WO2025108913A1

WO2025108913A1 - Method of detecting gene fusion events

Info

Publication number: WO2025108913A1
Application number: PCT/EP2024/082779
Authority: WO
Inventors: Jonathan Weiss; Miriam HESSE; Rink PHILIPP
Original assignee: Qiagen GmbH
Current assignee: Qiagen GmbH
Priority date: 2023-11-23
Filing date: 2024-11-19
Publication date: 2025-05-30
Anticipated expiration: 2026-05-23

Abstract

The invention relates to a method of detecting gene fusion events and to a kit adapted for carrying out said method.

Description

Method of detecting gene fusion events

[0001] The present invention relates to a method of detecting gene fusion events and to a kit adapted for carrying out said method.

FIELD OF THE INVENTION

[0002] The present invention relates to the field of molecular biology, more par- ticularly to the detection of gene fusion events.

BACKGROUND OF THE INVENTION

[0003] The term 'gene-fusion' commonly describes an event where parts of a genomic region, encoding for two different genes, are being fused together. These events can occur in different ways, especially in form of translocation events. Translocations oc- cur when in the process of genome replication prior to cell division two chromosomes ran- domly align close together. In most cases, these rearrangements are being recognized by special proteins which in turn signal to the affected cell to undergo apoptosis, i.e., pro- grammed cell death. In contrast, when these gene-fusions occur in cells which are already genetically damaged, apoptosis can be circumvented, and these newly formed genes be- come active. Some of these newly formed fusion-genes are able to drive oncogenic trans- formation of a cell ultimately leading to the formation of a tumor. In many cases, these gene fusions are therapeutic targets in cancer treatment and are therefore routinely tested for in molecular diagnostic laboratories around the world.

[0004] Prominent examples of gene fusion events are the BCR-ABL fusion in certain forms of leukemia which are susceptible to ABL kinase inhibitors like imatinib (Drucker et al., 2001 , Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia, N. Engl. J. Med. 344:1031-1037), and EML4-ALK fu- sion in non-small cell lung cancer being susceptible to treatment with e.g., crizotinib (Shaw et al., 2013, Crizotinib versus chemotherapy in advanced ALK-positive lung cancer, N. Engl. J. Med. 368:2385-2394).

[0005] Currently there are three prominent ways to test for presence of a gene fusion.

[0006] Fluorescence in-situ hybridization. Here, two differently fluorescence la- belled DNA probes are being incubated with a number of tumor cells on a glass slide and visually checked using a microscope. In cells without a gene-fusion, both signal can be detected separately, e.g. red and green. When a cell harbors the gene fusion of interest, the signals from both probes occur next to each other thereby overlapping the signals, e.g. yellowish when red and green colors are being used (Cruz-Rico, 2017, Diagnosis of EM L4-ALK translocation with FISH, immunohistochemistry, and real-time polymerase chain reaction in patients with non-small cell lung cancer, Am. J. Clin. Oncol. 40(6):631- 638). The drawback of the technology is in the throughput, FISH is a labor-intensive ap- proach and very costly in terms of reagents and labor costs. In addition, the operator can only look at a limited number of cells and it's dependent on established probes. [0007] Endpoint or quantitative PCR. Both types of PCR rely on the unique fea- ture that a de-novo gene fusion is not occurring in a normal genome. Therefore, it is possi- ble to design a PCR assay that is specific to the gene-fusion. By having each primer bind to one part of the newly formed gene-fusion, only in the presence of the gene-fusion the PCR reaction is able to amplify. The drawback of this approach is its specificity. Even though gene-fusion occur between the same genes, they can include different parts of each gene. For example, in BCR-ABL a number of different versions can occur and have to be tested for. That means these PCRs usually entail a highly multiplexed approach uti- lizing a large number of primer pairs (Bock et al., 2003, Multiplex RT-PCR for the detec- tion of common BCR-ABL fusion transcripts in paraffin-embedded tissues from patients with chronic myeloid leukemia and acute lymphoblastic leukemia, Diagn. Mol. Pathol. 12(3): 119-23). Multiplex PCRs are inherent difficult reactions as primers can interact with each other and give rise to false signals which makes data interpretation difficult. It also limits throughput and due to the low sensitivity for qPCR only one sample can be tested at a time, which then also has to be compared to a standard curve (Foroni et al., 2011 , Guidelines for the measurement of BCR-ABL1 transcripts in chronic myeloid leukaemia, Br. J. Haematol. 153(2): 179-90).

