Abstract
Purpose
This study aimed to evaluate shear-wave elastography (SWE)–derived stiffness values and key histopathologic features of soft-tissue sarcomas and to assess the potential diagnostic value of SWE for differentiating sarcomas from non-sarcoma soft-tissue lesions.
Methods
This retrospective study included 49 consecutive patients with clinically suspected soft-tissue tumors who underwent conventional ultrasonography (US) and SWE between April 2024 and March 2025. Quantitative SWE parameters, including mean and maximum stiffness values in kilopascals, were measured from at least four acquisitions for each lesion. Histopathologic grade was determined using the Fédération Nationale des Centres de Lutte Contre le Cancer (FNCLCC) system, and key histopathologic features relevant to SWE interpretation were assessed, including tumor necrosis, mitotic count, cellularity, and extent of myxoid stroma. Associations between SWE-derived stiffness values and histopathologic features were evaluated as the primary analysis, and diagnostic performance for sarcoma discrimination was evaluated as a secondary analysis. A P<0.05 was considered to indicate statistical significance.
Results
Microscopic necrosis was the only histopathologic feature significantly associated with SWE-derived stiffness; sarcomas without necrosis showed higher stiffness values than those with necrosis (P<0.05). SWE stiffness values showed a nonlinear relationship with FNCLCC grade, increasing from grade 1 to grade 2 and then decreasing in grade 3 sarcomas. Conventional US showed high sensitivity (84.4%) but low specificity (35.3%) for sarcoma discrimination, whereas SWE-derived stiffness showed low sensitivity (21.9%–31.3%) but relatively high specificity (82.4%–88.2%).
Conclusion
SWE-derived stiffness appears to be affected by intratumoral heterogeneity, particularly microscopic necrosis, which may help explain its nonlinear relationship with histologic grade. Although SWE alone showed limited sensitivity for sarcoma discrimination, its relatively high specificity may complement conventional US when markedly elevated stiffness is present.
Keywords: Soft-tissue neoplasms, Ultrasonography, Elasticity imaging techniques, Shear-wave elastography, Histopathology
Graphical abstract
Introduction
Accurate differentiation between benign and malignant soft-tissue tumors is essential. Although soft-tissue sarcomas account for less than 1% of all cancers [1], missed diagnosis followed by unplanned excision is associated with markedly higher local recurrence rates, reported at approximately 6%–11% versus 34%–46% [2–4]. Multiple imaging modalities, particularly ultrasonography (US) and magnetic resonance imaging (MRI), are therefore used for initial lesion characterization. Nevertheless, substantial overlap in imaging features persists, making consistent preoperative differentiation of soft-tissue sarcomas from benign mimickers challenging.
Shear-wave elastography (SWE) provides quantitative measurements of tissue stiffness and has been evaluated as an adjunct to conventional US in multiple clinical settings. In soft-tissue masses, SWE has been investigated as a complementary method for differentiating benign from malignant lesions [5–9]. Although sarcomas may be expected to be stiffer than benign lesions because of higher cellularity and collagen-rich stroma [10,11], studies of SWE have reported inconsistent findings, ranging from minimal added benefit [6] to improved characterization when stiffness data were incorporated [7]. Diagnostic performance may also be affected by lesion depth and patient age, and even with optimized stiffness thresholds, the sensitivity of SWE for malignancy has been reported to be as low as 46% [12]. These findings indicate that SWE alone may miss a substantial proportion of sarcomas.
Given these limitations, the clinical application of SWE as a standalone triage method remains constrained, particularly compared with its more established role in carcinomas, for which its diagnostic relevance has been more extensively studied. Further validation in dedicated sarcoma cohorts and refinement of interpretation strategies are therefore needed. In this context, the present study focused on histologic features specific to sarcomas, particularly the heterogeneous stroma that is critical for definitive pathologic diagnosis [13], as a basis for tailoring SWE interpretation. Unlike carcinomas, which form cohesive nests of epithelial cells within a relatively uniform host-derived stroma, sarcomas consist of disorganized proliferations of mesenchymal cells embedded in a tumor-secreted extracellular matrix [14]. Because this intrinsic matrix heterogeneity may lead to variable stiffness patterns on SWE, this study evaluated whether subtype-specific histologic context was associated with elastography measurements.
Therefore, the primary objective of this study was to evaluate associations between SWE-derived stiffness values and key histopathologic features of soft-tissue sarcomas, including microscopic necrosis, mitotic activity, cellularity, extent of myxoid stroma, and overall Fédération Nationale des Centres de Lutte Contre le Cancer (FNCLCC) histologic grade. The secondary objective was to assess the potential diagnostic contribution of SWE for differentiating sarcomas from non-sarcoma soft-tissue lesions.
Materials and Methods
Compliance with Ethical Standards
This retrospective study was approved by the Institutional Review Board of Yonsei University Severance Hospital (IRB No. 4-2025-0767), and the requirement for informed consent was waived. The study followed the Declaration of Helsinki and the Health Insurance Portability and Accountability Act.
