Nucleophosmin (NPM1) is essential for the maintenance of a healthy cellular state. If mutated, it is one of the most common molecular causes for acute myeloid leukemia (AML). This review systematically analyzes the current usage and potentials of anti-NPM1 antibodies in diagnostics and research around AML. Inclusion and exclusion criteria were applied to an initial pool of 142 papers, which produced a total of 5 studies that were examined in detail. Anti-NPM1 antibodies have only successfully been used for diagnostic purposes within studies backed by standard means of detection, such as polymerase chain reaction (PCR). The findings across the studies either directly stated or suggested, that the anti-NPM1 antibodies are to some extent unspecific, and cross-reaction takes place, especially between the NPM1 mutant and NPM1 wildtype protein. With further validation and an increased specificity however anti-NPM1 antibodies have the potential to be an option for diagnostics and research. Anti-NPM1 antibodies may be used for rapid diagnostics at first diagnosis by integration into already existing immunophenotyping processes. Additionally, especially monoclonal antibodies with a high sensitivity, may be used for the diagnosis of minimal residual disease (MRD). In research anti-NPM1 antibodies could be used as part of single-cell approaches to unveil the leukemic lineage and identify the cells of origin (leukemic stem cells) of AML. Studies such as that will be essential to the development of more personalized medicines against AML.
The NPM1 gene is located in chromosome 5q35 and made up of twelve exons. In its physiological form is a nucleo-cytoplasmic shuttle protein, which is mainly located in the nucleus . It plays an important role in several cellular processes such as the biogenesis of ribosomes , the maintenance of genomic stability , the organization of nucleosomes and the duplication of centrosomes . Additionally different publications show that NPM1 interacts with different tumor suppressors and proto-oncogenes, such as Adenosine diphosphate-Ribosylation Factor (ARF) . The subcellular localization of the protein is essential for its physiological function. The protein has different domains that ensure the right concentration of protein in different cell compartments, by establishing a balance between import and export processes. However, mutations occurring in NPM1 lead to a permanent translocation of NPM1 into the cytoplasm, which is a key driving force of leukemic pathogenesis.
Almost all of the approximately 50 known NPM1 mutations are found in exon 12. Out of the mutations described type A is the most common one accounting for 75 – 80% of all NPM1 mutations. Types A, B, C and D are caused by the insertion of four base pairs causing the loss of two tryptophan moieties at position 288 and 290. In types E, F, G and H there is an insertion of nine base pairs, leading to the loss of the tryptophan moiety at position 290. Both tryptophan moieties are essential parts of the nucleolar localization sequence. Additionally, a leucin rich  nuclear export signal is created by the mutation, so that the protein is found in the cytoplasm at an unusually high rate . Further mutations causing aberrant cytoplasmic localization have recently been found in exons 5, 9 and 11 . This shift to the cytoplasm disturbs the proteins function as a shuttle protein thereby disrupting its physiological functions and unleashing its cancerous potential.
Acute myeloid leukemia (AML) is a clonal neoplastic disorder that arises from malignant transformation of haematopoietic stem and progenitor cells. This leads to a maturation arrest and aberrant proliferation of myeloid progenitor cells resulting in the accumulation of AML blast cells in the bone marrow. One of the most prevalent driver mutations of AML is a frameshift mutation in the NPM1 gene, which is present in approximately 30% of AML patients. Analysis of NPM1 mutated cells revealed a multilineage involvement of myeloid, monocytic, megakaryocytic and erythroid cells . This suggests either that the pool of NPM1 mutated cells is fueled by a common myeloid progenitor or that the NPM1 mutation constitutes a growth advantage in proliferation of myeloid cells unlike in lymphoid cells .
Over the last 40 years, the treatment of AML has largely been based on chemotherapy involving cytotoxic drugs like cytarabine and daunorubicin as well as stem cell transplantation. However, despite these intensive and toxic therapies, the majority of patients dies, with modest increases in 5-year overall survival in under 60-year-olds from 10% in the 1970s to around 40 – 45% in the 2000’s with complete remission occurring in 75 – 80% of patients following initial treatment  , highlighting the need for more effective and selective drugs. In particular, for NPM1-mutated AML, there is a clear lack of development of more specific targeted therapies. This could be attributed to the fact that the precise mechanism of how NPM1 mutations drive AML pathogenesis and which cells lead to relapse remains largely unknown.
Although NPM1 mutated AML is generally chemotherapy sensitive, still a high fraction of patients relapses, especially if they harbor co-mutations such as in FLT3 . It is therefore of clinical and prognostic relevance if and which cells carry a mutated NPM1 gene. In order to improve patient outcomes, it is essential to find out which cells cause AML in the first place and which cells evade current therapy regimes to specifically target them.
