[1013] Metastatic Chemotherapy Resistant Urothelial Carcinoma (UC): Next Generation Sequencing (NGS) Reveals a Diverse Spectrum of Genomic Alterations (GA) and Targets of Therapy

Jeffrey S Ross, Tipu Nazeer, Christine E Sheehan, Gary Palmer, Geoff Otto, Sean Downing, John Curran, Roman Yelensky, Doron Lipson, Sohail Balasubramanian, Lazaro Garcia, Kristin Mahoney, Philip Stephens, Vincent Miller. Albany Medical Center, Albany, NY; Foundation Medicine Inc., Cambridge, MA

Background: Metastatic UC typically follows an aggressive clinical course and survival is modest despite use of systemic platin-based chemotherapy. The recent introduction of comprehensive NGS genomic profiling to the clinical setting has enabled the identification of novel and unanticipated drug targets of therapy for patients with metastatic chemoresistant UC.
Design: NGS was performed on hybridization-captured, adaptor ligation based libraries using DNA extracted from 4 FFPE sections cut at 10 microns from 15 UC that had relapsed and progressed after primary surgery and conventional chemotherapy. The exons of 182 cancer-related genes were fully sequenced using the Illumina HiSeq 2000 to at an average sequencing depth of 1153X and evaluated for all classes of GA: point mutations (mut), insertions, deletions, copy number alterations (amp or hom del), and gene fusions/rearrangements. Actionable GA were defined as impacting the selection of targeted anti-cancer therapies on the market or in registered clinical trials (CT).
Results: The 15 UC had a mean age of 63.5 with 4 female (27%) and 11 male (73%) patients. All 15 (100%) tumors were high grade and Stage IV at the time of NGS. NGS was performed on the primary UC in 8 (53%) including 6 TURBT (40%) and 2 cystectomies (13%) and on metastatic tumor in 7 UC (47%) including 2 lymph node (13%) and 1 (7%) each liver, lung, brain, psoas and abdominal wall biopsies, NGS discovered a total of 54 GA in the UC series with an average of 4.2 GA per tumor. Ten of fifteen (67%) patients' tumors harbored at least one GA potentially associated with clinical benefit of targeted therapies including: PIK3CA (20%); FGFR1 (20%); CDKN2A/B (13%); CCND3 (13%); NF1 (7%); NF2 (7%); PTEN (7%); ATR (7%); BRCA2 (7%), MYCL1 (7%); ERBB2 (7%); FBXW7 (7%); RAF1 (7%); EGFR (7%); KRAS (7%); MDM4 (7%) and FANCA (7%). Other biologically relevant GA were identified in: TP53 (60%); ARID1A (40%); RB1 (27%); KDM6A (13%); MCL1 (7%) and MYC (7%). The GA detected by NGS generated eligibility of the 15 UC for a total of 75 registered anti-UC CT (average 5 CT per patient).
Conclusions: Deep NGS of metastatic UC uncovers an unexpectedly high frequency of actionable GA that has the potential to influence therapy selection and can direct patients to enter clinical trials to study the benefit of anti-UC targeted therapies. The diversity and spectrum of actionable UC GA identified in this study would be challenging to characterize using other diagnostic methods.
Category: Genitourinary (including renal tumors)

Tuesday, March 5, 2013 1:00 PM

Proffered Papers: Section A, Tuesday Afternoon


Close Window