA polymerase possesses 5'-3' exonuclease activity that degrades the probe during the elongation phase, separating the reporter from the quencher. This results in the emission of a fluorescent signal suitable for the detection of the target sequence in the partition. In particular, the probe may be a core probe or a

[0038] WF / WF 241383WO 15 May 2026 control probe, whereby it is generally preferred that the core probe binds to the sequence of the target gene (e.g. the fusion transcript BCR::ABL1) and the control probe to the sequence of a control gene (e.g. the reference transcript ABL1).

[0039] For the purposes of this disclosure, the term "quencher" describes a molecule that absorbs the fluorescence energy of the fluorescent dye, thereby preventing the emission of a fluorescence signal as long as the probe is intact. From the known variety of dyes and quenchers, the specific combination of dye and quencher associated with the nuclear probe can be selected as desired. In particular, for example, FAM can be used as the dye and Iowa Black as the quencher associated with the nuclear probe.

[0040] For the purposes of this disclosure, the term "locked nucleic acids" (LNAs) refers to modified nucleic acid analogs stabilized by a chemical bridge between the 2'-O and 4'-C atoms of the ribose ring. This bridge fixes the ribose in a specific conformation, which improves the hybridization properties of the LNA probe, particularly resulting in increased stability against chemical and enzymatic degradation, high affinity for complementary nucleotide sequences, and improved specificity in binding to the target sequence.

[0041] The research underlying the invention has shown that a nuclear probe comprising or consisting of LNAs leads to a more efficient separation between positive and negative signals and, in particular, reduces the signal flux observed in conventional methods (see Fig. 1B). For example, the nuclear probe can comprise at least 2, at least 3, at least 4, or at least 6 LNA nucleotides, preferably no more than 12 or no more than 10 LNA nucleotides. Preferred nuclear probes can, for example, contain 5 or 6 LNA nucleotides. For example, the nuclear probe is configured for specific binding to the antisense strand of the DNA template.

[0042] WF / WF 241383WO 15 May 2026 By using LNAs in the nuclear probe, its sequence can be shortened compared to conventional probes, which usually have a length of 20 to 25 nucleotides. Therefore, nuclear probes according to the invention consist in particular of fewer than 20 nucleotides, namely of 12 to 18 nucleotides, preferably of 14 to 16 nucleotides, e.g., of 15 nucleotides. It has been shown that exceptionally high accuracy can be achieved with nuclear probes according to the invention, so that false positive signals can be excluded much more effectively than with conventional nuclear probes.

[0043] The method according to the invention is characterized by excellent accuracy and a particularly clear separation between positive and negative signals. In particular, the method exhibits a limit of white (LoB) of 0.000%, meaning that in a sample with 100,000 control transcripts, not a single false positive signal is to be expected.

[0044] The following describes various embodiments of the method, the DNA probe, and the use of the kit, with each embodiment applying independently to the method, the DNA probe, the use, and the kit, respectively. Furthermore, the individual embodiments can be combined in any way desired.

[0045] In one embodiment of the method according to the invention, the primers comprise a first primer pair configured to amplify a first gene segment of the target gene, and the nuclear probe is configured to bind specifically to the first gene segment. Preferably, the primers further comprise a second primer pair configured to amplify a second gene segment of the target gene or a control segment of a control gene. More preferably, at least one further DNA probe is a control probe configured to bind specifically to the second gene segment or the control segment. Optionally, the first and second primer pairs can also be present simultaneously in the same reaction.

[0046] WF / WF 241383WO 15 May 2026 In this embodiment, the method has the advantage that it is simple in design and implementation and, in particular, can be provided in the form of a kit with few components.

[0047] Within the scope of the present disclosure, a probe that binds specifically to a gene segment binds to a, preferably continuous, polynucleotide sequence within the gene segment. For example, a nuclear probe may have a sequence that is substantially complementary to a continuous sequence within the first gene segment, such that in the presence of the nuclear probe and the first gene segment, the nuclear probe binds specifically to the substantially complementary sequence within the first gene segment.

[0048] The specific design of the control probe is of secondary importance, since conventional DNA probes are also compatible with the method according to the invention. In particular, the control probe can therefore, like conventional probes, have a length of 20-25 nucleotides and be free of LNAs.

[0049] Alternatively, the control probe can have LNAs and a shorter length of, for example, 12 to 20 nucleotides.

[0050] Preferably, the determination in step b of the procedure is carried out by comparing a number of positive and negative signals from the core probe(s) and the control probe(s). Generally, a positive signal is a signal caused by the amplification of the corresponding sequence of the target gene or reference gene, and a negative signal is a signal caused by the lack of amplification.

[0051] The dPCR amplification reaction underlying the invention is carried out in a plurality of separated liquid volumes, so-called "partitions". For example, the partitions can be created by generating an oil-water emulsion system, as is done, for example, in known and commercially available dPCR systems. Alternatively, the partitions can be divided into

[0052] WF / WF 241383WO 15 May 2026 Partitions are performed on a microfluidic chip or well platform containing a defined number of reaction chambers; this approach is also known and products are commercially available.

