

differences, which were associated with patient survival. A
CpG islandmethylator phenotype was observed in a distinct
subgroup of type 2 pRCCs, many with a germline mutation
in the gene encoding fumarate hydratase (
FH
), character-
ized by early disease onset and poor survival
[17].
Oncocytomas are predominantly benign neoplasms
known for their accumulation of respiration-defective
mitochondria. Using exome and transcriptome sequencing,
Joshi et al
[20]identified two main subtypes of renal
oncocytoma, of which type 2 possibly progresses to more
aggressive eosinophilic chRCC, although data are limited.
Chromophobe RCCs can be difficult to distinguish from
oncocytoma and cannot be graded by the Fuhrman grading
system because of its innate nuclear atypia. Several
alternative grading systems have been proposed, some of
them validated
[2,21,22] .Losses of chromosomes Y, 1, 2, 6,
10, 13, 17, and 21 are typical genetic changes
[2] .Recently,
TCGA investigated the landscape of somatic genomic
alterations of 66 chRCCs based on mitochondrial DNA
and whole-genome sequencing. Mitochondrial DNA se-
quencing revealed loss-of-function mutations in nicotin-
amide adenine dinucleotide dehydrogenase subunits,
suggesting changes in mitochondrial function as a driver
of the disease biology. In addition, recurrent structural
breakpoints within the telomerase (
TERT
) promoter region
correlated with highly elevated
TERT
expression
[23,24].
Unclassified RCCs (uRCCs)—the subset of histologies
defying standard classification—present formidable diag-
nostic and management challenges. Chen et al
[25]reported
a molecular analysis of 62 high-grade primary uRCCs,
incorporating targeted cancer gene sequencing, RNA se-
quencing, single-nucleotide polymorphism array, fluores-
cence in situ hybridization, immunohistochemistry, and
cell-based assays. They identified recurrent somatic muta-
tions in 29 genes, including
NF2
(18%),
SETD2
(18%),
BAP1
(13%),
KMT2C
(10%), and
MTOR
(8%). Integrated platform
analysis reveals that 26% uRCCs characterized by
NF2
loss
and dysregulated Hippo-YAP signaling are associated with
worse survival, whereas 21% uRCCs with mutations of
MTOR
,
TSC1
,
TSC2
, or
PTEN
and hyperactive mTORC1 signaling have
a better clinical outcome. FH deficiency (6%), chromatin/DNA
damage regulator mutations (21%), and ALK translocation
(2%) distinguish additional cases. This study reveals distinct
molecular subsets for 76% of the uRCC cohort, with potential
diagnostic and therapeutic implications.
The value of data derived from the TCGA consortium lies
in resource building for future explorations of other kidney
tumors. Large consortia that catalog the metabolic, geno-
mic, and structural alterations and cellular factors of RKCs
may profit from these resources. The next step has already
begun: using the catalog to generate novel hypotheses
about how subtype-specific targeted therapy can be used to
treat these patients.
3.4.
Potential ‘‘druggable’’ pathways
The lower response rates to VEGF-targeted therapy likely
reflect the lack of
VHL
loss in RKCs and a more heteroge-
neous etiology. The
MET
oncogene pathway in pRCC is an
example of what could become a pathway-driven subtype-
specific approach. Initially thought to be responsible in
hereditary forms and to a lesser extent in sporadic pRCC
type 1,
MET
has recently been reported to be altered across
both sporadic pRCC types in a high percentage of 220 cases
analyzed
[26]. In a comprehensive analysis of rare kidney
tumor samples with next-generation sequencing, pRCCs
had frequent amplification, mutation, or overexpression of
MET
[18]. Remarkably, even among the phenotypically
distinct type 2 pRCCs,
MET
alterations have been described
[17]. Antibodies antagonizing the MET receptor–ligand
coupling, such as AMG102, has been tested in a phase I trial
of 40 patients with advanced solid MET-ligand hepatocyte
growth factor (HGF)–dependent tumors in six sequential
dose-escalation cohorts and one dose-expansion cohort
[27,28]. The maximum tolerated dose was not reached,
although dose-limiting hypoxia and dyspnea, and gastroin-
testinal hemorrhage were observed at lower dosages.
Sixteen of 23 (70%) evaluable patients had stable disease
(SD) for 7.9–40 wk but no Response Evaluation Criteria in
Solid Tumors (RECIST) objective responses were observed.
In a phase II study in patients with advanced RCCs including
11.5% papillary and 4.9% chromophobe subtypes
(NCT00422019), AMG102 was tolerated, but it remained
uncertain if the drug was growth inhibitory in this mixed-
subtype population
[29]. Monoclonal antibodies directed
against the MET receptor itself, MetMAb/onartuzumab
(PRO143966), have revealed promising results in preclinical
and phase I studies
[30]. A safety study of MetMAb
(PRO143966) has been completed in patients with locally
advanced or metastatic solid tumors (NCT01068977).
However, none of these monoclonal antibodies is currently
being tested in trials involving patients with type 1 pRCCs.
Foretinib, an oral broad kinase inhibitor targeting MET
among other receptors, was studied in a 74-patient phase II
biomarker study, and it resulted in a median PFS of 9.3 mo
[31]. The study showed that the objective response rate
(ORR) for patients with germline or somatic
MET
mutation
was 50% and 20%, respectively, versus only 8.8% in those
without
[31]. Disrupting the intracellular signaling path-
ways downstream of MET activity has been suggested as
another treatment strategy
[32] .Very recent data suggest
that MET activity predicts response to savolitinib in a phase
II single-arm study in 111 metastatic pRCC patient tumors.
Tumor tissue sequencing determined that 44 patients had
MET-driven pRCCs, which correlated with significantly
better PFS and ORR with savolitinib in patients with MET-
driven pRCCs compared with MET-independent disease
(6.2 mo and 18% vs 1.4 mo and 0%,
p
<
0.0001)
[33].
While the ESPN and ASPEN trials do not show significant
activity with everolimus, it was suggested that proliferation
could be inhibited by dual inhibition of phosphatidylino-
sitol 3-kinase (PI3K) and mTOR based on preclinical models
showing efficacy on cell proliferation, growth, cell survival,
and angiogenesis
[34]. A striking treatment benefit was
recently observed in a phase II study of everolimus plus
bevacizumab in patients with advanced RKCs. Interestingly,
the presence of papillary features was associated with a
benefit and correlated with ORR (43% vs 11%) and median
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 9 7 4 – 9 8 3
978