Radiomics of MRI for pretreatment prediction of pathologic complete response, tumor regression grade, and neoadjuvant rectal score in patients with locally advanced rectal cancer undergoing neoadjuvant chemoradiation: an international multicenter study

OBJECTIVE

To investigate whether pretreatment MRI-based radiomics of locally advanced rectal cancer (LARC) and/or the surrounding mesorectal compartment (MC) can predict pathologic complete response (pCR), neoadjuvant rectal (NAR) score, and tumor regression grade (TRG).

METHODS

One hundred thirty-two consecutive patients with LARC who underwent neoadjuvant chemoradiation and total mesorectal excision (TME) were retrospectively collected from 2 centers in the USA and Italy. The primary tumor and surrounding MC were segmented on the best available T2-weighted sequence (axial, coronal, or sagittal). Three thousand one hundred ninety radiomic features were extracted using a python package. The most salient radiomic features as well as MRI parameter and clinical-based features were selected using recursive feature elimination. A logistic regression classifier was built to distinguish between any 2 binned categories in the considered endpoints: pCR, NAR, and TRG. Repeated k-fold validation was performed and AUCs calculated.

RESULTS

There were 24, 87, and 21 T4, T3, and T2 LARCs, respectively (median age 63 years, 32 to 86). For NAR and TRG, the best classification performance was obtained using both the tumor and MC segmentations. The AUCs for classifying NAR 0 versus 2, pCR, and TRG 0/1 versus 2/3 were 0.66 (95% CI, 0.60-0.71), 0.80 (95% CI, 0.74-0.85), and 0.80 (95% CI, 0.77-0.82), respectively.

CONCLUSION

Radiomics of pretreatment MRIs can predict pCR, TRG, and NAR score in patients with LARC undergoing neoadjuvant treatment and TME with moderate accuracy despite extremely heterogenous image data. Both the tumor and MC contain important prognostic information.

KEY POINTS

• Machine learning of rectal cancer on images from the pretreatment MRI can predict important patient outcomes with moderate accuracy. • The tumor and the tissue around it both contain important prognostic information.

Ultrasound Guided versus Direct Vein Puncture in Central Venous Port Placement

Introduction

This study will report our experience on positioning of totally implanted venous catheter system (port-a-cath) as compared to ultrasound guidance versus blind technique.

Materials and Methods

From July 1996 to November 1999 in the vascular suite of the Europen Institute of Oncology, 427 port-a-cath were implanted in patients with neoplastic disease. All devices were implanted through the subclavian vein. 198 with ultrasound guided puncture and 229 following anatomical landmarks. All patients underwent a close and specific clinical and instrumental follow-up to evaluate possible complications.

Results

Use of Ultrasound (US) in subclavian vein catheterization has reduced the number of puncture attempts, with a better patient complicance, allowing a faster procedure and reducing peri-procedural complications. Ultrasound technique has shown reduction in early complications. In fact in our experience we had no pneumothorax events by using ultrasound guidance, versus 11 events with blind technique. US has shown no reduction in late complications: 3 thrombosis versus 6 thrombosis with US guidance, and 3 fractured and embolized catheter versus 2 cases. Fibrin-cuff percentage was the same in the two groups with only 1 case, as the dislocation of the catheter tip in the jugular vein with 2 cases in both groups. Moreover US let us to avoid the arterial puncture and to perform a more peripheral puncture of the subclavian vein, reducing the risk of “pinch-off” phenomenon and of haematoma, with no cases reported under US guidance versus 1 case respectively in direct vein puncture.

Conclusions

US guided puncture of subclavian vein for the implantation of venous catheter system is faster by reducing procedure time, it is easier for the operator and safer for the patient than blind technique by exposing anatomical structures. Morevover US guidance reduces early complications and limits costs.

Prospective comparison of computed tomography enterography and magnetic resonance enterography for assessment of disease activity and complications in ileocolonic Crohn's disease

BACKGROUND

Studies comparing magnetic resonance enterography (MRE) and computerized tomography enterography (CTE) for Crohn's disease (CD) are scarce.

METHODS

The aim of this study was to prospectively compare the sensitivity, specificity, and accuracy of abdominal MRE and CTE to assess disease activity and complications (fistulas, strictures) in ileocolonic CD. A total of 44 patients (23 male; 21 female; mean age 44) with ileocolonic CD underwent both MR and CT in a short time interval (mean 5 days). A 16-slice CT with intravenous contrast and an MRI with oral and paramagnetic intravenous contrast were performed. Ileocolonoscopy was used as the reference standard. Sensitivity values of CT and MR for detection of extraenteric signs of disease were compared with the McNemar test, with results of imaging studies, surgery, and physical examination as reference standards.

RESULTS

No significant differences in sensitivity, specificity, and accuracy were observed between MRE and CTE regarding the following parameters at the patient level: localization of CD (P = 1.0), bowel wall thickening (P = 1.0), bowel wall enhancement (P = 1.0), enteroenteric fistulas (P = 0.08), detection of abdominal nodes (P = 1.0), and perivisceral fat enhancement (P = 0.31). MR was significantly superior compared to CT in detecting strictures (P = 0.04). Per segment analysis showed that MRE was significantly superior to CTE in detecting ileal wall enhancement (P = 0.02).

CONCLUSIONS

MR and CT are equally accurate to assess disease activity and bowel damage in CD. MR may be superior to CT in detecting intestinal strictures and ileal wall enhancement. MR may represent an alternative technique to CT in assessing ileocolonic CD.

Transoral surgical removal of a giant fibrolipoma of the esophagus: case report

An esophageal mass of more than 20 cm in length was diagnosed in a patient who presented with persistent dysphagia. Diagnosis of an endo-esophageal tumour was made by barium swallow; esophagoscopy confirmed the presence of a capsulated pink endo-esophageal mass. MRI confirmed the presence of a large capsulated mass within the esophagus, that appeared to be adipose tissue; a small stalk originating at the level of the upper esophageal sphincter was described and the polyp extended down to the gastroesophageal junction. Demonstration of the site and length of the stalk allowed a transoral removal of the mass, performed through a Weerda diverticuloscope (Karl Storz Endoskopie Gmbh, Tuttlingen Germany), a technique that has never been described before. Histology confirmed the mass as a fibrolipoma. The authors discuss both the role of MRI in diagnosis and treatment planning and the technique of transoral excision.