[0053] The number of positive signals thus corresponds to the number of analyzed partitions in which amplification of the first gene segment was achieved. Similarly, the number of negative signals corresponds to the number of analyzed partitions in which no amplification of the first gene segment was achieved.

[0054] In general, both the core probe(s) and the control probe(s) can be present in the same reaction (so-called multiplexing).

[0055] In an exemplary embodiment of the method according to the invention, one or more DNA probes are fluorescently labeled, and the determination in step b. comprises detecting a fluorescence signal, wherein preferably in step b. a fluorescence signal is detected only if the DNA probe has been cleaved. In this embodiment, the method is compatible with known and commercially available systems for performing and fluorescence-based evaluation of dPCR reactions.

[0056] In dPCR methods such as the one according to the invention, a nucleic acid fraction is generally obtained from the biological sample before step a., and steps a. and b. are carried out with the nucleic acid fraction. Preferably, the nucleic acid fraction is a fraction comprising mRNA; more preferably, the nucleic acid fraction is the total RNA fraction of the biological sample.

[0057] In general, the biological sample can be a blood or bone marrow sample, e.g. whole blood or bone marrow aspirate, preferably anticoagulated (e.g. with EDTA).

[0058] In one embodiment of the method according to the invention, at least two core probes are included, each of the core probes being used for the specific detection of a

[0059] WF / WF 241383WO 15 May 2026, each alternative variant of the target gene is set up. For example, the reaction mixture can contain one primer pair for the amplification of both variants of the target gene (e.g., a first primer pair) and two different variant-specific core probes. Alternatively, in addition to the variant-specific core probes for the amplification of the respective variants of the target gene, the reaction mixture can contain two different primer pairs, or, for example, one forward primer and two different reverse primers, or, for example, two different forward primers and one reverse primer.

[0060] In this embodiment, the target gene is preferably a fusion gene, and at least one, preferably each, of the nuclear probes is configured to bind to a variant-specific sequence of the target gene. The variant-specific sequence is particularly preferably a fusion site of the fusion gene. A fusion site is understood to be the specific point at which two of the originally separate sequences underlying the fusion gene are joined together, such that the fusion site comprises sequence segments of both original sequences.

[0061] In one embodiment, the method has a limit of detection (LoD) of 0.01% or less (MR4), preferably 0.0032% or less (MR4.5), and even more preferably 0.001% or less (MR5). The method is preferably suitable for measuring minimal residual disease (MRD measurement) and thus for clinical application in diagnostics.

[0062] Surprisingly, it has been shown that when using a core probe according to the invention, the method leads to a better separation between positive and negative signals and at the same time has a sensitivity that allows its use in MRD measurements and meets even the strictest criteria currently used.

[0063] In a preferred embodiment of the method according to the invention, the target gene is a BCR::ABL1 fusion gene, wherein the first gene segment is a fusion site.

[0064] WF / WF 241383WO 15 May 2026 of BCR and ABL1. In this embodiment, the method can be used, for example, for MRD measurement of CML and / or ALL in a patient sample.

[0065] In this embodiment, the at least one core probe is preferably configured to bind to a variant-specific sequence, preferably at the fusion site, of BCR and ABL1. In this embodiment, the method enables the detection of different fusion variants of the BCR:ABL1 fusion gene.

[0066] In one embodiment, the first primer pair can comprise a first DNA primer—preferably binding on the sense strand—designated for specific binding in exon 13 of the BCR gene, and a second DNA primer—preferably binding on the antisense strand—designated for specific binding in exon 2 of the ABL1 gene. In this embodiment, a first nuclear probe is preferably configured for the detection of the gene variant el3a2 and a second nuclear probe for the detection of the gene variant el4a2. Since the fusion site of the two variants differs in the nucleotide sequence (see SEQ ID No. 1 for a gene sequence of the fusion site el3a2 and SEQ ID No. 6 for a gene sequence of the fusion site el4a2), the method in this embodiment enables the simultaneous detection of both variants. The respective variant-specific nuclear probes can be associated with the same dye or with different dyes.

[0067] In the embodiment described here, the method takes these fusion variants into account by providing variant-specific core probes and is therefore characterized by higher accuracy.

[0068] In another embodiment of the method, step a. involves performing dPCR in a multitude of droplets, and step b. comprises determining a multitude of signals, each originating from one of these multiple droplets. In this embodiment, the digital PCR is designed as digital droplet PCR (ddPCR). This makes the method compatible

[0069] WF / WF 241383WO 15 May 2026 with established and commercially available systems for performing and evaluating ddPCR analyses.

[0070] The problem addressed by the invention is further solved by using the method according to the invention for the diagnosis of chronic myeloid leukemia (CML) and / or acute lymphoblastic leukemia (ALL), preferably as MRD measurement, wherein the target gene is preferably a BCR::ABL1 fusion gene and at least one nuclear probe is further preferably configured to bind to a - preferably variant-specific - sequence, preferably a fusion site, of BCR and ABL1.

[0071] The problem addressed by the invention is further solved according to the invention by a kit for the detection of a target gene, comprising: a first primer pair configured to amplify a first gene segment of the target gene; at least one nuclear probe, preferably two nuclear probes, configured to bind specifically to the first gene segment, wherein at least one nuclear probe comprises or consists of linked nucleic acids (LNAs) and has a length of 12 to 18 nucleic acids.

