Introduction

Upper tract urothelial carcinoma (UTUC) is a relatively rare disease, accounting for only 5–10% of all urothelial tumors [1]. Although radical nephroureterectomy with bladder cuff excision is an established standard treatment for UTUC, local recurrence and distant metastasis rates remain high. The reported 5-year recurrence-free survival (RFS) and cancer-specific survival (CSS) rates are 69–72% and 72–74%, respectively; these rates are even worse for locally advanced cases [2–4].

In order to improve the oncological outcome for patients with UTUC, systemic chemotherapy is often administered after nephroureterectomy in select, high-risk patients [5–14]. However, due to the rarity of this disease, no randomized controlled trial (RCT) has addressed whether it significantly improves patient outcome. The efficacy of post-surgical adjuvant chemotherapy (AC) remains controversial [1].

In this retrospective study, we evaluated the efficacy of AC following radical nephroureterectomy in patients with locally advanced (pT3-4pN0/xM0) UTUC.

Materials and methods

Patient population

This study was approved by the institutional review board of each institute. We reviewed the medical records of all patients who underwent nephroureterectomy for UTUC with curative intent at our four institutions between 1996 and 2013. Patients with locally advanced disease without nodal involvement or distant metastasis (pT3-4pN0/xM0) were included in this study. The execution and the extent of lymph node dissection in conjunction with nephroureterectomy were based on presurgical imaging and intraoperative findings, and were determined at the physicians’ discretion. Patients who received neoadjuvant chemotherapy (n = 1), underwent adjuvant radiotherapy (n = 1), had pathology of non-urothelial carcinoma (n = 2), or who developed synchronous or metachronous invasive bladder cancer (n = 11) were excluded from this study. Patients with early recurrence or mortality (within 12 weeks following nephroureterectomy, n = 3) were also excluded. Thus, 109 eligible patients were identified (Fig. 1).

Fig. 1
figure 1

A diagram representing study cohort. UTUC upper tract urothelial carcinoma, UCB urothelial carcinoma of the bladder, UC urothelial carcinoma, RT radiation therapy, NAC neoadjuvant chemotherapy

Full size image

All nephroureterectomy specimens were subjected to routine histological examination. Histological diagnosis was determined according to the World Health Organization (WHO) classification system [15]. The primary tumors and lymph nodes were re-staged based on the 2009 Union for International Caner Control TNM system [16]. Glomerular filtration rate was estimated according to the revised formula for Japanese subjects [17].

While two or three cycles of AC are generally recommended for patients with locally advanced UTUC, the final decision for the patients to undergo AC is taken after a discussion between the patient and the physician. Some patients underwent four cycles of chemotherapy in accordance with the patients’ wishes or at the physician’s discretion. A few patients ceased treatment after completion of only one cycle. Until 2008, the AC protocol had predominantly consisted of a combination of methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC); thereafter, it became a combination of gemcitabine and cisplatin (GC). A combination of methotrexate, epirubicin, and cisplatin (MEC), or that of docetaxel, ifosfamide, and cisplatin (DIP) [18] was administered to a few patients. The chemotherapy dosage was reduced by 30% in cases of impaired renal function and/or severe adverse events in the previous cycles. Chemotherapy administered after development of metastasis was not included in the definition of AC in this study.

In general, patients were followed-up every 3–4 months for 2 years after nephroureterectomy, every 6 months for the next 3 years, and then every 6–12 months thereafter. Follow-up comprised physical examination, serum biochemical profile, urine cytology, cystoscopy, chest X-ray or computed tomography (CT), and abdominopelvic CT. Bone scintigraphy was performed when clinically indicated. Tumor recurrence was defined as the development of local recurrence or distant metastasis; this did not include recurrence in the retained urothelium (bladder, urethra, or ipsilateral upper tract).

Statistical analysis

Primary endpoints were RFS and CSS. Clinicopathological characteristics were compared between the AC group and surgery-alone (SA) group using the Chi square test or the Mann–Whitney U test. RFS and CSS from the date of surgery were calculated using the Kaplan–Meier method. Univariate and multivariate analyses with Cox proportional hazards regression models addressed associations between potential prognostic factors and survival rates. Variables with statistical significance in univariate analysis and those with significant differences in background characteristics were used in the multivariate analysis. The Harrell’s C-statistics for the identified Cox model was estimated. Statistical analysis was performed with JMP®Pro version 11.0.0 and SAS software version 9.4 (SAS Institute, Cary, NC, USA). A p value <0.05 was considered significant.

Results

One hundred and nine patients with pT3-4pN0/xM0 UTUC were identified and included in the present study. Patient characteristics are summarized in Table 1. The median patient age was 71 years [interquartile range (IQR), 64–77]. Forty-three patients (39.5%) underwent at least one cycle of adjuvant chemotherapy (AC group), starting 35 days (median; range 19–102) after surgery: 3, 23, 14, and 3 patients underwent 1, 2, 3, and 4 cycles of chemotherapy, respectively. MVAC, GC, MEC, and DIP regimens were administered to 21, 14, 7, and 1 patient, respectively (Supplementary Table 1). Patients in the AC group were significantly younger than those in the SA group (median 67 versus 75 years, p < 0.0001), but otherwise there were no significantly different characteristics between the two groups (Table 1). Median follow-up was 46.5 months (IQR 23.2–76.7) for all patients, and 53.3 months (IQR 27.9–91.1) for those who were alive at the last follow-up. The follow-up duration for those who were alive at the last follow-up was not significantly different between AC group and SA group.

