Chemotherapy in elderly patients with advanced breast cancer

Table 3. Studies with combination chemotherapy

RR = response rate; MTP = median time to progression; CMF = cyclophosphamide, methotrexate, 5-fluorouracil; *median age.

The authors stopped the study after the first 4 patients enrolled at the first dose-level due to toxicity and reporting no clinical response. They concluded that TXT at 75 mg/m² and over, every 21 days, is too toxic in the elderly (Zanetta et al 2000). On the contrary, Constenia et al (1999) treated 14 elderly patients (> 65 years, six of which frail) with TXT at 75 mg/m² and 100 mg/m², every 3 weeks, with lenograstim support. They reported a RR of 71% with acceptable hematological toxicity. Two trials used weekly TXT. D’hondt et al (2000) administered weekly TXT at the dose of 36 mg/m² in 29 elderly or younger unfit patients, mostly heavily pretreated. The median age was 60 years and the RR was 21% with a good tolerability. In another phase II study, TXT (36 mg/m²) was administered weekly for 6 weeks in 41 elderly or poor performance status patients with ABC. The reported RR was 36% with a 72% of disease control, a median TTP of 7 months, a median OS of 13 months and with 1- and 2-year actuarial survival rate of 61% and 29%, respectively. Most common toxicity was grade 3-4 fatigue occurring in 20% of patients (Hainsworth et al 2001).

Paclitaxel (TAX) was administered weekly (80 mg/m², i.v. day 1, 8 and 15 every 4 weeks), in a phase II study, in 29 elderly patients with ABC. The reported results were 3 complete and 13 partial responses for an overall RR of 55% with mild toxicity (Repetto et al 2002).

Two studies with capecitabine were reported. O’Shaughnessy et al randomised patients in a phase II trial to capecitabine (2510 mg/m²/b.i.d., p.o., days 1 to 14, every 3 weeks) or CMF (cyclophosphamide, methotrexate and 5-fluorouracil). The study accrued 95 women aged > 55 years. Objective response was 25% for capecitabine and 16% for CMF with a median TTP of 132 days and 94 days, respectively. The authors concluded that home-based monotherapy with capecitabine shows at least comparable efficacy to CMF (O’Shaughnessy et al 1998). Procopio et al (2001) treated 40 women older than 65 years with capecitabine (2500 and than 2000 mg/m²/b.i.d., p.o., day 1 to 14, every 3 weeks). They reported RR of 35% with a median TTP of 6 months, one patient died due to gastrointestinal toxicity among 31 evaluable patients.

Four studies reported on single-agent vinorelbine (VNR) in the treatment of elderly patients with ABC (Sorio et al 1997; Buonadonna et al 1998; Vogel et al 1999; Rossi et al 2003). Sorio et al (1997) treated 20 patients (> 65 years) with VNR at the dose of 30 mg/m², i.v., as first-, second- and third-line therapy, on days 1 and 8 every 3 weeks, reporting 30% OR. Buonadonna et al (1998) reported 39.5% RR with MDR of 7 months and a disease control of 60.5%. The schedule of VNR used was 25 mg/m², on days 1 and 8, every 3 weeks. Median age was 70 years and about 40% of patients were treated as second-line therapy. Vogel et al (1999) reported a RR of 38% with MDR of 9 months and a disease control of 76%. VNR was administered at the dose of 30 mg/m², weekly for the first 13 weeks and then every 2 weeks. Median dose intensity of VNR was 20.6 mg/m²/week. Recently, Rossi et al (2003), treated 24 elderly ABC patients with VNR 30 mg/m², i.v. day 1 and 8, every 3 weeks. Nine (37.5%) objective responses (2 complete and 7 partial responses) were observed with MDR and survival of 7 and 11 months, respectively. VNR given on day 1 and 8, recycled every 3 weeks, has a very similar dose-intensity and seems to be better tolerated as compared to weekly administration.