[0072] Optionally, the kit further includes: a second primer pair configured to amplify a second gene segment of the target gene or a control segment of a control gene; and at least one control probe configured to bind specifically to the second gene segment or the control segment.

[0073] Further features and advantages of the invention will become apparent from the following description of exemplary embodiments, with reference to the accompanying figures.

[0074] The figures show

[0075] WF / WF 241383WO 15 May 2026 Fig. 1 an evaluation diagram of a ddPCR analysis with a commercially available kit (Fig. 1A) and with the inventive method (Fig. 1B) of a positive patient sample (BCR::ABL1 p210; el4a2 variant),

[0076] Fig. 2 shows an evaluation diagram of a ddPCR analysis using a commercially available kit (Fig. 2A) and the inventive method (Fig. 2B) of a weakly positive patient sample (BCR::ABL1 p210; el4a2 variant).

[0077] Fig. 3 shows an evaluation diagram of a ddPCR analysis using a commercially available kit (Fig. 3A) and the inventive method (Fig. 3B) of a positive patient sample (BCR::ABL1 p210; el3a2 variant).

[0078] Fig. 4 shows an evaluation diagram of a ddPCR analysis with established primers and probes suitable for standard qRT-PCR and a positive control sample (Fig.

[0079] 4A) and a negative control sample (Fig.4B),

[0080] Fig. 5 shows a schematic representation of sample preparation for the LoD test (Fig.

[0081] 5A) and the LoD test measurements (Fig. 5B),

[0082] Fig. 6 shows a graph of the linearity determination for the fusion gene BCR:::ABL1 p210,

[0083] Fig. 7 shows a graph of the linearity determination for the reference gene ABL1,

[0084] Fig. 8 shows a graph of the linearity determination for the ratio of target gene transcript BCR::ABL1 el3a2 and reference gene ABL1,

[0085] Fig. 9 shows a graph of the linearity determination for the ratio of target gene transcript BCR::ABL1 el4a2 and reference gene ABL1.

[0086] WF / WF 241383WO 15 May 2026 Fig. 10 the comparison of the logarithmic values ​​of the BCR::ABL1 / ABL1 ratios from measurements with a commercially available 1VD certified product and a measurement with the method according to the invention, and

[0087] Fig. 11 shows an evaluation of one p90 positive (Fig. 11A) and one p210 positive (Fig.

[0088] 11B] and negative (Fig. 11C] samples using primers and core probes according to the invention, set up to detect the BCR::ABL1 fusion variants ela2 and ela3 (pl90).

[0089] Example 1: Development of a dPCR assay for the detection of the BCR::ABL1 fusion gene. To develop a dPCR assay, a combination of primers and probes was produced that leads to the amplification of a gene region from the fusion site of the BCR::ABL1 fusion gene.

[0090] An initial primer and probe combination was used to detect the el3a2 variant of the BCR:::ABL1 fusion gene. SEQ ID No: 1 shows the transition sequence, where nucleotides 1-298 represent the BCR portion of the fusion gene and nucleotides 299-922 represent the ABL1 portion. The transition site is located between nucleotides 298 and 299, as shown in SEQ ID No: 1. A combination of a forward primer (SEQ ID No: 3) and a reverse primer (SEQ ID No: 4) suitable for amplifying the transition site was prepared. The amplicon of length lOlnt is shown in SEQ ID No: 2 and corresponds to nucleotides 236-336 according to SEQ ID No: 1. The two primers have a calculated melting temperature of 62.0 °C (SEQ ID No: 3) and 61.9 °C (SEQ ID No: 4), respectively.

[0091] An exemplary nuclear probe is shown in SEQ ID No: 5. Its 13 nt long sequence is AGG+GC+T+T+CTTC+CT, where bridged nucleotides are indicated by a "+" before the letter. The nuclear probe shown in SEQ ID No: 5 thus has a total of 5 bridged nucleotides: a bridged guanine at position 4, bridged thymines at positions 6 and 7, and bridged cytosines at positions 8 and 12. The nuclear probe has a calculated melting point of 67.6 °C.

[0092] WF / WF 241383WO 15 May 2026°C and binds to the antisense strand in the region of nucleotides 293-305 according to SEQ ID No: 1. It is labeled at its 5' end with the fluorescent dye FAM and at its 3' end with the fluorescent quencher Iowa Black.

[0093] A second primer and probe combination was used to detect the el4a2 variant of the BCR:::ABL1 fusion gene. SEQ ID No: 6 shows the transition sequence, where nucleotides 1-298 represent exon 13 of the BCR gene, nucleotides 299-373 represent exon 14 of the BCR gene, and nucleotides 374-997 represent exon 2 of the ABL1 gene. The transition site is located between nucleotides 373 and 374. The same combination of forward primer (SEQ ID No: 3) and reverse primer (SEQ ID No: 4) was used as for amplification of the transition site of the el3a2 variant. The 176 nucleotide amplicon is shown in SEQ ID No: 7 and corresponds to nucleotides 236-411 according to SEQ ID No: 6.

