Cutaneous Leiomyosarcoma: Treatment and Outcomes With a Standardized Margin of Resection

Transcription

1 Good oncological control and excellent outcomes are possible in cases of cutaneous leiomyosarcoma. Tenzin Norbu Lama. Building the Temple. Cutaneous Leiomyosarcoma: Treatment and Outcomes With a Standardized Margin of Resection Jeremiah L. Deneve, DO, Jane L. Messina, MD, Marilyn M. Bui, MD, PhD, Suroosh S. Marzban, MS, G. Douglas Letson, MD, David Cheong, MD, Ricardo J. Gonzalez, MD, Vernon K. Sondak, MD, and Jonathan S. Zager, MD Background: Cutaneous leiomyosarcoma is primarily a low-grade malignancy that affects elderly male Caucasians. It is a rare dermal-based tumor for which treatment algorithms have been poorly defined. Methods: We retrospectively reviewed the use of a median -cm margin for resection to treat patients with cutaneous leiomyosarcoma referred for treatment between 2005 and 200. Results: Thirty-three patients with cutaneous leiomyosarcoma were treated. Of these, 76% were male, 97% were Caucasian (median age: 63.5 years), and 67% of tumors were located on the extremities. Preoperative staging was negative for distant metastasis in all patients. A majority of the tumors (88%) were low grade (median size:.3 cm). All of the tumors were positive for smooth-muscle actin. A total of 94% of patients underwent primary surgical resection with a median margin of cm. Final resection margin was negative in 97% of patients. Adjuvant radiotherapy was used in 5%. No metastatic spread or recurrences were present, and 00% of patients were alive at last follow-up (median: 5.5 months). Conclusions: Good oncological control and excellent outcomes are possible with a -cm resection margin in most cases of cutaneous leiomyosarcoma. From the Departments of Cutaneous Oncology (JLD, JLM, SSM, RJG, VKS, JSZ), Anatomic Pathology (JLM, MMB), and Sarcoma (MMB, GDL, DC) at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida. Dr Deneve is now with the Division of Surgical Oncology (JLD) at the University of Tennessee Health Science Center, Memphis, Tennessee. Submitted November 4, 202; accepted January 3, 203. Address correspondence to Jonathan S. Zager, MD, FACS, Department of Cutaneous Oncology, Moffitt Cancer Center, 2902 Magnolia Drive, Tampa, FL Jonathan.Zager@Moffitt.org No significant relationship exists between the authors and the companies/organizations whose products or services may be referenced in this article. Presented in part at the Society of Melanoma Research Annual Meeting in Tampa, Florida, November 9-3, 20. Introduction Cutaneous leiomyosarcoma (LMS) is a rare soft-tissue tumor believed to arise from the arrector pili muscle of the hair follicles and accounts for % to 3% of all soft-tissue sarcomas.,2 Although cutaneous LMS originates within the dermis of the skin, occasional extension into the subcutaneous tissue may be present. Histologically, LMS is characterized by a dermal-based proliferation of spindle cells with blunt-ended, cigar-shaped nuclei and increased mitotic figures. 3 Immunohistochemical staining using vimentin, smooth-muscle actin, and desmin often aids in the diagnosis of cutaneous LMS and excludes other cutaneous spindle-cell proliferations. 3,4 October 203, Vol. 20, No. 4 Cancer Control 307

2 The clinical behavior of cutaneous LMS is difficult to establish because most retrospective series do not separately distinguish these tumors from soft-tissue LMS. However, cutaneous LMS is thought to have an indolent course and a better prognosis than those of subcutaneous and deeper origins. 5 Cutaneous LMS has a risk of local recurrence as high as 40% in some earlier series. 6 Massi et al 7 reported a 6% recurrence for cutaneous LMS confined to the dermis but an 8.% incidence of recurrence for cutaneous LMS with a subcutaneous extension. Distant metastases are more commonly associated with subcutaneous LMS, whereas those with cutaneous LMS do not have the same risk of metastasis. Moreover, subcutaneous LMS, which is thought to arise from smooth muscle of the small vessels, may metastasize earlier and have a worse prognosis. 