B. Combination regimens

Our literature review showed an increasing interest for oral drugs. In 8 out of the 25 reported studies the Authors used an oral drug formulation. In the future novel biological agents should be an interesting new approach. The use of oral drugs in this setting as well is of great interest.

A. Oral drugs

Patient’s preferences and quality of life issues, which are becoming central considerations in palliative treatment regimens, request the development of oral drugs administration. Indeed, a work has suggested that i.v. lines were a major source of discomfort and stress for cancer patients and approximately 90% of them expressed a preference for oral versus i.v. chemotherapy, predominantly because of the convenience of administration outside a clinical setting or current concerns or previous problems with i.v. access (Liu et al 1997). For the mentioned reasons, if equivalent safety and efficacy can be demonstrated, the oral drugs formulation provides more convenience for patients; this added convenience may be particularly important in elderly and unfit patients. Anyway, the majority of drugs administered orally are intended to act systematically, and for these, absorption is a prerequisite for activity. Delays or losses of the drug during absorption may contribute to variability in the drug response, and occasionally, may result in the treatment failure. An ideal chemotherapeutic drug would have little interpatient variability in absorption and time curve (AUC) and, more importantly, little intrapatient variability with successive doses (DeMario and Ratain 1998).

In the present review, capecitabine is an interesting oral drug. Capecitabine is a selectively tumor activated fluoropyrimidine which is effective in a wide range of solid tumors, particularly in breast and colon cancer. In the two reported studies, capecitabine was active and well tolerated in the treatment of ABC elderly patients (O’Shaughnessy et al 1998; Procopio et al 2001).

Based on the available data and our experience as well, VNR seems to be one of the most active single-agent.

Bonneterre et al. conducted a dose-finding phase I study in advanced breast cancer patients. Three dose levels were evaluated on a weekly regimen basis: 60, 80 and 100 mg/m². Twenty-seven patients were enrolled in the study and the maximum tolerated dose was 100 mg/m² with a dose-limiting toxicities being neutropenia, nausea/vomiting and neuroconstipation. The recommended dose of oral VNR for further trials was defined at 80 mg/m² /week. The activity was observed at 80 and 100 mg/m² (Bonneterre et al 1996).

Following these results, the absolute bioavailability of oral VNR was determined in 24 patients receiving oral administration at 80 mg/m² or i.v. VNR at 25 mg/m² one week apart in a cross over design. The bioavailability factor calculated on blood exposure (AUC) was 43 + 14%. When data from the population pharmacokinetic analysis is taken into account (including all the patients from phase I studies) the bioavailability factor is 36 + 10% and this has formed the justification for a pragmatic use of about 40% as the basis for clinical equivalence studies. Based on these results the oral dose of 80 mg/m² was demonstrated to correspond to 30 mg/m² of the i.v. formulation and 60 mg/m² oral to 25 mg/m² i.v. (Marty et al 2001). Recently, a trial performed in elderly patients with advanced non small cell lung cancer, demonstrated that the pharmacokinetics of oral VNR is not altered in older patients who presented similar bioavalability and at least comparable inter-individual variability to younger patients (Puozzo et al 2001). Based on these data studies of oral VNR in ABC elderly patients are warranted.

B. New biological agents

The numerous molecular mechanisms implicated in the pathogenesis of breast cancer present exciting avenues for target-specific approaches to therapy. Based on the current data, the inhibitors of cell signalling seem to be the most promising class of agents for the treatment of breast cancer.

Her-2/neu, a member of the group I growth factor receptor family, is a tyrosine-kinase membrane receptor that, when activated, induces a phosphorylation cascade in cytoplasmic kinases leading to increased transcription of nuclear proteins and cellular growth. It is amplified and/or overexpressed in 20% to 30% of patients with breast cancer (Press et al 1990). Overexpression of this oncogene product is associated with increased rates of tumour growth, enhanced rates of metastasis, shorter disease-free survival, and overall survival (Slamon et al 1989; Press et al 1990; Liu et al 1995). Patients with HER-2/neu-overexpressing tumours have more aggressive and more malignant courses. HER-2/neu has been targeted by monoclonal antibodies, immunoconjugates, vaccines, antibody-directed enzyme prodrug therapy, antisense therapy and gene therapy (Hortobagyi 1990).