[0094] An exemplary nuclear probe is shown in SEQ ID No: 10. Its 15 nt long sequence is CT+CATC+A+T+CTT+C+CTT. The nuclear probe shown in SEQ ID No: 10 thus has a total of 6 bridged nucleotides: a bridged adenine at position 7, a bridged thymine at position 8, and bridged cytosines at positions 3, 9, 12, and 13. The nuclear probe has a calculated melting point of 66.1 °C and binds to the antisense strand in the region of nucleotides 291–306 according to SEQ ID No: 6. It is labeled at its 5' end with the fluorescent dye FAM and at its 3' end with the fluorescent quencher Iowa Black.

[0095] A third combination of a primer pair and a reference probe was produced for the reference gene ABL1 (see SEQ ID No: 11). The primer pair consists of a forward primer (SEQ ID No: 13) and a reverse primer (SEQ ID No: 14). The amplicon (SEQ ID No: 12) replicated by the primers is 114 nt long and comprises nucleotides 1223–1336 according to SEQ ID No: 11. The two primers have calculated melting points of 61.6 °C and 61.3 °C, respectively.

[0096] WF / WF 241383WO 15 May 2026 An exemplary reference probe is shown in SEQID No: 15. The reference probe has a length of 22 nt and the sequence TCTCCCCTACCAGGCAGTTTCG (calculated melting point: 67.2 °C). It has no bridged nucleic acids and binds to the sense strand of the ABL1 gene in the region of nucleotides 1291-1312. It is labeled at its 5' end with the fluorescent dye HEX and at its 3' end with the fluorescent quencher Iowa Black.

[0097] A dPCR assay was performed using the commercially available QX200 ddPCR system (Bio-Rad) according to the manufacturer's instructions, employing all three primer and probe combinations. The DNA samples analyzed were prepared by converting RNA samples of various origins (excess patient material, cell lines, proficiency testing samples, or commercial control material) to cDNA via reverse transcription (RT). In some experiments (e.g., for the linearity performance characteristic), samples that were negative for BCR::ABL1 were further modified with gBlocks containing the sequences for el3a2 and el4a2 (SEQ ID No: 1 and 6, respectively). These samples then served as templates for subsequent ddPCR applications. For this purpose, a master mix was first prepared, containing the commercially available ddPCR Multiplex Supermix (Bio-Rad) and a combination of the previously described primers and probes.For each individual reaction, the master mix and template were combined, and a droplet emulsion was prepared using the Automated Droplet Generator (Bio-Rad). PCR was performed in a thermocycler (C1000 Touch Deepwell Thermal Cycler, Bio-Rad), and the droplets were subsequently read using a Droplet Reader (QX200 Droplet Reader, Bio-Rad). Data analysis was performed using the QXManager 2.1 software (Bio-Rad).

[0098] For comparison, the same samples were analyzed using the commercially available QXDx BCR-ABL %1S kit (Bio-Rad). This kit is designed for use with the ddPCR system QX200 (Bio-Rad).

[0099] WF / WF 241383WO 15 May 2026. Figure 1 shows an evaluation of the positive patient sample (BCR::ABL1 p210; el4a2 variant) analyzed with the commercial kit (Fig. 1A) or a core probe according to the invention (Fig. 1B). The x-axis represents the fluorescence intensity in the HEX channel (ABL1 reference transcript). The y-axis represents the fluorescence intensity in the FAM channel (BCR::ABL1 fusion transcripts). Each point represents a measured droplet. The lines represent the threshold values ​​at which a droplet was considered positive. Droplets to the right of the vertical line are determined to be positive for ABL1, droplets above the horizontal line as positive for BCR::ABL1, and droplets in the upper right quadrant accordingly as positive for AB LI and BCR::ABL1. It can be seen that the separation of positive and negative droplets on the y-axis (FAM channel, BCR::ABL1) during measurement with a probe according to the invention (Fig.1A) is significantly improved compared to the commercial kit (Fig. 1B), and the signal rain is significantly reduced, which significantly facilitates the unambiguous differentiation of positive and negative droplets. The method according to the invention therefore exhibits higher accuracy than known dPCR methods.

[0100] Fig. 2 shows the evaluation of a weakly positive patient sample (BCR::ABL1 p210; el4a2 variant) (Fig. 2A: commercial kit; Fig.

[0101] 2B: method according to the invention). The representation in the figure is essentially identical to Fig. 1. It can be seen that the wide distribution of the BCR::ABL1 positive droplets makes it difficult to unambiguously set the threshold for channel 1 (horizontal line) and thus increases the risk of incorrectly classifying individual droplets as positive or negative, especially since the threshold in the commercial method shown is set using a BCR::ABL1 p210 el3a2 positive sample.

[0102] While individual droplets have very little influence on the result in highly positive samples, they can affect the result in clinically particularly relevant low-positive samples. The separation of positive and negative droplets on the y-axis (FAM channel, BCR::ABL1) is significantly improved in the inventive method (Fig. 2B) compared to the commercial kit (Fig. 2A), thus enabling the unambiguous classification of positive and negative droplets.

