Inherited Bone Marrow Failure Syndromes

Background

FA is inherited in an autosomal recessive manner in 98% of cases. In rare cases, it is transmitted in an X-linked recessive or autosomal dominant mode.

Although the original report of FA in 1927 by Dr. Guido Fanconi described pancytopenia combined with physical anomalies in three brothers, many publications thereafter have underscored the clinical variability of the condition. FA is a genomic instability disorder characterized by chromosomal fragility and breakage, a defect in DNA repair, progressive BM cell underproduction, peripheral blood cytopenias, developmental anomalies, and a strong propensity for hematologic and solid tumor cancers.

FA patients may present with either physical anomalies but normal hematology, or normal physical features but abnormal hematology, normal physical features and normal hematology, or physical anomalies and abnormal hematology ( Fig. 30.1 ). There can also be sibling heterogeneity in presentation with discordance in clinical and hematologic findings, even in affected monozygotic twins. Using published information, the median age at diagnosis of FA is about 6.5 years with a reported range from birth to 49 years. However, advances in genetic diagnostic tools may result in early diagnosis of young siblings or adults who do not have apparent clinical manifestations.>

Epidemiology

The overall prevalence of FA is 1 to 5 cases per million with a carrier frequency of 1 in 200 to 300 in most populations. Data from the CIMFR showed a prevalence of 11.4 cases per million live births per year. It occurs in all racial and ethnic groups. Spanish Gypsies have the world’s highest prevalence of FA with a carrier frequency of 1 in 64 to 1 in 70 for a common founder mutation. A founder effect has also been demonstrated in Afrikaners in South Africa in whom one specific mutation is common (frequency, 1 in 83), as well as in Ashkenazi Jews (1 in 89), Moroccan Jews, Tunisians, sub-Saharan African Blacks, Indian, Israeli Arabs, Brazilians, and Japanese.

Genetics

FA genes are all involved in DNA repair. The first clue for this defect was the abnormal chromosome fragility that is readily seen in metaphase preparations of peripheral blood lymphocytes or skin fibroblasts cultured with phytohemagglutinin (PHA) and is enhanced by adding a DNA interstrand cross-linking agent, such as mitomycin C (MMC) or diepoxybutane (DEB) (see Abnormal Chromosome Fragility section later). This feature was utilized to discover the first FA genes by complementation analysis. Complementation is considered when fusion of FA patient cells with cells from an individual who does not have mutations in the same gene (i.e., cell hybridization with cells from a healthy subject or a FA patient with mutations in another FA gene) result in correction of MMC or DEB hypersensitivity in growth inhibition or chromosomal fragility assays. Complementation has also been determined by transducing known FA gene cDNA into ungenotyped patient cells. The mutated gene was determined by a failure of fused cells with known mutation to correct (complement) the abnormal chromosome breakage in the patient’s T cells in culture on exposure to DEB/MMC. This is why FA genes are called FANC genes (FA complementation). Recently, next generation gene panel assays replaced complementation testing for identifying the mutant FA gene, and whole exome/genome sequencing replaced the complementation testing for FA gene discovery. So far, 22 genetic groups (termed types A, B, C, D1, D2, E, F, G, I, J, L, M, N, O, P, Q, R, S, T, U, V, W) have been proposed (see Table 30.1 ). The first FA gene, FANCC , was discovered in 1992 in Toronto, and then the other genes, corresponding to each of the other complementation groups, were subsequently cloned: FANCA, FANCB, FANCC, FANCD1/BRCA2, FANCD2, FANCE, FANCF, FANCG, FANCI, FANCJ/BACH1/BRIP1, FANCL, FANCM, FANCN/PALB2, FANCO/RAD51C, FANCP/SLX4, FANCQ/ECCR4, FANCR/EXCC2 , and FANCS/BRCA1, FANCT/UBE2T, FANCU/XRCC2, FANCV/REV7, FANCW .

The most commonly mutated genes based on data from the International Fanconi Anemia Registry (IFAR) are FANCA (∼61%), FANCC (∼16%), and FANCG (∼10). As of 2020, the most commonly used tool for genetic investigation of FA patients is by either next generation sequencing gene panel assays or single gene analysis. Partial or complete gene deletions are common in FA; hence, analysis of deletions should always be considered when nucleotide sequencing does not reveal pathogenic variants. Deletion analysis can be done by either subjecting next generation sequencing gene panel data to special copy number variation analysis software programs, or by utilizing multiplex ligation-dependent probe amplification) or microarray copy number variation (CNV).

Pathophysiology

Pathogenesis of Bone Marrow Failure

Hematologic abnormalities in FA are evident at the hematopoietic stem and progenitor cell (HSPC) level with marked reduction in both multipotent (HSC, and multipotent progenitors, MPP) and oligopotent cells (common myeloid progenitors, CMP, megakaryocyte-erythroid progenitors, MEP), but a modest reduction in granulocyte-monocyte progenitors (GMP). Cure of FA BM failure by HSC transplantation (HSCT) supports the hypothesis that the hematopoietic defect starts at the HSC level. Clonogenic assays show reduced frequencies of CFU-E (colony-forming unit-erythroid), BFU-E (burst-forming unit-erythroid), and CFU-GM (colony-forming unit-granulocyte macrophage) colony-forming cells in almost all patients after aplastic anemia ensues as well as in a few patients before the onset of aplastic anemia. Besides low progenitor numbers, decreased colony numbers in these studies can also be interpreted as faulty proliferative properties which lead to an inability to form colonies in vitro. Indeed, there is a defective proliferative response of CFU-GEMM (colony-forming unit granulocyte, erythrocyte, macrophage, megakaryocyte), BFU-E, and CFU-GM progenitors to GM-CSF plus stem cell factor (SCF) (c-kit ligand) or to IL-3 plus SCF.

Additional factors are operative in FA BM failure. Telomeres, the non-encoding DNA at each end of chromosomes, shorten with each round of cell division in normal human somatic cells. Their length is a reflection of the mitotic history of the cell. Telomerase, a ribonucleoprotein reverse transcriptase that can restore telomere length, is variably present in hematopoietic progenitors. Leukocyte telomere length is significantly shortened in FA patients despite increased telomerase activity. In parallel, increased BM cell apoptosis has been demonstrated in FA patients and in knock-out mouse models and is mediated by Fas, a membrane glycoprotein receptor containing an integral death domain. FA cells exposed to tumor necrosis factor-α (TNF-α), interferon-γ (INF-γ), MIP-1α, Fas ligand, and double-stranded RNA undergo exaggerated apoptotic responses.

Studies of cytokines in FA patients have shown varied abnormalities. FA fibroblasts showed no deficiencies in SCF or M-CSF (macrophage colony-stimulating factor) production. Importantly, IL-6 production was found to be reduced in FA patients and TNF-α generation was markedly increased.

Initial attempts to generate induced pluripotent stem cells (iPSCs) from FA patients have been difficult since reprogramming causes increased DNA double-strand breaks and the FA pathway needs to be activated. This barrier could be bypassed by either correcting the genetic defect before reprogramming or performing the reprogramming under hypoxic conditions. Successful reprogramming resulted in cells that recapitulate the hematopoietic defect and identify the early pathogenetic defect at the stage of hemoangiogenic progenitors.

Interestingly, transforming growth factor-β (TGF-β) signaling was found to suppress FA cells. Blocking this pathway improved the survival and proliferation of HSPCs derived from FA mice and from FA patients. Further, inhibition of TGF-β signaling in FA HSPCs resulted in elevated homologous recombination repair with a decrease in non-homologous end-joining.

Clinical Features

The diagnosis of FA can readily be made based on signs and symptoms related to aplastic anemia and the presence of characteristic congenital physical anomalies. However, about 30% of the patients have no physical anomalies, and about 25% may be diagnosed with FA based on physical anomalies without yet developing cytopenias. Interestingly, family screening may detect affected family members who have neither physical malformations nor cytopenias at the time of diagnosis (about 7% of the patients).

