Papillary renal cell carcinoma in contrast-enhanced ultrasound (CEUS) – A diagnostic performance study

BACKGROUND: Renal cell carcinomas (RCC) represent a heterogeneous group of hypo- and hypervascularized malignancies. Using contrast-enhanced ultrasound (CEUS) specific imaging features of clear cell (ccRCC), papillary (pRCC) and chromophobe RCC (chRCC) subtypes have been demonstrated. However, some RCCs show atypical imaging features making it difficult to distinguish between the subtypes. OBJECTIVE: This study was performed to evaluate the observed enhancement features of pRCC in CEUS and to assess the sensitivity in the diagnosis of suspected renal malignancies in a 10 year retrospective analysis at our institution. METHODS: The study population consisted of 60 patients with histologically confirmed pRCC. All patients underwent CEUS imaging between 2005 and 2015 as part of their diagnostic workup. RESULTS: In 45 out of 60 (75%) cases the examined pRCC showed typical hypoenhancement and wash-out. 15 out of 60 (25%) pRCC showed atypical enhancement features; in 14 cases the contrast enhancement indicated a ccRCC. 1 complex cyst was falsely reported as IIF lesion. 59 out of 60 malignancies were reported as malignant using CEUS resulting in a sensitivity of 98.4%. CONCLUSIONS: CEUS is an eligible imaging technique to visualize the contrast enhancement features of pRCC. However, up to 25% of pRCCs show an atypical enhancement pattern making it difficult to distinguish it from other renal lesions.

Keywords: Contrast-enhanced ultrasound; CEUS; renal cell carcinoma; RCC; enhancement features

Renal cell carcinoma (RCC) accounts for 2–3 % of all cancer cases and for 90% of all adult renal tumors. RCC represents not a single entity but a heterogeneous group of neoplasms. The most common subtypes are the clear cell RCC (ccRCC), papillary RCC (pRCC) and chromophobe RCCs (chRCC), they account for approximately 70%, 15–20% and 5%, respectively of all RCCs. Men are 2 to 3 times more affected of RCCs than women and median age at diagnosis is 65 years. Beside some rare genetic disorders the few established risk factors for RCCs are cigarette smoking, obesity and hypertension. Trichloroethylene or diabetes rank among the suspected risk factors [[1]].

Over the last decades the incidence of RCC increased at a rate of 2% per year which is mainly attributed to a more frequent use of imaging techniques as most RCCs are incidentally detected by ultrasound or cross-sectional imaging. This has led to a shift to smaller tumor size and early stage disease at the time of diagnosis [[3]]. Detecting incidental renal lesions in ultrasound or cross-sectional imaging is of course beneficial for patient treatment and outcome. However, physicians are confronted with the decision if and what kind of diagnostic workup is necessary for these incidental lesions. To avoid unnecessary biopsy or resection the primary goal is differentiating between benign renal lesions such as angiomyolipomas or oncocytomas and malignant neoplasms such as RCCs. In these cases contrast-enhanced ultrasound (CEUS) is a valid additional imaging technique without any radiation exposure and without any nephrotoxic or cardiotoxic effects. Furthermore, the contrast agent can be administered repeatedly and the imaging can be performed as bedside examination [[6]].

Over the past years multiple studies showed contrast-enhanced ultrasound to be a helpful imaging modality for characterizing complex cystic renal lesions applying the CEUS-Bosniak classification [[8]]. In relation to solid renal masses CEUS was shown to add great value to diagnosing clear cell RCC. However, distinguishing papillary RCC from lipid-poor angiomyolipoma or oncocytoma is difficult [[11]]. Since ccRCCs, pRCCs and chRCCs show a different outcome as well as localization and propensity to metastasize, research groups aimed for providing a non-invasive subtype diagnosis using CEUS [[3]]. Qualitative and quantitative analysis demonstrated that ccRCC show shorter time to peak enhancement, early hyperenhancement and wash-out in delayed phase compared to the adjacent renal parenchyma. pRCC and chRCC are described as homogeneously hypoenhancing lesions with fast wash-out compared to renal parenchyma in most of the cases. Thus, several studies conclude that CEUS is helpful to differentiate ccRCC from pRCC and chRCC but not to distinguish between pRCC and chRCC [[10], [15]].

