Trisomy
Maternal
Paternal
References
MI
MII
MI or MII
Mitotic
Totala
MI
MII
MI or MII
Mitotic
Totala
2
4
1
6
1
13
5
5
[30]
7
2
3
1
6
12
2
2
13
4
17
21
1
1
1
3
[24]
34
33
3
3
5
1
3
[33]
14
3
4
2
9
2
2
[24]
15
21
3
3
27
5
5
[16] (UPD study)
10
10
2
2
[18] (UPD study)
17
2
10
29
4
1
5
16
56
6
62
0
[25]
18
11
17
56
1
6
[28]
17
5
[21]
16
≥35
3
61
2
2
[26]
10
17
1
28
2
1
3
[31]
21
9
1
22
3
[17]
91
6
[18]
128
38
188
2
7
9
[19]
7
15
8
36
[23] (paternal study only)
174
58
79
311
9
15
8
32
[24]
62
81
10
13
[27]
67
22
97
4
4
10
[29]
22
20
1
15
37
1
1
The association between autosomal aneuploidy and maternal age has long been recognized. In 1933, Penrose demonstrated that maternal age was the key factor for the birth of Down syndrome children [34]. Why aneuploidy is maternal age dependent, and what constitutes the mechanism and etiology of chromosomal nondisjunction have been topics of much research, as summarized later in this chapter.
Nondisjunction can occur during either meiosis I (MI) or meiosis II (MII). In MI, homologous chromosomes pair and form bivalents (see Chap. 2). Malsegregation of homologous chromosomes can occur in one of two ways. The first involves nondisjunction of the bivalent chromosomes with both homologs going to the same pole (Fig. 8.1d,e). The second type of error involves premature separation of the sister chromatids of one homolog of a chromosome pair. Subsequent improper distribution of one of the separated chromatids results in its segregation with the other homolog of the chromosome pair [35] (Fig. 8.2d,e). In MII, sister chromatids separate. Malsegregation occurs when both chromatids go to the same pole (Fig. 8.3g,h). It has been shown that error involving premature sister chromatid separation, especially in the smaller chromosomes, is in fact a more common mechanism leading to aneuploidy than malsegregation of the whole chromosome [6, 36, 37].
Fig. 8.1
Schematic representation of meiosis I nondisjunction. (a) Prophase I. (b) Metaphase I. (c) Anaphase I. (d) Telophase I, with both homologs of one chromosome pair segregating together. (e) Products of meiosis I. (f) Metaphase II. (g) Anaphase II. (h) Meiotic products—two gametes lack one chromosome and two gametes contain two copies of one chromosome
Fig. 8.2
Schematic representation of meiosis I error resulting from premature sister chromatid separation. (a) Prophase I. (b) Metaphase I. (c) Anaphase I, with premature separation of centromere of one chromosome. (d) Telophase I, with one prematurely separated chromatid segregating with its homologous chromosome. (e) Products of meiosis I. (f) Metaphase II. (g) Anaphase II. (h) Meiotic products—two gametes with a normal chromosome complement, one gamete lacking one chromosome, and one gamete containing two copies of one chromosome
Fig. 8.3
Schematic representation of meiosis II nondisjunction. (a) Prophase I. (b) Metaphase I. (c) Anaphase I. (d) Telophase I. (e) Products of meiosis I. (f) Metaphase II. (g) Anaphase II, with both sister chromatids segregating together. (h) Meiotic products—two gametes with a normal chromosome complement, one gamete lacking one chromosome, and one gamete containing two copies of one chromosome
Earlier cytogenetic studies of oöcytes, performed mostly on unfertilized or uncleaved specimens obtained from in vitro fertilization programs, have provided conflicting results regarding whether the frequency of gamete aneuploidy actually increases with maternal age [38–41]. A FISH study of human oöcytes using corresponding polar bodies as internal controls demonstrated that nondisjunction of bivalent chromosomes (MI error) does in fact increase with maternal age, and a study using multiplex FISH on fresh, noninseminated oöcytes also indicated an increase in premature separation of the sister chromatids in MI with increasing maternal age [42, 43]. More recent data based on studies of large numbers of oöcytes further provided evidence for a direct correlation between advanced maternal age and increased aneuploidy frequency [6, 44].
Different mechanisms have been proposed to account for the observation of the correlation between maternal age and aneuploidy frequency. One example is the “production line” hypothesis [45, 46]. This hypothesis proposes that oöcytes mature in adult life in the same order as the corresponding oögonia entered meiosis in fetal life. Oögonia that enter meiosis later in development may be more defective in the formation of chiasmata and thus more likely to undergo nondisjunction. One direct cytological support for this hypothesis was provided by a study that examined the frequency of unpaired homologs in MI pachytene and diplotene in oöcytes obtained from abortuses at 13–24 weeks and 32–41 weeks of gestation [47]. Of the six chromosomes studied (chromosomes X, 7, 13, 16, 18, and 21), the rate of pairing failures in early specimens (0–1.2%) was significantly lower than that in later specimens (1.3–5.5%). No corroborating data are available. It remains an interesting question whether the oöcytes first committed to meiosis in fetal life are the first to ovulate in adult life. Another example is the “limited oöcyte pool” model [48]. At the antral stage of each menstrual cycle, multiple follicles at various stages of development are present. When stimulated with high levels of plasma follicle stimulating hormone (FSH), only one follicle, presumably the one at the most optimal stage, will complete MI and eventually achieve ovulation. The number of follicles in the antral stage decreases with increasing maternal age. When the number of these follicles is low, it is more likely that an oöcyte that is not at the optimal stage will be selected for ovulation. If such “less optimal” oöcytes are more likely to undergo MI nondisjunction, then the ovulated oöcytes of older women will have higher rates of aneuploidy. More recent data, however, does not appear to support this hypothesis [49, 50].
