A clinicopathologic study of 45 pediatric soft tissue tumors with an admixture of adipose tissue and fibroblastic elements, and a proposal for classification as lipofibromatosis

The tumor described here as lipofibromatosis is a rare pediatric neoplasm t hat has been variously interpreted as a type of infantile or juvenile fibro matosis, a variant of fibrous hamartoma of infancy, and a fibrosing lipobla stoma. This report details the clinicopathologic features associated with 4 5 cases of this soft tissue entity. The study group consisted of 32 males, 12 females, and one person of unstated gender. The patients presented with a soft tissue mass (range, 1-7 cm) involving the hand (n = 18), arm (n = 8) , leg (n = 7), foot (n = 6), trunk (n = 5, or head (n = 1). Eight tumors we re evident at birth. The individuals ranged in age from 11 days to 12 years (median age, 1 yr) at the time of initial biopsy or resection. Microscopic examination revealed abundant adipose tissue with a spindled fibroblastic element that chiefly involved the septa of fat and skeletal muscle. The pro cess generally did not cause extensive architectural effacement of fat as i s common with conventional fibromatoses, and it did not have a primitive no dular fibromyxoid component as is characteristic of fibrous hamartoma of in fancy. The fibroblastic element exhibited focal fascicular growth and typic ally had limited mitotic activity (less than or equal to1 mitosis/10 high-p ower fields) and cytologic atypia. Oftentimes, small collections of univacu olated cells were present at the interface between some of the fibroblastic fascicles and the mature adipocytes. The tumors entrapped vessels (n = 45) , nerves (n = 44), skin adnexa (n = 16), and skeletal muscle (n = 18). Foca l immunoreactivity was present in some tumors for CD99, CD34, alpha -smooth muscle actin, BCL-2, and less frequently, S-100 protein, muscle actin (HUC 1-1), and EMA. However, no reactivity was detected for desmin (D33 and D-E R-11 clones), keratins, or CD57. Follow-up data were available for 25 indiv iduals (median follow-up period, 6 yrs 7 mos) with regrowth of the tumor or persistent disease documented in 17 (72%). The following events were more common in the group with recurrent or persistent disease: congenital onset, male sex, hand and foot location, incomplete excision, and mitotic activit y in the fibroblastic element. Although it is likely this tumor comprises p art of the spectrum of what has been referred to in the literature as infan tile/juvenile fibromatosis, its clinicopathologic features and, in particul ar, its distinctive tendency to contain fat as an integral component, warra nt separate classification as a ''Lipofibromatosis."