The patients had unresectable, locally advanced or metastatic, and histologically-confirmed TCC of the bladder, ureter or renal pelvis, with documented progressive disease (PD) prior to registration and measurable disease according to the Response Evaluation Criteria In Solid Tumors (RECIST) [27]. All patients had to receive no more than one previous line of systemic chemotherapy for advanced disease. Prior radiotherapy was allowed, but a minimum of four weeks had to elapse between the end of the prior radiotherapy and patient enrolment. Patients were ≥ 18 years old, had an Eastern Cooperative Oncology Group (ECOG) performance status (PS) score of ≤ 2 and had recovered from any toxicity derived from previous treatments. Patients had an adequate organ function as defined by neutrophil count ≥ 1.5 × 10⁹ /L, platelet count ≥ 100 × 10⁹ /L, hemoglobin ≥ 9 g/dL, creatinine clearance ≥ 40 mL/min, serum bilirubin ≤1.5 mg/dL, alkaline phosphatase (AP) ≤ 2.5 × the institutional upper limit of normal (ULN) and up to ≤ 5 × ULN in case of extensive bone metastases, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) ≤2.5 × ULN and up to ≤ 5 × ULN in case of liver metastases, albumin ≥25 g/L and normal left ventricular ejection fraction (LVEF).

Patients were excluded if they were pregnant or lactating women or were not using adequate contraception methods, had a history of other neoplastic diseases (with the exception of adequately treated non-melanoma skin carcinoma or carcinoma in situ), had known brain or leptomeningeal involvement, suffered from any other relevant medical condition that may harm study participation, or had known hypersensitivity to plitidepsin or to any component of the reconstitution solution (cremophor EL, mannitol and ethanol).

Plitidepsin (Pharma Mar, Colmenar Viejo, Madrid, Spain) was infused at a dose of 5 mg/m² as a 3-hour i.v. infusion every 2 weeks (i.e., one cycle = two weeks). Treatment cycles were repeated for a maximum of 12 (~six months) or until disease progression, unacceptable toxicity, patient refusal or treatment delay for >2 weeks. In the event of specific toxicities after the first cycle, the dose of plitidepsin was to be reduced in subsequent cycles, first to 4.25 mg/m² and then to 3.6 mg/m². No further dose reductions were allowed.

Treatment with hematopoietic growth colony stimulating factors was allowed (except during the first cycle) in accordance with the guidelines of the American Society of Clinical Oncology (ASCO) [28]. Prophylactic treatment for emesis, which included glucocorticoids (dexamethasone 4–8 mg i.v.), H1-receptor and serotonin (5-HT₃) receptor antagonists [29], was administered 20–30 min before each plitidepsin infusion. Patients developing grade ≥2 muscular toxicity could be empirically treated with L-carnitine (1.5 g every 8 hours) until returning to grade ≤1 severity.

Radiological evaluation (chest X-ray, computer tomography [CT] scan or magnetic resonance imaging [MRI]) was required within four weeks before start of study treatment, and within six months prior to registration in order to document pre-baseline disease progression. While on treatment, a complete clinical examination was done at the end of each cycle (i.e., every two weeks). Blood samples for hematological and biochemistry analyses were obtained weekly during the first eight weeks of therapy and afterwards at least every two weeks. Creatinine clearance calculation and urine analysis were assessed before every cycle administration, while coagulation tests were performed every other cycle. LVEF measurements were repeated at 3–6 months after the start of treatment. All disease parameters were reassessed every eight weeks by using appropriate imaging procedures for the assessment of measurable and non-measurable lesions as per RECIST.

The primary efficacy endpoint was objective tumor response rate according to RECIST in all patients who received a minimum of two treatment cycles. Whenever the criteria of response were met, re-assessment had to be repeated at least four weeks later in order to confirm response. Any eligible patients who died or stopped treatment because of unmanageable toxicity prior to response evaluation were considered non-evaluable for response.

Secondary efficacy endpoints were the rate of stable disease (SD) lasting for at least six months and time-to-event variables (tumor response duration, time to progression [TTP], progression-free survival [PFS] and OS), and evaluation of the safety profile of plitidepsin. Patients were evaluable for toxicity if they had received at least one plitidepsin infusion. Treatment-related toxicities were graded according to National Institute of Health Common Toxicity Criteria (NCI-CTC), version 3.0 [30]. If toxicity persisted after the end of treatment, further assessments were performed 30 days after the administration of the last protocol treatment or at the resolution of all abnormalities that occurred during treatment.

