Congenital juvenile xanthogranulomas: Born to be wild (but ultimately mild)

“Lost to follow-up” has implications that go beyond the obvious. Following up with patients is an essential part of learning for physicians of all ages, especially when disease presentations are unusual. Years ago I was consulted on a neonate presenting with large, ulcerated plaques on the trunk and extremities — my clinical differential diagnosis was all over the map, ranging from bullous mastocytosis to a rhabdomyosarcoma. A biopsy revealed the diagnosis of a juvenile xanthogranuloma (JXG). I desperately wanted to follow this patient, but unfortunately, that never transpired. What becomes of such patients who present with such dramatic lesions at birth?

JXGs usually appear early in life, although they may occur in adults. Lesions may be solitary or multiple, and typically range in size from 0.5 to 2 cm, although giant lesions (up to 10 cm) have been described. JXGs are a non-Langerhans cell histiocytosis, unassociated with lipid abnormalities. Histologically, they are characterized by Touton giant cells; they are CD68+, but negative for S100 and CD1a. Cutaneous lesions are usually asymptomatic and most lesions spontaneously involute; however, some cases present at noncutaneous sites and can result in serious complications. (1)

Typically, lesions of JXG present at birth (20%) or during the first 6 months of life; they may persist or continue to erupt for years. (2) Congenital presentations with ulceration are rare. Logue et al reported a term 2-week-old girl with a solitary congenital mass on her left medial buttock. It had been present since birth without any appreciable changes in size, texture, or color, but developed a central painful ulceration. Adequate treatment was provided with topical petrolatum and occasional miconozole or zinc oxide; the mass spontaneously regressed. At one year, there was complete resolution of the lesion, with just a couple of residual sclerotic papules at the prior site of ulceration. (3)

Oza et al identified 31 cases of congenital JXGs (including 3 of their own cases) involving only the skin and 16 cases with systemic involvement. Congenital JXGs involving only the skin were large (> 3 cm), presented with various clinical morphologies, and showed signs of regression by 1 year of age. Atypical clinical presentations included exophytic tumors, infiltrative plaques, agminated plaques, and subcutaneous tumors. Complications included ulceration and anetodermic scarring. Infants with congenital JXGs who also had systemic involvement typically had multiple cutaneous tumors and hepatic involvement and showed signs of spontaneous regression independent of treatment. Two patients with systemic involvement died in infancy because of liver failure associated with multiorgan JXG involvement. The authors concluded that congenital JXGs behave similarly to that of typical JXGs of infancy or childhood. It is important to recognize congenital JXGs to avoid misdiagnosis, inappropriate examinations, and unnecessary treatments. Infants with multiple congenital cutaneous JXGs should be evaluated for systemic involvement, with a particular focus on the liver, because 72.2% of these children were found to have hepatic juvenile xanthogranulomas. (4)

Although this commentary focuses on congenital JXGs, it is worthwhile to consider some newer literature pertinent to JXGs in general.

The eye has been considered the most common organ of involvement second to the skin, but that may be debatable. The iris is the site most often involved, and potential complications of ocular JXG include hyphema, glaucoma, or blindness. Children at greatest risk for ocular JXG include those 2 years of age or younger and those with multiple skin lesions. (2)

Samuelov et al retrospectively reviewed ocular involvement and JXGS. Of 338 children with cutaneous JXGs: 76 (median age 6 months, 51% female) met inclusion criteria by having an ophthalmologic assessment. The most frequently involved site was the head and neck region (40%). In 39 patients (51%), there was a single lesion. Multiple lesions (>5) were evident in 20 patients (26%). Most cutaneous JXGs were micronodular (77%). None of the patients had ocular involvement. One patient had multiple asymptomatic hepatic nodules on imaging that regressed spontaneously within several months. Literature review of pediatric cutaneous JXGs, including their cohort, revealed that the incidence of ocular manifestations is 0.24% (7/2949) and of systemic manifestations is 0.75% (22/2949). The authors concluded that cutaneous JXGs are generally limited to the skin. Because eye involvement is rare, a routine eye examination is of low yield and probably not warranted in children with no ocular or visual symptoms. New recommendations for systemic screening could not be drawn from this study. (5)

I would be remiss in not mentioning the association of JXGs and childhood leukemia, most commonly with juvenile chronic myelogenous leukemia (JCML) and multiple JXGs. Several such reports were associated with café-au-lait macules and a family history of type 1 neurofibromatosis (NF). A systematic review of the literature revealed that the frequency of the triple association of JXG, JCML, and NF is 30- to 40-fold higher than expected; it is estimated that children with NF and JXG have a 20- to 32-fold higher risk for JCML than do patients with NF who do not have JXG. Regardless, the vast majority of patients with multiple JXGs, even those with associated NF, do not develop JCML (2) Liy-Wong, et al performed a retrospective study of retrospective case-control study comparing 739 children with NF-1 and malignancy with a control of sex- and age-matched children with NF-1 without malignancy. JXGs were found in 4/14 (28.5%) cases and 6/29 (21%) controls, a statistically insignificant finding, allowing the authors to conclude that JXGs do not appear to confer an increased risk for malignancy in children with NF-1. (6) Despite the rarity of the association of JCML, JXGs, and NF-1, I will still check a CBC in patients presenting with JXGs and café-au-lait macules.

The yellowish hue of JXG is accentuated by dermoscopy, and has been likened to a “setting sun”. (1) In 1787, on the final day of the Constitutional Convention, Benjamin Franklin pointed to George Washington’s chair and said: “I have often … in the course of the session … looked at that sun behind the President without being able to tell whether it was rising or setting. But now at length I have the happiness to know it is a rising and not a setting sun.” The latest data on JXGs appears to be good. Perhaps the dermoscopy sign should be considered a “rising sun”.

Point to remember: Congenital JXGs may present dramatically, but most will have an excellent prognosis.

1. Collie JS, Fillman EP. Xanthogranuloma, Juvenile (Nevoxanthoendothelioma, JXG). StatPearls [Internet]. Treasure Island (FL0: StatPearls Publishing 2018-2018 Aug 31.
2. Paller AS, Mancini AJ. Juvenile xanthogranuloma. In Hurwitz Clinical Pediatric Dermatology, 5th edition, Elsevier, 2015 (accessed via Expert Consult).
3. Logue ME, et al. Congenital juvenile xanthogranuloma with ulceration: A pediatric case report. Dermatol Online J 2017; 23 (7).
4. Oza VS, et al. Congenital-type juvenile xanthogranuloma: A case series and literature review. Pediatr Dermatol 2018; 35: 582-7.
5. Samuelov L, et al. Risk of intraocular and other extracutaneous involvement in patients with cutaneous juvenile xanthogranuloma. Pediatr Dermatol 2018; 35: 329-35.
6. Liy-Wong C, et al. The relationship between neurofibromatosis type 1, juvenile xanthogranuloma, and malignancy: A retrospective case-control study. J Am Acad Dermatol 2017; 76: 1084-87.