Soft-tissue sarcoma in adults: Imaging appearances, pitfalls and diagnostic algorithms

Evaluating the role and diagnostic performance of ultrasound in diagnosing soft tissue tumors and pseudotumors in adults: a systematic review and meta-analysis

OBJECTIVES

This systematic review and meta-analysis evaluates the combined accuracy of ultrasonic imaging in diagnosing soft tissue tumors and pseudotumors and identifies factors contributing to variability in its diagnostic accuracy.

METHODS

A literature search in significant databases used specific keywords and inclusion/exclusion criteria. Two authors independently performed cohort characterization and data analysis, with a third author resolving disagreements. A multivariate random effects model estimated pooled sensitivity, specificity, predictive values, and overall accuracy.

RESULTS

Ultrasound imaging demonstrated high pooled sensitivity and specificity, indicating effectiveness in identifying lesions. Variability in accuracy was attributed to factors such as tumor type and location, operator experience, and clinic location.

CONCLUSIONS

Ultrasonic imaging is a reliable diagnostic tool for soft tissue tumors and pseudotumors, providing high sensitivity and specificity. However, variability in accuracy underscores the need to consider operator experience and tumor characteristics. Further research should explore the impact of technological advancements and include non-English studies to enhance understanding of ultrasound imaging's diagnostic validity.

Radiomics and artificial intelligence for soft-tissue sarcomas: Current status and perspectives

This article proposes a summary of the current status of the research regarding the use of radiomics and artificial intelligence to improve the radiological assessment of patients with soft tissue sarcomas (STS), a heterogeneous group of rare and ubiquitous mesenchymal malignancies. After a first part explaining the principle of radiomics approaches, from raw image post-processing to extraction of radiomics features mined with unsupervised and supervised machine-learning algorithms, and the current research involving deep learning algorithms in STS, especially convolutional neural networks, this review details their main research developments since the formalisation of 'radiomics' in oncologic imaging in 2010. This review focuses on CT and MRI and does not involve ultrasonography. Radiomics and deep radiomics have been successfully applied to develop predictive models to discriminate between benign soft-tissue tumors and STS, to predict the histologic grade (i.e., the most important prognostic marker of STS), the response to neoadjuvant chemotherapy and/or radiotherapy, and the patients' survivals and probability for presenting distant metastases. The main findings, limitations and expectations are discussed for each of these outcomes. Overall, after a first decade of publications emphasizing the potential of radiomics through retrospective proof-of-concept studies, almost all positive but with heterogeneous and often non-replicable methods, radiomics is now at a turning point in order to provide robust demonstrations of its clinical impact through open-science, independent databases, and application of good and standardized practices in radiomics such as those provided by the Image Biomarker Standardization Initiative, without forgetting innovative research paths involving other '-omics' data to better understand the relationships between imaging of STS, gene-expression profiles and tumor microenvironment.

Sarcoma care in the era of precision medicine

Sarcoma subtype classification is currently mainly based upon histopathological morphology. Molecular analyses have emerged as an efficient addition to the diagnostic workup and sarcoma care. Knowledge about the sarcoma genome increases, and genetic events that can either support a histopathological diagnosis or suggest a differential diagnosis are identified, as well as novel therapeutic targets. In this review, we present diagnostic, therapeutic, and prognostic molecular markers that are, or might soon be, used clinically. For sarcoma diagnostics, there are specific fusions highly supportive or pathognomonic for a diagnostic entity-for instance, SYT::SSX in synovial sarcoma. Complex karyotypes also give diagnostic information-for example, supporting dedifferentiation rather than low-grade central osteosarcoma or well-differentiated liposarcoma when detected in combination with MDM2/CDK4 amplification. Molecular treatment predictive sarcoma markers are available for gastrointestinal stromal tumor (GIST) and locally aggressive benign mesenchymal tumors. The molecular prognostic markers for sarcomas in clinical practice are few. For solitary fibrous tumor, the type of NAB2::STAT6 fusion is associated with the outcome, and the KIT/PDGFRA pathogenic variant in GISTs can give prognostic information. With the exploding availability of sequencing technologies, it becomes increasingly important to understand the strengths and limitations of those methods and their context in sarcoma diagnostics. It is reasonable to believe that most sarcoma treatment centers will increase the use of massive-parallel sequencing soon. We conclude that the context in which the genetic findings are interpreted is of importance, and the interpretation of genomic findings requires considering tumor histomorphology.

