Small indeterminate superficial soft tissue masses: relationship between depth and histological grade

Abstract

Objective:

To determine whether the location of a small, indeterminate soft tissue mass within the subcutaneous compartment is related to its histological grade.

Methods:

All Sarcoma Service referrals over a 12 month period of small (<3 cm) superficial soft tissue masses, indeterminate by MRI evaluation which subsequently underwent primary excision biopsy were included. Lesions were categorised by their anatomical location in the subcutaneous compartment. Histopathological diagnoses were categorized according to ¹² WHO 2013. χ² statistical analysis was performed to determine the relationship between lesion depth and histological grade.

Results:

The study included 43 patients, mean age 42 years (range 15–71 years). Within the subcutaneous compartment, 16 lesions were categorized as superficial, 9 lesions central and 18 lesions deep, of which 9 were non-neoplastic, 29 benign, 1 intermediate-grade and 4 malignant. Location in the deep aspect of the subcutaneous compartment was associated with a higher risk of intermediate or malignant histology (p = 0.02).

Conclusion:

The location of a small, indeterminate soft tissue mass within the subcutaneous compartment may be an indicator of histological aggressiveness. Lesions in the deep subcutaneous compartment are more likely to be intermediate-grade/malignant lesions. Therefore, if considering excision biopsy as definitive treatment, a wider margin may be appropriate.

Advances in knowledge:

Small, indeterminate soft tissue masses can be aggressive and the anatomical depth within the subcutaneous tissue may be a potential indicator of histological aggressiveness.

Introduction

A large variety of lesions are referred to specialist musculoskeletal sarcoma centres as possible soft tissue sarcomas. Such masses have a wide differential diagnosis including benign neoplasms, malignant neoplasms (primary and metastatic) and non-neoplastic lesions. MRI is routinely used for diagnostic evaluation, local staging and biopsy planning ¹ in addition to determining subsequent surgical management. ^2–4 Many benign and non-neoplastic lesions such as lipomas, vascular malformations, nerve sheath tumors and dermoid cysts can be characterized well by MRI alone, ^5–8 but a significant proportion of soft tissue masses are indeterminate by MRI criteria.

According to the National Institute of Clinical Excellence guidelines, a soft tissue mass should be suspected as being malignant if large (>5 cms), growing or located deep to the fascia. ⁹ All such lesions should be referred early to a local Sarcoma Service. However, a significant proportion of soft tissue sarcomas arise superficial to the fascia. ¹⁰ Although MRI features that suggest a superficial lesion is more likely to be malignant have been previously reported, ¹¹ to our knowledge the relationship between the location of a soft tissue mass within the subcutaneous compartment and its likelihood of malignancy has not been thoroughly investigated, although involvement of the deep fascia is recognized in staging systems as a prognostic feature. ² Smaller superficial lesions can be difficult to biopsy using a needle and therefore often undergo excision biopsy, with lesions ≤3 cm considered appropriate. ² There is potential for the excision biopsy to be definitive treatment, if the surgical margins are adequate. However, this is at the risk of overtreating benign lesions.

The aim of the current study is to assess whether the anatomical location of small, indeterminate soft tissue lesions within the subcutaneous compartment can indicate the aggressiveness of a lesion, and therefore be helpful for pre-operative planning if clear margins are intended.

Methods

The study was passed by the local Research and Innovation Service with no requirement for informed patient consent.

All patients referred to a specialist Musculoskeletal Sarcoma Service in the year 2014 were identified from the new patient multidisciplinary team (MDT) meeting and retrospectively analyzed. Small (<3 cm) superficial soft tissue lesions considered indeterminate in nature by MRI criteria, which subsequently underwent primary surgical excision biopsy formed the study group. A lesion was considered indeterminate if an MR imaging diagnosis could not be suggested with confidence, such lesions typically being well-defined with intermediate T ₁ weighted turbo spin echo (T ₁W TSE), proton density weighted (PDW) and T ₂ weighted fast spin echo (T ₂W FSE) signal intensity (SI) characteristics and intermediate-to-increased SI on fat suppressed T ₂W FSE or short tau inversion recovery (STIR) sequences (Figures 1–3).

Figure 1.

(a) Axial PDW FSE and (b) sagittal T ₂W MR images of the elbow demonstrating a small soft tissue mass in the posterior subcutaneous tissues (white arrows) abutting the dermis and categorised as ‘superficial’. The subsequent histopathological diagnosis was of a glomus tumor. FSE,fast spin echo; PDW, proton density weighted; T ₂W, T ₂ weighted.

Figure 2.

(a) Axial T ₁W SE, (b) STIR and (c) sagittal T ₂W FSE MR images of the cervicothoracic region demonstrating a small superficial lesion (white arrow) centred in the posterior subcutaneous fat which was categorised as ‘central’. The subsequent histopathological diagnosis was of a spindle cell lipoma. FSE,fast spin echo; STIR, short tau inversion recovery; T ₁W, T ₁ weighted.

Figure 3.

