Combination of Gemcitabine and Carboplatin in Urothelial Cancer Patients Unfit for Cisplatin due to Impaired Renal or Cardiac Function

ARTICLE INFO _______________________________________________________________ _____________________

Recently, a large randomized phase III trial has established gemcitabine/carboplatin as the standard therapy for patients with metastatic urothelial cancer who are unfit for cisplatin-based therapy.An objective response rate of 41.2% was observed in 119 patients.The progression-free survival was 5.8 months, and the survival was 9.3 months.Efficacy was affected by the Bajorin risk groups: good prognosis, intermediate and poor prognosis -12.0, 9.3, and 5.5 months, respectively.Unfortunately, despite a higher response rate than the conventional therapy (carboplatin/vinblastine/methotrexate), the overall survival was similar (13,14).
In the present study, we focus on the effect of this combination in patients unfit for cisplatin due to decreased renal function or cardiac function, thus precluding the necessary forced hydration required for cisplatin therapy, irrespective of performance status and disease extent.

Patient Selection
Patients were required to have histologically proven locally advanced or metastatic urothelial TCC of the urinary tract, not amenable to curative surgical or radiotherapeutic treatment and with measurable disease as defined by the Response Evaluation Criteria in Solid Tumors (14).Ineligibility (unfit) for cisplatin was defined as measured creatinine clearance < 60 mL/min or cardiac ejection fraction < 50% or active cardiac ischemia proven by nuclear studies.Additional eligibility criteria included an estimated life expectancy of at least 12 weeks and Karnofsky performance status score ≥ 40%.Previous treatment with adjuvant or neoadjuvant chemotherapy was allowed if completed more than 6 months before the beginning of the study.Laboratory requirements included adequate bone-marrow (white blood cells > 3500/μL, absolute neutrophil count (AnC) > 1.5 ×10 3 /μL and platelets >125×10 3 /μL), bilirubin level ≤ 1.6-7.0mg/dL (15).Exclusion criteria included: the presence of brain metastases or other central nervous system lesions; a concomitant malignancy except for cured basal-cell skin cancer or carcinoma in situ of the cervix; and pregnancy.Patients were required to comply with all the requirements of the trial.The study was approved by the Ethics Committee of the center and each patient gave written informed consent before study entry.Response was prospectively documented while toxicity was retrospectively collected.

Therapy
Patients received gemcitabine at a dose of 1000 mg/m 2 as an intravenous infusion over 30 minutes on days 1 and 8, combined with carboplatin AUC = 4.5 on day1 every 3 weeks.
Treatment was continued until detected disease progression or intolerable toxicity.We attempted to deliver at least 6 cycles, with 2 cycles beyond documented complete response.
If patients required more than 2 weeks for hematologic recovery, or there was grade 4 neutropenia with fever, grade 4 thrombocytopenia for more than 3 days, or thrombocytopenia with active bleeding during the nadir, treatment was continued with 25% dose-reduction of both drugs.Granulocyte colony-stimulating factors were not routinely used.They were introduced, without treatment dose modification, in the event of prolonged (> 2 weeks) neutrophile recovery or grade 4 neutropenia with fever.In the event of subsequent grade 3-4 hematologic toxicity, doses were reduced by 25%.Similar dose modification was utilized for any grade 3-4 non-hematologic toxicity.Only 2 dose modifications were allowed.

Therapy Management & Assessment
Toxicity was evaluated using Common Toxicity Criteria, Version 2.0.All patients underwent a complete medical history, physical examination, blood count, blood chemistries and measured creatinine clearance before each treatment cycle.
Response to treatment was assessed by computed tomography every 2 cycles using Response Evaluation Criteria in Solid Tumors criteria 1.0 (15).
Sample size was based on Response Rate (RR).The smallest response probability (rejected threshold) was set to 20% (p0 = 0.2), as we believed that an RR of at least 20% would be beneficial to the patient, the largest response probability was based on the assumption that a 40% response rate -a desirable response (p1 = 0.4), would be detected.According to Simon, the sample size for detection of the desired width of 20% -40% RR (alpha = 0.1, one tail; beta = 0.2) at the first stage is 23 patients (16).Additional observations included safety and overall survival (16).We report here the results of the first stage.