Table 30.2 lists the characteristic physical abnormalities and their approximate frequency based on more than 2000 published case reports. The two most common anomalies are skin hyperpigmentation and short stature , each with a frequency of 40% of cases. Characteristically, the hyperpigmentation is a generalized brown melanin-like splattering that is most prominent on the trunk, neck, and intertriginous areas that becomes more obvious with age. Café-au-lait spots are also common. Hypopigmentation and vitiligo may also be seen. In the minority of cases with short stature, growth failure is associated with endocrine abnormalities. In one report, spontaneous overnight growth hormone secretion was abnormal in all patients tested, and 44% had a subnormal response to growth hormone stimulation. Approximately 40% of patients also have overt or compensated hypothyroidism, sometimes in combination with growth hormone deficiency.

Malformations involving the upper limbs are common, especially hypoplastic, supernumerary, bifid, or absent thumbs. Hypoplastic or absent radii are always associated with hypoplastic or absent thumbs in contrast to TAR syndrome in which thumbs are always present. Less often, anomalies of the feet are seen, including toe syndactyly, short toes, a supernumerary toe, clubfoot, and flat feet. Congenital hip dislocation and leg abnormalities are occasionally seen. Male patients often have gonadal and genital abnormalities, including undescended, atrophic, or absent testes, hypospadias, an underdeveloped penis or micropenis, phimosis, and an abnormal urethra. Female patients occasionally have malformations of the vagina, uterus, or ovary. Renal anomalies such as ectopic, pelvic, or horseshoe kidneys are detected often, as are duplicated, hypoplastic, dysplastic, or absent organs. Occasionally, hydronephrosis or a hydroureter is present.

Many patients have a “ Fanconi facies,” and unrelated patients can resemble each other almost as closely as siblings. The head and facial changes vary but commonly consist of microcephaly; small eyes; epicanthal folds; and abnormal shape, size, or positioning of the ears (see Fig. 30.1 ). Anomalies in the tympanic membrane and middle ear ossicles are seen in almost 70% of patients, resulting in hearing loss in most affected patients. Approximately 10% of FA patients have cognitive deficiencies.

Laboratory Findings

Predisposition to Malignancy

A major feature of the FA phenotype is the propensity to develop cancer. The chromosome fragility, defects in DNA repair, genomic instability, oxidative stress, and the cellular damage that occur in FA patients translate into a significant predisposition to develop a malignancy. Because many genes are associated with FA and because alterations in the FA pathway are relevant to the pathogenesis of common types of cancers, the disorder is a critical human model of the genetic determinants of hematologic cancers and solid tumors.

FA can be considered as a member of two families of cancer predisposition syndromes. The first is composed of genetic disorders of DNA repair that include ataxia telangiectasia, xeroderma pigmentosum, and Bloom syndrome. The close relationship between FA and these syndromes is underscored by data showing convergence of signaling pathways in these conditions and the identification of ERCC4 mutations in patients who manifest a complex phenotype of FA, xeroderma pigmentosa and Cockayne syndrome. The second family of predisposition syndromes consists of other inherited BM failure disorders described herein, including SDS and DC that have a propensity for malignant myeloid transformation or development of solid tumors.

The magnitude of the risk of developing malignancy in FA has been evaluated in several studies including literature reviews and analyses of registry data. In the National Cancer Institute study the observed to expected ratio of any malignancy was 19 in FA patients who have not received HSCT and was 55 in patients who received HSCT. The median patient age for the development of all cancers in a literature review was 16 years of age, which is strikingly different from the median age of 68 years for the same types of cancer in the general population. The risk of developing solid tumors in FA increases with advancing age.

In a study of the Canadian cohort, the cumulative risk of clonal and malignant myeloid transformation (of MDS and AML) after censoring patients transplanted for severe BM failure was 75% by the age of 18 years. The IFAR data indicate that the risk of acquiring clonal cytogenetic abnormalities was 67% by 30 years of age; the actuarial risk of MDS or AML was 52% by 40 years of age; and the cumulative incidence of leukemia was 33% by 40 years of age. This steady tempo of leukemic evolution implies a stepwise acquisition over time of additional, critical genetic “hits” before overt MDS/AML develops. In a literature review of 320 published FA cases, 14 patients developed leukemia. The risk of MDS was estimated as 5% to 10% and of leukemia as 5% to 10%. Studies from the IFAR showed that the risk of developing MDS and AML was higher for patients in whom a prior clonal BM cytogenetic abnormality had been detected. Monosomy 7, rearrangement or partial loss of 7q, rearrangements of 1p36 and 1q24-34, and rearrangements of 11q22-25 are frequent recurring cytogenetic abnormalities. Additional data indicate a strong correlation in FA BM cells of chromosome 3q26q29 partial trisomies and tetrasomies and rapid progression to MDS or AML. When interpreting the significance of clonal cytogenetic abnormalities in FA patients, note that clonal variation is frequent, including appearances of new clones, inability to detect established clones on repeat examination, and clonal evolution. In a study from France the investigators used SNP arrays to analyze whole marrow cells of FA patients with MDS/leukemia. They identified a relatively high frequency of somatic RUNX1 gene disruption compared to that typically seen in patients with de novo MDS/AML. In a study of FA patients without apparent MDS/AML the GMP-cell population was reduced, but to a lesser degree than other multipotent and oligopotent progenitors. These cells harbored higher mutation rates than control subjects with a nucleotide variant signature that was similar to that seen in AML; increased G>A/C>T variants, decreased A>G/T>C variants, increased trinucleotide mutations at Xp(C>T)pT, and lower mutation rates at Xp(C>T)pG sites compared to other Xp(C>T)pX sites. The same pattern was seen in blast cells from a patient with FA and AML. These data suggest that the CFU-GM represent a cellular reservoir for clonal evolution.

The IFAR data indicates that by the age of 40 years, the cumulative incidence of nonhematologic cancers is 28%. In a literature review of 320 published FA cases, 25 non-transplanted cases had one to three separate types of solid tumors. Among the NCI cohort 21 of 163 FA patients developed solid tumors with an observed/expected ratio of 19. The most frequent solid tumor reported was squamous cell carcinoma involving the head and neck (mostly tongue) and upper and lower esophagus followed by the vulva or anus, cervix, brain, and skin. There were additional cases of tongue and oral squamous cell carcinoma as well as thyroid cancer and lymphoproliferative diseases that occurred after HSCT. Liver tumors, benign and malignant, were the second most frequent. Most of the patients with hepatomas and adenomas had received prior androgen therapy for aplastic anemia. Androgen administration has therefore been implicated in liver tumor pathogenesis. In descending order of frequency, cancers were also reported in the brain, kidney, breast, and adrenal gland.

Heterozygote Phenotype

Heterozygote carriers of FA gene mutations do not develop peripheral blood cytopenias or aplastic anemia, and cell lines from heterozygote carriers do not show excessive chromosome fragility in culture when exposed to DEB or MMC. The mean chromosomal breakage level of lymphocytes from FA carriers tested in cultures with a clastogenic agent may be higher than controls, but individual carrier testing may show overlap with normal values and severely limits its diagnostic utility. Literature from the early 1980s describes congenital anomalies of the hand and the genitourinary system in relatives of patients with FA, and parents of children with FA may have short stature. FA carriers may have increased levels of HbF, decreased natural killer (NK) cell counts, and diminished reactivity to mitogen stimulation. However, these results need to be reexamined using modern genetic diagnostic testing to exclude homozygosity and coexisting other conditions.

Monoallelic carriers for several FA genes are at increased risk of developing cancer, including for example, FANCD1, FANCS , FANCN, FANCJ, FANCM , FANCP , and FANCO . Female carriers of FANCD1/BRCA2 and FANCS/BRCA1 have an increased risk of breast cancer ranging from 40% at age 80 to a lifetime risk of about 80% and of ovarian cancer with a risk of up to 20% at age 70. Male carriers have a 7% risk of breast cancer and a 20% risk of prostate cancer before age 80. Other FA genes that are more prevalent in patients with breast cancer than in the general population, but to a much lower degree than FANCD1 and FANCS, are FANCN, FANCJ, FANCO, FANCP, FANCU, FANCD2.

Heterozygous mutations in FANCN are associated with pancreatic cancer, medulloblastoma, and high-grade glioma. Heterozygous mutations in FANCJ , FANCP , and FANCO are also associated with ovarian cancer.

