Biopsy of bone and soft tissue tumours: hints and hazards

Radiologists with assistance of deep learning can achieve overall accuracy of benign-malignant differentiation of musculoskeletal tumors comparable with that of pre-surgical biopsies in the literature

PURPOSE

The purpose of this study was to assess if radiologists assisted by deep learning (DL) algorithms can achieve diagnostic accuracy comparable to that of pre-surgical biopsies in benign-malignant differentiation of musculoskeletal tumors (MST).

METHODS

We first conducted a systematic review of literature to get the respective overall diagnostic accuracies of fine-needle aspiration biopsy (FNAB) and core needle biopsy (CNB) in differentiating between benign and malignant MST, by synthesizing data from the articles meeting our inclusion criteria. To compared against the accuracies reported in literature, we then invited 4 radiologists, respectively with 2 (A), 6 (B), 7 (C), and 33 (D) years of experience in interpreting musculoskeletal MRI to perform diagnostic tests on our own dataset (n = 62), with and without assistance of a previously developed DL algorithm. The gold standard for benign-malignant differentiation was histopathologic confirmation or clinical/radiographic follow-up.

RESULTS

For FNAB, a meta-analysis containing 4604 samples met the inclusion criteria, with the overall diagnostic accuracy reported to be 0.77. For CNB, an overall accuracy of 0.86 was derived by synthesizing results from 7 original research articles containing a total of 587 samples. On our internal MST dataset, the invited radiologists, respectively, achieved diagnostic accuracies of 0.84 (A), 0.89 (B), 0.87 (C), and 0.90 (D), with the assistance of DL.

CONCLUSION

Use of DL algorithms on musculoskeletal dynamic contrast-enhanced MRI improved the benign-malignant differentiation accuracy of radiologists to a level comparable to that of pre-surgical biopsies. The developed DL algorithms have a potential to lower the risk of miss-diagnosing malignancy in radiological practice.

Deep Learning Assisted Diagnosis of Musculoskeletal Tumors Based on Contrast-Enhanced Magnetic Resonance Imaging

BACKGROUND

Misdiagnosis of malignant musculoskeletal tumors may lead to the delay of intervention, resulting in amputation or death.

PURPOSE

To improve the diagnostic efficacy of musculoskeletal tumors by developing deep learning (DL) models based on contrast-enhanced magnetic resonance imaging and to quantify the improvement in diagnostic performance obtained by using these models.

STUDY TYPE

Retrospective.

POPULATION

Three hundreds and four musculoskeletal tumors, including 212 malignant and 92 benign lesions, were randomized into the training (n = 180), validation (n = 62) and testing cohort (n = 62).

FIELD STRENGTH/SEQUENCE

A 3 T/T₁ -weighted (T₁ -w), T₂ -weighted (T₂ -w), diffusion-weighted imaging (DWI), and contrast-enhanced T1-weighted (CET₁ -w) images.

ASSESSMENT

Three DL models based, respectively, on the sagittal, coronal, and axial MR images were constructed to predict the malignancy of tumors. Blinded to the prediction results, a group of specialists made independent initial diagnoses for each patient by reading all image sequences. One month after the initial diagnoses, the same group of doctors made another round of diagnoses knowing the malignancy of each tumor predicted by the three models. The reference standard was the pathological diagnosis of malignancy.

STATISTICAL TESTS

Sensitivity, specificity, and accuracy (all with 95% confidential intervals [CI]) corresponding to each diagnostic test were computed. Chi-square tests were used to assess the differences in those parameters with and without DL models. A P value < 0.05 was considered statistically significant.

RESULTS

The developed models significantly improved the diagnostic sensitivities of two oncologists by 0.15 (95% CI: 0.06-0.24) and 0.36 (95% CI: 0.24-0.28), one radiologist by 0.12 (95% CI: 0.04-0.20), and three of the four orthopedists, respectively, by 0.12 (95% CI: 0.04-0.20), 0.29 (95% CI: 0.18-0.40), and 0.23 (95% CI: 0.13-0.33), without impairing any of their diagnostic specificities (all P > 0.128).

