A simple CT enhancement measure is able to differentiate a benign form of kidney cancer from an occasionally lethal subtype of renal cell carcinoma (RCC) with virtually 100% accuracy, sparing patients the need for either partial or full nephrectomy to ascertain which type of tumor is present, new research shows.The study was published in the August issue of Clinical Cancer Research.

Study Details

The initial part of the study was conducted in a restrospective discovery cohort composed of 87 patients, including 53 patients with renal oncocytoma and 34 patients with chromophobe RCC (ChRCC).

The investigators explain that at least 70% of all renal oncocytomas and ChRCC tumors express the CD117/KIT biomarker. Because the CD117/KIT biomarker is not present in other RCC subtypes, CD117/KIT expression significantly narrows the differential diagnosis to either of these two tumor types.

An analysis was conducted of the retrospective discovery cohort to identify both clinical and radiographic features that distinguish renal oncocytoma from ChRCC tumors.

The researchers then prospectively followed 37 nephrectomy patients with either renal oncocytoma or ChRCC over a 4-year period to validate clinical variables of each of the two tumor types. They also analyzed the tumor:cortex peak early-phase enhancement ratio (PEER) using multiphase CT.

Clinical features that helped predict CD117-positive renal oncocytomas included multifocality; patient's age less than 50 years enabled correct classification of most CD117-positive ChRCC tumors.

However, "the variable most reliably associated with renal oncocytoma versus ChRCC diagnosis was the CT peak signal intensity within a tumor, particularly when expressed as a tumor:cortex ratio," the investigators report.

The radiographic measurement that best discriminated between the two types of tumors was the PEER of tumor to cortex, they elaborate. The tumor-to-cortex PEER measurement reflects the level of brightness of the tumor on the CT scan compared to the brightness of the cortex of the kidney.

To assess this, a CT scan is first performed without contrast dye. Another scan is then performed with contrast as the dye is entering the kidney, reflecting "early-phase enhancement." When a CD117-positive tumor appears brighter than the rest of the kidney, it is more likely to be benign, the investigators explain.

Using this radiographic measurement, Kauffman and colleagues determined that all renal oncocytoma tumors were relatively hyperenhancing with a tumor:cortex PEER in excess of 0.50.

"In contrast, all but four ChRCC tumors were relatively hypoenhancing with a tumor:cortex PEER <0.50," they state. Interestingly, the four ChRCC tumors for which the tumor:cortex PEER exceeded 0.50 were negative for the CD117 biomarker.

Prospective Validation

During a 4-year prospective follow-up, the study authors analyzed 40 additional tumors, half of which were renal oncocytomas and the other half, ChRCC tumors.

For 28 of these tumors, preoperative multiphase CT scans were available in which the PEER could be measured. Twenty-two of these tumors were CD117-positive, and the remaining six were CD117-negative.

"All 22 CD117(+) tumors were correctly classified using a PEER threshold value of 0.50 by each of the three independent reviewers," the researchers report. In contrast, the same three reviewers misclassified three of the six CD117-negative tumors using the same approach.

Reliable Scoring

Kauffman noted that among the most promising and exciting aspects of the current research is that fact that the PEER does not require any special type of contrast CT, provided an early contrast phase is included, which is routine.

"In fact, we evaluated CT scans from many different imaging facilities, all using somewhat different protocols for the early contrast phase, and it didn't matter — the scoring approach was reliable in all cases," Kauffman observed.

The fact that no specific CT protocol is needed suggests that the PEER approach could be easily adapted at other centers, he added. Importantly as well, the image assessment itself is easy to do and adds only an extra few minutes to standard protocols.

"We have gone through validation testing at our institute, including independent testing among several radiologists, which has reliably confirmed 100% accuracy," Kauffman concluded.

"So I have confidence and comfort using this approach routinely in my clinical practice, and if reproducibility between different centers can be confirmed, [PEER] will be ready for prime time," he said.