Novel Therapies for Advanced Urologic Cancers

Axel Heidenreich


Curr Opin Urol. 2020;30(4):594-601. 

In This Article

Urothelial Carcinoma of the Bladder

Neoadjuvant systemic chemotherapy (NAC) with gemcitabine and cisplatin followed by radical cystectomy with an extended pelvic lymphadenectomy and urinary diversion represents the guideline-recommended standard of care for patients with muscle-invasive or recurrent, high-risk nonmuscle-invasive urothelial carcinoma of the urinary bladder.[5] However, up to 50% of patients are considered to be cisplatin-ineligible due to preexisting impairment of renal function, significant comorbidities or refusal to receiving any systemic chemotherapy.[6]

Despite NAC and its oncological benefit, up to 50% of patients exhibit residual vital disease at time of radical cystectomy, which is associated with worse 5-year relapse-free, cancer-specific and OS rates of 50 versus 90%, 40 versus 82% and 33 versus 82%, respectively, as compared to patients with pT0 disease at the time of radical cystectomy.[7]

The development and introduction of immune checkpoint inhibitors has revolutionized the treatment of urothelial bladder cancer. Currently, there are five inhibitors of the PD-1/PD-L1 pathway, which have been approved for the first or second-line treatment of advanced bladder cancer: azetolizumab, avelumab, durvalumab, nivolumab and pembroluzimab.[8]

Currently, two prospective phase-II trials of neoadjuvant PD-1/PD-L1 inhibitors in muscle-invasive bladder cancer have been reported: the ABACUS trial (atezolizumab) and the PURE-01 trial (pembroluzimab).[9–11] Both trials included cisplatin-unfit patients or patients who refused standard chemotherapy with locally advanced or lymph node positive bladder cancer with the primary study endpoint of complete pathological response (pCR) following two cycles of atezolizumab or three cycles of pembrolizumab.

In PURE-01, 50 patients were enrolled and 46 (92%) patients would have been eligible for cisplatin therapy.[9] A total of 21 (42%) patients achieved a pT0-stage and an additional six patients had been downstaged to nonmuscle-invasive disease. PD-L1 expression with a CPS at least 10% was associated with a significantly higher response rate (54.3 versus 13.3%, P = 0.011) as was a tumour mutation burden at least 15 mut/MB. In addition, DNA damage repair and RB1 gene alterations were associated with pT0 disease as was a mutation in PMBR1. In ABACUS, 68 cisplatin ineligible patients with muscle-invasive bladder cancer had been recruited and the authors observed overall pT0 rate of 29% with a statistically significant difference of 40 versus 16% in patients with high or low PD-L1 expression.[10] As compared to NAC, the safety profile of all drugs used for immune-oncological therapy (IOT) is quite low with less than 10% grade 3/4 toxicities as compared to 30–40% for NAC. In the updated version, it was demonstrated that the pT0 status could be achieved in 37% of patients with muscle-invasive bladder cancer if variant histologies. These data demonstrate the oncological benefit of neoadjuvant IOT.

Both studies demonstrated a high frequency of complete pathological responses, which was even enriched in high PD-L1 expressing tumours. However, both studies also have significant limitations: lack of mature follow-up data to ensure that pT0 stage following IOT is associated with an improved survival, small numbers of lymph node positive patients, exclusion of patients with locally advanced disease such as pT4 disease in PURE-01, lack of reliable predictive biomarkers to identify patients with a high likelihood of response and lack to stratify treatment response with the current TGCA-molecular subclassification of urothelial bladder cancer.

Quite recently, a significant discordance of high PD-L1 expression was demonstrated not only in primary and metastatic lesions but also between tumour cell and immune cell compartments. High PD-L1 expression was observed in primary and metastatic lesions in 6% and 7.7% of tumour cells and in 14.5 and 11.5% of immune cells. PD-L1 expression in primary lesions was not associated with PD-L1 expression in paired metastatic lesions. In a similar approach, high PD-L1 expression was identified in 23 and 55% of tumour cells and tumour infiltrating immune cells, respectively.

In this context, the surgical safety of radical cystectomy following neoadjuvant treatment with pembroluzimab was evaluated in the context of the PURE-01 trial.[12] A total of 68 patients underwent radical cystectomy of whom 77 and 23% were performed as robotic-assisted and open surgery, respectively. Grade at least 2 complications and Clavien-Dindo at least 3a were observed in 69 and 32% of the patients, respectively. The most common complications were fever in 52% and ileus in 32% of the cases. The readmission rate during the first 90 postoperative days was 32%. On the basis of these data, radical cystectomy following neoadjuvant immunooncological therapy is not associated with an increased surgical complication rate.