(a) Axial T ₁W and (b) STIR MR images of the left lower leg demonstrating a small indeterminate superficial soft tissue mass (white arrows) which is abutting the deep fascia, which was categorised as ‘deep’. The subsequent histopathological diagnosis was of a MPNST. FSE,fast spin echo; MPNST, malignant peripheral nerve sheath tumor; STIR, short tauinversion recovery; T ₁W, T ₁weighted.

The anatomical location of each lesion within the subcutaneous compartment was categorized into superficial (involving skin or lying in the superficial third of the subcutaneous compartment) (Figure 1), central (lying within the middle third of the subcutaneous compartment without contacting the skin or deep fascia) (Figure 2), or deep (lying within the deep third of the subcutaneous compartment or in contact with deep fascia) (Figure 3). If lesions encroached on the boundaries of these categories, the region within which the lesion was centred was chosen. The assessment was performed by two dedicated Consultant Musculoskeletal Radiologists independently with consensus reading of any discrepancies. Lesions of the fingers and toes were excluded as it was not possible to categorise lesion depth in these locations. The histological diagnoses were categorized into non-neoplastic lesions, benign neoplasms, and intermediate-grade neoplasms and malignant neoplasms according to the 2013 World Health Organization classification of soft tissue tumors. ¹²

Statistical analysis was performed using the Fisher’s exact test. All statistical calculations were performed by a dedicated statistician using Stata v.13.1 (Statecorp, Texas) to assess the relationship between lesion depth and degree of histological aggressiveness. A p-value of </=0.05 was considered statistically significant.

Results

43 patients met the inclusion criteria, 24 males and 19 females with a mean age of 42 years (range 15–71 years). The maximal lesion dimension ranged from 0.6 to 3.0 cm (mean 1.5 cm). 23 lesions (53.5%) arose in the upper limb, 14 (32.5%) in the lower limb and 6 (14%) on the trunk.

When considering anatomical depth, 16 (37.2%) were categorized as superficial, 9 (20.9%) as central and 18 ((41.9%) as deep within the subcutaneous compartment (Table 1). Following primary excision biopsy, 9 lesions (20.9%) were diagnosed as non-neoplastic, 29 (67.4%) as benign neoplastic and 5 (11.6%) as intermediate-grade/malignant neoplasms. (Table 2).

Table 1.

Fisher’s exact test statistical analyses between the anatomical depth of the indeterminate superficial soft tissue lesions and their histopathological biopsy result categorization

	Anatomical depth of superficial indeterminate lesion
Biopsy result	Superficial (%)	Central (%)	Deep (%)	p-value
Analysis 1
Non-neoplastic	4 (25%)	1 (11%)	4 (22%)	0.08
Benign tumor	12 (75%)	8 (89%)	9 (50%)
Intermediate/malignant tumors	0 (0%)	0 (0%)	5 (28%)
Analysis 2
Non-neoplastic/Benign tumor	16 (100%)	9 (100%)	13 (72%)	0.02
Intermediate/Malignant tumor	0 (0%)	0 (0%)	5 (28%)

Table 2.

Histopathological diagnoses categorised into non-neoplastic, benign and intermediate/malignant lesions (in numerical order (number in parentheses) and alphabetical order if equal in number)

Non-neoplastic (9)	Benign (29)	Intermediate & malignant (5)
Endometriosis (2)	Pilomatrixoma (6)	Malignant melanoma (1)
Epidermal cyst (2)	Schwannoma (4)	Malignant peripheral nerve sheath tumor (1)
Benign squamoid cyst (1)	Angioleiomyoma (3)	Soft tissue metastasis (unknown GI origin) (1)
Foreign body abscess (from vegetable / plant material) (1)	Angiolipoma (2)	Soft tissue giant cell tumour (1)
Ganglion (1)	Benign fibrous histiocytoma (dermatofibroma) (2)	Spindle cell sarcoma (low-grade) (1)
Non-specific inflammation (1)	Hemangioma (2)
Thrombus (1)	Nodular fasciitis (2)
	AVM (1)
	Capillary hemagioma (1)
	Cutaneous leiomyoma (1)
	Fibroma of tendon sheath (1)
	Glomus tumor (1)
	Myopericytoma (1)
	Myxoma (1)
	Spindle cell lipoma (1)

AVM, arteriovenous malformation; GI, gastrointestinal.

χ² analysis comparing anatomical depth (superficial, central and deep) and histopathological diagnosis (non-neoplastic, benign and intermediate-grade/malignant) yielded a p-value of 0.08 (Table 1, analysis 1). However, secondary analysis when non-neoplastic and benign lesions were combined as a single group of non-aggressive lesions, yielded statistically significant results (p = 0.02) (Table 1, analysis 2).