The goal of this review is to systematically review current knowledge and methods regarding diagnostics and research using anti-NPM1 antibodies. Core topics being covered are the different types of antibodies currently in use and different methods that use NPM1 antibodies. Limitations will be discussed and potential new uses will be explored.
A thorough literature search involving AML and NPM1 was conducted at the beginning of the project. The literature review was performed between 15th December 2021 and 18th December 2021. The search terms “NPM1”, “AML” and “antibody” constituted the baseline of the research. Later the search was expanded with the terms “flow cytometry” and “immunohistochemistry”. These preliminary searches resulted in 142 results.
The following criteria were applied for the review:
Only English and German studies were recognized.
A protocol for the use of the antibody is available.
The antibody is anti-human.
No reviews were included.
The antibody was used for flow cytometry and / or immunohistochemistry.
The full paper is available.
After these criteria were met, the publication’s reliability, completeness and relevance was determined individually.
Initially 142 results were found. Of these papers 102 were related to flow cytometry, 30 were related to immunohistochemistry and 10 to both. After application of inclusion and exclusion criteria I was left with 11 studies. In a last step the individual relevance of each paper was examined. An example of this examination was the exclusion of some papers that did have a protocol, which was however found to be incomplete upon closer inspection. In summary the literature search yielded 5 papers.
NPM1 mutations are routinely detected by polymerase chain reaction (PCR). Further common means of detection are high resolution fragment analysis, melting curve analysis and denaturing high-performance liquid chromatography . Additionally, NPM1 mutations with aberrant cytoplasmic localization coupled with fluorescent proteins were detected by immunohistochemistry during different studies. The scientific community however has been looking for additional methods for diagnostics and research of NPM1 mutations. The NPM1 protein and / or its mutant form may be specifically labeled using antibodies. These antibodies are either conjugated with a fluorochrome themselves or a secondary conjugated antibody will be added to them. The fluorochromes are then excited by a laser with a certain wavelength. They will then emit light with a known wavelength itself which is then detected. This way cells labeled by the antibody can be distinguished from other cells. Overall, the use of NPM1 specific antibodies in flow cytometry and immunohistochemistry can be divided into four possibilities: Polyclonal as well as monoclonal antibodies have been used by different groups. Of these antibodies some were NPM1-mutant specific while others targeted the wild type as well as the mutated protein.
Oelschlegel and colleagues have devised a protocol for the flow cytometric detection of NPM1 with aberrant cytoplasmic localization. They were testing five monoclonal antibodies, which detected both the wild type and mutant form of the protein. The clone 5E3 antibody, directed at the C-terminus of NPM1, achieved the highest difference in mean fluorescence intensity and was therefore chosen for all further experiments. Eleven out of 298 samples were discrepant from the PCR results (6 false positive and 5 false negative) .
Furthermore, two groups have independently developed monoclonal antibodies specific to NPM1 mutants only, the results however were inconsistent between the publications. Gruszka and colleagues produced an antibody (T26) specific to the mutated C-terminus of the most common mutations (95% of all NPM1 mutations). It produced specific results in immunohistochemistry and flow cytometry applications, with a minimum percentage of blasts necessary for the mutation to be detected of 0,001% . Tan and colleagues reported, that although monoclonal antibodies were produced and used in immunohistochemistry, they did cross-react with the wild type form of NPM1 . Gruszka and colleagues included Fc-blocking in their protocol , while Tan and colleagues did not take it into account .
Du Pisani and colleagues were using a rabbit polyclonal antibody from Abcam, UK, which specifically binds to the mutated C-terminus of NPM1. A secondary conjugated goat anti-rabbit antibody was used for the detection of the bound primary antibody. NPM1 mutated and non-mutated cell lines as well as patient samples were reliably distinguished, if they contained more than 10% NPM1 mutated cells. For the distinction between mutated and wildtype NPM1 the normalized mean fluorescence intensity (mean fluorescence intensity compared to the wild type negative control) rather than the mean fluorescence intensity was used. An important economical finding of this study was the fact that the flow cytometric assay was ten times cheaper than a standard cytogenetic panel .
Recently El-Gamal and colleagues were using a polyclonal antibody from NOVUS Biologicals, since the antibody manufactured by Abcam is not commercially available anymore. In the published paper NPM1-mutant and non-mutant AML was distinguished using two individual statistical methods: Samples were analyzed using the normalized mean fluorescence intensity and the percentage of cells expressing the NPM1 mutant. In both cases cut-off values were established to distinguish the mutant from the wildtype protein. The obtained results were compared to PCR. Both means of analysis produced discordant results in four out of 89 cases (percentage NPM1 mutant positive cells: 1 false negative, 3 false positive; normalized MFI: 4 false positive). The group proposed a cross-reaction of the antibody between wild type and mutant NPM1 as a reason for the false positive results .