This study was performed as a 10 year retrospective analysis at our institution to evaluate contrast enhancement features of pRCC in CEUS and compare the findings with current literature; and to assess the sensitivity in the diagnosis of suspected renal malignancies compared to histopathology.

The study population of this retrospective analysis consisted of 60 patients with histologically confirmed papillary renal cell carcinoma. All these patients had undergone CEUS imaging as part of their diagnostic workup of a suspected renal malignancy ahead of biopsy or tumor resection.

Prior to each imaging examination oral and written consent of the patient was obtained. The local ethic committee approved the study protocol and study data were collected according to the principles of the Helsinki/Edinburgh Declaration of 2002. The authors followed the ethical guidelines for publication in Clinical Hemorheology and Microcirculation [[17]].

CEUS imaging was performed between July 2005 and November 2015 and mean period of time between the ultrasound and the histological diagnosis was 81.6 days. CEUS imaging was performed using high-end ultrasound systems equipped with up-to date CEUS specific protocols available at the time of the examinations (GE Healthcare: LOGIQ E9; Philips Ultrasound HDI 5000, iU22, EPIQ 7, Affiniti; Samsung: RS80A 50 Prestige, RS80A Prestige; Siemens Ultrasound: Sequoia, S2000, S3000). The ultrasound probes used included CA1–7A, C6–1 HD, C5–1, C4–1 and V4–1 probes available at the time of the examinations. During the examination a low mechanical index (always < 0.4) was used to avoid unintentional early destruction of the microbubbles.

In all CEUS examinations a second-generation blood pool contrast agent (SonoVue®, Bracco, Milan, Italy) was used. The contrast agent was administered using a peripheral 18–22 G needle. In most cases 1.6 to 2.4 ml (min. 1.0 ml; max. 5.0 ml) contrast agent were administered as a bolus injection followed by a flush of 5 to 10 ml of 0.9% sodium chloride solution [[18]]. A single dose of contrast agent was sufficient in the majority of cases, in individual cases the injection of contrast agent was repeated up to two times. Images were acquired during the arterial phase 10 to 30 seconds after contrast administration and after 30 to 90 seconds for the late phase of kidney enhancement [[15]]. During the imaging acquisition patients were asked to suspend respiration if possible for up to 30 seconds. Patient with respiratory impairment were asked to breathe lightly and slowly. The mean overall examination time ranged from 3 to 5 minutes. No critical adverse reactions were observed after the administration of SonoVue®.

CEUS imaging studies were completed successfully in all 60 patients with satisfactory imaging quality. During the CEUS examination cine loops were acquired and stored in the picture archiving and communication system (PACS) of our institution. For the retrospective analysis imaging studies and patient record files were retrieved from the archiving system of our institution. We included all patients with histologically diagnosed papillary RCC.

All CEUS examinations were performed and interpreted by a single experienced radiologist with more than 15 years of experience in CEUS.

Out of 60 patients included in this retrospective study 50 patients (83.3%) were male and 10 patients (16.7%) were female. Mean age at the time of CEUS was 63.3 years (SD±11.1 years) ranging between 37.1 years and 83.0 years.

Contrast enhancement features of the examined pRCC were retrospectively analyzed and classified according to current literature, data given in percent. Diagnostic accuracy of CEUS compared to histopathology was tested using sensitivity. Due to patient selection based on the histopathology reports the data allows no statement regarding false positive or right negative reports.

In 56.7% of the cases the examined suspect lesion was located in the right kidney. 37.5% of the lesions were described in the lower pole with 31.3% in the middle and upper pole, respectively. In 45 out of 60 (75%) cases the examined renal lesion showed a rather homogeneous hypoenhancement and a wash-out of the contrast agent compared to the adjacent renal parenchyma. These lesions were reported as highly suspicious for pRCC (Figs. 1 and 2). 15 (25%) examined suspected renal lesions showed contrast enhancement features varying from the above-mentioned hypoenhancement and wash-out. In detail, 9 lesions showed an areal and another 5 lesions a circular hyperenhancement and fast wash-out. These lesions were reported as highly suspicious for a malignant renal lesion in particular for a ccRCC. 1 cystic lesion showed thickening of the cystic wall and septa as well as calcification but only a few scattered microbubbles. This cystic lesion was reported as CEUS-Bosniak IIF representing the one false negative report compared to histopathology as gold standard. Resulting in a sensitivity of 98.4%.