Discussion

The differential diagnosis of a superficial soft tissue lesion is wide and can vary depending on its location in relation to the skin (epidermis and dermis) or the subcutaneous fat, as reviewed by Blacksin et al. ¹³ Hung et al ¹⁴ performed an ultrasound study of 714 superficial soft tissue masses which included 247 with histological diagnosis, of which only 11 (4.5%) were malignant. Of the benign lesions, 105 (42.5%) were lipomas, 30 (12.1%) vascular malformations, 30 (12.1%) epidermoid cysts, 16 (6.5%) nerve sheath tumors and 8 (3.2%) abscesses. Based on current MRI experience, it would be expected that these lesions could have been diagnosed purely on imaging features, leaving approximately 25% of cases which may have had indeterminate MRI appearances and would possibly have required percutaneous needle biopsy or primary excision biopsy for diagnosis. A recent study by Pham et al ¹⁵ described the differential diagnosis of small (<2 cm) MRI indeterminate soft tissue masses referred to a tertiary orthopaedic oncology clinic, finding 7 of 39 (18%) superficial tumours to be malignant. They found no significant difference between non-malignant and malignant lesions with regards relationship to vessels, tendon or fascia, or enhancement characteristics.

Current European Guidelines suggest that superficial soft tissue tumours less than 3 cm in maximal dimension can be treated with primary surgical excision, ² and a recent report has indicated that this is a safe practice when undertaken in a specialist sarcoma service. ¹⁶ However, the removal of small superficial lesions in a non-specialist centre is known to be associated with a high incidence of incomplete excision ¹⁷ which may require subsequent re-excision and/or radiotherapy, ¹⁸ and potentially increased morbidity as a result.

The MRI features of superficial soft tissue masses that suggest malignancy include size >5 cm, a lobular contour, fascial edema and skin involvement. ¹¹ The relationship between tumor and fascia is also relevant, lesions either crossing the fascia or forming an obtuse angle with the fascia being >6 times more likely to be malignant. ¹⁹ This finding was confirmed in a more recent study, which also showed that tumor size >5 cm was a significant indicator of malignancy. ²⁰ However, superficial sarcomas may be smaller than 5 cm, with a recent study indicating that up to 7% of indeterminate superficial soft tissue tumors measuring 3 cm or less in maximal dimension were malignant. ¹⁶ Maximal lesion size may also be a poor discriminator between benign, intermediate and malignant tumors, but it has been suggested that the ratio of axial to lateral diameter is a more reliable measurement. ²¹ Other findings have been described which may aid in the differentiation of benign and malignant soft tissue lesions. T ₂W hypointensity has been shown to be highly specific for a benign lesion. ²² Contrast enhancement patterns may be of value, with non- or uniform enhancement more commonly seen in benign tumors. ²³ However, the routine use of contrast enhancement may need to be reviewed in the light of current evidence of gadolinium deposition in the brain. ²⁴ The ability of diffusion-weighted imaging in the differentiation of benign and malignant soft tissue tumours has now been extensively investigated, with malignant lesions showing significantly reduced apparent diffusion coefficient values compared to benign lesions, ^22,25–27 and this has also been shown to be true for superficial soft tissue tumours. ²⁸

The current study suggests that in the setting of relatively small (<3 cm) indeterminate superficial soft tissue masses, the depth of the lesion within the superficial compartment may provide additional information as to its degree of aggressiveness, with lesions located deeper being more likely to represent an aggressive lesion. No intermediate-grade or malignant lesions were located in the superficial or central compartments, with these tumours representing 28% of all deeply located lesions. When the histopathological diagnoses are considered as three separate categories (Table 1, analysis 1) the p-value reached close to statistical significance, but when the categories were simplified to non-aggressive (non-neoplastic and benign neoplasm) compared to aggressive (intermediate-grade and malignant tumours) in the secondary analysis, the p-value was statistically significant at 0.02. (Table 1, analysis 2).

The current study is differentiated from a study by Galant et al ¹⁹ in several ways. The inclusion criteria were small (<3 cm) MRI indeterminate lesions, whereas they included lesions of all size and those which could be diagnosed by MRI evaluation alone. It is more likely that larger tumors included within their study would be malignant since greater size of superficial lesions is related to risk of malignancy. ¹¹ Also, the emphasis of their study was to look at the relationship of tumor to fascia and how this impacted upon diagnosis, while the current study looked at location within the subcutaneous compartment and did not include lesions that had invaded the fascia.

Therefore, as well as previously mentioned criteria, consideration should also be given to the depth of a small indeterminate lesion which is contained within the subcutaneous compartment prior to consideration of appropriate surgical excision technique.

A limitation of this study is the small number of sarcomas and overall aggressive lesions within the study group, but this reflects the relative rarity of such lesions as has been previously documented. ^14,16 It should also be noted that only lesions suspected to be sarcomas, and therefore referred to a specialist musculoskeletal sarcoma service have been included resulting in selection bias, with the assumption that lesions suspected of being primary dermal or epidermal tumors would have been referred to a skin MDT.

In conclusion, the results of the current study suggest that anatomical location of small superficial MRI indeterminate soft tissue lesions within the subcutaneous compartment may be a potential indicator of histological aggressiveness. Lesions in the deep subcutaneous compartment are more likely to be intermediate-grade/malignant lesions. Therefore, if considering primary excision biopsy as definitive treatment, a wider margin may be appropriate. Since the study was undertaken in the setting of a specialist musculoskeletal oncology service, we believe that results should only apply to similar services rather than supporting the management of such lesions within non-specialist centres.