A Review of CEBPA's Role in Hereditary Leukemia

ABSTRACT

Background

Advances in genetic testing have allowed for extensive testing and identification of previously unrecognized inherited conditions; a prominent area of this research includes familial patterns of acute myeloid leukemia (AML). Recognition of familial related hematologic malignancies can have lasting implications for both patients and their treatment plans, as well as their families.

Aims

This review summarizes the specific familial AML disease, CCAAT/enhancer binding protein‐alpha (CEBPA) associated AML.

Content

We highlight CEBPA's function, pathogenesis, and potential treatment considerations after identification of the pathogenic or likely pathogenic (P/LP) germline mutations in cases of AML. Discussion of testing and screening for affected family members is also reviewed.

Summary

Identification of CEBPA familial AML is necessary for appropriate treatment planning and has additional implications for familial testing and screening. Clinical practice is still evolving treatment paradigms for these patients and their families.

Trial Registration: The authors have confirmed clinical trial registration is not needed for this submission

1. Introduction

With the rapid advances of the whole genome sequencing leading to increased genetic testing of tumor samples, the recognition and diagnosis of familial hematologic malignancies, and especially myeloid malignancies, has exploded. Multiple P/LP germline mutations have been identified in patients with AML that are considered to have contributed to their disease course. Careful evaluation of a patient's family medical history, as well as the standard of care genetic sequencing, is imperative when evaluating a patient with new AML. As per National Comprehensive Cancer Network (NCCN) guidelines, genetic testing is recommended to assess for hereditary myeloid malignancy in patients suggestive of a genetic predisposition, newly diagnosed aplastic anemia, hypocellular myelodysplastic syndrome (MDS), as well as those aged less than 50 years with MDS or AML. The Surveillance, Epidemiology, and End Results Program (SEER) cites that the incidence of patients diagnosed under the age of 55 is 22.2%, representing a significant portion of patients [1]. The identification of patients with a familial AML syndrome is paramount to establishing an appropriate treatment plan and prognosis, as well as having marked implications for a patient's family members. Here, we describe one of the well‐documented genes associated with familial AML, CCAAT/enhancer binding protein‐alpha (CEBPA). This review will discuss normal function of the CEBPA protein, resultant implications for P/LP germline mutations, as well as treatment outcomes and potential strategies for patients with confirmed hereditary leukemia.

2. Normal Function and Pathophysiology

The CCAAT/enhancer binding protein (C/EBP) transcription factors are members of the basic leucine zipper (bZIP) family of transcription factors, and have been characterized as key regulators of myeloid development and activity [2]. They function by dimerizing with other transcription factors and activate transcription by binding at the consensus sequence 5'‐TT/GNNGNAAT/G‐3' on the promoter regions of target genes [3]. These proteins are expressed in various degrees in a multitude of different tissues; they are vital in the development of normal tissue, cellular function, as well as cellular proliferation and differentiation [4]. While there are six members in this family of transcription factors, mutations in one specific transcription factor, CEBPA, have been associated with profound hematopoietic abnormalities [4]. The CEBPA gene is located on the long arm of chromosome 19q13.1; its bZIP domain is notably at the C‐terminal, amino acids 272‐358 [5]. It encodes two protein isomers, a full‐length 42 kDa protein with full transcription activity, and a shorter 30 kDa isoform [6]. The truncated isomer has been shown to have an inhibitory effect on the full‐length peptide. These proteins function to activate transcription from lineage‐specific gene promoters and regulate the expression of many myeloid genes, including granulocyte colony‐stimulating factors [4, 7, 8]. CEBPA has also been implicated in the overall granulocyte differentiation of myeloid precursors by upregulating miR223 [7]. The majority of CEPBA expression in the myeloid lineage has been seen in hematopoietic stem cells, myeloid progenitors, and granulocytes.

Early data regarding CEBPA's role in cellular differentiation initially resulted from mouse knockout models. Mice with targeted disruption of CEBPA showed a selective block in the differentiation of neutrophils, and most of the white cells in the peripheral blood had the appearance of myeloid blasts [9]. Further investigation into the downstream effects of CEBPA silencing with hypermethylation revealed a reciprocal upregulation of CCAAT/enhancer binding protein gamma (CEBPG). Increased expression of CEBPG demonstrated a block in granulocyte differentiation and subsequent myeloid cell arrest [10]. Downstream effects of CEBPA knockout lead to further investigation within myeloid malignancies, specifically AML, as a driver mutation of the disease.