[0008] NGS-based approaches. With the development of massive parallel RNA sequencing, it became possible to identify gene-fusions on the RNA level. At first gene- fusions were detected using whole transcriptome RNA sequencing, where millions of RNA fragments are being sequenced. Some of them will contain a sequence where at the be- ginning of the fragment encodes for gene 1 and the end for gene 2, covering the break- point (Ge et al., 2011 , FusionMap: detecting fusion genes from next-generation sequenc- ing data at base-pair resolution, Bioinformatics 27(14): 1922-1928). Later, this approach was developed further to only sequencing certain RNA fragments of interest, by specifi- cally PCR amplifying these fragments prior to sequencing (Heyer et al., 2019, Diagnosis of fusion genes using targeted RNA sequencing, Nature Communications volume 10, Arti- cle number: 1388). Using this approach, some of the disadvantages of the whole RNAseq approach are being remedied. Namely the larger amount of sequencing and its associated costs and the fact that low abundant RNA fragments might not be covered sufficiently. However, both approaches suffer from relative high costs, complicated data analysis, long turn-around times, limited scalability and the need for highly specialized (and therefore of- ten expensive) personal.

[0009] Against this background it is an object to the invention to provide a method of detecting gene fusion events by means of which the disadvantages of the methods of the art are improved or even avoided. In particular, such a method should be provided which is fast, easy to use and relatively cheap.

[0010] The present invention satisfies these and other needs.

SUMMARY OF THE INVENTION

[0011] This problem is solved by a method of detecting gene fusion events, comprising the following steps:

(i) providing a composition comprising mRNA molecules suspected of encod- ing at least a first gene A (RIRNAA) and at least a second gene B (mRNAB),

(ii) determining in the composition

- the amount of 5' ends of OIRNAA ^ITIRNAA), and

- the amount of 3' ends of OIRNAA ^ITIRNAA) and/or

- the amount of 5' ends of mRNAB (⁵'mRNAB), and

- the amount of 3' ends of mRNAB (³'mRNAB)

(iii) detecting a fusion of gene A and gene B if

⁵'mRNA_A # ³'mRNA_A or

⁵'mRNA_B # ³'mRNA_B or ⁵'mRNA_A « ³'mRNA_B or

³'mRNA_A « ⁵'mRNA_B.

[0012] According to the invention, a 'composition' is a mixture of substances of any embodiment which contains the mRNA molecules which are suspected of coding for the at least first gene A and the at least second gene B. A 'composition' can be any me- dium, such as a buffered solution, which contains the mRNA molecules in question. It can also comprise total RNA from a biological cell or cell line.

[0013] 'Gene A' and 'Gene B' are symbolic for any genes that are either sus- pected of fusing with each other or are known to be able to fuse with each other, such as the well-described oncogenes BCR (gene A) and ABL1 (gene B).

[0014] According to the invention, 'mRNAA¹ stands for an mRNA molecule which codes for gene A, and 'mRNAB¹ stands for an mRNA molecule which codes for gene B.

[0015] According to the invention '⁵'mRNA_A means the amount of 5' ends of mRNAA, '³'mRNA_A means the amount of 3' ends of mRNAA, '⁵'mRNA_B means the amount of 5' ends of mRNAB and '³'mRNA_B means the amount of 3' ends of mRNAB.

[0016] According to the invention, the 5'-end of the respective mRNA, i.e. mRNAA and mRNAB, is understood to be the section of the mRNA which is located up- stream of the suspected or known fusion site in the direction of the 5'-terminus, and the 3'- end is understood to be the section of the mRNA which is located downstream of the sus- pected or known fusion site in the direction of the 3'-terminus.

[0017] In step (iii), it is detected whether an imbalance in the expression of the two termini of the respective mRNA, i.e., the 5' end and the 3' end, is present. An imbal- ance in the amounts or expression of both termini of the mRNA is symbolized by the sign If both termini, i.e., the 5'-end and the 3'-end of the mRNA of the same gene, are expressed differently, i.e., if different amounts of the 5'-ends and 3'-ends of the OIRNAA and/or of the mRNAs are present in the composition, a gene fusion event has taken place and each section is under the control of a different promoter, which controls a different ex- pression in each case. The different amounts of mRNA sections can be measured with methods known to the skilled person, for example by means of a polymerase chain reac- tion (PCR). In step (iii) of the method according to the invention, surprisingly even the smallest imbalances in the quantities of the 5' and 3' ends of the mRNA are sufficient to detect a gene fusion.

