Even though tumors in children are rare, they cause the second most deaths under the age of 18 years. More often than in other age groups, underage patients suffer from malignancies of the bones, and these mostly occur in the area around the knee. One problem in the treatment is the early detection of bone tumors, especially on X-rays. The rarity and non-specific clinical symptoms further prolong the time to diagnosis. Nevertheless, an early diagnosis is crucial and can facilitate the treatment and therefore improve the prognosis of affected children. A new approach to evaluating X-ray images using artificial intelligence may facilitate the detection of suspicious lesions and, hence, accelerate the referral to a specialized center. We implemented a Vision Transformer model for image classification of healthy and pathological X-rays. To tackle the limited amount of data, we used a pretrained model and implemented extensive data augmentation. Discrete parameters were described by incidence and percentage ratio and continuous parameters by median, standard deviation and variance. For the evaluation of the model accuracy, sensitivity and specificity were computed. The two-entity classification of the healthy control group and the pathological group resulted in a cross-validated accuracy of 89.1%, a sensitivity of 82.2% and a specificity of 93.2% for test groups. Grad-CAMs were created to ensure the plausibility of the predictions. The proposed approach, using state-of-the-art deep learning methodology to detect bone tumors on knee X-rays of children has achieved very good results. With further improvement of the algorithm, enlargement of the dataset and removal of potential biases, this could become a useful additional tool, especially to support general practitioners for early, accurate and specific diagnosis of bone lesions in young patients.

Background/Aim: Ewing sarcoma is a highly malignant tumour predominantly found in children. The radiological signs of this malignancy can be mistaken for acute osteomyelitis. These entities require profoundly different treatments and result in completely different prognoses. The purpose of this study was to develop an artificial intelligence algorithm, which can determine imaging features in a common radiograph to distinguish osteomyelitis from Ewing sarcoma. Materials and Methods: A total of 182 radiographs from our Sarcoma Centre (118 healthy, 44 Ewing, 20 osteomyelitis) from 58 different paediatric (≤18 years) patients were collected. All localisations were taken into consideration. Cases of acute, acute on chronic osteomyelitis and intraosseous Ewing sarcoma were included. Chronic osteomyelitis, extra-skeletal Ewing sarcoma, malignant small cell tumour and soft tissue-based primitive neuroectodermal tumours were excluded. The algorithm development was split into two phases and two different classifiers were built and combined with a Transfer Learning approach to cope with the very limited amount of data. In phase 1, pathological findings were differentiated from healthy findings. In phase 2, osteomyelitis was distinguished from Ewing sarcoma. Data augmentation and median frequency balancing were implemented. A data split of 70%, 15%, 15% for training, validation and hold-out testing was applied, respectively. Results: The algorithm achieved an accuracy of 94.4% on validation and 90.6% on test data in phase 1. In phase 2, an accuracy of 90.3% on validation and 86.7% on test data was achieved. Grad-CAM results revealed regions, which were significant for the algorithms decision making. Conclusion: Our AI algorithm can become a valuable support for any physician involved in treating musculoskeletal lesions to support the diagnostic process of detection and differentiation of osteomyelitis from Ewing sarcoma. Through a Transfer Learning approach, the algorithm was able to cope with very limited data. However, a systematic and structured data acquisition is necessary to further develop the algorithm and increase results to clinical relevance.

Background/Objectives: The quality of surgical margins has been shown to be a prognostic factor in many sarcoma entities, yet its role in osteosarcoma remains controversial. While previous studies have shown that the outcome was not related to the margin width in bone, the impact of the extraosseous margin width (margin at the soft tissue invasion)—which needs to be close sometimes due to neurovascular structures—needs to be assessed. This study aims to evaluate the influence of soft tissue surgical margins on local recurrence and overall survival in patients with high-grade osteosarcoma. Methods: We conducted a retrospective, single-center study including 75 patients treated for high-grade osteosarcoma. All patients underwent standardized neoadjuvant chemotherapy followed by complete surgical resection. Patients were stratified into three groups based on the histological margin width of the extraosseous parts: group 1 (<1 mm), group 2 (1–5 mm), and group 3 (≥5 mm). Primary endpoints were local recurrence and overall survival (OS), analyzed using Kaplan–Meier estimates, log-rank tests, and Cox regression. Results: Local recurrence occurred in seven patients (9%). Although the overall comparison between the three groups was not statistically significant (p = 0.074), a subgroup analysis revealed a significantly higher recurrence rate in patients with margins < 1 mm compared to those with wider margins (p = 0.024). No significant differences in overall survival (OS) were observed between the groups (p = 0.896). Tumor location, metastatic status, and UICC stage were significant predictors for both endpoints in univariate analysis. However, none of these association were confirmed in multivariate analyses. Conclusions: Very close surgical margins (<1 mm) may increase the risk of local recurrence in high-grade osteosarcoma; however, they do not appear to affect overall survival.