Patient Population and Study Design
Between April 2024 and March 2025, 85 consecutive patients with suspected soft-tissue tumors were assessed for eligibility. These patients were referred to the tertiary sarcoma center where the authors are affiliated for tissue diagnosis based on one or more established high-risk clinical features, such as increasing size, size >5 cm, deep location, or pain, and/or suspicious preoperative MRI findings prompting consideration of biopsy. The inclusion criteria were as follows: (1) completion of both B-mode US and SWE, (2) no prior treatment of the target lesion, (3) availability of preoperative contrast-enhanced MRI, and (4) histopathologic confirmation by core-needle biopsy or surgical excision.
The exclusion criteria were as follows: (1) lesions other than primary soft-tissue masses (n=25: metastasis, n=14; diffuse large B-cell lymphoma, n=3; T-cell lymphoma, n=1; polymorphic lymphoplasmacytic infiltration, n=1; reactive lymphoid follicular hyperplasia, n=1; proliferative myositis, n=1; focal vascular proliferation, n=1; hemorrhage, n=1; synovial cyst, n=1; cartilage lesion, n=1); (2) SWE studies with a reliable measurement index (RMI) <0.6 (n=8), indicating insufficient measurement reliability, including very superficial or deep lesions; and (3) cases in which the pathology specimen was insufficient or the results were inconclusive (n=3). Consequently, 49 patients were included in the final cohort (Fig. 1).
Fig. 1. Study design flowchart.
Eighty-five patients with clinically suspected soft-tissue tumors who underwent B-mode ultrasonography (US) and shear-wave elastography (SWE) were screened. After exclusions for non–soft-tissue masses, unreliable SWE data, or inconclusive pathology, 49 patients, including 32 with sarcomas and 17 with non-sarcoma lesions, were included in the final analysis. MRI, magnetic resonance imaging.
SWE Acquisition and Analysis
All patients underwent US using an RS85 Prestige system with an LA2-14A linear-array transducer (2–14 MHz, Samsung Medison, Seoul, Korea). Examinations were performed by a musculoskeletal fellowship-trained radiologist with 7 years of experience. B-mode and power Doppler imaging were first performed to document each lesion’s dimensions in three orthogonal axes, echogenicity relative to adjacent muscle, echotexture as homogeneous or heterogeneous, vascularity pattern as absent, increased, or markedly increased, and anatomic location, including specific body site and lesion compartment/depth relative to the superficial fascia (superficial vs. deep).
SWE was then performed using the same linear-array transducer and system, which provided a real-time dual display of an elasticity map and an RMI overlay. The elasticity map encoded stiff regions in red and compliant regions in blue, with the dynamic range fixed at 0–200 kPa. The RMI map displayed confidence levels on a color scale, with green indicating high reliability (near 1.0) and red indicating low reliability (near 0). Representative examples of different RMI values are shown in Fig. 2.
Fig. 2. Representative reliable measurement index maps showing measurement confidence during shear-wave elastography (SWE).
An 82-year-old man had pathologically confirmed dedifferentiated liposarcoma in the right axilla. A. B-mode ultrasound image shows a heterogeneous hypoechoic mass. B. Power Doppler image demonstrates internal vascularity within the lesion. C, D. Real-time dual-mode displays show the reliable measurement index (RMI) map on the left and the corresponding elasticity map on the right, obtained with the same linear-array transducer. The elasticity map encodes stiffer regions in red and softer regions in blue within a 0–200 kPa scale, whereas the RMI map depicts measurement confidence, with green indicating high reliability and red indicating low reliability. SWE measurements were performed at RMI values of 0.6 (C) and 0.8 (D) using a fixed 5-mm region of interest, with automatic display and recording of the mean, standard deviation, minimum, and maximum stiffness values.
A rectangular sampling box was placed over the lesion, and within this box, a circular 5-mm region of interest (ROI) was positioned in areas with an RMI ≥0.6 under real-time RMI map guidance. To minimize intraobserver variability and better reflect intratumoral heterogeneity, at least four independent SWE measurements were obtained for each lesion rather than relying on a single focal measurement. ROIs were placed in viable-appearing portions of the lesion while avoiding cystic or hemorrhagic areas, guided by grayscale US and preoperative MRI. Within each ROI, the system automatically recorded mean, standard deviation, minimum, and maximum stiffness values, along with the corresponding RMI. For each lesion, the final SWE values used for statistical analysis were calculated by averaging stiffness parameters across repeated measurements, with at least four measurements required to have acceptable reliability (RMI ≥0.6).
B-Mode and Doppler US for Risk Assessment
B-mode and power Doppler images of each lesion were retrospectively reviewed by two musculoskeletal fellowship-trained radiologists (J.P. and Y.H.L., with 7 and 18 years of experience, respectively), both of whom were blinded to SWE findings and histopathology. For each lesion, the reviewers recorded maximal lesion diameter and the presence of any of four suspicious sonographic features: irregular margins, heterogeneous echotexture, marked internal vascularity, and deep fascial location. Lesions measuring ≥5 cm or showing two or more suspicious features were classified as high risk [15–18], and high- versus low-risk categories were then compared with final histologic diagnoses.
