Clinical characteristics of the subjects
Clinical and molecular characteristics of these 24 subjects are described in Table 1. Representative photographs of nine isolated macrodactyly patients are shown in Fig. 1. Twenty-two of 24 patients displayed asymmetric and disproportionate overgrowth in their hands or feet at birth. The remaining two patients had a later onset between six and twelve months after birth. Our sample ranged in age from 1 to 34 years, though except one 34-year-old patient, most are less than 15 years of age, with a median age of 5. A slight male predominance was observed (16 males versus eight females, binomial P = 0.152) in our cohort. Fourteen patients had exclusive involvement in the lower extremities, and nine had macrodactyly exclusively in the upper extremities, suggesting a slight predominance of the lower extremities (binomial P = 0.405). Only one patient had overgrowth of both the upper and the lower extremities. Twenty of 24 patients had unilateral involvement, including 12 V.S. 8 patients with right V.S. left side affected (binomial P = 0.503). Four of 24 patients had bilateral involvements. The number of affected digits ranged from one to seven, with an average of 2.7. The second digit is the most frequently affected digit (N = 22), followed by the third (N = 17). More than half of our patients (14/24) had two affected digits, and the combination of their respective locations at the second- and third-digit overgrowth (N = 10) was more frequently observed than double-digits enlargement of the first and the second (N = 4; binomial P = 0.180). In addition, no other combinations were observed when patients have two affected digits. Five of 24 patients had syndactyly, and all of them were presented with syndactyly of 2–3 toes.
Table 1
Summary of the clinical and molecular findings of studied subjects.
Patient No. | Gender | Age | Syndactyly | Affected limb | Affected digits | Gene | Variant | VAF |
1 | M | 11 | N | R-H | 2,3 | AKT1 | c.49G > A (p.Glu17Lys) | 22.03% |
2 | M | 11 | N | L-F | 1,2,3,4,5 | AKT1 | c.49G > A (p.Glu17Lys) | 11.16% |
3 | M | 2 | N | R-H | 2,3 | AKT1 | c.49G > A (p.Glu17Lys) | 9.93% |
4 | F | 10 | N | L-H | 3,4,5 | AKT1 | c.49G > A (p.Glu17Lys) | 20.57% |
5 | F | 6 | N | B-F, R-H | L-F: 2,3; R-F:1.2; R-H:2,3 | PIK3CA | c.1357G > A (p.Glu453Lys) | 11.10% |
6 | M | 3 | N | R-F | 2 | PIK3CA | c.1624G > A (p.Glu542Lys) | 24.48% |
7 | M | 4 | Y | L-F | 2,3 | PIK3CA | c.1624G > A (p.Glu542Lys) | 17.15% |
8 | M | 5 | Y | L-F | 1,2,3,4 | PIK3CA | c.1624G > A (p.Glu542Lys) | 20.95% |
9 | M | 4 | N | L-F | 2,3 | PIK3CA | c.1624G > A (p.Glu542Lys) | 17.10% |
10 | F | 4 | N | R-F | 2,3 | PIK3CA | c.1624G > A (p.Glu542Lys) | 27.58% |
11 | M | 3 | N | R-F | 1,2 | PIK3CA | c.1624G > A (p.Glu542Lys) | 17.79% |
12 | M | 2 | N | L-F | 1,2,3 | PIK3CA | c.1633G > A (p.Glu545Lys) | 19.11% |
13 | M | 2 | Y | R-F | 2,3 | PIK3CA | c.1633G > A (p.Glu545Lys) | 27.31% |
14 | M | 11 | N | B-F | L: 1,2,3; R:1,2,3,4 | PIK3CA | c.1636C > A (p.Gln546Lys) | 24.50% |
15 | F | 13 | N | B-F | 2 | PIK3CA | c.3139C > T (p.His1047Tyr) | 18.94% |
16 | M | 1 | N | R-H | 2,3 | PIK3CA | c.3140A > G (p.His1047Arg) | 25.63% |
17 | M | 2 | Y | R-F | 1,2 | PIK3CA | c.3140A > G (p.His1047Arg) | 23.29% |
18 | F | 5 | Y | L-F | 2,3 | PIK3CA | c.3140A > G (p.His1047Arg) | 21.45% |
19 | M | 2 | N | R-H | 1,2 | PIK3CA | c.3140A > G (p.His1047Arg | 25.57% |
20 | F | 10 | N | R-H | 1,2,3,4,5 | PIK3CA | c.3140A > G (p.His1047Arg) | 10.36% |
21 | M | 2 | N | L-F | 2,3 | PIK3CA | c.3140A > G (p.His1047Arg) | 20.03% |
22 | F | 6 | N | B-H | L:5; R:1,2,3 | PIK3CA | c.3140A > G (p.His1047Arg) | 15.82% |
23 | F | 3 | N | R-H | 5 | PIK3CA | c.3140A > T (p.His1047Leu) | 33.38% |
24 | M | 34 | N | R-H | 1,2 | PIK3CA | c.3140A > T (p.His1047Leu) | 18.99% |
M: male; F: female; Y: Yes or present; N: Not present; L: left; R: right; B: bilateral; VAF: variant allele frequency. NA: Not available. |
Patients 1 to 4 had variable presentations of cerebriform connective tissue nevi adjacent to overgrown digits. Therefore, diagnoses of Proteus syndrome were established in these patients.
