Reviewers: John Han-Chih Chang, MD
Source: International Journal of Radiation Oncology Biology Physics, December 1998, Volume 42 (Number 5): pages 1007?1014
Desmoid tumors are aggressive fibromatoses. Local invasiveness characterizes this low grade tumor. They arise in skeletal muscles or their fascia. They are of connective tissue origin. Growth is relatively slow compared with most tumors, but aggressive accelerated growth is also seen, as this is a heterogeneous group. Desmoid tumors are almost always non-metastatic. Deeply infiltrating lesions can invade into surrounding muscle. The mainstay of treatment has always been surgery. Treatment aims to have a curative effect with a good functional and cosmetic outcome. The problem lies therein, because these tumors are very infiltrative by nature and potentially require a large volume of resection to obtain a negative margin (if using surgery alone). As an attempt to maintain functional and cosmetic outcome, many have tried to use radiation therapy (RT) as an adjunct for patients who have a positive margin, subtotal resection or unresectable disease.
In this article the authors from MD Anderson Cancer Center have reviewed their experience in 75 patients with desmoid tumors treated with RT. These patients were treated from 1965 to 1994. Median follow-up was 7.5 months.
Twenty-three had radical RT for gross unresectable or residual disease following subtotal resection. Fifty-two underwent gross total resection either followed by post-operative RT or pretreated with RT (7 patients). The distribution of tumor location was 15% head and neck, 40% trunk, 27% upper extremity and 19% lower extremity. RT was delivered with a cobalt-60 machine or megavoltage linear accelerator.
Overall the results of treatment on all patients demonstrated an actuarial freedom from relapse of 78% at 5 years and 75% at 10 years. Seventeen total relapses were seen with 14 in the field of RT and 3 in the edge or just beyond the field of RT. No significant factors were found to correlate with relapse rates on univariate analysis. Variety of salvage treatments yielded a salvage rate of 82% (14 patients). Uncorrected overall survival was 99%, 88% and 85% at 5, 10 and 15 years.
Of the radical RT patients actuarial relapse rate was only 31% at 5 years. Seven had a local relapse with a median time to relapse of 14 months and none after 5 years. In a univariate analysis, dosage was found to be significantly correlated with relapse rate: 60% relapsed if given doses of 50 Gy or less, while 23% relapsed with doses greater than 50 Gy. It was evident that doses above 56Gy were not any more beneficial in this population of patients. For the 16 that had no relapses, 6 had documented complete regression of the tumor over a median time of 12 months. Four had regression that was not complete, but very significant. Six also had prolonged freedom from disease, but length of time to complete regression was not known.
Post-operatively treated (RT) patients had a relapse rate of 18% and 23% at 5 and 10 years, respectively. Ten have relapsed and 7 were salvaged. In both an univariate and multivariate analysis, positive margins, dose and other factors were not significantly associated with outcome (ie. relapse).
The crude complication rate from RT was 17%, and the 5 year actuarial rate was 22%. The complication rate did significantly vary with dose: 5% at dose of 56Gy or less but increased to 30% at higher doses.
Historically, aggressive resection of desmoids yields a recurrence rate of approximately 25% - 35% in most surgical series. Recurrences are salvaged with more surgery ? RT. Taking into account margin status, surgery on average yields a 13% relapse rate with negative margins, but 52% with positive margins. After RT is added, the relapse rates are respectively 7% and 26%. RT as an adjunct to surgery appears to improve outcome.
RT has been investigated quite successfully. Also inoperable lesions have been treated with localized RT. A series from Duke University showed local control in 15 of 16 patients treated with primary RT. University of Florida had 100% local control in 6 primary lesions and 80% in 10 recurrent lesions with RT alone or after subtotal resection. Massachusetts General Hospital (Harvard) had similar results. The amount of disease, whether it is bulky gross disease or microscopic residual, does not appear to have a bearing on local control (75% ? 85%) for radiation therapy alone. This article seems to confirm the earlier findings in this disease.
Tumor responses are slow, with occurring with any 6 months after treatment. Response can be seen 1?2 years after treatment. The MD Anderson Cancer Center experience seems to support this with their median time to complete regression in their gross disease patients of 12 months.
Some other areas of pursuit in alternative treatment modalities are non-steroidal anti-inflammatories, hormonal treatment and chemotherapy.