Upcoming event

Molecular subtyping of clinically localized urothelial carcinoma reveals lower rates of pathological upstaging at radical cystectomy among luminal tumors

  • Yair Lotan 1,
  • Stephen A. Boorjian 2,
  • Jingbin Zhang 3,
  • Trinity J. Bivalacqua 4,
  • Sima P. Porten 5,
  • Thomas Wheeler 6,
  • Seth P. Lerner 6,
  • Ryan Hutchinson 1,
  • Franto Francis 1,
  • Elai Davicioni 3,
  • Robert S. Svatek 7,
  • Chun-Liang Chen 7,
  • Peter C. Black 8,
  • Ewan A. Gibb 3
1 Department of Urology, University of Texas Southwestern Medical Center, Dallas, TX, USA 2 Mayo Clinic, Rochester, MN, USA 3 GenomeDx Inc, Vancouver, BC, Canada 4 John Hopkins Medical Institute, Baltimore, MD, USA 5 University of California San Francisco, San Francisco, CA, USA 6 Baylor College of Medicine, Houston, TX, USA 7 University of Texas Health San Antonio, San Antonio, TX, USA 8 Vancouver Prostate Centre, University of British Columbia, Vancouver, BC, USA

Publication: European Urology, April 2019

DOI: doi.org/10.1016/j.eururo.2019.04.036


Upstaging of clinical T1-T2 urothelial carcinoma (UC) to non-organ-confined (NOC) pathological stage ≥T3 or N+ at radical cystectomy (RC) is common. Tools for stratifying patients who may have NOC disease are limited.

To determine an association of a genomic subtyping classifier (GSC) with pathological upstaging in multi-institutional cohort of patients with cT1-T2 UC treated with RC.

Design, setting, and participants
Precystectomy transurethral specimens from 206 patients with high-grade, cT1-T2, N0M0 UC, who underwent RC without neoadjuvant chemotherapy, underwent GSC testing.

Outcome measurements and statistical analysis
Uni- and multivariable logistic regression analyses evaluated GSC for upstaging, defined as pT3/T4 and/or pTanyN1–3 disease at RC.

Results and limitations
Pathological upstaging occurred in 23% of cT1 and 57% of cT2 cases. Lower rates of upstaging to NOC was seen for luminal versus nonluminal tumors (34% vs 51%, p = 0.02). The differences in upstaging were confined to T stage, with no difference in node positivity for luminal versus nonluminal patients (cT1: 13% for both [p > 0.9], cT2: 15% and 23% [p = 0.6], respectively). Fewer patients with luminal tumors were upstaged to ≥pT3Nany compared with nonluminal tumors (Mantel-Haenszel p = 0.002; cT1: 13% vs 30%, cT2: 34% vs 58%). On multivariable logistic regression analysis, nonluminal patients were more likely to be upstaged to ≥pT3 at RC (p < 0.001). Limitations include retrospective design and sample size.

Molecular subtyping of clinically localized UC demonstrated that luminal tumors have lower rates of upstaging to non-organ-confined disease compared with nonluminal tumors. If validated, these data can help inform which patients may need multimodal therapy.

Patient summary
Determining whether bladder cancer has spread beyond the bladder is challenging at diagnosis. In this paper, genomics helped identify patients who were more likely to have aggressive disease that has spread outside the bladder. These patients may benefit from chemotherapy prior to surgery.

Expert's summary

The standard treatment for T2-T4aN0M0 bladder cancer (BCa) is represented by cisplatin-based neoadjuvant chemotherapy (NAC) and radical cystectomy (RC) with pelvic lymph node dissection (PLND). Based on the results of randomized clinical trials, NAC provides an overall survival benefit between 6% and 9% compared to surgery alone. However, this advantage is more evident in patients with clinical stage ≥T3 and, given the non-negligible NAC-related toxicity rates, the real benefit of NAC in patients with clinical T2 without features of tumor aggressiveness remains questionable. However, clinical staging, mainly based on transurethral resection of the bladder and CT/MRI imaging is mostly inaccurate and upstaging at RC is a common event occurring in about 40% of patients undergoing surgery. Therefore, decision-making based on clinical stage (i.e. not administering NAC to cT2 patients) may lead to possible under-treatment with dramatic impact on oncological outcomes. In this study, Lotan and colleagues evaluated the role of molecular markers in providing new insights for treatment stratification. Luminal tumors were less-likely upstaged at RC and had a significant lower rate of upstaging to non-organ confined disease compared to non-luminal ones. Moreover, as already shown in the literature, luminal tumors had better cancer-specific survival compared to non-luminal patients. These findings, if prospectively validated, could play a major role, together with standard clinico-pathologic factors, in differentiating patients who are more likely to benefit from NAC (i.e. non-luminal subtypes) from those in whom systemic treatment and related toxicity could be safely spared without affecting survival.

Expert's comment

The aim of this retrospective multicenter study was to evaluate the impact of bladder cancer molecular subtyping on pathologic upstaging to non-organ confined disease at radical cystectomy in patients with localized disease. Overall, 206 patients with clinical T1 (42%) or T2 (58%) who did not receive any type of neoadjuvant therapy and who did not present features of tumor aggressiveness (i.e. hydronephrosis, variant histology, cancer in diverticulum) were included in the study. A whole-transcriptome assay was used to generate a genomic subtyping classifier score for each tumor, thus stratifying tumors in luminal, basal, infiltrated luminal and claudin low subtypes. Luminal tumors showed a lower rate of upstaging to non-organ confined disease compared to non-luminal ones (34% vs 51%, p=0.02). These data, if externally validated, may pave the way towards a personalized treatment approach, thereby identifying those patients who may need a multimodal therapy.