Histopathologic Diagnosis
Histopathologic diagnoses were obtained by 14-gauge or 16-gauge core-needle biopsy or subsequent surgical excision and were used as the reference standard for the final cohort. For tumors proven to be soft-tissue sarcomas on biopsy, definitive wide excision was performed by a sarcoma-specialized orthopedic surgeon (S.H.K., with 14 years of experience) who is a core member of the multidisciplinary sarcoma clinic at the authors’ institution. For the radiologic–pathologic correlation analysis, FNCLCC grade and detailed histopathologic parameters were reassessed only in soft-tissue sarcoma cases with wide excision specimens by a musculoskeletal pathologist (S.K.K., with 16 years of experience) from the same multidisciplinary sarcoma clinic.
The following pathologic parameters were reexamined microscopically: mitotic count, cellularity, tumor necrosis, and extent of myxoid stroma. Mitotic activity was scored as 1 for 0–9 mitoses per 10 high-power fields (HPFs), 2 for 10–19 per 10 HPFs, or 3 for ≥20 per 10 HPFs. Tumor necrosis was graded as 0 for absent, 1 for ≤50% of the tumor, or 2 for >50% of the tumor. Cellularity was classified as low, intermediate, or marked. Myxoid stroma was considered present if myxoid material was observed in ≥5% of the tumor area and absent if it was observed in <5%. For statistical analysis, mitotic activity was dichotomized as score 1, corresponding to 0–9 mitoses per 10 HPFs, versus score ≥2, corresponding to ≥10 mitoses per 10 HPFs; necrosis as score 0, absent, versus score ≥1, present; cellularity as low versus intermediate or marked; and myxoid stroma as <5% versus ≥5% of the tumor area.
Statistical Analysis
Statistical analyses were performed according to the predefined primary and secondary study endpoints. The primary endpoint was the association between SWE-derived stiffness values and histopathologic features of soft-tissue sarcomas. SWE stiffness values were compared according to histopathologic parameters using Student’s t-test or one-way analysis of variance, as appropriate.
The secondary endpoint was the exploratory diagnostic performance of conventional US and SWE for differentiating sarcomas from non-sarcoma soft-tissue lesions. For each modality, sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were calculated with corresponding 95% confidence intervals (CIs). For SWE, receiver operating characteristic curve analysis was also performed to evaluate the discriminative performance of mean and maximum stiffness values, and the area under the curve was reported with 95% CIs.
Categorical variables were compared using the chi-square test. Statistical significance was defined as a two-sided P-value <0.05. All analyses were performed using MedCalc version 23.3.2 (MedCalc Software, Ostend, Belgium), and Microsoft Excel version 16.99.2 (Microsoft, Redmond, WA, USA).
Results
Demographic Characteristics
Patient demographics and tumor distributions are summarized in Table 1. The study population had a mean age of 63.57±24.45 years, with an age range of 21–85 years; patients with malignant lesions tended to be older, with a mean age of 74.0±12.33 years.
Table 1.
Patient demographics and histopathologic diagnoses
| Sarcomas (n=32) | Non-sarcomas (n=17) | |
|---|---|---|
| Age (year), mean±SD | 74.0±12.33 | 49.7±26.6 |
| Male:female (male %) | 15:17 (46.9%) | 12:5 (63.2%) |
| Histopathology | Malignant peripheral nerve sheath tumor (n=7) | Schwannoma (n=4) |
| Myxofibrosarcoma (n=5) | Desmoid-type fibromatosis (n=3) | |
| Undifferentiated spindle cell sarcoma (n=5) | Angiofibroma (n=2) | |
| Leiomyosarcoma (n=4) | Neurofibroma (n=2) | |
| Myxoid liposarcoma (n=2) | Hemangioma/vascular malformation (n=2) | |
| Rhabdomyosarcoma (n=2) | Lipoma (n=1) | |
| Synovial sarcoma (n=1) | Hibernoma (n=1) | |
| Histiocytic sarcoma (n=1) | Desmoplastic fibroblastoma (n=1) | |
| Myxoinflammatory fibroblastic sarcoma (n=1) | Angioleiomyoma (n=1) | |
| NTRK-rearranged spindle cell neoplasm (n=1) | ||
| Angiosarcoma (n=1) | ||
| Solitary fibrous tumor (n=1) | ||
| Chordoma (n=1) |
Total (n=49): age 63.57±24.45 years; male:female 27:22 (male 55.1%).
SD, standard deviation; NTRK, neurotrophic tyrosine receptor kinase.
