More than 50% of patients with advanced urothelial carcinoma are not eligible for the standard treatment with cisplatin-based chemotherapy. In general, cisplatin-ineligible patients with metastatic urothelial cancer experience poor outcomes with standard treatment, although substantial heterogeneity exists. Baseline variables associated with poor prognosis include borderline performance status, presence of visceral metastases, liver metastases, and low hemoglobin. Although no standard treatment has been defined for cisplatin-ineligible patients, recommendations regarding carboplatin-based combination chemotherapy versus single-agent chemotherapy versus best supportive care are typically based on performance status and renal function. The clinical development of novel agents is of considerable interest.
Key points
- •
Patients not eligible for cisplatin are defined by one of the following: Eastern Cooperative Oncology Group performance status (PS) greater than or equal to 2, creatinine clearance (glomerular filtration rate [GFR]) less than 60 mL/min, hearing loss greater than or equal to grade 2, peripheral neuropathy greater than or equal to grade 2, heart failure greater than or equal to New York Heart Association class III.
- •
Patients unfit for cisplatin generally have a poor prognosis. Within this group PS 2, presence of visceral metastases, liver metastases, and low baseline hemoglobin are prognostic factors of poor outcome.
- •
Treatment decisions in patients who are unfit for cisplatin are mainly based on 2 factors: PS and renal function. In case of PS greater than or equal to 2 and GFR less than 60 mL/min, treatment consists of best supportive care or single-agent chemotherapy. In case of 0 or 1 risk factor, carboplatin-based combination chemotherapy is the preferred option with the highest level of evidence. Patient inclusion in clinical trials is highly recommended.
Introduction
Cisplatin-based combination chemotherapy is effective and can considerably improve outcomes in patients with advanced urothelial carcinoma (UC), with response rates of more than 50% and 5-year survival rates of 33% in patients with good performance status (PS; ie, Karnofsky score >80%) and no evidence of visceral metastases. Patients fit for cisplatin, but with 1 or both of these poor prognostic factors, experience considerably worse outcomes. Poor outcomes have also been shown when carboplatin is substituted for cisplatin. Although no adequately powered randomized trials have been completed comparing cisplatin-based with carboplatin-based chemotherapy combinations, a meta-analysis showed a significantly increased likelihood of achieving an objective response with cisplatin-based chemotherapy. All practice guidelines therefore support the use of cisplatin-based regimens in advanced UC. In clinical practice, greater than 50% of all patients with advanced UC have contraindications for treatment with cisplatin and alternative treatment options are necessary. There is no consensus on the standard chemotherapy treatment of patients who are unfit for cisplatin. This article reviews the definition of patients not eligible for cisplatin; clinical, pathologic, and molecular prognostic factors for this patient group; as well as different treatment options based on baseline patient characteristics.
Introduction
Cisplatin-based combination chemotherapy is effective and can considerably improve outcomes in patients with advanced urothelial carcinoma (UC), with response rates of more than 50% and 5-year survival rates of 33% in patients with good performance status (PS; ie, Karnofsky score >80%) and no evidence of visceral metastases. Patients fit for cisplatin, but with 1 or both of these poor prognostic factors, experience considerably worse outcomes. Poor outcomes have also been shown when carboplatin is substituted for cisplatin. Although no adequately powered randomized trials have been completed comparing cisplatin-based with carboplatin-based chemotherapy combinations, a meta-analysis showed a significantly increased likelihood of achieving an objective response with cisplatin-based chemotherapy. All practice guidelines therefore support the use of cisplatin-based regimens in advanced UC. In clinical practice, greater than 50% of all patients with advanced UC have contraindications for treatment with cisplatin and alternative treatment options are necessary. There is no consensus on the standard chemotherapy treatment of patients who are unfit for cisplatin. This article reviews the definition of patients not eligible for cisplatin; clinical, pathologic, and molecular prognostic factors for this patient group; as well as different treatment options based on baseline patient characteristics.