[0018] Preferably, the two genes A and B are expressed at different levels in normal, healthy cells in which no gene fusion event has taken place. The more the two genes differ from each other in their expression level in such normal, healthy cells, the more reliably and accurately a gene fusion event in an anormal, non-healthy cell in which a gene fusion event has taken place, can be detected with the method according to the invention. Preferably, the expression levels of both genes in normal, healthy cells differ by at least 5%, preferably by at least 10%, further preferably by at least 20%, and most pref- erably by at least 30%.

[0019] The inventors further recognized that the amount of the 5' and 3' ends of the mRNA of the first gene A and the second gene B are then approximately equal when a gene fusion has occurred. In this case,

stands for approximately equal. The expres- sion ' or "approximately equal" in this description refers specifically to the comparison of mRNA amounts within a reaction mixture. It denotes that the actual amount of mRNA is within an acceptable deviation from the specified amount, with a maximum allowable devi- ation of 30%. Preferably, however, the deviation is less than 25%, more preferably less than 20%, even more preferably less than 15%, and most preferably less than 10%. This tolerance range reflects typical variability in mRNA quantification and is recognized as an acceptable approximation within the relevant technical field.

[0020] Surprisingly, the method according to the invention is highly sensitive such that a gene fusion can still be reliably detected even if, along with mRNA molecules encoding fused genes, large amounts of mRNA molecules encoding non-fused genes are present in the composition. For example, detection is still possible even if only approx. 25% of the mRNA transcripts in the composition code for gene fusion A-B and approx. 75% of the mRNA transcripts code for the non-fused genes A and/or B.

[0021] Compared to the current state-of-art methods of gene fusion detection, the method according to the invention is fast, easy to use and relatively cheap.

[0022] In an embodiment of the invention in step (iii) a fusion of gene A and B is detected if

⁵‘ mRNA A is higher or lower ³‘ mRNA A or

⁵'mRNAB is higher or lower ³'mRNAB, preferably each by at least 20%, preferably by at least 30%, further preferably by at least 50%, fur- ther preferably by at least 100%, further preferably by at least 150%, further preferably by at least 200%, further preferably by at least 250%, further preferably by at least 300%, fur- ther preferably by at least 350%, further preferably by at least 400%, further preferably by at least 450%, and further preferably by more than 450%.

[0023] This measure has the advantage that even small imbalances in the ex- pression of the 5' and 3' ends of the mRNA enable the detection of a gene fusion. This is possible in particular, but not only, when using highly sensitive detection methods such as digital PCR (dPCR). The indicated percentages can be volume percent (vol.%), weight percent (wt.%) or mole percent (mol.%).

[0024] In still another embodiment in the method according to the invention in that in step (iii) a fusion of gene A and B is detected if

⁵'mRNAA and ³'mRNAB differ by less than 40%, preferably less than 35%, more preferably less than 30%, more preferably less than 25%, more preferably less than 20%, more preferably less than 15%, more preferably less than 10%, more preferably less than 5%, or

³'mRNAA and ⁵'mRNAB differ by less than 40%, preferably less than 35%, more preferably less than 30%, more preferably less than 25%, more preferably less than 20%, more preferably less than 15%, more preferably less than 10%, more preferably less than 5%.

[0025] The inventors have recognized that the amount of the 5' and 3' ends of the mRNA of the first gene A and the second gene B are then approximately equal when the indicated deviations are present. In such a case, gene fusion has occurred.

[0026] Accordingly, in an embodiment of the method according to the invention, the determination in step (ii) is carried-out by means of polymerase chain reaction (PCR), preferably by means of digital PCR (dPCR).

[0027] This measure has the advantage that such detection method is em- ployed which allows a reliable and accurate determination of the amount of the respective mRNA sections. Digital polymerase chain reaction (dPCR) is a biotechnological refine- ment of conventional polymerase chain reaction methods that can be used to directly quantify and clonally amplify nucleic acids strands including RNA, but also DNA and cDNA. dPCR offers certain advantages of conventional PCR. In dPCR, samples are parti- tioned in thousands of individual reactions prior to a PCR cycling reaction. This effectively decreases the concentration of target molecules within each individual partitioned PCR re- action, in contrast to traditional PCR where a bulk sample is being analyzed. End-point PCR then occurs in each of the partitions independently, producing a signal only in parti- tions where the target of interest is present. Due to the small volume the reaction is more tolerant towards contaminants or PCR inhibitors that might be introduced by the analyzed sample. dPCR detects PCR signals from individual partitions and calculates the absolute amount of target molecules present using Poisson statistics, eliminating the need for a standard curve. Due to these features, dPCR allows for absolute quantification of target molecules with a higher sensitivity and specificity compared to qPCR, down to a single target molecule. In addition, dPCR removes the challenges associated with bioinformatic analysis and interpretation of NGS data, is easily scalable to run in cost efficient manner, doesn’t need overly expensive machines or reagents and has a fast turn-around time.