Histopathologic evaluation confirmed 32 malignant (65.3%), three intermediate (6.1%), and 14 benign (28.5%) soft-tissue tumors. Among benign lesions, schwannoma was the most frequent diagnosis (n=4, 8.2%), whereas desmoid-type fibromatosis predominated among intermediate neoplasms (n=3, 6.1%). The most common sarcoma subtype was malignant peripheral nerve sheath tumor (n=7, 14.3%). To facilitate statistical analysis and reflect the study focus on differentiating sarcomas from other lesions, benign and intermediate tumors were combined into a non-sarcoma category, particularly given the small number of intermediate cases.
SWE-Derived Stiffness Values in Relation to Histopathologic Features of Soft-Tissue Sarcomas
Association of SWE stiffness values with FNCLCC histologic grade in soft-tissue sarcomas
Among the 32 sarcoma cases with wide excision specimens, the association between SWE-derived stiffness values and pathologist-confirmed FNCLCC histologic grades 1–3 was evaluated in 28 cases: grade 1, n=11; grade 2, n=7; and grade 3, n=10. Four cases without definitive microscopic grading were excluded.
Contrary to the expected progressive increase in stiffness with higher histologic grade, grade 2 tumors had the highest SWE values. Mean stiffness was 37.90±15.02, 67.72±34.71, and 49.97±36.52 kPa for FNCLCC grades 1, 2, and 3, respectively, and maximum stiffness was 61.96±19.09, 114.94±54.79, and 89.71±42.18 kPa, respectively (Table 2, Fig. 3A). Although one-way analysis of variance showed no statistically significant difference among grades for mean stiffness (P=0.153) or maximum stiffness (P=0.108), both mean and maximum stiffness values increased from grade 1 to grade 2 and then decreased in grade 3.
Table 2.
SWE-derived stiffness values according to FNCLCC grade and histopathologic features of soft-tissue sarcomas
| Category | Mean SWE (kPa) | P-value | Max SWE (kPa) | P-value |
|---|---|---|---|---|
| Diagnosis | ||||
| Sarcoma | 49.79±30.67 | 0.715 | 89.85±54.10 | 0.500 |
| Non-sarcoma | 46.76±25.28 | 80.48±54.12 | ||
| FNCLCC grade | ||||
| Grade 1 | 37.90±15.02 | 0.153 | 61.96±19.09 | 0.108 |
| Grade 2 | 67.72±34.71 | 114.94±54.79 | ||
| Grade 3 | 49.97±36.52 | 89.71±42.18 | ||
| Histologic features | ||||
| Necrosis | ||||
| Low | 57.21±33.19 | 0.036* | 100.32±48.59 | 0.016* |
| High | 38.14±16.11 | 69.82±19.58 | ||
| Cellularity | ||||
| Low | 46.95±30.82 | 0.628 | 82.44±46.77 | 0.642 |
| High | 78.61±19.55 | 137.22±43.89 | ||
| Mitosis | ||||
| Low | 37.15±12.94 | 0.405 | 64.22±7.15 | 0.290 |
| High | 57.94±31.52 | 102.85±45.44 | ||
| Myxoid stroma | ||||
| Low | 54.21±36.44 | 0.411 | 97.68±62.15 | 0.420 |
| High | 48.77±30.91 | 83.15±41.22 |
Values are presented as mean±standard deviation.
SWE, shear-wave elastography; FNCLCC, Fédération Nationale des Centres de Lutte Contre le Cancer.
*P<0.05 indicates a statistically significant difference.
Fig. 3. Comparison of shear-wave elastography (SWE) stiffness values according to histopathologic features of soft-tissue sarcomas.
Bar graphs show the mean±standard deviation for mean and maximum SWE stiffness values, expressed in kilopascals. A. A comparison of SWE stiffness values was performed according to Fédération Nationale des Centres de Lutte Contre le Cancer (FNCLCC) histologic grade in sarcomas. B. A comparison of SWE stiffness values was performed according to microscopic tumor necrosis and other histopathologic parameters. Statistically significant differences (P<0.05) are indicated by asterisks and corresponding P-values on the plot.
Association of SWE stiffness values with microscopic tumor necrosis and other histopathologic parameters in soft-tissue sarcomas
Microscopic necrosis was the only histopathologic feature significantly associated with SWE-derived stiffness. Sarcomas without microscopic necrosis showed significantly higher mean stiffness than sarcomas with necrosis (57.21±33.19 kPa vs. 38.14±16.11 kPa, P=0.036) and significantly higher maximum stiffness (100.32±48.59 kPa vs. 69.82±19.58 kPa, P=0.016).
In contrast, no significant associations were observed between SWE-derived stiffness and the other histopathologic parameters. Mean SWE stiffness did not differ significantly according to cellularity (P=0.628), mitotic count (P=0.405), or extent of myxoid stroma (P=0.411). Similarly, maximum SWE stiffness did not differ significantly according to cellularity (P=0.642), mitotic count (P=0.290), or extent of myxoid stroma (P=0.420).
Detailed results are presented in Table 2 and Fig. 3B, and representative cases showing the association between SWE stiffness values and microscopic tumor necrosis are shown in Fig. 4.
