Abstract
Retinoblastoma is a primary intraocular cancer that develops in the eyes of children. Tumorigenesis is initiated
by a mutation in the RB1 gene, which was the first tumor suppressor gene described. Investigating the tumor
suppressor pathway regulated by RB1 has provided unprecedented insights into the genetic mechanisms of
tumorigenesis, not only for retinoblastoma, but virtually all human cancers. The evaluation of specific
mutations, oncogenes and suppressors, tumor markers and gene expression profiles have revolutionized the
care of breast, lung, prostate and other cancer patients. However, despite the fact that retinoblastoma-related
research launched the field of cancer genetics, leveraging this knowledge to personalize the care for
retinoblastoma patients has been elusive in large part because we cannot safely biopsy this tumor due to
concerns of extraocular spread. Thus, we currently cannot correlate the genetic and genomic changes at the
level of the tumor with clinical outcomes. Identifying these genomic changes, and providing clinical correlation
is critically needed for retinoblastoma patients. The objective of this proposal is to develop a surrogate tumor
biopsy for retinoblastoma using the aqueous humor, which is the clear fluid in a separate part of the eye from
where the tumor forms. Starting in 2012, extractions of aqueous humor have been routinely and safely done as
a standard part of the procedure to inject chemotherapy into the vitreous cavity of the eye; no cases of
extraocular spread have been reported. Our preliminary data from evaluation of the aqueous revealed that
tumor-derived cell-free DNA is present and whole genome sequencing demonstrates a profile of chromosomal
gains and losses that corroborate changes in the tumor. The specific aims of this study are designed to test the
following hypothesis: the aqueous can serve as a surrogate tumor biopsy – more precisely, the aqueous
harbors tumor-derived genetic material which can be assayed and correlated with clinical outcomes. Aim 1 is
to define highly recurrent chromosomal changes in the aqueous, such as 2p gain/MYCN amplification, that are
known to occur in retinoblastoma tumors. Aim 2 is to identify the specific RB1 mutation(s) that underlie
tumorigenesis from the aqueous humor. Finally, Aim 3 will evaluate RNA expression profiles to define
subgroups in the aqueous, and correlate these with clinical outcomes such as tumor recurrence. Currently, the
only way any of these Aims are possible is from evaluation of tumor tissue, which is only available if an eye is
removed (e.g. enucleated). However, by defining these genetic changes in the aqueous we will, for the first
time, be able to characterize tumors in situ during therapy. This will allow for the development of prognostic
markers for therapeutic response and, in the future, may guide therapy. Use of the aqueous humor as a
surrogate tumor biopsy holds tremendous potential, from directly impacting the way we manage retinoblastoma
patients, to ultimately allowing for the development of personalized medicine for this blinding childhood cancer.
!
!