Defining patients not eligible for cisplatin
Although the human condition and the heterogeneity of disease necessitates a degree of vagueness and instinct in the practice of medicine, uniformly defining disease states, conditions, and end points is critical to drug development and integral to regulatory science. Several prior clinical trials, discussed in this article, have explored the development of therapeutic regimens in patients with metastatic UC who were considered ineligible for cisplatin. However, these trials generally used variable eligibility criteria, complicating interpretation of the results. In 2011, an expert panel was convened to establish a uniform definition of cisplatin ineligibility to facilitate clinical trials in this patient population for the future. Through consensus, this panel established the definition presented in Box 1 . Although clinical judgment is essential when applying rigid guidelines such as these, to routine patient management, consistency in how patient populations are defined is critical for clinical trial purposes. The definition shown in Box 1 has been adopted in the design of several ongoing clinical trials.
At least 1 of the following:
- •
ECOG PS of 2 (KPS of 60%–70%)
- •
Creatinine clearance less than 60 mL/min
- •
CTCAE v4 grade greater than or equal to 2 audiometric hearing loss
- •
CTCAE v4 grade greater than or equal to 2 peripheral neuropathy
- •
NYHA class III heart failure
- •
Abbreviations: CTCAE v4, common terminology criteria for adverse events, version 4; ECOG, Eastern Cooperative Oncology Group; KPS, Karnofsky performance status; NYHA, New York Heart Association.
Prognostic factors in patients not eligible for cisplatin
Similar to cisplatin-eligible patients with metastatic UC, cisplatin-ineligible patients also represent a heterogeneous group with variable outcomes to treatment. Several analyses have defined baseline prognostic factors in patients with metastatic UC treatment with cisplatin-based chemotherapy. However, baseline prognostic factors have been explored less commonly in the cisplatin-ineligible patient population. Collette and colleagues analyzed the cohort of patients enrolled in EORTC (European Organisation for Research and Treatment of Cancer) 30986, a phase II/III trial of gemcitabine plus carboplatin (GCa) versus methotrexate, carboplatin, plus vinblastine, to identify baseline factors associated with patient survival. On multivariate analysis, the risk of death was significantly increased in patients with a PS of 2, visceral metastases, liver metastases, and a lowered hemoglobin level (≤12 g/dL for women and ≤13.6 g/dL for men). The median survival was 17.4 months in patients with no poor prognostic factors versus 5.5 months in patients with 3 or more poor prognostic factors. Whether impaired renal function is an independent poor prognostic factor in patients with metastatic UC is unclear, given the limitations of the datasets available to address this question.
First-line treatment options in patients not eligible for cisplatin
Although a considerable number of patients with metastatic UC are unfit for cisplatin, few trials in general, and only 1 randomized phase III trial in particular, have been performed. Interpretation of the published trials is difficult in view of the lack of a consensus definition for this patient group until 2011. In addition, assessment of PS is complicated because it can be related to the underlying malignancy or related to comorbidities and age. It is unclear whether these different underlying causes of functional impairment confer similar prognostic and treatment implications. A response to chemotherapy could theoretically result in an improvement in the former but not necessarily the latter. In general, constitutional symptoms such as fatigue, weight loss, anorexia, and failure to thrive are typical signs of advanced or metastatic disease and denote a poor prognosis. In rare cases, patients may have constitutional symptoms caused by potentially reversible complications of cancer or other comorbidities, such as renal failure caused by bilateral ureteral obstruction. In such situations, identifying the cause of the deterioration may help to improve symptoms and functional status, and increase the likelihood of delivery of standard anticancer therapy with subsequent improved prognosis. Therefore, exploring the cause of a patient’s symptoms, rather than assuming that all symptoms in patients with advanced cancer are direct systemic manifestations of the cancer itself, is important for optimizing care.
Clinical trials in cisplatin-ineligible patients have explored both single-agent and combination regimens, which are discussed in further detail later.