Fig. 4. Representative cases demonstrating the relationship between microscopic necrosis and shear-wave elastography stiffness in malignant peripheral nerve sheath tumors.
From left to right, each row presents B-mode ultrasound, shear-wave elastography (SWE), and the corresponding T2-weighted magnetic resonance image of histologically confirmed malignant peripheral nerve sheath tumors. A–C. A heterogeneous hypoechoic tumor located in the deep subcutaneous tissue of the left upper back abuts the fascia and displaces the adjacent muscle (A), with intermediate signal intensity on T2-weighted magnetic resonance imaging (C). B. SWE demonstrated relatively high stiffness (mean, 81.9±4.4 kPa; maximum, 91.8 kPa). Microscopic examination revealed no necrosis (score 0). D–F. A tumor located in the deep subcutaneous layer of the left posterior thigh, similar to the tumor in case A, displaced the adjacent muscle and showed heterogeneous high signal intensity on fat-suppressed T2-weighted magnetic resonance imaging (F). Histopathology demonstrated microscopic necrosis involving more than 50% of the tumor (score 2). E. SWE showed substantially lower stiffness than in case A (mean, 31.4±2.6 kPa; maximum, 36.9 kPa), illustrating that necrosis within sarcomas may contribute to lower SWE values despite similar histologic grades.
Diagnostic Performance of Conventional US and SWE in Differentiating Sarcomas from Non-sarcoma Lesions
Diagnostic performance of conventional US
A total of 49 soft-tissue lesions were stratified by conventional B-mode and power Doppler criteria into high-risk and low-risk groups, with high risk defined as size ≥5 cm or the presence of at least two suspicious features [15,19]. On conventional US, 38 lesions (77.6%) were classified as high risk, including 27 of 32 sarcomas (84.4%) and 11 of 17 non-sarcomas (64.7%). The remaining 11 lesions (22.4%) were classified as low risk and comprised six non-sarcomas and five sarcomas. Compared with histopathologic diagnosis, conventional US had a sensitivity of 84.4% (95% CI, 67.2% to 94.7%), specificity of 35.3% (95% CI, 14.2% to 61.7%), accuracy of 67.3% (95% CI, 52.5% to 80.1%), positive predictive value of 71.1% (95% CI, 54.1% to 84.6%), and negative predictive value of 54.5% (95% CI, 23.4% to 83.3%) (Table 3). The association between US risk category and histopathologic diagnosis did not reach statistical significance (P=0.068).
Table 3.
Diagnostic performance of conventional US and SWE-derived stiffness values for sarcoma discrimination
| Modality/Parameter | Optimal cut-off | Sensitivity (%) | Specificity (%) | Accuracy (%) | PPV (%) | NPV (%) | AUC (95% CI) |
|---|---|---|---|---|---|---|---|
| Conventional US | Size ≥ 5 cm or 2 featuresa) | 84.4 (67.2–94.7) | 35.3 (14.2–61.7) | 67.3 (52.5–80.1) | 71.1 (54.1–84.6) | 54.5 (23.4–83.3) | - |
| Mean SWE | >73.84 kPa | 21.9 (9.3–40.0) | 88.2 (63.6–98.5) | 44.9 (30.7–59.8) | 77.8 (40.0–97.2) | 37.5 (22.7–54.2) | 0.518 (0.371–0.663) |
| Maximum SWE | >120.0 kPa | 31.3 (16.1–50.0) | 82.4 (56.6–96.2) | 49.0 (34.4–63.7) | 76.9 (46.2–95.0) | 38.9 (23.1–56.5) | 0.545 (0.396–0.688) |
US, ultrasonography; SWE, shear-wave elastography; PPV, positive predictive value; NPV, negative predictive value; AUC, area under the receiver operating characteristic curve; CI, confidence interval.
a)The optimal criteria for conventional US high-risk classification were defined as size ≥5 cm or ≥2 suspicious features. Data are presented as percentages, with 95% CIs shown in parentheses where applicable. The AUC for conventional US was not calculated because the risk stratification was evaluated as a fixed, dichotomous categorical variable (high- vs. low-risk) rather than a continuous scale.
Diagnostic performance of SWE
SWE-derived stiffness values were compared between sarcoma (n=32) and non-sarcoma (n=17) lesions. Although both mean and maximum stiffness tended to be higher in sarcomas, the distributions overlapped substantially, and the differences were not statistically significant. Mean SWE stiffness was 49.79±30.67 kPa for sarcomas versus 46.76±25.28 kPa for non-sarcomas (P=0.715), and maximum SWE stiffness was 89.85±54.10 kPa versus 80.48±54.12 kPa, respectively (P=0.500) (Table 2).