Table 1 Characteristics of the patients
Full size table

At the final follow-up, 11 (25.6%) and 29 (43.9%) patients had developed recurrence and/or metastasis in the AC and SA groups, respectively. On univariate analysis, administration of AC was associated with better RFS (p = 0.0262, Fig. 2a), as well as preoperative hydronephrosis, lower nuclear grade (G2), and negative lymphovascular invasion (LVI) (Table 2). On multivariate analyses, AC was the sole independent predictive factor for better RFS [p = 0.0178, hazard ratio (HR) = 0.41]. Harrell’s C-statistics of the model that does not include AC was 0.78 (95% CI 0.69–0.87). The predictive accuracy of the model slightly improved by including AC, with C-statistics of 0.81 (95% CI 0.72–0.90).

Fig. 2
figure 2

Kaplan–Meier survival curves for 109 patients according to treatment with adjuvant chemotherapy: a progression-free survival, and b cancer-specific survival after nephroureterectomy. AC adjuvant chemotherapy, SA surgery-alone

Full size image
Table 2 Univariate and multivariate analyses of the associations of clinicopathological variables with recurrence-free survival
Full size table

Six (14.0%) and 18 (27.3%) patients ultimately died of disease progression in the AC and SA groups, respectively. On univariate analysis, CSS was significantly better in the AC group than in the SA group (p = 0.0395, Fig. 2b). Lower Charlson Comorbidity Index (CCI), preoperative hydronephrosis, lower nuclear grade (G2), and negative LVI were also associated with improved CSS (Table 3). On multivariate analysis, AC (p = 0.0375, HR = 0.33), lower nuclear grade (p = 0.0070), and the absence of hydronephrosis (p = 0.0493) were independently associated with better CSS. C-statistics of the multivariate models with and without AC were 0.73 (95% CI 0.62–0.83) and 0.70 (95% CI 0.60–0.81).

Table 3 Univariate and multivariate analyses of the associations of variables with cancer-specific survival
Full size table

Discussion

The role of post-surgical AC in the treatment of UTUC is controversial. Although some large-scale multi-institutional studies revealed negative results [6, 7, 9], a recent meta-analysis of several retrospective studies suggested that AC might demonstrate efficacy in the treatment of advanced UTUC [10, 14].

Our study has several advantages over previous large-scale studies [6, 7, 9, 13]. First, as many as 40% of the patients in the current study underwent AC. In the four previous large-scale multi-institutional studies, only 19 to 23% of the cohort underwent AC, implying strong selection bias. Second, all AC regimens employed in our cohort were cisplatin-based, although the dosage needed to be reduced in some patients due to impaired renal function. Cisplatin-based regimens demonstrate greater effects than other regimens including carboplatin-based or non-platinum regimens. A previous meta-analysis study of AC in UTUC reported that a statistically significant survival benefit was seen only in the pooled analysis of cisplatin-based studies [10]. Third, we restricted the study population to patients with locally advanced disease without nodal involvement. The majority of the previous studies included both locally advanced (pT3-4) and node-positive cases. Nodal involvement implies metastatic potential, suggesting that these tumors may be more aggressive. The multi-institutional study from France included 107 cases with distant metastasis in a total study population of 627 [7]. By excluding such cases, the heterogeneity of the study population was reduced and the statistical power of the study may have been strengthened. Fourth, the follow-up period of our study (46.5 months) was much longer than those of the previous three studies (22–26 months) [6, 7, 9].

Currently, there is no clear evidence that systemic chemotherapy is effective in the treatment of UTUC; the use of systemic chemotherapy depends on an extrapolation of evidence from studies in bladder cancer. In bladder cancer, two RCTs revealed that neoadjuvant chemotherapy demonstrated a clear benefit in the treatment of muscle-invasive disease [19, 20]. Similar to urothelial bladder cancer, UTUC is chemosensitive. Therefore, it would be reasonable to incorporate perioperative systemic chemotherapy into the management strategy for locally advanced UTUC.

A number of groups have advocated the presurgical use of chemotherapy for patients with advanced UTUC [21]. In UTUC, the postoperative decrease of renal function is more profound than that in bladder cancer due to the loss of the renal unit, which may preclude patients from receiving cisplatin-based AC [22]. Concerns regarding neoadjuvant chemotherapy include the delay in surgery, potential primary chemoresistance, and overtreatment due to lack of accurate preoperative staging tools.

Currently, cytotoxic chemotherapy is a standard treatment for metastatic UTUC. Recently, newer targeted drugs and immunotherapies have been gaining attention and rigorously studied in urothelial carcinoma, including immune checkpoints inhibitors and fibroblast growth factor receptor (FGFR) inhibitors. [23] Although most of them are being studied for metastatic diseases, anti-PD-1/PD-L1 antibody agents have now being evaluated in adjuvant or neoadjuvant settings [23]. In the future, there may be a paradigm shift in the treatment of advanced UTUC, including adjuvant treatment strategy.

Our study was limited by its small sample size and retrospective design. In particular, treatment strategies that included lymph node dissection and chemotherapy regimens had not been standardized during the long study period. In addition, the current cohort lacked the data on several potential prognostic factors, including concomitant carcinoma in situ and tumor size. A well-organized prospective RCT is needed in order to further clarify the role of AC in the management of patients with advanced UTUC.

Conclusions

Cisplatin-based postoperative AC improved RFS and CSS in patients with locally advanced (pT3-4pN0/xM0) UTUC. Although a further confirmatory prospective study is required, AC may be considered a reasonable treatment option for applicable patients.