Differential Diagnosis

About 30% of FA patients do not have physical anomalies, and such individuals may not be recognized until they present with aplastic anemia, MDS, AML, unilineage cytopenias, or macrocytic RBCs. Thus, FA should be part of the differential diagnosis in children and adults with unexplained cytopenias; characteristic birth defects; a diagnosis of aplastic anemia, MDS, or AML in patients mainly up to the age of 40 years, but sometimes also higher; unusual sensitivity to chemo- or radiotherapy; cancers typical of FA but at an atypical age such as cancer of the cervix when younger than 30 years or squamous cell carcinoma of the head and neck when younger than 50 years of age. Any of these should prompt consideration of FA as the underlying problem. All patients with idiopathic aplastic anemia who are younger than 40 years should have chromosomal fragility testing. However, if the test was not performed at diagnosis, patients with “idiopathic” aplastic anemia who fail to respond to immunosuppressive therapy with ATG and cyclosporine should be tested.

Although neutropenia is a consistent feature of SDS , anemia or thrombocytopenia (or both) is seen in more than 50% of the patients and can be confused with FA. Because growth failure is also a manifestation of SDS, differentiating between the two disorders can initially be difficult. The major difference between them is that SDS is a disorder of exocrine pancreatic dysfunction that may or may not produce gut malabsorption. This can be confirmed by fecal fat analysis, by showing reduced levels of serum trypsinogen, serum isoamylase, or fecal elastase, and by reduced levels of fat-soluble vitamins such as A, D, and E. Nowadays pancreatic stimulation studies using intravenous secretin or cholecystokinin and measuring enzyme secretion is rarely done. Ultrasonography, MRI of the pancreas may also demonstrate fatty changes within the pancreas. Other skeletal features found in some patients with SDS include short flared ribs, thoracic dystrophy at birth, delayed bone maturation, and metaphyseal dysostosis of the long bones. Chromosomal analyses do not show spontaneous breaks in SDS, and increased breakage after clastogenic stress testing using DEB or MMC does not occur in SDS. Genetic testing for mutations in one of the SDS genes can help distinguish SDS from FA.

Dyskeratosis congenita (DC) shares some features with FA, including development of pancytopenia, a predisposition to develop solid tumors and leukemia, and skin pigmentary changes. However, the pigmentation pattern is somewhat different in DC and manifests with a lacy reticulated pattern affecting the face, neck, chest, and arms, often with a telangiectatic component. At some point, usually in the first decade of life, DC patients also develop dystrophic nails of the hands and feet and, somewhat later, leukoplakia involving the oral mucosa, especially the tongue. Other findings seen only in DC and not in FA are teeth abnormalities with dental decay and early tooth loss, hair loss, and hyperhidrosis of the palms and soles. Chromosomal fragility with DEB testing is typically normal in DC patients, which contrasts sharply with FA patients. Molecular analysis of DC genes is positive in about three-quarters of the patients (see Table 30.1 ).

Congenital amegakaryocytic thrombocytopenia (CAMT) and TAR syndrome both present in the neonatal period with thrombocytopenia. Patients with CAMT develop impairment in other blood cell lineages soon after presentation. A neonatal hematologic presentation is atypical for FA; fewer than 5% of patients are diagnosed during the first year of life. Neither CAMT nor the various thrombocytopenia syndromes above show chromosome fragility, which separates them from FA. Genetic testing is available for many of these disorders (see Table 30.1 ). In the TAR syndrome, thumbs are always preserved and intact despite the absence of radii, but in FA, the thumbs are hypoplastic or absent when the radii are absent.

Seckel syndrome , or “bird-headed dwarfism” manifests with short stature; microcephaly; cognitive delay; sinopulmonary infections; and a predisposition to developing lymphomas, pancytopenia, and AML. Some patients may show increased chromosomal breakage in lymphocyte cultures with DEB or MMC and mimic FA. There are several genes that have been linked to Seckel syndrome: mutant ATR has been associated with Seckel Type 1, RBBP8 with Type 2, CENPJ with Type 4, CEP152 with Type 5, CEP63 with Type 6, and ATRIP with Type 8. Genotyping will distinguish FA from Seckel syndrome.

Nijmegen breakage syndrome (NBS) is an autosomal recessive disorder caused by mutations in the NBS1 gene and is characterized by stunted growth, microcephaly, a distinctive facies, café-au-lait spots, immunodeficiency, and a predisposition to develop lymphoid malignancies. Some patients resemble those with FA, have BM failure, and may show increased chromosome breakage in lymphocyte cultures with MMC. The genetic defect is a mutant NBS1 gene whose wild-type protein product is involved in DNA repair. Because NBS can mimic and be confused with FA, genotyping is essential and diagnostic.

Cells from patients with Bloom syndrome show abnormal spontaneous breakage, but unlike FA cells, the breakage does not increase in vitro in response to DEB. Ataxia telangiectasia is characterized by sister chromatid exchange without hypersensitivity to DEB or BM failure.

Natural History and Prognosis

The most serious early consequence in most FA patients is BM failure. The exceptions are patients with biallelic FANCD1 / BRCA2 mutations who have a cumulative probability of 97% of developing a malignancy by age 6 years, including AML, Wilms tumor, and medulloblastoma. By age of 40 years, patients with mutations in genes other than FANCD1 have a cumulative risk of marrow failure, AML, and solid tumors of about 60%, 10%, and 15%, respectively. The risk of marrow failure plateaus at 70% at about 60 years of age, but the risk of solid tumors continues to rise. Treatment for cancer imposes additional problems and probably increases the risk for additional cancers secondary to therapy. Thus, the major causes of death in FA patients are sepsis and bleeding from BM failure, complications of HSCT, and progressive cancer or consequences of its treatment.

Despite these serious issues, the prognosis for FA patients is improving. Based on a literature review of more than 2000 FA case reports, the median survival from 1927 to 1999 was 21 years. In contrast, the median survival age from 2000 to 2009 was 29 years and 39 years in 2018. More than 80% of patients reach age 18 years or more. Earlier diagnosis, especially of mild cases, diagnosis of FA in young adults with AML or a solid tumor, comprehensive clinical and laboratory surveillance programs, timely therapeutic interventions, and HSCT are attributed to the improved outlook.

Therapy

Because of their clinical and psychosocial complexity, patients with FA should be supervised by a hematologist at a tertiary care center using a comprehensive and multidisciplinary approach. On the initial visit, the practitioner should take a detailed personal and family history, a careful physical examination with emphasis on physical anomalies, complete blood counts and chemistries, a BM biopsy for cellularity and morphology, and an aspirate for additional morphology, cytogenetics, and an iron stain looking for ringed sideroblasts. Chromosome fragility testing with and without DEB and/or MMC on peripheral blood lymphocytes on patients should be arranged. If FA is confirmed by DEB or MMC testing, genetic diagnosis should be offered to the family. Subsequently, imaging studies should be requested to search for internal anomalies. Because of the carcinogenic risk, imaging using radiation should be limited to those that may change management and may provide information that cannot be obtained by other methods or in a timely fashion. When all the results from the workup have been compiled, a discussion should be held with the patient and guardians about the diagnosis, management options, and prognosis. A referral to a genetic counselor should ensue. Blood counts, hemoglobin variant analysis, and chromosome fragility testing of siblings should be offered. High-resolution human leukocyte antigen (HLA) typing of the patient and immediate family members is recommended shortly after the diagnosis is established to determine potential matched-related donors in case HSCT becomes necessary.

If the patient is stable, has only minimal to moderate hematologic changes, and does not have transfusion requirements, a period of observation is indicated. During this time, subspecialty consultations (e.g., with orthopedic surgeons, urologists, gynecologists, and otolaryngologists) can be arranged. Blood counts should be frequently monitored to determine their stability. In a stable patient with mild cytopenias, blood counts can be monitored every 3 months, and BM evaluation should be performed annually. Falling counts, a clonal BM cytogenetic abnormality, or prominent multilineage dysplasia require more frequent clinic visits and blood and BM sampling to monitor for progression to severe aplastic anemia, MDS, or AML. Spectral karyotyping (SKY), fluorescent in situ hybridization (FISH), and comparative genomic hybridization of BM cells can enhance the diagnostic capability ( Chapter 57 ). Analysis of mutations in specific cancer-related genes may be introduced into clinical practice in the future, but more research is necessary to determine their ability to predict progression to MDS/AML.

