This is the author’s version of the review article: Angulo-Urarte, Ana; Graupera, Mariona. «When, where and which PIK3CA mu- tations are pathogenic in congenital disorders». Nature Cardiovascular Research, Vol. 1 Núm. 8 (2022), p. 700-714. DOI 10.1038/s44161-022-00107-8 This version is available at https://ddd.uab.cat/record/267090 under the terms of the license 1 When, where and which PIK3CA mutations are pathogenic in congenital 2 disorders 3 Ana Angulo-Urarte1 and Mariona Graupera1,2 1 Endothelial Pathobiology and Microenviroment Group, Josep Carreras Leukaemia Research Institute (IJC), 08916 Badalona, Barcelona, Catalonia, Spain 2 Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), Instituto de Salud Carlos III, Av. de Monforte de Lemos, 5, 28029, Madrid, Spain 4 5 Corresponding Authors: 6 aangulo@carrerasresearch.org and mgraupera@carrerasresearch.org 7 8 9 Abstract 10 PIK3CA encodes for the class I PI3Kα isoform and is frequently mutated in cancer. 11 Activating mutations in PIK3CA also cause a range of congenital disorders featuring 12 asymmetric tissue overgrowth, known as the PIK3CA-related overgrowth spectrum (PROS), 13 with the vasculature frequently involved. In PROS, PIK3CA mutations arise postzygotically 14 during embryonic development leading to a mosaic distribution resulting in a variety of 15 phenotypic features. A clear skewed pattern of overgrowth favouring some mesoderm and 16 ectoderm-derived tissues is observed but is not understood. Here, we summarize current 17 knowledge on the determinants of PIK3CA-related pathogenesis in PROS, including 18 intrinsic factors such as cell lineage susceptibility and PIK3CA variant bias and extrinsic 19 factors which refers to the environmental modifiers. Gaining biological understanding of 20 PIK3CA mutations in PROS will contribute to unravel the onset and progression of these 21 conditions, and ultimately impact on their treatment. Given that PIK3CA mutations are 22 similar in PROS and cancer, deeper insight into one will also inform about the other. 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 PIK3CA encodes for p110α, one of the four class I phosphatidylinositol 3-kinase 41 (PI3K) catalytic subunits. p110α is an obligate heterodimer with a p85-type regulatory 42 subunit, with no evidence of the existence of p85-free p110α1–4. For simplicity, we will use 43 below PI3K when referring to the p110s isoforms. PI3Kα is ubiquitously expressed and is 44 activated by tyrosine kinases. PIK3CA is the most frequently mutated oncogene across all 45 human cancers with the high prevalence in breast and endometrial cancers5,6. Activating 46 mutations in PIK3CA span almost the entire PIK3CA gene product, with most frequently 47 mutated hotspots found in the helical (E542K and E545K) and kinase (H1047R) domains. 48 PIK3CA mutations are acquired in a somatic fashion and are largely present in 49 heterozygosity. Nevertheless, there is evidence of the presence of double PIK3CA 50 mutations in cis which further increase its PI3K activity7. 51 Our Review stems from the remarkable discovery of oncogenic mutations in PIK3CA 52 being causative of sporadic mosaic congenital disorders characterised by tissue overgrowth, 53 with the vascular compartment as the most frequently affected. These conditions have 54 become widely known as the PIK3CA-related overgrowth spectrum (PROS) and they can 55 range from isolated (e.g. skin-related lesions, vascular malformations, brain or muscle 56 overgrowth) to complex and syndromic phenotypes, where several tissues are affected (Fig. 57 1). Enigmatically, a clear biased pattern of disease manifestation, favouring some 58 mesoderm-derived tissues, is observed but is not understood8. PROS are considered 59 monogenic diseases; albeit emerging evidence indicate that co-occurrence of several 60 genetic events, at least in the vasculature, is more frequent than previously anticipated9–12. 61 This Review focuses on the pathogenic effects of somatic activating PIK3CA mutations 62 when are acquired at different developmental stages. We will discuss how the interplay 63 between genetics, cell identity and the environment explain the onset, progression, and 64 severity of these disorders. Also, we will provide an overview about the impact of distinct 65 PIK3CA variants in these congenital conditions. Finally, we include a dedicated section on 66 vascular malformations given that the vascular compartment appears most affected in 67 PROS. For congenital disorders caused by other PI3K signaling components we refer the 68 reader to Box 1. Of note, mirroring the similarities between RAS and PI3K congenital 69 manifestations, the term PIK3Copathies has been proposed when referring to all PI3K- 70 related conditions13. 71 72 Class I PI3Ks 73 PI3Ks are a large family of lipid kinases that catalyse the phosphorylation of the 3-hydroxyl 74 group of the inositol ring of different phosphatidylinositol (PtdIns) lipid substrates present at 75 the cellular membranes. In vertebrates, PI3Ks are divided into three classes (class I, class 76 II, and class III) based on their structure, substrate preference, distribution, mechanism of 77 activation and function (Box 2 includes extended information on class II and class III)3,14,15. 78 79 Basic concepts on class I isoforms. Class I PI3Ks are heterodimers composed of a 80 catalytic and a regulatory subunit4. The p110 (here refer to as PI3K) subunit confers the lipid 81 kinase activity while the regulatory subunit modulates the activity, stability, and subcellular 82 localization of the complex. Class I PI3Ks are subdivided into class IA and class IB 83 depending on their ability to bind to different regulatory subunits14–16. Class IA catalytic 84 subunits PI3Kα, PI3Kβ, and PI3Kδ (encoded by PIK3CA, PIK3CB and PIK3CD respectively) 85 interact with one of five p85-type regulatory subunits p85α (or its splice variants p55α and 86 p50α), p85β, and p55γ (encoded by PIK3R1, PIK3R2 and PIK3R3 respectively). PI3Kα and 87 PI3Kβ evenly interact with p85α and p85β. Instead, PI3Kδ preferentially binds to p85α17. 88 p85 stabilizes but inhibits PI3K kinase activity in the basal state. Upon stimulation, p85 89 allows PI3K activation by promoting their recruitment to pTyr residues in receptor tyrosine 90 kinases (RTK) and adaptor molecules1. Both, p85-mediated inhibition and recruitment to 91 pTyr residues occur via the same Src homology 2 (SH2) domains (in p85)1,15. Of note, the 92 (basal) inhibition of PI3Kα involves the nSH2 and iSH2 domains of p85, whereas PI3Kβ and 93 PI3Kδ also require inhibition by the sSH2 domain18–20. This may explain why mutations in 94 PI3Kα easier result in PI3K activation (loss of p85-dependent inhibitory effect) compared to 95 other catalytic subunits16,21. Class IB is solely composed by the PI3Kγ catalytic subunit 96 (encoded by PIK3CG gene) which may interact with one of two regulatory proteins, p101 or 97 p84/p87 (encoded by PIK3R5 and PIK3R6 genes respectively)22. Class I PI3Ks catalytic 98 subunits show specific expression patterns, being PI3Kα and PI3Kβ ubiquitously expressed 99 and PI3Kδ and PI3Kγ enriched in some cell lineages such immune cells, neurons, and 100 heart3,14,15. 101 All class I PI3K are activated by extracellular signals at the plasma membrane. PI3Kα 102 and PI3Kδ are recruited to the plasma membrane via binding of the SH2 domains of p85 to 103 tyrosine-phosphorylated proteins. Instead, the PI3Kγ heterodimer is activated by the Gβγ 104 subunits released by activated G protein-coupled receptor (GPCRs)22–24. PI3Kβ is unique in 105 that multiple active membrane receptors, including both RTKs and GPCRs, may potentially 106 recruit it and activate it25–27. This has led to the interpretation that full activation of this isoform 107 likely involves cooperation of several inputs, albeit further evidence is required to fully 108 demonstrate this. All class I PI3K catalytic isoforms contain a RAS-binding domain (RBD) 109 which allows them to interact with membrane-bound small GTPases and provide an extra 110 input of activation. This includes RAS for PI3Kα, PI3Kδ and PI3Kγ28–30 or RAC1 and CDC42 111 for PI3Kβ30. 112 113 Canonical class I PI3K signalling. Activated class I PI3Ks phosphorylate 114 phosphatidylinositol 4,5-bisphosphate (PtdIns(4,5)P2) at the plasma membrane and 115 generate the second messenger phosphatidylinositol 3,4,5-trisphosphate (PtdIns(3,4,5)P3; 116 also known as PIP3)31. A transient rise in PIP3 levels engages a signalling cascade that is 117 required for the regulation of the broad range of cellular functions including growth, 118 proliferation, metabolism, migration, and survival3,14. How and when PIP3 favours one or 119 another cellular function is not well understood but it quite likely involves (1) the activation 120 of PI3K by different extracellular inputs, (2) the intensity and duration of PI3K activation, (3) 121 the specific localization and amount of PIP3 produced, and (4) the rate of phosphate 122 hydrolysis; all ultimately leading to the activation of distinct downstream effectors32. PIP3 123 serves as ligand and functional regulator of a group of proteins which contain a pleckstrin 124 homology (PH) domain such as phosphoinositide-dependent kinase 1 (PDK1) and the 125 serine-threonine kinase AKT (also known as protein kinase B (PKB))2,33,34. Other PI3K 126 effectors with PH domains are tyrosine kinases (e.g., BTK in B-lympocytes), several GEFs 127 and GAPs that regulates small-GTPases of the RAC, RAS, RHO and ARF families (e.g., 128 GRP1, ARAP3) and protein adaptors (e.g., GAB1, GAB2, TAPP1 or DAPP)35. Their 129 recruitment and activation are isoform-selective and cell-type dependent comparing to a 130 more universal activation of AKT. Phosphatase and tensin homolog (PTEN) and Src 131 homology 2 (SH2) domain containing inositol polyphosphate 5-phosphatase 1 and 2 (SHIP1 132 and SHIP2) counterbalance the transient increase in PIP3. PTEN converts PIP3 back to 133 PI(4,5)P2 while SHIP dephosphorylates PIP3 into PI(3,4)P2, which is then further 134 dephosphorylated by the INPP4B phosphatase36,37. 135 AKT is the most widely studied effector of PI3K and comprises three isoforms (AKT1, 136 AKT2 and AKT3) which have different patterns of expression and localization38,39. Upon 137 binding to PIP3, AKT is recruited to the plasma membrane through its PH domain. There, 138 AKT is phosphorylated on Thr308 by PDK1, which is also recruited to the membrane through 139 its PH domain. Nevertheless, full activation of AKT requires an additional phosphorylation 140 on Ser473 by mammalian target of rapamycin complex 2 (mTORC2). Activated AKT can 141 exert its function in the cytoplasm and nucleus, where it phosphorylates and consequently 142 activates or inhibits different downstream substrates. More than 100 non-redundant 143 substrates of AKT have been identified39. Among them, we highlight tuberous sclerosis 144 complex 2 (TSC2) and forkhead box protein O1 (FOXO1) for their importance in cancer 145 biology. Activated AKT phosphorylates, and in turn inhibits both TSC2 and FOXO1. Upon 146 phosphorylation, TSC2 loses its ability to inhibit mTOR complex 1 (mTORC1). Of relevance, 147 activation of mTORC1 occurs at multiple levels, thus, it is incorrect to assume that PI3K 148 signalling encompasses full activation of mTORC140. FOXO1 is phosphorylated by AKT at 149 3 serine/threonine residues, which results in its nuclear exclusion and in turn in the 150 inactivation of its transcriptional activity41. 151 152 Oncogenic PIK3CA mutations beyond cancer 153 In 2012, activating PIK3CA mutations were linked for the first time to mosaic, congenital, 154 and progressive overgrowth disorders for which the name PROS was coined42–45. PROS 155 features anatomically variable admixture of overgrown tissues, with vasculature and adipose 156 tissue most severely affected macroscopically. Previously described disorders that now are 157 grouped under the umbrella of PROS are: Congenital Lipomatous Overgrowth, Vascular 158 malformations, Epidermal nevi, Scoliosis/ skeletal and spinal (CLOVES) syndrome44; 159 Capillary malformation of the lower lip, Lymphatic malformation of the face and neck, 160 Asymmetry of the face and limbs, and Partial or generalized Overgrowth (CLAPO) 161 syndrome46; Klippel-Trenaunay Syndrome (KTS)47; Dysplastic MegalEncephaly 162 (DMEG)/HemiMegalEncephaly(HME)/Focal cortical dysplasia (FCD) ; FibroAdipose 48 163 hyperplasia or Overgrowth (FH/FAO)43; Fibroadipose Infiltrating Lipomatosis/facial 164 infiltrative lipomatosis (FIL)49; HemiHyperplasia Multiple Lipomatosis (HHML)50; 165 Macrodactyly51; Muscular HemiHyperplasia (MHH)52; Diffuse Capillary Malformation with 166 Overgrowth (DCMO)53; Lipomatosis Of Nerve (LON)54; Megalencephaly-Capillary 167 malformation syndrome/macrocephaly-capillary malformation (MCAP/M-CM)42; and 168 FibroAdipose Vascular Anomaly (FAVA)55 (Fig. 1). 169 A few years later somatic activating PIK3CA mutations were discovered as a cause 170 of congenital sporadic venous malformations (VMs) and lymphatic malformations (LMs)55– 171 .This exposed that, beyond being associated with complex phenotypes, PIK3CA-related 58 172 vascular malformations may also occur in isolation. Since then, different subtypes of new 173 PIK3CA-related disorders have been described in the literature. This has confused the field 174 as it is not clear whether all or only some pertain to the so-called PROS. Strictly speaking, 175 all conditions involve tissue overgrowth (beyond other phenotypes); thereby suggesting that 176 all should be grouped under the umbrella of PROS. Nevertheless, we favour the 177 subclassification proposed by Mirzaa and colleagues which distinguishes between isolated 178 or syndromic PROS59. The former includes any clinical manifestations which occurs as a 179 focal lesion affecting only one tissue or body part. Instead, syndromic PROS are those 180 conditions in which tissue overgrowth is not focal, affects several tissues and is presented 181 with others features (Fig. 1). The terms isolated or syndromic PROS will be used across this 182 Review. Of note, Victor Martinez-Glez et all., have recently described PIK3CA mutations in 183 patients that present segmental undergrowth (in length or volume) of musculoskeletal 184 tissues together with vascular malformations and with or without associated overgrowth60. 185 This reflects that the understanding of PIK3CA-related congenital disorders is still in its 186 infancy and indicates that current classification may need to be revisited in the future. 187 188 Determinants of clinical phenotypes in PROS. Overgrowth in PROS is characteristically 189 present at birth, progressing during childhood and sometimes adulthood. Activating PIK3CA 190 mutations in PROS arise postzygotically and stochastically during embryonic development 191 leading to a mosaic distribution where only a subset of cells carries the mutation resulting in 192 a variety of phenotypic features. 