5,8 By contrast, findings from the Scandinavian Sarcoma Group, 9 which reviewed 225 patients with nonvisceral soft-tissue LMS, revealed no patients with strictly cutaneous LMS whose cancer metastasized or who died of tumor-related causes. Some treatment-related issues remain unresolved for cutaneous LMS. For instance, although many published series recommend wide excision, few specify the actual surgical margin used. A variety of margins have been described as appropriate for resection in the medical literature, including narrow excision, wide excision, and Mohs micrographic surgery. 0-4 However, no consensus currently exists with regard to the most appropriate margin of resection. We have found that standardization of the surgical margin of resection is useful when treating other rare soft-tissue tumors (eg, dermatofibrosarcoma protuberans). 5,6 This article explores our experience at a single institution in the diagnosis and management of cutaneous LMS using a median of cm for a margin of resection. Methods This study was an Institutional Review Board approved, single-institution, retrospective review of patients who underwent resection for cutaneous LMS at the Moffitt Cancer Center in Tampa, Florida, from 2005 to 200. Patients with insufficient data for analysis, LMS of a visceral origin (gastrointestinal, gynecological, or retroperitoneal), atypical fibroxanthoma, dermatofibrosarcoma protuberans, or other noncutaneous LMS were excluded. Patients with cutaneous LMS arising from the dermis were further analyzed. Those with cutaneous or dermal LMS with a subcutaneous extension were included in the study, while those with LMS originating in the subcutaneous tissues or from a blood vessel were excluded from analysis. Preoperative staging evaluation, method of biopsy, and treatment were determined from the electronic medical record or outside documents when patients were referred to the Moffitt Cancer Center for consultation or further treatment. The margin of resection was recorded as follows: < cm,.0 cm,..5 cm,.6 2 cm, or other as determined from the dictated operative note when available. Tumor size was recorded in centimeters with mitotic figures noted when present. The diagnosis of cutaneous LMS was verified by a sarcoma pathologist (M.M.B.) or a dermatopathologist (J.L.M.), and immunohistochemistry was performed when necessary to support or confirm the diagnosis of cutaneous LMS or exclude other entities. Smooth-muscle tumors with evidence of nuclear pleomorphism and a greater than mitosis/0 high-power field (HPF) were diagnosed as LMS. Tumors were evaluated for nuclear atypia (mild, moderate, or severe), French Fédération Nationale des Centres de Lutte Contre le Cancer grade (, 2, or 3), mitotic activity (count of the most mitotically active area in 0 consecutive HPFs, ie, 400), and presence (and/or percentage) of tumor necrosis. Immunohistochemical results and morphology were recorded with immunoreactivity to smooth-muscle actin and/or desmin as confirmation of smooth-muscle origin. Adjuvant treatment, when administered, was noted. Recurrence and length of follow-up were recorded in months from the date of original diagnosis. Follow-up status was defined as no evidence of disease, alive with disease, or dead of disease or other causes. Descriptive statistics were measured using SPSS 9.0 software (IBM Software, Cary, NC). Results Patient Demographics Of 33 patients who underwent evaluation and treatment for cutaneous LMS between 2005 and 200 (Table ), 32 (97%) were Caucasian and 25 (76%) were men. The median age of the patients was 63.5 years (range, 20 9 years). Primary tumor location involved the lower extremities in 5 patients (46%), the head and neck in 9 (27%), the upper extremities in 7 (2%), and the trunk in 2 (6%). Shave biopsy was the most frequent method of diagnosis, followed by excision and punch biopsy. Preoperative staging studies, which were performed at the discretion of the treating physician, included chest computed tomography (CT) in 4 patients (42%), diagnostic chest radiography in 3 (39%), and whole-body positron emission tomography (PET) imaging in 4 (2%), all of whom were negative for metastatic disease. No patient received radiation to the primary tumor site prior to surgical resection. Primary Tumor Characteristics All tumors were cutaneous or dermal LMS in origin. The median tumor size was.3 cm (range, cm), with a majority of tumors having low-grade 308 Cancer Control October 203, Vol. 20, No. 4

