Fibroadipose vascular anomaly (FAVA) and phosphatase and tensin homolog (PTEN) hamartoma of the soft tissue (PHOST) are characteristically intramuscular vascular anomalies with predilection for the extremities. They are known to exhibit clinical, histologic, and radiologic similarities that present a diagnostic challenge to clinicians and both may be difficult to treat. “Loss-of-function” PTEN mutations have been recognized in the pathogenesis of PHOST, while “gain-of-function” phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) mutations have been linked to the etiology of FAVA. However, genetic testing is not 100% specific in the diagnosis of either of these entities, and families may defer genetic testing for various reasons. Differentiating PHOST from FAVA is important, as the diagnosis of PHOST carries the risk of PTEN-driven neoplastic development, and the diagnosis of FAVA makes available a specific medical therapy that targets PIK3CA mutations.1
Both conditions were only recently identified, with PHOST first described in 2012 by Kurek et al and FAVA first reported in 2014 by Alomari et al.1,2 In the initial description of PHOST, it was noted that the lesions presented with pain and swelling by 15 years of age, and typically affected the lower extremity. Of the 34 patients with PHOST included, 18 had a positive PTEN mutation. On histology, PHOST demonstrated an encapsulated mass with a mix of fatty or myxoid fibrous tissues, clustered venous structures with muscularized walls, occasional arteriovenous communications, lymphoid follicles, bone foci, and hypertrophic nerves. On magnetic resonance imaging (MRI), the original PHOST description was of T1 contrast-enhancing and T2 hyperintense, serpiginous lesions that were occasionally multifocal and often extended through tissue planes. Alomari’s description of FAVA included 16 patients presenting at a median of 10.5 years of age with pain and occasional contractures affecting the extremities. On MRI, the lesions showed heterogeneous high signal intensity on T1 and T2 weighted images. Histopathologic findings revealed dense fibrous tissue and fat with lymphoplasmacytic aggregates and muscularized, clustered vessels. No genetic information was available in the original description of FAVA. Alomari and Hill have since commented on distinguishing features of FAVA compared to PHOST.2,3 In particular, FAVA patients lack syndromic features common in PHOST such as penile pigmentation, macrocephaly, and developmental delay. FAVAs may not display the hypervascularity or arteriovenous communications sometimes seen in PHOST.
Direct comparison of FAVA and PHOST cases is lacking in the literature. The objective of this case series is to provide a direct comparison of the clinical, histologic, radiologic, and genetic data of patients with FAVA and PHOST seen in 1 tertiary care center, to add to the growing body of literature on these rare conditions.
Materials and methods
This study was approved by the Human Research Protections Office at the University of New Mexico. All patients included have been seen at the University of New Mexico’s multidisciplinary Vascular Anomalies Clinic, which serves a population of approximately 2.3 million people. Patients were identified by searching the clinic’s patient database for the terms “FAVA” or “PHOST” during the time-period of 2014 to 2020. This retrospective case series includes 7 patients identified with onset of FAVA or PHOST between 2 and 18 years of age, all symptomatic after experiencing minor trauma. A retrospective review of the subjects’ electronic medical record was performed to extract variables including diagnosis, relevant pathology and genetic testing, age at presentation, deidentified clinical photographs, radiologic images, and treatments. Each variable was analyzed by members of our interdisciplinary vascular anomalies team, including 2 interventional radiologists, a diagnostic radiologist, a pediatric dermatologist, and a dermatopathologist according to respective areas of expertise.
Review of cases at our institution included 5 cases of FAVA and 2 cases of PHOST. Four of the 5 FAVA patients were female (80%), and both of the PHOST patients were female (100%). All patients presented during childhood. Mean age at presentation was 4 years old for the PHOST patients and 11 years old for the FAVA patients. All patients who underwent biopsies had genetic testing performed. Of the 5 patients with FAVA, 2 had biopsies taken of their lesion with subsequent genetic testing positive for PIK3CA mutations. Both patients with PHOST had biopsies positive for pathogenic PTEN mutations. The patients who did not have genetic testing did not have histopathologic specimens collected.