One probable factor that predisposes gametes to nondisjunction is aberrant recombination [51] (see Chap. 2). Data on recombination patterns are available for trisomies 15, 16, 18, and 21. Studies of chromosome 15 nondisjunction in uniparental disomy (see Chap. 20) revealed that there was a mild reduction in recombination in association with maternal nondisjunction, with an excess of cases that have zero or one crossover and a deficiency of cases that have multiple crossovers [20]. In contrast, in a study of trisomy 18, approximately one-third [5/16] of maternal MI nondisjunctions were associated with a complete absence of recombination, whereas the remaining MI and all MII nondisjunctions appeared to have normal rates of recombination [26]. Studies of trisomy 16 and trisomy 21 reported similar findings between the two. In trisomy 16, it was shown that recombination was reduced, but not absent, and that distribution of recombination was altered, with rare crossovers in the proximal regions of the chromosome [25]. A recent study performed on oöcytes from young egg donors after hormone-induced superovulation demonstrated that 2.5% of chromosome 16 bivalents had no crossovers and a high percentage (19.8%) had only a single recombination [37]. In trisomy 21, there was an overall reduction in recombination with an increase in both zero and one crossover in maternal MI nondisjunction [52]. Lamb et al. showed that in maternal MI nondisjunction for chromosome 21, the average number of recombination events was decreased, with approximately 35–45% of cases having no crossovers [53]. When at least one crossover was present, it occurred largely at distal 21q. This study, together with one on trisomy 16, suggests that, at least for trisomies 16 and 21, distal chiasmata are less efficient in preventing nondisjunction in MI [25]. In contrast, in maternal MII nondisjunction, the number of recombination events appeared to be increased, especially in proximal 21q. These proximal recombinations may cause an “entanglement” effect. Entanglement of the two homologs can cause the bivalent to move to the same pole at MI, and then at MII the two homologs finally separate, resulting in two disomic gametes each having two chromatids with identical centromeres. Alternatively, the entanglement may disrupt sister chromatid cohesion resulting in premature separation of the sister chromatids at MI. If the two separated sister chromatids travel to the same pole at MI and again at MII, an apparent MII nondisjunction would be observed. Thus, these data suggest that all nondisjunction events may be initiated during MI. The observation that for chromosome 21, MI error is associated with distal recombination while MII error is associated with proximal recombination has been independently confirmed recently in a study of a population in India [54]. Lamb et al. showed that the alteration in recombination pattern was not maternal age dependent. They proposed a “two hit” model and hypothesized that certain recombination configurations are less likely to be processed properly in older women [53, 55]. This could result from, for example, an age-dependent loss of spindle forming ability, thus explaining their observation for trisomy 21 that although an altered recombination pattern is not maternal age dependent, meiotic disturbance is age dependent [56]. The same argument was used by Hassold et al. to explain their findings with trisomy 16 [51].
It has been proposed that the cellular mechanism assuring correct segregation of chromosomes into daughter cells is provided by a four-protein complex (SMC1, SMC3, SCC1/RAD21, SCC3/SA/STAG) that together form a ring-like structure known as the cohesin complex [57]. The cohesin complex acts as “chromosome glue” and thus mediates cohesion of the two sister chromatids during cell division. Additional proteins are needed for the establishment and maintenance of cohesion. Loss of cohesion of both arms, telomere, and centromere during the metaphase/anaphase transition is also tightly controlled by various proteins including a specific protease separin/separase.
It has also been suggested, at least for chromosome 21, that chiasmata and recombination in oöcyte are as efficient as in spermatocyte [58]. The less effective meiotic checkpoint mechanism in oöcytes allowing aneuploid oöcytes to progress through meiosis appeared to be the basis for the observation that the majority of trisomy 21 conceptions are of maternal origin. On the other hand, evidence against a defective spindle assembly checkpoint being the cause of aneuploidy associated with advanced maternal age has also been reported [59].
The possibility of the presence of a genetic predisposition to nondisjunction has also been proposed. One study involving consanguineous families in Kuwait showed that the relative risk for the occurrence of Down syndrome was approximately four times greater for closely related parents (first cousins, first cousins once removed, second cousins) than for unrelated parents [60]. As consanguinity is usually perpetuated in certain families or sections of the population, these results were taken as evidence for the existence of an autosomal recessive gene that facilitates meiotic nondisjunction in homozygous parents. Thus, in a subgroup of trisomy 21 patients, nondisjunction may be genetically determined. In a study of trisomy recurrence based on North American data, a significantly increased risk for recurrent of a different trisomy was observed [61]. This supports the hypothesis of possible genetic predisposition to nondisjunction.
While maternal age and altered recombination remain the only well-established risk factors for nondisjunction, our understanding of the underlying mechanism of this observation is still not complete. It is possible that more than one mechanism, including possibly environmental and hormonal factors, contributes to the observed maternal age effect [62].
Nondisjunction occurring at mitosis, on the other hand, will result in mosaicism, usually with both normal and abnormal cell lines.
Discussion of autosomal aneuploidies in this chapter will be limited largely to those observed in liveborns only.
Autosomal Trisomies
Trisomy 21
Incidence
Trisomy 21 [47,XX or XY,+21] (Fig. 8.4) was the first chromosome abnormality described in man by Lejeune et al. in 1969 [63]. The phenotype was delineated by John Langdon Down (1828–1896) in 1866 and is referred to today as Down syndrome [64]. It is the most common single known cause of mental retardation. The frequency in the general population is approximately 1 in 700. Down syndrome is more frequent in males, with a male-to-female ratio of 1.2:1. A recent study using multicolor FISH showed that among sperm disomic for chromosome 21, significantly more were Y-bearing than X-bearing [65]. This finding was consistent with earlier reports showing an excess of males among trisomy 21 conceptuses that resulted from paternal meiotic errors [23]. This preferential segregation of the extra chromosome 21 with the Y chromosome contributes to a small extent to the observed sex ratio in trisomy 21 patients. Other mechanisms, such as in utero selection against female trisomy 21 fetuses, must also exist.
Fig. 8.4
Trisomy 21 Down syndrome male karyogram (47,XY,+21)
Trisomy 21 accounts for approximately 95% of all cases of Down syndrome. Mosaicism and Robertsonian translocations (see Chap. 9) comprise the remaining 5%. As described previously, the incidence of trisomy 21 in newborns is closely associated with maternal age (Table 8.2).