3 Table. Clinical Characteristics of 33 Patients: Results of a Single-Institution Study Variable No. of Patients (%) Age (yrs) 63.5 (range, 20 9) Caucasian 32 (97) Sex Female Male Location Head and neck Trunk Upper extremities Lower extremities Biopsy method Shave Excision Punch Preoperative staging method Chest radiography Computed tomography Positron emission tomography Magnetic resonance imaging of the brain Magnetic resonance imaging of the extremity 8 (24) 25 (76) 9 (27) 2 (6) 7 (2) 5 (46) 6 (49) (33) 6 (8) 3 (39) 4 (42) 4 (2) 5 (5) histology (88%; Figs and 2). The remaining were higher-grade tumors. Mitoses were recorded in 76% of patients (range, 58/0 HPF; Table 2). Immunohistochemistry was performed in 27 patients (82%). Smooth-muscle actin was positive in all tumors tested. Desmin was tested in 20 patients (6%), with expression present in 4 (42%). S00 was tested in 24 patients (73%) and was negative in 23 (96%; Fig 3). Primary Tumor Resection The definitive margin of resection used following referral to the Moffitt Cancer Center was cm in the majority of patients (22 of 33; 67%). One patient (3%) underwent resection with a.2-cm margin, 4 patients (2%) with a.5-cm margin, and 4 patients (2%) with a 2-cm surgical margin. Two patients (6%) underwent resection with a margin smaller than cm (Fig 4). One of the patients who underwent resection with an initial -cm margin required temporary regenerative tissue matrix coverage of the scalp defect. Once negative margins were confirmed, we performed a planned extirpation of the temporary biological dressing with definitive skin graft coverage. We have used this technique for other pathologies prior to definitive coverage while awaiting permanent pathological margin assessment in situations in which further resection may be required. 7 Two patients (6%) had a positive margin following initial resection using a -cm margin (Table 3). Table 2. Histological Characteristics of the Tumors in 33 Patients: Results of a Single-Institution Study Variable No. of Patients (%) Median tumor size, cm.30 (range, ) Fig. Cutaneous leiomyosarcoma showing fascicular growth of ovoid spindle cells with mild nuclear pleomorphism and mitotic activity (hematoxylin-eosin stain, 200). Tumor grade Low High 29 (88) 4 (2) Recorded mitotic activity 25 (76) Median no. of mitoses (range) 0 ( 58) Atypical mitotic figures 8 (24) Residual tumor after biopsy 20 (6) Fig 2. Cutaneous leiomyosarcoma demonstrating diffuse positivity for desmin ( 200). Re-excision Planned re-excision with regenerative tissue matrix Positive margin after initial excision Unknown resection margin at outside Final margin Negative Positive 32 (97) (3) October 203, Vol. 20, No. 4 Cancer Control 309

4 30 Immunohistochemistry Number of Cutaneous LMS Patients Negative Positive SM Actin Myosin Desmin Vimentin Caldesmon Pro-collagen CD0 CD3 CD34 CD68 Factor XII S00 Mart- HMB45 Melan-A CK34 Cytokeratin Cytokeratin-903 Pankeratin Fig 3. Immunohistochemical profile of the tumor. LMS = leiomyosarcoma, SM = smooth muscle. The first patient underwent re-excision to negative margins using an additional -cm margin, while the second patient had a -cm tumor located at the nasolabial fold. A -cm margin was used and the area was covered with a rotational cheek flap. On final pathological analysis, the nasal mucosal margin was microscopically positive. This patient did not undergo further resection but was instead treated with adjuvant radiation. Therefore, the final margin of resection was negative in 32 patients (97%). Adjuvant Therapy Adjuvant radiation was administered to 5 patients (5%) following resection. All of the patients received radiation therapy based on recommendations from a multidisciplinary tumor board. One patient received adjuvant radiation for the presence of a subcutane- 2% Margin of Resection 6% Table 3. Treatment Outcomes of 33 Patients: Results of a Single-Institution Study Variable No. of Patients (%) Surgical margin (cm) <.0 (0.6 and 0.4) Autologous skin graft Full thickness Split thickness Regenerative tissue matrix Adjuvant radiation Positive margin (initial/final) Multidisciplinary recommendation Cutaneous leiomyosarcoma with subcutaneous extension Adjuvant chemotherapy 0 Recurrence 0 2 (6) 22 (67) (3) 4 (2) 4 (2) 0 (30) 5 (5) 2 (6) 5 (5) 2 2 2% 3% 67% <.0 cm.0 cm. cm.5 cm 2.0 cm Complication Nonhealing wound Cellulitis Self-limiting facial asymmetry after resection Length of follow-up, mos (median) 5.5 (range, 2 49) Fig 4. Surgical margin of resection for cutaneous leiomyosarcoma. No evidence of disease at follow-up 33 (00) 30 Cancer Control October 203, Vol. 20, No. 4