[0028] In still another embodiment of the invention the PGR or dPCR involves PGR primers or probes capable of hybridizing to

- 5' end-located sections of mRNAAand

- 3' end-located sections of mRNAA and/or

- 5' end-located sections of mRNAsand

- 3' end-located sections of mRNAB.

[0029] With this measure, the technical prerequisites are created so that the 5' and 3' termini of the respective mRNA molecules can be reliably and quantitatively deter- mined by means of PGR or dPCR, respectively.

[0030] In yet another embodiment of the method according to the invention the 5'-end-located sections are located upstream of the site of fusion of gene A and gene B, i.e., 'before' of the fusion site or 'at the beginning' of the respective mRNA, and the 3'-end- located sections are located downstream of the site of fusion of gene A and gene B, i.e., 'behind' the fusion site or at the end of the respective mRNA. With this measure, the skilled person receives instructions on how to design the PGR primers and probes in order to selectively and specifically amplify only the 5'-ends or 3'-ends before or behind the fu- sion site of the respective mRNA molecules. For the amplification of the 5'-end of the re- spective mRNA, the PGR primers and the hybridization probe are selected so that they bind to the mRNA template before the fusion site, i.e., in the direction of the 5'-end. For the amplification of the 3' end, on the other hand, they are selected so that they bind to the mRNA template after the fusion site, i.e., towards the 3' end. In the example of BCR-Abl described by the inventor, the PGR primers and hybridization probes to detect the 5'-end must bind at or before the end of ABL_Exon1 , i.e., in a section of the mRNA that extends from the 5'-end to the end of ABL_Exon1 , since in some BCR-Abl fusions described, exon 2 is already part of the fusion gene. This can be different for other fusion genes. For ex- ample, if exons 1-3 of gene A are never part of the resulting fusion transcript, the PCR pri- mers and hybridization probes can hybridize to detect the 5' end of the mRNA at any posi- tion in this area, which extends in the mRNA from the 5' end to the end of exon 3 of gene A.

[0031] In another embodiment of the invention the composition provided in step (i) is total RNA of biological cells suspected of carrying a fusion of genes A and B.

[0032] This measure has the advantage that optimal source material is pro- vided for the determination of fusion events. It is a further advantage that total RNA does not have to be purified further and, in particular, isolation of mRNA molecules is not nec- essary in order to be able to reliably determine a gene fusion.

[0033] In another embodiment of the invention the biological cells are tumor cells, preferably white blood tumor cells.

[0034] Although the method according to the invention is suitable for detecting any gene fusions in any biological cell, this measure has the advantage that clinically par- ticularly significant gene fusions, such as the BCR-ABL1 gene fusion, which occurs pri- marily in chronic myeloid leukemia, can be reliably detected.

[0035] In yet another embodiment of the invention the biological cells stem from one individual. In an alternative embodiment the biological cells stem from at least two in- dividuals.

[0036] These two embodiments enable the method according to the invention to be implemented as a single case test method or as a pool test method. In the latter em- bodiment, biological samples from different individuals or patients that are to be tested for the presence of a gene fusion are combined and tested together 'in one go'. Thereby, ac- cording to the invention, 'at least two individuals' means two, three, four, five, six, seven, eight, nine, ten, twenty, thirty, forty, fifty, one hundred, two hundred, three hundred or more individuals.

[0037] In another embodiment of the invention gene A is BCR or ABL1 and/or wherein gene B is ABL1 or BCR. In still further embodiments gene A is selected from the group consisting of EML4, Tmprss2, and CD74, and gene B is selected from the group consisting of ALK, ERG, and NRG1.

[0038] As mentioned furthermore above, the method according to the invention can be used to detect any gene fusion. However, the above measures have the ad- vantage that the method according to the invention is adapted for the detection of clinically particularly significant gene fusions.

[0039] In another embodiment of the method according to the invention, in step (iii) the detection is made using an integrity calculation algorithm.

[0040] In digital PCR each template is randomly distributed into individual parti- tions. Dependent on the number of templates and partitions, a certain percentage of parti- tions will contain 2 or more templates. When these templates are then being accessed us- ing different fluorophore labeled hydrolysis probes for example, partitions containing two or more templates will appear fluorescent in two or more detection channels. When digital PCR is being performed using RNA extracted from a normal cell with no gene fusion and using two assays, each targeting one fusion partner of interest. Depending on the number of available partitions and RNA templates, some partitions will contain templates from both potential fusion partners. In digital PCR, this is commonly referred to as multiple oc- cupancy and the number of partitions that are expected to randomly contain more than one template can be calculated using poison distribution statistics. Now assuming RNA extracted from a cancer cell containing a gene fusion is being used in the same setup as above. In this case, the total RNA containing the mRNA encoding the fusion gene (and hence the target sequence for both assays) is being distributed as a single template and not individually as in the setup before. The result being a number of double positive parti- tions that is significantly higher than what can be expected based on poison statistics. This then indicated a non-random distribution which can only be explained by the pres- ence of a gene fusion.