[0103] WF / WF 241383WO 15 May 2026 Especially with clinically relevant low positive samples, this reduces the risk of obtaining an erroneous result.

[0104] Fig. 3 shows the evaluation of a positive patient sample (BCR::ABL1 p210; el3a2 variant). The representation in the figure is essentially identical to Fig. 1. It can be seen that the separation of positive and negative droplets on the y-axis (FAM channel, BCR::ABL1) is significantly improved when measuring with a probe according to the invention (Fig. 3A) compared to the commercial kit (Fig. 3B), and significantly less "rain" occurs, which significantly facilitates the unambiguous differentiation of positive and negative droplets and thus increases the accuracy of the method. The method according to the invention therefore also exhibits higher accuracy with regard to the el3a2 fusion gene than known dPCR methods.

[0105] The inventive method was also compared with ddPCR and the use of established primers and probes for qRT-PCR. This was carried out with the following primers and probes:

[0106] Forward primer SEQ ID No: 24,

[0107] Reverse primer SEQ ID No: 25,

[0108] Probe SEQ ID No: 26,

[0109] the probe was coupled at its 5' end to the fluorescent dye FAM and at its 3' end to the quencher MGB (Minor Groove Binder).

[0110] The method was carried out under the same conditions as the method using the primers and probes according to the invention.

[0111] Figure 4 shows the evaluation of one positive (Fig. 4A) and one negative (Fig. 4B) sample using a conventional BCR::ABL1 probe without bridged nucleotides, which was used in qRT-PCR before the establishment of ddPCR, as well as the ABL1 reference probe and primer. The sequences of the primers and probe used for BCR::ABL1 are published in Gabert et al. (Leukemia 2003). The axes show, analogous to Figure 1, positive and negative droplets for BCR::ABL1 and ABL1.

[0112] WF / WF 241383WO 15 May 2026 It can be seen that the target transcript BCR::ABL1 (β-axis) is detected sufficiently well in the positive sample, and the separation of positive and negative signals is also satisfactory with an acceptable amount of rain. However, 11 false positive droplets were detected in the negative sample. This indicates that the primers and probe of the qRT_PCR assay do not have the specificity required for ddPCR and generate an unacceptable number of false positive signals, thus rendering them unsuitable for use.

[0113] The data presented here show that the inventive method enables a significantly more sensitive and specific detection of the BCR::ABL1 fusion transcript in a sample than known methods based on qRT-PCR and dPCR technology.

[0114] Example 2: Validation of the blank value

[0115] The method according to the invention was validated by determining the limit of blank (LoB). This indicates up to which detection level the measurement method provides background signal values ​​that are not attributable to actual target molecules such as the transcript of a BCR::ABL1 fusion gene.

[0116] To determine the LoB, approximately 100 patient samples identified as negative (CML follow-up examinations over 200 days; no detected BCR::ABL1 p210 transcripts) were analyzed using the inventive method according to Example 1. The measurements were performed under the following variable conditions:

[0117] 5 different employees;

[0118] 5 different batches of multiplex mix containing dNTPs, DNA polymerase and buffer solutions;

[0119] 4 different batches of ABL1 assay, containing primers and probes of sequences SEQ ID No: 11 to 13; and

[0120] 3 different batches of p210 assay, containing primers and probes of sequences SEQ ID No: 3 to 5 and 10.

[0121] WF / WF 241383WO 15 May 2026 A total of 104 patient samples were measured using the inventive method, and the number of BCR::ABL1 fusion transcripts (target sequence) was determined relative to the reference gene ABL1 (average 125,000 ABL1 transcripts). In 99 measurements, not a single BCR::ABL1 transcript was detected (95.2%). In 5 measurements, one BCR::ABL1 transcript was detected in each (4.8%).

[0122] The value for L0B95, described by the highest measured value that occurs in 95% of cases for a negative sample, is therefore 0 transcripts BCR::ABL1 per measurement. In other words, the LoB value of the method according to the invention was determined to be 0.000%.

[0123] Example 3: Determining the limit of detection (LoD)

[0124] To validate the inventive method, the limit of detection (LoD) was determined. The limit of detection indicates the lowest concentration of a target gene that can be reliably detected.

[0125] To determine the detection limit of the method, three positive samples, each containing approximately 30 transcripts in 40 pL (undiluted base samples), were prepared, corresponding to detection level MR4. Three negative pools (approximately 2000 ABL1 transcripts per pL, corresponding to approximately 160,000 ABL1 transcripts per measurement) were mixed with low-positive control samples (cell lines) in a predetermined ratio, resulting in approximately 30 BCR::ABL1 transcripts in each 40 pL base sample. The base sample was then divided. 20 pL were used for the base sample measurement itself, and 10 dilutions, each in a 1:10 ratio, were prepared from the remaining portion. All samples, undiluted and diluted, were analyzed using the method described in Example 1, and the number of target gene transcripts was determined. Figure 5A shows a schematic representation of the sample preparation.

[0126] The aim of this procedure is to detect or rediscover all BCR::ABL1 transcripts present in the sample by measuring the entire volume of the base sample and performing individual measurements of all dilutions.