5 ous extension of cutaneous LMS on final pathology. Two patients with positive margins following tumor resection (one of whom underwent re-excision) received adjuvant radiation therapy. Per recommendations from the multidisciplinary tumor board, 2 patients received radiation therapy, neither of whom had high-grade tumors nor positive margins. These two patients were treated earlier in the current series. No patient received adjuvant chemotherapy. Complications and Outcomes Complications were observed in 3 patients (9%). One patient with a nasolabial fold tumor underwent rotational flap advancement with a positive mucosal margin and then subsequently developed facial asymmetry following resection. It was self-limiting and resolved after 3 months. One patient developed cellulitis and responded to oral antibiotics. The remaining patient developed a nonhealing scalp wound that required serial office debridement that healed by secondary intention. No locoregional recurrences or distant metastases were present, and no deaths were observed during the treatment period (median follow-up, 5.5 months; range, 2 49 months). All patients were without disease at the last follow-up. Discussion Cutaneous soft-tissue sarcomas are rare tumors and account for fewer than % of all soft-tissue malignancies. 2,9,8 They most commonly affect Caucasian men, with a peak incidence in the 5th and 6th decades of life. 7,9 These tumors present as solitary, erythematous, or brownish well-circumscribed dermal nodules ranging in size from 0.5 to 3 cm. Because they are dermal-based tumors, they may appear fixed to the epidermis and ulceration may be present. 20 The lower extremities and head and neck locations are the most common sites of involvement as identified in the literature. 3,4,2 Preoperative staging with cross-sectional imaging (CT of the thorax, PET imaging, or both) appeared to add minimal clinical impact as more than one-half of the patients in this series underwent imaging. Imaging results for all patients were negative. The utility of preoperative staging for cutaneous LMS has not been previously described in the literature. Multiple institutions have helped describe the behavioral and outcome differences between cutaneous and subcutaneous LMS. 5,6,22 Pijpe et al 22 and others 23,24 identified that more superficial tumors may have a better prognosis than those with a deeper origin. In the Pijpe series, deeper LMS had a significantly larger tumor size, which may explain the worse outcome observed with the deeper tumors. Massi et al 7 noted the impact of tumor grade on outcome; the cases of cutaneous LMS that recurred or metastasized in their series were all high-grade tumors. Other prognostic variables, such as DNA ploidy, 25 Ki-67 proliferative index, 26,27 and mitotic figures,,4 have also been identified as markers of worse outcome. However, the optimal width and impact of the margin of resection on outcome are not as well defined. Moreover, the majority of publications concerning cutaneous LMS have made little or no comment regarding the margin of resection performed at the time of excision. Although negative margins are the optimal management for these tumors, no specific consensus statement exists regarding the preferred or most appropriate margin. Retrospective, single-center series and collective reviews have provided some insight into this matter. Bernstein and Roenigk 4 recommended a wide margin excision when reporting their experience with 34 patients whom they followed for more than 30 years. Others have suggested using a 3- to 5-cm margin of excision because of the potential risk of local recurrence and distant metastasis. 