Institute of Cancer Research, United Kingdom
This was a phase III, multi-center open-label randomized controlled trial in which patients who had undergone nephroureterectomy for UTUC, with pT2-T4, N0 disease or pTany N1-3 M0 disease, and fit for adjuvant chemotherapy, were randomized (1:1) to either four cycles of platinum-based adjuvant chemotherapy or surveillance. Patients with microscopically positive margins on pathology were permitted as long as all gross disease was resected. Chemotherapy consisted of four 21-day cycles of gemcitabine-cisplatin, however, in patients with GFR 30-49 mL/min, carboplatin was substituted for cisplatin and was initiated within 90 days of surgery.
The benefit of peri-operative chemotherapy for urothelial cancer of the bladder is well established for neoadjuvant therapy and less so for adjuvant therapy due to the difficulties in accrual to many of these trials. Although UTUC is related to urothelial cancer of the bladder, the role of chemotherapy and the proper sequencing of surgery with chemotherapy in UTUC is largely unknown and frequently extrapolated from the management of bladder cancer. This trial specifically evaluates the impact of adjuvant chemotherapy given after nephroureterectomy for high grade UTUC.
The primary endpoint of the trial was disease-free survival (DFS) at three years. The secondary endpoints included overall survival, metastasis-free survival, incidence of bladder second primary tumors, incidence of contralateral primary tumors, acute and late toxicity, treatment compliance, and quality of life.
Although renal function may be less optimal after nephroureterectomy, the rationale of adjuvant chemotherapy is that it would minimize overtreatment given more accurate staging with final pathology. However, many patients (estimated to be up to a third) will be ineligible for cisplatin-based adjuvant chemotherapy (GFR<50 mL/min) and subset analysis demonstrated no benefit with carboplatin-based chemotherapy. Nevertheless, this trial firmly establishes the benefit and role of adjuvant gemcitabine-cisplatin in patients with GFR>50 mL/min. This trial does not address neoadjuvant chemotherapy which conceivably would be more tolerable to patients (given better renal function with both kidneys in place) and demonstrates excellent overall survival from retrospective series. However, the risk of over-treatment of patients with neoadjuvant chemotherapy is significant given that chemotherapy is based on limited staging information obtained from ureteroscopic biopsies.
Ultimately, 132 patients were assigned chemotherapy and 129 were assigned surveillance. Adjuvant chemotherapy significantly improved DFS (HR = 0.45, 95% CI 0.3-0.68, p=0.0001) at a median follow-up of 30.3 months with 3-year DFS estimates of 71% (chemotherapy) vs. 46% (surveillance). In addition, the metastasis-free survival rate at two years was 74% vs. 60% for chemotherapy vs. surveillance patients (p=0.002), respectively. However, overall survival was not significantly different between the groups in the early published analysis. On subset analysis, chemotherapy did not demonstrate a benefit in lymph node positive patients or patients with microscopic positive margins. Most importantly, subset analysis did not show a benefit with gemcitabine-carboplatin chemotherapy. Among all chemotherapy-treated patients, 44% had acute grade 3 or greater treatment-related adverse events (TRAEs). This is in comparison to a 4% acute grade 3 or greater TRAEs in the surveillance patients.