Table 8.2
Maternal age-specific risks for trisomy 21 at birth
Maternal age (years) | Incidence at birth (1 in) | Maternal age (years) | Incidence at birth (1 in) |
|---|---|---|---|
15 | 1,580 | 33 | 570 |
16 | 1,570 | 34 | 470 |
17 | 1,565 | 35 | 385 |
18 | 1,555 | 36 | 307 |
19 | 1,540 | 37 | 242 |
20 | 1,530 | 38 | 189 |
21 | 1,510 | 39 | 146 |
22 | 1,480 | 40 | 112 |
23 | 1,450 | 41 | 85 |
24 | 1,400 | 42 | 65 |
25 | 1,350 | 43 | 49 |
26 | 1,290 | 44 | 37 |
27 | 1,210 | 45 | 28 |
28 | 1,120 | 46 | 21 |
29 | 1,020 | 47 | 15 |
30 | 910 | 48 | 11 |
31 | 800 | 49 | 8 |
32 | 680 | 50 | 6 |
Phenotype
The clinical phenotype of Down syndrome has been well described [68, 69]. Briefly, there is a characteristic craniofacial appearance with upward-slanting palpebral fissures, epicanthal folds, flat nasal bridge, small mouth, thick lips, protruding tongue, flat occiput, and small and overfolded ears. Hands and feet are small and may demonstrate clinodactyly, hypoplasia of the midphalanx of the fifth finger, single palmar crease (Fig. 8.5), and a wide space between the first and second toes. Hypotonia and small stature are common, and mental retardation is almost invariable. Cardiac anomalies are present in 40–50% of patients, most commonly endocardial cushion defects, ventricular septal defects (VSD), patent ductus arteriosus (PDA), and atrial septal defects (ASD). Other observed major malformations include duodenal atresia, annular pancreas, megacolon, cataracts, and choanal atresia. In addition, a 10- to 20-fold increase in the risk for leukemia, most commonly acute megakaryoblastic leukemia, has been observed in Down syndrome patients of all ages, with a bimodal age of onset in the newborn period and again at 3–6 years [70]. Moreover, a transient abnormal myelopoiesis (TAM) with clinical and morphologic findings indistinguishable from acute myeloid leukemia occurs in approximately 10% of Down syndrome newborns [71]. Spontaneous remission within the first 3 months of life is observed in most patients; occasional relapse and life-threatening disease have also been noted. Of interest is the observation of the presence of a trisomy 21 clone in association with TAM in 15 phenotypically normal children, at least 4 of whom were determined to be constitutional mosaics for Down syndrome [72].
Fig. 8.5
The hand of a Down syndrome child showing small hand, clinodactyly, only one crease in the fifth finger, and single palmar crease
Overall, the clinical phenotype is typically milder in mosaic Down syndrome patients, but there is no clear correlation between the percentage of trisomy 21 cells and the severity of clinical presentation. This can be as severe in mosaic patients as in nonmosaic trisomy 21 individuals.
Delineation of the regions of chromosome 21 responsible for the Down syndrome phenotype has been attempted using molecular methods to study patients with partial trisomy 21 who present clinically with various features of the syndrome [73–80]. Studies by Korenberg et al. in a panel of cell lines derived from 16 partial trisomy 21 individuals suggest that, instead of a single critical region, many chromosome 21 regions are responsible for various Down syndrome features [79]. The study was expanded to include a total of 30 subjects carrying rare segmental trisomies of various regions of chromosome 21. By using current genomic technologies including high-density isothermal oligonucleotide DNA tiling arrays, a high-resolution genetic map of Down syndrome phenotype was constructed corresponding to discrete regions of 1.8–16.3 Mb likely to be involved in the development of eight Down syndrome phenotypes: acute megakaryocytic leukemia, transient myeloproliferative disorder, Hirschsprung disease, duodenal stenosis, imperforate anus, severe mental retardation, Down syndrome-Alzheimer disease, and Down syndrome-specific congenital heart disease [81]. The map also provided evidence against both the existence of a single Down syndrome consensus region and the previous supposition that a synergistic role of DSCR1, DYRK1A, and/or APP was sufficient for many of the Down syndrome phenotypes.
The additional copy of chromosome 21 is proposed to result in the increased expression of many of the genes encoded by this chromosome. The knowledge of which of the genes, when present in three copies, leads to each of the different Down syndrome-associated phenotype, together with research using Down syndrome mouse models, may provide insight into possible pharmacological approach to improving some of the symptoms [82].
Recurrence
Various estimates of the recurrent risk for trisomy 21 have been reported. The overall empirical recurrence risk is about 1% in women under 30 years of age and includes trisomies other than 21. For women over 30, the recurrence risk may not be significantly different from the age-specific risk [83]. A more recent study reported 5,960 women with a previous trisomy 13, 18, or 21 pregnancy; 75 of the 3,713 subsequent pregnancies were trisomic [84]. The relative risk of a subsequent trisomy 21 compared to the expected number of trisomies based on maternal age-related risk alone was 2.2. The risk of a different trisomy subsequent to trisomy 21 might also be increased (relative risk 1.4). The increase in risk was greater for women under age 35 at the first trisomic pregnancy. A similar increase in the rate of trisomy pregnancy following an initial trisomy pregnancy was reported in a study of trisomy recurrence based on North American data [61].
One study of 13 families with two trisomy 21 children showed that three had a parent who was mosaic for trisomy 21 (by cytogenetic studies), and two had a parent who was potentially mosaic (by DNA polymorphism analysis) [85]. In a family with three trisomy 21 children, Harris et al. reported that the mother was mosaic for trisomy 21 in lymphocytes and skin fibroblasts [86]. In another single case report involving a family with four trisomy 21 children, the mother was found to have a trisomy 21 cell line in an ovarian biopsy specimen [87]. In a study compiling data from 80 families with either maternal (61 families) or paternal (19 families) gonadal mosaicism for trisomy 21, a total of 142 Down syndrome offspring were reported [88]. Among these offspring, mosaicism was observed in 12 families and the proportion of mosaics among affected female offspring (14%) was significantly higher compared to that among affected male offspring (0%). Based on these observations, it was proposed that female-specific trisomy rescue might be a mechanism of formation of both gonadal mosaicism and somatic mosaicism. Gonadal mosaicism in one parent is an important cause of recurrent trisomy 21 and should be looked for in families with more than one affected child.
The recurrence risk for mosaic trisomy 21 that results from mitotic nondisjunction should, in general, not be increased. However, several studies investigating the mechanism and origin of mosaic trisomy 21 have shown that in a relatively high proportion of cases (probably over 50%), the mosaicism results from the loss of one chromosome 21 during an early mitotic division in a zygote with trisomy 21 [89, 90]. In such cases, the recurrence risk for nondisjunction will be the same as for nonmosaic trisomy 21.
Trisomy 18
Incidence
Trisomy 18 [47,XX or XY,+18] was first described by Edwards et al. in 1960 [91]. The incidence is 1 in 6,000–8,000 births. It is more frequent in females, with a male-to-female ratio of 1:3–4. The risk for trisomy 18 also increases with maternal age.
Phenotype
The most common features of trisomy 18 include mental and growth deficiencies, neonatal hypotonicity followed by hypertonicity, craniofacial dysmorphism (prominent occiput, narrow bifrontal diameter, short palpebral fissures, small mouth, narrow palate, low-set malformed ears, micrognathia) (Fig. 8.6), clenched hands with a tendency for the second finger to overlap the third and the fifth finger to overlap the fourth, short dorsiflexed hallux, hypoplastic nails, rocker bottom feet, short sternum, hernias, single umbilical artery, small pelvis, cryptorchidism, hirsutism, and cardiac anomalies (mainly ventricular septal defect [VSD], atrial septal defect [ASD], and patent ductus arteriosus [PDA]). Studies show that median survival averages approximately 5 days, with 1-week survival at 35–45%; one later study indicated a median survival of 14.5 days [92–96]. Fewer than 10% of patients survive beyond the first year of life. A few patients over 10 years of age, all females with one exception, have been described; however, the presence of a normal cell line in these patients was not always searched for [97–99].