A surveillance program for solid cancers should be initiated at least annually. After the age of 10 years or one year after HSCT, the oral cavity should be examined every 6 months for signs of malignant changes because the risk in untransplanted FA patients is 700-fold that of the general population. Dentists, oral surgeons, or head and neck surgeons should be periodically recruited after the age of 10 years or after HSCT to screen for head and neck squamous cell carcinomas by rhinopharyngoscopy using a flexible endoscope. Beginning at age 13 years, all women with FA should be offered an annual gynecologic screening because the relative risk of vulvar squamous cell carcinoma is 4000-fold higher and cervical cancer is 200-fold higher than that of the general population. Human papilloma virus (HPV) DNA can be detected in 84% of FA squamous cell carcinoma specimens from various anatomic sites. Although the role of HPV in FA carcinogenesis is controversial, quadrivalent HPV vaccine is still recommended for boys and girls with FA at 9 years of age as a possible preventive approach.

Growth should be serially documented, and if growth velocity or stature falls below expectations, endocrine evaluation is needed to identify growth hormone deficiency. Diabetes mellitus occurs more commonly in FA, and random glucose levels should be evaluated annually or biannually. Based on the degree of hyperglycemia found on initial testing, fasting glucose levels and glucose tolerance tests should be performed. Screening for hypothyroidism should also be performed annually.

Genetic Counseling

Genetic counseling should be offered to all patients and families with FA. The discussion should include the implications of the identified (or not identified) underlying genetic lesion, important genotype phenotype correlations (e.g., high risk of early cancer in patients with FANCD1/BRAC2 mutations), mode of transmission, risk of having the disease in family members, risk of recurrence in future pregnancies, available diagnostic tests during pregnancy, and PGD/IVF and selection of HLA-matched embryos who do not have FA as potential donors for HSCT. Screening of all first-degree relatives should be offered according to the methods available: medical history, physical examination, complete blood counts, hemoglobin F level, chromosome fragility, and mutation analysis.

During pregnancy, gene testing and/or the abnormal chromosome breakage patterns characteristic of FA can be used to make a prenatal diagnosis of FA. Diagnostic testing can be performed on fetal amniotic fluid cells obtained at week 16 of gestation or on chorionic villus biopsy specimens at 9 to 12 weeks of gestation.

A manual for the management of FA patients is available by the Fanconi Anemia Research Fund (see www.fanconi.org for Fanconi Anemia Guidelines for Diagnosis and Management , 4th edition, 2014).

Future Directions

The premise for gene therapy in FA is based on the assumption that corrected hematopoietic cells would have a growth advantage. Strengthening this supposition are FA patients with hematopoietic somatic mosaicism who show spontaneous disappearance of cells with the FA phenotype. These mosaic patients may show spontaneous hematologic improvement, suggesting that hematopoiesis was derived from stem cells with a normal phenotype.

Despite encouraging preclinical studies since the early 2000s using retroviral vectors showing that wild-type FANCC and FANCA can be integrated into normal and FA CD34 ⁺ cells, the ensuing clinical trials in FANCC and FANCA patients using retroviral vectors were disappointing. A central problem was suboptimal wild-type gene integration into FA cells in culture. Because of the apoptotic phenotype and the sensitivity to oxidative stress, FA cells die rapidly in vitro before efficient gene transfer is accomplished. Changing the tissue culture conditions (e.g., usage of low oxygen condition) and introducing lentiviral vectors that can infect noncycling human cells were deemed the solutions. A few clinical trials for FA-A were opened in the late 2010s. Rio and colleagues reported promising results demonstrating successful engraftment of CD34 ⁺ hematopoietic stem and progenitor cells carrying lentivector-mediated corrected FANCA gene in FA patients without conditioning. The procedure was well tolerated, hematopoiesis from CD34 ⁺ cells was improved, but clinical efficiency could not be assessed due to the small number of patients. One caveat: a successful FA gene therapy protocol may correct BM failure and possibly the propensity for MDS and AML, but the predisposition for cancer in other tissues will continue unchecked.

Epidemiology

SDS has been reported among all ethnic groups. Older studies suggested a higher incidence in males. However, recent data suggest an equal distribution between genders as expected from a mainly autosomal recessive disorder. Based on data from the CIMFR, SDS is the third most common IBMFS with an incidence of 1 in 118,000 to 1 in 175,000 live births.

Genetic Basis

SDS is considered a ribosome-related disorder (ribosomopathy). The vast majority of patients with an SDS phenotype have mutations in genes involved in the last stages of pre-60S ribosome subunit maturation ( SBDS , DNAJC21 , EFL1 ). Some patients have a mutation in a gene related to the cotranslation protein-targeting pathway ( SRP54 ), which also cause severe congenital neutropenia phenotype.

The first identified SDS gene is the Shwachman-Boyden-Diamond-syndrome ( SBDS ) gene that is mutated in 85% to 90% of SDS patients. The described mutations in this gene include null and splicing alterations, but never biallelic null, suggesting that a complete absence of SBDS protein is lethal. The common SBDS mutations are composed of sequences that are homologous to its pseudogene, SBDSP , and are believed to result from genetic recombination events in which SBDSP1 acts as the donor. These recombinational events result in three common gene conversion mutations in exon 2 that account for 75% of SDS alleles: (1) a splice-site mutation, c.258+2 T>C, which may either cause premature truncation of the SBDS protein by frameshift (p.C84fs3) or use an alternative splice site; (2) a nonsense mutation, c.183_184TA>CT that introduces an in-frame stop codon (p.K62X); and (3) an extended conversion mutation, c.183_184TA>CT and c.258+2 T>C, that encompasses both mutations. In the Toronto database of 210 SDS families, 89% of unrelated SDS individuals carry a gene conversion mutation on one allele, and 60% carry conversion mutations on both alleles. The majority of patients are compound heterozygotes with respect to p.K62X and p.C84fsx3. Additional rare mutations in the SBDS gene have been identified in SDS patients. These include dozens of insertion, deletion, and missense mutations that have not arisen from gene conversion events. Most SBDS mutations alter the N-terminal domain of the protein and lead to markedly reduced protein levels.

DNAJC21 is the second gene identified to be associated with SDS. Biallelic mutation in DNAJC21 can be identified in about 10% of SDS patients. DNAJC21 -associated SDS is autosomal recessive. Both biallelic missense mutations and biallelic null mutations have been reported. Additional groups identified rarer SDS genetic groups with biallelic mutations in EFL1 and monoallelic mutations in SRP54. All reported patients with biallelic EFL1 mutations had at least one missense mutations.

Pathobiology

SDS genes encode highly conserved proteins that are ubiquitously expressed. SBDS protein has three main domains (N-terminal, middle, and C-terminal) with predicted protein-protein, protein-DNA, and protein-RNA binding motifs. SBDS protein is essential for life because patients with homozygous null mutations have not been reported, and small levels of residual protein can usually be detected in SDS patients. Furthermore, a complete loss of the protein in mice causes developmental arrest before embryonic day 6.5 and early lethality. In contrast to SBDS , biallelic null mutations in DNAJC21 have been identified in SDS patients, indicating the complete loss of this protein is still compatible with life.

SBDS, DNAJC21, and EFL1 play a direct role in ribosome biogenesis. SBDS binds EFL1 on the 60 S large ribosome subunit, which through GTP hydrolysis causes release of the dissociation factor eIF6 and binding of 60 S to a mRNA-loaded 40 S subunit. Patient-related mutations in SBDS and in EFL1 have been shown to disturb this function. DNAJC21 is required for the release of the dissociation factor PA2G4 from the pre-60S ribosome subunit, which is critical for 60 S subunit maturation and association with 40 S. This function of DNAJC21 homolog in yeast (Jjjj1) is also required for the release of eIF6 homolog (Tif6) by SBDS homolog (Sod1).

Yeast strains deleted in SBDS and DNAJC21 homologs accumulate their targets in the cytosol (eIF6 and Arx1, respectively), grow poorly at low temperatures, accumulate halfmer ribosomes and have reduced levels of mature ribosomes, all hallmarks of dysfunctional 60 S ribosomal subunit biogenesis. Loss of SBDS or DNAJC21 in human cells results in markedly reduced global translation.

