Annals of Surgical Oncology

, Volume 16, Issue 6, pp 1642–1649

Desmoid Tumors of the Anterior Abdominal Wall: Results from a Monocentric Surgical Experience and Review of the Literature

Authors

    • Division of General and Laparoscopic SurgeryEuropean Institute of Oncology
  • Antonio Chiappa
    • Division of General and Laparoscopic SurgeryEuropean Institute of Oncology
  • Alessandro Testori
    • Division of Melanoma and Muscle-Cutaneous SarcomasEuropean Institute of Oncology
  • Giovanni Mazzarol
    • Division of PathologyEuropean Institute of Oncology
  • Roberto Biffi
    • Division of Abdomino-pelvic SurgeryEuropean Institute of Oncology
  • Stefano Martella
    • Division of Plastic and Reconstructive SurgeryEuropean Institute of Oncology
  • Ugo Pace
    • Division of General and Laparoscopic SurgeryEuropean Institute of Oncology
  • Javier Soteldo
    • Division of Melanoma and Muscle-Cutaneous SarcomasEuropean Institute of Oncology
  • Paolo Della Vigna
    • Division of RadiologyEuropean Institute of Oncology
  • Rosalba Lembo
    • Division of General and Laparoscopic SurgeryEuropean Institute of Oncology
  • Bruno Andreoni
    • Division of General and Laparoscopic SurgeryEuropean Institute of Oncology
Bone and Soft Tissue Sarcomas

DOI: 10.1245/s10434-009-0439-z

Cite this article as:
Bertani, E., Chiappa, A., Testori, A. et al. Ann Surg Oncol (2009) 16: 1642. doi:10.1245/s10434-009-0439-z

Abstract

Background

Desmoid tumor, also known as aggressive fibromatosis, is a rare soft tissue tumor. For those cases localized in the anterior abdominal wall, radical resection and reconstruction with a mesh is indicated. Because the rarity of the disease, randomized trials are lacking, but in reported retrospective series, it is clear that although it is considered a benign lesion, local recurrence is not uncommon.

Methods

We analyzed the records of 14 consecutive patients (3 men, 11 women, mean age 36 years, range 25–51 years) with desmoid tumor of the anterior abdominal wall treated at the European Institute of Oncology. The surgical strategy was the same in all cases: wide surgical excision and immediate plastic reconstruction with mesh after intraoperative confirmation by frozen sections of disease-free margins of >1 cm. We considered long-term outcomes by using the European Organization for the Research and Treatment of Cancer QLQ-C30 as an instrument to evaluate the overall quality of the treatment delivered to these patients.

Results

No immediate postoperative complication was registered, and no patient developed recurrence after a median follow-up period of 55 months. Two women experienced mesh bulging within 1 year after the operation. The long-term mean global health status registered was 97 out of 100.

Conclusions

Radical resection aided by intraoperative margin evaluation via frozen sections followed by immediate mesh reconstruction is a safe procedure and can provide definitive cure without functional limitations for patients with desmoid tumors of the anterior abdominal wall.

Desmoid tumor, also called aggressive fibromatosis, is an uncommon neoplasm of soft tissues caused by a monoclonal proliferation arising in musculoaponeurotic structures that locally infiltrate but lack metastatic potential.1 This rare entity accounts for <3% of soft tissue tumors, or 0.03% of all neoplasms, and is more common in women. Its peak incidence is in patients between 25 and 35 years of age, and most cases occur between puberty and 40 years of age.2 Many studies have shown that between 37 and 50% of desmoid tumors arise in the abdominal region.3 Abdominal desmoid tumors occur sporadically or in association with some familial syndromes, such as familial adenomatous polyposis (FAP).4 Mutations in either the adenomatous polyposis coli and beta-catenin genes are likely the driving force in the formation of these tumors.3,4 Because local recurrence rates of 25 to 65% have been reported, despite its being considered a benign lesion, treatment requires wide excision followed by complex abdominal wall reconstruction in some cases.2 We analyzed the data of 14 consecutive patients treated at the European Institute of Oncology over the last 10 years and performed a review of the literature concerning desmoid tumors of the anterior abdominal wall.

Patients and Methods

This is a retrospective single-institution study of 14 consecutive patients with desmoid tumors of the anterior abdominal wall treated between October 1999 and June 2008 at the European Institute of Oncology.