193 194 Germline vs. Mosaicism. Within PROS, most activating PIK3CA mutations have been 195 detected in a mosaic fashion with very low allelic frequency in the affected tissues, and 196 absent in blood cells. Likely, activating mutations in PIK3CA in the human zygote cause 197 early embryonic death61. This has been shown in mice where expression of the Pik3caH1047R 198 variant in the germline leads to embryonic lethally62,63. This is not unique of PIK3CA 199 mutations, as many oncogenes have a dominant lethal activity that can only survive through 200 mosaicism13,64. Of note, germline mutations in PIK3CA have been reported in 13 cases of 201 PROS with macrocephaly. Ten of these cases carried missense mutations of uncertain 202 significance, and none were identified in the cancer hotspot sites42,65–67. There is a recent 203 case of a child showing a mild PROS phenotype which presented a PIK3CA G364R germline 204 mutation. This variant is annotated as functional activating mutation in cancer5 and was 205 detected in 50% variant allelic frequency (VAF) in peripheral blood cells, buccal smears, and 206 skin fibroblasts. Patient-derived fibroblasts carrying this mutation showed increase PI3K 207 signalling67; albeit it is not clear whether the increase in PI3K activity occurs at the same 208 level as cancer hotspots. Together, these data suggest that there is a threshold of PI3Kα 209 activity that a living organism can tolerate, with quite likely only weak PIK3CA variants 210 surviving in the germline. While PIK3CA mutations are primary presented in a mosaic 211 fashion in PROS, it is still not clear whether overgrown lesions are exclusively composed of 212 mutant cells, or they are also mosaic. 213 214 Does the time of mutation acquisition define PROS clinical severity? Clinical severity 215 is defined by simultaneous presence of pain and disability. Without exceeding the threshold 216 of PI3Kα activity that it is compatible with life, it is expected that postzygotic activating 217 PIK3CA mutations which arise at early developmental stages affect a higher number of cell 218 lineages leading to a more widespread, pleiotropic, and severe condition. On the other hand, 219 if PIK3CA mutations were acquired later in development or after birth it would result in a 220 lineage-specific pathogenesis, such as that found in isolated vascular malformations. This 221 has led to the assumption that the latter is a less clinically severe PROS. The implementation 222 of next generation sequence (NGS) into the clinical practice to diagnose PROS has allowed 223 to study and follow large cohorts of patients. This has provided substantial evidence that 224 there is not always a clear correlation between the type of cell lineages which carry the 225 mutation, the VAF of the mutation in the affected tissue and the severity of the clinical 226 outcome68. In fact, there are patients who develop an isolated, but very severe lesion and 227 other patients with a widespread overgrowth, but with lower severity. This indicates that 228 severity primary relies on the anatomic location and extension of the overgrown tissue rather 229 than the degree of widespread. This also suggests that severity and phenotypes (number of 230 tissues affected) are not synonymous in PROS. In addition, it poses the notion that it is not 231 accurate to assume that the earlier a mutation appears the more severe the pathogenic 232 outcome is. Instead, we believe that intrinsic (cell-autonomous) and extrinsic factors to which 233 mutated clones are exposed to, are also key determinants to PROS severity; including cell 234 lineage that acquired the mutation (e.g., mesoderm vs endoderm precursor, progenitor vs. 235 differentiated cell), the degree and mechanism of PI3Kα activation (PIK3CA variant) and the 236 spatiotemporal environmental modifiers of PI3Kα signalling (availability of growth factors, 237 paracrine activation of wild-type cells surrounding mutant cells, cell-cell and cell-extracellular 238 matrix (ECM) interactions and mechanical signals among others). We propose that the final 239 phenotypic outcome of PROS would be the consequence of a unique combination of all 240 these parameters (which, when and where). Below we further develop each of these aspects 241 (Fig. 2). 242 243 Tissue patterns of PIK3CA pathogenesis. A remarkable observation within the variable 244 spectrum of PROS is the clear biased pattern of tissues that present phenotypic traits, 245 including adipose, muscle, bone, nervous, vascular and, skin (epidermis and dermis)44. Yet, 246 the exact cell type or types that carry the mutation in each case is not always clear. Most of 247 genetic testing has been done using biopsies of affected tissues in which many different cell 248 types are found, including non-mutated cells. In addition, the VAF within these biopsies 249 range from 0.5% to 50% indicating for instance that in cases where 1% VAF is detected only 250 2 out of 100 cells carry the mutation. However, there is an intrinsic variability based on 251 sample handling; often patients need to be biopsied several times before the presence of a 252 PIK3CA mutation is detected69. Cell type specific isolation and in vitro culture of patient 253 derived cells have clarified that PIK3CA mutations are present in keratinocytes70, blood 254 endothelial cells (BECs) and lymphatic endothelial cells (LECs)69,71–73, fibroblasts43,74–76 and 255 adipose-derived stem cells and adipocytes77,78. Most of the tissues carrying a PIK3CA 256 mutation are mesodermal derivates (vasculature, adipose, muscle, and bone) and/or 257 ectodermal derivates (nervous tissue, epidermis, and connective tissues of the head)48,54,79. 258 Within the nervous tissue, it is not clear which specific cell linages carry PIK3CA mutations 259 as biopsies have not discriminated between neurons, macroglia and microglia (being 260 neurons and macroglia of neuroectodermal origin and microglia derived from the mesoderm 261 line)48,78,80,81. Also, it is incorrect to consider that all PROS-related neuropathies involve 262 overgrowth of neuroectodermal derivates. For example, LON (lipomatosis of nerve) is a 263 subtype of isolated PROS in which patients suffer from enlargement of nerve bundles 264 primarily caused by overgrown adipose and fibrous tissue54. In line with this, a recent case 265 report has confirmed that PIK3CA mutations in LON are prominently found in mesoderm- 266 derivate lineages81. 267 No endodermal-derived tissues (e.g., epithelial lining of the gastrointestinal and 268 respiratory tract, the parenchyma of tonsils, liver, thymus, thyroid, parathyroid and pancreas) 269 are usually found phenotypically affected in PROS. This is not unique of congenital 270 disorders, as PIK3CA mutations also have a dominant role in ectodermal and mesoderm- 271 derived cancers such as breast and endometrial 82,83. An enigmatic aspect about the lineage 272 skewing pattern in PROS is whether PIK3CA mutations are present, but silent, in non- 273 pathogenic tissue, or instead they are not present in those tissues. One possibility is that 274 mesoderm and ectoderm-derived tissues are more sensitive to PI3K overactivation while in 275 endoderm-derived tissues PIK3CA mutations are not enough to cause pathogenesis. It is 276 also possible that mutation acquisition favours differentiation into specific lineages, as 277 shown in breast cancer84. In fact, homozygous PIK3CA mutations induce self-sustained 278 stemness and resistance to spontaneous differentiation in human induced pluripotent stem 279 cells (iPSC)85,86. This suggests that PIK3CA mutations persist better in less differentiated 280 cell states. Nevertheless, we cannot rule out that PIK3CA mutant clones undergo negative 281 selection (either by out competition or by cell death) in specific cell lineages. Of note, 282 PIK3CA mutations have been found in healthy adult tissues from endoderm-derived tissue 283 such as oesophagus87,88. While PIK3CA mutant clones outcompete their wild-type 284 neighbours in that context, they do not lead to abnormal tissue growth87. This would fit with 285 PIK3CA mutations being silent in endoderm-derived tissues. 286 287 The bias of PIK3CA variants in PROS. Missense activating mutations in PIK3CA span 288 almost the entire gene in cancer and PROS. In cancer, more than 80% of somatic mutations 289 are found in three hotspots located in the helical (E545K, E542K) and kinase (H1047R) 290 domains5. While the mutational profile of PIK3CA in PROS is similar than in cancer, the 291 occurrence of mutations other than these three hotspots is much higher68 (Fig. 3). Mutations 292 with lesser gain-of-function activity are quite likely no that frequent in cancer because of their 293 lower oncogenic potential. Instead, it seems that mild and weak activating PIK3CA mutations 294 are enough to generate a pathogenic response when acquired at embryonic stages. Indeed, 295 G914R and E726K, which are likely non-strong activating PIK3CA mutations, are also 296 hotspots in PROS7,68. Intriguing, sporadic cases of isolated PROS with two mutations have 297 been identified, where a hotspot mutation is combined with a non-hotspot mutation48,89. Yet, 298 it is not clear whether these mutations are presented in the same clone in cis or trans or in 299 different clones. It is important to bear in mind that cancer hotspot mutations have been 300 preferentially mapped for genetic testing in PROS which has quite likely underestimated the 301 occurrence of mutations beyond the hotspots mentioned above. In line with this, there is 302 also a bias in the clinical visibility of severe cases which tend to overrepresented for genetic 303 testing90. 304 The emerge of numerous genetic studies in the context of PROS is allowing for the 305 first time to conceptualize phenotypes from genotypes. For example, the majority of MCAP 306 (megalencephaly-capillary malformation syndrome) patients with a reported genetic 307 diagnosis carry a non-hotspot mutation in PIK3CA with G914R and E726K being the most 308 common variants42,68,76,91–95. Also, MCAP is a subtype of PROS disorder in which affected 309 tissues derive from two different developmental layers, ectoderm (neurons and macroglia) 310 and mesoderm (vasculature and microglia); thereby suggesting that weakly activating 311 mutations in PIK3CA are compatible with their existence before the divergence of the 312 germline layers. Another study with a large cohort of patients with lymphatic malformations 313 (LMs) has revealed genotype-phenotype associations; with cancer hotspots (E545K, 314 E542K, H1047R) being overrepresented in localized LMs and KTS. 69 . On the other hand, 315 non-hotspot mutations were found significantly more frequently and at higher VAFs in LMs 316 presented in CLOVES and unclassified PROS. Many clinical units worldwide are currently 317 implementing in their routine pipeline genetic studies for PROS patients. We anticipate that 318 this will shed light into new genotype-phenotype correlations. 319 320 Output of PIK3CA variants. Based on the impact of each mutation on the protein 321 conformation of PI3Kα 96,97, PIK3CA mutations display quantitative differences in the intrinsic 322 PI3Kα lipid kinase activity. For example, the H1047R variant (1) leads to increase interaction 323 of PI3Kα with lipid membranes, (2) enhances PI3Kα kinase activity under growth factors 324 stimulation and (3) becomes insensitive to RAS binding98–102. On the other hand, the 325 E545K/E542K helical variants require RAS-GTP binding to be fully activated but are no 326 longer inhibited by p85 (the regulatory subunit). This propels PI3Kα in a basal active state 327 that mimics the activation induced by RTK98,102,103 and explains why growth factors’ 328 stimulation does not add greater activity to PI3Kα compared to basal state100. Instead, 329 mutants in the C2 domain, such as the C420R variant, result in increased positive surface 330 charge; thereby leading to an enhanced recruitment of PI3Kα to cellular membranes96. 331 Indeed, C420R shows a greater increase in PI3K signalling than E545K/E542K upon growth 332 factor stimulation96,104. Of relevance, there is a large amount of less frequent mutations in 333 PIK3CA for which there is very little knowledge. Several groups have shown that most of 334 non-hotspot mutations are also gain-of-function mutations and signal constitutively through 335 AKT104,105. Yet, structural insights and biochemical insights of the impact of these less 336 frequent variants are lacking which hampers the understanding of the mechanisms by which 337 they promote high PI3Kα activity. 338 New evidence has emerged that distinct PIK3CA variants induce different molecular 339 programs in glioblastoma106. In breast cancer, instead, the expression of the same variant 340 in different mammary gland populations results in different molecular programs which cause 341 different tumour types and clinical outcomes84. While these differences are yet to be 342 described in PROS, the variety of clinical manifestations in these conditions suggests that 343 PIK3CA variants exhibit qualitative differences by means of variant-specific molecular 344 signals106. Another important unresolved question is whether PIK3CA variant-related 345 different pathogenesis confer differential susceptibility to classical PI3K inhibitors. 346 347 Extrinsic factors in PROS pathogenesis. The comparison between PIK3CA-related 348 phenotypes in PROS and cancer suggests that the timing when the mutation is acquired 349 (development vs. adult) confers different susceptibilities to PIK3CA mutations. Tissue 350 growth is chiefly taking place during embryogenesis and early postnatal periods, during 351 which most cells in the organism divide and growth extensively. It is during these growing 352 phases when activating mutations in PIK3CA favour PROS onset. In line with this, tissues 353 with high plasticity which are in constant adaptation to the microenvironmental needs, such 354 as the vasculature and adipose tissues, are highly affected in PIK3CA-related congenital 355 disorders. Recent data have shown that onset and growth of Pik3ca-related vascular 356 malformations relies on the synergy between Pik3ca mutations and growth factors71,107. This 357 explains why in the adulthood when the growth factor signals are residual, these lesions do 358 not form the novo or existing ones progress very slowly. This also fits with the observation 359 that many lesions regrowth after incomplete surgical removal, when the body reacts locally 360 busting the production of growth factors to promote wound closure and explains why some 361 asymptomatic lesions ignite its growing during injury, adolescence and pregnancy (hormonal 362 changes). Thus, patients may benefit from therapies in which, at specific timing windows, 363 for example after a resection, microenvironment-derived paracrine specific signals are 364 inhibited107. The FAVA disorder is an example of a PROS condition in which patients are 365 asymptomatic at birth with lesions developing in the extremities during late childhood and 366 adolescence. In fact, some patients have reported that FAVA lesions appear after an 367 accidental event causing physical injury (Eulàlia Baselga’s personal communication) which 368 is coherent with the push-growth notion, as damage often results in hypoxia and acute 369 production of growth factors. The notion that growth factors are critical for PROS 370 pathogenesis opens the discussion if the degree of widespread is dependent on in situ 371 production of growth factors at the time that a PIK3CA mutation is acquired. This would imply 372 that the penetrance (proportion of cells carrying a particular variant) of PROS would be 373 primary linked to the local production of lineage-specific growth factors when the mutation 374 occurs. 