[0041] Another subject-matter of the invention relates to a kit for detecting gene fusion events between a first gene A and a second gene B, comprising:

- PCR primers and/or probes capable of hybridizing to 5' end-located sections and 3' end-located sections of mRNA molecules encoding gene A (mRNAA) and/or gene B (mRNAB), and

- a manual comprising instructions for carrying out the method according to the in- vention.

[0042] The properties, features, advantages and embodiments of the method according to the invention apply equally to the kit according to the invention.

[0043] The advantage of providing a kit is that all components, reagents and in- formation are provided together to enable error-free operation of the method according to the invention, even for untrained or semi-skilled personnel.

[0044] In the kit the PCR primers and/or hybridization probes may be provided in separate containers or a common container, each in solution or in lyophilized and/or crystalline formulation, and/or as part of a preformulated master mix. The kit may com- prise another container containing a diluent or reconstituting solution for the lyophilized and/or crystalline formulation.

[0045] The kit may further comprise one or more of a buffer, a diluent, a filter, a needle, or a syringe. The container(s) is/are preferably a bottle, a vial, a syringe or test tube; and it may be a multi-use container. The container(s) may be formed from a variety of materials such as glass or plastic. Preferably the kit and/or container contain/s instruc- tions on or associated with the container(s) that indicate directions for reconstitution and/or use. [0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Although any methods and materials similar or equivalent to those described herein can also be used in the practice or testing of the present disclo- sure, the detailed methods and materials are now described. All publications mentioned herein are incorporated herein by reference to disclose and describe the methods and/or materials in connection with which the publications are cited. The publications discussed herein are provided solely for their disclosure prior to the filing date of the present applica- tion. Nothing herein is to be construed as an admission that the present disclosure is not entitled to antedate such publication by virtue of prior disclosure. Further, the dates of publication provided can be different from the actual publication dates, which can need to be independently confirmed.

EXAMPLES

[0047] Before the present invention is further described, it is to be understood that this disclosure is not strictly limited to particular embodiments or examples described herein, as such can of course vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments and examples only, and is not intended to be limiting, since the scope of the present invention will be limited only by the claims.

[0048] The invention is now further explained by means of examples resulting in additional features, characteristics and advantages of the invention. The features men- tioned in the specific examples are features of the invention and may be seen as general features which are not applicable in the specific example but also in an isolated manner in the context of any embodiment of the invention.

[0049] It is appreciated that certain features of the disclosure, which are, for clarity, described in the context of separate embodiments or examples, can also be pro- vided in combination in a single embodiment. Conversely, various features of the disclo- sure, which are, for brevity, described in the context of a single embodiment or example, can also be provided separately or in any suitable sub-combination. All combinations of the embodiments are specifically embraced by the present disclosure and are disclosed herein just as if each and every combination was individually and explicitly disclosed. In addition, all sub-combinations are also specifically embraced by the present disclosure and are disclosed herein just as if each and every such sub-combination was individually and explicitly disclosed herein.

[0050] The invention is now further described and explained in further detail by referring to the following non-limiting examples and figures:

Fig. 1: Principal concept. In a normal cell, mRNA from gene A and B are expressed full length, i.e., the amount of 5' end equals the amount of the 3' end and both genes are overall differently expressed (left side). In case of the fusion event, the 3' end of gene B now shows an increased rate of expression (similar to the 5' end of gene A) and very different the original 5' end of gene B.

Fig. 2: RNA from healthy donors. Each bar represents the copies/pl reaction from one assay. Bars 1-3 represent the assays targeting the 5' end and bars 4-6 repre- sent the 3' end of the ABL mRNA.

Fig. 3: Fusion detection assays on K562 RNA. Bars 1-3 represent assays targeting the 5' end of the mRNA, whereas bars 4-6 represent the 3' end. The copies/pl from the 5' to the 3' are clearly out of balance and indicate the presence of a gene fusion.