Single-agent Chemotherapy
Several chemotherapeutic agents have shown activity in UC when used as single agents. Few trials have included predominantly patients with impaired renal function, reduced PS, or both. All these trials are small and the publications date back 20 years. The patient populations included were heterogeneous and some factors, such as renal function, are not uniformly reported across trials. Several patients included in these trials may be considered cisplatin eligible based on contemporary standards. No randomized trials comparing single-agent therapy with best supportive care or trials comparing one agent with another have been published. The results of these trials must be interpreted with caution and cross-trial comparisons are problematic.
Although some of the monotherapy trials included a large number of patients with PS 2, few patients with PS 3 were included. In these latter patients, the risks of cytotoxic chemotherapy likely outweigh any potential benefits and these patients may be best managed with supportive care alone.
The trials discussed are summarized in Table 1 .
| Combination (Doses mg/m 2 ) | N | Design | Cisplatin Unfit | ORR (%) | Median PFS (mo) | Median OS (mo) | Reference |
|---|---|---|---|---|---|---|---|
| Gem (1000) day 1, day 8, day 15, q4w | 40 | Phase II Single arm | 25% Karnofsky ≤80% S-Creat ≤2 mg/dL for inclusion | 28 | 4.6 | 12.4 | |
| Gem (1200) day 1, day 8, q3w | 25 | Phase II Single arm | 52% PS 2 Median age 76 y | 46 | 5 | 8 | |
| Gem (1200) day 1, day 8, day 15, q4w vs | 22 | Phase II randomized | 41% PS 2 84% GFR<60 | 43 | 3.8 | 5.4 | |
| Gem (1000) day 1, day 15 Oxali (100) day 2, day 16, q4w | 22 | 25% both factors | 28 | 3.4 | 8.1 | ||
| Docetaxel (100) d1, q3w (+G-CSF) | 11 | Phase II Single arm | 26% PS ≥2 100% Creat >1.6 mg/dL | 36 | NR | 11 | |
| Paclitaxel (225) d1, 24-h infusion q3w | 26 | Phase II Single arm | 14% PS 2 S-Creat up to ≤2 mg/dL for inclusion | 42 | NR | 8.4 | |
| Paclitaxel (175) d1, q3w | 13 | Phase II Single arm | 69% PS ≥2 54% S-Creat >1.5 mg/dL | 31 | NR | 9 |
Single-agent chemotherapy with gemcitabine
Several trials explored the efficacy and safety of gemcitabine monotherapy : response rates of up to 46% have been reported, usually short lived and few were complete. The most recent and stringent data derive from the trial of Culine and colleagues. This randomized phase II trial tested monotherapy for gemcitabine against the combination of gemcitabine and oxaliplatin. The population was restricted to patients unfit for cisplatin, and almost half of the patients included had PS 2. The results show the limited efficacy of single-agent gemcitabine, with an overall survival (OS) of only 5.4 months despite a response rate of 43%. In addition, patients in this trial experienced substantial toxicity.
Single-agent chemotherapy with taxanes
Both docetaxel and paclitaxel have been tested as monotherapy in cisplatin-ineligible patients with results similar to those of gemcitabine. High response rates were reported in these studies; however, OS remained short. Roth and colleagues reported complete remissions in 27% of patients treated with paclitaxel lasting up to 16 months. The number of patients with impaired renal function was not reported in this trial. Monotherapy with taxanes causes toxicity with febrile neutropenia in up to 23% of patients. In the reported trials growth factors were not used prophylactically.
Summary of single-agent chemotherapy
Monotherapy with either gemcitabine or taxanes can achieve objective responses in a large proportion of chemonaive patients with advanced UC; however, OS remains poor. In cisplatin-ineligible patients with advanced UC who are also considered poor candidates for carboplatin-based combination chemotherapy, decisions regarding single-agent chemotherapy versus best supportive care require careful consideration of the risks and benefits of each approach.
Combination Chemotherapy
For patients eligible for cisplatin, combination chemotherapy regimens such as methotrexate, doxorubicin, vinblastine, and cisplatin or gemcitabine plus cisplatin have become the standard of care. In view of the modest benefits with single-agent chemotherapy, the use of combination regimens has also been explored in patients unfit for cisplatin. A logical choice was to replace cisplatin with the less nephrotoxic, and generally better tolerated, platinum agent, carboplatin. Another strategy has been to use split doses or lower doses of cisplatin in patients with only mildly impaired renal function. Non–platinum-containing combination regimens have also been explored.