3. CEBPA Mutations in AML Pathogenesis

CEBPA mutations are found in 5%–14% of all cases of AML [11], most often seen with a normal karyotype, and both monoallelic (CEBPA^sm) and biallelic (CEBPA^dm) mutations can occur. Most mutations occur at two separate locations in the gene. Mutations in the N‐terminal result in translation of a 30 kDa protein, which has a dominant negative effect over the full p42 protein [12]. Other mutations occur as C‐terminal in‐frame insertions/deletions in the DNA binding or leucine zipper domain, and disrupt binding and dimerization [12, 13].

CEBPA was previously identified as a potential pathogenic gene for hereditary AML in 2004, after an analysis of 187 patients with AML found that nearly 10% had mutations in CEBPA. Two of the patients were found to have P/LP germline N‐terminal frameshift CEBPA mutations and additional somatic mutations in CEBPA; this suggested that the P/LP germline CEBPA mutations predispose patients to AML but require additional somatic mutations before development of the disease [14]. Another well‐known example of hereditary AML was published in 1978 [15]; in a family of 239 members, and across three generations, 13 individuals had been diagnosed with AML. At the time, the authors were unable to identify the genetic component of the disease; however, they concluded that germline inheritance was the likely etiology. It wasn't until a member of a later generation had been diagnosed with AML that subsequent investigators were able to analyze a peripheral blood, somatic DNA sample.[16]. This revealed a heterozygous base pair deletion (c.68delC) in the N‐terminal of CEBPA, as well as a three‐base pair duplication (c.937_939dupAAG) in the C‐terminal of CEBPA. This suggested that the N‐terminal mutation had been inherited, as the base pair duplication was only present in a proportion of the peripheral blood cells, and thus deemed to be likely an acquired mutation [16]. In 2015, there was an in‐depth investigation into 10, CEBPA –mutated families; 24 of these members had been diagnosed with AML [17]. Whole exome and deep sequencing were performed, confirming that the P/LP germline CEBPA mutations were clustered within the N‐terminal. All 18 diagnostic tumors tested had CEBPA^dm, with second mutations primarily located at the C‐terminal.

4. Penetrance

In a number of familial studies, germline mutations in the N‐terminal of CEBPA have been shown to be highly penetrant and inherited in an autosomal dominant pattern [18, 19]. These patients will also tend to have a younger age of onset with AML, as young as 1.8–50 years old [20]; pathogenesis ultimately depends on the duration of time until additional mutations occur. In the previously mentioned analysis of ten different CEBPA‐affected families, there were three unrelated adult carriers who remained unaffected at the time of evaluation [17]. This raised the possibility of variations in penetrance; however at the time of investigation, the individuals were 41, 24, and 19 years of age, remaining still within a time frame to develop AML.

5. Co‐Occuring Mutations

Multiple studies have sought to further identify possible deleterious mutations that co‐occur with CEBPA^dm. A study published in 2012 used whole exome sequencing on five patients with known CEBPA^dm; they detected somatic GATA2 zinc finger one mutations in 2 of the 5 patients [21]. GATA2 is a zinc finger transcription factor that is important in the proliferation of hematopoietic stem cells; mutations in GATA2 have also been implicated in case of familial AML [21, 22]. Based on these whole‐exome sequencing results, a further analysis was performed with mutational screening in 33 CEBPA^dm patients. Of these, 13 patients (39.4%) had additional heterozygous missense, somatic mutations in the coding Exon 4 of GATA2's N‐terminal zinc finger domain. Interestingly, the germline‐associated GATA2 mutations are instead described at a separate, C‐terminal zinc finger domain [22]. An analysis published in 2016 that sought to further investigate many driver mutations of AML showed results that verified the co‐occurrence between CEBPA^dm and GATA2 [23]. It is theorized that this relates to a unique type of leukemogenesis for this subset of patients.

While the GATA2 co‐mutation appears to be the most common co‐occurrence with CEBPA^dm, other genes have also been implicated as well. Other relatively common somatic mutations that have co‐occurrences with CEBPA^dm are CSF3R, TET2, and WT1 mutations [24, 25]. Additional mutations have been found to occur mutually exclusive with CEBPA^dm, including FLT3, DNMT3A, IDH 1/2, NPM1, and RUNX1 [24].

In multiple evaluations of patients with CEBPA^dm and GATA2 mutations, no change in clinical presentation nor worsening of clinical outcome was found [26, 27, 28], though the data is not yet conclusive [24, 29].