As the first identified disease-related gene, SBDS functions have been studied more extensively than other SDS genes. It has been shown that SBDS is critical for several cellular pathways, including cell survival, telomere maintenance, mitotic spindle stabilization, chemotaxis, and marrow stromal function, besides ribosome biogenesis.

The SBDS protein can be detected in human cell nuclei and cytoplasm. It concentrates in the nucleolus during G1 and G2. SBDS interacts with multiple proteins with diverse molecular functions; many of them are involved in ribosome biogenesis, such as RPL4, and DNA metabolism, such as RPA70.

SBDS is critical for cell survival. When SBDS is lost in SDS BM cells or in SBDS -knockdown K562 and HeLa cells, the cells undergo accelerated apoptosis. The accelerated apoptosis in BM cells and SBDS -knockdown cells seems to be through the Fas pathway and not through the Bax/Bcl-2/Bcl-XL pathway. SBDS deficiency in primary SDS cells and in SBDS -knockdown cells results in abnormal accumulation of functional Fas (transcript 1) at the plasma membrane level.

Interestingly, knocking down SBDS in CD34 ⁺ cells and in cell lines increased the levels of ROS, and antioxidants reduced Fas-mediated cell death and improved hematopoiesis from primary SDS cells. This suggests that SBDS promotes balanced levels of ROS, thereby protecting hematopoietic cells from cell death.

Patients with SDS have a defect in leukocyte chemotaxis. Consistent with this observation, the SBDS homologue in amoeba was found to localize to the pseudopods during chemotaxis. These observations suggest that the SBDS protein deficiency in SDS causes a chemotaxis defect in patients.

SBDS has been shown to colocalize to the mitotic spindle and bind microtubules and stabilize them. Its deficiency results in centrosomal amplification and multipolar spindles.

Clinical Features

The many clinical manifestations that occur in varying combinations have been reported in several publications and are shown in Table 30.4 . Most patients present in infancy with evidence of growth failure, feeding difficulties, diarrhea, and infections. Steatorrhea and abdominal discomfort are frequent. Approximately 50% of patients exhibit a modest improvement in pancreatic function and do not require further pancreatic enzyme replacement therapy. Hepatomegaly is a common physical finding in young children but typically resolves with age and does not have clinical significance.

Patients with SDS are particularly susceptible to bacterial and fungal infections, including otitis media, bronchopneumonia, osteomyelitis, septicemia, and recurrent furuncles. Overwhelming sepsis is a well-recognized fatal complication of this disorder, particularly early in life.

Short stature is a common feature of the syndrome; the 50th percentile of SDS charts for height is positioned on the 3rd percentile of regular charts, both for boys and girls. When treated with pancreatic enzyme replacement, most patients show a normal growth velocity yet remain consistently below the third percentile for height and weight, indicating an intrinsic growth defect. Although metaphyseal dysplasia is a common radiologic abnormality (44% to 77% of patients), particularly in the femoral head and the proximal tibia, in most patients it fails to produce any symptoms. Occasional patients have clinical joint deformities, resulting in pain, functional impairment, or cosmetic problems, necessitating surgery. Some patients present at birth with respiratory distress caused by thoracic dystrophy. Others may have asymptomatic short and flared ribs.

The majority of the patients have deficits in cognitive abilities at varying levels of severity. These include delayed language development, low intellectual ability, impaired visual-motor integration, and failure to achieve higher order language functioning and problem solving. About one-fifth of the children have behavioral challenges such as attention deficit hyperactivity disorder, pervasive developmental disorder, or oppositional defiant disorder.

Some additional clinical features are seen very infrequently in SDS. Endocrine abnormalities include insulin-dependent diabetes, growth hormone deficiency, hypogonadotropic hypogonadism, hypothyroidism, and delayed puberty. Cardiomyopathies have been noted in some cases. Urinary tract anomalies, renal tubular acidosis, and cleft palate also occur. Retinopathy has been reported in SDS patients with DNAJC21 mutations.

It is important to note that SDS phenotype might be mild and a diagnosis may be made only as part of a comprehensive genetic screening of adults with MDS. Nevertheless, to see if a subset of previously undiagnosed SDS patients presented for the first time with AML, 48 BM samples at remission were studied for mutations in SBDS, but none were found.

Laboratory Findings

Peripheral Blood and Bone Marrow Findings

The spectrum of hematologic findings has been reported in several publications and is summarized in Table 30.4 . Neutropenia is present in almost all patients on at least one occasion. The neutropenia can be chronic or intermittent. Neutropenia has been identified in some SDS patients in the neonatal period during an episode of sepsis. Anemia is recorded in about half of the patients, but may fluctuate. RBC MCV and fetal hemoglobin are elevated in 60% and 75% of the patients, respectively, after the age of 1 year, which represents stress hematopoiesis. The combination of isolated neutropenia and high MCV or high HbF after the first year of life is seen in up to 28% of SDS patients and rarely in other IBMFSs. Reticulocyte responses are inappropriately low for the levels of hemoglobin in 75% of patients. Thrombocytopenia can be seen in about 40% of the patients and similar to anemia and neutropenia can also appear intermittently.

More than one lineage can be affected, and multi-lineage cytopenias have been observed in up to 65% of cases. Multi-lineage cytopenias can be due to profound severe aplastic anemia ( Fig. 30.2 ). However, BM biopsies and aspirates vary widely with respect to cellularity; varying degrees of BM hypoplasia and fat infiltration are the usual findings. BM with normal or even increased cellularity has also been observed, typically in young children. The severity of neutropenia does not always correlate with BM cellularity, nor is the severity of the pancreatic insufficiency concordant with the hematologic abnormalities.

SDS neutrophils may have defects in mobility, migration, and chemotaxis. There appears to be a diminished ability of SDS neutrophils to orient toward a gradient of N -formyl-methionyl-leucyl-phenylalanine. An unusual surface distribution of concanavalin A has also been reported that reflects a cytoskeletal defect in SDS neutrophils. Whatever the magnitude of the chemotaxis abnormality is in vitro in SDS, neutrophil recruitment into abscesses or empyemas ensues robustly in vivo.

Immune Dysfunction

Impaired immune function can be significant in SDS and contribute to recurrent infections even if adequate numbers of neutrophils are present. Patients have various B-cell abnormalities, including one or more of the following: low immunoglobulin G (IgG) or IgG subclasses, low percentage of circulating B lymphocytes, decreased in vitro B-cell proliferation, and lack of specific antibody production. Patients may also have T-cell abnormalities, including a low percentage of circulating T lymphocytes or subsets or NK cells, and decreased in vitro T-cell proliferation. Inverted CD4:CD8 ratios have also been described.

Exocrine Pancreatic Tests

The exocrine pancreatic pathology is caused by failure of pancreatic acinar development ( Fig. 30.3 ). Pathologic studies reveal normal ductular architecture but extensive fatty replacement of pancreatic acinar tissue, which can be visualized by ultrasonography, CT, or MRI.

During the first 3 years of life, serum trypsinogen is typically reduced and can be used for diagnostic purposes. After the age of 3 years serum trypsinogen levels in SDS patients may normalize, which reduces its diagnostic usefulness at this stage. Serum isoamylase levels are reduced in SDS patients of all ages. However, normal children younger than 3 years have low isoamylase levels, so its measurement is not diagnostically useful at this age. Fecal elastase is another pancreatic enzyme that is reduced in SDS. Approximately 50% of patients exhibit a modest improvement in enzyme secretion with advancing age and normal fat absorption when assessed by 72-hour fecal fat balance studies. These patients do not require further pancreatic enzyme replacement therapy.

Pancreatic function studies using intravenous secretin or cholecystokinin can confirm the presence of markedly impaired enzyme secretion averaging 10% to 14% of normal but with preserved ductal function. Because of its invasive nature, this test has been replaced by measuring the levels of pancreatic enzymes in the serum.

Imaging Studies

Radiographs of the bone are useful as a screening diagnostic test for SDS. Osteopenia is seen in most patients but rarely results in clinical osteoporosis. Metaphyseal dysplasia has been reported in about 50% of the patients, particularly of the femoral heads, knees, humeral heads, wrists, ankles, and vertebrae. Rib-cage abnormalities can be found in 30% to 50% of patients. These include a narrow rib cage, short ribs, flared anterior rib ends, and costochondral thickening. Digital abnormalities such as clinodactyly, syndactyly, and supernumerary thumbs have been reported but are rare. Spinal deformities, including kyphosis and scoliosis, have been reported.

