Ultrasound Matches Advanced Imaging in Cervical Cancer

Preoperative imaging helps define tumour extent, lymph-node status and treatment selection in cervical cancer. The CANNES trial, published in Ultrasound in Obstetrics & Gynecology, compared ultrasound, PET/CT and diffusion-weighted MRI for preoperative pelvic lymph-node assessment. It assessed whether ultrasound could match the diagnostic accuracy of the other two methods before surgery. The trial included women with histologically confirmed cervical cancer across early and more advanced stages. Ultrasound and PET/CT were mandatory, while DW-MRI was optional. Imaging performance was checked against final histopathology, with postoperative imaging used in selected discordant cases.

Imaging Before Surgical Staging

The trial enrolled women treated at three European gynaecological oncology centres in Prague, Rome and Madrid. Eligible patients had cervical cancer confirmed on histopathology, were suitable for surgical lymph-node staging and underwent surgery within six weeks after imaging. Patients with pregnancy, non-primary cervical cancer, another active malignancy, FIGO Stage IA1 disease without lymphovascular space invasion or FIGO Stage IV disease were not included.

Must Read: Parametrial Endometriosis Detection with Transvaginal US

All patients underwent ultrasound and PET/CT before surgery. Most also underwent DW-MRI, provided that it was available and not contraindicated. Each imaging method was interpreted independently, and examiners were blinded to the results of the other modalities. This reduced the risk that one imaging result would influence another.

Ultrasound assessment combined transabdominal imaging with a transvaginal or transrectal approach. Grayscale and Doppler evaluation were used to assess lymph-node morphology and vascular features. PET/CT assessment relied on 18F-FDG uptake together with CT morphology. DW-MRI assessment used lymph-node size and morphological features, including shape, signal intensity, borders and cortical asymmetry.

Surgical staging included sentinel lymph-node biopsy, pelvic lymph-node dissection, node sampling or removal of suspicious nodes. Lymph nodes considered positive on preoperative imaging were intended for removal. Final histopathology served as the main reference standard. When imaging suggested nodal involvement, but histopathology did not confirm it, postoperative imaging could be used to clarify whether disease persisted.

Comparable Accuracy Across Modalities

A total of 120 patients were included in the final pelvic lymph-node assessment. All underwent ultrasound and PET/CT, and most also underwent DW-MRI. Final histopathology confirmed pelvic nodal involvement in a substantial minority of patients. Most positive cases involved macrometastases, while a smaller group had micrometastases only. Postoperative imaging identified persistent nodal disease in two additional discordant cases.

For pelvic lymph-node macrometastases, ultrasound reached diagnostic accuracy close to PET/CT and DW-MRI. Sensitivity and specificity were also broadly similar across the three methods. Ultrasound met the predefined non-inferiority margin when compared with both PET/CT and DW-MRI. This applied to the main per-patient assessment and was also seen in most more detailed pelvic analyses.

For overall pelvic lymph-node involvement, including both micro- and macrometastases, diagnostic accuracy remained comparable between the three modalities. Ultrasound again met the non-inferiority threshold against PET/CT and DW-MRI. DW-MRI showed the highest specificity in the overall pelvic assessment, but ultrasound and PET/CT remained close in overall performance.

The results indicate that ultrasound can provide a reliable preoperative assessment of pelvic lymph-node status when performed by experienced examiners using specialised abdominopelvic techniques. The results also show that PET/CT and DW-MRI do not clearly outperform ultrasound for the main pelvic nodal outcomes assessed in the trial.

Limits in Low-Volume Disease

Low-volume nodal disease remained difficult to detect with imaging. Patients with micrometastases only were frequently missed by all three modalities. This pattern was particularly relevant in early-stage disease, where nodal metastases were less common and metastatic deposits were often smaller. Sensitivity was lower in this setting, while specificity remained high.

In more advanced cervical cancer, nodal involvement was more frequent, and metastatic deposits were generally larger. Sensitivity improved for ultrasound, PET/CT and DW-MRI. However, specificity was lower than in early-stage disease. The trial linked some false-positive findings to reactive or inflammatory changes, including cases after cervical diagnostic conisation.

False-negative findings were most often associated with micrometastases or small macrometastases. The diameter of metastatic foci was smaller in false-negative cases than in true-positive cases. This reinforces the limitation that imaging may miss small nodal deposits even when overall diagnostic accuracy is good.

Para-aortic assessment was also performed, but only a small number of patients had para-aortic lymph-node involvement. This limited the strength of the para-aortic results. Specificity remained high across the modalities, but the small number of positive cases meant that sensitivity estimates were less stable. No adverse events related to ultrasound, PET/CT or DW-MRI were recorded.

Ultrasound achieved non-inferior diagnostic accuracy compared with PET/CT and DW-MRI for preoperative pelvic lymph-node assessment in cervical cancer. The results support ultrasound as a reliable and accessible option when performed in expert settings. However, low-volume nodal disease remains difficult to detect with imaging, particularly in early-stage disease. Imaging alone therefore cannot replace sentinel lymph-node biopsy with ultrastaging when accurate nodal assessment is required. False-positive findings after cervical diagnostic conisation also require careful interpretation.

Source: Ultrasound in Obstetrics & Gynecology

Image Credit: iStock