Desmoid-type fibromatosis of the head and neck in children: a case report and review of the literature

We performed a review of the literature by searching PubMed and using the keywords “desmoid,” “fibromatosis,” and “pediatric.” We found 97 articles from 1982 to 2015 when we searched cross-sectionally. The established exclusion criteria were as follows: details in individual cases not described, patients who were >19-years old, reports not written in English, and tumor sites not in the head and neck. We examined the following factors in all cases: age, sex, tumor site, tumor size, treatment, margin status, recurrence, complication, disease condition, and follow-up duration. The descriptive terms of incomplete resection (MIR), residual tumor, positive, and CR, which relate to margin status, were defined as follows: MIR, microscopic positive margin but no remaining gross tumor; residual tumor, remaining gross tumor; positive, positive resection margin, but we could not determine if it was MIR or a residual tumor; CR, negative microscopic margin; and NR, data not shown. The cases were reviewed in detail and are discussed in this case report.

We identified 141 patients with DF in the head and neck [7, 8, 13‐72]. Age at diagnosis ranged from birth to 18 years, and the mean age was 4 years 3 months. The sex of the patient was reported in 92 % of all cases (67 males and 63 females). Regarding the tumor sites, a majority of tumors arose from the mandible (25 %). Other sites were as follows: submandibular area, infratemporal fossa, neck, peritracheal area, and paraspinal area. A list of tumor sites is presented in Table 1. We were able to identify the tumor size in 78 (55 %) patients. The average tumor size was 43.6 mm (range, 5 to 100 mm). Surgery was performed in 125 (88.6 %) patients during the treatment period, and the remaining patients were treated only with chemotherapy (12 patients), the watch and wait strategy (3 patients), or a combination of radiation and chemotherapy (1 patient). Margin status was identified in 78 (62.4 %) of the 125 patients. The number of patients who had MIR, residual tumors, positive status, and CRs was 24, 19, 19, and 16, respectively. Recurrence was observed in 34 (27.2 %) of the 125 patients with surgery. The most common margin status was residual, which was observed in 14 patients. The number of patients who had positive status, MIR, CR, and NR was 7, 8, 1, and 4, respectively (Fig. 5). Postoperative complications were identified in detail in 13 patients. Trismus was the most common complication (n=6). Two patients experienced secondary papillary carcinoma due to radiation therapy. Osteomyelitis (n=2), slight ptosis due to a sectioned facial nerve (n=1), restrictive neck movement (n=1), and Claude–Bernard–Horner syndrome (n=1) were confirmed. The disease condition was identified in 126 of the 140 patients (Fig. 6). Details were not reported for 15 patients. Ninety-seven patients showed no evidence of disease (NED). Stable disease (SD), partial response (PR), progressive disease (PD), and death occurred in 15, 6, 5, and 3 patients, respectively. The average follow-up duration was 53.3 months (range, 2 to 298 months).

The most common tumor site in the recurrent patients was the cervical paraspinal region (seven patients). The next most common tumor site was the neck (five patients). Other sites we could identify were as follows: peritracheal area, infratemporal fossa, submandibular area, and tongue. All tumor sites in the recurrent patients are shown in Table 1. The recurrence rates for each margin status were MIR, 7 (29.1 %) of 24; residual, 14 (73.6 %) of 19; positive, 8 (42.1 %) of 19; and CR, 1 (6.2 %) of 16 patients (Fig. 7). There were 29 patients with positive margins without recurrence; “positive” status as used here includes MIR, residual, and positive status. Adjuvant therapies for these patients were as follows. In the case of MIR without recurrence (n=15), observation (n=9), chemotherapy (n=4), and radiation (n=2) were performed. In the case of residual tumors without recurrence (n=4), all patients spontaneously regressed. In positive patients without recurrence (n=10), observation (n=3), only chemotherapy (n=5), and administration of NSAIDs plus chemotherapy (n=2) were performed.