[0127] WF / WF 241383WO 15 May 2026 should show that even when measuring dilute samples containing very few BCR::ABL1 fusion transcripts, these few fusion transcripts can be reliably detected.

[0128] The PCR assays were each measured in quadruplicate (4 wells).

[0129] The following table shows the number of BCR::ABL1 fusion transcripts found in the analyzed samples. The values ​​are given as mean values ​​[MW] of four measurements. Figure 5B also shows a graphical representation of the measured values.

[0130] Sample 1 2 3 Total target transcripts of the measurement Measurement Measurement 3 Measurements (MW) Undiluted sample 11 22 13 46

[0131] Sum of 10 14 17 14 45

[0132] diluted samples

[0133]

[0134] The table and figure show that the set quantity of target gene transcripts present in the samples (15 transcripts in the undiluted base sample, 15 transcripts in total in the diluted samples) was reliably detected in all three measurement runs by completely measuring the entire sample. This demonstrates that the method according to the invention can reliably detect even a single BCR::ABL1 transcript in a sample. The measurement deviation averaged 3-4 transcripts across all three measurement runs, which is within the expected range. The measurements are performed at the absolute limits of the method and are subject to expected statistical fluctuations.

[0135] The experiment shows that a single target transcript BCR::ABL1 in a negative matrix with at least 100,000 reference transcripts ABL1 (at least MR5)

[0136] WF / WF 241383WO 15 May 2026 can be reliably proven and thus the best possible LoD of 1 target transcript BCR::ABL1 is achieved.

[0137] Example 4: Determination of the limit of quantification (LoQ)

[0138] The method according to the invention was further validated by determining the limit of quantification (LoQ). This value indicates the lowest concentration of a target transcript that can not only be reliably detected but also quantified with a defined precision and accuracy.

[0139] To determine the limit of quantification (LoQ) of the method, seven negative patient pools were spiked with a positive control sample (cell line) such that they exhibited MR levels between MR3 and MR5. The samples were then analyzed in six separate assays (each assay: quadruplicates, 4 wells).

[0140] The aim of this approach was to reliably quantify the detection of BCR::ABL1 p210 fusion gene transcripts in the range of the MR level of at least 4.5 using the method according to the invention, or to determine the highest MMR value (i.e., the lowest concentration of BCR::ABL1 transcripts compared to ABL1 reference transcripts) at which the requirements for the relative standard deviation (coefficient of variation CV% <= 25%) specified for a LoQ test are still met.

[0141] The following tables illustrate the measured values ​​for the LoQ determination. For each of the seven LoQ samples (LoQ 1-7), the mean values ​​of the six measurements, as well as the corresponding standard deviations (StDev) and coefficients of variation (CV), are listed for the determined MMR level and the number of BCR::ABL1 transcripts. The highest MMR level that meets the CV requirements is 4.81%, thus fulfilling the previously defined requirements for the procedure. Another relevant value for considering the LoQ is the detected

[0142] WF / WF 241383WO 15 May 2026BCR::ABL1 Transcripts. Here, the lowest value at which a CV of < 25% is met is 7.7 transcripts.

[0143] M MR BCR::ABL1 Transcripts Mean Mean

[0144] StDev VK [%] StDev VK [%] (n=6) (n=6)

[0145] LoQ 1 3.19 0.05 1.54 62.32 7.21 11.57

[0146] LoQ 2 3.65 0.10 2.68 22.64 5.21 23.02

[0147] LoQ 3 4.13 0.14 3.33 7.57 1.94 25.67

[0148] LoQ 4 4.12 0.09 2.30 7.70 1.45 18.87

[0149] LoQ 5 4.29 0.21 5.00 5.92 2.60 43.99

[0150] LoQ 6 4.68 0.25 5.30 2.12 1.36 64.09

[0151]

[0152] LoQ 7 4.81 0.21 4.34 1.37 1.24 90.64

[0153] The LoQ is achieved at an MMR level of > 4.5 or < 8 BCR::ABL1 transcripts per 100,000 reference transcripts. This demonstrates that quantification of BCR::ABL1 fusion transcripts using nuclear probes according to the invention is performed with high accuracy even at very low levels, and the method thus provides reliable results.

[0154] Example 5: Linearity

[0155] To validate the method according to the invention, linearity was determined. This indicates the concentration range in which the measurement results are proportional to the actual target concentration. A test is linear if doubling the target concentration also leads to a doubling of the measured signal.

[0156] To determine linearity, target gene DNA (gBlocks) was added (spiked) to a negative patient pool (i.e., without target gene transcripts) and serially diluted with water (Figs. 6 and 7) or a negative patient pool (Figs. 8 and 9). These so-called gBlocks are double-stranded DNA that maps the sequence of the BCR::ABL1 p210 fusion transcripts. They are approximately 500 bp long and are available for both fusion variants, el3a2 and el4a2. These were used in the

[0157] WF / WF 241383WO 15 May 2026 Linearity analysis was also considered separately. As in the previous experiment, the different dilutions were used as samples in the inventive method according to Example 1 and analyzed. The number of fusion gene transcripts was determined and a linear regression was performed.