Background: A pathological/inflamed cellular microenvironment state is an additional risk factor for any cancer type. The importance of a chronic inflammation state in most diffuse types of tumour has already been analysed, except for in Ewing’s sarcoma. It is a highly malignant blue round cell tumour, with 90% of cases occurring in patients aged between 5 and 25 years. Worldwide, 2.9 out of 1,000,000 children per year are affected by this malignancy. The aim of this retrospective study was to analyse the role of C-reactive protein (CRP) as a prognostic factor for Ewing’s sarcomas. Methods: This retrospective study at Klinikum rechts der Isar included 82 patients with a confirmed Ewing’s sarcoma diagnosis treated between 2004 and 2019. Preoperative CRP determination was assessed in mg/dL with a normal value established as below 0.5 mg/dL. Disease-free survival time was calculated as the time between the initial diagnosis and an event such as local recurrence or metastasis. Follow-up status was described as death of disease (DOD), no evidence of disease (NED) or alive with disease (AWD). The exclusion criteria of this study included insufficient laboratory values and a lack of information regarding the follow-up status or non-oncological resection. Results: Serum CRP levels were significantly different in patients with a poorer prognosis (DOD) and in patients who presented distant metastasis (p = 0.0016 and p = 0.009, respectively), whereas CRP levels were not significantly different in patients with local recurrence (p = 0.02). The optimal breakpoint that predicted prognosis was 0.5 mg/dL, with a sensitivity of 0.76 and a specificity of 0.74 (AUC 0.81). Univariate CRP analysis level >0.5 mg/dL revealed a hazard ratio of 9.5 (95% CI 3.5–25.5). Conclusions: In Ewing’s sarcoma cases, we consider a CRP pretreatment value >0.5 mg/dL as a sensitive prognostic risk factor indication for distant metastasis and poor prognosis. Further research with more data is required to determine more sensitive cutoff levels.

Background/Objectives: One of the most important diagnostic tools in bone tumors is X-rays. Preliminary and, in the case of some benign lesions, definitive diagnoses are formed using this basic tool. Part of the decision making in this stage is based on statistical probability using the patient’s age, as well as the incidence and predilection sites of different entities. The information used today is based on older and fragmented data. To verify the underlying principles, we retrospectively evaluated all bone tumors in children and adolescents treated by our tertiary center in the last 20 years. Methods: For this retrospective study, patients under the age of 18 years suffering from histopathologically verified bone tumors were evaluated. Data were retrieved from our local musculoskeletal tumor database. Results: We were able to include 420 children treated for bone tumors in our tertiary center. The cohort consisted of 335 benign and 85 malignant lesions. The most common lesions were 137 osteochondromas; the malignant tumors consisted mainly of osteosarcomas (53) and Ewing’s sarcomas (28). The primary predilection sites were the metaphyses of long bones. Conclusions: We were able to confirm and supplement the fragmentary data of these rare diseases using our own cohort.

Background: Chondrosarcomas (CS) are a rare and heterogenic group of primary malignant bone tumors. In the literature, data on prognostic factors in chondrosarcomas are scarce, and most studies are limited by a short follow-up. The aim of this retrospective study was therefore to determine factors associated with the survival and local recurrence of chondrosarcomas and to compare the results with previous studies. Methods: We retrospectively evaluated 77 patients who were treated for chondrosarcoma of the extremities or pelvis at our tertiary sarcoma center between 1998 and 2007. Patient-related data (age, sex, etc.), tumor characteristics (localization, grading, presence of metastases, etc.), and treatment-related data (previous surgical treatment, type of local treatment, surgical margins, etc.) were evaluated and analyzed for possible correlation with patients’ outcomes. A statistical analysis was performed, including multivariate analysis. Results: The mean survival in our patients was 207 months, which resulted in a five-year survival rate of 76%. Negative prognostic factors for survival were histopathological grading, a patient aged over 70 years, and metastatic disease. The quality of the resection (clear or contaminated margins) negatively influenced both the development of local recurrence and survival too, at least in the univariate analysis. In contrast, factors such as tumor localization (extremities vs. pelvis), pathological fractures, or an initial inadequate resection elsewhere had no significant effect on survival. Conclusions: In accordance with results in the literature, the survival of patients with chondrosarcomas is mainly influenced by factors such as tumor grading, age, and metastases. However, complete resection remains paramount for the outcome in patients with chondrosarcoma—a primary malignant bone tumor with limited alternative treatment options.