375 It is still not fully understood to at what extend tissue overgrowth in PROS relates to 376 cell growth and/or cell proliferation. While PI3K signalling mediates both cellular functions, 377 cell growth and cell cycle can be also independently regulated. Modelling Pik3ca-related 378 single and complex PROS in mice has shown that overactivated PI3Kα results in an obvert 379 hyperplasia with an increase in the projected surface area; thereby indicating that PIK3CA 380 mutations alter both proliferation and growth108,109. However, data on PROS patients treated 381 with alpelisib (an allosteric PI3Kα specific inhibitor) have showed that, upon treatment, 382 overgrowth lesions essentially reduced their volume without cell death which suggests that 383 treatment primary interferes with cell size . Albeit it is also possible that such prominent 108 384 effect on lesion size relates to reduce swelling. Given that PROS hyperplastic phenotypes 385 largely depend on the presence of growth factors, it is tempting to speculate that onset of 386 PROS relies on cell proliferation and that the slow progression during the lifetime of the 387 patient is more dependent on the intrinsic cell growth. This would fit with the observation 388 that PROS lesions are considered as non-proliferative lesions at the time of diagnose110. 389 Cells are in constant exposure to biomechanical cues (shear, tensile and 390 compressive stresses)111, being at foremost play during developmental stages112,113. Indeed, 391 mechanotransduction (the cellular response induced by biomechanical cues) is believed to 392 contribute to cell fate decisions114–116. This is particularly relevant for those cells/lineages 393 which co-exist in cell-cell or cell-extracellular matrix contact. For example, endothelial cells 394 (ECs) which establish adherent junctions to one another and are in constant interaction with 395 both their luminal and abluminal extracellular space are extremely dependable of 396 mechanobiology signalling117. Intriguing, the anatomical location of a mutant clone has been 397 recently identified as critical factor for tumorigenesis118. Hence, it is tempting to speculate 398 that tissue architecture and mechanical forces contribute to define such anatomical-related 399 pathogenesis. Several evidence suggest that oncogenic signalling synergises with 400 mechanotransduction to promote pathogenesis119–121. Specifically, PI3K signalling 401 cooperates with several components of the mechanobiology machinery such as YAP/TAZ, 402 cadherins and actin remodelling proteins122–124. Based on these novel concepts, it is 403 tempting to speculate that aberrant crosstalk between biomechanical cues and PIK3CA 404 contribute to the clinical manifestation of PROS. 405 Other reports have identified that PIK3CA-related pathogenesis is supported by non- 406 cell autonomous mechanisms. For example, Martin-Corral et al. have showed that the 407 presence of Pik3ca mutant clones in lymphatic vessels results in the accumulation of 408 immune cells, including macrophages, which then become a major source of VEGF-C than 409 can further promote pathological lymphangiogenesis107. Other such studies have provided 410 evidence that PIK3CA mutant clones in tumours, through paracrine communication, interfere 411 with wild-type or HER2 mutant clones in close proximity and thus catalyse their aberrant 412 behaviour125,126. While this has not been reported in the context of PROS yet, it is possible 413 that mosaic lesions also require to attract non-PIK3CA mutant cells for pathogenesis. 414 Indeed, this has been reported in some vascular malformations127,128. We anticipate that 415 understanding how mutant clones hijack these other clones may open new therapeutic 416 opportunities. 417 418 PIK3CA mutations in vascular malformations. 419 Vascular malformations occur in both isolated and syndromic PROS. Even within the 420 syndromic PROS, which tend to feature a variable admixture of overgrown tissues, the 421 vascular compartment/tissue is most commonly affected. This indicates that vascular 422 malformations are a hallmark of PROS. This is coherent with PI3Kα being a master regulator 423 of endothelial cell biology (Box 3). Vascular malformations are abnormal vessels that grow 424 aberrantly during embryonic development and are manifested at birth (congenital) or 425 throughout the life of affected individuals. They slowly grow and do not regress 426 spontaneously over time. Depending on the type(s) and localization of the affected vessels, 427 patients’ symptoms can range from mild to severe, even life-threatening. These vascular 428 lesions often cause pain, swelling or bleeding, together with cosmetic deformities that can 429 interfere with the normal function of the affected areas. Recent evidence shows that 430 occasionally vascular malformations are not clinically manifested until a pathophysiological 431 condition such as adolescence, pregnancy, or injury triggers their growth129,130. Due to the 432 variability in the clinical manifestations, the diagnosis and treatment are not easy and require 433 a multidisciplinary team of specialists. 434 Subtypes of PIK3CA-related vascular malformations. Vascular malformations are 435 divided in low-flow (venous, lymphatic and capillary) and fast-flow (arteriovenous) lesions. 436 They can also be classified as simple, when only a specific vascular bed is affected such 437 as capillary, venous, lymphatic, or arteriovenous anomalies, or combined, when a lesion 438 presents two or more vascular malformations or mixed vascular beds characteristics such 439 as capillary-venous malformations (CVMs), lymphatic-venous malformations (LVMs) or 440 capillary-lymphatic venous malformations (CLVMs), among others. Intriguing, PIK3CA- 441 related vascular malformations are restricted to low-flow lesions, including isolated venous 442 malformations (VMs), capillary malformations (CMs), and lymphatic malformations (LMs) or 443 in combination. It is not clear why the presence of PIK3CA mutations has not been reported 444 in arteriovenous malformations (AVM). One possibility is that PIK3CA mutations behave as 445 silent mutations in arteries. Indeed, arterial ECs exhibit molecular refractoriness to other 446 vascular-related mutations131. However, targeted sequencing for the PIK3CA gene in more 447 than 100 surgically resected human brain AVMs did not identify any positive case9,10. 448 Another possibility is that arterial mutant clones are negatively selected during vascular 449 development or homeostasis. Studies using inducible genetic models which allow to express 450 PIK3CA mutations in different endothelial cell populations would help to clarify this 451 conundrum. Below we summarize the most recent relevant aspects of PIK3CA-related 452 specific vascular malformations subtypes. For more general aspects, we refer the reader to 453 specialized reviews on vascular malformations132–135. 454 Lymphatic malformations (LMs): a prototypical example of PIK3CA dominance. LMs 455 are debilitating vascular anomalies classified as cystic LMs (micro- or macro-cystic) and 456 complex lymphatic anomalies. While the former appears in isolation, complex lymphatic 457 anomalies show diffuse and multifocal pattern and may cause defects in the central 458 collecting lymphatic channels such as generalized lymphatic anomalies (GLA), Gorham- 459 Stout disease (GSD), kaposiform lymphangiomatosis (KLA), and central conducting 460 lymphatic anomalies (CCLA). Activating mutations in PIK3CA have been detected in the 461 majority of cystic LMs and GLA 55,136. Of significance, there is a clear genotype to phenotype 462 association in PIK3CA-related LMs, with the so-called cancer hotspots (H1047R, E545K and 463 E542K) being dominant. While this could be explained by the notion that variants with lesser 464 gain-of-function activity are not sufficient to induce pathogenesis in the lymphatic 465 endothelium, it is also possible that this vascular compartment is more tolerant than other 466 linages to high PI3K signalling. Of note, mutations in the helical domain (E545K and E542K) 467 are more common than H1047R in LMs , albeit the significance of this remains to be 69 468 determined. 469 The generation of mouse models of PIK3CA-related LMs has provided significant 470 insights on the molecular and cellular factors that determine the onset and the subtype of 471 LMs107,108,136. For example, the expression of Pik3caH1047R mutation in VEGFR3-positive cells 472 during early embryonic development recapitulates traits of macro-cystic LMs. Instead, if the 473 same mutation is expressed in VEGFR3-positive cells during late embryonic or early 474 postnatal stages, mice develop micro-cystic LMs107. Another study showed that the 475 expression of P110* (a dominant active Pik3ca transgene with 20 times higher kinase 476 activity than H1047R) in VEGFR3 positive cells in adult mice causes multifocal LMs108. 477 Similarly, the expression of the Pik3caH1047R mutation in PROX1-positive lymphatic cells after 478 weaning resulted in GLA136. These data suggest that the type of cell/precursor, the time of 479 activation and the mouse modelling genetic approach used to activate of PI3Kα signalling 480 are key factors in the onset of different subtypes of LMs. Another important lesson learnt 481 from these mouse models is that, rapamycin, an allosteric inhibitor of mTOR, alone is not 482 sufficient to revert LMs107,108. Martin-Corral et al. showed that VEGFC-VEGFR3 dependent 483 activation of mutated PI3Kα promotes the growth of microcystic LMs in mouse. This explains 484 why combined inhibition of mTOR and VEGFC leads to the regression of microcystic LMs 485 107 . Recently, it has been shown that alpelisib ameliorates LM symptoms in mouse models 486 and in patients with cystic LMs that previously do not respond to rapamycin108. Currently, it 487 is not clear why rapamycin and alpelisib induced different responses in LMs. However, it is 488 tempting to speculate that the latter provides an overall better response due to a direct 489 inhibition of the mutant protein. Also, given that rapamycin only targets a branch of the PI3K 490 pathway, it is possible that PIK3CA-related pathogenesis occurs through mTOR dependent 491 and independent mechanisms. Collectively, this emphasises the importance of generating 492 faithful preclinical models for each subtype of LMs towards the so-called personalized 493 medicine. 494 495 PI3K overactivation in venous malformations (VMs): a matter of PIK3CA and TEK. VMs 496 are bluish lesions caused by aberrant EC proliferation that show enlargement, tortuosity, 497 reduced mural coverage and impaired functionality. VMs are classified into Common VMs, 498 Familial VM cutaneo-mucosal (VMCM), Blue rubber bleb nevus syndrome (BRBNS), 499 Glomuvenous malformations (GVM), Cerebral cavernous malformation (CCM), Familial 500 intraosseous vascular malformation (VMOS), and verrucous venous malformation (VVM). 501 Common VMs are the most frequent VMs (90% of all VMs). They are caused by activating 502 mutations in TEK (60%)137,138 or PIK3CA (20-25%)56–58 with both mutations being largely 503 mutually exclusive. While PIK3CA mutations only occur in a somatic fashion, both somatic 504 and germline mutations in TEK can cause VMs. PIK3CA and TEK mutations lead to 505 increased PI3Kα signalling, albeit TEK mutations activate the pathway to a lower extent 506 . This quite likely explains why some TEK mutations are compatible with their survival 56,71,139 507 in the germline. Of note, there is a clear tissue-genotype association between TEK and 508 PIK3CA, with TEK-related VMs being mostly found in the skin surface while PIK3CA-related 509 VMs are preferentially located in intramuscular areas56,57. An important aspect to consider 510 is that TEK-related lesions tend to be purer VMs than PIK3CA-related lesions which often 511 also express some lymphatic markers. Data on the co-occurrence of PIK3CA and TEK 512 mutations in the same lesion are also emerging57,139. Yet, it is too early to say whether 513 second events play a wider role in the severity of these diseases and whether multi-genetic 514 events co-occur in the same cells. 515 Mouse models of Pik3ca-related common VMs have also flourished. For example, 516 Castillo et al. reproduced the aetiology of VMs by widespread mosaic induction of 517 Pik3caH1047R under the endogenous promoter in the mouse lateral plate mesoderm during 518 embryonic development. At present, it is still enigmatic why this mouse model solely 519 develops vascular malformations, predominantly VMs, while avoiding major alterations in 520 other tissues. Based on these data, it is tempting to speculate that the type of PROS also 521 relates to the subtype of (mesoderm) precursor where the mutation occurs. Another study 522 has shown that ubiquitous, but mosaic, expression of Pik3caH1047R in adult mice also leads 523 to a rapid and exclusive development of VMs57. Collectively, these mouse models have 524 shown that Pik3ca-related VM form through proliferation and are deprived of mural cells57,58. 525 Xenographs models using human - ECs derived from patients have also emerged139. While 526 these models become a relevant tool for preclinical testing, they have limitations for the 527 understanding of the onset and biology of these diseases. Taken together, data from 528 modelling PROS in mice support the notion that BECs are particularly sensitive to PIK3CA 529 pathogenesis. 530 531 Cerebral cavernous malformations and PIK3CA: the advent of multigenic events in 532 vascular malformations. Activating PIK3CA mutations have been also found in cerebral 533 cavernous malformations (CCMs), that are capillary-venous malformations specifically 534 located in the brain and spinal cord. There are two types of CCM diseases, familiar CCMs 535 (20% of CCMs) and sporadic CCMs (80% of CCMs). Until recently it was believed that CCM 536 were monogenic diseases, largely caused by inherited or somatic loss-of-function mutations 537 in one of the three genes (KRIT1 (also known as CCM1), CCM2 or PDCD10 (also known as 538 CCM3)) that encode for the heterotrimetic CCM protein complex140,141. In 2021, somatic 539 gain-of-function mutations in MAP3K3 (encoding for MEKK3) were also found in sporadic 540 orphan CCMs142–144. In CCM, MAP3K3 mutations largely lay in the I441M spot, and lead to 541 enhanced activation of MEKK3 signalling. In addition, several studies have showed that 542 CCMs with a prominent and rapid growth and associated with strokes and seizures, carry 543 an additional somatic genetic hit in PIK3CA, chiefly on a cancer hotspot10,142,143. While 544 mutations in MAP3K3 and CCMs genes are mutually exclusive, PIK3CA mutations may co- 545 occur with any (MAP3K3, KRIT1, CCM2 and PDCD10). This is coherent with the observation 546 that KRIT1, CCM2, PDCD10 and MAP3K3 mutations all lead to activation of MEKK3 547 signalling. 548 Modelling CCMs in mice using endothelial-specific CreERT2 mouse models has 549 confirmed that the synergy between loss of Krit1 and expression of Pik3ca occurs by 550 interaction within or between ECs10. In line with this, isolation of ECs from human CCM has 551 confirmed that these cells carry PIK3CA mutation142. However, another such study in mice 552 has proposed that Pik3ca mutations in CCMs occur in pericytes, a subtype of mural cells 553 which adhere to and support capillary endothelial function9,145. Yet, these data remain 554 controversial as the Cre mouse line employed to activate the expression of Pik3caH1047R is 555 neither inducible nor pericyte-specific. In addition, no proof that ECs in the mouse model do 556 not carry Pik3ca mutations is provided146. In support of a possible involvement of mural cells 557 in the CCM disease, another report showed that specific deletion of Pdcd10 in mural cells, 558 including both pericytes and smooth muscle cells, resulted in CCM. While proof that human 559 brain pericytes carry PIK3CA mutations would be required to validate these findings, it is 560 possible that PIK3CA mutations in endothelial cells and mural cells account for different 561 subtypes of CCMs. Of note, pericytes do not rely on PI3Kα, but PI3Kβ to activate PI3K 562 signalling which would be coherent with no involvement of PIK3CA mutations in pericytes in 563 the progression of the CCM147. 