Patients Studied

Eleven patients were women (mean age 36 years, range 25–51 years) and three were men (mean age 47 years, range 16–57 years) (Table 1). In all these patients, FAP was excluded by rigid sigmoidoscopy. Among the 11 women, 10 had previous pregnancies, and 5 got pregnant within 2 years. Patients were studied with ultrasound (Fig. 1) and/or magnetic resonance imaging (MRI) (Fig. 2). All patients except one were studied by MRI, which revealed infiltration of the uterus by the tumor in one patient and of the iliac crest in another. Two patients (14%) were candidates for radical excision after nonradical removal of the desmoid tumor at another institutions, and two were referred with radiological diagnosis of recurrent disease and underwent surgery without biopsy. The remaining 10 patients underwent ultrasound-guided core biopsy (Gallini, Mantova, Italy). Nine patients had received a diagnosis of aggressive fibromatosis that had been histologically confirmed before surgery; in one patient, the biopsy was not diagnostic. No patient underwent any hormone manipulation or any course of radiochemotherapy before or after surgery.
Table 1

Patient characteristics

Patient no.

Age (y)

Sex

Primary or recurrent disease

Abdominal quadrant

Mode of diagnosis

Biopsy performed

Maximum size (cm)

DFM size (cm)

Organ involved

Reconstruction

Follow-up (mo)

QoLa

1

25

F

Primary

Right upper + lower

MRI

Yes

14

1.5

Vicryl + Marlex

106

100

2

36

F

Primary

Right upper

MRI

Yes

2

2

Marlex

87

100

3

31

F

Recurrent

Left lower

MRI

No

3

2

Uterus fundus

Vicryl + Marlex

81

97

4

51

F

Primary

Left upper + lower

MRI

Yes

3.2

2

Marlex

76

97

5

32

F

Primary

Left upper + lower

MRI

Yes

3.5

3

Vicryl + Marlex

64

97

6

43

F

Primary

Right lower

CT

Yes

4.2

4

Vicryl + Marlex

58

100

7

41

F

Recurrent

Left upper + lower

MRI

No

3.5

1.5

Vicryl + Marlex

57

97

8

59

M

Primary

Right upper + lower

MRI

No

4.8

2

Bard Composix

54

97

9

67

M

Primary

Lower right + left

MRI

Yes

7.0

1

Bard Composix

41

100

10

30

F

Primary

Right upper + lower

MRI

Yes

3.5

0

Iliac spine

Bard Composix

23

100

11

38

F

Primary

Right lower

MRI

Yes

2.6

2.5

Bard Composix

10

83

12

33

F

Primary

Right upper + lower

MRI

Yes

2.5

3

Bard Composix

9

91

13

16

M

Primary

Lower right + left

MRI

Yes

6.5

1.5

Bard Composix

9

100

14

34

F

Primary

Right upper + lower

MRI

No

5.5

1.5

Bard Composix

8

100

DFM disease-free margin, QoL quality of life, MRI magnetic resonance imaging, CT computed tomography

aGlobal health status score, 1–100

https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0439-z/MediaObjects/10434_2009_439_Fig1_HTML.jpg
Fig. 1

Transverse ultrasound scan shows a large nonhomogeneous hypoechoic mass located at the rectus abdominis muscle

https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0439-z/MediaObjects/10434_2009_439_Fig2_HTML.jpg
Fig. 2

Sagittal T1-weighted magnetic resonance image shows a nodular lesion of the rectus abdominis muscle with well-defined margins characterized by homogeneous contrast enhancement after contrast medium administration

Surgical Procedure

All operations were performed by a surgical team composed of a member of the permanent staff of the Division of General Surgery and a member of the Division of Plastic and Reconstructive Surgery of the European Institute of Oncology. All 14 patients underwent the same surgical procedure, which comprised a wide excision of the mass and removal of all gross disease, together with a normal-tissue rim of at least 1 cm whenever possible. The closest margin was intraoperatively confirmed to be free of disease by frozen section. Surgical clearance was obtained by including the excision of the oblique internal abdominal muscle and fascia or the rectum abdominal muscle in all cases. The peritoneal surface was sacrificed in all cases because it was invariably inseparable from the overlying muscle plane. In one case where the tumor involved the uterus, clearance was achieved by continuing the dissection to resect the fundus of the uterus. In another case, the dissection continued to the iliac crest. In two cases, the greater omentum was sutured to the margins of the defect to produce a neoperitoneum; reconstruction was done with one layer of Marlex mesh (Bard, Galway, Ireland). In five cases, two layers of mesh were used: the first by Vicryl (polyglactin; Ethicon, Somerville, NJ) to cover the peritoneal defect, and the superficial one by Marlex. In the remaining seven cases, Bard Composix mesh was placed after the greater omentum fixation. In these cases, the mesh was sutured drum tight to prevent the development of any prosthetic bulging overlapping the posterior abdominal fascia. In all cases, two crossing drains were positioned over the mesh and were removed after a mean time of 4.4 days (range 3–8 days). All operations were performed under antibiotic cover with third-generation cephalosporin, which was continued until the suction drains were removed. Low-molecular-weight heparin was administered starting from the first preoperative day and was continued for 1 month after surgery to prevent thromboembolic disease.