To further assess the diagnostic performance of SWE, receiver operating characteristic curve analysis was performed for mean and maximum stiffness values. For mean SWE stiffness, the area under the curve was 0.518 (95% CI, 0.371 to 0.663). The estimated cutoff value based on the Youden index was >73.84 kPa, with a sensitivity of 21.9% (95% CI, 9.3% to 40.0%) and specificity of 88.2% (95% CI, 63.6% to 98.5%). For maximum SWE stiffness, the area under the curve was 0.545 (95% CI, 0.396 to 0.688), with an estimated cutoff value of >120.0 kPa, sensitivity of 31.3% (95% CI, 16.1% to 50.0%), and specificity of 82.4% (95% CI, 56.6% to 96.2%). The overall diagnostic performance of conventional US, mean SWE stiffness, and maximum SWE stiffness is summarized in Table 3.
Discussion
SWE-derived stiffness in soft-tissue tumors is variable, and its clinical interpretation remains challenging. This study systematically investigated associations between SWE-derived stiffness and key histopathologic features of soft-tissue sarcomas, including microscopic necrosis, mitotic activity, cellularity, extent of myxoid stroma, and FNCLCC histologic grade.
The principal finding was a significant inverse association between microscopic necrosis and SWE-derived stiffness: sarcomas without necrosis had higher mean and maximum stiffness values than those with necrosis. Similar necrosis-related radiologic–pathologic relationships have been reported in carcinoma models, in which increasing tumor necrosis was associated with lower SWE stiffness and treatment-induced softening corresponded to the emergence of necrosis on pathology [13]. To the authors’ knowledge, this relationship has not been quantitatively demonstrated in a clinical cohort of human soft-tissue sarcomas, in which mesenchymal tumor heterogeneity is often more pronounced than the relatively uniform stromal architecture of epithelial tumors. By correlating SWE values with standardized FNCLCC necrosis scores reassessed on wide excision specimens, the findings extend this concept to the clinically heterogeneous sarcoma setting and provide a plausible pathophysiologic explanation for the nonlinear FNCLCC grade pattern described below.
SWE-derived stiffness did not increase progressively with higher FNCLCC grade. Instead, stiffness values increased from grade 1 to grade 2 and then declined in grade 3 sarcomas. This pattern may reflect biological changes that occur as high-grade soft-tissue sarcomas progress. As sarcomas enlarge and become more aggressive, rapid tumor proliferation may exceed the available vascular supply, leading to secondary degenerative changes such as hemorrhage, cystic degeneration, and extensive necrosis [20,21]. Consequently, although grade 3 sarcomas show aggressive histologic features, the increasing proportion of nonviable tissue may paradoxically reduce measured SWE stiffness. These findings may help explain why previous studies evaluating SWE-derived stiffness in soft-tissue sarcomas have yielded inconsistent or conflicting results [6,7,12], suggesting that elastographic stiffness should not be interpreted as a simple surrogate marker of tumor grade.
In contrast, the other evaluated histopathologic parameters—cellularity, mitotic activity, and extent of myxoid stroma—were not significantly associated with SWE-derived stiffness. Although increased cellularity or mitotic activity might be expected to increase stiffness, these microscopic proliferative features may exert a smaller macroscopic mechanical effect than necrosis-related architectural disruption. Experimental studies suggest that tissue stiffness is influenced more strongly by large-scale extracellular matrix architecture and structural degeneration [20,22]. Similarly, although myxoid components are intrinsically soft, the lesion-level averaging approach used in this study may have diluted focal myxoid effects in tumors with mixed fibrous, cellular, and myxoid composition. These findings suggest that SWE-derived stiffness may be influenced more by larger-scale structural alterations and tissue viability than by isolated microscopic cellular or stromal features. Future studies using spatially matched radiologic–pathologic correlation may better define the localized contributions of cellular and stromal components to SWE-derived stiffness.
This study also evaluated the diagnostic performance of conventional US and SWE for differentiating sarcomas from non-sarcoma lesions. Conventional US showed high sensitivity (84.4%) but low specificity (35.3%), reflecting a high rate of false-positive high-risk classifications among benign or intermediate lesions. SWE showed low sensitivity but relatively high specificity at the estimated cutoff values: for mean stiffness >73.84 kPa, sensitivity was 21.9% and specificity was 88.2%; for maximum stiffness >120.0 kPa, sensitivity was 31.3% and specificity was 82.4%. These complementary patterns suggest a potential adjunctive role for SWE. Conventional US may be useful for initial screening, whereas markedly elevated SWE stiffness may increase suspicion for sarcoma when present. However, these cutoff values should be interpreted as exploratory estimates derived from a modestly sized single-center cohort. Larger multicenter studies are therefore needed to validate these thresholds and determine whether an integrated US-plus-SWE diagnostic model provides incremental value for sarcoma discrimination beyond conventional US alone.