DATA CONCLUSION

The DL models developed can significantly improve the performance of doctors with different training and experience in diagnosing musculoskeletal tumors.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Primary synovial sarcomas in the paediatric and young adult population: A pictorial review

PURPOSE

To present a pictorial essay of paediatric primary synovial sarcomas from common and less documented anatomical locations. To review the literature for the imaging characteristics and prognostic factors of this rare but important childhood malignancy.

METHOD

24 primary synovial sarcoma cases (17 male, 7 female with an age range 4-21 years) were reviewed in a collaborative effort between St Jude Children's Research Hospital and Great Ormond Street Hospital for Children. Images from 19 cases were selected for inclusion, to demonstrate the spectrum of appearances across imaging modalities, in a range of different anatomical locations (upper limb, lower limb, chest/abdomen/pelvis, and head and neck). A literature review depicting the typical radiological features and the prognostic significance of these features, was also conducted.

RESULTS AND CONCLUSIONS

Primary synovial sarcoma can occur in any anatomical location, but typically within the extremities and often in close association with joints. Rarer anatomical locations described in our essay include the gastrohepatic ligament and femoral nerve sheath. We detail the salient imaging characteristics, including the T2 'triple signal' pattern which is believed to be highly specific for this particular sarcoma and in many cases predicts a poor outcome. Other poor prognostic factors include haemorrhage, lack of calcification and tumour size >10 cm. A broad range of radiological appearances are described, and in some cases related to anatomical position and size, however the presence of a soft tissue mass close to a joint in a young patient are suggestive of this diagnosis.

Bone reconstruction after malignant tumour resection using a motorized lengthening intramedullary nail in adolescents: preliminary results

PURPOSE

Bone reconstruction after malignant tumour of the lower limbs is a challenge. Our aim was to investigate the results of bone lengthening and transport using the Fitbone motorized lengthening nail.

METHODS

Eight patients were included. Two patients had had a tumour involving soft tissue only without bone resection. In six patients, the initial bone defect was 15.5 cm (8 to 24). The first step of reconstruction had consisted of temporary fixation and a cement spacer. The second step had consisted of bone grafting in five patients. One patient was managed with bone transport. Eventually, five patients had a limb-length discrepancy (LLD) managed with bone lengthening only. Two patients had a LLD and a nonunion managed with bone transport and lengthening. Mean age at bone lengthening was 15.2 years (11 to 19). Mean follow-up was 30.5 months (10 to 48).

RESULTS

In all, 11 nails were implanted in eight patients (eight femurs, three tibias). Mean length gain per procedure was 54.5 mm (30 to 80). Mean healing index was 48.4 days per cm (22.6 to 85.7). The complication rate was 18%. In two cases we observed a loose locking screw, which was revised. In all cases the lengthening involved the short bone (femur or tibia). Mean Musculoskeletal Tumor Society (MSTS) score improved from 52.7 (16.6 to 73.3) to 79.9 (63.3 to 96.6).

CONCLUSIONS

Bone reconstruction using a motorized intramedullary nail stands as a safe and reliable alternative after malignant tumour. It allows biological reconstruction with satisfactory clinical and radiological outcome and low complication rate.

LEVEL OF EVIDENCE

III.

The presence of tumor extrachomosomal circular DNA (ecDNA) as a component of liquid biopsy in blood

In molecular biology covalently closed circular DNAs are able to passthrough double layer of eukaryotic cellular membrane. Very recently the presence of circular extra chromosomal DNA (ecDNA) has been shown which are different in seventeen different types of cancers. In fact, ecDNA are the tricky way of oncogenes to increase their copy number. We hypothesis the presence of ecDNA in the blood of cancer patients as a subpopulation of liquid biopsy. On the occasion of their presence in blood they will be very beneficial to cover the small amount of cell frees DNA (cfDNA). Isolation and characterization of ecDNA will be possible by a sensitive method entitled Circle-Seq. The origin of tumor more than its prognosis and diagnosis will be possible in the easiest way by using ecDNA.