564 Intriguingly, sporadic CCMs are frequently found in close proximity to developmental 565 venous anomalies (DVAs)148,149. DVAs are the most common vascular malformations 566 (present in about 10% of the adult population) and are largely develop before the age of 20. 567 Emerging data suggest that sporadic CCMs may, at least some, derive from DVAs. While 568 comparing the mutational status of DVA and its paired CCM (present in the same patient), 569 it was identified that DVA and CCM carried both a somatic activating PIK3CA mutation, while 570 CCM lesions harboured mutations only in MAP3K3. Based on these finding, it has been 571 proposed that individuals who have a PIK3CA-related DVAs are predisposed to develop 572 sporadic CCM in close proximity to the DVA144. Collectively, these studies have catapulted 573 PIK3CA as a critical genetic hit for the CCM disease. Yet, it is not clear whether PIK3CA 574 mutations are required for the formation of CCMs, or they serve as endothelial clonal 575 amplifiers. 576 577 Capillary malformations (CMs): the least pathogenic vascular malformations within 578 PROS. Low-flow CMs are anomalies composed of dilated capillaries near the surface of the 579 skin which normally present a macular, pink to red stain. PIK3CA-related CMs have been 580 largely described as part of some PROS, such as MCAP and DCMO, where they are largely 581 caused by activating non-hotspot mutations in PIK3CA53,93. Also, a recent case with an 582 acquired capillary malformation (after birth) associated with V344M PIK3CA variant has 583 been reported150. Overall, these lesions tend to be largely cosmetic; thereby indicating that 584 they are less severe than LMs or VMs. Mutations in GNAQ and GNA11 (encoding for 585 Gαq and Gα11 respectively) are overrepresented in cutaneous CMs. Intriguing, co- 586 occurrence of PIK3CA and GNAQ has also been reported in combined vascular 587 malformations11. Whether this co-existence relates to higher severity is not clear yet. 588 589 Treatment options for PROS: the era of repurposing drugs used in oncology. 590 Yet, there is no approved molecular treatment for PROS patients. The broad clinical 591 manifestations in PROS together with the fact that they pertain to the category of rare 592 diseases have compromised the implementation of effective and safety targeted therapies 593 for their treatment. Until very recently, the standard care was surgical debulking (including 594 amputation, lesion debulking/resection among others) and/or scleroembolization of vascular 595 malformations. However, these treatments are not curative and have high risk of recurrence 596 (hypertrophy). In line with this, treatments to diminish symptoms such as pain (steroids or 597 antihistamines) or seizures (epilepsy medication) have been prescribed in some cases. With 598 the discovery of PIK3CA mutations being causative of PROS, treatment possibilities 599 emerged. Initial efforts have been centered on the repurpose of PI3K inhibitors used in 600 oncology. Below we summarized most promising attempts (Box1). Of note, with the emerge 601 of pharmacotherapy for PROS, the efficacy and the best regimen is also being established. 602 We refer the reader to Table 1 for specific details on completed and ongoing clinical trials 603 for PROS. 604 605 Sirolimus: the first targeted therapy to treat PROS. Sirolimus (also known as rapamycin) 606 is an allosteric mTOR inhibitor approved by both the food and drug administration (FDA) and 607 European medicine agency (EMA). Hence, it is no surprising that siroliums was first 608 proposed for the treatment of PROS. The very first clinical trial was planned in patients with 609 vascular anomalies, even prior to the implementation of the genetic diagnosis as a 610 precondition for trial inclusion (NCT00975819). Together with other follow up clinical trials, 611 it showed that sirolimus reduced the overall volume of vascular malformations with LMs 612 exhibiting the highest susceptibility to the drug151–154. However, several adverse events 613 (AEs) and lesion regrowth upon treatment withdrawal were also identified on those studies 614 . The PROMISE study, which tested a lower dose of sirolimus, arose as the first clinical 153 615 trial specifically for syndromic PROS patients 45. In that context, sirolimus showed a modest 616 reduction in the volume of the overgrown tissues which was also accompanied with frequent 617 AEs. Although these results were not as good as the initial expectations, it is fair to 618 acknowledge that many patients have benefited from sirolimus. This explains why current 619 standard of care includes sirolimus (with and without surgery) for PROS patients. However, 620 should an individual risk-benefit evaluation by the physician be considered45. Currently, 621 there are several ongoing clinical trials which are assessing the potential benefits of 622 sirolimus for PROS (NCT04598204, NCT03987152, NCT04128722, NCT03972592, 623 NCT02638389, NCT03767660). Topic administration on superficial lesions is being also 624 considered (NCT03972592), given that reduced side-target effects are expected. However, 625 it is important to consider that topic administration is not an option for syndromic PROS with 626 internal lesions. 627 628 Miransertib (currently known as MK-7075): an AKT inhibitor for the treatment of 629 PROS. Miransertib is an allosteric highly selective AKT inhibitor which was initially 630 developed for oncology 155. Given that AKT is the main effector of PI3Kα signaling pathway, 631 repurposing of miransertib was propose not only for Proteus Syndrome (PS, caused by gain 632 of function mutations in AKT1, Box 1) but also for PROS. This was first used in a 633 compassionate basis with partial therapeutic efficacy and no major toxicities156,157. An 634 important aspect to bear in mind about miransertib is that it seems most efficient in severe 635 PROS. Preclinical studies in isolated vascular malformations have also showed promising 636 data, even when using half of the dose used in oncology71. Clinical trials testing the safety 637 and effectiveness in PS and PROS patients with a confirmed genetic diagnosis will remain 638 open until June 2022 (NCT04316546, NCT03094832, NCT03317366). 639 640 The use of PI3K inhibitors for PROS: pan vs. PI3Kα selective targeting. PROS are 641 caused by activating PIK3CA mutations, hence selective inhibition of PI3Kα shall be the 642 most accurate targeted treatment. Nevertheless, a clinical trial using taselisib, a selective 643 pan class I PI3K (PI3Kα, β δ, γ) inhibitor was first approached (NCT03290092)158. PROS 644 patients treated with taselisib showed clinical improvement with reduced pain, chronic 645 bleeding resolution and functional improvement. However, presentation of severe drug- 646 related AEs in some patients led to the early termination of the trial. These adverse effects 647 were thought to be caused by the impact of inhibition of PI3Kδ/γ on the immune system; 648 thereby suggesting that selective inhibition of the PI3Kα isoform would be a better choice in 649 that context. Since then, expectations have been put in alpelisib, a PI3Kα selective inhibitor 650 which has been recently approved by the FDA for metastatic breast cancer159. Alpelisib was 651 first tested in mouse models of CLOVES-like syndrome showing promising results with 652 higher efficacy than sirolimus94. Similar results have been recently obtained in a mouse 653 model of LMs108. However, withdrawal of the treatment led to the recurrence of the lesions, 654 indicating that a chronic treatment should be considered for these patients94. The exciting 655 observations that topic administration of alpelisib in xerograph mouse models induced a 656 prominent regression in vascular skin lesions57, enhances the prospects for topical 657 administration of this inhibitor in humans. 658 An unregistered case series of patients with PROS with confirmed PIK3CA mutations 659 treated with alpelisib on a compassionate basis, reported evidence of efficacy with minor 660 side effects94. However, neither safety nor efficacy endpoints were pre-specified in that 661 study94. Later, other studies also confirmed promising results while using of alpelisb in a 662 compassionate basis160–162. Nevertheless, the lack of evidence of benefit-risk assessments 663 in a long-term basis pushed a retrospective and non-interventional study of 32 patients 664 treated with this inhibitor (EPIK-P1). Current reported analysis on the first endpoint (after 24 665 weeks of treatment) exposed that 37.5% of alpelisib-treated patients exhibit clinical benefits 666 and 38.6% of the patients suffer of hyperglycemia, aphthous ulcer and stomatitis163. EPIK- 667 P3 is now continuing EPIK-P1 to evaluate the long-term safety and efficiency 668 (NCT04980833; 2020-005896-12). Alpelisib treatment has also been tested in patients with 669 LMs who showed a general improvement with decrease volume of LMs and reduced 670 tiredness and pain while reporting moderate AEs such as aphthous and diarrhea108. The 671 promising data on alpelisib for compassionate has finally catalyzed an ongoing prospective, 672 multicenter, randomized, double-blind and, with placebo-controlled period clinical trial 673 (EPIK-P2) to demonstrate the efficacy, tolerability, and safety of alpelisib (NCT04589650). 674 While we are still learning from all the pharmacotherapy studies described above, it 675 is now clear that these drugs have become essential tools to treat PROS. Collectively, they 676 have shown a clear impact on the quality of life of these patients while different penetrance 677 in the reduction of volume lesion and symptoms were observed. It is important to consider 678 that PROS patients exhibit high heterogeneity in their clinical manifestations (from isolated 679 vascular malformations like VM or LMs to complex and syndromic cases such as CLOVES 680 or KTS among others) which quite likely explains the heterotypic responses to the different 681 treatments. Thus, once safety of these treatments is well established, physicians will have 682 to evaluate the risk-benefits of each patient based on their specific needs and responses. 683 We believe preclinical studies are being instrumental to define dosing regiments (e.g. high 684 vs low dose; intermittent vs continuous) as well as to test combination therapies. 685 686 Conclusions and perspectives 687 With the advent of NGS PIK3CA has emerged as frequently mutated oncogene in congenital 688 disorders with asymmetric overgrowth. This has accelerated the generation of mouse 689 models of these diseases towards the understanding of the mechanisms which lead to their 690 onset and progression. Lessons learnt suggest that the pathogenic score of activating 691 PIK3CA mutations in congenital disorders is a combination of the developmental time 692 (when) that these mutations appear together with intrinsic factors (cell-autonomous), such 693 as the mechanisms and degree of activation of PI3Kα (which) and the cell lineage specificity 694 of the mutated cell (where) and extrinsic factors (non-cell autonomous), which refers to 695 environmental modifiers of the PI3Kα outcome. Unraveling the specific contribution of each 696 of these elements will be required to define better treatments for these patients. In line with 697 this, mapping the entire PIK3CA gene should be considered when genotyping patients with 698 a suspicion of PROS. 699 With regard to the clear skewing tissue pattern in PROS, it has become clear that 700 some lineages are more sensitive to PI3K overactivation. Yet, it has to be learned the basis 701 of such context-dependent pathogenesis. We propose that scRNAseq approaches in 702 combination with genetic mouse models which allow to express Pik3ca mutations in specific 703 subpopulations will be critical to identify cell lineage histories of mutant clones and to 704 understand how PI3Kα activation subverts cellular processes in a context-dependent 705 manner. Second or even triple genetic hits in vascular malformations are emerging in 706 aggressive clinical cases. This overrules the original idea that congenital vascular 707 malformations are mostly monogenic disorders that behave differently from cancer. This 708 highlights the urge to look for multiple genetic hits in other PIK3CA-related disorders that 709 may explain the severity or tissue-specific pathogenesis. Finally, it is of vital importance to 710 consider non-cell autonomous mechanisms for future treatments. Response to co-inhibition 711 of specific upstream or downstream signalling pathways may be also different among the 712 PIK3CA variants. Targeted studies of molecular programs underlying distinct types of 713 PIK3CA-related overgrowth are needed to inform precision medicine strategies. 714 715 716 BOX1. Non-PIK3CA-related congenital disorders caused by aberrant PI3K signalling. 717 Mutations in several components of the PI3K pathway other than PIK3CA have been 718 identified in congenital disorders characterized by aberrant activation of PI3K signalling and 719 tissue overgrowth164. This includes loss of PTEN165 and TSC1 and TSC2166,167 and germline 720 or somatic activating mutations in AKT1168, AKT2169, AKT342,170,171, PIK3R1172, PIK3R242 721 and mTOR79,173–176. Heterozygous loss of PTEN and TSC1/TSC2 and activating mutations 722 in AKT3, PIK3R2 and mTOR may occur in the germline or in a somatic fashion. The germline 723 (heterozygous) loss of PTEN is known as PTEN hamartoma tumour syndrome (PHTS) and 724 includes Bannayan-Riley-Ruvalcaba syndrome (BRRS), Cowden syndrome (CS) PTEN- 725 related Proteus syndrome (PS), and PTEN-related Proteus-like syndrome165. While PHTS 726 patients are prone to develop benign and malignant tumours, evidence has emerged that 727 50% of them also develop vascular malformations. Autosomal dominant loss of TSC1/TSC2 728 leads to Tuberous sclerosis complex (TSC), a neurocutaneous disorder frequently 729 associated with abnormalities in the brain177. Somatic activating mutations in AKT1 lead to 730 the so-called Porteous syndrome (PS)168 and in PIK3R1 have been associated with vascular 731 malformations and overgrowth similar to PROS/PS42,170,172. Mutations in AKT3 (highly 732 expressed in brain and heart) and in PIK3R2 have been associated with brain overgrowth. 733 Only very few cases with AKT2 mutations have been found and they have been associated 734 with severe fasting hypoglycemia and asymmetrical growth169. Although there are 735 overlapping characteristics between these genetic conditions, the specific expression and 736 regulation of each of the member define the ultimate clinical outcome178. The correct genetic 737 diagnosis is relevant to find proper treatments as well as to better understand the gene- 738 specific functions during development. It is important to bear in mind that overactivation of 739 the PI3K signalling pathway may not be only promoted by the PI3Kα isoform. Figure BOX1 740 integrated in this box. 741 742 BOX2. Class II and Class III PI3Ks 743 Class II and class III PI3K still remain quite enigmatic. Class II PI3K is composed of three 744 catalytic isoforms PI3KC2α, PI3KC2β and PI3KC2γ (encoded by PIK3C2A, PIK3C2B and 745 PIK3C2G respectively) that lack dedicated regulatory subunits. These enzymes can 746 generate PI(3)P and PI(3,4)P2 and regulate vesicular trafficking, including receptor 747 endocytosis, endosomal trafficking, neurosecretory granule release and insulin secretion. 