Pathologic Evaluation

A histologically confirmed diagnosis of aggressive fibromatosis (Fig. 3) was achieved in 12 cases; no microscopically evident tumor was found in the two patients who underwent radicalization after an initial marginal surgery performed at another institute. Tumor characteristics studied included site and size. Size was obtained by measuring the largest dimension of the tumor in the surgical specimen or by measurement of MRI images. Surgical-free margins on the definitive specimen were confirmed to be >1 cm in all but one case. Whenever deemed necessary, appropriate immunohistology was performed as an adjunct to the morphological evaluation of the neoplasia.
https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0439-z/MediaObjects/10434_2009_439_Fig3_HTML.jpg
Fig. 3

Loosely arranged mesenchymal neoplasia infiltrates the skeletal muscle fibers (right) (hematoxylin and eosin; original magnification, ×20)

Follow-Up

Disease-free survival (DFS) was recorded for all patients and was obtained by outpatient clinical appointments, with clinical and radiological assessment as indicated. Follow-up ranged from 11 to 108 months (median 55 months). No patient was lost to follow-up.

Quality of Life

Quality of life was measured by the European Organization for the Research and Treatment of Cancer (EORTC) QLQ-C30, developed by the EORTC study group. It is a frequently used (nationally and internationally) and validated 30-question cancer-specific health-related questionnaire.5 In this study, we analyzed the global quality-of-life score and the five functional scales (physical, role, cognitive, emotional, and social). Each item has four answer categories, as follows: 1 = not at all, 2 = a little, 3 = rather a lot, 4 = very much. Scores were transformed into a scale from 0 to 100 according to the manual, so that a higher global quality-of-life score and higher functional scores corresponded with better quality of life.6 The questionnaire was administered during the last follow-up visit.

Results

There were no perioperative complications. The median hospital stay was 6 days (range 4–10 days). At pathologic examination, one patient whose tumor reached the iliac crest showed microscopic margin infiltration. In the remaining 13 cases, surgical disease-free margins of >1 cm were histologically demonstrated on definitive staining, in all cases confirming the data of the intraoperative frozen section. The mean tumor diameter was 4.7 cm (range 2–14 cm). All but one patient had a single neoplastic nodule; in the remaining patient, who had recurrent disease at the time of referral, four nodules were pathologically confirmed. No patient experienced recurrence (Table 1). Two women developed mesh bulging 8 and 12 months, respectively, after operation. In one of these two cases, Marlex mesh was fixed to the iliac crest during the first surgical procedure.

The EORTC QLQ-C30 questionnaire was completed by all the patients during the last follow-up visit after a median time from operation of 55 months (range 11–108 months). The mean global health status, on a 0- to 100-point scale, was rated as 97 ± 5.9; for specific functional scales, results were as follows: physical functioning 93 ± 11.1, role functioning 89 ± 16.7, emotional functioning 87 ± 19.1, cognitive functioning 94 ± 8.3, and social functioning 93 ± 14.7.

Discussion

To our knowledge, the present study is the largest reported series of patients with abdominal wall fibromatosis homogeneously treated at a single institution without recurrence after a lengthy follow-up period. Surgical radicality was not achieved to the detriment of functional outcomes because none of the patients reported any marked change in their lifestyle or habits.

Because of the lack of randomized, controlled trials studying desmoid tumors, we have tried to extrapolate from the published series those cases with tumors localized to the abdominal wall to compare with our data (Table 2). In all but one previously reported series of desmoid tumors, at least one patient experienced tumor recurrence. In one small series of seven patients, one patient developed recurrent disease that was surgically resected; the patient was free of disease 6 years after surgery.7 Yet in a larger series of 23 patients treated at the same institution and published 5 years later, the local recurrence rate was even higher (3 of 23 patients) after a median follow-up of 30 months.8 In a small series of four patients with desmoid tumors of the anterior abdominal wall who underwent surgery after neoadjuvant therapy with preoperative doxorubicin and radiotherapy, no long-term recurrences were reported after a median follow-up time of approximately 6 years.9
Table 2

Published series of cases with tumors localized to the abdominal wall

Study

Monocentric or multicentric study

No. of patients with abdominal wall desmoid

Primary/recurrent disease (n)