All patients had anomalies confined to the leg, except one of the FAVA patients who had forearm involvement. All patients endorsed progressively disabling pain at the lesion site. Clinical exam revealed subtle to moderate muscular enlargement in all patients without other cutaneous stigmata (Figure 1). None of the FAVA patients experienced contractures. The PHOST patients exhibited the common phenotypic feature of macrocephaly, but neither presented with other physical characteristics associated with PHOST, such as mucocutaneous lesions, genital lentiginosis, or other dermatologic lesions.1
On histology of mostly core biopsies, FAVA and PHOST exhibited very similar components (Figure 2). Proliferation of fibroadipose tissue in muscle is a hallmark feature compatible with either FAVA or PHOST. Additional findings that we noted in both lesions included capillaries in a background of myxoid stroma, muscularized vessels, background of sclerosis with small lymphoid aggregates and clusters of veins within adipose tissue. We did not find any features on histology that confidently differentiated the 2 entities.1
All patients received various imaging studies of their lesions; however, MRI in our cohort demonstrated subtle differences favoring a diagnosis of PHOST versus FAVA and was considered the most useful modality. Though both lesions are known as intramuscular vascular anomalies, MRI studies revealed that FAVAs exhibited more exclusive intramuscular involvement, with muscle replacement by intralesional fat compared with PHOSTs. FAVAs also tended to respect fascial planes, largely occupying a single muscular compartment (Figure 3). In contrast, PHOST lesions displayed a predilection for trans-spatial involvement with a poorly marginated, infiltrative appearance. PHOSTs additionally presented with more avid, mass-like enhancement to a greater extent than FAVAs. On ultrasound, FAVAs had hyperechoic fatty components with intermittent venous lakes and prominent phlebectasia, whereas PHOSTs had a less conspicuous venous component. FAVAs had a slow flow vascular supply compared to PHOSTs, which appeared relatively more hypervascular on Doppler ultrasound (Figure 4).
All of the aforementioned patients are being managed in a multidisciplinary fashion. The patients with FAVA have undergone combinations of sclerotherapy, embolization, and excision. Our FAVA patients for various reasons were not amenable to starting oral sirolimus. Both of our PHOST patients were started on sirolimus, however. At follow-up visits, these patients reported significantly decreased pain and symptomatic improvement of their PHOST lesion after the initiation of sirolimus, with one not requiring further treatment to date.
This is a retrospective case series of 7 patients with the diagnosis of FAVA or PHOST from a tertiary center for vascular anomalies identified between 2014 and 2020. Although clinical and histologic features were similar between these 2 entities, distinguishing features were found on imaging studies that could help in differentiating these diagnoses. Since the first descriptions of PHOST (2012) and FAVA (2014), we did not find published literature directly comparing the findings of these 2 strikingly similar diagnoses. FAVA and PHOST are commonly misdiagnosed, not only for one another, but also for other vascular anomalies, including venous malformations, hemangiomas, and arteriovenous malformations.2 Pain management of these entities can also prove challenging.
In the review of our cohort, we found that both FAVA and PHOST tend to present in early childhood. However, as all of our patients presented after an episode of minor trauma, it is conceivable that the associated pain prompted them to seek care earlier than would otherwise be typical for patients with these diagnoses. This possibility is supported by a comparison of the mean age at time of presentation in our study, 4 for PHOST and 11 for FAVA, versus 15.1 for PHOST and 10.5 for FAVA in other published series.1,4 Additionally, the small number of PHOST cases in this series may skew our average age at presentation lower. Clinically, both entities have a predilection to affect females. This sex-predominance persisted in our series, with females accounting for >80% of our patients.
Our clinical experience aligned with that described in other publications, with the most commonly reported symptom being localized pain, often significant and debilitating, with corresponding subtle to moderate enlargement of the affected muscle on physical exam. We did not observe associated contractures or other cutaneous stigmata in either group, potentially due to early intervention in our younger patient cohort. Although physical examination alone is not a reliable diagnostic tool, it should be noted that both of our PHOST patients were macrocephalic, a commonly reported feature of PHOST, and an important exam finding to elicit. Other phenotypic characteristics that have also been associated with PHOST include penile lentigines and mucocutaneous lesions1; these were not found in our patients, indicating that the vascular anomaly and increased head size may be the only clinical features seen.