748 PI3KC2α and PI3KC2β are broadly expressed while PI3K-C2γ expression is limited to some 749 tissues such as liver, pancreas, prostate, and breast. Due to the lack of a specific regulatory 750 subunit, the mechanisms through which class II PI3Ks are recruited to specific sites and 751 activated are still unclear14,179,180. Class III PI3K is composed by a solely catalytic subunit 752 Vps34 (encoded by PIK3C3) associated with the regulatory protein Vps15 (encoded by 753 PIK3R4) which controls Vps34 localization and activation. Vps34 catalyzes the 754 phosphorylation of phosphatidylinositol (PtdIns) to phosphatidylinositol-3-phosphate 755 (PtdIns3P) and, principally, it is involved in vesicle trafficking and autophagy controlling 756 nutrient acquisition pathways14,181. 757 758 BOX3. PI3Kα in the endothelium 759 PI3Kα has emerged as master regulator of vascular morphogenesis in blood and lymphatic 760 vessels182. Genetically engineering mouse models have shown that both inactivation and 761 overactivation of PI3Kα activity in the germline or specifically in BECs cause profound 762 defects in the blood vasculature ultimately resulting in embryonic lethality57,62,63,183–185. This 763 is explained by PI3Kα being the main producer of PIP3 upon RTK stimulation in ECs. In line 764 with this, inactivation or deletion of other class I PI3K isoforms does not interfere with 765 vascular development nor with embryonic development183. Lymphatic endothelial-specific 766 depletion of PI3Kα results in perinatal lethally due to selective alterations in mesenteric and 767 intestinal lymphatic vessels186. Stanczuk et al. also showed that LECs differently rely on 768 VEGFR3-PI3Kα signaling pathway depending on their origin and tissue location. An 769 intriguing aspect of PI3Kα is that it is activated through distinct mechanism in the BECs and 770 LECs. BECs utilise the regulatory subunits to recruit PI3Kα to the plasma membrane upon 771 RTK stimulation. Instead, the LECs also require the RBD domain of PI3Kα intact. Hence, 772 mice carrying mutations in Pik3ca that block PI3Kα binding to RAS exhibit selective defects 773 lymphatic vessel development187. PI3Kα regulates vessel growth through both AKT 774 dependent and independent mechanisms188,189, including (1) control of cell cycle 775 progression through the inactivation of FOXO1188,190; (2) stimulation of vein specification by 776 increasing COUP-TFII levels upon TIE2 activation191; (3) regulation of junctional remodelling 777 and cell rearrangements through the control of NUAK1/MYPT1/MLCP122. Endothelial- 778 specific loss of Pten in mice also result in aberrant vascular development and early 779 embryonic failure. Mechanistically, PTEN fine tunes endothelial cell proliferation during the 780 early steps of the angiogenic process192,193. Of note, loss of Pten metabolically rewires 781 endothelial cells towards lipid consumption194. 782 783 Acknowledgements 784 We would like thank Sandra D Castillo and Taija Mäkinen for their valuable feedback. 785 Figures were created with BioRender.com. 786 787 Funding 788 M.G. laboratory is supported by the research grant from the Spanish Ministry of Science and 789 Innovation (PID2020-116184RB-110); PTEN RESEARCH Foundation (BRR-17-001); and 790 La Caixa Foundation (HR18-00120); A. A-U and M.G are recipients of a CLOVES syndrome 791 community research grant. 792 793 References 794 1. Fox, M., Mott, H. R. & Owen, D. Class IA PI3K regulatory subunits: p110- 795 independent roles and structures. Biochem. Soc. Trans. 48, 1397–1417 (2020). 796 2. Vanhaesebroeck, B., Stephens, L. & Hawkins, P. PI3K signalling: the path to 797 discovery and understanding. Nat. Rev. Mol. Cell Biol. 13, 195–203 (2012). 798 3. Fruman, D. A. et al. Leading Edge The PI3K Pathway in Human Disease. Cell 170, 799 605–635 (2017). 800 4. Geering, B., Cutillas, P. R., Nock, G., Gharbi, S. I. & Vanhaesebroeck, B. Class IA 801 phosphoinositide 3-kinases are obligate p85-p110 heterodimers. Proc. Natl. Acad. 802 Sci. U. S. A. 104, 7809–7814 (2007). 803 5. Tate, J. G. et al. COSMIC: the Catalogue Of Somatic Mutations In Cancer. Nucleic 804 Acids Res. 47, D941–D947 (2019). 805 6. Millis, S. Z., Ikeda, S., Reddy, S., Gatalica, Z. & Kurzrock, R. Landscape of 806 Phosphatidylinositol-3-Kinase Pathway Alterations Across 19 784 Diverse Solid 807 Tumors. JAMA Oncol. 2, 1565–1573 (2016). 808 7. Vasan, N. et al. Double PIK3CA mutations in cis increase oncogenicity and 809 sensitivity to PI3Ka inhibitors. Science (80-. ). 366, 714–723 (2019). 810 8. Madsen, R. R., Vanhaesebroeck, B. & Semple, R. K. Cancer-Associated PIK3CA 811 Mutations in Overgrowth Disorders. Trends in Molecular Medicine vol. 24 856–870 812 (2018). 813 9. Peyre, M. et al. Somatic PIK3CA Mutations in Sporadic Cerebral Cavernous 814 Malformations . N. Engl. J. Med. 385, 996–1004 (2021). 815 10. Ren, A. A. et al. PIK3CA and CCM mutations fuel cavernomas through a cancer-like 816 mechanism. Nat. 2021 5947862 594, 271–276 (2021). 817 11. Ten Broek, R. W. et al. Comprehensive molecular and clinicopathological analysis 818 of vascular malformations: A study of 319 cases. Genes Chromosom. Cancer 58, 819 541–550 (2019). 820 12. Siegel, D. H. et al. Analyzing the Genetic Spectrum of Vascular Anomalies with 821 Overgrowth via Cancer Genomics. J. Invest. Dermatol. 138, 957–967 (2018). 822 13. Castel, P., Rauen, K. A. & McCormick, F. The duality of human oncoproteins: 823 drivers of cancer and congenital disorders. Nature Reviews Cancer vol. 20 383–397 824 (2020). 825 14. Bilanges, B., Posor, Y. & Vanhaesebroeck, B. PI3K isoforms in cell signalling 826 and vesicle trafficking. Nat. Rev. Mol. Cell Biol. 2019 209 20, 515–534 (2019). 827 15. Vanhaesebroeck, B., Guillermet-Guibert, J., Graupera, M. & Bilanges, B. The 828 emerging mechanisms of isoform-specific PI3K signalling. Nat. Rev. Mol. Cell Biol. 829 2010 115 11, 329–341 (2010). 830 16. Thorpe, L. M., Yuzugullu, H. & Zhao, J. J. PI3K in cancer: divergent roles of 831 isoforms, modes of activation and therapeutic targeting. Nat. Rev. Cancer 2015 151 832 15, 7–24 (2014). 833 17. Tsolakos, N. et al. Quantitation of class IA PI3Ks in mice reveals p110-free-p85s 834 and isoform-selective subunit associations and recruitment to receptors. Proc. Natl. 835 Acad. Sci. U. S. A. 115, 12176–12181 (2018). 836 18. Burke, J. E. et al. Dynamics of the phosphoinositide 3-kinase p110δ interaction with 837 p85α and membranes reveals aspects of regulation distinct from p110α. Structure 838 19, 1127–1137 (2011). 839 19. Zhang, X. et al. Structure of lipid kinase p110β/p85β elucidates an unusual SH2- 840 domain-mediated inhibitory mechanism. Mol. Cell 41, 567–578 (2011). 841 20. Miller, M. S. et al. Structural basis of nSH2 regulation and lipid binding in PI3Kα. 842 Oncotarget 5, 5198–5208 (2014). 843 21. Burke, J. E. Dynamic structural biology at the protein membrane interface. J. Biol. 844 Chem. 294, 3872–3880 (2019). 845 22. Nürnberg, B. & Beer-Hammer, S. Function, Regulation and Biological Roles of 846 PI3Kγ Variants. Biomol. 2019, Vol. 9, Page 427 9, 427 (2019). 847 23. Stephens, L. R. et al. The G beta gamma sensitivity of a PI3K is dependent upon a 848 tightly associated adaptor, p101. Cell 89, 105–114 (1997). 849 24. Rathinaswamy, M. K. et al. Structure of the phosphoinositide 3-kinase (PI3K) 850 p110γ-p101 complex reveals molecular mechanism of GPCR activation. Sci. Adv. 7, 851 (2021). 852 25. Kurosu, H. et al. Heterodimeric Phosphoinositide 3-Kinase Consisting of p85 and 853 p110β Is Synergistically Activated by the βγ Subunits of G Proteins and 854 Phosphotyrosyl Peptide. J. Biol. Chem. 272, 24252–24256 (1997). 855 26. Murga, C., Fukuhara, S. & Gutkind, J. S. A Novel Role for Phosphatidylinositol 3- 856 Kinase β in Signaling from G Protein-coupled Receptors to Akt. J. Biol. Chem. 275, 857 12069–12073 (2000). 858 27. Bresnick, A. R. & Backer, J. M. PI3Kβ-A Versatile Transducer for GPCR, RTK, and 859 Small GTPase Signaling. Endocrinology 160, 536–555 (2019). 860 28. Kurig, B. et al. Ras is an indispensable coregulator of the class IB phosphoinositide 861 3-kinase p87/p110γ. Proc. Natl. Acad. Sci. U. S. A. 106, 20312–20317 (2009). 862 29. Rodriguez-Viciana, P., Warne, P. H., Vanhaesebroeck, B., Waterfield, M. D. & 863 Downward, J. Activation of phosphoinositide 3-kinase by interaction with Ras and by 864 point mutation. EMBO J. 15, 2442–2451 (1996). 865 30. Fritsch, R. et al. XRAS and RHO families of GTPases directly regulate distinct 866 phosphoinositide 3-kinase isoforms. Cell 153, 1050 (2013). 867 31. Whitman, M., Downes, C. P., Keeler, M., Keller, T. & Cantley, L. Type I 868 phosphatidylinositol kinase makes a novel inositol phospholipid, 869 phosphatidylinositol-3-phosphate. Nat. 1988 3326165 332, 644–646 (1988). 870 32. Madsen, R. R. & Vanhaesebroeck, B. Cracking the context-specific PI3K signaling 871 code. Science Signaling vol. 13 2940 (2020). 872 33. James, S. R. et al. Specific binding of the Akt-1 protein kinase to 873 phosphatidylinositol 3,4,5-trisphosphate without subsequent activation. Biochem. J. 874 315, 709–713 (1996). 875 34. Singh, N. et al. Redefining the specificity of phosphoinositide-binding by human PH 876 domain-containing proteins. Nat. Commun. 2021 121 12, 1–13 (2021). 877 35. Lien, E. C., Dibble, C. C. & Toker, A. PI3K signaling in cancer: beyond AKT. Curr. 878 Opin. Cell Biol. 45, 62–71 (2017). 879 36. Liu, S. L. et al. Quantitative Lipid Imaging Reveals a New Signaling Function of 880 Phosphatidylinositol-3,4-Bisphophate: Isoform- and Site-Specific Activation of Akt. 881 Mol. Cell 71, 1092-1104.e5 (2018). 882 37. Lee, Y. R., Chen, M. & Pandolfi, P. P. The functions and regulation of the PTEN 883 tumour suppressor: new modes and prospects. Nat. Rev. Mol. Cell Biol. 2018 199 884 19, 547–562 (2018). 885 38. Sugiyama, M. G., Fairn, G. D. & Antonescu, C. N. Akt-ing up just about everywhere: 886 Compartment-specific Akt activation and function in receptor tyrosine kinase 887 signaling. Front. Cell Dev. Biol. 7, 70 (2019). 888 39. Manning, B. D. & Toker, A. AKT/PKB Signaling: Navigating the Network. Cell 169, 889 381–405 (2017). 890 40. Liu, G. Y. & Sabatini, D. M. mTOR at the nexus of nutrition, growth, ageing and 891 disease. Nat. Rev. Mol. Cell Biol. 2020 214 21, 183–203 (2020). 892 41. Webb, A. E. & Brunet, A. FOXO transcription factors: key regulators of cellular 893 quality control. Trends Biochem. Sci. 39, 159–169 (2014). 894 42. Rivière, J. B. et al. De novo germline and postzygotic mutations in AKT3, PIK3R2 895 and PIK3CA cause a spectrum of related megalencephaly syndromes. Nat. Genet. 896 44, 934–940 (2012). 897 43. Lindhurst, M. J. et al. Mosaic overgrowth with fibroadipose hyperplasia is caused by 898 somatic activating mutations in PIK3CA. Nat. Genet. 44, 928–933 (2012). 899 44. Kurek, K. C. et al. Somatic mosaic activating mutations in PIK3CA cause CLOVES 900 syndrome. Am. J. Hum. Genet. 90, 1108–1115 (2012). 901 45. Parker, V. E. R. et al. Safety and efficacy of low-dose sirolimus in the PIK3CA- 902 related overgrowth spectrum. Genet. Med. 21, 1189–1198 (2019). 903 46. Rodriguez-Laguna, L. et al. CLAPO syndrome: identification of somatic activating 904 PIK3CA mutations and delineation of the natural history and phenotype. Genet. 905 Med. 20, 882–889 (2018). 906 47. Vahidnezhad, H., Youssefian, L. & Uitto, J. Klippel-Trenaunay syndrome belongs to 907 the PIK3CA-related overgrowth spectrum (PROS). Exp. Dermatol. 25, 17–19 908 (2016). 909 48. Jansen, L. A. et al. PI3K/AKT pathway mutations cause a spectrum of brain 910 malformations from megalencephaly to focal cortical dysplasia. Brain 138, 1613– 911 1628 (2015). 912 49. Maclellan, R. A. et al. PIK3CA activating mutations in facial infiltrating lipomatosis. 913 Plast. Reconstr. Surg. 133, (2014). 914 50. Keppler-Noreuil, K. M. et al. Clinical delineation and natural history of the PIK3CA- 915 related overgrowth spectrum. Am. J. Med. Genet. Part A 164, 1713–1733 (2014). 916 51. Rios, J. J. et al. Somatic gain-of-function mutations in PIK3CA in patients with 917 macrodactyly. Hum. Mol. Genet. 22, 444–451 (2013). 918 52. Frisk, S. et al. Early activating somatic PIK3CA mutations promote ectopic muscle 919 development and upper limb overgrowth. Clin. Genet. 96, 118–125 (2019). 920 53. Goss, J. A. et al. Diffuse capillary malformation with overgrowth contains somatic 921 PIK3CA variants. Clin. Genet. 97, 736–740 (2020). 922 54. Blackburn, P. R. et al. PIK3CA mutations in lipomatosis of nerve with or without 923 nerve territory overgrowth. Mod. Pathol. 2019 333 33, 420–430 (2019). 924 55. Luks, V. L. et al. Lymphatic and other vascular malformative/overgrowth disorders 925 are caused by somatic mutations in PIK3CA. J. Pediatr. 166, 1048-1054.e5 (2015). 926 56. Limaye, N. et al. Somatic Activating PIK3CA Mutations Cause Venous 927 Malformation. Am. J. Hum. Genet. 97, 914–921 (2015). 928 57. Castel, P. et al. Somatic PIK3CA mutations as a driver of sporadic venous 929 malformations. Sci. Transl. Med. 8, 332ra42-332ra42 (2016). 930 58. Castillo, S. D. et al. Somatic activating mutations in Pik3ca cause sporadic venous 931 malformations in mice and humans. Sci. Transl. Med. 8, (2016). 932 59. Mirzaa, G., Conway, R., Graham, J. M. & Dobyns, W. B. PIK3CA-Related 933 Overgrowth Spectrum. GeneReviews® (2021). 934 60. Martinez‐Glez, V. et al. Segmental undergrowth is associated with pathogenic 935 variants in vascular malformation genes: A retrospective case‐series study. Clin. 936 Genet. (2021) doi:10.1111/cge.14095. 937 61. De Graer, C. et al. Novel features of PIK3CA-Related Overgrowth Spectrum: 938 Lesson from an aborted fetus presenting a de novo constitutional PIK3CA mutation. 939 Eur. J. Med. Genet. 63, 103775 (2020). 940 62. Berenjeno, I. M. et al. Oncogenic PIK3CA induces centrosome amplification and 941 tolerance to genome doubling. Nat. Commun. 2017 81 8, 1–15 (2017). 942 63. Hare, L. M. et al. Heterozygous expression of the oncogenic Pik3caH1047R 943 mutation during murine development results in fatal embryonic and extraembryonic 944 defects. Dev. Biol. 404, 14–26 (2015). 945 64. Happle, R. Lethal genes surviving by mosaicism: A possible explanation for 946 sporadic birth defects involving the skin. J. Am. Acad. Dermatol. 16, 899–906 947 (1987). 948 65. Orloff, M. S. et al. Germline PIK3CA and AKT1 mutations in cowden and cowden- 949 like syndromes. Am. J. Hum. Genet. 92, 76–80 (2013). 950 66. Yeung, K. S. et al. Identification of mutations in the PI3K-AKT-mTOR signalling 951 pathway in patients with macrocephaly and developmental delay and/or autism. 952 Mol. Autism 8, 1–11 (2017). 953 67. Zollino, M. et al. Germline pathogenic variant in PIK3CA leading to symmetrical 954 overgrowth with marked macrocephaly and mild global developmental delay. Mol. 955 Genet. Genomic Med. 7, e845 (2019). 956 68. Mussa, A. et al. Genotypes and phenotypes heterogeneity in PIK3CA-related 957 overgrowth spectrum and overlapping conditions: 150 novel patients and systematic 958 review of 1007 patients with PIK3CA pathogenetic variants. J. Med. Genet. 0, 959 jmedgenet-2021-108093 (2022). 