The PURE-01 study tested the activity of preoperative pembrolizumab (pembro) in patients (pts) with muscle-invasive bladder UC, resulting in 38.5% pathologic complete responses (pT0N0). Within this study, we explored the activity of preoperative pembro in a cohort of pts with UTUC (PURE-02). At present, no data have been reported with the use of the immune checkpoint inhibitors (ICI) in localized UTUC.
The PURE-02 cohort was aimed to assess the feasibility of neoadjuvant pembro before RNU. This cohort was represented by a pre-determined sample size of 10 pts who received 200 mg pembro, intravenously (IV), every 3 weeks, preceding RNU (planned within 14 days of the last pembro dose). The study included pts with clinical stage N0M0 UTUC, with high-risk features according to modified EAU guidelines, defined by the presence of either: high-grade disease (urinary cytology and/or biopsy), multifocal disease, T≥2cm tumor mass, hydronephrosis. Pts with previous/concomitant bladder UC or variant histology were excluded. The endpoints were to prospectively test the reliability of the EAU high-risk UTUC criteria within a prospective study, to assess the clinical activity (number of ypT0N0 responses) and the surgical and medical safety (CTCAE version 4.03). Hybrid capture-based genomic profiling tests were done on tissue and blood samples, and multiparametric magnetic resonance imaging (mpMRI) tests of the renal pelvis and ureter were also investigated.
Between 05/2018 and 10/2020, PURE-02 enrolled 10 pts evaluable for the study endpoints: 33% were female, 2 had a renal pelvis, 6 an ureteral tumor and 2 both of them. Tumor biopsy was not performed in 7 pts (70%) due to the presence of positive cytology and tumor mass (N=4) or multifocal disease (N=3). 9 pts completed the pembro administrations, and 1 pt died due to the development of severe myasthenia gravis, myocarditis, myositis, and hepatitis after the 1st pembro course. The remaining pts did not show any grade 2-4 AE. Two pts (20%) achieved a radiological partial response, 2 received subsequent therapy due to nonresponse (stable disease). In total, 7 pts underwent RNU: one (14.3%) achieved a ypT1N0 response, the remaning pts were nonresponders. 3/6 baseline ctDNA assays were test failure due to the small amount of DNA. mpMRI findings did not help discriminating a response besides dimensional changes.
The preliminary, feasibility results of the PURE-02 study provided us with multiple limitations on a strategy to administer single-agent pembro preoperatively in high-risk UTUC. Safety may be a major concern in these pts, and activity of single-agent therapy is minimal. Assessing ctDNA for therapeutic targets and mpMRI to evaluate tumor response are critical goals to achieve in early-disease UTUC.