Diagnosis and treatment of soft-tissue sarcomas of the extremities and trunk

The relatively low incidence and often atypical clinical presentation of soft-tissue sarcomas (STS) impedes early and adequate diagnosis. Patients may report on recently enlarged soft-tissue swellings, infrequently complain of painful lesions, or even have no symptoms at all.

A thorough diagnostic work-up is essential in order to distinguish between benign soft-tissue tumours and STSs. Patient history, clinical features and radiological findings all help in assessing the underlying pathology. ‘Worrying’ features such as recent increase in size, deep location relative to the fascia, a tumour exceeding 4 cm in size, and invasive growth patterns seen on imaging should prompt verification by biopsy.

Even though acquisition of biopsy material may be incomplete, one should bear in mind some essential rules. Regardless of the biopsy technique applied, the most direct route to the lump in question should be identified, contamination of adjacent structures should be avoided and a sufficient amount of tissue acquired.

Treatment of STS is best planned by a multidisciplinary team, involving experts from various medical specialities. The benchmark therapy consists of en bloc resection of the tumour, covered by a safety margin of healthy tissue. Depending on tumour histology, grade, local extent and anatomical stage, radiotherapy, chemotherapy and isolated hyperthermic limb perfusion may be employed.

Due to the complexity of treatment, any soft-tissue swelling suspected of malignancy is best referred directly to a sarcoma centre, where therapeutic management is carefully planned by an experienced multidisciplinary team.

Cite this article: EFORT Open Rev 2017;2:421-431. DOI: 10.1302/2058-5241.2.170005

Osteolytic lesions of the calcaneus: results from a multicentre study

PURPOSE

Tumours of the calcaneus are exceedingly rare and the correct diagnosis is often missed. X-rays are the standard clinical examination tool and therefore we wanted to discover whether X-rays alone were a sufficient diagnostic tool for these tumours. Diard's classification was applied to define whether different types of lesions were characteristically distributed in the bone and in addition we analysed whether type and/or duration of symptoms were possible indicators of malignancy.

METHODS

Ninety-two patients' files (59 men and 33 women) were retrospectively reviewed. Seventy-five patients with a mean age at surgery of 28 years (range five to 78) were surgically treated. Parameters analysed were sex, age at surgery, side, type and duration of symptoms, tentative diagnosis, biopsy prior to surgery, operative procedure, recurrence rate, revision and localisation of the lesion according to Diard. For each lesion the first documented radiological diagnosis and-in cases of malignancy-Enneking's classification was applied.

RESULTS

Discrepancies between the radiological and definitive histological diagnosis occurred in 38 (41 %) of 92 cases. In eight (osteosarcoma n = 5, Ewing's sarcoma n = 2, metastases n = 1) of 17 malignant cases radiological examination initially gave no evidence of malignancy, resulting in an unplanned excision ("whoops procedure") in three cases of osteosarcoma. Applying Diard's system trabecular area 6 (radiolucent area) was highly affected in 64 (80 %) of 80 investigated plain X-rays, whereas areas 1 and 5 were affected in nine (11 %) and 16 (20 %) cases only.

CONCLUSIONS

In each case of an osteolytic lesion of the calcaneus a malignant tumour must be ruled out, and thus preoperative plain X-rays in two planes alone are not sufficient and should therefore be followed by magnetic resonance imaging. Applying the Diard system different types of lesions are not characteristically distributed in the bone. Increasing pain for more than ten days without previous trauma should always justify further examinations.

[Partial pelvic resection (internal hemipelvectomy) and endoprosthetic replacement in periacetabular tumors]

OBJECTIVE

Treatment of tumors of the pelvic girdle by resection of part or all of the innominate bone with preservation of the extremity. Implantation and stable fixation using a custom-made megaprosthesis to restore painless joint function and loading capacity. The surgical goal is to obtain a wide surgical margin and local tumor control.

INDICATIONS

Primary bone and soft tissue sarcomas, benign or semi-malignant aggressive lesions, metastatic disease (radiation resistance and/or good prognosis).

CONTRAINDICATIONS

Limited life expectancy and poor physical status, extensive metastatic disease, persistent deep infection or recalcitrant osteomyelitis, poor therapeutic compliance, local recurrence following a previous limb-sparing resection, extensive infiltration of the neurovascular structures and the intra- and extrapelvic soft tissues.