Fig. 8.6
Profile of a trisomy 18 child showing prominent occiput, low-set malformed ear, and micrognathia
Mosaic trisomy 18 patients have, in general, milder phenotypes. At least six mosaic trisomy 18 patients, again all females, with normal intelligence and long-term survival have been reported [100–105].
Two molecular studies, performed on a total of 10 patients with partial trisomy 18, suggest that the region proximal to band 18q12 does not contribute to the syndrome, while two critical regions, one proximal (18q12.1→q21.2) and one distal (18q22.3→qter), may work in cooperation to produce the typical trisomy 18 phenotype [106, 107]. In addition, severe mental retardation in these patients may be associated with trisomy of the region 18q12.3→q21.1.
Recurrence
Single case reports of trisomy 18 in sibs (e.g., [105]), and of trisomy 18 and a different trisomy in sibs or in prior or subsequent abortuses (e.g., [108–110]) are recorded. In the same studies referenced for trisomy 21, an increased risk of trisomy 18 subsequent to a previous pregnancy with trisomy 18 was observed [61, 84]. The relative risk was 1.7–3.8. Again, the increase in risk was greater for women under age 35 at the first trisomic pregnancy. Given the low baseline age-related risk, the absolute risk of recurrence is nonetheless quite low.
Trisomy 13
Incidence
Trisomy 13 [47,XX or XY,+13] was first described by Patau et al. in 1960 [111]. The incidence is estimated to be 1 in 12,000 births. It is seen slightly more in females than in males. Again, the risk for trisomy 13 increases with maternal age.
Phenotype
The most prominent features of trisomy 13 include the holoprosencephaly spectrum (Fig. 8.7), scalp defects, microcephaly with sloping forehead, large fontanels, capillary hemangioma (usually on the forehead), microphthalmia, cleft lip, cleft palate, abnormal helices, flexion of the fingers, polydactyly, hernias, single umbilical artery, cryptorchidism, bicornuate uterus, cardiac abnormalities in 80% of patients (mostly VSD, PDA, and ASD), polycystic kidneys, increased polymorphonuclear projections of neutrophils, and persistence of fetal hemoglobin. Prognosis is extremely poor, with a median survival of 2.5–7 days and a 6-month survival of 5% [94, 96]. Severe mental deficiencies, failure to thrive and seizures are seen in those who survive. Mosaic trisomy 13 patients are, again, in general less severely affected; however, the degree is very variable and can be as severe as in nonmosaic trisomy 13 individuals.
Fig. 8.7
Trisomy 13 stillborn with midline cleft lip and holoprosencephaly
Development of a karyotype-phenotype correlation by studying partial trisomies for different segments of chromosome 13 has also been attempted [112, 113]. These studies were based on cytogenetic methods and suggested that the proximal segment (13pter→q14) contributes little to the trisomy 13 phenotype, while the distal segment (all or part of 13q14→qter) is responsible for the complete trisomy 13 features. A prenatally diagnosed pure partial trisomy 13 involving 13q14→qter with breakpoints delineated by array comparative genomic hybridization (aCGH; see Chap. 18) analysis was reported recently. The fetus had agenesis of the corpus callosum and the diaphragm, severe pulmonary hypoplasia, and generalized hydrops [114].
Recurrence
Trisomy 8
Trisomy 8 [47,XX or XY,+8] was first reported by Grouchy et al. in 1971 [115]. It is rare, with an unknown incidence. More than 100 cases have been reported in the literature, most of them mosaics [47,+8/46] [116–121]. The male-to-female ratio is 2–3:1.
Growth and the degree of mental deficiency are variable. Mild to severe retardation is seen, while a proportion of patients have normal IQs. Craniofacial dysmorphism (Fig. 8.8) includes prominent forehead, deep-set eyes, strabismus, broad nasal bridge, upturned nares, long upper lip, thick and everted lower lip, high arched or cleft palate, micrognathia and large dysplastic ears with prominent antihelices. Skeletal abnormalities include a long, thin trunk, hemivertebrae, spina bifida, kyphoscoliosis, hip dysplasia, multiple joint contractures, camptodactyly, dysplastic nails, and absent or dysplastic patella. The presence of deep palmar and plantar furrows is characteristic. Renal and ureteral anomalies and congenital heart defects are common. A case with extremely elevated maternal serum alpha-fetoprotein noted prenatally without open defect was recorded [121]. A few cases of hematological malignancy have been reported in mosaic trisomy 8 patients [122, 123]. This is of particular interest because trisomy 8 is a frequently acquired cytogenetic abnormality in myeloid neoplasms (see Chap. 15). When studied, the abnormal cells in these patients appeared to have developed from the trisomic cell population. The significance of this is not clear, but the possibility remains that constitutional trisomy 8 may predispose individuals to myeloid neoplasia.
Fig. 8.8
An infant with mosaic trisomy 8. Note prominent forehead, strabismus, broad nasal bridge, upturned nares, long upper lip, and everted lower lip
There is no direct correlation between the proportion of the trisomy 8 cells and the severity of the phenotype. The percentage of trisomic cells is usually greater in skin fibroblasts than in blood lymphocytes. In addition, the proportion in lymphocytes usually decreases with time.
The risk for recurrence is not known.
Trisomy 9
The first cases of trisomy 9 in either nonmosaic [47,XX or XY,+9] or mosaic [47,+9/46] form were reported in 1973 [124, 125]. More than 40 cases of liveborns or term stillborns with trisomy 9 have been reported. Most were mosaics [126–130]. The male-to-female ratio is close to 1:1.
Clinical features include craniofacial anomalies (high narrow forehead, short upward-slanting palpebral fissures, deep-set eyes, microphthalmia, low-set malformed auricles, bulbous nose, prominent upper lip, micrognathia), skeletal malformations (abnormal position/function of various joints, bone dysplasia, narrow chest, 13 ribs), overlapping fingers, hypoplastic external genitalia, and cryptorchidism. Cardiac anomalies are seen in more than 60% of cases, most frequently VSD. Renal malformations are present in 40% of patients. A case of mosaic trisomy 9 with holoprosencephaly and another case with XX sex reversal were reported [131, 132]. The majority of patients die in the early postnatal period. With rare exceptions, all survivors have severe mental deficiency. Mosaic patients tend to survive longer, but the proportion of trisomy 9 cells does not predict the severity of the condition or the length of survival. It is possible that a normal cell line could be present in some tissues in apparently nonmosaic patients.
The mean maternal age of women bearing trisomy 9 offspring was reported to be significantly increased over that of the general population [127]. This suggests that the occurrence of trisomy 9 may also be associated with advanced maternal age. The risk for recurrence is not known.