960 69. Brouillard, P. et al. Non-hotspot PIK3CA mutations are more frequent in CLOVES 961 than in common or combined lymphatic malformations. Orphanet J. Rare Dis. 16, 1– 962 12 (2021). 963 70. Hafner, C. et al. Oncogenic PIK3CA mutations occur in epidermal nevi and 964 seborrheic keratoses with a characteristic mutation pattern. Proc. Natl. Acad. Sci. 965 104, 13450–13454 (2007). 966 71. Kobialka, P. et al. Low dose AKT inhibitor miransertib cures PI3K-related vascular 967 malformations in preclinical models of human disease. bioRxiv (2021) 968 doi:10.1101/2021.07.16.452617. 969 72. Boscolo, E. et al. AKT hyper-phosphorylation associated with PI3K mutations in 970 lymphatic endothelial cells from a patient with lymphatic malformation. Angiogenesis 971 18, 151–162 (2015). 972 73. Le Cras, T. D. et al. Constitutively active PIK3CA mutations are expressed by 973 lymphatic and vascular endothelial cells in capillary lymphatic venous malformation. 974 Angiogenesis 23, 425–442 (2020). 975 74. Kuentz, P. et al. Molecular diagnosis of PIK3CA-related overgrowth spectrum 976 (PROS) in 162 patients and recommendations for genetic testing. Genet. Med. 19, 977 989–997 (2017). 978 75. Kłaniewska, M. et al. CLOVES syndrome caused by mosaic mutation in the PIK3CA 979 gene identified in fibroblasts. Pediatr. Pol. 96, 148–152 (2021). 980 76. Ranieri, C. et al. In vitro efficacy of ARQ 092, an allosteric AKT inhibitor, on primary 981 fibroblast cells derived from patients with PIK3CA-related overgrowth spectrum 982 (PROS). Neurogenetics 19, 77–91 (2018). 983 77. Sun, B. et al. Activating PIK3CA mutation promotes adipogenesis of adipose- 984 derived stem cells in macrodactyly via up-regulation of E2F1. Cell Death Dis. 2020 985 117 11, 1–11 (2020). 986 78. Couto, J. A. et al. Somatic PIK3CA mutations are present in multiple tissues of facial 987 infiltrating lipomatosis. Pediatr. Res. 2017 825 82, 850–854 (2017). 988 79. Lee, J. H. et al. De novo somatic mutations in components of the PI3K-AKT3-mTOR 989 pathway cause hemimegalencephaly. Nat. Genet. 44, 941–945 (2012). 990 80. Wu, J. et al. An investigation of PIK3CA mutations in isolated An investigation of 991 PIK3CA mutations in isolated macrodactylymacrodactyly. J. Hand Surg. Eur. Vol. 992 43, 756–760 (2018). 993 81. Ebner, B. A. et al. Brachial plexus lipomatosis with perineurial pseudoonion bulb 994 formation: Result of a mosaic PIK3CA mutation in the para‐axial mesoderm state. 995 Brain Pathol. (2022) doi:10.1111/bpa.13057. 996 82. Ercan, C., J. van Diest, P. & Vooijs, M. Mammary Development and Breast Cancer: 997 The Role of Stem Cells. Curr. Mol. Med. 11, 270–285 (2011). 998 83. Spina, E. & Cowin, P. Embryonic mammary gland development. Semin. Cell Dev. 999 Biol. 114, 83–92 (2021). 1000 84. Van Keymeulen, A. et al. Reactivation of multipotency by oncogenic PIK3CA 1001 induces breast tumour heterogeneity. Nature 525, 119–123 (2015). 1002 85. Madsen, R. R. et al. Oncogenic PIK3CA promotes cellular stemness in an allele 1003 dose-dependent manner. Proc. Natl. Acad. Sci. U. S. A. 116, 8380–8389 (2019). 1004 86. Madsen, R. R. et al. NODAL/TGFβ signalling mediates the self-sustained stemness 1005 induced by PIK3CA H1047R homozygosity in pluripotent stem cells. Dis. Model. 1006 Mech. 14, (2021). 1007 87. Herms, A. et al. Levelling out differences in aerobic glycolysis neutralizes the 1008 competitive advantage of oncogenic <em>PIK3CA</em> mutant 1009 progenitors in the esophagus. bioRxiv 2021.05.28.446104 (2021) 1010 doi:10.1101/2021.05.28.446104. 1011 88. Martincorena, I. & Campbell, P. J. Somatic mutation in cancer and normal cells. 1012 Science (80-. ). 349, 1483–1489 (2015). 1013 89. Usui, H. et al. A novel method for isolating lymphatic endothelial cells from 1014 lymphatic malformations and detecting PIK3CA somatic mutation in these isolated 1015 cells. Surg. Today 2020 513 51, 439–446 (2020). 1016 90. Zenner, K. et al. Genotype correlates with clinical severity in PIK3CA-Associated 1017 lymphatic malformations. JCI Insight 4, (2019). 1018 91. Mirzaa, G. et al. PIK3CA-associated developmental disorders exhibit distinct 1019 classes of mutations with variable expression and tissue distribution. JCI Insight 1, 1020 (2016). 1021 92. McDermott, J. H. et al. Hypoglycaemia represents a clinically significant 1022 manifestation of PIK3CA- and CCND2-associated segmental overgrowth. Clin. 1023 Genet. 93, 687–692 (2018). 1024 93. Park, H. J. et al. Detailed analysis of phenotypes and genotypes in 1025 megalencephaly-capillary malformation-polymicrogyria syndrome caused by 1026 somatic mosaicism of PIK3CA mutations. Orphanet J. Rare Dis. 15, 1–10 (2020). 1027 94. Venot, Q. et al. Targeted therapy in patients with PIK3CA-related overgrowth 1028 syndrome. Nature 558, 540–546 (2018). 1029 95. Döcker, D. et al. Germline PTPN11 and somatic PIK3CA variant in a boy with 1030 megalencephaly-capillary malformation syndrome (MCAP) - pure coincidence? Eur. 1031 J. Hum. Genet. 2015 233 23, 409–412 (2014). 1032 96. Burke, J. E., Perisic, O., Masson, G. R., Vadas, O. & Williams, R. L. Oncogenic 1033 mutations mimic and enhance dynamic events in the natural activation of 1034 phosphoinositide 3-kinase p110α (PIK3CA). Proc. Natl. Acad. Sci. U. S. A. 109, 1035 15259–15264 (2012). 1036 97. Dornan, G. L. & Burke, J. E. Molecular mechanisms of human disease mediated by 1037 oncogenic and primary immunodeficiency mutations in class IA phosphoinositide 3- 1038 kinases. Front. Immunol. 9, 575 (2018). 1039 98. Zhao, L. & Vogt, P. K. Helical domain and kinase domain mutations in p110α of 1040 phosphatidylinositol 3-kinase induce gain of function by different mechanisms. Proc. 1041 Natl. Acad. Sci. 105, 2652–2657 (2008). 1042 99. Chaussade, C., Cho, K., Mawson, C., Rewcastle, G. W. & Shepherd, P. R. 1043 Functional differences between two classes of oncogenic mutation in the PIK3CA 1044 gene. Biochem. Biophys. Res. Commun. 381, 577–581 (2009). 1045 100. Miled, N. et al. Mechanism of two classes of cancer mutations in the 1046 phosphoinositide 3-kinase catalytic subunit. Science (80-. ). 317, 239–242 (2007). 1047 101. Gkeka, P. et al. Investigating the Structure and Dynamics of the PIK3CA Wild-Type 1048 and H1047R Oncogenic Mutant. PLOS Comput. Biol. 10, e1003895 (2014). 1049 102. Zhao, L. & Vogt, P. K. Hot-spot mutations in p110α of phosphatidylinositol 3-kinase 1050 (PI3K): Differential interactions with the regulatory subunit p85 and with RAS. 1051 http://dx.doi.org.sire.ub.edu/10.4161/cc.9.3.10599 9, 596–600 (2010). 1052 103. Hao, Y. et al. Gain of Interaction with IRS1 by p110α-Helical Domain Mutants Is 1053 Crucial for Their Oncogenic Functions. Cancer Cell 23, 583–593 (2013). 1054 104. Gymnopoulos, M., Elsliger, M. A. & Vogt, P. K. Rare cancer-specific mutations in 1055 PIK3CA show gain of function. Proc. Natl. Acad. Sci. 104, 5569–5574 (2007). 1056 105. Dogruluk, T. et al. Identification of Variant-Specific Functions of PIK3CA by Rapid 1057 Phenotyping of Rare Mutations. Cancer Res. 75, 5341–5354 (2015). 1058 106. Yu, K. et al. PIK3CA variants selectively initiate brain hyperactivity during 1059 gliomagenesis. Nature 578, 166–171 (2020). 1060 107. Martinez-Corral, I. et al. Blockade of VEGF-C signaling inhibits lymphatic 1061 malformations driven by oncogenic PIK3CA mutation. Nat. Commun. 11, (2020). 1062 108. Delestre, F. et al. Alpelisib administration reduced lymphatic malformations in a 1063 mouse model and in patients. Sci. Transl. Med. 13, (2021). 1064 109. Roy, A. et al. Mouse models of human PIK3CA-related brain overgrowth have 1065 acutely treatable epilepsy. Elife 4, (2015). 1066 110. International Society for the Study of Vascular Anomalies. ISSVA classification for 1067 vascular anomalies - 2018. 20th ISSVA Work. (2018). 1068 111. Angulo-Urarte, A., van der Wal, T. & Huveneers, S. Cell-cell junctions as sensors 1069 and transducers of mechanical forces. Biochim. Biophys. Acta - Biomembr. 1862, 1070 183316 (2020). 1071 112. Petzold, J. & Gentleman, E. Intrinsic Mechanical Cues and Their Impact on Stem 1072 Cells and Embryogenesis. Front. Cell Dev. Biol. 9, 3112 (2021). 1073 113. Valet, M., Siggia, E. D. & Brivanlou, A. H. Mechanical regulation of early vertebrate 1074 embryogenesis. Nat. Rev. Mol. Cell Biol. 2021 233 23, 169–184 (2021). 1075 114. Li, D. et al. Role of mechanical factors in fate decisions of stem cells. 1076 http://dx.doi.org/10.2217/rme.11.2 6, 229–240 (2011). 1077 115. Carley, E., King, M. C. & Guo, S. Integrating mechanical signals into cellular 1078 identity. Trends Cell Biol. 0, (2022). 1079 116. De Belly, H., Paluch, E. K. & Chalut, K. J. Interplay between mechanics and 1080 signalling in regulating cell fate. Nat. Rev. Mol. Cell Biol. 2022 1–16 (2022) 1081 doi:10.1038/s41580-022-00472-z. 1082 117. Gordon, E., Schimmel, L. & Frye, M. The Importance of Mechanical Forces for in 1083 vitro Endothelial Cell Biology. Front. Physiol. 11, (2020). 1084 118. Weiss, J. M. et al. Anatomic position determines oncogenic specificity in melanoma. 1085 Nat. 2022 1–8 (2022) doi:10.1038/s41586-022-04584-6. 1086 119. Yu, H., Mouw, J. K. & Weaver, V. M. Forcing Form and Function: Biomechanical 1087 Regulation of Tumor Evolution. Trends Cell Biol. 21, 47 (2011). 1088 120. Broders-Bondon, F., Ho-Bouldoires, T. H. N., Fernandez-Sanchez, M. E. & Farge, 1089 E. Mechanotransduction in tumor progression: The dark side of the force. J. Cell 1090 Biol. 217, 1571–1587 (2018). 1091 121. Huang, W. et al. Regulatory networks in mechanotransduction reveal key genes in 1092 promoting cancer cell stemness and proliferation. Oncogene 2019 3842 38, 6818– 1093 6834 (2019). 1094 122. Angulo-Urarte, A. et al. Endothelial cell rearrangements during vascular patterning 1095 require PI3-kinase-mediated inhibition of actomyosin contractility. Nat. Commun. 9, 1096 1–16 (2018). 1097 123. Di-Luoffo, M., Ben-Meriem, Z., Lefebvre, P., Delarue, M. & Guillermet-Guibert, J. 1098 PI3K functions as a hub in mechanotransduction. Trends Biochem. Sci. 46, 878– 1099 888 (2021). 1100 124. Di-Luoffo, M., Delarue, M. & Guillermet-Guibert, J. The biomechanical context 1101 influences the output signaling, independently of PIK3CA mutations in breast cancer 1102 cells. bioRxiv 2021.10.18.464825 (2021) doi:10.1101/2021.10.18.464825. 1103 125. Yum, M. K. et al. Tracing oncogene-driven remodelling of the intestinal stem cell 1104 niche. Nat. 2021 5947863 594, 442–447 (2021). 1105 126. Karthikeyan, S. et al. Hierarchical tumor heterogeneity mediated by cell contact 1106 between distinct genetic subclones. J. Clin. Invest. 131, (2021). 1107 127. Malinverno, M. et al. Endothelial cell clonal expansion in the development of 1108 cerebral cavernous malformations. Nat. Commun. 2019 101 10, 1–16 (2019). 1109 128. Detter, M. R., Snellings, D. A. & Marchuk, D. A. Cerebral cavernous malformations 1110 develop through clonal expansion of mutant endothelial cells. Circ. Res. 123, 1143– 1111 1151 (2018). 1112 129. Hassanein, A. H. et al. Venous malformation: risk of progression during childhood 1113 and adolescence. Ann. Plast. Surg. 68, 198–201 (2012). 1114 130. Hiruma, H., Kitsukawa, K., Ogawa, Y. & Mimura, H. Venous malformation of the 1115 foot: Spontaneous regression postpartum on MRI. Radiol. Case Reports 16, 62–65 1116 (2021). 1117 131. Orsenigo, F. et al. Mapping endothelial-cell diversity in cerebral cavernous 1118 malformations at single-cell resolution. Elife 9, 1–34 (2020). 1119 132. Castillo, S. D., Baselga, E. & Graupera, M. PIK3CA mutations in vascular 1120 malformations. Current Opinion in Hematology vol. 26 170–178 (2019). 1121 133. Queisser, A., Seront, E., Boon, L. M. & Vikkula, M. Genetic Basis and Therapies for 1122 Vascular Anomalies. Circ. Res. 129, 155–173 (2021). 1123 134. Le Cras, T. D. & Boscolo, E. Cellular and molecular mechanisms of PIK3CA-related 1124 vascular anomalies. Vasc. Biol. 1, H33–H40 (2019). 1125 135. Mäkinen, T., Boon, L. M., Vikkula, M. & Alitalo, K. Lymphatic Malformations: 1126 Genetics, Mechanisms and Therapeutic Strategies. Circ. Res. 129, 136–154 (2021). 1127 136. Rodriguez‑Laguna, L. et al. Somatic activating mutations in PIK3CA cause 1128 generalized lymphatic anomaly. J. Exp. Med. 216, 407–418 (2019). 1129 137. Limaye, N. et al. Somatic mutations in angiopoietin receptor gene TEK cause 1130 solitary and multiple sporadic venous malformations. Nat. Genet. 2008 411 41, 118– 1131 124 (2008). 1132 138. Soblet, J., Limaye, N., Uebelhoer, M., Boon, L. M. & Vikkula, M. Variable Somatic 1133 TIE2 Mutations in Half of Sporadic Venous Malformations. Mol. Syndromol. 4, 179– 1134 183 (2013). 1135 139. Goines, J. et al. A xenograft model for venous malformation. Angiogenesis 21, 725– 1136 735 (2018). 1137 140. Zhou, Z. et al. Cerebral cavernous malformations arise from endothelial gain of 1138 MEKK3–KLF2/4 signalling. Nat. 2016 5327597 532, 122–126 (2016). 1139 141. Fischer, A., Zalvide, J., Faurobert, E., Albiges-Rizo, C. & Tournier-Lasserve, E. 1140 Cerebral cavernous malformations: from CCM genes to endothelial cell 1141 homeostasis. Trends Mol. Med. 19, 302–308 (2013). 1142 142. Hong, T. et al. Somatic MAP3K3 and PIK3CA mutations in sporadic cerebral and 1143 spinal cord cavernous malformations. Brain 144, 2648–2658 (2021). 1144 143. Weng, J. et al. Somatic MAP3K3 mutation defines a subclass of cerebral cavernous 1145 malformation. Am. J. Hum. Genet. 108, 942–950 (2021). 1146 144. Snellings, D. A. et al. Developmental venous anomalies are a genetic primer for 1147 cerebral cavernous malformations. Nat. Cardiovasc. Res. 2022 13 1, 246–252 1148 (2022). 1149 145. Holm, A., Heumann, T. & Augustin, H. G. Microvascular Mural Cell Organotypic 1150 Heterogeneity and Functional Plasticity. Trends Cell Biol. 28, 302–316 (2018). 1151 146. Cellular Origin of Sporadic CCMs. N. Engl. J. Med. 386, 1290–1292 (2022). 1152 147. Figueiredo, A. M. et al. Phosphoinositide 3-kinase-regulated pericyte maturation 1153 governs vascular remodeling. Circulation (2020) 1154 doi:10.1161/CIRCULATIONAHA.119.042354. 1155 148. Zhang, P., Liu, L., Cao, Y., Wang, S. & Zhao, J. Cerebellar cavernous 1156 malformations with and without associated developmental venous anomalies. BMC 1157 Neurol. 13, 1–8 (2013). 1158 149. Idiculla, P. S., Gurala, D., Philipose, J., Rajdev, K. & Patibandla, P. Cerebral 1159 Cavernous Malformations, Developmental Venous Anomaly, and Its Coexistence: A 1160 Review. Eur. Neurol. 83, 360–368 (2020). 1161 150. Rosenthal, J., Sibbald, C., Jen, M., Deardorff, M. A. & Treat, J. A PIK3CA mutation 1162 in an acquired capillary malformation. Pediatr. Dermatol. 37, 246–247 (2020). 1163 151. Ricci, K. W. et al. Efficacy of systemic sirolimus in the treatment of generalized 1164 lymphatic anomaly and Gorham–Stout disease. Pediatr. Blood Cancer 66, (2019). 1165 152. Maruani, A. et al. Sirolimus (Rapamycin) for Slow-Flow Malformations in Children: 1166 The Observational-Phase Randomized Clinical PERFORMUS Trial. JAMA 1167 Dermatology (2021) doi:10.1001/JAMADERMATOL.2021.3459. 1168 153. Hammer, J. et al. Sirolimus is efficacious in treatment for extensive and/or complex 1169 slow-flow vascular malformations: A monocentric prospective phase II study. 1170 Orphanet J. Rare Dis. 13, 1–13 (2018). 1171 154. Adams, D. M. et al. Efficacy and safety of sirolimus in the treatment of complicated 1172 vascular anomalies. Pediatrics 137, (2016). 1173 155. Yu, Y. et al. In-vitro and in-vivo combined effect of ARQ 092, an AKT inhibitor, with 1174 ARQ 087, a FGFR inhibitor. Anticancer. Drugs 28, 503–513 (2017). 