FAP (%)

Biopsy performed, n (%)

Other organ involvement, n (%)

Closure with mesh, n (%)

Adjuvant treatments, n (%)

5-year local recurrence (%)

Median follow-up (y)

Heiskanen et al.19

Multicentric

12

12/0

100%

2 (17%)

0%

45%

Sutton et al.7

Monocentric

7

6/1

0%

6 (86%)

2 (28%)

7 (100%)

0%

17%

3.5

Ballo et al.14

Monocentric

18

33 (85%)

1 (6%)

24%a

9.4

Merchant et al.16

Monocentric

21

23%b

4.1

Stojadinovic et al.23

Monocentric

39

35/4

5%

9 (27%)

1 (3%)

8%

4.4

Baliski et al.9

Monocentric

4

3/1

4 (100%)c

0%

6

Gronchi et al.11

Monocentric

44

44%

11.2

Phillips et al.8

Monocentric

23

20/3

0%

17 (74%)

23 (100%)

0%

13%

2.5

Latchford et al.28

Monocentric

12

12/0

100%

100% NSAID

33%

5

Lev et al.12

Monocentric

30

10%a

5.2

Present series

Monocentric

14

12/2

0%

10 (71%)

21 (7%)

14 (100%)

0%

0%

4.6

FAP familial adenomatous polyposis, NSAID nonsteroidal anti-inflammatory drugs

aAbdominal and thoracic wall

bExtremity and trunk

cPreoperative adjuvant chemotherapy with doxorubicin + 3000 cGy radiotherapy

Recurrence is not uncommon after primary treatment of desmoid tumors and still remains a major problem. Most recurrences are usually observed within 3 years, and nearly all by 6 years.10 In our series, two patients had previously been treated elsewhere 6 and 16 months before, respectively. Both underwent radical surgery for tumor recurrence, and both were free of disease at least 4 years after the second operation. Age may affect the recurrence rate; although information is unclear, it may be more likely in younger patients with extra-abdominal desmoid tumors.10

A limitation of this study is the relatively small number of patients. However, many of the largest series in literature reported study populations with desmoid tumors not only of the anterior abdominal wall, but also of the extremities, girdles, and trunk.8,11,12 Moreover, these studies, as well as those by others, are biased by several factors, such as inclusion in the same series of both primary lesions and recurrent disease, and inclusion of patients who may have received different kinds of treatments.1315 In one series, we can extrapolate the existence of 44 patients with desmoid tumor of the anterior abdominal wall.11 The 5-year DFS rate for those with tumors to the wall of the trunk was 56% in this series. An interesting finding of the Istituto Nazionale dei Tumori study was that the presence of microscopic disease in one surgical margin proved to be relevant in patients with recurrent disease, but not in those with newly diagnosed disease. Others have made the same observations.2,10,16,17

Conversely, other authors identified positive margins at resection as the most important independent predictive factor of local recurrence.1315 In a study by the University of Texas, DFS was equivalent in margin-positive and margin-negative patients for a recent patient subset, in comparison to an earlier series at the same institution, where the DFS difference was evident.13 The authors concluded that it could be possible that margin positivity variably influenced subsequent adjuvant therapies chosen in different centers, or even within the same center. However, as outlined before, all these large studies did not show a subset analysis for patients with abdominal wall tumors. A strong point of our study is that even though the number of patients is small, all of them were treated at the same single institution according to the same surgical strategy.

Our opinion is that obtaining a wide disease-free margin is crucial for reducing the recurrence rate. Moreover, one-stage treatment with immediate mesh reconstruction is cost-effective and increases the chances of definitive cure, enhancing the patient’s perceived quality of treatment. However, mesh placement has to be performed in the presence of a microscopically negative margin, and for this reason, we routinely prepared frozen sections of the closest margin, widening the surgical dissection until the margins were pathologically confirmed to be free of disease and were >1 cm in size. This practice did not take into account any possible technical problems regarding abdominal wall reconstruction. Disease-free margins of >1 cm were achieved in all but one case, where the accuracy of diagnosis via frozen section was confirmed postoperatively by definitive pathologic staining. For these reasons, we strongly believe that the multidisciplinary management of our patients—here, involving plastic surgeons in all cases and a gynecologist in one case—could account for the absence of recurrence after a long follow-up period.

