Introduction: High-risk non-muscle invasive bladder cancer (HR-NMIBC) patients represent a challenging group as they have the highest risk of treatment failure despite transurethral tumor resection (TUR) and adjuvant intravesical Bacillus Calmette-Guérin (BCG) instillations. Treatment failure is associated with an increased risk of progression and unnecessary exposure to toxicity of BCG, while continuing BCG shortage demands selective use of limited resources. The current risk stratification is insufficient at predicting BCG failure in HR-NMIBC patients. Therefore, we aim to investigate whether pathologic depth of tumor invasion (i.e., T1 substaging) is able to improve patient stratification.

Methods: HR-NMIBC patients who received a TUR and at least induction BCG (5 of 6 instillations) were included at four European centers between 2000−2016. Clinicopathologic data were collected and all diagnostic TURs, re-TURs, and subsequent TURs of tumor recurrences were reviewed, which included T1 substaging (micro vs. extensive [pT1e]). Primary endpoint was BCG failure according to EAU guidelines. Progression was defined as muscle-invasive and/or metastatic disease. Logistic regression analysis and Cox regression models were used to identify predictive markers of BCG failure on diagnostic TUR and first recurrence.

Results: N=513 HR-NMIBC patients were included: median age 70 years, 78% men. Median follow-up was 52 months. In total, 316 (62%) patients had pT1 disease, 144 (28%) pTa, and 53 (10%) pTis. BCG failure occurred in 28% (18% progression, 6% metastasis). Median time to progression was 18 months. Strongest predictors of BCG failure on diagnostic TUR were pT1e substaging (HR: 2.1; 95% CI 1.1–4.0; p=0.023), concomitant carcinoma in situ (CIS) (HR: 3.0; 95% CI 1.5–5.7; p=0.003), and lymphovascular invasion (LVI) (HR: 3.3, 95% CI 1.0–11, p=0.05). pT1e substaging combined with either concomitant CIS or LVI increased the risk of BCG failure (HR: 3.6; 95% CI 1.8–6.9; p<0.001). Interestingly, 5/12 patients with micropapillary disease were in this high-risk subgroup. Predictors of progression on diagnostic TUR: concomitant CIS (HR 2.1: 95% CI 1.1–3.9; p=0.018), LVI (HR: 2.6, 95% CI 1.1–6.4, p=0.036) and T1e (HR: 3.6; 95% CI 1.5–9.4; p=0.006). Risk of progression at first recurrence increased in case of a pT1 recurrence (HR: 4.1; 95% CI 2.3–7.2; p<0.001) and adding pT1e substaging increased the risk even more (HR: 5.9, 95% CI 3.2–10.8, p<0.001).

Conclusion: Substaging has the potential to improve the current risk stratification model by identification of HR-NMIBC patients at the highest risk of BCG failure and is specifically able to identify patients at high risk of progression. Substaging can be performed by every pathologist with minimal additional training required. Combined with other pathologic markers, substaging can be used as a stratification tool for early cystectomy or to include patients in clinical trials.

Citation Format: Florus C. de Jong, Robert F. Hoedemaeker, Jolien T.M. Mensink, Vebjørn Kvikstad, Egbert R. Boevé, Deric K.E. van der Schoot, Ellen C. Zwarthoff, Tahlita C.M. Zuiverloon. Pathologic T1 substaging improves the identification of high-risk non-muscle invasive bladder cancer patients at risk of BCG failure [abstract]. In: Proceedings of the AACR Special Conference on Bladder Cancer: Transforming the Field; 2019 May 18-21; Denver, CO. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(15_Suppl):Abstract nr B06.

©2020 American Association for Cancer Research.
2020
American Association for Cancer Research.