Upcoming event

Multicenter phase II clinical trial of gemcitabine and cisplatin as neoadjuvant chemotherapy for patients with high-grade upper tract urothelial carcinoma

  • Jonathan A. Coleman,
  • Wesley Yip,
  • Nathan C. Wong,
  • Daniel D. Sjoberg,
  • Bernard H. Bochner,
  • Guido Dalbagni,
  • S. Machele Donat,
  • Harry W. Herr,
  • Eugene K. Cha,
  • Timothy F. Donahue,
  • Eugene J. Pietzak,
  • A. Ari Hakimi,
  • Kwanghee Kim,
  • Hikmat A. Al-Ahmadie,
  • H. Alberto Vargas,
  • Ricardo G. Alvim,
  • Soleen Ghafoor,
  • Nicole E. Benfante,
  • Anoop M. Meraney,
  • Steven J. Shichman,
  • Jeffrey M. Kamradt,
  • Suresh G. Nair,
  • Angelo A. Baccala Jr,
  • Paul Palyca,
  • Bradley W. Lash,
  • Muhammad A. Rizvi,
  • Scott K. Swanson,
  • Antonio F. Muina,
  • Andrea B. Apolo,
  • Gopa Iyer,
  • Jonathan E. Rosenberg,
  • Min Y. Teo,
  • Dean F. Bajorin

Publication: Journal of Clinical Oncology, January 2023


Neoadjuvant chemotherapy (NAC) has proven survival benefits for patients with invasive urothelial carcinoma of the bladder, yet its role for upper tract urothelial carcinoma (UTUC) remains undefined. We conducted a multicenter, single-arm, phase II trial of NAC with gemcitabine and split-dose cisplatin (GC) for patients with high-risk UTUC before extirpative surgery to evaluate response, survival, and tolerability.


Eligible patients with defined criteria for high-risk localized UTUC received four cycles of split-dose GC before surgical resection and lymph node dissection. The primary study end point was rate of pathologic response (defined as < ypT2N0). Secondary end points included progression-free survival (PFS), overall survival (OS), and safety and tolerability.


Among 57 patients evaluated, 36 (63%) demonstrated pathologic response (95% CI, 49 to 76). A complete pathologic response (ypT0N0) was noted in 11 patients (19%). Fifty-one patients (89%) tolerated at least three complete cycles of split-dose GC, 27 patients (47%) tolerated four complete cycles, and all patients proceeded to surgery. With a median follow up of 3.1 years, 2- and 5-year PFS rates were 89% (95% CI, 81 to 98) and 72% (95% CI, 59 to 87), while 2- and 5-year OS rates were 93% (95% CI, 86 to 100) and 79% (95% CI, 67 to 94), respectively. Pathologic complete and partial responses were associated with improved PFS and OS compared with nonresponders (≥ ypT2N any; 2-year PFS 100% and 95% v 76%, P < .001; 2-year OS 100% and 100% v 80%, P < .001).


NAC with split-dose GC for high-risk UTUC is a well-tolerated, effective therapy demonstrating evidence of pathologic response that is associated with favorable survival outcomes. Given that these survival outcomes are superior to historical series, these data support the use of NAC as a standard of care for high-risk UTUC, and split-dose GC is a viable option for NAC.

Dr. Laura Mertens

Radical nephroureterectomy (RNU) is the standard treatment for high-risk upper tract urothelial cancer (UTUC). The POUT study provided level-1 evidence supporting adjuvant chemotherapy to improve disease-free survival in patients with pT2-pT4 or lymph-node-positive (pTanyN1-3M0) UTUC (Lancet 2020). However, only a small subset of patients will be eligible for full dose cisplatin-based adjuvant chemotherapy due to the postoperative renal function after surgery.

Therefore, more patients may be eligible for cisplatin-based therapy in the neoadjuvant setting. Several retrospective studies have shown promising pathologic response rates after neoadjuvant chemotherapy, and one prospective phase II trial (Margulis, J Urol. 2020), demonstrated a pathologic complete response of 14% and <ypT1 in more than 60% after MVAC followed by RNU.

Coleman et al., now performed a prospective phase II trial evaluating neoadjuvant gemcitabine and split-dose cisplatin for patients with high-risk UTUC. The study was performed at four US hospitals between 2010 and 2019. A total of 57 (cisplatin-eligible) patients were included. Primary inclusion criteria were high-risk patients (defined as histologically confirmed diagnosis of high-grade UTUC (93% of patients) and/or radiographically visible invasive disease (cT2-T4aN0M0 with positive selective urinary cytology (7% of patients). The primary outcome was: pathologic response; secondary outcomes were: a) treatment safety; b) cancer progression; c) survival.

In this study, patients received split-dose gemcitabine-cisplatin. The rationale for this split-dose scheme was provided by previous studies showing improved tolerability and toxicity profile (which was considered of particular importance given the renal involvement and/or hydronephrosis in UTUC). The vast majority (n=51, 89%) were able to receive at least three cycles. Most patients (95%) underwent RNU including an ipsilateral lymph node dissection, whereas 5% underwent only distal ureterectomy (because of only distal ureteral involvement).

Regarding the primary endpoint, 63% achieved a pathologic response (<ypT2N0) and 19% had pathologic complete response (ypT0N0) – meeting the pre-specified criteria.

Regarding the secondary endpoints: Treatment-related toxicities were observed in 93% of patients, 58% required toxicity-related dose modifications, but early discontinuation occurred in only 14% and all patients were able to undergo surgery. Furthermore, at a median follow-up of 3.1 years among survivors, there were 19 progression events and 11 patients who died. Two-year progression-free survival was 89% and two-year overall survival was 93%.

Because this was not a randomised study, without a control arm, a direct comparison with an upfront RNU group could not be made in terms of survival. Instead, the authors compared the survival rates with historical series and refer to previous analyses indicating that pathologic response may be a surrogate for survival. Another limitation is the long accrual period. This is explained by the rarity of the disease and the higher proportion of cisplatin-ineligibility inherent to UTUC. Furthermore, disease prognosis of UTUC is predominantly driven by advanced stage and high-grade disease at presentation. In neoadjuvant studies for UTUC, this is a challenging issue because accurate preoperative staging is difficult. There are no standardised imaging criteria and biopsy from the upper tract is unable to reliably determine depth of invasion. This must be considered in the interpretation of the pathologic response rate.

Altogether, this prospective study by Coleman et al., adds to the scarce current literature on neoadjuvant chemotherapy in UTUC. They conclude that split-dose-based neoadjuvant chemotherapy is safe and effective in UTUC prior to surgery and that this regimen may ultimately contribute to the feasibility of neoadjuvant chemotherapy for a larger group of patients with this disease.