Fig. 4: The assays described here are shown from left to right. For each assay the mutant/wildtype RNA mixture is also shown from left to right. The bars repre- sent the concentration of RNA molecules in absolute copies/pl reaction. At 25% K562 (mutant) RNA, the 3' end fragments are distinguishable imbalanced to the 3' end copies. Fig. 5: Comparison of observed versus expected double-positive partitions in digital

PCR analysis. Results show a 33% higher number of double-positive parti- tions in RNA extracted from K562 cells (indicating the presence of the BCR- Abl fusion gene) compared to the expected number, while Leucocyte RNA shows alignment between observed and expected values, confirming separate mRNA presence.

1. Concept

[0051] The method according to the invention is based on the idea that in un- perturbed gene expression all parts of the mRNA are equally present. Otherwise, no com- plete protein can be translated. When a gene-fusion is present, parts of the same mRNA are under control of two different gene expression promoters and therefore parts of the same original mRNA are presents in different absolute amounts. For example, a gene is not expressed at all in a particular cell and no mRNA can be detected. When this cell now contains a gene-fusion encompassing parts of the gene (for example from the middle to the end), suddenly this part of the mRNA can be detected. In contrast, the "front end" part of the mRNA is still not expressed as it is still under the control of its original promoter. This leads to a detectable imbalance between the two parts of the gene’s mRNA. This basic principle illustrated exemplarily in Fig. 1 for genes A and B can be used for all gene fusions. Limitations can occur in cases where the native expression of both genes is too close to accurately determine differences in expression.

2. Material and methods

Sample material

[0052] Total RNA, extracted from the BCR-ABL positive cell line K562 and treated with DNAsel, was purchased from Biochain. For normal controls, total RNA ex- tracted from buffy coats was used. Post extraction, RNA was treated with DNAsel and cleaned using Qiagen RNAeasy kits. PCR primers and probes

Assay sequences are listed in Table 1 below.

Table 1 : PCR primers and probes

Digital PCR

[0053] The dPCR reactions were set up according to standard QIAcuity One- Step advanced probe PCR Kit Quick Start Protocol. TaqMan probe-based assays were used for multiplex detection of target levels. The nanoplates were sealed and placed in a QIAcuity Digital PCR instrument according to the instrument's user manual. Standard QI- Acuity dPCR cycling program (annealing was set to 60°C) was selected. Standard gain and exposure were used for imaging. Results were analyzed using QIAcuity Software Suite (Suite 2.2). The assays were run in a duplex reaction to eliminate potential experimental bias. Assays detecting the 5' end were FAM labelled, the 3' end assays HEX labelled.

Sample dilution

[0054] K562 derived RNA was serially diluted using RNA derived from healthy donor leucocytes. The absolute amount of RNA was kept stable and the ration between mutant and wildtype was set at 100, 90,75, 50,25,10 and 0% mutant RNA (Table 2).

3. Results

RNA derived from healthy donor material

[0055] The inventor tested 100ng of healthy donor RNA in order to establish the assays and test the differential expression. All assays performed as expected on RNA ex- tracted from healthy leucocyte RNA (Figure 2). As expected, the absolute counts of RNA molecules on the 5' end (Assays 1, 2 and 3) are in balance to the counts obtained from assays targeting the 3' end (Assays 4, 5 and 6). The assay detecting ABL_1b quantifies higher than the other assays, due to also detecting pre-processed RNA and does not ac- curately reflect the mature mRNA pool.

RNA derived from the K562cell line

[0056] Next total RNA extracted from the K562 cell line was tested, which car- ries the BCR-ABL fusion (Ge et al., 2011 , /.c.) following the approach of the invention to detect the BCR-ABL fusion gene. All 6 assays were tested and compared to each other. In the BCR-ABL fusion gene, exon 1 is always retained and exon 2 onwards is fused to the 5' end of the BCR gene locus. For the approach of the invention this means the as- says 1-3 targeting exon 1 will show different counts when comparing to the assays 4-6. The results in Figure 3 illustrates this. The copies/pl reaction for assays 1-3 average out around 230, whereas assays 4-6 count 2680 copies/pl on average. The clear imbalance between the 5' and 3' end stems from the presence of the BCR-ABL fusion mRNA. Serial dilution of K562 with normal RNA

[0057] K562 RNA is derived from a tumor cell line that is cultured in solitude.

Therefore, all cells contain the BCR-ABL fusion gene. Serial dilutions of the K562 with healthy RNA allows to artificially simulate a situation where the tumor content is not 100%, as expected when using for example patient material. The inventor kept the total amount of added RNA stable at 10ng RNA I reaction and increased the ratio of K562 and wildtype RNA according to Table 2.

Table 2: Serial dilution of K562 with healthy RNA.