In previously published series, similar language has been used to describe histologic and radiologic findings in FAVA and PHOST. Histologically, we were not able to confidently identify characteristics specific to FAVA relative to PHOST. This observation is supported in the literature, with both FAVA and PHOST described as soft tissue intramuscular masses with dense fibrofatty, myxoid tissue, lymphoid aggregates and abnormal, irregularly muscularized veins.1,2,4,5 Although other groups have noted the presence of arteriovenous communication in PHOST on excisional specimens as a helpful diagnostic clue,1,6 this feature was not seen in our patients. Of note, most of our specimens were derived from core biopsies; it is possible that larger, excisional specimens may have proven more elucidating.
Radiologically, FAVA and PHOST have both been described as heterogeneous intramuscular masses that display high T1 and T2 signal intensity on MRI.2,4,5 However, in our review, a number of radiologic features distinguished PHOST from FAVA. Specifically, MRI revealed that well-defined replacement of muscle by intralesional fat was more characteristic of lesions found in FAVA compared with PHOST. Our MRI results for FAVA are also consistent with published reports showing intramuscular fat admixed with heterogeneous T2/short tau inversion recovery hyperintense signal as well as mild to moderate postcontrast enhancement.2 Sonographically, FAVA showed a more well-defined venous component with phlebectasia, characteristics that have also been reported previously.5 Interestingly, a breach of fascial planes with a more infiltrative, trans-spatial appearance and relatively avid, mass-like enhancement were striking features on MRI examinations in PHOST that were not reliably seen in our FAVA cases. These findings are consistent with previously published descriptions of PHOST.1,2,4 Although underlying osseous abnormalities have been described in both entities, this was only demonstrated in one of our FAVA cases.1,2,5
Patients with FAVA and PHOST are often misdiagnosed, and clinical data can be combined with mutational analysis to aid in rectifying that. When biopsies or excisions of either entity are performed, next generation sequencing for genetic mutations is recommended to identify PIK3CA mutations associated with FAVA and PTEN mutations associated with PHOST. In the 2 FAVA patients that underwent biopsy of their lesion, mutational analysis showed positive PIK3CA mutations. PTEN mutations were present in both PHOST cases. These conditions are likely related through their effects on a common signaling pathway. Specifically, activation of PI3 kinase by a variety of ligands interacting with cell surface receptors, such as the tyrosine kinase receptor, serves to phosphorylate phosphatidylinositol-4,5-bisphosphate (PIP2) to phosphatidylinositol-3, 4, 5-triphosphate (PIP3) with ATP serving as the phosphate donor. PIP3 drives various downstream physiologic effects, including increased cell survival and proliferation via PIP3 binding proteins such as Akt. Conversely, PTEN works in the opposite direction by dephosphorylating PIP3 to PIP2 to shut off the signaling pathway (Figure 5). In FAVA, a “gain-of-function” mutation in PIK3CA would lead to more PIP3, driving Akt-dependent expression of genes involved in cell survival and proliferation. In contrast, pathophysiology of PHOST includes a “loss of function” mutation in PTEN that would lead to less conversion of PIP3 to PIP2, and more PIP3 to drive Akt-dependent signaling.1,7 This explains the similar outcome of greater expression of genes involved in cell survival and proliferation.
The shared pathophysiologic relationship in the cellular signaling pathway may explain the overlap in phenotypic features found in PHOST and FAVA and may provide insight into the efficacy of sirolimus in our PHOST patients. Although not described as a first-line treatment option in the majority of published cases, both of our PHOST patients were started on oral sirolimus (1.0-1.5 mg BID) with symptomatic improvement. This finding aligns with another recently published case where a lower dose of sirolimus was used with good efficacy in a 17-year-old female with PHOST.3 Sirolimus is a rapamycin inhibitor most commonly used in vascular anomalies linked with PIK3CA mutations, such as lymphatic malformations and FAVA. The mechanism of action can be explained by its inhibitory action on mammalian target of rapamycin, a downstream kinase involved in the PIK3CA pathway.8 As PTEN is a tumor suppressor gene involved in regulation of mammalian target of rapamycin and cellular proliferation, the apparent efficacy of sirolimus in our PHOST patients demonstrates the connectedness of the genetic pathways involved in vascular anomalies.