[0158] Figures 6 and 7 show the results of the analysis. The graphs plot the determined number of BCR::ABL1 p210 fusion gene transcripts (Fig. 6) and the number of ABL1 reference gene transcripts (Fig. 7), respectively. The regression line fitted to the measured values ​​exhibits a correlation coefficient (R) in both cases. 2 ) of 0.9999 and indicates that the inventive method exhibits excellent linearity over the measured range (dilution factor 1 to 1000).

[0159] In a second approach, sample preparation and procedure were identical, with the exception that the spiked patient pools were diluted in a further negative patient pool. Furthermore, the two variants el3a2 and el4a2 of the fusion gene were analyzed separately. This strategy also allows for the determination of the linearity of the ratio, since, with a constant concentration of the reference gene ABL1, only the concentration of the target gene transcript changes. Figures 8 (el3a2) and 9 (el4a2) show the results.

[0160] The figures show that the ratio of fusion gene to reference gene also exhibits very good linearity in both variants (R). 2 > 0.995).

[0161] Example 6: Comparison with established procedure

[0162] The inventive method was further compared with a method certified according to 1VD for determining MRD. The established method was the Bio-Rad QXDx BCR-ABL %1S Kit (1VD / CE-1VD).

[0163] The established procedure was carried out according to the manufacturer's instructions: The previously isolated total RNA from patient material, the RNA samples from the WHO

[0164] WF / WF 241383WO 15 May 2026 Reference panels and internal quality controls were reverse-transcribed into cDNA using Bio-Rad iScript reverse transcriptase. Droplet generation was then prepared by mixing cDNA, QXDx ddPCR Supermix, and QXDx 20x BCR-ABL primers / probes. The droplets were subsequently generated in the Bio-Rad Automated Droplet Generator. Internal controls were analyzed in duplicate (2 wells of approximately 20,000 drops each), and patient samples in quadruplicate (4 wells of approximately 20,000 drops each). All reactions were performed using the specified PCR protocol in a Bio-Rad T1000 Touch Deepwell Cycler. Droplet counts were measured in the Bio-Rad QX200 Droplet Reader (1VD). Analysis was performed according to the manufacturer's instructions.

[0165] Fig. 10 shows the comparison of the logarithmic values ​​of the BCR::ABL1 / ABL1 ratios from measurements with a commercially available 1VD-certified product and a measurement using probes according to the invention. A total of 21 patient samples, 5 internal quality control samples of the commercial kit, and 4 reference samples (WHO Genetic Reference Panel, for the quantitation of BCR-ABL translocation by RQ-PCR (1st lS) 09 / 138) were individually measured. It can be seen that both measurements show a very good correlation and yield comparable results (R 2 > 0.99).

[0166] Example 7: Development of a dPCR assay for the detection of the variants ela2 and ela3 (p!90) of the fusion gene BCR::ABL1

[0167] Another combination of primers and probes was used to detect the ela2 and ela3 (pl90) variants of the BCR:::ABL1 fusion gene. In rare cases, this fusion variant can also occur in patients with CML, but it also appears in acute lymphoblastic leukemia (ALL).

[0168] SEQ ID No: 16 shows the transition sequence of the ela2 fusion, where nucleotides 1-454 represent exon 1 of the BCR gene, nucleotides 455-628 represent exon 2 of the ABL1 gene, and nucleotides 629-925 represent exon 3 of the ABL1 gene. The transition point is located between nucleotides 454 and 455. SEQ ID No: 21 shows the transition sequence of the ela3 fusion, where nucleotides 1-454 represent exon 1.

[0169] WF / WF 241383WO 15 May 2026 of the BCR gene, and nucleotides 455-751 show exon 3 of the ABL1 gene. The junction is located between nucleotides 454 and 455. The same combination of forward primer (SEQ ID No: 18) and reverse primer (SEQ ID No: 19) was used for both variants. The amplicon of variant ela2, with a length of 278 nt, is shown in SEQ ID No: 17, and the amplicon of variant ela3, with a length of 103 nt, is shown in SEQ ID No: 22. Exemplary nuclear probes for BCR::ABL1 pl90 are shown in SEQ ID No: 20 and 23. Their lengths of 13 nt and 103 nt, respectively, are shown in SEQ ID No: 23. The 14 nt long sequences are AGG+GC+T+T+CTGC+GT (ela2) and TTT+CA+C+C+TGCGT+CT (ela3), where a "+" before a nucleotide indicates a bridged nucleic acid. The nuclear probes shown in SEQ ID No: 20 and 23 thus each have a total of 5 bridged nucleotides. The nuclear probes have a calculated melting temperature of 68.7 °C and 68.7 °C, respectively.71.4 °C and bind to the antisense strand in the region of nucleotides 449-462 according to SEQ ID No: 16 or nucleotides 447-461 according to SEQ ID No: 21. It is labeled at its 5' end with the fluorescent dye FAM and at its 3' end with the fluorescent quencher Iowa Black.

[0170] The experiment was carried out and evaluated as described in Example 1.

