Nearly all malignant and many benign solid renal masses typically do not contain macroscopic fat. The malignancies include all RCC types, including clear cell (most common), papillary, chromophobe, collecting duct, and medullary cancers, and the more recently described XP11.2 translocation RCC. Upper-tract urothelial cancers and primitive neuroectodermal neoplasms also can present as solid renal masses.

Benign solid renal masses that contain little or no fat include the aforementioned mfAMLs, oncocytomas, papillary adenomas, and renomedullary interstitial cell tumors, and more rarely, metanephric neoplasms, hemangiomas and lymphangiomas, leiomyomas, juxtaglomerular tumors, and mixed epithelial and stromal tumors.

Many papers have attempted to distinguish among various non- or minimal-fat- containing solid renal masses on CT and MRI. A few characteristic imaging features for the different neoplasms are summarized here.

On contrast-enhanced CT or MRI, the most common two papillary subtypes, the better-prognosis type 1 tumors and the more aggressive type II tumors both tend to demonstrate better well-defined margins, greater homogeneity, and lesser and greater delayed enhancement than other RCC cell types (Fig. 3) [24, 25]. Papillary cancers do not demonstrate peak enhancement until NP and/or EP. On MRI, many papillary cancers demonstrate characteristic T2-signal hypointensity [26]. In comparison, only a small minority of clear cell RCC are T2 hypointense [26].

In comparison with other types of RCC, medullary, collecting duct, and XP11.2 translocation cancers generally arise in the medulla. Whereas collecting-duct cancers frequently occur in older adults, renal medullary and XP11.2 cancers are usually encountered in young patients, with XP11.2 RCC accounting for about one third of pediatric RCC [27]. Medullary RCC typically develops in patients with sickle cell trait [28].

One cannot rely on differences in growth rates to distinguish among different RCC cell types. In one series, for example, neither the initial size of a detected RCC nor its cell type predicted the likelihood that the cancer would be more or less likely to grow quickly.

As with complex cystic renal masses, benign and malignant solid renal masses can remain stable in size or enlarge, with growth rates of both types of lesions usually being slow and similar [30]. It is suggested that growth of a renal mass can only be considered suggestive of malignancy when such growth is unusually rapid (>5 mm within 12 months) [31].

Due to overlap of many of the above-described imaging findings, we believe that, at the present time, imaging differentiation of malignant from benign non-fatcontaining solid renal masses (or of various types of malignant solid renal masses) is not possible in any given case. This once was not felt to be an important problem, as it was believed that the vast majority of solid renal masses were RCC. It has become apparent that a sizeable minority of small solid renal masses are benign: about 20% of solid renal masses <4 cm in diameter are benign [32], with the frequency of benign tumors being even greater for smaller lesions [32].

Given the substantial overlap in imaging features of benign and malignant lesions, an increasing number of institutions perform biopsies of many, or even all, detected solid renal masses. Although there was some hesitation about performing biopsies of masses that could represent RCC due to concerns about diagnostic errors, tract seeding, or bleeding, it is now widely accepted that percutaneous biopsy of a solid renal mass can be performed accurately and safely in the vast majority of patients [33]. In particular, percutaneous biopsy differentiation of oncocytomas is now considered reliable in most patients, with only a few biopsies of oncocytic tumors being indeterminate [34, 35]. Even if biopsy is generally accurate in determining whether a renal mass is benign or malignant and in assessing RCC cell type, debate about whether a percutaneous biopsy might occasionally underestimate the grade of an RCC by failing to sample the portion that contains the most aggressive cells. It is not yet clear whether undergrading will be a clinically significant problem.