Introduction

In 2000, approximately 36,000 new cases of kidney cancer were estimated to occur in North America and 46,000 in the European Union. In the Scandinavian countries, the number is about 3,000. At the time of diagnosis, about 30% of patients have metastatic renal cell cancer (MRCC), and furthermore, approximately 50% of patients with renal cell cancer who undergo resection with curative intent subsequently develop local relapse or metastatic disease.

So far, traditional cytotoxic chemotherapeutic regimens have failed significantly to change the natural course of MRCC [1]. The aim of the present article is to review the latest single, multiagent phase II/III trials, and to give a short overview of novel drugs to be tested in patients with MRCC.

General remarks

Simultaneous resistance of malignant cells against multiple antineoplastic agents which are structurally and functionally unrelated from each other is known as multidrug resistance (MDR), one of the main causes of chemotherapy failure in kidney cancer [2].

The prototypic transporter protein responsible for this phenomenon is MDR1-P-glycoprotein (MDR1-Pgp), discovered in 1976 by Juliano and Ling [3]. Overexpression of MDR1-Pgp protects kidney cancer cells and normal tissue from penetration by drugs and toxins [4, 5].

Other factors involved in MDR are lung cancer multidrug resistance protein (LRP), glutathione increase, p53 mutations or amplification of topoisomerase II.

The activity of detoxifying enzymes such as glutathione-S-transferase might contribute to cellular MDR in kidney cancer [6]. Elevated intracellular glutathione levels add to resistance against doxorubicin, cisplatin, carboplatin or oxaliplatin [2]. Further, slow tumor proliferation may be another factor responsible for the observed lack of chemotherapeutic efficacy.

To overcome chemoresistance in kidney cancer, several MDR1-Pgp modulators have been used with the aim of reducing MDR [7, 8]. Agents such as medroxyprogesterone acetate, tamoxifen, cyclosporine and verapamil can reverse the MDR-mechanism to a certain degree, but do not enhance the response rates to cytotoxic drugs in MRCC.

Most of the sporadic kidney cancers arise from the proximal tubule and show inactivation of the von Hippel Lindau gene (VHL) [9]. The VHL gene maps to the short arm of chromosome 3. The gene encodes for the VHL protein (pVHL), which is implicated in the regulation of the cellular response to hypoxia and the production of angiogenic peptides, such as the vascular endothelial growth factor (VEGF) [10]. Under physiological conditions, angiogenic peptides are upregulated only after hypoxia. Hypoxia inducible factor-1 (HIF-1) is the major transcription factor that activates a diversity of genes including VEGF [11]. Tumor growth depends on angiogenesis. The hypervascularity of kidney cancers may be related to the overproduction of angiogenic peptides due to pVHL deficient cells. In contrast to the physiological expression of angiogenic peptides, the upregulation of such peptides in kidney cancers is independent of environmental oxygen tension [11].Thus, hypoxia and the overexpression of HIF-1 have been associated with an increased risk of invasion and metastases, as well as a poor outcome in certain malignancies [12]. Hypoxia might be an additional factor in the lack of efficacy of many anticancer drugs used in MRCC. This relationship should be investigated in the design of future trials.

Several prognostic models for predicting survival in patients with MRCC have been developed [13]. Impaired functional status seems to be one major factor, in addition to low hemoglobin, increased serum calcium and a high erythrocyte sedimentation rate.

Since the introduction of targeted therapy, alone or in combination with conventional chemotherapy, difficulties in measuring metastatic lesions have become obvious [14]. Recent experience in pulmonary lesions from non-small cell carcinoma of the lung showed considerable interobserver misclassification of progression [15]. The need for a central review of response evaluation in phase II trials has to be addressed. It is also questionable whether response according to the RECIST criteria [16] is the only valid outcome measure in MRCC. “Stabilisation” of previously progressing disease may represent a favourable clinical outcome in future trials [17].

Conventional chemotherapy

In the 1990s, the most extensively studied drugs were floxuridine and fluoruracil (5-FU). Interestingly, 5-FU is not removed from the cell by an MDR-like mechanism. In one trial, a 20% response rate was reported with continuous infusion of floxuridine administered according to a circadian schedule [18]. A comprehensive review on these regimes by Yagoda et al. in 1995 revealed an overall response rate of 6% among 4,093 patients with advanced MRCC who received adequate treatment, with only a slight improvement (response rate 14.6%) in a subgroup of patients who were treated with an antimetabolite (floxuridine or 5-FU) [19].