Carboplatin-containing combination chemotherapy
Combination chemotherapy in cisplatin-ineligible patients may be associated with an increase in toxicity compared with combination regimens in fitter patient groups. This association was shown by the dose finding study for the trial EORTC 30986. Based on phase II data obtained in patients with non–small cell lung cancer, an initial dose level of gemcitabine 1000 mg/m² on days 1 and 8 and carboplatin area under the curve (AUC) 5 on day 1 every 21 days was selected. This regimen proved poorly tolerated in a frail patient population with metastatic UC: dose-limiting myelotoxicity was observed in 6 of 8 patients, requiring dose reduction or delay in 5 patients. After reducing the carboplatin dose to AUC 4.5, while maintaining the gemcitabine dose at 1000 mg/m², hematological toxicity was less pronounced and this dose and schedule were chosen for the randomized study. Another trial using carboplatin and gemcitabine in a frail and elderly patient population used an even lower dose of carboplatin (AUC 4). Despite these lower doses, patients experienced considerable toxicity. The EORTC performed the first randomized phase III trial in cisplatin-ineligible patients with metastatic UC. Patients were eligible for this trial based on a World Health Organization PS of 2, and/or impaired renal function with GFR less than 60 mL/min, or both. This pivotal trial was performed in 2 steps as a phase II/phase III trial allowing for premature termination in case of inadequate response rate or excessive toxicity. Patients were randomized to receive methotrexate, carboplatin, vinblastine (M-CAVI) or GCa. The main results are shown in Table 2 . Both regimens showed objective responses in a large proportion of patients, with a numerical advantage for GCa. A statistically significant difference was seen only for the confirmed response rate in favor of GCa. OS and progression-free survival (PFS) were disappointingly low in both treatment arms (9.3 and 8.1 months for GCa and M-CAVI, respectively) in the context of contemporary studies using cisplatin-based chemotherapy. No statistically significant difference between GCa and M-CAVI was seen in either PFS or OS. Severe acute toxicity was higher in the M-CAVI arm. Therefore the use of carboplatin in combination with gemcitabine is a reasonable choice in patients unfit for cisplatin with either PS 2 or impaired renal function with GFR less than 60 mL/min. For patients with both of these factors the response rate was low and OS short. Moreover, severe acute toxicity was high (25%) and the likelihood of receiving only 1 cycle of therapy was 20%.
Combining chemotherapy with antiangiogenic treatment has also been studied in cisplatin-ineligible patients with metastatic UC. Balar and colleagues explored the combination of gemcitabine, carboplatin, and bevacizumab (beva) followed by beva maintenance in patients defined as being unfit for cisplatin based on at least 1 of the following: impaired renal function, solitary kidney, Karnofsky score 60% to 70%, or visceral metastases. In this trial, the response rate was 49% and OS 13.9 months. Similar results were seen in a randomized phase II screening trial comparing vinflunine (VFL) and carboplatin with VFL and gemcitabine. All patients had impaired renal function but a PS of 0 to 1. The OS rates were 12.8 versus 13.9 months for VFL-carboplatin and VFL-gemcitabine, respectively. Whether or not these combination regimens represent an improvement compared with GCa, or whether the outcomes in these trials are a result of more favorable baseline prognostic factors, is unknown and requires assessment in definitive randomized trials.
Further results of carboplatin combination chemotherapy are summarized in Table 2 .
Related posts:
Bladder Cancer
Erratum
Nonurothelial Bladder Cancer and Rare Variant Histologies
Second-Line Therapies in Metastatic Urothelial Carcinoma
Management of Low-risk and Intermediate-risk Non–Muscle-invasive Bladder Carcinoma
First-Line Treatment and Prognostic Factors of Metastatic Bladder Cancer for Platinum-Eligible Patients
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