A Simon’s two-stage MiniMax design was used to test the null hypothesis that the probability of objective response (confirmed PR or CR) was H0 ≤5% versus the alternative hypothesis that it was HA ≥ 20% [31]. After testing the drug on 18 patients in a first stage, the trial was to be terminated if no responses were observed. Otherwise, the trial was to go onto a second stage and other 14 patients were to be evaluated to reach a total of 32 patients evaluable for efficacy. If the total number of patients with an objective tumor response was ≤ 3, this schedule was not to be considered for further evaluation. Descriptive statistics were used to characterize demographics, response and toxicity rates and laboratory observations. Time-to-event efficacy endpoints and their fixed time point rates were analyzed according to the Kaplan-Meier method [32].

Patient and disease characteristics at baseline are shown in Table 1. The median age was 64 years, 76.2% were males and 95.2% had an ECOG PS score of 0 or 1.

All patients had metastatic and/or locally advanced TCC, except for one who had poorly differentiated urothelial carcinoma. Patients had a median of one metastatic site (range, 1–3), with lymph nodes as the most common tumor site involved at baseline (52.4% of patients), followed by the liver (28.6%) and the lung (14.3%).

All patients had previously received chemotherapy (Table 2); additionally, three patients also received biological therapy (BCG vaccine two patients and trastuzumab one patient). Most patients (76.2%) had previously received cisplatin (81.0%) and gemcitabine (71.4%). All patients had undergone surgery, mostly consisting of radical cystectomy (71.4%), while radiotherapy had been administrated to 5 patients (23.8%) in the palliative setting.

A total of 57 cycles (median: two per patient; range: 1–8) were administered during the study. The median total cumulative dose delivered was 10.0 mg/m² (range, 5.0–40.0 mg/m²) and the median dose intensity was 2.5 mg/m²/wk (range, 1.3–2.5 mg/ m²/wk), which corresponded to a median relative dose intensity of 99.4% (range, 52.1–101.0%).

Cycle delays occurred in four patients (19.0%); all due to reasons unrelated to plitidepsin. The length of delay was 3–7 days. Thirteen patients (61.9%) discontinued treatment early because of disease progression, while other patients discontinued due to death as a result of disease progression (n = 3, 14.3%), treatment refusal (n = 2, 9.5%) and other reasons (n = 2, 9.5%) such as general condition deterioration and worsening of Parkinson’s disease. Only one patient (4.8%) discontinued due to plitidepsin-related increase in creatine phosphokinase (CPK) levels after three cycles.

Three of 21 treated patients were non-evaluable for efficacy due to not having undergone any tumor assessments before discontinuing the study as a result of patient refusal (n = 2) and worsening of Parkinson’s disease (n = 1). No objective tumor responses were found among the 18 evaluable patients. Two patients (11.1%) achieved SD according to RECIST that lasted less than six months (2.2 months and 3.7 months). With a median follow-up of 4.6 months, the median PFS was 1.4 months (95% CI, 0.9–1.6 months) and the median OS was 2.3 months (95% CI, 1.4–4.6 months).

All 21 treated patients were assessable for toxicity. Nausea (38.1% of patients), fatigue (28.6%), myalgia (28.6%) and anorexia (23.8%) were the most frequent adverse events (AEs) related to plitidepsin. No plitidepsin-related grade four AEs were observed. No plitidepsin-related hospitalizations or deaths occurred during the study. Table 3 shows all the AEs reported during this study regardless of relationship observed in at least 5% of patients. One patient received only one administration of L-carnitine due to muscular weakness and increased CPK levels.

The most frequent biochemical abnormalities were transient and asymptomatic increase in AP (76.2% of patients), creatinine (61.9%) and hepatic transaminases (ALT: 61.9% and AST: 52.4%) (Table 4). Grade 3 AP increases occurred in 4 patients, whereas one patient had grade 3 ALT increase. Additionally, one patient had grade 3 bilirubin increase after one cycle. No patients had grade 3/4 increases in AST, creatinine or CPK levels.

The most common hematological abnormalities were anemia and lymphopenia, which were observed in 21 (100.0%) and 11 (55.0%) patients, respectively (Table 4). Anemia and lymphopenia were grade 3 in one patient each; however grade 3 lymphopenia was already present at baseline. Additionally, one patient had grade 4 lymphopenia after cycle 3. No patients showed leukopenia of any grade, whereas the cases of neutropenia and thrombocytopenia found during the study were grade 1 only.