[0058] The result is depicted in Fig. 4. Column 1 shows the absolute amounts that were mixed together, whereas column 2 shows the resulting tumor content. As can be seen at 10%, preferably 25% K562 (mutant) mRNA, the 3' end fragments are distinguisha- ble imbalanced to the 3' end copies.

[0059] In another embodiment, mRNA encoding for a fusion gene is being de- tected using an integrity calculation algorithm. The idea here being that in digital PCR each template is being randomly distributed into individual partitions. Dependent on the number of templates and partitions, a certain percentage of partitions will contain 2 or more templates. When these templates are then being accessed using different fluorophore labeled hydrolysis probes for example, partitions containing two or more tem- plates will appear fluorescent in two or more detection channels.

[0060] Now assuming digital PCR is being performed using RNA extracted from a normal cell with no gene fusion and using two assays, each targeting one fusion partner of interest. Depending on the number of available partitions and RNA templates, some partitions will contain templates from both potential fusion partners. In digital PCR, this is commonly referred to as multiple occupancy and the number of partitions that are ex- pected to randomly contain more than one template can be calculated using poison distri- bution statistics.

[0061] Now assuming RNA extracted from a cancer cell containing a gene fu- sion us being used in the same setup as above. In this case, the mRNA containing the mRNA encoding the fusion gene (and hence the target sequence for both assays) is being distributed as a single template and not individually as in the setup before. The result be- ing a number of double positive partitions that is significantly higher than what can be ex- pected based on poison statistics. This then indicated a non-random distribution which can only be explained by the presence of a gene fusion.

[0062] The inventors have tested this approach by reanalyzing their BCR-Abl dataset using their integrity calculation module in the QIAcuity Software suite. Here, the inventors tested pure Leucocyte RNA from a healthy donor and RNA extracted from K562 cells with two assays in a duplex reaction. Each assay targeted one of the fusion partners, in this case BCR (Exon 1) and ABL (Ex8/9). The results from the Leucocyte RNA show that when BCR and Abl are separated mRNAs the number of observed double positive partitions is nearly identical to the calculated expected (Figure 5, second column). When looking at K562 RNA (Figure 5, first column) the number of observed double positive parti- tions is 33% higher compared to the expected calculated number. This difference is due to the presence of the BCR-Abl fusion gene in the K562 cells.

[0063] In another embodiment the presence of a fusion gene event can be de- termined using a different approach. In a cell without a gene fusion, the absolute number of 5’ and 3’ fragments of each gene fusion partner should be equal. To demonstrate this the inventors have amplified both ends of both fusion partners of the BCR-Abl fusion on total RNA from normal leucocyte cells. As expected, the absolute counts of both, the 5’ and 3’ fragments of both genes are very close, indicating that in this particular sample, no BCR-Abl gene fusion is present and the experimentally observed differences are most likely sample variation (Table 3).

PCR and absolute counts compared to each other. The observed percentage dif- ference indicate no presence of a BCR-Abl fusion gene.

[0064] Now the idea is that the presence of a gene fusion event can be deter- mined by comparing the absolute counts of the 5’ and 3’ ends of both gene fusion part- ners to each other. When testing the RNA extracted from normal cells, the absolute counts of both partners should be significantly different when compared to each other. This is because the mRNA molecules of both potential fusion partners are intact and differ from each other when naturally differently expressed. As expected, when tested the 5’ end of the BCR gene together with the 3’ end of Abl gene in a duplex reaction the difference of absolute fragments was -450% as seen in Table 4, middle column. When the inventors used the same approach on RNA extracted from K562 cells harboring a BCR_Abl gene fusion, the observed number of fragments of BCR 5’ and ABL 3’ ends were highly similar (Table 4 right column). This is in stark contrast to the observed 450% in normal leucocytes and clearly indicated the presence of a BCR_Abl fusion transcript.

Table 4: The absolute number of the 5’ and 3’ end of the potential fusion partners was counted using digital PCR. Middle column, in RNA from healthy cells the counts are vastly different showing both genes being intact and differently ex- pressed. In comparison, in the presence of BCR_ABL the counts are very closely together as both fragments are expressed together.