Trisomy 16
Trisomy 16 is the most frequently observed autosomal aneuploidy in spontaneous abortuses (see Chap. 13). Full trisomy 16 is almost always lethal during early embryonic or fetal development, although a single case of a stillborn at 35 weeks gestation has been recorded [133].
Mosaic trisomy 16 fetuses, however, may occasionally survive to term. More than ten such cases have been reported [134–141]. Intrauterine growth restriction is nearly invariable. An elevated maternal serum hCG or alpha-fetoprotein level during pregnancy was noted in more than 50% of cases. Congenital cardiac defects (mainly VSD or ASD) were present in 60% of patients. Other clinical findings included postnatal growth retardation, mild developmental/speech delay, craniofacial asymmetry, ptosis, flat broad nasal bridge, low-set dysplastic ears, hypoplastic nipples, umbilical hernia, deep sacral dimple, scoliosis, nail hypoplasia, and single transverse palmar crease. One patient had normal growth and development at 11 months of age [141]. Approximately 50% of the patients died within the first year of life. Long-term follow-up is not available; however, survival to more than 5 years has been observed (Hajianpour and Wang, personal observation).
The risk for recurrence is probably negligible.
Trisomy 20
Although mosaic trisomy 20 is one of the most frequent autosomal aneuploidies detected prenatally, its occurrence in liveborns is very rare [142]. The majority of prenatally diagnosed cases are not cytogenetically confirmed in postnatal life. It appears that in conceptuses capable of surviving to the second trimester, trisomy 20 cells are largely confined to extraembryonic tissues. Liveborns with documented mosaic trisomy 20 have been reported and most were phenotypically normal at birth [143–150]. In cases with long-term follow-up, hypopigmentation, mild psychomotor delay, and facial dysmorphism have been observed in some cases. The possibility of a more consistent phenotype associated with mosaic trisomy 20 has been recently suggested, including spinal abnormalities (spinal stenosis, vertebral fusion, kyphosis), hypotonia, lifelong constipation, sloped shoulders, and significant learning disabilities [150]. No case of liveborn nonmosaic trisomy 20 has been recorded.
Phenotypic abnormalities in abortuses with cytogenetically confirmed mosaic trisomy 20 include microcephaly, facial dysmorphism, cardiac defects, and urinary tract anomalies (megapelvis, kinky ureters, double fused kidney) [151].
Trisomy 20 cells have been found in various fetal tissues including kidney, lung, esophagus, small bowel, rectum, thigh, rib, fascia, and skin [142, 151, 152]. Postnatally, they have been detected in cultured foreskin fibroblasts and urine sediments [143–148]. The detection of trisomy 20 cells in newborn cord blood has been reported in one case, but subsequent study of peripheral blood at 4 months of age produced only cytogenetically normal cells [145]. There are no other reports of trisomy 20 cells in postnatal blood cultures.
The risk for recurrence is probably negligible.
Trisomy 22
Trisomy 22 was first reported in 1971 [153]. Since then, more than 20 liveborns have been reported in the literature [154–161]. Although most cases were apparently nonmosaic full trisomies, the presence of an undetected, normal cell line confined to certain tissues cannot be excluded, as pointed out by Robinson and Kalousek [162].
The most consistent phenotypic abnormalities include intrauterine growth restriction, low-set ears (frequently associated with microtia of varying degrees plus tags/pits), and midfacial hypoplasia. Other frequently seen abnormalities are microcephaly, hypertelorism with epicanthal folds, cleft palate, micrognathia, webbed neck, hypoplastic nails, anal atresia/stenosis, and hypoplastic genitalia. Cardiac defects, complex in some cases, are seen in 80% of patients. Renal hypoplasia/dysplasia is also common. Skin hypopigmentation (hypomelanosis of Ito) is usually present in mosaic cases. Intestinal malrotation and Hirschsprung disease were recently reported in a prenatally diagnosed mosaic trisomy 22 infant with normal development [160]. A 4-year-old girl with confirmed trisomy 22 mosaicism in skin had normal cognitive, behavioral, and physical development [161]. Prolonged survival to over 20 years has been observed in mosaic patients.
Most nonmosaic patients die in the first months of life. The longest survival reported is 3 years [163]. That patient had severe growth and developmental delay and died a few days before his third birthday.
Trisomy 22 cells can be detected in both blood lymphocytes and skin fibroblasts. The risk for recurrence is unknown and probably negligible.
Other Rare Autosomal Trisomies
As noted in the introduction, mosaic or nonmosaic autosomal trisomies for chromosomes other than 1 and 11 have been reported in liveborns. Trisomies are detected much more frequently in spontaneous abortuses or in prenatal diagnostic specimens, following which elective terminations are often performed. Thus, the occurrence of such trisomies in liveborns is extremely rare and only isolated case reports are available. The risks for recurrence for these rare trisomies are probably negligible. The following discussion will include cytogenetically confirmed postnatal cases only.
At least two cases of liveborn mosaic trisomy 2 have been reported [164, 165]. In one case, the mosaicism was detected in amniocytes and confirmed postnatally in liver biopsy fibroblasts (4 of 100 cells) but not in blood, skin fibroblasts, or ascites fluid cells. At 16 months of age, the child had hypotonia, microcephaly, and growth and developmental delay. In the second case, mental retardation, multiple congenital anomalies, and dysmorphic findings similar to Pallister-Killian syndrome were observed. Another case of possible mosaic trisomy 2, detected at amniocentesis and observed in a single cell of a foreskin fibroblast culture following the birth of a dysmorphic child, was reported in an abstract [166].
Three cases of mosaic trisomy 3 have been reported; one of these, a severely mentally retarded woman, was alive at age 32 [14, 167, 168]. Clinical features in the three cases vary, except all had prominent forehead, ear, and eye anomalies.
At least two cases of postnatally confirmed mosaic trisomy 4 have been reported [169, 170]. In both cases, the trisomic cells were detected in prenatal amniocytes and confirmed postnatally in skin fibroblasts, but not in blood lymphocytes. One of the cases also had low-level mosaicism for trisomy 6 with clinical features of prenatal growth restriction, right facial hypoplasia, dysplastic and posteriorly rotated right ear, high vaulted palate, retrognathia, aplasia of the right thumb, hypoplasia of the fingernails, deep sacral dimple, and patchy skin hypopigmentation of the right leg [170]. Long-term follow-up was available on the other case [171]. The patient had right hand and ear anomalies. At age 14, she had delayed puberty with no menarche, asymmetrical breast development, and low normal intelligence.