Calculating Tumor Volume Using Three-Dimensional Models in Preoperative Soft-Tissue Sarcoma Surgical Planning: Does Size Matter?

This feasibility study aims to explore the use of three-dimensional virtual surgical planning to preoperatively determine the need for reconstructive surgery following resection of an extremity soft-tissue sarcoma. As flap reconstruction is performed more often in advanced disease, we hypothesized that tumor volume would be larger in the group of patients that had undergone flap reconstruction. All patients that were treated by surgical resection for an extremity soft-tissue sarcoma between 1 January 2016 and 1 October 2019 in the University Medical Center Groningen were included retrospectively. Three-dimensional models were created using the diagnostic magnetic resonance scan. Tumor volume was calculated for all patients. Three-dimensional tumor volume was 107.8 (349.1) mL in the group of patients that had undergone primary closure and 29.4 (47.4) mL in the group of patients in which a flap reconstruction was performed, p = 0.004. Three-dimensional tumor volume was 76.1 (295.3) mL in the group of patients with a complication following ESTS treatment, versus 57.0 (132.4) mL in patients with an uncomplicated course following ESTS treatment, p = 0.311. Patients who had undergone flap reconstruction had smaller tumor volumes compared to those in the group of patients treated by primary closure. Furthermore, a larger tumor volume did not result in complications for patients undergoing ESTS treatment. Therefore, tumor volume does not seem to influence the need for reconstruction. Despite the capability of three-dimensional virtual surgical planning to measure tumor volume, we do not recommend its utilization in the multidisciplinary extremity soft-tissue sarcoma treatment, considering the findings of the study.

Imaging Features of Alveolar Soft Part Sarcoma: Single Institution Experience and Literature Review

Alveolar soft part sarcoma (ASPS) is an extremely rare and aggressive soft-tissue sarcoma (STS) subtype with poor prognosis and limited response to radiation therapy and chemotherapy. Prompt recognition and referral to sarcoma centers for appropriate management are crucial for patients' survival. The purpose of this study was to report ASPS pre-treatment imaging features and to examine the existing literature on this topic. Twelve patients (7 women, 5 men-mean age 27.1 ± 10.7 years) were included from our single-center experience. Ultrasonography (US), computed tomography (CT), and magnetic resonance imaging (MRI) available were reviewed according to an analysis grid incorporating features from the latest research on STS. Clinical, histological, and outcome data were collected. MRI was available in 10 patients (83.3%), US in 7 patients (58.3%), and CT in 3 patients (25%). Mean longest tumor diameter was 7.6 ± 2.9 cm, and all tumors were deeply seated. Large peritumoral feeding vessels were systematically found and identified on ultrasonography (7/7), MRI (10/10), and CT (3/3). US revealed a well-defined heterogeneous hypoechoic pattern, with abundant flow signals in all patients (7/7). In all patients, MRI showed mildly high signal intensity (SI) on T1-WI and high SI on T2-WI and peritumoral edema. Moreover, flow-voids (due to arteriosus high-flow) into the peritumoral/intratumoral feeding vessels were detected in the MRI fluid-sensitive sequences of all patients. At baseline, whole-body contrast-enhanced CT revealed metastases in 8/12 (66.7%) patients. A pre-treatment longest diameter > 5 cm was significantly associated with distant metastases at diagnosis (p = 0.01). A maximum diameter > 5 cm represents a risk of metastatic disease at diagnosis (odds ratio = 45.0000 (95% CI: 1.4908-1358.3585), p = 0.0285). In the comprehensive literature review, we found 14 articles (case series or original research) focusing on ASPS imaging, with a total of 151 patients included. Merging our experience with the data from the existing literature, we conclude that the hallmark of ASPS imaging at presentation are the following characteristics: deep location, a slight hyperintense MRI SI on T1-WI and a hyperintense SI on T2-WI, numerous MRI flow voids, high internal vascularization, and large peritumoral feeding vessels.