20 Narrow margin excision and Mohs micrographic surgery have also been used. 0,,3,28 Although several small case series using Mohs resection have been described, recurrences have been reported with long-term follow-up. Huether et al 2 reported a 4% incidence of recurrence when treating spindle-cell neoplasms of the skin (eg, cutaneous LMS) using the Mohs micrographic technique. Standardizing an appropriate margin of resection for cutaneous LMS should be a priority. For example, in an analysis of 84 cases of atypical intradermal smooth-muscle neoplasms, Kraft and Fletcher 9 identified margin status as the most important predictor of recurrence. In that series, 8 local recurrences (2%) were observed following the initial excision (2 with positive margins, with a close [< 0.2 cm] margin, and 5 with unknown margins). No recurrences were observed when the initial excisions were performed with negative margins. At the Moffitt Cancer Center, we attempted to standardize the surgical margin by taking at least a -cm margin at the time of resection. Early in the study period ( ), resection margins were more variable, ranging from 0.4 to 2 cm. Since 200, we have more consistently utilized a resection margin of cm for cutaneous LMS, including patients who undergo an initial excision at another facility with unknown margins and subsequent reexcision with a -cm margin following a referral to the Moffitt Cancer Center. Two of the 33 patients in this series had microscopic positive margins following a -cm resection. One patient underwent further re-excision with an additional -cm margin to obtain negative margins, and another patient received adjuvant radiation. Although the follow-up period was short for our study, no local recurrences were reported during the trial period. October 203, Vol. 20, No. 4 Cancer Control 3

6 The role of adjuvant therapy for cutaneous LMS has similarly remained nonstandardized and largely anecdotal. Most institutions reserve the use of adjuvant radiation, chemotherapy, or both for the development of recurrence or distant metastasis. 6,7 Local recurrence reports following excision of cutaneous LMS range from 6% to 40%. 6,7,2,9 Furthermore, many series have included both cutaneous and subcutaneous LMS in their descriptions, confounding this issue further and making the interpretation of these recurrence rate results less clear. Jensen et al 5 reported no incidences of recurrence after 5 years of follow-up for intradermal LMS, but 4% of those patients with subcutaneous LMS had tumors that metastasized, and these patients ultimately died of pulmonary metastases. Fields and Helwig 6 similarly reported no distant metastases for those with cutaneous LMS, whereas 4 of 2 patients with subcutaneous LMS eventually developed metastases and progressed to tumor-related death. We did not observe distant metastases in any patient, and no patient received adjuvant chemotherapy. We administered adjuvant radiation on an individualized basis, such as in situations in which positive margins were present on final pathology but limited options existed for further resection or those who had larger cutaneous LMS with an extensive subcutaneous extension. In situations of local recurrence development, some have suggested that simple re-excision is sufficient therapy. Kraft and Fletcher 9 noted that 0 of 8 tumors studied with local recurrence could be re-excised to negative margins without additional recurrence or metastasis. Because many of the patients treated in that series underwent initial excision at outside institutions and were referred to their center after the development of recurrence, they further surmised that appropriate local therapy is most likely sufficient therapy to prevent recurrence. Conclusions Cutaneous leiomyosarcoma is a low-grade malignancy predominantly affecting elderly male Caucasians in the lower extremities and the head and neck. Immunohistochemistry may be utilized to confirm cutaneous leiomyosarcoma or rule out other spindle-cell proliferations. Good oncological local control and favorable overall outcomes are possible with a standardized, -cm surgical margin of resection. The benefit of preoperative chest imaging and/or adjuvant radiation to the primary tumor site remains unknown within the setting of negative margins. The choice for adjuvant radiation should be individualized; therefore, we recommend a multidisciplinary discussion of the need for adjuvant radiation in patients with larger tumors, persistently positive margins following multiple resections, or those patients who have cutaneous leiomyosarcoma tumors with an extensive subcutaneous extension. References. S tout AP, Hill WT. Leiomyosarcoma of the superficial soft tissues. Cancer. 