Ultrasound and Magnetic Resonance Imaging of Burned-Out Testicular Tumours: The Diagnostic Keys Based on 48 Cases

Abstract

The spontaneous regression of testicular germ-cell tumours is a rare event whose mechanisms have yet to be elucidated. In the majority of published cases, tumour regression is concomitant with the metastatic development of the disease. Residual lesions, often referred to as burned-out testicular tumours (BOTTs), are difficult to diagnose due to the paucity of published data, especially in the field of imaging. The aim of this article is to describe the radiological signs of BOTTs on multimodal ultrasound and multiparametric MRI from a series of 48 patients whose diagnosis was confirmed histologically. The demographic, clinical and laboratory characteristics of the patients are studied, as well as the data of the imaging examinations, including conventional scrotal ultrasound, shear-wave elastography, contrast-enhanced ultrasound (CEUS) and multiparametric MRI. A total of 27 out of 48 patients were referred for investigation of primary testicular lesion following the discovery of retroperitoneal metastases, 18/48 patients were referred because of lesions suspected on an ultrasound that was performed for an infertility work-up, and 3/48 were referred because of scrotal clinical signs. Of these last 21 patients (infertility work-up/scrotal clinical sign), 6 were found to be metastatic on the extension work-up. Of the 48 orchiectomy specimens, tumour involution was complete in 41 cases, and a small active contingent remained in 7 cases, with 6 suspected upon advanced US and MRI. Typically, BOTTs appear on a conventional ultrasound as ill-delineated, hypoechoic and hypovascular nodular areas. Clustered microliths (60.4%) and macrocalcifications (35.4%) were frequent. Shear-wave elastography showed areas of focal induration (13.5 ± 8.4 vs. 2.7 ± 1.2 kPa for normal parenchyma, p < 0.01) in 92.5% of the patients for whom it was performed, and contrast ultrasonography demonstrated hypoperfusion of these lesions. Of the 42 MRIs performed, BOTTs corresponded to nodules on T2-weighted sequences (hyposignal) with significantly increased ADC values compared with healthy parenchyma (2 ± 0.3 versus 1.3 ± 0.3 × 10−3 mm2/s, p < 0.01) and an enhancement defect after injection. This enhancement defect overlapped the lesions visible on T2-weighted sequences in most cases. In the case of predominant partial regression, an enhanced portion after contrast injection was visible on MRI in all seven patients of our series, and in six of them a focal diffusion restriction zone was also present. Spontaneously involuted testicular germ-cell tumours have specific radiological signs, and all of the mentioned examinations contribute to this difficult diagnosis, even histologically, because there is no tumour cell left. These signs are similar whether the patient is initially symptomatic metastatic or whether the discovery is fortuitous on the occasion of an infertility work-up, and whatever the seminomatous or non-seminomatous nature of the germ-cell tumour, when this can be determined. The appearance of regressed germ-cell tumours is often trivialized, which can lead to the wrong diagnosis of an extra gonadal germ-cell tumour (in metastatic patients) or of scarring from an acute event such as trauma or infection, which is not recognized or forgotten. In our series, two patients had an unrecognized diagnosis in their history, with local and/or distant recurrence. An improvement in diagnosing burned-out tumours, combining advanced US and MRI, is necessary in order to optimize patient management, with special attention paid to asymptomatic patients, to prompt extension screening and orchiectomy with analysis of the whole testis. This may reveal a persistent viable tumour or lesions of germinal neoplasia in situ, which are precursors of testicular germ-cell tumours.