Patients with SDS do not have macroscopic brain malformations by MRI testing. However, they may have a decreased global brain volume (both gray matter and white matter) and a smaller posterior fossa, cerebellar vermis, corpus callosum, brainstem, and occipitofrontal head circumference compared with control participants. These anomalies might be the basis for the neurocognitive and neurobehavioral difficulties.

The French registry found cardiac anomalies in 11% of SDS patients. These include dilated and non-dilated cardiomyopathy, and structural malformations such as atrial septal defect, ventricular septal defect, coarctation of the aorta, and tetralogy of Fallot. Circumferential strain as measured by echocardiography was found to be decreased by the USA SDS registry, suggesting systolic dysfunction. Endocardial fibrosis and reduced left ventricular strain were also reported.

Cancer Predisposition

SDS is characterized by a high propensity to develop MDS and leukemia, particularly AML. The published crude rate for MDS/AML in patients with SDS ranges from 8% to 33%. In data from the CIMFR and the French Severe Chronic Neutropenia Registry, the cumulative risk of MDS/AML by the age of 18 and 20 years was 20% and 19%, respectively. The risk of leukemia in the French registry was 36% by the age of 30 years.

There is an increased frequency of BM clonal cytogenetic abnormalities as the sole evidence for a clonal disease in an otherwise hypocellular BM without excess blast counts or prominent multilineage cellular dysplasia. The incidence is roughly estimated to be 7% to 41% based on pooled published data; however, not all clones progress and some clones may not be detected on subsequent testing. Isochromosome 7q [i(7q)], an extremely uncommon finding rarely described in primary MDS or AML, was seen in up to 44% of SDS patients. This high occurrence suggests that it is a fairly specific marker for SDS and might be related to the mutant gene on 7q(11). Other chromosome 7 abnormalities are seen in 33% of SDS patients and include monosomy 7, i(7q) combined with monosomy 7 and deletions or translocations involving part of 7q. The prognostic significance of the cytogenetic changes requires prospective monitoring for clarification. Of the patients with i(7q), progression to advanced MDS with excess blasts or to AML has rarely been reported, but development of severe cytopenias and additional clones was described in three of four patients after long-term follow-up in the Canadian registry. Among a group of six patients with i(7q) from several hospitals in the United Kingdom, none progressed to advanced MDS/AML. In contrast, approximately 40% of patients with the other chromosomal 7 abnormalities progressed to either advanced MDS or to AML. Similarly, only rare cases with SDS patients and del(20q) evolve into advanced MDS/AML.

The pathophysiologic link between SBDS mutations and propensity to MDS and AML is unknown. Patients with SDS cells develop more frequent mutations than healthy subjects, possibly because of genomic instability due to mitotic spindle dysregulation or telomere shortening. It is also possible that impaired ribosome biogenesis and accelerated apoptosis cause a growth disadvantage for SDS BM cells, allowing for a growth advantage and expansion of malignant clones. Although molecular and cellular parameters do not distinguish SDS patients with transformation from SDS patients without transformation, it is remarkable that SDS BM demonstrates many features characteristic of MDS. These include impaired BM stromal support of normal hematopoiesis, increased BM cell apoptosis mediated by the Fas pathway, telomere shortening of leukocytes, increased BM neovascularization, high frequency of clonal cytogenetic abnormalities, and abnormal leukemia-related gene expression in BM progenitor cells, e.g. overexpression of the oncogenes TAL1 and LARG .

Similar to clonal evolution in FA, deep immunotyping analysis of early progenitors in SDS BM samples demonstrated relative selective survival of cells with GMP markers. Whole-exome and targeted sequencing of GMP-like cells in leukemia-free patients revealed a higher mutational load than in healthy controls and molecular changes that are characteristic of AML including increased G>A/C>T variants, decreased A>G/T>C variants, increased trinucleotide mutations at Xp(C>T)pT (X indicates any nucleotide), and decreased mutation rates at Xp(C>T)pG sites compared with other Xp(C>T)pX sites. Serial analysis of GMPs from an SDS patient who progressed to leukemia revealed a gradual increase in the mutational burden, enrichment of G>A/C>T signature, and emergence of new clones. Importantly, the molecular signature of marrow cells from a SDS patient with leukemia was similar to that of SDS patients without transformation. The predicted founding clones in SDS-derived AML harbored mutations in several genes, including TP53. The results suggest that GMP-like cells might represent a cellular reservoir for clonal evolution.

Monoallelic deletion of 20q10-11 and point mutations in EIF6 are commonly seen in SDS, and alleviate the SDS ribosome joining defect and translation by inactivating eIF6. In contrast, somatic monoallelic TP53 mutations, which are also frequently seen in SDS, decrease checkpoint activation and may progress to leukemia after developing a second mutation on the other allele, as demonstrated in one SDS patient and AML.

The vast majority of published cases of SDS-associated MDS/AML developed without previous G-CSF therapy. None of the six patients with SDS-associated MDS/AML from our institution were treated with G-CSF before transformation. However, it is still unclear whether G-CSF increases the risk of developing leukemia or promotes the expansion of existing malignant clones. Because G-CSF might increase neutrophil counts and prevent infections in SDS, a fraction of the reported patients with SDS-associated MDS/AML had been previously treated with G-CSF.

SDS patients with MDS/AML have common SBDS mutations, and no specific mutations have been shown to be associated with a higher risk of MDS/AML.

Several cases of solid tumors have been described in SDS. These include two cases of pancreatic ductal adenocarcinoma, one brain frontal lobe B-cell lymphoma, one dermatofibrosarcoma protuberans, and one breast cancer. However, more data are needed to determine whether the risk of solid tumors is higher than in the general population.

Genotype Phenotype Correlation

Despite the very similar function of SBDS and DNAJC21 in ribosome biogenesis, some important phenotypic differences between these two SDS genetic groups have been observed. (1) Many SBDS -associated SDS have neutropenia without prominent anemia and thrombocytopenia. By contrast, hematopoietic failure in DNAJC21 -associated SDS is often more global and affects more lineages and can be more severe. (2) Retinal disease is associated with DNAJC21 mutations, but not with SBDS mutations. ELF1 mutations are associated with multi-lineage cytopenia, pancreatic insufficiency, and skeletal dysplasia. Patients with a SRP45 mutation have a variable phenotype ranging from severe congenital neutropenia resembling ELANE-related disease to classical SDS.

Differential Diagnosis

The introduction of single gene analysis and later of comprehensive genomic testing has improved the ability to diagnose the disorder and particularly has helped identify cases with an atypical presentation. The diagnostic criteria have been summarized in an international consensus document and include having at least two of the following: (1) chronic BM failure, (2) exocrine pancreatic insufficiency, (3) positive genetic testing results or a first degree-relative with SDS. Several syndromes with overlapping features have to be excluded.

The syndrome of refractory sideroblastic anemia with vacuolization of BM precursors, or Pearson syndrome , is clinically similar to SDS but characterized by very different BM morphology. Severe anemia requiring transfusions rather than neutropenia is often present at birth and by 1 year of age in all cases. In contrast to SDS, the major BM morphologic findings are ringed sideroblasts with decreased erythroblasts and prominent vacuolation of erythroid and myeloid precursors. The disorder shares clinical similarities with SDS because of exocrine pancreatic dysfunction. Malabsorption and severe failure to thrive occur in approximately half of cases within the first 12 months of life. Qualitative pancreatic function tests show depressed acinar function and reduced fluid and electrolyte secretion. Approximately 50% of reported patients die early in life from sepsis, acidosis, and liver failure; the others appear to improve spontaneously with reduced transfusion requirements. At autopsy, the pancreas shows acinar cell atrophy and fibrosis, but fatty infiltration as seen in SDS is not a prominent feature. Patients may need pancreatic enzyme replacement. These patients have a diagnostic deletion of mitochondrial deoxyribonucleic acid (mtDNA). mtDNA encodes enzymes in the mitochondrial respiratory chain that are relevant to oxidative phosphorylation, including the reduced form of nicotinamide adenine dinucleotide dehydrogenase (NADH), cytochrome oxidase, adenosine triphosphatase (ATPase), mitochondrial transfer ribonucleic acids (tRNAs), and mitochondrial ribosomal RNAs. The degree of heteroplasmy affects the disease expression.