Responses were generally short, lasting a few months. The addition of fluoruracil modulators, such as calcium folinate did not increase the response rates.

The response rate to vinblastine alone was 3% in 277 patients [20]. Keeping in mind the peculiar behaviour of MRCC with a natural course ranging from a few months to several years, and a spontaneous remission rate of 5% [21], vinblastine must be regarded as ineffective.

Conventional chemotherapy in combination with novel drugs

From 2000 through July 2004, a variety of chemo-, chemoimmuno- and immunotherapeutic agents were studied in phase II-III trials (Table 1). The best response rates were reported in phase II trials of combinations using all previously described modalities. Ryan et al. tested the antitumor activity of gemcitabine combined with 5-FU or with 5-FU plus immunotherapy and found overall response rates of 17% and 14.6%, respectively [2]. Formerly described prognostic factors for survival in MRCC using gemcitabine plus 5-fluorouracil were confirmed in a recently published article by Stadler et al. [23].

Table 1 Metastases in renal cell carcinoma. New agents and new combination schedules (2000–2004). IFN: interferon-alpha, IL-2: interlekin-2, 5-FU:5-fluoruracil; CRA: 13-cis-retinoic acid
Full size table

The efficacy and safety of the FOLFOX-4 regimen was investigated in 59 patients by Bennouna et al. who concluded that oxaliplatin had no role in the treatment of MRCC [24]. Combinations with vinblastine and estramustine in patients with MRCC had minimal activity [25]. The substitution of capacetabine for 5-FU in pre-existing triple regimes (5-FU, vinblastine, IFN) showed significant antitumor activity. Thirty patients were enrolled in a study by Oevermann et al. and the therapy was well tolerated [26].

Despite a high percentage of disease stabilization in a phase II study by Fizazi et al., capacetabine had limited activity in patients with MRCC [27].

New agents and combination schedules in the treatment of MRCC

Gemcitabine is not a known substrate for p-glycoprotein. This novel agent leads to moderate responses [28], either alone or in combination with 5-FU, but no further improvements were recorded when either cisplatin, interferon, or interleukin-2 were added. However, other results of gemcitabine monotherapy have been less favorable, as shown in a recent report from a Dutch phase II study [29]. In recent phase II trials, a combination of gemcitabine and oxaliplatin achieved partial responses in up to 14% of patients, with acceptable toxicity profiles in patients with immunotherapy resistant advanced MRCC [30]. Weekly gemcitabine with continuous 5-FU infusion is an active combination in patients with MRCC. This regime is well tolerated and produced an improvement in progression free survival in a phase II trial conducted by Rini et al. [31]. These authors reported an objective partial response in 17% of patients.

Studies including the topoisomerase I inhibitors irinotecan and topotecan have shown that these substances are not efficacious in the treatment of MRCC [32]. So far, the taxanes have demonstrated no significant activity [33].

In phase I studies, the alkylating agent temozolomide exhibited broad antitumor activity in patients with MRCC. Park et al. investigated its efficacy in a phase II trial on 12 patients [34]. This trial ended up with disappointing results. Temozolomide’s ineffectiveness might be due to high levels of alkylguanine-DNA alkyltransferase measured in some biopsies of these patients.

The NCI undertook a multi-institution phase II study of troxacitabine (BCH-4556) to evaluate its efficacy and safety [35]. Between June 1999 and March 2000, 35 patients were treated with troxacitabine given as an intravenous infusion over 30 min at a dose of 10 mg/m2, once every 3 weeks. This nucleoside analog had modest activity against MRCC (two PR of 33 patients, 21 patients had stable disease), but larger studies are needed to confirm this.

The antitumor activity of irofulven was determined in 13 patients with MRCC by Berg et al. [36]. Irofulven was administered at a dose of 11 mg/m2 by 5-min intravenous infusion, on five consecutive days. Toxicity included myelosuppression and gastrointestinal side effects. No major responses were achieved. Irofulven did not produce a clinical response in patients with MRCC.

Recent phase II and III studies clearly support the overall, well documented finding that MRCC is mainly refractory to chemotherapeutic agents. However, some MDR-independent agents, such as 5-FU and gemcitabine, have shown a modest activity of limited duration.