Graph: Fig.1 a) B-mode ultrasound shows an exophytic growing, hyperechoic renal lesion (white arrow). b) Color-Doppler shows no signs of hypervascularization (white arrow). c) During the arterial phase of CEUS the lesion is hypoenhanced (white arrow) in comparison to the adjacent renal parenchyma. d) In the late phase of CEUS the tumor shows a wash-out of the contrast agent compared to the adjacent renal parenchyma (white arrow). This lesion shows typical CEUS features of a papillary RCC which was confirmed by histopathology.

Graph: Fig.2 a) B-mode ultrasound shows a small hyperechoic renal lesion (white arrow). b) Color-Doppler shows no signs of hypervascularization (white arrow). c) The lesion shows hypoenhancement in the arterial phase of CEUS (white arrow). d) The lesion shows a wash-out in comparison to the adjacent renal parenchyma in the late phase of CEUS. In summary, the lesion shows the typical features of a papillary RCC. The diagnosis was confirmed by histopathology after tumor resection.

Over the past years contrast-enhanced ultrasound has been shown to add great value to the characterization of renal lesions. Especially in cases of incidentally detected lesions distinguishing between a benign and malignant etiology is most important [[19]]. To evaluate the malignant potential of complex cystic lesions the Bosniak classification has been modified and well established for CEUS. Depending on the imaging features of the complex cyst the sonographer can recommend if and what additional workup is necessary [[8], [20]]. Studies about the diagnostic accuracy of CEUS regarding solid renal masses draw differing conclusions. While some studies point out the diagnostic value of CEUS distinguishing between benign and malignant solid lesions, other studies showed how challenging it could be to make a distinction between pRCC and lipid-poor angiomyolipoma or oncocytoma [[11]]. While patients profit of the early detection of renal lesions, 10 to 17% of the histopathologically analyzed renal lesions are proofed to be benign [[3]]. Although false positive cases were not analyzed in our study due to histopathology based patient selection, the high sensitivity of CEUS in distinguishing between benign and malignant lesion is confirmed by other studies [[12], [21]]. However, it is remarkable that the false negative case was a cystic lesion characterized by CT and CEUS as IIF lesion. Furthermore, a number of studies analyzed the potential of using CEUS for subtype characterization of RCCs. While Haendl et al. presumed that RCCs show no characteristic enhancement features in CEUS some research groups demonstrated significantly different enhancement patterns comparing ccRCCs and non ccRCCs [[13], [22]]. However, even within the histologically subtypes CEUS has been demonstrated to show remarkable divergences in enhancement features. Our results regarding the atypical contrast enhancement features of 25% of the histopathologically confirmed pRCC are in line with Xue et al. who analyzed in detail the varying enhancement patterns of pRCC [[16]]. Since RCCs represent a very heterogeneous group of tumors, diverse genetic mutations and biologic behaviors have been described. These tumor characteristics affect patient treatment and outcome demonstrating histopathological evaluation to be inevitable [[3], [5]].

To summarize, these results suggest that the varying contrast enhancement patterns of papillary renal cell carcinoma limit the diagnostic potential of CEUS regarding the characterizing of RCC subtypes. CEUS shows noteworthy sensitivity distinguishing benign from malignant renal lesions. However, in some cases atypical enhancement features do not allow a valid classification. If the imaging features indicate a malignant lesion further diagnostic workup using cross-sectional imaging or biopsy is inevitable.

Due to the study design there are limiting factors of this work that need to be addressed. This study was designed as a retrospective, mono-centre analysis with one experienced radiologist evaluating the renal lesions using CEUS imaging. As a result of the histopathology based patient selection the case number is limited to 60 patients without a control group with benign lesions. During period of time different ultrasound systems and probes were used for the imaging studies in our institution.

CEUS is an eligible imaging technique to visualize the contrast enhancement features of pRCC. However, up to 25% of pRCCs show an atypical enhancement pattern making it difficult to distinguish it from other renal lesions. And histopathological evaluation of tumor genetics remains inevitable for treatment decisions.