REsuLTs
During the last 7 years we treated 23 patients who fulfilled the required inclusion crite-
Analysis according to Memorial Bajorin's prognostic groups revealed that the median Therapy was not devoid of toxicity (Table-3).nine patients (39%) developed infection, including one patient who died of pneumonia.This patient had a low performance status (Karnofsky performance status -40%) with creatinine clearance of 34 mL/min.Additional toxicity included: bleeding ≥ grade 2 in 3 (13%) patients, fatigue grade 2 in 5 (22%) patients and fatigue
We have also demonstrated a better outcome among Bajorin's good prognosis patients compared to the intermediate and poor prognosis groups.However, the small sample size prevented distinguishable separation between the intermediate and poor prognostic categories (3,14) not allowing a randomized trial study.
These findings re-emphasize the need for personalized therapy for the Bajorin's good prognostic group of the cisplatin-unfit population.This group is characterized by higher treatment efficacy and the lowest toxicity profile with the gemcitabine/carboplatin combination (13).Thus, survival prolongation is a legitimate objective in these patients.Methods to expand the cisplatinbased therapy options in this group should include improving the selection of patients who may safely received cisplatin-based therapy, optimizing the drug administration and developing effective chemo-protective approaches (17).
Various methods for estimating renal function have been explored, many found to be sub-optimal (17).This may explain the widely accepted threshold of 60 mL/min creatinine clearance for cisplatin administration.It appears that measured creatinine clearance is more accurate than calculation with the Cockcroft and Gault equations (18).It is conceivable that utilizing measured creatinine clearance, or adminis- tering cisplatin over a longer period, it would be possible to lower this threshold (19).non-nephrotoxic regimens have not yet emerged as potential replacements for cisplatinbased therapy in cisplatin-unfit patients.A recent publication of first line therapy with the combination gemcitabine/paclitaxel reported a response rate of 37%.This trial recruited patients with creatinine clearance > 40 m/min, which falls within the definition of cisplatin-unfit patients (20).Unfortunately, the authors did not report the efficacy of this regimen in this population.
Despite the low response rate of gemcitabine/carboplatin for the cisplatin-unfit population (Table-4), the acceptable toxicity pattern signals this regimen as a potential therapy for such patients.
Angiogenesis, another non-nephrotoxic treatment alternative, has been shown to be important in the development and progression of urothelial cancer (21).Thus, several trials with sunitinib, sorafenib and pazopanib have been performed (21)(22)(23)(24).Sunitinib is preclinically active against urothelial carcinoma (25).When given as second-line therapy after chemotherapy in patients fit for cisplatin, it failed to achieve its initial expectations.However, a clinical benefit was observed in some patients (21).Recently, promising results have been obtained with sunitinib in patients unfit for cisplatin.Of the 21 evaluable patients, 3 (8.1%)achieved a partial response, and an additional 14 (53.8%)patients had prolonged (> 3 months) stable disease, with clinical benefits to 62% and a progression-free survival of 5.9 month (22).This is within the efficacy range of gemcitabine/carboplatin regimens (Table - 4).A potential approach is a sequential combination of sunitinib to patients who have not progressed on gemcitabine/carboplatin therapy.Preliminary data suggested that such a scheme is feasible in fit-forcisplatin-based therapy patients (26).
In our experience, the therapy with gemcitabine in combination with carboplatin was associated with significant toxicity.Our sample size prevented a reliable statistical categorization according to Bajorin prognostic factors.However, a larger trial with this combination demonstrated recently that toxicity was the lowest in the Bajorin good prognostic group.This suggests that with the lack of more effective and less toxic therapeutic options, gemcitabine/carboplatin therapy should be limited to patients with Bajorin good prognostic group.Therapy with poor prognostic features should focus on palliative endpoints, while therapy in the intermediate group should be determined on a case by case review.
Finally, 3 of our long-term survivors who achieved a partial response to therapy underwent resection of residual disease.Our data suggested that this is feasible also in cisplatin-unfit patients (27).

CONCLusIONs
Our experience with the gemcitabine/carboplatin combination is within the range of the published data.This regimen is active mostly in the Bajorin's good prognosis group.It is associated with considerable toxicity.Patients with a partial response should be considered for resection of residual disease.

ANC
= absolute neutrophil count Carbo = carboplatin CCT = creatinine clearance time CR = complete response G = gemcitabine PS = WHO performance status RR = response rate TCC = transitional cell carcinoma TTP = time to disease progression CONFLICT OF INTEREsT none declared.

Table 4 -gemcitabine/carboplatin combination in patients with metastatic urothelial cancer unfit for cisplatin.
G = Gemcitabine; Carbo = Carboplatin; CCT =, Creatinine clearance time; RR = Objective response rate; PS = WHO performance status; TTP = time to disease progression.