SURGICAL TECHNIQUE

Levels of osteotomy are defined preoperatively by a CT-controlled manufactured three-dimensional 1:1 model of the pelvis. Using these data, the custom-made prosthesis and osteotomy templates are then constructed by the manufacturer. The anterior (internal, retroperitoneal) and posterior (extrapelvic, retrogluteal) aspects of the pelvis are exposed using the utilitarian incision surgical approach. The external iliac and femoral vessels are mobilized as they cross the superior pubic ramus. The adductor muscles, the rectus femoris and sartorius muscle are released from their insertions on the pelvis and the obturator vessels and nerve are transected. If the tumor extends to the hip joint, the femur is transected at a level distal to the intertrochanteric line to ensure hip joint integrity and to prevent tumor contamination. A large myocutaneous flap with the gluteus maximus muscle is retracted posteriorly. The pelvitrochanteric and small gluteal muscles are divided near their insertion in the upper border of the femur. To release the hamstrings and the attachment of the sacrotuberous ligament, the ischial tuberosity is exposed. After osteotomy using the prefabricated templates, the pelvis is released and the specimen is removed en bloc. The custom made prosthesis can either be fixed to the remaining iliac bone or to the massa lateralis of the sacrum. The released muscles are refixated on the remaining bone or the implant.

POSTOPERATIVE MANAGEMENT

Time of mobilization and degree of weight-bearing depends on the extent of muscle resection. Usually partial loading of the operated limb with 10 kg for a period of 6-12 weeks, then increased loading with 10 kg per week. Thrombosis prophylaxis until full weight bearing. Physiotherapy and gait training. At follow-up, patients are monitored for local recurrence and metastases using history, physical examination, and radiographic studies.

RESULTS

Between 1994 and 2008, 38 consecutive patients with periacetabular tumors were treated by resection and reconstruction with a custom-made pelvic megaprosthesis. The overall survival of the patients was 58% at 5 years and 30% at 10 years. One or more operative revisions were performed in 52.6% of the patients. The rate of local recurrence was 15.8%. Deep infection (21%) was the most common reason for revision. In two of these cases (5.3%), a secondary external hemipelvectomy had to be performed. There were four cases of aseptic loosening (10.5%) in which the prosthesis had to be revised. Six patients had recurrent hip dislocation (15.8%). In four of them a modification of the inserted inlay and an implantation of a trevira tube had to be performed respectively. Peroneal palsy occurred in 6 patients (15.8%) with recovery in only two. There were 4 operative interventions because of postoperative bleeding (10.5%). The mean MSTS score for 12 of the 18 living patients was 43.7%. In particular, gait was classified as poor and almost all patients were reliant on walking aids. However, most patients showed good emotional acceptance.

[Consensus diagnosis and therapy of soft tissue sarcoma]

Soft tissue sarcomas are heterogeneous tumours and relatively uncommon. There have been advances over the past years concerning pathology, clinical behaviour, diagnosis strategies and the treatment. To summarize these advances as well as making it public is one of the goals of the following consensus guidelines. But why do we need special guidelines for Austria? There are international guidelines published by the European Society of Medical Oncology (ESMO) and the National Comprehensive Cancer Network (NCCN). The cause is that we need an explanation of the matrix the ESMO and the NCCN gave according to our clinical practice, the local requirements and facilities in Austria. The following recommendations were drawn up following a consensus meeting of sarcoma specialists from the three high volume centres located at the medical universities in Austria. All fields of involved physicians from diagnosis to therapy worked together to know that soft tissue sarcomas are an interdisciplinary challenge and multimodal treatment is essential. For this reason, these guidelines not only explain but also give the state of the art and clear recommendations. One of the most important guidelines is that any patient with a suspected soft tissue sarcoma should be referred to one of the three university centres and managed by a specialist sarcoma multidisciplinary team. We hope that the consensus is helpful for the clinical practice and improves the quality of care for patients with soft tissue sarcomas in Austria.