Biology Reference
In-Depth Information
disease. There is evidence in the literature that loss of
platinum sensitivity in ovarian cancer is associated
with regain of BRCA function;
145,146
however these
findings were not borne out in the study described
above. Clearly the findings in these small non-randomized
studies require confirmation in the phase III trial, but
they do represent an exciting indication that PARP
inhibitors will be of benefit in these patients with
low observed toxicity. Phase II studies in this indica-
tion are also ongoing with AG014699 and ABT-888.
It is clear that mutations in BRCA1/2 are just the “tip
of the iceberg” when it comes to HR defects in cancer.
Epigenetic silencing of BRCA1 through promoter meth-
ylation and upregulation of inhibitors of BRCA2 have
been shown to contribute to a “BRCAness” phenotype
in breast and ovarian cancer.
147
However, this term is
too narrow, implying that HR depends principally on
BRCA1 and 2 and is largely restricted to breast and
ovarian cancer, when in fact it is a complex pathway,
involving a panoply of proteins that include damage sig-
nalling and checkpoint kinases (e.g. ATR and Chk1), the
FANC and Rad51 homologs and many other compo-
nents, some of which remain to be identified, and is
defective in a wide variety of cancers. For example, the
finding that PARP inhibitors were synthetically lethal
in cells with PTEN mutations was unexpected until it
was discovered that PTEN regulates Rad51 function.
Mutations/defects in many of these proteins are associ-
ated with cancers in various tissues so the potential for
selective PARPi therapy is vast.
a HR deficiency (HRD) phenotype, it was imminent
that the gene expression profile of HRD would be
explored to identify this group of cancers. Recently,
Konstantinopoulos
et al.
149
used a publicly available
micro-array dataset including 61 patients with epithelial
ovarian cancers (sporadic/ BRCA germ line mutant) to
develop a “BRCAness” profile.
149
The profile was vali-
dated in 70 patients enriched for sporadic disease to
assess its association with clinical outcome. The
BRCAness profile correlated with responsiveness to
platinum and PARP inhibitors and was predictive of
a subset of sporadic EOCs with improved outcome.
Another alternative is to use immunofluorescence-
based assays to studyRAD51 focus formation, aHR repair
protein after
ex vivo
DNA damage . Willers
et al.
described
a prospective pilot study evaluating immunofluorescence
based BRCA1, FANCD2, and RAD51 foci formation in
seven fresh breast tumor biopsies, which were irradiated
ex vivo
and could be divided into RAD51 foci positive and
defective tumors.
150
Three out of 4 foci defective tumors
were triple negative breast cancers, a phenotype that has
been associated with BRCA1 deficiency. In a larger
prospective study
151
involving 60 patients with primary
breast cancer receiving neo-adjuvant epirubicin and
cyclophosphamide, quantification of four HR related
proteins
BRCA1,
g
-H2AX, conjugated ubiquitin, and
RAD51 foci in paired tumor biopsies, before and after first
cycle of chemotherapy, was undertaken to study the role
of HR functional proteins as a predictive biomarker for
chemosensitivity. The presence of BRCA1-positive base-
line foci and presence of either baseline or chemotherapy
induced RAD51 foci inversely correlatedwith response to
chemotherapy. Conversely, proficient DNA repair indices
based on all four markers correlatedwith drug resistance,
supporting the hypothesis that DNA damage response
competency predicts for reduced tumor responses to
DNA damaging agents. Mukhopadhyay
et al.
investi-
gatedRad51 focus formation as amarker ofHR inprimary
ovarian cancer cell cultures and found that an increase in
foci could accurately predict sensitivity to the PARP inhib-
itorAG014699.
152
It is important tonote that these samples
were not from BRCA1/2 mutation carriers and they
reveal a sizeable population of HR-defective tumors
(50
e
FUNCTIONAL ASSAY/PREDICTIVE
BIOMARKER FOR HOMOLOGOUS
REC
OMBINATION (HR) DEFE
CTS
As the above data suggest, many cancer patients
outside of the BRCA mutation carrier population could
benefit from PARP inhibitor therapy. The problem is
how to identify them given the heterogeneous mecha-
nisms by which sporadic cancers can develop defective
HR. The development of assays to determine HR defects
is the next big challenge for researchers. High
throughput sequencing of sporadic tumors would be
difficult as HR involves multiple genes and such
processes would be costly; therefore, the need to
develop a functional assay for HR deficiency or HRD.
One method is to use gene expression arrays to deter-
mine a
BRCA
-like profile in tumors. In a study by Jazaeri
et al.
,
148
gene expression profiles of
BRCA1/ 2
mutated
cancers were compared with sporadic epithelial ovarian
cancers from formalin fixed tissues. Within the sporadic
group, cancers with BRCA-1 like and BRCA-2 like
(named BRCA linked) gene expression profile were
identified. With further development of the concept of
60% of epithelial ovarian cancers) that could poten-
tially benefit from PARPi therapy. Similarly, in a recently
published study by Graeser
et al.
,
153
core biopsies were
obtained 24 hours after first cycle of neo-adjuvant anthra-
cycline based chemotherapy in 68 patients with sporadic
breast cancers. RAD51 foci formation by immunofluores-
cencewas assessed as amarker forHRdefect and geminin
as a marker for proliferating cells. Low RAD51 score
correlated with high histological grade, triple negative
breast cancers and was a strong predictor for pathologic
complete response to chemotherapy.
e
In a study by
Gaymes
et al.
154
it was shown that PARP-inhibitor