958;(4): Z ahm SH, Fraumeni JF Jr. The epidemiology of soft tissue sarcoma. Semin Oncol. 997;24(5): K addu S, Beham A, Cerroni L, et al. Cutaneous leiomyosarcoma. Am J Surg Pathol. 997;2(9): B ellezza G, Sidoni A, Cavaliere A, et al. Primary cutaneous leiomyosarcoma: a clinicopathological and immunohistochemical study of 7 cases. Int J Surg Pathol. 2004;2(): J ensen ML, Jensen OM, Michalski W, et al. Intradermal and subcutaneous leiomyosarcoma: a clinicopathological and immunohistochemical study of 4 cases. J Cutan Pathol. 996;23(5): F ields JP, Helwig EB. Leiomyosarcoma of the skin and subcutaneous tissue. Cancer. 98;47(): M assi D, Franchi A, Alos L, et al. Primary cutaneous leiomyosarcoma: clinicopathological analysis of 36 cases. Histopathology. 200;56(2): Farshid G, Pradhan M, Goldblum J, et al. Leiomyosarcoma of somatic soft tissues: a tumor of vascular origin with multivariate analysis of outcome in 42 cases. Am J Surg Pathol. 2002;26(): Svarvar C, Bohling T, Berlin O, et al. Clinical course of nonvisceral soft tissue leiomyosarcoma in 225 patients from the Scandinavian Sarcoma Group. Cancer. 2007;09(2): Humphreys TR, Finkelstein DH, Lee JB. Superficial leiomyosarcoma treated with Mohs micrographic surgery. Dermatol Surg. 2004;30(): Starling J III, Coldiron BM. Mohs micrographic surgery for the treatment of cutaneous leiomyosarcoma. J Am Acad Dermatol. 20;64(6): Huether MJ, Zitelli JA, Brodland DG. Mohs micrographic surgery for the treatment of spindle cell tumors of the skin. J Am Acad Dermatol. 200;44(4): Tsutsumida A, Yoshida T, Yamamoto Y, et al. Management of superficial leiomyosarcoma: a retrospective study of 0 cases. Plast Reconstr Surg. 2005;6(): Bernstein SC, Roenigk RK. Leiomyosarcoma of the skin: treatment of 34 cases. Dermatol Surg. 996;22(7): Zager JS, Bui MM, Farma JM, et al. Proper margins of excision in dermatofibrosarcoma protuberans: wide or narrow? Ann Surg Oncol. 2008; 5(9): Farma JM, Ammori JB, Zager JS, et al. Dermatofibrosarcoma protuberans: how wide should we resect? Ann Surg Oncol. 200;7(8): Deneve JL, Turaga KK, Marzban SS, et al. Single institution outcome experience using AlloDerm as temporary coverage or definitive reconstruction for cutaneous and soft tissue malignancy defects. Ann Surg. 203;79(5): Rouhani P, Fletcher CD, Devesa SS, et al. Cutaneous soft tissue sarcoma incidence patterns in the U.S.: an analysis of 2,4 cases. Cancer. 2008;3(3): Kraft S, Fletcher CD. Atypical intradermal smooth muscle neoplasms: clinicopathologic analysis of 84 cases and a reappraisal of cutaneous leiomyosarcoma. Am J Surg Pathol. 20;35(4): Fish FS. Soft tissue sarcomas in dermatology. Dermatol Surg. 996; 22(3): Annest NM, Grekin SJ, Stone MS, et al. Cutaneous leiomyosarcoma: a tumor of the head and neck. Dermatol Surg. 2007;33(5): Pijpe J, Broers GH, Plaat BE, et al. The relation between histological, tumor-biological and clinical parameters in deep and superficial leiomyosarcoma and leiomyoma. Sarcoma. 2002;6(3): Konomoto T, Fukuda T, Hayashi K, et al. Leiomyosarcoma in soft tissue: examination of p53 status and cell proliferating factors in different locations. Hum Pathol. 998;29(): Weiss SW, Goldblum JR. Enzinger and Weiss s Soft Tissue Tumors. 4th ed. St Louis, MO: Mosby; Alvegard TA, Berg NO, Baldetorp B, et al. Cellular DNA content and prognosis of high-grade soft tissue sarcoma: the Scandinavian Sarcoma Group experience. J Clin Oncol. 990;8(3): Ueda T, Aozasa K, Tsujimoto M, et al. Prognostic significance of Ki- 67 reactivity in soft tissue sarcomas. Cancer. 989;63(8): Huuhtanen RL, Blomqvist CP, Wiklund TA, et al. Comparison of the Ki-67 score and S-phase fraction as prognostic variables in soft-tissue sarcoma. Br J Cancer. 999;79(5-6): Iacobucci JJ, Stevenson TR, Swanson NA, et al. Cutaneous leiomyosarcoma. Ann Plastic Surg. 987;9(6): Cancer Control October 203, Vol. 20, No. 4