1175 156. Forde, K. et al. Clinical experience with the AKT1 inhibitor miransertib in two 1176 children with PIK3CA-related overgrowth syndrome. Orphanet J. Rare Dis. 16, 1–7 1177 (2021). 1178 157. Biesecker, L. G. et al. Clinical report: one year of treatment of Proteus syndrome 1179 with miransertib (ARQ 092). Cold Spring Harb. Mol. case Stud. 6, (2020). 1180 158. Luu, M. et al. Safety and efficacy of low-dose PI3K inhibitor taselisib in adult 1181 patients with CLOVES and Klippel–Trenaunay syndrome (KTS): the TOTEM trial, a 1182 phase 1/2 multicenter, open-label, single-arm study. Genet. Med. 2021 2312 23, 1183 2433–2442 (2021). 1184 159. Narayan, P. et al. FDA approval summary: Alpelisib plus fulvestrant for patients with 1185 hr-positive, HER2-negative, PIK3CA-mutated, advanced or metastatic breast 1186 cancer. Clin. Cancer Res. 27, 1842–1849 (2021). 1187 160. Pagliazzi, A. et al. PIK3CA-Related Overgrowth Spectrum From Diagnosis to 1188 Targeted Therapy: A Case of CLOVES Syndrome Treated With Alpelisib. Front. 1189 Pediatr. 9, 732836 (2021). 1190 161. Fontelles, G. G., Pardo, J., Carme, | & Moreillo, G. Alpelisib to treat CLOVES 1191 syndrome, a member of the PIK3CA-related overgrowth syndrome spectrum. Br. J. 1192 Clin. Pharmacol. (2022) doi:10.1111/BCP.15270. 1193 162. Morin, G. et al. Treatment of two infants with PIK3CA-related overgrowth spectrum 1194 by alpelisib. J. Exp. Med. 219, (2022). 1195 163. Canaud, G. et al. LBA23 EPIK-P1: Retrospective chart review study of patients (pts) 1196 with PIK3CA-related Overgrowth Spectrum (PROS) who have received alpelisib 1197 (ALP) as part of a compassionate use programme. Ann. Oncol. 32, S1297 (2021). 1198 164. Nathan, N., Keppler-Noreuil, K. M., Biesecker, L. G., Moss, J. & Darling, T. N. 1199 Mosaic Disorders of the PI3K/PTEN/AKT/TSC/mTORC1 Signaling Pathway. 1200 Dermatol. Clin. 35, 51–60 (2017). 1201 165. Mester, J. & Eng, C. PTEN hamartoma tumor syndrome. Handb. Clin. Neurol. 132, 1202 129–137 (2015). 1203 166. Napolioni, V. & Curatolo, P. Genetics and Molecular Biology of Tuberous Sclerosis 1204 Complex. Curr. Genomics 9, 475–487 (2008). 1205 167. Randle, S. C. Tuberous sclerosis complex: A review. Pediatr. Ann. 46, e166–e171 1206 (2017). 1207 168. Lindhurst, M. J. et al. A Mosaic Activating Mutation in AKT1 Associated with the 1208 Proteus Syndrome . N. Engl. J. Med. 365, 611–619 (2011). 1209 169. Hussain, K. et al. An activating mutation of AKT2 and human hypoglycemia. 1210 Science (80-. ). 334, 474 (2011). 1211 170. Alcantara, D. et al. Mutations of AKT3 are associated with a wide spectrum of 1212 developmental disorders including extreme megalencephaly. Brain 140, 2610–2622 1213 (2017). 1214 171. Nellist, M. et al. Germline activating AKT3 mutation associated with 1215 megalencephaly, polymicrogyria, epilepsy and hypoglycemia. Mol. Genet. Metab. 1216 114, 467–473 (2015). 1217 172. Cottrell, C. E. et al. Somatic PIK3R1 variation as a cause of vascular malformations 1218 and overgrowth. Genet. Med. 23, 1882–1888 (2021). 1219 173. Reijnders, M. R. F. et al. Variation in a range of mTOR-related genes associates 1220 with intracranial volume and intellectual disability. Nat. Commun. 2017 81 8, 1–12 1221 (2017). 1222 174. Mroske, C. et al. Germline activating MTOR mutation arising through gonadal 1223 mosaicism in two brothers with megalencephaly and neurodevelopmental 1224 abnormalities. BMC Med. Genet. 16, 1–11 (2015). 1225 175. Mirzaa, G. M. et al. Association of MTOR Mutations With Developmental Brain 1226 Disorders, Including Megalencephaly, Focal Cortical Dysplasia, and Pigmentary 1227 Mosaicism. JAMA Neurol. 73, 836–845 (2016). 1228 176. Møller, R. S. et al. Germline and somatic mutations in the MTOR gene in focal 1229 cortical dysplasia and epilepsy. Neurol. Genet. 2, (2016). 1230 177. Martin, K. R. et al. The genomic landscape of tuberous sclerosis complex. Nat. 1231 Commun. 2017 81 8, 1–13 (2017). 1232 178. Dobyns, W. B. & Mirzaa, G. M. Megalencephaly syndromes associated with 1233 mutations of core components of the PI3K-AKT–MTOR pathway: PIK3CA, PIK3R2, 1234 AKT3, and MTOR. American Journal of Medical Genetics, Part C: Seminars in 1235 Medical Genetics vol. 181 582–590 (2019). 1236 179. Margaria, J. P., Ratto, E., Gozzelino, L., Li, H. & Hirsch, E. Class II PI3Ks at the 1237 Intersection between Signal Transduction and Membrane Trafficking. Biomol. 2019, 1238 Vol. 9, Page 104 9, 104 (2019). 1239 180. Koch, P. A., Dornan, G. L., Hessenberger, M. & Haucke, V. The molecular 1240 mechanisms mediating class II PI 3-kinase function in cell physiology. FEBS J. 288, 1241 7025–7042 (2021). 1242 181. Backer, J. M. The intricate regulation and complex functions of the Class III 1243 phosphoinositide 3-kinase Vps34. Biochemical Journal vol. 473 2251–2271 (2016). 1244 182. Kobialka, P. & Graupera, M. Revisiting PI3-kinase signalling in angiogenesis. Vasc. 1245 Biol. 1, H125–H134 (2019). 1246 183. Graupera, M. et al. Angiogenesis selectively requires the p110α isoform of PI3K to 1247 control endothelial cell migration. Nat. 2008 4537195 453, 662–666 (2008). 1248 184. Lelievre, E., Bourbon, P. M., Duan, L. J., Nussbaum, R. L. & Fong, G. H. Deficiency 1249 in the p110α subunit of PI3K results in diminished Tie2 expression and Tie2-/-–like 1250 vascular defects in mice. Blood 105, 3935–3938 (2005). 1251 185. Bi, L., Okabe, I., Bernard, D. J., Wynshaw-Boris, A. & Nussbaum, R. L. Proliferative 1252 Defect and Embryonic Lethality in Mice Homozygous for a Deletion in the p110α 1253 Subunit of Phosphoinositide 3-Kinase. J. Biol. Chem. 274, 10963–10968 (1999). 1254 186. Stanczuk, L. et al. CKit lineage hemogenic endothelium-derived cells contribute to 1255 mesenteric lymphatic vessels. Cell Rep. 10, 1708–1721 (2015). 1256 187. Gupta, S. et al. Binding of Ras to Phosphoinositide 3-Kinase p110α Is Required for 1257 Ras- Driven Tumorigenesis in Mice. Cell 129, 957–968 (2007). 1258 188. Lee, M. Y. et al. Endothelial Akt1 mediates angiogenesis by phosphorylating 1259 multiple angiogenic substrates. Proc. Natl. Acad. Sci. U. S. A. 111, 12865–12870 1260 (2014). 1261 189. Gambardella, L. et al. PI3K signaling through the dual GTPase-activating protein 1262 ARAP3 is essential for developmental angiogenesis. Sci. Signal. 3, (2010). 1263 190. Potente, M. et al. Involvement of Foxo transcription factors in angiogenesis and 1264 postnatal neovascularization. J. Clin. Invest. 115, 2382–2392 (2005). 1265 191. Chu, M. et al. Angiopoietin receptor Tie2 is required for vein specification and 1266 maintenance via regulating COUP-TFII. Elife 5, (2016). 1267 192. Serra, H. et al. PTEN mediates Notch-dependent stalk cell arrest in angiogenesis. 1268 Nat. Commun. 6, (2015). 1269 193. Ola, R. et al. PI3 kinase inhibition improves vascular malformations in mouse 1270 models of hereditary haemorrhagic telangiectasia. Nat. Commun. 7, (2016). 1271 194. Monelli, E. et al. Angiocrine polyamine production regulates adiposity. Nat. Metab. 1272 (2022) doi:10.1038/S42255-022-00544-6. 1273 195. Drolet, B. A. et al. Consensus-derived practice standards plan for complicated 1274 Kaposiform hemangioendothelioma. J. Pediatr. 163, 285–291 (2013). 1275 196. Canaud, G. et al. EPIK-P2: A prospective phase 2, double-blind, randomized, 1276 placebo-controlled study of alpelisib in pediatric and adult patients (Pts) with 1277 PIK3CA-related overgrowth spectrum (PROS). J. Clin. Oncol. 39, TPS3160– 1278 TPS3160 (2021). 1279 197. Martinez-Glez, V. et al. Segmental undergrowth is associated with pathogenic 1280 variants in vascular malformation genes: A retrospective case-series study. Clin. 1281 Genet. (2021) doi:10.1111/CGE.14095. 1282 198. Loconte, D. C. et al. Molecular and functional characterization of three different 1283 postzygotic mutations in PIK3CA-related overgrowth spectrum (PROS) patients: 1284 Effects on PI3K/AKT/mTOR signaling and sensitivity to PIK3 inhibitors. PLoS One 1285 10, e0123092- (2015). 1286 199. Osborn, A. J. et al. Activating PIK3CA alleles and lymphangiogenic phenotype of 1287 lymphatic endothelial cells isolated from lymphatic malformations. Hum. Mol. Genet. 1288 24, 926–938 (2015). 1289 200. Castiglioni, C. et al. Activating PIK3CA somatic mutation in congenital unilateral 1290 isolated muscle overgrowth of the upper extremity. Am. J. Med. Genet. Part A 164, 1291 2365–2369 (2014). 1292 201. Marty, M. et al. Gingival Biopsy to Detect Mosaicism in Overgrowth Syndromes: 1293 Report of Two Cases of Megalencephaly-Capillary Malformation Syndrome with 1294 Periodontal Anomalies. Case Rep. Dent. 2020, (2020). 1295 202. Al-Qattan, M. M., Hadadi, A., Al-Thunayan, A. M., Eldali, A. A. & AlBalwi, M. A. 1296 Upper limb muscle overgrowth with hypoplasia of the index finger: A new over- 1297 growth syndrome caused by the somatic PIK3CA mutation c.3140A>G. BMC Med. 1298 Genet. 19, 1–7 (2018). 1299 203. Hafner, C. et al. Oncogenic PIK3CA mutations occur in epidermal nevi and 1300 seborrheic keratoses with a characteristic mutation pattern. Proc. Natl. Acad. Sci. 1301 104, 13450–13454 (2007). 1302 204. Zenner, K. et al. Somatic activating <em>BRAF</em> variants cause 1303 isolated lymphatic malformations. medRxiv 2021.11.03.21265682 (2021) 1304 doi:10.1101/2021.11.03.21265682. 1305 205. Hori, Y. et al. PIK3CA mutation correlates with mTOR pathway expression but not 1306 clinical and pathological features in Fibfibroipose vascular anomaly (FAVA). Diagn. 1307 Pathol. 17, 1–7 (2022). 1308 206. Koutlas, I. G. et al. Orofacial overgrowth with peripheral nerve enlargement and 1309 perineuriomatous pseudo-onion bulb proliferations is part of the PIK3CA-related 1310 overgrowth spectrum. Hum. Genet. Genomics Adv. 1, 100009 (2020). 1311 207. Li, J. F. et al.

An Analysis of the Pathogenic Genes and Mutation Sites of 1312 Macrodactyly

. Pharmgenomics. Pers. Med. 15, 55–64 (2022). 1313 208. Ten Broek, R. W. et al. Kaposiform hemangioendothelioma and tufted angioma – 1314 (epi)genetic analysis including genome-wide methylation profiling. Ann. Diagn. 1315 Pathol. 44, 151434 (2020). 1316 209. Park, S., Kim, S. H., Youn, S. E., Lee, S. T. & Kang, H. C. A Child With 1317 Lymphangioma Due to Somatic Mutation in PIK3CA Successfully Treated With 1318 Everolimus. Pediatr. Neurol. 91, 65–67 (2019). 1319 210. Song, W., Kwee, T. C. & Suurmeijer, A. J. Macrodactyly with a complex 1320 glomuvenous malformation in congenital lipomatous overgrowth with vascular 1321 malformations, epidermal naevi and skeletal anomalies (CLOVES) syndrome. 1322 Histopathology 73, 705–708 (2018). 1323 211. İli, E. G. et al. Phenotypic and molecular characterization of five patients with 1324 PIK3CA-related overgrowth spectrum (PROS). Am. J. Med. Genet. Part A (2022) 1325 doi:10.1002/AJMG.A.62709. 1326 212. McDermott, J. H., Byers, H. & Clayton-Smith, J. Detection of a mosaic PIK3CA 1327 mutation in dental DNA from a child with megalencephaly capillary malformation 1328 syndrome. Clin. Dysmorphol. 25, 16–18 (2016). 1329 213. Swarr, D. T. et al. Expanding the differential diagnosis of fetal hydrops: an unusual 1330 prenatal presentation of megalencephaly-capillary malformation syndrome. Prenat. 1331 Diagn. 33, 1010–1012 (2013). 1332 214. Stutterd, C. A. et al. Genetic heterogeneity of polymicrogyria: study of 123 patients 1333 using deep sequencing. Brain Commun. 3, (2021). 1334 215. Sinn, D. I., Shaik, R., Miglis, M. G., Muppidi, S. & Jaradeh, S. Orthostatic 1335 intolerance with Klippel-Trenaunay syndrome. Clin. Auton. Res. 2021 314 31, 577– 1336 579 (2021). 1337 216. Maguolo, A. et al. Clinical pitfalls in the diagnosis of segmental overgrowth 1338 syndromes: A child with the c.2740G > A mutation in PIK3CA gene. Ital. J. Pediatr. 1339 44, 1–6 (2018). 1340 217. Stutterd, C. et al. Polymicrogyria in association with hypoglycemia points to 1341 mutation in the mTOR pathway. Eur. J. Med. Genet. 61, 738–740 (2018). 1342 218. Nolte, J. W., Alders, M., Karssemakers, L. H. E., Becking, A. G. & Hennekam, R. C. 1343 M. Unilateral condylar hyperplasia in hemifacial hyperplasia, is there genetic proof 1344 of overgrowth? Int. J. Oral Maxillofac. Surg. 49, 1464–1469 (2020). 1345 219. Hiraide, T. et al. Genetic and phenotypic analysis of 101 patients with 1346 developmental delay or intellectual disability using whole-exome sequencing. Clin. 1347 Genet. 100, 40–50 (2021). 1348 220. Blesinger, H. et al. PIK3CA mutations are specifically localized to lymphatic 1349 endothelial cells of lymphatic malformations. PLoS One 13, e0200343 (2018). 1350 221. Yates, C. L. et al. Whole-exome sequencing on deceased fetuses with ultrasound 1351 anomalies: expanding our knowledge of genetic disease during fetal development. 1352 Genet. Med. 2017 1910 19, 1171–1178 (2017). 1353 222. Jin, S. C. et al. Exome sequencing implicates genetic disruption of prenatal neuro- 1354 gliogenesis in sporadic congenital hydrocephalus. Nat. Med. 2020 2611 26, 1754– 1355 1765 (2020). 1356 223. Turner, T. N. et al. Genome Sequencing of Autism-Affected Families Reveals 1357 Disruption of Putative Noncoding Regulatory DNA. Am. J. Hum. Genet. 98, 58–74 1358 (2016). 1359 224. Baldassari, S. et al. Dissecting the genetic basis of focal cortical dysplasia: a large 1360 cohort study. Acta Neuropathol. 138, 885–900 (2019). 1361 225. Palmieri, M. et al. A pilot study of next generation sequencing–liquid biopsy on cell- 1362 free DNA as a novel non-invasive diagnostic tool for Klippel–Trenaunay syndrome. 1363 Vascular 29, 85–91 (2021). 1364 226. Fortin, O., Ashour, M., Lacroix, C., Sabapathy, C. A. & Myers, K. A. 1365 Megalencephaly–Capillary Malformation–Polymicrogyria with Cerebral Venous 1366 Thrombosis. Can. J. Neurol. Sci. 47, 828–829 (2020). 1367 227. Mirzaa, G. M., Rivière, J. B. & Dobyns, W. B. Megalencephaly Syndromes and 1368 Activating Mutations in the PI3K-AKT Pathway: MPPH and MCAP. Am. J. Med. 1369 Genet. Part C Semin. Med. Genet. 163, 122–130 (2013). 1370 228. Cohen, A. S. A. et al. Somatic mosaicism for the p.His1047Arg mutation in PIK3CA 1371 in a girl with mesenteric lipomatosis. Am. J. Med. Genet. Part A 164, 2360–2364 1372 (2014). 1373 229. Rasmussen, M., Sunde, L., Weigert, K. P., Bogaard, P. W. & Lildballe, D. L. 1374 Segmental overgrowth syndrome due to an activating PIK3CA mutation identified in 1375 affected muscle tissue by exome sequencing. Am. J. Med. Genet. Part A 164, 1376 1318–1321 (2014). 1377 230. Tapper, W. J. et al. Megalencephaly Syndromes: Exome Pipeline Strategies for 1378 Detecting Low-Level Mosaic Mutations. PLoS One 9, e86940 (2014). 1379 231. Bornstein, E. et al. Isolated Fetal Macrodactyly: Phenotypic and Genetic Disparities 1380 in Mosaic Overgrowth Syndrome. J. Ultrasound Med. 33, 1305–1307 (2014). 1381 232. Emrick, L. T. et al. Prenatal diagnosis of CLOVES syndrome confirmed by detection 1382 of a mosaic PIK3CA mutation in cultured amniocytes. Am. J. Med. Genet. Part A 1383 164, 2633–2637 (2014). 1384 233. D’Gama, A. M. et al. Mammalian target of rapamycin pathway mutations cause 1385 hemimegalencephaly and focal cortical dysplasia. Ann. Neurol. 77, 720–725 (2015). 1386 234. Udomchaiprasertkul, W., Kitidumrongsook, P., Suphapeetiporn, K. & Shotelersuk, 1387 V. A somatic pik3ca p.h1047l mutation in a Thai patient with isolated macrodactyly: 1388 A case report. Asian Biomed. 13, 33–36 (2019). 1389 235. Tian, W. et al. Phenotypic and genetic spectrum of isolated macrodactyly: somatic 1390 mosaicism of PIK3CA and AKT1 oncogenic variants. Orphanet J. Rare Dis. 15, 1–7 1391 (2020). 1392 236. Gripp, K. W. et al. Nephroblastomatosis or Wilms tumor in a fourth patient with a 1393 somatic PIK3CA mutation. Am. J. Med. Genet. Part A 170, 2559–2569 (2016). 