In one series, 51 abdominal wall tumors in FAP patients were resected with no mortality or marked morbidity, but the recurrence rate was 41%.18 In a study of 11 patients with FAP who developed abdominal wall desmoid tumors who underwent surgical resection, 5 experienced local recurrence, and 4 of them experienced a second recurrence after the excision of the first one.19 Complete clearance of the abdominal wall was achieved in eight desmoid tumors of that series that were operated on, and there was one case of complete spontaneous regression. Prosthetic mesh was used to repair the abdominal wall defect in only two patients, and musculocutaneous flap reconstruction was performed in one. This may lead to the apparent conclusion that insufficient margins were achieved during surgery because recurrences were observed even after radical excisions. The authors concluded that excisional surgery for desmoid tumors cannot be regarded as a great success, although it plays a central role in the treatment of abdominal wall desmoid tumors.

The findings of two recent studies seem contradictory regarding the effect of R1 surgery for desmoid tumors.20,21 In one case, obtaining disease-free margins does not seem to influence outcome with comparable recurrence rates after R0 and R1 resections.20 In such cases, the authors do not recommend re-excision, especially for pelvic girdle tumors, where excessive surgery risks functional impairment. In contrast, the other study showed a statistically significant deleterious impact of marginal resection in comparison to R0 and even to nonsurgical strategies.21 This is in relationship to tumor location, where abdominal wall desmoid tumors seem to have a favorable prognosis.

No patient with FAP was referred to our institute with an abdominal wall desmoid. However, it is unclear whether having a FAP could represent a risk factor for developing local recurrence for patients who underwent surgical excision of desmoid tumors localized within the anterior abdominal wall. Local recurrence was also reported to be high in patients who underwent surgery for sporadic desmoid tumors, although this recurrent disease was successfully resected at its appearance.8

Imaging is important for the diagnosis and treatment of these solid lesions. Ultrasound is useful for diagnosis because of its relatively low cost, because of its lack of radiation, and because it is fundamental for guiding biopsy. MRI best demonstrates the extent of lesions and works particularly well for assessing visceral involvement.22 Achieving preoperative diagnosis is important for planning the surgical procedure. In fact, not all soft tissue tumors of the abdominal wall are desmoid tumors, and sarcomas must be considered in the diagnosis and management of neoplasms of this site.23

Reconstruction of large abdominal wall defects is complicated and technically demanding.7,24,25 Abdominal wall reconstruction can be achieved by direct repair (with sutures), and by the use of synthetic material (meshes) or myocutaneous flaps when the defect is extensive. Our study reports only two long-term minor complications of the abdominal wall reconstruction with mesh: two women who experienced mesh bulging. The bulge, which resulted from the relaxation of the mesh inserts, could be an indication for surgery, mainly because of its resulting chronic pain syndrome, or because of the cosmetic problems it could cause.26 In our patients, surgery was not indicated because of the observed benefit of symptoms relief after some courses of physiotherapy. However, the treatment of these large defects of the anterior abdominal wall is similar to the one adopted for large incisional hernias.

Although no comment can be made from our data, there are claims for the efficacy of radiotherapy. However, it is rarely used for abdominal wall and intra-abdominal disease because of the risk of developing radiation enteritis.15,27,28 It has been used mainly for the treatment of extra-abdominal desmoid tumors and is associated with an observed reduction of local recurrence rates.

Anecdotal reports or small series have been published on treatments with anti-inflammatory drugs, antiestrogens, imatinib, interferon alfa, and vitamin C.2931 Cytotoxic chemotherapy is usually indicated after noncytotoxic therapy fails, or in patients with inoperable or unresectable tumors. Moreover, it may be used as neoadjuvant therapy.9,31

Questionnaires administered to patients revealed that all of them had a long-term subjective perception of wellness. To our knowledge, this issue has never been investigated for patients treated for desmoid tumors. A limitation of this study may be its lack of preoperative quality-of-life assessment. Nevertheless, the usefulness of a comparison analysis is questionable because of the high scores reported on the EORTC QLQ-C30 for our patients free of disease years after the operation and the small size of the group.

Although hard data on abdominal wall desmoid tumors are difficult to amass, given their uncommon nature, many other published series contend that recurrence is a major issue and that some molecular determinants are probably involved in its developement.32,33 In contrast, this study seems to support the idea that after excision to achieve a wide disease-free margin followed by mesh reconstruction, and within a framework of multidisciplinary management, recurrence of abdominal wall desmoid tumors should be unusual. This principle requires that clinicians perform difficult surgery, and accurate intraoperative frozen section margin assessment must be made. However, our results in terms of quality of life are favorable in these cases, which is important because this rare disease predominantly manifests in women of childbearing age.

Acknowledgment

We thank Valentina Lazzati for her contribution to this study.

Copyright information

© Society of Surgical Oncology 2009