NPM1 antibodies may in the future be used for a flow cytometric NPM1 mutational screening. AML blasts undergo immunophenotyping at first diagnosis. The protocols followed by the studies introduced in this review can easily be integrated into this process with just a few extra steps. No further training or specialized equipment and personnel is required. NPM1 mutational status may also be examined using the introduced antibodies in immunohistochemistry, which would however be more time consuming. The defining factor of AML is the presence of 20% or more blasts in the bone marrow or in the peripheral blood. The monoclonal antibody raised by Gruszka and colleagues  and the polyclonal antibody manufactured by Abcam  may be used at first diagnosis due to their high sensitivity. The other publications did unfortunately not specify the minimum percentage of cells with NPM1 mutation needed for the mutation to be detected and therefore further validation would be necessary.
Anti-NPM1 antibodies may also be used in follow-up examinations and the determination of minimal residual disease (MRD). The minimum percentage of mutated cells for the mutation to be detected using the assays introduced ranges over a wide range (0,001% to 10%). The commonly used MRD detection method in the case of AML with NPM1 mutation is real-time quantitative PCR. It detects the disease if 0.1 to 0,001% of the cells are mutated . The antibody introduced by Gruszka and colleagues  may therefore be used in an MRD flow cytometric assay due to its high sensitivity, which lies within the range of sensitivity for commonly used MRD detection methods. The Abcam antibody used in the flow cytometric assay of Du Pisani and Shires  should not be used since its sensitivity is too low and because it is not yet clear whether cross-reaction occurs. The antibodies used by Tan and colleagues  as well as Oelschlegel and colleagues  should not be used either because they react with the wildtype or mutated protein. The NOVUS Biologicals antibody used by El-Gamal and colleagues  needs further validation since the minimum percentage of cells with NPM1 mutation needed for the mutation to be detected was not disclosed.
Furthermore, a deletion or insertion mutation in exon twelve is strongly associated with a normal cytogenetic karyotype  and they seem to be mutually exclusive with mutations in Runt-related transcription factor 1 (RUNX1) and CCAAT / enhancer-binding protein alpha (CEBPA) . This may render cytogenetic analysis obsolete in cases where mutated NPM1 is detected using anti-NPM1 antibodies.
An important economic outcome of the studies examined is that, the cost of a flow cytometric assay was reported to be ten times cheaper than a standard cytogenetic panel thus providing a huge saving potential if performed on the large scale .
Although knowledge about the role that NPM1 plays in the development of AML has recently been expanding leading to more accurate prognosis and despite an increase in treatment options, the origin of the mutation amongst the cell lineage itself remains unknown. In order to continue to improve patient outcomes by personalizing therapy options to the individual AML entity it is essential to find the leukemic stem cells that cause AML at first diagnosis and relapse. Anti-NPM1 antibodies could be the key tool in unveiling the leukemic lineage for AML populations. They could be used to specifically mark leukemic cells, which could then be sorted using fluorescence-activated cell sorting (FACS). This enables scientists to target and analyze the leukemic cells specifically rather than all cells provided in a patient sample. Concentrated NPM1 mutated cell populations could then be further analyzed using single-cell techniques as Tirana and colleagues  have used at different time points such as first diagnosis, with MRD and at relapse. For this purpose, very high sensitivities must be achieved. MFI and percentage of positive cells are irrelevant since each cell is analyzed individually rather than on a population level. New analysis protocols are therefore necessary. Cut-off values for each subpopulation of cells (e. g. blasts, leukemic stem cells, lymphocytes, macrophages, etc.) would have to be established to determine which cells are NPM1 mutated and which populations represent wild-type populations.
Anti-NPM1 antibodies have successfully been used in immunohistochemistry and flow cytometry to distinguish NPM1-mutant from NPM1-wildtype AML. The reliability of the results of flow cytometric and immunohistochemistry analysis ranges from 100% concordant results in smaller study populations to approximately 95% in larger study populations compared to the current gold standard PCR. However, the findings suggest that the mutant-specific monoclonal and polyclonal antibodies may cross-react with the mutant and wild type NPM1. This is highlighted by T26 . This antibody provided the most promising results of all the antibodies introduced in this paper, yet it has not made the leap from the lab to being commercially available in more than ten years, which is most likely due to a lack of reproducibility of results. The mutant specific antibody manufactured by Abcam used by Du Pisani and Shires  is not commercially available anymore, which further supports this theory. Considering all data included in this review it is safe to say, anti-NPM1 antibodies should only be used in diagnostics, such as AML rapid diagnostics, within clinical trials backed by conventional reliable methods of detection such as PCR, since patient lives are directly affected by the results of these analyses. After optimization of staining protocols however, anti-NPM1 antibodies may be the key for new advancements, especially in single-cell approaches, to unveil the cell lineage of AML and pave the way for personalized medicines to target the disease and cells remaining after conventional therapy more specifically.
The author declares no conflict of interest.
No raw data was generated for this manuscript.