1394 237. Suzuki, Y. et al. The effect of rapamycin, NVP-BEZ235, aspirin, and metformin on 1395 PI3K/AKT/mTOR signaling pathway of PIK3CA -related overgrowth spectrum 1396 (PROS). Oncotarget 8, 45470–45483 (2017). 1397 238. Biderman Waberski, M. et al. Urine cell-free DNA is a biomarker for 1398 nephroblastomatosis or Wilms tumor in PIK3CA-related overgrowth spectrum 1399 (PROS). Genet. Med. 20, 1077–1081 (2018). 1400 239. Morandi, G. et al. Limb hypertrophy: a skin vascular malformation and bilateral 1401 hydroureteronephrosis in a neonate. Arch. Dis. Child. - Educ. Pract. 104, 97–98 1402 (2019). 1403 240. Valentini, V. et al. PIK3CA c.3140A>G mutation in a patient with suspected Proteus 1404 Syndrome: a case report. Clin. Case Reports 6, 1358–1363 (2018). 1405 241. Roth, G. M., Ferguson, N. & Wanat, K. A. Segmental epidermal nevus and mucosal 1406 neuromas associated with PIK3CA-related overgrowth spectrum disorder. JAAD 1407 case reports 4, 1080–1082 (2018). 1408 242. Quindipan, C. et al. Custom Pediatric Oncology Next-Generation Sequencing Panel 1409 Identifies Somatic Mosaicism in Archival Tissue and Enhances Targeted Clinical 1410 Care. Pediatr. Neurol. 114, 55–59 (2021). 1411 243. Garde, A. et al. Clinical and neuroimaging findings in 33 patients with MCAP 1412 syndrome: A survey to evaluate relevant endpoints for future clinical trials. Clin. 1413 Genet. 99, 650–661 (2021). 1414 244. Fraissenon, A. et al. Prenatal imaging diagnosis of PIK3CA-related overgrowth 1415 spectrum disorders in first trimester with emphasis on extremities. Ultrasound 1416 Obstet. Gynecol. 56, 780–781 (2020). 1417 245. Keppler-Noreuil, K. M. et al. Thrombosis risk factors in PIK3CA-related overgrowth 1418 spectrum and Proteus syndrome. Am. J. Med. Genet. Part C Semin. Med. Genet. 1419 181, 571–581 (2019). 1420 246. Youssefian, L. et al. Fibroadipose Hyperplasia versus Proteus Syndrome: 1421 Segmental Overgrowth with a Mosaic Mutation in the PIK3CA Gene. J. Invest. 1422 Dermatol. 135, 1450–1453 (2015). 1423 247. Tripolszki, K. et al. Somatic mosaicism of the PIK3CA gene identified in a Hungarian 1424 girl with macrodactyly and syndactyly. Eur. J. Med. Genet. 59, 223–226 (2016). 1425 248. Leiter, S. M. et al. Hypoinsulinaemic, hypoketotic hypoglycaemia due to mosaic 1426 genetic activation of PI3-kinase. Eur. J. Endocrinol. 177, 175–186 (2017). 1427 249. Quinlan-Jones, E. et al. Prenatal detection of PIK3CA-related overgrowth spectrum 1428 in cultured amniocytes using long-range PCR and next-generation sequencing. 1429 Pediatr. Dev. Pathol. 20, 54–57 (2017). 1430 250. Hanafusa, H. et al. A girl with CLOVES syndrome with a recurrent PIK3CA somatic 1431 mutation and pancreatic steatosis. Hum. Genome Var. 2019 61 6, 1–3 (2019). 1432 251. Denorme, P. et al. Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit 1433 alpha (PIK3CA)-related overgrowth spectrum: A brief report. Pediatr. Dermatol. 35, 1434 e186–e188 (2018). 1435 252. Briand, C. et al. Congenital infiltrating lipomatosis of the face with lingual mucosal 1436 neuromas associated with a PIK3CA mutation. Pediatr. Dermatol. 37, 1128–1130 1437 (2020). 1438 253. Di Donato, N. et al. Identification and Characterization of a Novel Constitutional 1439 PIK3CA Mutation in a Child Lacking the Typical Segmental Overgrowth of “PIK3CA- 1440 Related Overgrowth Spectrum”. Hum. Mutat. 37, 242–245 (2016). 1441 254. Weinstein, B., Henderson-Jackson, E., Cruse, C. W. & Brohl, A. S. PIK3CA-Related 1442 Overgrowth Syndrome (PROS) and Angiosarcoma: A Case Report. Eplasty 20, ic6 1443 (2020). 1444 255. Yan, W. et al. Somatic frameshift mutation in PIK3CA causes CLOVES syndrome 1445 by provoking PI3K/AKT/mTOR pathway. Hereditas 158, 1–8 (2021). 1446 256. Michel, M. E. et al. Causal somatic mutations in urine DNA from persons with the 1447 CLOVES subgroup of the PIK3CA-related overgrowth spectrum. Clin. Genet. 93, 1448 1075–1080 (2018). 1449 257. Garneau, A. P., Haydock, L., Tremblay, L. E. & Isenring, P. Somatic non-cancerous 1450 PIK3CA-related overgrowth syndrome treated with alpelisib in North America. J. 1451 Mol. Med. 99, 311–313 (2021). 1452 258. Tous-Romero, F., Quesada-Espinosa, J. F., Sánchez-Calvín, M. T., Ortiz-Romero, 1453 P. L. & Palencia-Pérez, S. PIK3CA-related overgrowth spectrum: concurrence of 1454 multiple anomalies in one patient. JDDG J. der Dtsch. Dermatologischen 1455 Gesellschaft 16, 603–605 (2018). 1456 259. Di Stasi, M. et al. Cerebellar dysplasia related to PIK3CA mutation: a three-case 1457 series. neurogenetics 2020 221 22, 27–32 (2020). 1458 260. Collins, M., Krochmalnek, E., Alsubhi, S. & Srour, M. Teaching NeuroImages: 1459 CLOVES Syndrome. Neurology 96, e1487–e1488 (2021). 1460 261. Davis, S. et al. Growth hormone deficiency in megalencephaly-capillary 1461 malformation syndrome: An association with activating mutations in PIK3CA. Am. J. 1462 Med. Genet. Part A 182, 162–168 (2020). 1463 262. Chang, F. et al. Molecular Diagnosis of Mosaic Overgrowth Syndromes Using a 1464 Custom-Designed Next-Generation Sequencing Panel. J. Mol. Diagnostics 19, 613– 1465 624 (2017). 1466 263. Hucthagowder, V. et al. Utility of clinical high-depth next generation sequencing for 1467 somatic variant detection in the PIK3CA-related overgrowth spectrum. Clin. Genet. 1468 91, 79–85 (2017). 1469 264. Boccara, O. et al. Soft tissue angiomatosis: another PIK3CA-related disorder. 1470 Histopathology 76, 540–549 (2020). 1471 265. Yeung, K. S. et al. Somatic PIK3CA mutations in seven patients with PIK3CA- 1472 related overgrowth spectrum. Am. J. Med. Genet. Part A 173, 978–984 (2017). 1473 266. Piacitelli, A. M. et al. Characterization of a severe case of PIK3CA-related 1474 overgrowth at autopsy by droplet digital polymerase chain reaction and report of 1475 PIK3CA sequencing in 22 patients. Am. J. Med. Genet. Part A 176, 2301–2308 1476 (2018). 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 Type of administration and Number Title Phase Nº Patients/Pathology Age Results Ref dose Sirolimus/Rapamycin Clinical Trial Assessing Efficacy and Phase 2; -Efficacious and well tolerated. 60; Complicated Vascular Safety of the mTOR Inhibitor Interventional; up to Oral; Liquid dosing based on -Effective at stabilizing or reducing NCT00975819 Anomalies without genetic 151,154 Sirolimus in the Treatment of Single Group 31Y trough levels signs/symptoms in GLA and GSD testing Complicated VA. Assignment patients. Oral solution or tablets, Phase 2; Treatment of Superficial Voluminous 59; Slow-flow vascular starting with - Decrease LM volume together with Interventional; 6Y to NCT02509468 Complicated Slow-flow Vascular malformations including 0.08mg/kg/twice/d, with dose oozing, and bleeding. 152 Crossover 18Y Malformations With Sirolimus. syndromic forms* adjustments (6mg/d - Less efficient for VMs. Assignment maximum) - Beneficial especially for PROS with Phase 2; NCT02443818 PROMISE: Sirolimus Effect on 16;11;32; PROS with Adult: tablet 1mg/d; Children: progressive adipose tissue overgrowth. Interventional; 3Y to 2014-000484-41 Hypertrophic Syndromes Related confirmed PIK3CA oral solution 0.5mg/twice/d - High rate of discontinuations 45 Single Group 65Y NCT02428296 Gene PIK3CA. mutations tablet/solution. Max 1.5mg -No clear decrease in AE rates Assignment compared with higher-dose therapy. Clinical Study on Efficacy and Phase 3; 3Y to 19; microcystic LM, GLA or Children under 12: Oral - Partial response in all patients, NCT01811667; Safety of the mTOR Rapamycin Interventional; 64Y complex vascular solution 0.8 mg/m2 /twice/d; reducing symptoms and increasing 153 2012–001262-15 Inhibitor Found in the Complex Single Group (media malformations* Adult: tablet, 2mg/d QoL. Vascular Malformations. Assignment n: 15) A Phase II Clinical Study to Phase 2/3; Oral. Children under 3: 30; patients with KHE and TA Evaluate the Efficacy and Safety of Interventional; 1M to 0.8mg/m2/daily; Above 3: NCT04598204 and complicated vascular - Ongoing trial. 195 Rapamycin in Complex Vascular Single Group 14Y twice/d for a dosage of 10- malformations Anomalies in Pediatric Patients Assignment 15ng/ml in plasma Phase 3; Treatment of Congenital Vascular Interventional; 75; Congenital VM, or LM or NCT03987152 Malformations Using Sirolimus: Challenge- ≥1Y Daily intake - Ongoing trial. - combined. Improving Quality of Life. Dechallenge- Rechallenge Phase 2; Oral solution; 1mg/mL- TOPGUN: TOPical Sirolimus in Interventional; 12; LMLMs not associated to 0.5 mL to 1 mL/d on LMLM NCT04128722 linGUal Microcystic Lymphatic ≥5Y - Ongoing trial. - Crossover CLAPO according to the size of the Malformation. Assignment lesion Phase 2; 0.1% Topical Sirolimus in the Topical 0.1% sirolimus daily Interventional; NCT03972592 Treatment of CMLM in Children and 55; Primary CMLM ≥ 6Y on the randomly allocated - Ongoing trial. - Parallel Adults. area Assignment Phase 3; Efficacy and Safety of Sirolimus in 250; VM, LM or complex Interventional; 3M to NCT02638389 VA That Are Refractory to Standard vascular malformations (KTS, Not specified - Ongoing trial. - Single Group 70Y Care. PTEN, etc.) Assignment Phase 4; Efficacy of Rapamycin (Sirolimus) in Oral; Children: 1 mg/m2 body Interventional; NCT03767660 the Treatment of BRBNS, 20; BRBNS and VMs all surface area/d; Adults: 2 - Ongoing trial. - Single Group Hereditary or Sporadic VM. mg/m2/d Assignment Miransertib Phase 2; A Multi-Cohort Phase 2 Dose- 45; Proteus syndrome with Interventional; NCT04316546 Escalation Study of MK-7075 confirmed somatic AKT1 ≥3Y Not specified - Ongoing trial. - Single Group (Miransertib) in Proteus Syndrome. mutation Assignment 85; PIK3CA-Related MOSAIC: Study of Miransertib (MK- Phase1/2; Overgrowth Spectrum 7075) in Participants With PIK3CA- Interventional; Oral capsules; 15 mg/m2; up NCT03094832 (PROS)/Proteus Syndrome ≥2Y - Ongoing trial. - related Overgrowth Spectrum and Parallel to 25 mg/m2 with confirmed PI3KCA or Proteus Syndrome. Assignment AKT1 mutation. Overgrowth diseases and/or vascular anomalies with NCT03317366 Expanded Access of MOSAIC. - ≥2Y Capsules - Ongoing trial. - confirmed PI3KCA or AKT1 mutation. Taselisib Phase 1/2; - 76.4% reported clinical improvement. TOTEM: Trial of Taselisib in Interventional; 19: PROS with confirmed 16Y to - No reduction of lesion volume. NCT03290092 1 or 2 mg/d 158 Overgrowth. Single Group PI3KCA mutations 65Y - Unfavorable safety profile in KTS and Assignment CLOVES, Alpelisib - Reduction in lesion volume - Improve QoL due to reduction of EPIK-P1: Retrospective Chart Observational major symptoms/sings in the majority of Review Study of Patients With 57; PROS with confirmed Adult:250 mg/d; Children: 50 NCT04285723 : ≥2Y cases. 163 PROS Who Have Received PIK3CA mutation mg/d case only - 39% alpelisib-related AE: Alpelisib. hyperglycaemia, aphthous ulcer and stomatitis. EPIK-P3: Study Assessing Long- Phase 2; term Safety and Efficacy of Alpelisib NCT04980833/ Interventional; 50; PROS with confirmed Doses permitted are 50, 125, in Patients With PROS Who ≥2Y - Ongoing trial. - 2020-005896-12 Single Group PI3KCA mutation 200 and 250 mg. Previously Participated in Study Assignment EPIK-P1. Phase 2 multi- EPIK-P2: Study Assessing the center; Efficacy, Safety and PK of Alpelisib 150; PROS with confirmed Oral tablet; Children 2-17: 50 NCT04589650 Interventional; ≥2Y - Ongoing trial. 196 (BYL719) in Pediatric and Adult PIK3CA mutation mg/d; Adults: 125 mg/d Parallel Patients With PROS. Assignment 1493 1494 Table 1: Summary of previous and ongoing PROS clinical trials with PI3K-related inhibitors. 1495 (Y) Years; (M) months; (d) daily; (AE) Adverse events; (PROS) PIK3CA-related overgrowth spectrum; 1496 (CLAPO) Capillary malformation of the lower lip, lymphatic malformation of the face and neck, 1497 asymmetry and partial/generalized overgrowth; (LMLM) Lingual Microcystic Lymphatic Malformation; 1498 (CMCM) Cutaneous microcystic lymphatic malformations; (LM) lymphatic malformations; (VM) 1499 venous malformations; (GLA) Generalized lymphatic anomaly; (GSD) Gorham-Stout 1500 disease;(BRBNS) Blue rubber bleb nevus syndrome. *Some patients included in the clinical trial have 1501 a genetic diagnosis. 1502 1503 1504 1505 Figure1. Classification of PIK3CA-related overgrowth spectrum (PROS) disorders. PROS 0B 1506 patients share a variety of clinical manifestations, with overgrowth being found in different tissues 1507 such as brain, skin, vasculature, bone, muscle, adipose and connective. PROS disorders are divided 1508 in: isolated PROS, when overgrowth is locally found affecting only one tissue or body part, and 1509 syndromic PROS, when overgrowth is not focal, and it is presented in at least two different systems 1510 together with other features59. Belong to isolated PROS: (1) Brain overgrowth: DMEG- Dysplastic 1511 Megalencephaly; HMEG- HemiMegalencephaly; FCD- Focal Cortical Dysplasia; (2) Vascular 1512 overgrowth/malformations: VMs-Venous Malformations; LMs-Lymphatic Malformations; CCMs- 1513 Cerebral Cavernous Malformations; LVMs- Lymphatic-Venous Malformations; CVLMs- Capillary- 1514 Venous-Lymphatic Malformations; GLA- Generalized Lymphatic Anomaly; (3) Skin lesions: BLK- 1515 Benign Lichenoid Keratosis; EN- Epidermal Nevi; SK- Seborrhoeic Keratosis; (4) Combined lesions 1516 (2 or more tissues locally affected): FAVA- Fibro Adipose Vascular Anomaly, LON- Lipomatosis of 1517 Nerve (Macrodactyly and others), MHH- Muscular HemiHyperplasia and FIL- Facial Infiltrating 1518 Lipomatosis. Belong to syndromic PROS: MCAP- Megalencephaly-Capillary malformation (CM) 1519 syndrome; DCMO- Diffuse Capillary Malformation (CM) with Overgrowth; KTS- Klippel-Trenaunay 1520 syndrome; CLAPO syndrome- Capillary vascular malformation of the lower lip, Lymphatic 1521 malformations of the head and neck, Asymmetry and Partial/generalized Overgrowth; CLOVES 1522 syndrome- Congenital Lipomatous Overgrowth, Vascular malformations, Epidermal nevi, and 1523 Skeletal/Spinal abnormalities; FH/FAO- FibroAdipose Overgrowth; HHML- HemiHyperplasia-Multiple 1524 Lipomatosis; Segmental undergrowth with vascular malformations. *Recently, some patients with 1525 activating PIK3CA mutations presented tissue undergrowth together with vascular malformations with 1526 or without the presence of tissue overgrowth197. 1527 1528 1529 Figure 2: Intrinsic and extrinsic factors that define PIK3CA-related pathogenesis in PROS. 1B 1530 Somatic activating PIK3CA mutations are acquired during embryonic development. However, a 1531 variety of elements are at play when defining the ultimate clinical outcome: the developmental time 1532 when the genetic error occurs, the type of cell lineage that acquire a PIK3CA mutation, the degree 1533 and mechanism of PI3Kα activation (PIK3CA variant) and the spatiotemporal environmental modifiers 1534 of PI3Kα signalling. 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 Figure 3. PIK3CA variants in PROS. Summary of the documented mutations in PIK3CA that have 2B 1549 been found in 1173 PROS patients12,42–58,61,65–69,72–74,76,78–80,89–91,93–95,108,136,150,160,161,197–229,229–266. Two 1550 patients showed two different mutations in PIK3CA. 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 Figure BOX1. PI3K-related congenital disorders and tested inhibitors. Tested or current 3B 1565 inhibitors in clinical trials for PROS are alpelisib (PI3Kα isoform specific inhibitor), taselisip (pan-PI3K 1566 inhibitor), miransertib (pan-AKT inhibitor) and sirolimus (mTOR inhibitor). GF- Growth factor; RTK- 1567 Receptor tyrosine kinase; PROS- PIK3CA-related overgrowth spectrum; PHTS- PTEN hamartoma 1568 tumor syndrome. 1569