[0171] Fig. 11 shows an evaluation of one pl90 positive (Fig. 11A), one p210 positive (Fig. 11B), and one negative (Fig. 11C) sample each, using primers and core probes according to the invention, which were prepared for the detection of the BCR::ABL1 fusion variants ela2 and ela3, as well as the ABL1 reference probe and primer. The axes show, analogous to Fig. 1, positive and negative droplets for BCR::ABL1 and ABL1. It can be seen that the target transcript BCR::ABL1 pl90 (y-axis) is detected sufficiently well in the positive sample, and the separation of positive and negative signals is also satisfactory with an acceptable amount of rain. In Fig. 11B, the measurement of a p210 positive sample with the pl90 assay is shown. No false positive signals occur, which indicates a high specificity of the assay.

[0172] WF / WF 241383WO 15 May 2026 The example shows that the variants ela2 and ela3 can also be detected with the present invention, so that the method according to the invention is also suitable for the detection of ALL.

[0173] WF / WF 241383WO 15 May 2026

Claims

May 15, 2026 Claims 1. Method for detecting a target gene in a biological sample, comprising: a. Performing a digital polymerase chain reaction (dPCR) with primers and one or more DNA probes targeting the target gene; and b. Determining the presence and / or quantity of the target gene in the sample, characterized in that at least one of the DNA probes is a nuclear probe comprising or consisting of bridged nucleic acids (LNAs) and having a length of 12 to 18 nucleic acids.

2. Method according to claim 1, characterized in that the primers comprise a first primer pair configured to replicate a first gene segment of the target gene and the nuclear probe is configured to bind specifically to the first gene segment, wherein preferably the primers comprise a second primer pair configured to replicate a second gene segment of the target gene or a control segment of a control gene, and further preferably at least one further DNA probe is a control probe configured to bind specifically to the second gene segment or the control segment; wherein optionally the first and second primer pairs are present simultaneously in the same reaction, wherein the determination in step b. is preferably carried out by comparing a number of positive and negative signals of the core probe(s) and the control probe(s), wherein optionally both the core probe(s) and the control probe(s) are present in the same reaction.

3. Method according to one of the preceding claims, characterized in that the one or more DNA probes are fluorescently labeled and the determination in step b. comprises detecting a signal for the fluorescence, wherein preferably in step b. a signal for the fluorescence is detected only if the DNA probe has been cleaved.

4. A method according to any of the preceding claims, comprising at least two nuclear probes, wherein each of the nuclear probes is configured for the specific detection of a respective alternative variant of the target gene, wherein preferably the target gene is a fusion gene and at least one, preferably each, of the nuclear probes is configured to bind to a variant-specific sequence, preferably a fusion site, of the target gene.

5. Method according to one of the preceding claims, characterized in that the method has a detection limit (LoD) of 0.01% or less, wherein the method is preferably suitable for measuring minimal residual disease (MRD measurement).

6. Method according to any one of claims 2 to 5, characterized in that the target gene is a BCR::ABL1 fusion gene, wherein the first gene segment comprises a fusion site of BCR and ABL1; wherein preferably the at least one nuclear probe is configured to bind to a variant-specific sequence, preferably to the fusion site, of BCR and ABL1.

7. Method according to claim 6, wherein the first primer pair comprises a first DNA primer - preferably on the sense strand binding end - which is configured for specific binding in the BCR gene in exon 13, and a second DNA primer - preferably on the antisense strand binding end - which is configured for specific binding in the ABL1 gene in exon 2; WF / WF 241383WO 15 May 2026, preferably including a first nuclear probe for the detection of the gene variant el3a2 and a second nuclear probe for the detection of the gene variant el4a2.

8. Method according to one of the preceding claims, characterized in that in step a. the dPCR is carried out in a plurality of droplets and the determination in step b. comprises the detection of a plurality of signals, each originating from one of the plurality of droplets.

9. Method according to one of the preceding claims, characterized in that the biological sample is a patient sample, preferably a blood sample.

10. DNA probe for use as a nuclear probe in a method according to any of the preceding claims, comprising or consisting of bridged nucleic acids (LNAs) and having a length of 12 to 18 nucleic acids, wherein the DNA probe is optionally configured to bind to the antisense strand of a target gene or gene segment, wherein the DNA probe is preferably configured to bind to a variant-specific sequence, preferably a fusion site, of a fusion gene.

11. Use of a method according to any one of claims 1 to 9 for the diagnosis of chronic myeloid leukemia (CML) and / or acute lymphoblastic leukemia (ALL), preferably as MRD measurement, wherein the target gene is preferably a BCR::ABL1 fusion gene and at least one nuclear probe is further preferably configured to bind to a variant-specific sequence, preferably a fusion site, of BCR and ABL1.

12. Kit for the detection of a target gene, comprising: a first primer pair that is set up to replicate a first gene segment of the target gene; WF / WF 241383WO 15 May 2026 at least one nuclear probe, preferably two nuclear probes, which is / are configured to bind specifically to the first gene segment, wherein at least one nuclear probe comprises or consists of bridged nucleic acids (LNAs) and has a length of 12 to 18 nucleic acids; - optionally a second primer pair designed to replicate a second gene segment of the target gene or a control segment of a control gene; and optionally, at least one control probe is configured to bind specifically to the second gene segment or the control segment. WF / WF 241383WO 15 May 2026