SDS shares some manifestations with FA such as BM dysfunction and growth failure, but patients with SDS can usually be distinguished because of malabsorption syndrome, fatty changes within the pancreatic body that can be visualized by imaging, and characteristic skeletal abnormalities not seen in patients with FA. In difficult cases with incomplete disease expression, the distinction relies on normal clastogenic stress-induced chromosome fragility testing and genetic testing of SDS and FA genes.

Atypical SDS cases with only little evidence of pancreatic changes can be difficult to distinguish from early-onset dyskeratosis congenita with no mucocutaneous manifestations. Telomere length testing showing very short telomeres (below 1% for age) is far more commonly seen in DC; nevertheless, when the diagnosis is unclear, genetic testing can help define the diagnosis.

Prognosis

Because of the carried clinical phenotype in SDS, the number of undiagnosed patients with mild or asymptomatic disease is unknown. Hence, the overall prognosis may be better than previously thought.

From a literature review, the projected median survival of SDS patients was calculated as 35 years. During infancy, morbidity and mortality are mostly related to infections, thoracic dystrophy, and malabsorption. Later in life, the major problems are hematologic or complications related to their treatment. Cytopenias tend to fluctuate in severity but do not fully resolve spontaneously. The most common cause of death in late childhood or adulthood is related to MDS/AML.

The impact of different SDS genetic groups on long-term outcome may vary, but needs to be further studied.

Therapy

Patient management is ideally shared by a multidisciplinary team consisting of a hematologist and a gastroenterologist as core members and other subspecialists such as an orthopedic surgeon, a dentist, and a psychologist as required. All patients with DNAJC21 mutations should also have an eye examination by an ophthalmologist. The malabsorption component of SDS responds to treatment with oral pancreatic enzyme replacement with meals and snacks using guidelines similar to those for cystic fibrosis. Supplemental fat-soluble vitamins are also usually required. When monitored over time, approximately 50% of patients convert from pancreatic insufficiency to sufficiency because of spontaneous improvement in pancreatic enzyme secretion. This improvement is particularly evident after 4 years of age.

A long-term plan should be initiated for early detection of severe cytopenias that require corrective action or malignant myeloid transformation. There are currently no data about the cost effectiveness of a specific leukemia surveillance program in SDS. However, in a consensus document on SDS management it is generally recommended that surveillance should include periodic blood counts with differentials and blood smears every 3 to 4 months, a clinical evaluation by a hematologist every 6 months, and BM testing every 1 to 3 years. The latter includes aspirates for smears and cytogenetics analyses. Concomitant BM biopsies are recommended when the patient’s clinical status changes. Recently it has been shown that an SDS BM testing may reveal evidence of MDS before changes are seen in the peripheral blood counts.

Future Directions

Several other clinical and basic research questions in SDS must be addressed. First, the various biochemical functions of SDS genes require further studies. How the protein products maintain normal hematopoiesis and protect from apoptosis as well as cancer is unclear. The natural history and risk factors for the development of complications need to be clarified. There is also a need to understand the mechanism for the heightened sensitivity of SDS patients to chemotherapy and irradiation and to develop effective and less toxic regimens for HSCT. Determining risk factors and molecular events during malignant myeloid transformation might prompt strategies for prevention and screening for complications. Further research is needed to develop new drugs that improve growth potential of HSCs and relieve the severity of the cytopenias. Clinical trials to explore efficacy and safety of agents that target pathways that were shown to be abnormal in SDS (e.g., translation, oxidative stress, and TGF-beta signaling) may help identify such drugs. Recently Ataluren, which promotes mRNA reading by the ribosome through premature stop codons, has shown to increase SBDS protein levels, improve CFU-GEMM and CFU-GM colony formation, and neutrophil maturation from SDS BM mononuclear cells as well as neutrophil survival. Ataluren has been approved for the treatment of patients with Duchenne muscular dystrophy and nonsense mutations, and might also be effective in SDS which will require evaluation in a clinical trial.

Background

DC is an inherited multisystem disorder of the mucocutaneous and hematopoietic systems in association with a wide variety of other somatic abnormalities. Originally, it was considered a dermatologic disease, and was termed Zinsser-Cole-Engman syndrome after the first investigators who described it. The term “dyskeratosis congenita with pigmentation” was first proposed by Cole and colleagues, Rauschkolb, and Toomey in 1930. The traditional diagnostic ectodermal triad consists of reticulate skin pigmentation of the upper body, mucosal leukoplakia, and nail dystrophy. The skin and nail findings usually become apparent during the first 10 years of life, but the oral leukoplakia is commonly observed later. These manifestations tend to progress as patients get older. Age of presentation varies according to the genetic group.

Hematologic manifestations were reported in the 1950s, and in the 1960s/1970s BMF was recognized to be a major component of the syndrome and responsible for substantial morbidity and mortality. Indeed, the full diagnostic dermatologic triad is present in only about 46% of the patients, but BM failure of varying severity is reported in up to 90% of cases. With the recent advances in understanding the molecular basis of the disease, patients with hematologic abnormalities but without dermatologic findings have been identified that dramatically changed the historical definition of the disease. DC patients also have a predisposition to develop solid tumors and MDS/AML.

Epidemiology

The estimated incidence of DC in childhood is about 4 cases per million per year. In older literature, most DC patients were reported as males. However, with better understanding and broadening of the genetic and clinical spectrum of the disease and with more autosomal cases being identified, the proportion of males is much lower.

Genetics

DC is a telomere-related disorder (telomeropathy), and all the identified DC genes play a role in telomere homeostasis (see Table 30.1 ). DC genes encode components of the telomerase complex ( TERT , DKC1 , TERC , NOP10 , and NHP2 ), T-loop assembly protein ( RTEL1 ), telomere capping ( CTC1 ), the telomere shelterin complex ( TINF2 ), and the telomerase trafficking protein ( TCAB1 ); all of which are critical for telomere maintenance.

The X-linked recessive disease is a common form of DC. It was originally estimated to comprise as many as 75% of DC cases, but with the identification of more DC genes and more patients with autosomal dominant inheritance, the true incidence is approximately 30%. The X-linked disease is caused by mutations in DKC1 on chromosome Xq28. The encoded protein, dyskerin, is part of a complex (with TERC ) that stimulates the telomerase (TERT). It is also involved in rRNA biogenesis.

The autosomal recessive forms of DC are caused by biallelic mutations in RTEL1 , PARN , NOP10 , NHP2, TCAB1, CTC1, ACD ( TPP1 ), or TERT . Interestingly, several DC gene ( RTEL1 , PARN or TERT, ACD ( TPP1 )) mutations can be associated with a severe disease (Hoyeraal-Hreidarsson syndrome) when both alleles are mutated, or a milder disease when one allele is affected.

Several genes are mutated in families with autosomal dominant inheritance. TINF2 is probably the most commonly mutated gene in this group and accounts for approximately 11% to 25% of the DC families. The vast majority of TINF2 mutations are de novo, probably due to the severity of disease. Heterozygous mutations in TERT, TERC, PARN, ACD (TPP1), or RTEL1 also result in autosomal dominant disease. In autosomal dominant DC where family members from several generations are affected, younger generations tend to have longer telomeres and more severe disease than their ancestors.