In recurrent patients, the relationship between margin status (n=31) and disease condition was confirmed as follows (Fig. 8). In the case of MIR (n=7), six (85.7 %) of seven patients had NED because second surgeries had been performed. One of the seven patients with SD was treated only with chemotherapy. In the case of residual tumors (n=14), there were two patients with NED, seven with SD, two with PR, two with PD, and one who died. Only 14.2 % of the patients showed controlled NED. In the case of positive status (n=9), there were two patients with NED, five with SD, one with PD, and one who died. There was NED in 11.1 % of the patients. The patient having CR with recurrence (n=1) had controlled NED. The patients having MIR (85.7 %) and CR (100 %) with recurrence had high rates of NED. It is notable that all recurrent patients with controlled NED were rescued by second surgery.

The number of patients for each disease condition in all recurrent patients (n=34) with surgery was 14 with NED, 13 with SD, 2 with PR, 3 with PD, and 2 who died (Fig. 5). Of all recurrent patients with surgery, 41.1 % had controlled NED. Consequently, in all cases in which surgery was performed, 91 (72.8 %) patients had controlled NED. The mean follow-up duration for patients with surgery was 53.3 months. The result of our literature review was summarized in Table 2.

The main finding of our study was that the tumor control rate tended to be high in cases of MIR. Moreover, there were 29 margin-positive patients who were without recurrence. A majority of these patients had MIR (n=15), and 15 (62.5 %) of 24 patients with MIR did not show recurrence. In our view, these results suggest that a finding of MIR could be acceptable during local aggressive DF treatment in children. A paper on EORTC’s position on DF in all populations [5] proposed that if positive microscopic surgical margins were found at a pathological examination, no further treatment should be considered. Our review results supported this position in pediatric patients with head and neck DF. Wide resection may induce long-term functional morbidity, especially in children, and may require reconstruction. Even though microvascular reconstruction is useful in pediatric patients for well-experienced surgeons [73, 74], the decision to proceed with wide resection should be carefully considered; the results of this study may be useful when making decisions in such cases.

If progression occurs after the watch and wait strategy has been pursued, medical therapy is clearly recommended in patients with head and neck DF according to the paper on EORTC’s position [5]. However, a standard treatment is yet to be established. Previous prospective studies in pediatric patients with DF, including a phase II trial of vinblastine plus methotrexate and tamoxifen plus NSAIDs, have reported that the CR and PR rates were 19.2 % and 8 %, respectively [75, 76]. Even when CRs are included, the total response rate is insufficient. The restrictive effectiveness of chemotherapy regimens, such as liposomal doxorubicin, anthracycline-based regimen, imatinib, and sorafenib, in the total population has been confirmed [77‐80]. However, no high-grade evidence study (that is, a randomized phase III trial) has been reported. In the present study, in only one patient, chemotherapy using Adriamycin (doxorubicin) and dacarbazine achieved CR. Furthermore, in the recurrent patients, there were none with controlled NED who had been treated only with subsequent chemotherapy. They were rescued by second surgery. Therefore, at present, the therapeutic effect of chemotherapy may be limited in pediatric DF treatment. A randomized trial would be needed to confirm the efficacy of chemotherapy in pediatric DF treatment.

Although previous reports have indicated that treatments with surgery and radiotherapy in all populations improve the progression-free survival rate relative to that of surgery alone [3, 81], the use of radiotherapy during DF treatment is controversial. However, in pediatric patients, radiotherapy may be less available because of late adverse effects and lower effectiveness. Side effects including secondary carcinoma related to radiotherapy for DF in pediatric and young-adult patients have been reported [53, 82]. A medium-sized retrospective study involving 30 patients in a single institution who were treated with radiotherapy for pediatric and young-adult DF reported a lower regional control rate for patients who were <18-years old (20 %) than for those who were 18 to 30-years old (63 %) [53]. In the present study, 13 patients received radiotherapy, and complications were identified in three patients: two patients developed secondary papillary thyroid cancer after total doses of 55 Gy and 50 to 60 Gy (not specified), and one developed osteoradionecrosis after a total dose of 55 Gy. There is no clearly standardized evidence-based radiation strategy for pediatric head and neck DF. Some refractory patients may need radiation therapy. However, it may not be reasonable to routinely consider radiotherapy for pediatric patients with DF having positive margins.