Spontaneous Regression of Myxofibrosarcoma of the Thigh after Open Biopsy

Unexpected regression of cancer is a well-known phenomenon, albeit rare. Spontaneous regression of cancer has been defined as partial or complete disappearance of a malignant tumor without any kind of treatment [1]. Carcinomas account for majority of spontaneous regression reported to date but encountering such phenomenon in sarcomas is extremely rare. Although various mechanism of the remission has been reported to date, little is known behind this rare phenomenon. In this report, we present a rare case of spontaneous regression of myxofibrosarcoma associated with an open biopsy, with a review of the literature.

A 62-year-old man was referred to our clinic with a gradually enlarging mass from one year ago in his right thigh. On physical examination, there was a 10 cm non-tender mass on his anterior thigh. The overlying skin was taut and adherent with slight hotness. On X-ray, there was no obvious calcification or scalloping of the femur. MRI revealed an elongated 8.6 × 13.3 × 1.9 cm bland mass which showed low intensity on T1WI and high intensity on T2WI with a tail like sign in the subcutaneous tissue. Gadolinium administration showed uniform enhancement of the tumor without any peripheral inflammation or edema (Fig. 1). The laboratory data including white blood cells, neutrophils, and C-reactive protein were all in the normal range. On positron emission tomography – computed tomography (PET – CT), there was no accumulation of ¹⁸F – FDG besides the thigh. Differential diagnosis included benign lesion such as a nodular fasciitis and malignant tumors such as myxofibrosarcoma and undifferentiated pleomorphic sarcoma; therefore, an open biopsy was performed. On histological examination, the lesion was composed of spindle and pleomorphic tumor cells with atypical nuclei admixed with myxoid stroma (Fig. 2). On immunohistochemistry, the lesion was positive for vimentin, CD34 and Ki67 (21.5%), suggestive of myxofibrosarcoma. During the preoperative period for wide resection, repeat MRI was performed to assess the extent of the hemorrhage 21 days after the open biopsy. Although the pattern of signal intensities was the same, there was a significant decrease in size from 8.6 × 13.3 × 1.9 to 6.0 × 8.5 × 0.9 cm. Since we didn’t perform any medical treatments other than open biopsy, we determined it to be a spontaneous regression of the tumor; however, due to the unpredictable nature of the phenomenon, wide resection with skin graft was performed. At the time of the surgery 34 days after the open biopsy, the tumor has regressed even further where it was difficult to palpate the mass. Margin of resection was determined from the MRI of the initial onset, 3 cm from the edge of the tumor. On gross examination, resected specimen was tan white residing in the subcutaneous tissue measuring 3.7 × 2.5 × 1.2 cm. Histologically, over 95% of the residual lesion was composed of prominent fibrosis and granulation with myxomatous change by hematoxylin and eosin (HE) staining. There were abundant foamy cells, lymphocytes and plasma cell proliferation around small vessels, and sporadic atypical cells, most likely degenerated tumor cells, were present in a very small part of the lesion (Fig. 3). Immunohistochemically, CD68, vimentin and Ki67 (5.4%) were positive. Based on these histological and radiographic findings, spontaneous regression of myxofibrosarcoma was chosen as the final diagnosis. The postoperative period was uneventful, and at the final follow-up after one year postoperatively, no local recurrence or metastasis has been observed.

Myxofibrosarcoma is a subtype of soft tissue sarcoma characterized by its infiltrative growth often seen in elderly patients. It was first described in 1977 as a myxoid variant of malignant fibrous histiocytoma [2], and subsequently, World Health Organization reclassified the entity as a distinct tumor in light of histological and immunohistochemical findings in 2002. It has a predilection to occur in the lower extremity of a male and is considered a rare tumor representing approximately 5% of all soft tissue tumor diagnosis. Surgical resection is recommended for myxofibrosarcoma due to its insensitivity to chemotherapy and radiation. Although the prognosis of myxofibrosarcoma is relatively good with a 5-year survival rate reported to be approximately 60%, it has a propensity to recur locally [3].

Spontaneous regression of cancer has been reported to occur at a frequency of approximately 1 in 60,000–100,000 [4]. It has been postulated that the frequency of natural regression of cancer may differ depending on race and histology. Reported number of spontaneous regressions varies widely depending on the type of cancer, with large number of reports in renal cell cancer, malignant melanoma and lymphoma [5]. Although the mechanism behind spontaneous regression is unknown, it has been hypothesized as due to influence of hormone, involvement of immune modulation, trauma, blood flow disorder, infection and emergence of paraneoplastic syndrome [6, 7]. Among the various possibilities, enhancement of the immune system has been suggested to play an important role in both the progression and regression of cancers. Our pathological results showing multiple invasion of lymphocytes, neutrophils and plasma cells also suggest the presence of some kinds of immunologic response. According to Jessy ‘s report, by accelerating the immune system, antigen presenting cells such as dendritic cells are activated, thereby increasing the recognition of cancer cells and pathogens. Furthermore, lymphocytes and NK cells efficiently approach cancer cells due to hyperpermeability to cancer cells which leads to spontaneous regression of cancer [6]. The cause of acceleration of the immune system is thought to be various, and in our case, open biopsy seems to be the only cause that could have triggered this phenomenon.

It is difficult to predict the frequency of spontaneous regression in sarcomas. In Challis’s report, among the 504 cases with spontaneous regressions in malignant tumor, only 5 cases were sarcomas [5]. To date, there have been 17 reports of spontaneous regression of sarcomas: 4 osteosarcomas, 4 angiosarcomas, 2 fibrosarcomas, 2 leiomyosarcomas, and one each in alveolar soft tissue sarcoma, endometrial stromal sarcoma, synovial sarcoma, extraskeletal myxoid chondrosarcoma and myxofibrosarcoma [8-14]. Of the 17 cases, 10 regressions occurred in the metastatic site, and 7 in the primary lesion. Complete disappearance of the tumor was reported in 4 metastases and 3 primary lesions.

The treatment strategy is controversial regarding spontaneously regressed cases of the primary lesion. Of the seven cases in whom regression of the primary tumor was observed, 4 cases were treated by surgery and 3 cases were followed up with strict observation. Surgery was withheld from 3 cases due to multiple distant metastases in one case and patient requests in 2 cases. In 2 out of 4 cases where surgery was selected, it was because of regrowth after spontaneous regression. Furthermore, according to Cole, only 22 out of the 176 cancer regressions lasted for more than 10 years, and majority of the patients succumbed to their disease after partially or temporally suspended tumor growth [15]. Although common factors in these cases are difficult to ascertain given the limited number and various histology, re-growth and metastasis are often seen after spontaneous regression; therefore, implementing surgical resection with adequate margin and standard follow-up protocol for each histologically different sarcoma should be considered to ensure cure. Because there was no distant metastasis in our case, wide resection was performed for a definitive treatment. The extent of the resection margin is controversial especially in regressed cases, but due to its uncertain nature, we determined the margin with reference to the maximum diameter of the tumor from the initial MRI.

In conclusion, a rare spontaneous regression of myxofibrosarcoma in the thigh was observed in a 62-year-old male. Although there has been no local recurrence after wide resection, the rarity of this phenomenon has not enabled the prediction of the outcome, therefore further follow-up is warranted.

The authors have no ethical conflicts to disclose. Informed consent was obtained from the patient for this case report and any accompanying images.

The authors have no conflicts of interest to declare.