One case of postnatally confirmed mosaic trisomy 5 has been reported [172]. The trisomic cells were detected in prenatal amniocytes and confirmed postnatally in skin fibroblasts, but not in blood lymphocytes. The patient had multiple dysmorphic features and congenital anomalies, including eventration of the diaphragm and ventricular septal defect.
At least two cases of mosaic trisomy 6 have recently been reported. The first patient was born at 25 weeks of gestation. Clinical features included heart defects (ASD and peripheral pulmonary stenosis), large ears, cleft right hand, cutaneous syndactyly, overlapping toes of irregular shape and length, and epidermal nevi. Growth was considerably delayed, but development was relatively normal at age 2¾. Trisomy 6 cells were detected in skin fibroblasts but not in blood [173]. Mosaic trisomy 6 was prenatally diagnosed in fetal urine in the second case. The infant was born at term with normal growth parameter, heart defect, and malformations of hands and feet [174].
At least seven cases of cytogenetically documented mosaic trisomy 7 in skin fibroblasts have been recorded [175–178]. All patients were phenotypically abnormal. Common features included growth and developmental delay, skin pigmentary dysplasia with hypo- and hyperpigmentation, facial or body asymmetry, and facial dysmorphism. One mentally retarded male was 18 years old at time of report. A few cases of liveborn mosaic trisomy 10 have been reported [179–181]. One patient was mosaic for trisomy 10 and monosomy X in skin fibroblasts, whereas only monosomy X cells were present in blood. This infant died at 7 weeks of age from heart failure. Another patient was mosaic for trisomy 10 and had maternal uniparental disomy for chromosome 10 in the diploid cell line [181]. The common clinical phenotype included growth failure, craniofacial dysmorphism (prominent forehead, hypertelorism, upslanted palpebral fissures, blepharophimosis, dysplastic large ears, retrognathia), long slender trunk, deep palmar and plantar fissures, cardiac defects, and short survival.
At least seven cases of cytogenetically confirmed trisomy 12 have been reported in liveborns; all were mosaics [110–113, 115–187]. The earliest reported case was that of an infertile man. A more recent case was a girl with pituitary malformation associated with growth retardation responding to growth hormone therapy. The patient also had a polycystic right ovary. Phenotypic presentation was variable among patients and included facial dysmorphism, scoliosis, ASD, PDA, dysplastic pulmonary and tricuspid valves, short stature, and mental retardation. Trisomy 12 cells have been found in lymphocytes, skin fibroblasts, urine sediments, and internal organs including liver, spleen, adrenal, ovary, and thymus.
More than 20 cases of mosaic trisomy 14 have been reported in liveborns [188–190]. The most consistent phenotypic abnormalities were growth and mental retardation, broad nose, low-set dysplastic ears, micrognathia, congenital heart defects, and micropenis/cryptorchidism in males. One prenatally diagnosed patient had alobar holoprosencephaly and died at 36 days of age [189]. Survival varied from days to more than 29 years. Trisomy 14 cells were detected in both lymphocytes and fibroblasts, with a generally higher percentage in lymphocytes. There was no clear correlation between the proportion of trisomic cells and the severity of the phenotype. In patients with body asymmetry, trisomic cells were usually limited to the atrophic side.
At least ten cases of liveborn trisomy 15 have been recorded, two of them were reportedly nonmosaics [191–197]. In some cases, the trisomy 15 cell line was present only in skin fibroblasts and not in peripheral blood lymphocytes. The concurrent finding of maternal uniparental disomy 15 (see Chap. 20) in the normal cell line was reported in two of the cases [192, 194]. These cases appeared to have the most severe phenotype. Phenotypic abnormalities include hypotonia, various craniofacial dysmorphisms, minor skeletal anomalies, congenital heart defects, and short survival. One patient with longer survival had short stature, mild mental retardation, hemihypotrophy, atrial septal defect, bilateral branchial cleft fistulas, and abnormal skin pigmentation [195].
At least four cases of confirmed mosaic trisomy 17 have been reported [198–200]. The trisomic cells were not seen in lymphocytes but were found in high percentage in skin fibroblasts. One patient, age 8½ years at the time of reporting, had mental and growth retardation, microcephaly, minor dysmorphism, seizures, hearing loss, attention deficit hyperactivity disorder, and autistic behavior. Peripheral motor and sensory neuropathy, hypoplastic cerebellar vermis, zonular cataract, and body asymmetry have also been reported.
Autosomal Monosomies
As noted in the introduction, autosomal monosomies are extremely rare in either liveborns or abortuses, reflecting the severity of the genetic imbalance resulting from the loss of an entire chromosome. The only monosomies that have been reported are monosomy 21 and mosaic monosomy 22.
Monosomy 21
Mosaic monosomy 21 was reported in four liveborns in the early literature [203–206]. The most prominent features included intrauterine growth restriction, postnatal growth and mental retardation, hypertonia, facial dysmorphism with downward slanting palpebral fissures, large low-set ears, and micrognathia. A more recent report described pathological findings of an electively terminated 20-week female fetus after mosaic monosomy 21 was diagnosed by repeated amniocenteses [207]. The facial abnormalities previously described were present in this abortus. In addition, a complex cardiac malformation, malrotation of the bowel, uterus didelphys, small dysplastic ovaries, and focal cystic dysplasia of the lung were noted.
Approximately ten cases of apparently nonmosaic monosomy 21 have been reported in liveborns [208–211]. Some of these cases have subsequently been shown to represent partial monosomy 21 resulting from an undetected subtle translocation [212–214] with part of chromosome 21 material attached to a derivative chromosome, explaining the observation that mosaic monosomy 21 is less commonly observed than apparently nonmosaic monosomy 21 and indicating that complete monosomy 21 is almost always incompatible with life. The phenotypic features were similar to those observed in the mosaics and included intrauterine growth restriction, postnatal growth and mental deficiencies, microcephaly, hypertelorism with downward slanting palpebral fissures, large low-set ears, prominent nose, cleft lip/palate, micrognathia, cardiac anomalies, and abnormal muscle tone. Most patients died before 2 years of age. A case of full nonmosaic monosomy 21 confirmed by fluorescence in situ hybridization analysis was reported in a liveborn who died shortly after birth [211]. The phenotype of this infant included severe intrauterine growth restriction, microcephaly, semilobar holoprosencephaly, hypotonia, bilateral microphthalmia, facial dysmorphism, agenesis of the external auditory meatus, redundant skin in the neck, narrow chest, cryptorchydism, hypospadias, micropenis, camptodactyly, congenital heart disease, and agenesis of the right kidney.
Monosomy 22
At least four cases of mosaic monosomy 22 in liveborns have been reported [215–218]. All four were male. One was a 34-week premature infant with gastroschisis who died from intracranial hemorrhage shortly after birth. No dysmorphic features were noted, and autopsy was not performed [217]. Two patients had growth and developmental deficiencies, microcephaly, and mild facial dysmorphism. The fourth patient was a 30-week premature infant with facial features of DiGeorge syndrome, hypertonicity, limited extension of major joints, and flexion contractures of all fingers.