Synovial Sarcoma in the Extremity: Diversity of Imaging Features for Diagnosis and Prognosis

Synovial sarcomas are rare and highly aggressive soft-tissue sarcomas, primarily affecting adolescents and young adults aged 15-40 years. These tumors typically arise in the deep soft tissues, often near the large joints of the extremities. While the radiological features of these tumors are not definitely indicative, the presence of calcification in a soft-tissue mass (occurring in 30% of cases), adjacent to a joint, strongly suggests the diagnosis. Cross-sectional imaging characteristics play a crucial role in diagnosing synovial sarcomas. They often reveal significant characteristics such as multilobulation and pronounced heterogeneity (forming the "triple sign"), in addition to features like hemorrhage and fluid-fluid levels with septa (resulting in the "bowl of grapes" appearance). Nevertheless, the existence of non-aggressive features, such as gradual growth (with an average time to diagnosis of 2-4 years) and small size (initially measuring < 5 cm) with well-defined margins, can lead to an initial misclassification as a benign lesion. Larger size, older age, and higher tumor grade have been established as adverse predictive indicators for both local disease recurrence and the occurrence of metastasis. Recently, the prognostic importance of CT and MRI characteristics for synovial sarcomas was elucidated. These include factors like the absence of calcification, the presence of cystic components, hemorrhage, the bowl of grape sign, the triple sign, and intercompartmental extension. Wide surgical excision remains the established approach for definitive treatment. Gaining insight into and identifying the diverse range of presentations of synovial sarcomas, which correlate with the prognosis, might be helpful in achieving the optimal patient management.

Percutaneous CT-Guided Bone Biopsies: Indications, Feasibility and Diagnostic Yield in the Different Skeletal Sites-From the Skull to the Toe

CT-guided bone biopsies are currently the diagnostic tool of choice for histopathological (and microbiological) diagnoses of skeletal lesions. Several research works have well-demonstrated their safety and feasibility in almost all skeletal regions. This comprehensive review article aims at summarizing the general concepts in regard to bone biopsy procedures, current clinical indications, the feasibility and the diagnostic yield in different skeletal sites, particularly in the most delicate and difficult-to-reach ones. The choice of the correct imaging guidance and factors affecting the diagnostic rate, as well as possible complications, will also be discussed. Since the diagnostic yield, technical difficulties, and complications risk of a CT-guided bone biopsy significantly vary depending on the different skeletal sites, subdivided analyses of different anatomical sites are provided. The information included in the current review article may be useful for clinicians assisting patients with possible bone neoplasms, as well as radiologists involved in the imaging diagnoses of skeletal lesions and/or in performing bone biopsies.

Imaging presentation of extraskeletal osteosarcomas on CT and MRI and correlation with patients outcome: A two-center retrospective study of 54 patients

PURPOSE

The purpose of this study was to analyze the imaging features of extraskeletal osteosarcomas (ESOS) on computed tomography (CT) and magnetic resonance imaging (MRI) and to investigate their associations with overall survival (OS) using uni- and multivariable survival analyses.

MATERIALS AND METHODS

This two-center retrospective study included all consecutive adult patients between 2008 and 2021 with histopathologically-proven ESOS who underwent pre-treatment CT and/or MRI. Clinical and histological characteristics, ESOS presentation on CT and MRI, treatment and outcomes were reported. Survival analyses were performed using Kaplan-Meier analysis and Cox regressions. Associations between imaging features and OS were searched using uni- and multivariable analyses.

RESULTS

Fifty-four patients were included (30/54 [56%] men, median age: 67.5 years). Twenty-four died of ESOS (median OS: 18 months). ESOS were mostly deep-seated (46/54, 85%) in the lower limb (27/54, 50%) with a median size of 95 mm (interquartile range: 64, 142; range: 21-289 mm). Mineralization was seen on 26/42 (62%) patients, mainly gross-amorphous (18/26; 69%). ESOS were generally highly heterogeneous on T2-weighted images (38/48; 79%) and contrast-enhanced (CE) T1-weighted images (29/40; 72%), with necrosis (39/40; 97%), well-defined or focally infiltrative margins (39/47; 83%), with moderate peritumoral edema (39/47; 83%) and rim-like peripheral enhancement (17/40; 42%). Size, location, mineralization on CT, signal intensity heterogeneity on T1-, T2- and CE-T1-weighted images and hemorragic signal on MRI were associated with poorer OS (range of log-rank P = 0.0069-0.0485). At multivariable analysis, hemorragic signal and signal intensity heterogeneity on T2-weighted images remained predictive for poorer OS (hazard ratio [HR] = 2.68, P = 0.0299; HR = 9.85, P = 0.0262, respectively) CONCLUSION: ESOS typically presents as mineralized heterogeneous and necrotic soft tissue tumor with a possible rim-like enhancement and limited peritumoral abnormalities. MRI may help estimate outcome of patients with ESOS.