Combined immunotherapy with chemotherapy

A promising breakthrough in the treatment of advanced MRCC was made when the interferons showed sustained activity against this disease. The natural glycoprotein interferon-alpha (IFN) stimulates host mononuclear cells and is clearly involved in host immunosurveillance. Furthermore, it has been reported to influence angiogenesis, in part through down-regulating basic fibroblast growth factor [1]. Interleukin-2 (IL-2) stimulates natural killer cells and helper T-cells without direct effects on the tumor. In MRCC, the highest response rates were reported in a high-dose regimen (600,000–720,000 IU/kg every 8 h for 5 days), which resulted in a 19% response rate with 5% complete response [37].

Stimulated by the observed response rates when applying single drugs, trials have been performed using combinations of IFN or IL-2 with the best known active chemotherapeutics. IFN in combination with vinblastine has failed to improve survival compared to IFN alone [38]. The Atzpodien regime, consisting of 5-FU, IL-2 and IFN, has been regarded by some investigators as the gold standard for patients with MRCC. Comparisons of this regimen with IFN alone are ongoing in large randomized phase III trials in Europe [39].

A vitamin A metabolite, 13-cis retinoic acid (CRA), has shown antitumor activity. In the treatment of carcinoma, CRA may be important for inhibiting tumor cell expression of IL-6 receptors and may have some immunologic and antiangiogenic effects. A recent report by EORTC supports the promising results of this agent in combination with IFN [40]. On the other hand, Atzpodien and colleagues could not confirm the usefulness of 13-CRA in MRCC when compared to IL-2 combined with IFN [38]. Neither could Motzer et al. show that CRA increases the response rate of IFN [41].

Sunklara et al. investigated the efficacy of temozolamide in combination with IFN in 14 evaluable patients. The results were rather disappointing with only one minor response [42].

The usefulness of IFN-gamma combined with vinblastine with or without 13-CRA was investigated by Bacoyiannis et al. [43]. IFN-gamma did not enhance the low response rate of vinblastine based chemotherapy. In addition, administration of CRA to the schedule had no effect.

The French immunotherapy group investigated the role of 5-FU in combination with subcutaneous IL-2 and IFN [44]. They could not see any benefit for their patients with MRCC in terms of improved survival. Further, this group stated that neither of these regimens could be recommended as a standard treatment.

The pyrimidine analogue gemcitabine was tested in combination with IFN and IL-2. Neri et al. found that gemcitabine combined with standard doses of IFN and low doses of IL-2 resulted in objective responses and relatively long-term survival [45].

Ryan et al. studied the response rate and toxicity of 5-FU and gemcitabine intravenously with IL-2 and IFN in 41 patients [22]. The addition of gemcitabine and 5-FU to subcutaneous IL-2 and IFN resulted in a similar response rate to that observed in previous studies of IL-2 based therapy. The toxicity of this four drug regimen was significant.

The drug paclitaxel was added to the known schedule of IFN and 13-CRA by Vaishampayan et al. in a phase II evaluation [46]. This combination was well tolerated but had minimal efficacy in advanced MRCC.

Hormone therapy

The discovery of estrogen receptors in MRCC encouraged the use of hormones in animals studies. The gestational agent medroxyprogesterone acetate and antiestrogens such as tamoxifen have been tested in several trials in the treatment of MRCC in the last 30 years [4, 47]. Nevertheless, the clinical response rates have been rather low (5%).

Recently, toremifene showed encouraging response rates in 19 RCC patients who received 360 mg per day of this antiestrogen [8]. Similar results were reported in an earlier trial using 300 mg toremifene per day [48]. These results await confirmation in a large phase III study. Further approaches with the new class of aromatase inhibitors have not been conducted to our knowledge.

In a phase I-II study, the tolerability of high dose toremifene in combination with vinblastine was evaluated in 26 patients [8]. The regimen was well tolerated and serum concentrations of toremifene reached levels which were believed to reverse MDR in vivo.

To overcome MDR induced chemoresistance, Liu et al. tried to combine tamoxifen and colchicine modulated vinblastine followed by 5-FU [49]. Of 17 eligible patients, one achieved a complete response and three a partial response with an overall response rate of 23.5%. The treatment toxicity was limited.

Conclusion

In recent years, clinical trials have confirmed that MRCC is a malignancy resistant to todays available systemic treatment. In addition, most of todays approaches have substantial side effects. Novel approaches with targeted therapy are promising. Their clinical significance has to be documented in clinical studies.