Polyploidy
Polyploidies are numerical chromosome abnormalities with changes in the number of complete sets of chromosomes. They are usually incompatible with fetal survival and are extremely rare in liveborns.
Triploidy
The chromosome number in triploidy is 3n = 69 (Fig. 8.9). It is estimated to occur in approximately 1% of all human conceptions and is found in 17–18% of all chromosomally abnormal abortuses [219, 220]. Only very rarely do triploid conceptuses survive to term. Two distinct phenotypes have been recognized [221]. One type presents as a relatively well-grown fetus with or without microcephaly, and an abnormally large and cystic placenta usually classified as a partial hydatidiform mole. The parental origin of the extra haploid set of chromosomes in such cases is determined to be paternal (diandry) by analysis of cytogenetic heteromorphisms or DNA polymorphisms [221, 222, 180]. Diandry results from the fertilization of a normal ovum with either two sperm (dispermy) or a sperm that has a diploid chromosome complement resulting from a failure of meiotic division. The other type is characterized by severe intrauterine growth restriction with relative macrocephaly and a small and noncystic placenta. The extra haploid set of chromosomes in such cases is maternal (digyny) [221–224]. Digyny can result from a failure of the first maternal meiotic division, generating a diploid egg, or from retention of the second polar body. While the occurrence of triploidy does not appear to be associated with maternal age, digyny may play a major role in the generation of triploidy in the advanced maternal age group [220]. Early cytogenetic studies indicated that the majority of triploid conceptuses were diandric partial moles [222, 225]. Later studies based on DNA polymorphisms have suggested that a maternal contribution to triploidy may occur more frequently than was previously realized [223, 226]. Yet in a more recent study of 87 informative cases of triploid spontaneous abortuses at 5–18 weeks of gestation, Zaragoza et al. showed that approximately two-thirds are androgenetic in origin and that many, but not all, androgenetic triploids developed a partial molar phenotype [227]. The sex chromosome complement in triploidy is either XXX or XXY, with XYY occurring only rarely. For example, the reported numbers of XXX:XXY:XYY cases in two studies performed on spontaneous abortuses were 82:92:2 and 26:36:1, and in one study performed on amniotic fluid cells this ratio was 6:8:0 [3, 177, 228]. It has been suggested that 69,XYY triploid conceptuses are incompatible with significant embryonic development [3].
Fig. 8.9
Karyogram of a triploid fetus (69,XXX)
The observation that the phenotype of triploidy depends on the parental origin of the extra set of chromosomes is an example of genomic imprinting, or the differential expression of paternally and maternally derived genetic material [229, 230]. It correlates well with observations obtained from mouse embryo studies using nuclear transplantation techniques, which demonstrated that maternal and paternal genomes function differently and are both required for normal development [231–233]. See Chap. 20.
More than 50 cases of apparently nonmosaic triploidy, either 69,XXX or XXY, have been reported in liveborns. Most patients died shortly after birth. Eight patients with survival longer than 2 months have been reported, with the longest being 10½ months [234, 235]. The origin of the extra set of chromosomes was determined by cytogenetic polymorphisms or human leukocyte antigen (HLA) to be maternal in three cases and paternal in one case [236]. One study based on DNA polymorphism in an infant who survived for 46 days indicated a maternal meiosis II failure as the origin of the triploid [236]. These findings suggest that in general digynic triploids survive longer than diandric triploids. The most frequent phenotypic abnormalities include intrauterine growth restriction, hypotonia, craniofacial anomalies (macro/hydrocephalus, low-set dysplastic ears, broad nasal bridge), syndactyly, malformation of the extremities, adrenal hypoplasia, cardiac defects, and brain anomalies.
Mosaic triploidy (diploid/triploid mixoploidy) has been reported in approximately 20 patients. Triploid cells were found in both lymphocytes and fibroblasts, although in a number of cases the triploid cell line was limited to fibroblasts [237]. Patients with such mixoploidy are less severely affected than nonmosaics, and survival beyond 10 years has been observed. Usual clinical features include intrauterine growth restriction, psychomotor retardation, asymmetric growth, broad nasal bridge, syndactyly, genital anomalies, and irregular skin pigmentation [238]. Truncal obesity was seen in some patients [239]. A recent case of a 46,XX/69,XXY diploid/triploid mixoploid 8-year-old girl with normal female genital and ovarian development despite normal expression of SRY expression was reported [240].
Mitotic nondisjunction cannot readily explain the occurrence of diploid and triploid cell lines in the same individual. One possible mechanism is double fertilization of an ovum by two sperm; one sperm nucleus fuses with the ovum nucleus producing the diploid line, followed by a second sperm fertilizing one of the early blastomeres producing the triploid line. Cytogenetic evidence for such a mechanism has been reported in at least one case [241]. Another proposed mechanism supported by molecular evidence is delayed incorporation of the second polar body into one of the early blastomeres. The triploid cell line in this case is digynic [242].
Tetraploidy
The chromosome number in tetraploidy is 4n = 92. It is rarer than triploidy in spontaneous abortuses, seen in approximately 6–7% of such specimens with chromosome abnormalities [219, 220]. Tetraploid conceptuses usually abort spontaneously early in gestation and only rarely do they survive to term. A probable origin of tetraploidy is chromosome duplication in the zygote resulting from a failure of cytoplasmic division during the first division. Other theoretically possible mechanisms require the occurrence of two independent, rare events and are thus highly unlikely.
At least nine apparently nonmosaic tetraploid liveborns have been reported [243, 244]. The sex chromosome complement was either XXXX or XXYY. No 92,XYYY or XXXY conceptuses have been reported. The most frequent abnormalities were growth and developmental delay, hypotonia, craniofacial anomalies (short palpebral fissures, low-set malformed ears, high arched/cleft palate, micrognathia), and contracture/structural abnormalities of the limbs, hands and feet. Cardiac defects were present in four cases. Urinary tract abnormalities, such as hypoplastic kidneys, have also been recorded. Most patients died before 1 year of age. One girl had survived to 22 months at the time of report [245].
Mosaic tetraploidy (diploid/tetraploid mixoploidy) has been reported in at least 12 liveborns [246, 247]. This can occur as a result of postzygotic nondisjunction with failure of cytoplasmic division in a diploid conceptus. Tetraploid cells were seen in peripheral blood lymphocytes, skin fibroblasts, and bone marrow cells. In one severely malformed patient who died at 2 days of age, tetraploid cells were found in 95% of bone marrow cells [248]. In two females, aged 11 and 21 years, with severe intellectual handicaps and skin pigmentary dysplasia, tetraploid cells, were found only in skin fibroblasts [247]. In lymphocytes, the proportion of tetraploid cells decreases with age [249]. Overall, clinical features are similar to, but less severe than, those in nonmosaic tetraploidy patients. In addition to the longer survivals already mentioned, survivals to 6 years at the time of reporting have also been recorded [247, 250].