4. Summary

[0065] The invention presented here aims to establish a general concept for a simple and rapid way to detect gene fusion events on the RNA level, preferably by using digital PCR. The concept is based on the notion that when a fusion event occurs, parts of the original (normal) mRNA will be expressed differently. This imbalance can then be ac- curately detected, preferably when using the exact quantification capability of digital PCR. Here the inventor uses the fusion gene BCR-ABL as an example. Using the approach de- scribed in this application, the inventor shows that in gene fusion events the 5' (front) end of ABL is differently expressed than the 3' (back) end of the mRNAscR or mRNA/iBL in total RNA extracted from a BCR-ABL expressing cell line (K562). The inventor has further found that in the case of gene fusions, the expressions of 5' (front) end of mRNAscR and the 3' (back) end of mRNA/iBL, as well as the 5' (front) end of mRNA/iBL and the 3' (back) end of mRNAscR, are approximately the same. The detection of the gene fusion can be done, for example, by means of an 'integrity calculation algorithm'. In addition, the inven- tors diluted the BCR-ABL containing RNA with increasing amounts of RNA derived from healthy leucocytes. These exemplary results show that this method can detect a BCR- ABL fusion event even in a background of 90% normal RNA. This means that with the method according to the invention can detect a tumor population of 10% in a background of 90% wild type RNA. When assuming a high tumor content, several samples can be pooled. [0066] This method according to the invention can be extended to all fusion gene events, provided the gene expression of both fusion partners is different after the fu- sion event and the genetic sequence allows the design of a specific primers and probes.

Claims

1. A method of detecting gene fusion events, comprising the following steps:

(i) providing a composition comprising mRNA molecules suspected of encod- ing at least a first gene A (PIRNAA) and at least a second gene B (mRNAs),

(ii) determining in the composition

- the amount of 5' ends of mRNAA ^ITIRNAA), and

- the amount of 3' ends of mRNAA ^ITIRNAA) and/or

- the amount of 5' ends of mRNAs (⁵'mRNAB), and

- the amount of 3' ends of mRNAs (³'mRNAB)

(iii) detecting a fusion of gene A and gene B if

⁵'mRNA_A # ³'mRNA_A or

⁵'mRNA_B # ³'mRNA_B or

⁵'mRNA_A « ³'mRNA_B or

³'mRNA_A « ⁵'mRNA_B.

2. The method of claim 1 , characterized in that in step (iii) a fusion of gene A and B is detected if

⁵‘ mRNAA is higher or lower ³‘ mRNAA or

⁵'mRNA_B is higher or lower ³'mRNA_B, preferably each by at least 20%, further preferably by at least 30%, further preferably by at least 50%, further preferably by at least 100%, further preferably by at least 150%, further preferably by at least 200%, further preferably by at least 250%, further preferably by at least 300%, further preferably by at least 350%, further preferably by at least 400%, further preferably by at least 450%, and further preferably by more than 450%.

3. The method of claim 1 or 2, characterized in that in step (iii) a fusion of gene A and B is detected if

4. The method of any of claims 1 to 3, characterized in that in step (ii) said determin- ing is carried out via polymerase chain reaction (PCR).

5. The method of claim 4, wherein said PCR is digital PCR (dPCR).

6. The method of claim 4 or 5, characterized in that in said PCR primers and/or probes are used capable of hybridizing to

- 5' end-located sections of mRNAAand

- 3' end-located sections of mRNAA and/or

- 5' end-located sections of mRNAsand - 3' end-located sections of mRNAs.

7. The method of claim 6, characterized in that the 5'-end-located sections are lo- cated upstream of the site of fusion of gene A and gene B, and the 3'-end-located sections are located downstream of the site of fusion of gene A and gene B.

8. The method of any of the preceding claims, characterized in that the composition is total RNA of biological cells suspected of carrying a fusion of genes A and B.

9. The method of claim 8, characterized in that the biological cells are tumor cells, preferably white blood tumor cells.

10. The method of claim 8 or 9, characterized in that the biological cells stem from one individual.

11. The method of claim 8 or 9, characterized in that the biological cells stem from at least two individuals.

12. The method of any of the preceding claims, characterized in that gene A is BCR or ABL1 and/or wherein gene B is ABL1 or BCR.

13. The method of any of claims 1-11 , characterized in that gene A is selected from the group consisting of EML4, Tmprss2, and CD74, and gene B is selected from the group consisting of ALK, ERG, and NRG1.

14. The method of any of claims 5-13, wherein in step (iii) the detection is made using an integrity calculation algorithm.

15. A kit for detecting gene fusion events between a first gene A and a second gene B, comprising: - PCR primers and/or probes capable of hybridizing to 5' end-located sections and 3' end-located sections of mRNA molecules encoding gene A (mRNAA) and/or gene B (mRNAB), and

- a manual comprising instructions for carrying out the method of any of claims 1- 14.

16. The kit of claim 15, characterized in that the 5'-end-located sections are located upstream of the site of fusion of gene A and gene B, and the 3'-end-located sec- tions are located downstream of the site of fusion of gene A and gene B.