6. Outcomes

AML with in‐frame mutations affecting bZIP domain of CEBPA (CEBPA^Bzip) have most recently been categorized as an AML with recurrent genetic abnormalities by the World Health Organization's 2022 report. If a patient were to present with this mutation, they would only require 10% blasts in the bone marrow or peripheral blood in order to be diagnosed with AML. Notably, this was changed from prior classification requiring CEBPA^dm mutations; it now allows for single mutations if the mutation affects this specified region. This change was made after multiple studies revealed that the unique nature of the bZIP mutations that define the prognostic benefits and genetic expression [30]. These mutations, either sporadic or germline, are categorized as favorable prognosis without the presence of additional mutations. Even so, there are significant differences in overall outcomes between patients with germline CEBPA mutations, CEBPA^sm, and CEBPA^dm.

Investigation into survival outcomes showed that the overall survival (OS) in familial CEBPA AML at 10 years was superior to sporadic CEBPA^sm (57% vs. 29%, respectively, p = 0.003), but there was no difference between familial and CEBPA^dm [17]. The rate of first remission for familial AML was also favorable at 91%, in 21 out of 23 patients; however, patients often experienced late “relapse” with retained sensitivity to chemotherapy. For patients with familial CEBPA AML with “relapsed” disease, the median survival was 8 years; median survival was 16 months for patients with sporadic AML at relapse [31]. The team went on to investigate whole genome sequencing and deep sequencing for patients with CEBPA familial AML at “relapse.” Their evaluation revealed novel mutations that instead suggest these patients had a second, de novo AML, unrelated to their initial leukemia. This differs from somatic CEBPA^dm, where relapse often harbors the same CEBPA mutations from diagnosis to relapse [23].

7. Treatment of CEBPA Familial AML

As AML with CEBPA^Bzip without additional cytogenetic abnormalities is classified as better risk disease, treatment modalities focus on achieving cure with chemotherapy alone. In regard to patients with familial CEBPA AML, while chemotherapy may induce a prolonged remission, the P/LP germline mutation remains. As discussed above, familial CEBPA AML has high rates of a late, second leukemia, and additional de novo mutations at the time of relapse can occur, which may alter prognosis of relapsed disease [31]. Allogeneic hematopoietic stem cell transplantation works not only to cure patients of AML, but also to rid the patient of the stem cell population harboring this mutation [32]. Given this, some have argued to pursue transplant in first remission of these patients [33].

When evaluating potential donors, careful genetic testing should be considered in siblings or other family members. If the donor harbors the same germline mutations, it could result in donor‐derived leukemia, as demonstrated in a case published by Xiao et al. This report in 2011 was the first showcasing that multiple CEBPA mutations contributed to a donor‐origin leukemia after allogeneic transplantation [34]. Careful consideration of genetic testing should be performed in patients who have a family history of AML or who harbor CEBPA^dm mutations.

8. Implications for Family

When a P/LP germline mutation is identified in a patient, it is important to discuss testing for all at‐risk family members and refer family members for genetic counseling [35]. Expert consensus suggests that once an individual is diagnosed with a leukemia predisposition syndrome, they should obtain consultation with a hematologist/oncologist, genetic counseling, baseline complete blood count and differential, as well as bone marrow aspirate and biopsy with cytogenetic and molecular testing [36]. Continued clinical evaluation at a regular frequency is also recommended to assess for signs or symptoms of malignancy, generally every 6–12 months [36]. There has also been suggestion of using sequential next generation sequencing (NGS) in these patients to provide insight into disease processes and help inform patient and family member care [37]. In addition, in partnership with the American Society of Hematology, ClinGen created a specific Myeloid Malignancy Variant Curation Expert Panel to encourage standardization and variant reporting [35].

9. Conclusion

While the recognition of familial AML has only increased in recent years, clinical practice is still evolving treatment paradigms for patients and their families. P/LP germline mutations in the bZIP region of CEBPA are highly penetrant and inherited in an autosomal dominant pattern. After identification of CEBPA^Bzip mutations with genetic sequencing, careful assessment of family history is imperative prior to treatment planning. As stated, CEBPA^Bzip AML has a favorable prognosis and is traditionally treated with upfront high‐dose chemotherapy without plans for stem cell transplant. However, patients with familial CEBPA AML who pursue up‐front chemotherapy often are plagued by a second, de novo leukemia later in their lives. Careful consideration of treatment strategy should be given to these patients, as allogeneic hematopoietic stem cell transplant in first remission can provide cure with added benefit of P/LP germline mutation removal in the stem cells. Given the recent changes in the ELN 2022 Criteria with CEBPA in frame bZIP mutations replacing biallelic CEBPA mutations in the favorable category AML, special attention is necessary in this population. Co‐occurrence of associated mutations should also be considered in treatment planning, as they may impact overall prognosis. Further investigations are needed for all types of familial AML to continue to help educate providers, patients, and their family members regarding these complex inherited hematologic malignancies.

Funding

The authors have nothing to report.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The authors have nothing to report.