849PD - Comparison of tumor mutational burden (TMB) in relevant molecular subsets of metastatic urothelial cancer (MUC)

Date 10 September 2017
Event ESMO 2017 Congress
Session Genitourinary tumours, non-prostate
Topics Urothelial Cancers
Genitourinary Cancers
Translational Research
Presenter Sumanta Pal
Citation Annals of Oncology (2017) 28 (suppl_5): v295-v329. 10.1093/annonc/mdx371
Authors S.K. Pal1, N. Agarwal2, T.K. Choueiri3, P.J. Stephens4, J.S. Ross5, V.A. Miller4, S.M. Ali4, J. Chung6, P. Grivas7
  • 1Medical Oncology, City of Hope, 91010 - Duarte/US
  • 2Medical Oncology, Huntsman Cancer Institute, 84112 - Salt Lake City/US
  • 3Medical Oncology, Dana-Farber Cancer Institute, 02215 - Boston/US
  • 4R & D, Foundation Medicine, 02141 - Cambridge/US
  • 5Pathology, Albany Medical Center, 12208 - Albany/US
  • 6Pathology, Foundation Medicine, 02141 - Cambridge/US
  • 7Medical Oncology, Cleveland Clinic, 44195 - Cleveland/US



Phase I and II studies suggest potential benefit with targeted therapy (TT) (e.g., FGFR3, ERBB2/3 and CDK4/6 inhibitors) in relevant molecular subsets of MUC. Given increasing data supporting PD-1/PD-L1 inhibitors in MUC, rationale for combinations of TT and immunotherapy (IO) is sought. TMB is a putative biomarker of IO response; we investigated differences in TMB in relevant molecular subsets of MUC.


DNA was extracted from 40 microns of FFPE sections from 2024 consecutive patients with MUC. Comprehensive genomic profiling (CGP) was performed on hybridization-captured, adaptor ligation based libraries to a mean coverage depth of 688X for up to 315 cancer-related genes plus 37 introns from 14 genes frequently rearranged in cancer. The CGP assay included base substitutions (SUB), INDELs, copy number alterations (CNA) and fusions/rearrangements. TMB was determined on 1.2 million Mb of sequenced DNA; results are reported for the overall cohort and in subsets segregated separately by presence or absence of FGFR3, ERBB2/3, PIK3CA and CDKN2A/B alteration.


2024 consecutive pts (1457:567 M:F) with MUC were assessed with a median age of 67 years. Median TMB in the overall cohort was 7.2 mutations/Mb. FGFR3, ERBB2, ERBB3, PIK3CA, and CDKN2A/B alteration were identified in 23%, 14%, 4%, 19% and 37% of pts, respectively. TMB was significantly different in pts segregated based on ERBB2 alteration (P =  1.8x10−7), PIK3CA alteration (P =  1.7x10−6) and ERBB3 mutation (P =  0.01) (Table). ERBB2 and FGFR3 genomic alterations (GAs) were significantly mutually exclusive, while FGFR3 GAs significantly co-occurred with PIK3CA and CDKN2A/B. Further differences in CGP amongst these subsets will be presented at the meeting.Table:


TMB, muts/Mb Median (IQR 25%-75%)
All patients (n = 2024)7.2 (3.6, 11.7)
FGFR3 alteration
Yes (n = 461)6.3 (3.8, 10.8)
No (n = 1564)7.2 (3.6, 12.6)
ERBB2 alteration
Yes (n = 277)9.9 (5.4, 18.0)
No (n = 1747)6.3 (3.6, 11.3)
ERBB3 alteration
Yes (n = 72)10.8 (6.3, 20.3)
No (n = 1952)6.3 (3.6, 11.7)
PIK3CA alteration
Yes (n = 383)9.0 (4.8, 17.1)
No (n = 1641)6.3 (3.6, 10.8)
CDKN2A/B alteration
Yes (n = 742)7.2 (3.6, 12.6)
No (n = 1282)6.3 (3.8, 11.3)


Given the proposed correlation between TMB and IO response, these data may inform the utility of combination strategies. Specifically, given higher TMB in pts with ERBB2/ERBB3 or PIK3CA alteration, combination studies exploring IO with TT directed at these targets may be warranted.

Clinical trial identification

Legal entity responsible for the study

Sumanta K. Pal, MD




S.K. Pal: Consulting: Genentech, Aveo, Eisai, Roche, Pfizer, Novartis, Exelixis, Ipsen; Honoraria: Genentech. P.J. Stephens, J.S. Ross, V.A. Miller, S.M. Ali, J. Chung: Employee of and holds equity interest in Foundation Medicine All other authors have declared no conflicts of interest.