DC is a complex disorder and several clinicogenetic sub-types have been described. Classical DC that includes BMF and typical extra hematological manifestations can be caused by mutations in any of the DC genes ( Table 30.3 ). DC with either hemizygous DKC1 mutations or a heterozygous TINF2 mutation or biallelic mutations in TERT , PARN, RTEL1, WRAP53 (TCAB1), or ACD (TPP1) can result in a severe form of DC called Hoyeraal-Hreidarsson syndrome . It is characterized by hematologic and dermatologic manifestations of DC in addition to cerebellar hypoplasia. Immune deficiency is common when this syndrome is caused by DKC1 mutations. Revesz syndrome is a combination of classical manifestations of DC and exudative retinopathy. It is caused by mutations in TINF2 and is an autosomal dominant form of the disease. Some cases of moderate/severe aplastic anemia without physical anomalies are caused by mutations in TINF2, TERT , and TERC . In contrast to cases with biallelic TERT mutations, heterozygosity for mutations in TERT is associated with a milder phenotype, late presentation, severe aplastic anemia without physical malformations, isolated pulmonary fibrosis, isolated hepatic fibrosis, or a combination of these clinical sequelae. Heterozygosity for mutations in TERC is associated with a milder phenotype, late presentation and severe aplastic anemia, or MDS without physical malformations . Coats plus syndrome is caused by mutations in the CTC1 gene. It is characterized by retinal telangiectasia and exudates, intracranial calcification, leukodystrophy, brain cysts, osteopenia, gastrointestinal bleeding, and portal hypertension caused by the development of vascular ectasias in the stomach, small intestine, and liver. Some patients with this disease have the additional manifestations of DC, which include sparse and gray hair, dystrophic nails, and anemia. Telomeres are short.

It is noteworthy that mutations in certain telomere-related genes can cause partial DC phenotype without BMF; for example, monoallelic mutations in NAF1 or ZCCHC8 causing pulmonary fibrosis and biallelic mutations in STN1 causing Coats plus.

Pathobiology

The hallmark of DC is short telomeres. Most children have severely short telomeres (less than the first percentile for age). DC genes play a role in telomere length maintenance in several ways and can be divided into those who belong to telomerase complex ( TERT, TERC) , other telomere replication-related factors ( RTEL1 , CTC1 , RPA1 ), telomerase trafficking ( WRAP53 /TCAB1), TERC-associated factors that stimulate telomerase activity ( DKC1, NOP10, NHP2 ), TERC-maturation factors ( PARN ), shelterin complex ( TINF2, POT1, TPP1 )

Telomerase complex ( TERT, TERC) TERT encodes for the enzyme component of telomerase. Telomerase is a ribonucleoprotein polymerase that maintains telomere ends by synthesis and addition of G-rich telomere repeat TTAGGG at the 3′-hydroxy DNA terminus of telomeres using the TERC RNA as a template. TERC encodes for the RNA component of telomerase and has a 3′ H/ACA small nucleolar RNA -like domain. The H/ACA box at the 3′-end of TERC is essential for its stability and assembly into the telomerase pre-ribonucleoprotein complex.

Other telomere replication-related factors ( RTEL1 , CTC1 , RPA1 ). RTEL1 is a helicase that resolves DNA secondary structures. It is recruited to telomers by TRF2 and unwinds the telomeric t-loop, thereby allowing telomere replication. CTC1 is a member of the CTC1-STN1-TEN1 (CST) trimeric complex that helps synthesize telomeric C-strand by binding to the telomeric 3′ overhang and stimulating DNA polymerase a-primase. RPA1 binds to telomeric 3′ overhangs during late S-phase to unfold G-quadruplexes and facilitate telomerase activity. A gain of function mutation was identified in a patient with classic DC.

WRAP53 (TCAB1) is involved in telomerase trafficking to telomeres. TCAB1 interacts with the CAB box on TERC, promoting localized telomerase in Cajal bodies and trafficking it to telomeres. Mutations in the TCAB1 gene ( WRAP53 ) impair this trafficking activity and lead to misdirection of telomerase RNA to nucleoli, thereby preventing elongation of telomeres by telomerase.

Several DC genes encode TERC-associated factors that stimulate telomerase activity: DKC1, NOP10, NHP2. GAR1, TEP1, and NAF1 are also associated with TERC, but have not been shown to be mutated in DC as of December 2020. Dyskerin (encoded by DKC1 ) binds H/ACA classes of RNA. It binds to the 3′ H/ACA small nucleolar RNA-like domain of the TERC . This stimulates telomerase to synthesize telomeric repeats during DNA replication. Dyskerin is also involved in maturation of nascent rRNA. It binds to small nucleolar RNA through the 3′ H/ACA domain and catalyzes the isomerization of uridine to pseudouridine through its pseudouridine synthase homology domain. This might be the mechanism for impaired translation from internal ribosome entry sites seen in mice and human DC cells.

In the telomerase complex, the H/ACA domain of nascent human TERC forms a pre-ribonucleoprotein with dyskerin, NOP10, NHP2, and NAF1. Initially, the core trimer dyskerin-NOP10-NHP2 forms to enable incorporation of NAF1, and efficient reverse transcription of telomere repeats. NOP10 and NHP2 also play an essential role in the assembly and activity of the H/ACA class of small nucleolar ribonucleoproteins that catalyze the isomerization of uridine to pseudouridine in rRNAs.

Poly(A)-specific ribonuclease (PARN) is a TERC-maturation factor. It is one of the major mammalian deadenylases that trims single-stranded poly(A) tails of mRNAs, but also of oligoadenylated tails of H/ACA box snoRNAs and microRNAs. As such it trims excessive adenylate residues that are added to TERC by PARD5. In patient cells oligoadenylated TERC forms are more abundant in PARN-deficient patient cells than those of controls, although its total levels were slightly reduced.

The shelterin complex is composed of TINF2, POT1 , and TPP1 , which are mutated in DC, as well as TRF1, TRF2 , and RAP1 that as of December 2020 have not been shown to harbor mutations that cause DC. TINF2 protein binds to and protects telomeres by allowing cells to distinguish between telomeres and regions of DNA damage. In the complex, TINF2 binds to TRF1, TRF2, and TPP1.

In several acquired and IBMFSs besides DC, telomeres are shorter than in age-matched control participants. However, because the telomere maintenance pathway is profoundly impaired in DC, the telomeres in this disease are much shorter than in other disorders. Most children with DC have telomeres that are very short (lower than the first percentile of the normal range). Average telomere length in adults with DC is shorter than that of healthy subjects, but it is commonly at the low range of normal for age. Shortening of telomeres results in cellular senescence, apoptosis (“cellular crisis”), or chromosome instability. However, some cells may survive the crisis by harboring compensatory genetic mutations that confer proliferative advantage and neoplastic potential.

DC is a chromosome “instability” disorder of a different type than FA. Results of clastogenic stress studies of DC cells are typically normal. There is no significant difference in chromosomal breakage between patient and normal lymphocytes with or without exposure to bleomycin, DEB, MMC, or γ-radiation. This contrasts sharply with FA cells and distinguishes one disorder from the other. However, metaphases of cultured patient peripheral blood cells, BM cells, and fibroblasts show numerous spontaneous unbalanced chromosome rearrangements such as dicentrics, tricentrics, and translocations. These are probably caused by short telomeres.

Most studies of the pathogenesis of the aplastic anemia in DC have shown a marked reduction or absence of CFU-GEMM, BFU-E, CFU-E, and CFU-GM. Long-term DC BM cultures have shown that hematopoiesis is severely defective in all patients with a low frequency of colony-forming cells. The function of DC BM stromal cells is normal in their ability to support growth of hematopoietic progenitors from normal BM, but generation of progenitors from DC BM cells seeded over normal stroma is reduced, suggesting that the defect in DC is of stem cell origin. Telomerase is activated in HSCs and might be necessary for HSC self-renewal capacity and prevention of senescence. The BM failure in this disorder may be a result of a progressive attrition and depletion of HSCs. This is supported by studies showing reduced numbers of CD34 ⁺ /CD38 ⁻ cells in patient BMs. Alternatively, the BM dysfunction may represent a failure of replication, maturation, or both.

iPSCs from DC patients have been shown to have defects in telomere elongation during programming in a manner that is concordant with the mutated gene in the patients. In iPSCs from patients with heterozygous mutations in TERT , telomerase activity is directly affected. iPSCs from a patient with TCAB1 mutations featured mislocalization of telomerase from Cajal bodies to nucleoli. iPSCs from patients with mutant DKC1 display reduced telomerase activity because of impaired telomerase assembly. Extended culture of DKC1 -mutant iPSCs leads to progressive telomere shortening and eventual loss of self-renewal. In contrast, another group studied telomerase reactivation and TERC regulation during reprogramming and showed that reprogramming restores telomere elongation in DC cells despite genetic lesions affecting telomerase. This group showed that TERC upregulation is a feature of the pluripotent state and that several telomerase components are targeted by pluripotency-associated transcription factors.