Partial Autosomal Aneuploidies
Partial duplication/deletion as a result of structural rearrangement is discussed in Chap. 9. Only those partial autosomal aneuploidies that result from the presence of a supernumerary chromosome and have been detected in postnatal specimens will be presented in this chapter.
Tetrasomy 5p
Tetrasomy 5p [47,XX or XY,+i(5)(p10)] resulting from the presence of a supernumerary isochromosome for the entire short arm of chromosome 5 is rare and has been reported in only five liveborns, all of whom were mosaics with both normal and abnormal cell lines [251, 252]. The abnormal cell line has been found in lymphocytes, skin fibroblasts, and chondrocytes. The phenotype appears to be similar to that of trisomy 5p. This includes hypotonia, seizures/abnormal electroencephalogram (EEG), psychomotor retardation, macrocephaly, facial dysmorphism, and respiratory difficulties. Skin hyperpigmentation was observed in two patients. Survival was variable; the most recent case reported was a 35-year-old male with a normal phenotype [252]. One patient died at 6 months of age, and another was 5 years old at the time of reporting.
Tetrasomy 8p
Tetrasomy 8p [47,XX or XY,+i(8)(p10)] usually results from the presence of a supernumerary isochromosome for the entire short arm of chromosome 8. All except one of the cases reported were mosaics, with both normal and abnormal cell lines. The abnormal cell line was found in lymphocytes and skin fibroblasts. In some cases, the origin of the abnormal isochromosome was confirmed by molecular cytogenetic (FISH) studies [253–255]. At least 12 cases have been reported [255–257]. A few patients died before the first year of life, but survival beyond 5 years was not uncommon. Weight and head circumference were normal at birth. The most frequently observed phenotypic features include mental retardation, speech and motor delay, dilatation of cerebral ventricles, mild facial dysmorphism (depressed nasal bridge, short nose, upturned nares, low-set and posteriorly rotated ears), and vertebral abnormalities. Agenesis of the corpus callosum was noted in six patients and cardiac defects in five. Deep palmar and plantar creases have also been reported. The phenotype resembles, to some degree, that of mosaic trisomy 8. A single apparently nonmosaic case was recorded with isochromosome 8p present in all blood lymphocytes while prenatal amniocytes showed a normal karyotype [257]. The girl had congenital ventricular septal defect, agenesis of corpus callosum, and facial, ear and bone anomalies.
Tetrasomy 9p
Tetrasomy 9p [47,XX or XY,+i(9)(p10)], resulting from the presence of a supernumerary isochromosome, has been reported in more than 20 liveborns [258–262]. The isochromosome consists of either the entire short arm of chromosome 9 as previously described, the entire short arm and part of the heterochromatic region of the long arm, or the entire short arm and part of the long arm extending to the euchromatic region. No consistent phenotypic differences have been observed among the three types. Both mosaic and apparently nonmosaic patients have been reported. The tetrasomy 9p cells were seen in both lymphocytes and skin fibroblasts. In contrast to tetrasomy 12p (described later), the 9p isochromosomes were present only in lymphocytes in five patients and in fibroblasts at a much lower percentage than in lymphocytes in two others [258, 259, 263, 264, 265, 266]. The mechanism for this observed tissue-limited mosaicism for different chromosomes is not clear.
Survival is variable, ranging from a few hours to beyond 10 years. The most frequent phenotypic abnormalities include low birth weight, growth and developmental delay, craniofacial anomalies (microphthalmia, low-set malformed ears, bulbous tip of the nose, cleft lip/palate, micrognathia), short neck, skeletal anomalies, joint contracture, nail hypo-plasia, and urogenital anomalies. Cardiac defects are present in more than 50% of patients. Diaphragmatic hernia was reported in an apparently nonmosaic patient [262].
Overall, nonmosaic patients are more severely affected. One patient, who had the i(9p) present in 75% of lymphocytes but not in skin fibroblasts, had only mild developmental delay and minor anomalies [258].
Tetrasomy 12p
Tetrasomy 12p (Pallister-Killian syndrome) results from the presence of a supernumerary isochromosome for the entire short arm of chromosome 12 [i(12)(p10) or i(12p)] (Fig. 8.10). The syndrome was first described in 1977 by Pallister et al. in two adults, a 37-year-old man and a 19-year-old woman [267]. In 1981, Killian and Teschler-Nicola reported a 3-year-old girl with similar clinical manifestations [268]. Subsequently, many cases have been reported, and many more have been observed but not reported in the literature [269, 270]. All cases were mosaics, with a normal cell line in addition to cells containing i(12p). Maternal age for reported cases has been shown to be significantly higher than that for the general population [271]. This observation has been taken to suggest that the isochromosome arises from a meiotic error and that the normal cell line results from subsequent loss of the i(12p) from some cells. In 6 of 7 cases studied by molecular analysis, the meiotic error was determined to be maternal [272, 273]. Tissue specificity and both the in vivo and in vitro age dependencies of the i(12p) have been well demonstrated [274]. The i(12p) is found in a high percentage of skin fibroblasts and amniocytes but is rarely seen in blood lymphocytes. The percentage of cells containing the isochromosome also decreases with age. The presence of tetrasomy 12p in 100% of bone marrow cells has been reported in at least two newborn infants and in only 6% of marrow cells in a 3½-year-old child [275–277]. In lymphocytes it has been found in fetal blood, but has never been seen beyond childhood [274, 278]. In a case reported by Ward et al., the i(12p) was present in 10% of lymphocytes initially but was not seen in these cells when the patient was 2 months old [275]. The isochromosome is more stable in skin fibroblasts and can be found in adults, usually at lower percentage than in younger patients. When fibroblast cultures were examined, the percentage of cells containing the isochromosome decreased with increasing numbers of cell passages [272, 274–276, 279]. One study using FISH showed that in lymphocytes, the i(12p) was present in a significantly higher proportion of interphase nuclei than in metaphase cells [280]. With the availability of array CGH (see Chap. 18), gain of 12p has been detected in total genomic DNA from blood specimens [281]. These indicate that lymphocytes containing i(12p) may fail to divide upon phytohemagglutinin (PHA) stimulation. These observations suggest that tissue-limited mosaicism in Pallister-Killian syndrome may result from differential selection against cells containing i(12p) in different tissues and that this selection can occur both in vivo and in vitro.