This study has several limitations. First, the cohort was drawn from a single center and evaluated using one US platform and acquisition protocol. Although this design may limit generalizability, it also has important advantages: all examinations were performed using the same US platform, identical software version, and a consensus-based acquisition protocol, which helped ensure internally consistent and technically reliable SWE data. Although the sample size was relatively small, reflecting the low incidence of soft-tissue sarcomas even in a dedicated multidisciplinary sarcoma clinic, future multicenter studies with larger cohorts are needed to further assess generalizability. Second, the cohort was derived from a tertiary sarcoma center and included patients with clinically suspected sarcoma who underwent preoperative MRI and subsequent US-guided biopsy; therefore, referral-related spectrum bias is possible. However, given prior reports showing variable and often limited incremental value of SWE for benign–malignant differentiation, the present study intentionally screened a consecutive sarcoma-referral population to investigate determinants of SWE variability, particularly histopathologic correlates among sarcomas, and to evaluate the clinical utility of SWE in this setting. Third, the cohort represented a heterogeneous, real-world sarcoma-referral population with a broad histopathologic spectrum of soft-tissue tumors, in which substantial overlap in stiffness values across entities is expected. Intermediate lesions, consisting of only three cases of desmoid-type fibromatosis, were grouped with benign lesions as a non-sarcoma category for statistical analysis. Nevertheless, such histologic heterogeneity and intermediate features may contribute to overlapping stiffness measurements and affect diagnostic discrimination. Finally, exact point-to-point correspondence between SWE measurement sites and histopathologic assessment cannot be guaranteed. Although multiple independent SWE measurements were obtained for each lesion and radiologic-pathologic associations were evaluated at the lesion level using parameters reassessed on wide excision specimens, residual spatial mismatch may have influenced the observed associations. Future validation studies using co-registration techniques could improve spatial alignment between imaging and pathology, enabling more precise tumor-component–level correlation.
In conclusion, SWE-derived stiffness in soft-tissue sarcomas appears to be affected by intratumoral heterogeneity, particularly microscopic necrosis, which may contribute to the nonlinear relationship between stiffness and FNCLCC histologic grade. Although SWE alone showed limited sensitivity for sarcoma discrimination, markedly elevated stiffness had relatively high specificity and may provide complementary information when interpreted alongside conventional US.
Key points
Microscopic necrosis was the only histopathologic feature significantly associated with shear-wave elastography (SWE)–derived stiffness in soft-tissue sarcomas, with necrotic tumors showing lower stiffness values. SWE stiffness showed a nonlinear relationship with Fédération Nationale des Centres de Lutte Contre le Cancer histologic grade, increasing from grade 1 to grade 2 and decreasing in grade 3 sarcomas, likely reflecting intratumoral degeneration and heterogeneity. Although SWE alone showed limited sensitivity for sarcoma discrimination, markedly elevated stiffness values were relatively specific for sarcoma and may provide complementary information alongside conventional US.
Footnotes
Author Contributions
Conceptualization: Park J, Lee YH. Data acquisition: Park J, Hahn S, Lee YH. Data analysis or interpretation: Kim SK, Kim SH, Lee YH. Drafting of the manuscript: Park J, Kim SK, Kim SH, Hahn S, Lee YH. Critical revision of the manuscript: Park J, Lee YH. Approval of the final version of the manuscript: all authors.
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Acknowledgments
This work was supported by a National Research Foundation of Korea grant funded by the Korean government, Ministry of Science and ICT (2022R1F1A1071702).
References
- 1.Gamboa AC, Gronchi A, Cardona K. Soft-tissue sarcoma in adults: an update on the current state of histiotype-specific management in an era of personalized medicine. CA Cancer J Clin. 2020;70:200–229. doi: 10.3322/caac.21605. [DOI] [PubMed] [Google Scholar]
- 2.Chandrasekar CR, Wafa H, Grimer RJ, Carter SR, Tillman RM, Abudu A. The effect of an unplanned excision of a soft-tissue sarcoma on prognosis. J Bone Joint Surg Br. 2008;90:203–208. doi: 10.1302/0301-620x.90b2.19760. [DOI] [PubMed] [Google Scholar]
- 3.Potter BK, Adams SC, Pitcher JD Jr, Temple HT. Local recurrence of disease after unplanned excisions of high-grade soft tissue sarcomas. Clin Orthop Relat Res. 2008;466:3093–3100. doi: 10.1007/s11999-008-0529-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Qureshi YA, Huddy JR, Miller JD, Strauss DC, Thomas JM, Hayes AJ. Unplanned excision of soft tissue sarcoma results in increased rates of local recurrence despite full further oncological treatment. Ann Surg Oncol. 2012;19:871–877. doi: 10.1245/s10434-011-1876-z. [DOI] [PubMed] [Google Scholar]