Genetic characteristics
After genetic testing and interpretation, molecular diagnoses were achieved in all of our patients. In all four patients who met the diagnostic criteria of Proteus syndrome, we identified an AKT1 c.49G > A (p.Glu17Lys) variant, which is the only known variant leading to Proteus syndrome [10]. Variant allele frequencies (VAFs) of AKT1 in affected tissues range from 10–22%, with an average variant frequency of 16%. No variant read was identified in blood DNA (Table 1).
In 20 patients with isolated macrodactyly who do not met the diagnostic criteria of Proteus syndrome, we identified and confirmed pathogenic variants in PIK3CA from 20 patients (Table 1). The most commonly observed variants are PIK3CA p.His1047Arg (N = 7), followed by p.Glu542Lys (N = 6), p.Glu545Lys (N = 2), p.His1047Leu (N = 2), p.Glu453Lys (N = 1), p.Gln546Lys (N = 1) and p.His1047Tyr (N = 1) variations (Table 1). VAFs in affected tissues identified in range from 10–33% with the average of 21% (Table 1). None of these variants was identified in peripheral blood samples.
All of the seven variants have been previously reported to cause developmental disorders [5, 6, 11, 12]. These variants were either predicted or validated to have a gain-of-function mechanism. In vivo studies demonstrated that PIK3CA congenic variants could induce oncogenic transformation in chicken embryo fibroblasts by enhancing lipid kinase activity and activating mTOR and AKT1 signaling [13]. All seven variants were absent from the Deciphering Disorders Involving Scoliosis and COmorbidities (DISCO, http://discostudy.org/) study composed of 4000 exome sequencing data of Chinese population [7–9]. PIK3CA p.Glu453Lys, p.Glu542Lys, p.Gln546Lys and p.His1047Tyr were absent from the Genome Aggregation Database (gnomAD, https://gnomad.broadinstitute.org). PIK3CA p.Glu545Lys, p.His1047Leu, p.His1047Arg and AKT1 p.Glu17Lys were present at extremely low frequencies in gnomAD, with an allele frequency of 4e-6. Though the PIK3CA p.Gln546Lys, p.His1047Tyr and p.Glu453Lys variant had been observed in other forms of PROS [6, 11, 14–16], this is the first time for them to be identified in isolated macrodactyly.
PIK3CA protein has five functional domains, i.e. PI3K-ABD, PI3K-RBD, C2 PI3K-type, PIK helical and PI3K/PI4K kinase domain. p.Glu453Lys variant was located in the C2 domain. p.Glu542Lys, p.Glu545Lys and p.Gln546Lys variants occurred in adjacent amino acids of the helical domain. p.His1047Arg, p.His1047Tyr and p.His1047Leu variants were located at the kinase domain of PIK3CA (Fig. 2). The majority of established functional variants of PIK3CA cluster were found in the kinase and helical domains [17], which is in consistency with our finding in this study. In the 20 patients carrying pathogenic PIK3CA variants, nine had variants in the helical domain and ten were located in the kinase domain, and the remaining one variant located in the C2 domain.
Genotype-phenotype correlation
We then analyzed the potential correlation between subjects’ phenotypes (i.e. macrodactyly of the upper or the lower limb, the number of affected digits, with or without syndactyly) and the protein location of variants in patients carrying the PIK3CA variant (Table 2).
Table 2
Genotype-phenotype correlation
| In helical domain | In kinase domain |
Affected limb | | |
Upper limb | 0 | 6 |
Lower limb | 9 | 4 |
P-Value | 0.011 | |
Affected digits | | |
༜3 | 5 | 8 |
≥ 3 | 4 | 2 |
P-Value | 0.350 | |
All six (100%) patients presented with upper limb overgrowth carried variants in the kinase domain. In contrast, only four of thirteen (31%) patients who had overgrowth in the lower limbs had their variants located in the kinase domain. Therefore, there is a strong evidence for an association between the location of a genetic variant and the locus of macrodactyly, i.e. in the upper or lower limbs (P = 0.011; determined by Fisher’s exact test). We also compared the VAFs between the two phenotypically different groups, but no significant correlation was observed.
Thirteen patients have one to two affected digits, and eight of them carried variants located in the kinase domain. Six patients have more than two affected digits, and two of them had a variant in the kinase domain. Though it seemed that in patients with less than three affected digits, variants were more likely to be located in the kinase domain, and vice versa, this observation did not prove to be statistically significant (P = 0.350; determined by Fisher’s exact test).