FAVA and PHOST are both rare vascular anomalies typically presenting with pain and local enlargement of an extremity, and diagnosis may be challenging. Differentiating FAVA from PHOST is important given the potential of neoplastic development in PHOST patients. In this retrospective case series, we found that PHOST and FAVA both presented symptomatically in early to midchildhood. MRI and Doppler ultrasonography can distinguish these 2 entities by identifying respect for fascial planes (FAVA not PHOST), replacement of muscle by fat (FAVA not PHOST) and low venous flow (FAVA not PHOST). Histopathology of the 2 entities is usually indistinguishable, perhaps especially on core biopsies (Table 1). Genomic testing of the pathology specimen is most valuable, as PHOST typically will show “loss-of-function” PTEN mutation and FAVA demonstrates a “gain-of-function” PIK3CA mutation. By directly comparing our experience with these 2 rare entities, we hope to add to the growing body of literature so that clinicians can be best guided in their diagnosis and management of FAVA and PHOST.
Table 1. -
Clinical Description, Results of Genetic Testing When Present and Relevant Therapies Received in Included Patients Diagnosed With Either FAVA or PHOST
||Left lower leg malformation
c.1633G>A, p.E545K; (7%) mutation
||Right hand/forearm malformation
||Right thigh malformation
c.1633G>AG, p.E545K; (5%) mutation
|Embolization + cryoablation ×1
||Left thigh malformation
||Embolization + cryoablation ×1
||Right thigh malformation
||Right lower leg malformation
c.697C>T, p.Arg233Ter; (57%) mutation
|Sirolimus: 1.0 mg BID
||Right thigh malformation
Numerous multifocal hamartomas
|Sirolimus: 1.5 mg BID
Histopathologic Findings (Core Biopsies)
||Muscular enlargement of extremities
Muscle replaced by intralesional fat
Well-defined venous component
Low venous flow
|Dense fibrous tissue/fat within atrophied skeletal muscles
Large, irregular muscularized venous channels
Small lymphoid aggregates
Fibrous, tissue-encircled, nerves
Small clustered channels within myxoid stroma
Ill-defined venous component
Hypervascular venous flow
Summary of clinical, radiologic and histopathologic findings observed in this case series.
FAVA indicates fibroadipose vascular anomaly; PHOST, PTEN hamartoma of the soft tissue.
1. Kurek KC, Howard E, Tennant LB, et al. PTEN hamartoma of soft tissue: a distinctive lesion in PTEN syndromes. Am J Surg Pathol. 2012;36:671–687.
2. Alomari AI, Spencer SA, Arnold RW, et al. Fibro-adipose vascular anomaly: clinical-radiologic-pathologic features of a newly delineated disorder of the extremity. J Pediatr Orthop. 2014;34:109–117.
3. Hill LR, Duis J, Kulungowski AM, et al. A challenging diagnosis: PTEN hamartoma tumor syndrome presenting as isolated soft-tissue vascular anomalies. J Vasc Anom. 2021;2:e011.
4. Wang KK, Glenn RL, Adams DM, et al. Surgical management of fibroadipose vascular anomaly of the lower extremities. J Pediatr Orthop. 2020;40:e227–e236.
5. Greene AK, Goss JA. Vascular anomalies. Plast Reconstr Surg. 2018;141:709e717e.
6. Chism CB, Crawford L, Tchakarov A, Al-Ibraheemi A, Beckmann NM. PTEN hamartoma of the soft tissue: the initial manifestation of an underlying PTEN hamartoma tumor syndrome in a 4-year-old female. Skeletal Radiol. 2017;46:1591–1595.
7. Limaye N, Kangas J, Mendola A, et al. Somatic activating PIK3CA
mutations cause venous malformation. Am J Hum Genet. 2015;97:914–921.
8. Erickson J, McAuliffe W, Blennerhassett L, Halbert A. Fibroadipose vascular anomaly treated with sirolimus: Successful outcome in two patients. Pediatr Dermatol. 2017;34:e317–e320.