- 5.Wu M, Ren A, Xu D, Peng X, Ye X, Li A. Diagnostic performance of elastography in malignant soft tissue tumors: a systematic review and meta-analysis. Ultrasound Med Biol. 2021;47:855–868. doi: 10.1016/j.ultrasmedbio.2020.12.017. [DOI] [PubMed] [Google Scholar]
- 6.Bodard S, Lassalle L, Larousserie F, Guinebert S, Hacquart T, Campagna R, et al. Performance of sonoelastography for predicting malignancy in soft tissue. BMC Cancer. 2022;22:1305. doi: 10.1186/s12885-022-10300-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tavare AN, Alfuraih AM, Hensor EM, Astrinakis E, Gupta H, Robinson P. Shear-wave elastography of benign versus malignant musculoskeletal soft-tissue masses: comparison with conventional US and MRI. Radiology. 2019;290:410–417. doi: 10.1148/radiol.2018180950. [DOI] [PubMed] [Google Scholar]
- 8.Davis LC, Baumer TG, Bey MJ, Holsbeeck MV. Clinical utilization of shear wave elastography in the musculoskeletal system. Ultrasonography. 2019;38:2–12. doi: 10.14366/usg.18039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Yeoh HJ, Kim TY, Ryu JA. The feasibility of shear wave elastography for diagnosing superficial benign soft tissue masses. Ultrasonography. 2019;38:37–43. doi: 10.14366/usg.17059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.WHO Classification of Tumours Editorial Board . WHO classification of tumours. 5th ed. Vol. 3. Soft tissue and bone tumours. Lyon: International Agency for Research on Cancer; 2020. [Google Scholar]
- 11.Taljanovic MS, Gimber LH, Becker GW, Latt LD, Klauser AS, Melville DM, et al. Shear-wave elastography: basic physics and musculoskeletal applications. Radiographics. 2017;37:855–870. doi: 10.1148/rg.2017160116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Winn N, Baldwin J, Cassar-Pullicino V, Cool P, Ockendon M, Tins B, et al. Characterization of soft tissue tumours with ultrasound, shear wave elastography and MRI. Skeletal Radiol. 2020;49:869–881. doi: 10.1007/s00256-019-03363-1. [DOI] [PubMed] [Google Scholar]
- 13.Saggioro M, D'Angelo E, Bisogno G, Agostini M, Pozzobon M. Carcinoma and sarcoma microenvironment at a glance: where we are. Front Oncol. 2020;10:76. doi: 10.3389/fonc.2020.00076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Radu I, Scripcariu V, Panuta A, Rusu A, Afrasanie VA, Cojocaru E, et al. Breast sarcomas: how different are they from breast carcinomas? Clinical, pathological, imaging and treatment insights. Diagnostics (Basel) 2023;13:1370. doi: 10.3390/diagnostics13081370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shu H, Ma Q, Li A, Wang P, Gao Y, Yao Q, et al. Diagnostic performance of US and MRI in predicting malignancy of soft tissue masses: using a scoring system. Front Oncol. 2022;12:853232. doi: 10.3389/fonc.2022.853232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhang Y, Zhao C, Lv H, Dong L, Xie L, Tian Y, et al. Benefit of using both ultrasound imaging and clinical information for predicting malignant soft tissue tumors. Ultrasound Med Biol. 2023;49:2459–2468. doi: 10.1016/j.ultrasmedbio.2023.08.005. [DOI] [PubMed] [Google Scholar]
- 17.Chiou HJ, Chou YH, Chiu SY, Wang HK, Chen WM, Chen TH, et al. Differentiation of benign and malignant superficial soft-tissue masses using grayscale and color Doppler ultrasonography. J Chin Med Assoc. 2009;72:307–315. doi: 10.1016/s1726-4901(09)70377-6. [DOI] [PubMed] [Google Scholar]
- 18.Morii T, Kishino T, Shimamori N, Motohashi M, Ohnishi H, Honya K, et al. Differential diagnosis between benign and malignant soft tissue tumors utilizing ultrasound parameters. J Med Ultrason (2001) 2018;45:113–119. doi: 10.1007/s10396-017-0796-3. [DOI] [PubMed] [Google Scholar]
- 19.Griffith JF. Practical approach to ultrasound of soft tissue tumors and the added value of MRI: how I do it. J Ultrason. 2023;23:e299–e312. doi: 10.15557/jou.2023.0036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chamming's F, Latorre-Ossa H, Le Frere-Belda MA, Fitoussi V, Quibel T, Assayag F, et al. Shear wave elastography of tumour growth in a human breast cancer model with pathological correlation. Eur Radiol. 2013;23:2079–2086. doi: 10.1007/s00330-013-2828-8. [DOI] [PubMed] [Google Scholar]
- 21.Chamming's F, Le-Frere-Belda MA, Latorre-Ossa H, Fitoussi V, Redheuil A, Assayag F, et al. Supersonic shear wave elastography of response to anti-cancer therapy in a xenograft tumor model. Ultrasound Med Biol. 2016;42:924–930. doi: 10.1016/j.ultrasmedbio.2015.12.001. [DOI] [PubMed] [Google Scholar]
- 22.Huang JX, Liu FT, Sun L, Ma C, Fu J, Wang XY, et al. Comparing shear wave elastography of breast tumors and axillary nodes in the axillary assessment after neoadjuvant chemotherapy in patients with node-positive breast cancer. Radiol Med. 2024;129:1143–1155. doi: 10.1007/s11547-024-01848-1. [DOI] [PMC free article] [PubMed] [Google Scholar]