Trastuzumab is a humanized monoclonal antibody against the extracellular domain of HER-2/neu (Hudziak et al 1989). As a single agent, trastuzumab resulted in 15% objective response in ABC, as second-line treatment (Cobleigh et al 1999). Trastuzumab is well tolerated, low-grade fever, chills, fatigue and constitutional symptoms occur primarily with the first infusion and serious adverse effects are infrequent (Hortobagyi 1999). Trastuzumab has been showed well tolerated in elderly women (> 60 years) with HER-2-positive ABC and produces significant benefits added to chemotherapy (Fyfe et al 2001).

The EGFR (Epidermal Growth Factor Receptor), another member of the group I growth factor receptor family, is a 170 kDa transmembrane glycoprotein that consists of an extracellular domain, a hydrophobic transmembrane domain and an intracellular region containing the tyrosine kinase domain. The EGFR exists as inactive monomers, which dimerize after ligand activation. This causes homodimerization or heterodimerization between EGFR and another member of the erb receptor family. After the ligand binding, the tyrosine kinase intracellular domain of the receptor is activated, with autophosphorylation of the intracellular domain, which initiates a cascade of intracellular events. Several studies have demonstrated that EGFR-mediated signals also contribute to other processes that are crucial to cancer progression, including angiogenesis, metastatic spread, and inhibition of apoptosis (Ciardiello and Tortora 2001).

ZD1839, a synthetic anilinoquinazoline, is a p.o. active, selective reversible inhibitor of EGFR tyrosine kinase. Recently, 2 phase II trials tested ZD1839 in ABC patients. Robertson et al (2002) treated with ZD1839 500 mg, 22 patients with either ER-negative or ER-positive breast cancer that became clinical resistant to tamoxifen. The median age was 61 years (range 32-85 years). The Authors reported a partial response in 9% of patients. Albain et al, (2002) treated with an oral daily 500 mg dose of ZD1839 until disease progression, intolerable disease or consent withdrawal, 63 pretreated ABC patients (ages 34-80 years) with interesting results. The use of oral drugs as well among new biological agents is of great interest.

The treatment of elderly patients is an emerging issue. Unfortunately, among the published studies, one only is a randomised phase III trial while several phase II trials have been performed. The most part of the studies enrolled elderly patients with a cut-off age ranging from 60 to 65 years old, probably not representative of real elderly population.

Many referenced studies included in the same series patients treated as first-, second- and third-line chemotherapy confounding the reported results. Some of the cited trials used anthracycline-based chemotherapy. It is widely believed that the incidence and severity of toxic effects from anthracyclines are greater in older patients than in younger ones, and clinical experience often reinforces this belief. Anyway, Ibrahim et al confirmed the possibility to administer anthracyclines in elderly women with ABC. They performed a retrospective analysis on 1011 women over 65 years (24%) or 50-64 years, all treated with a doxorubicin-based chemotherapy. Although OR was higher for the younger patients (67% versus 51%, p = 0.001), no significant difference in terms of dose-intensity, TTP, OS and toxicity was observed (Ibrahim et al 1996). While the response to chemotherapy and clinical outcome are certainly poorer in elderly patients with chronic comorbidity than in younger and healthier patients, the evidence to date suggests that the benefits and toxic effects of chemotherapy in otherwise-healthy older patients are comparable to those in younger patients. Age is not predictive of treatment failure and chemotherapy is not necessarily less effective or less tolerable in older patients.

Based on the described studies, single-agent chemotherapy seems to determine superimposible results as compared to polichemotherapy. To date even if there is no specifically randomised study, single-agent chemotherapy probably might be considered the standard treatment for ABC in the elderly.

Great interest is for oral drugs if well tolerated and usually with a good patient’s compliance.

One of the main research-line to explore is the introduction of new biological agents in the treatment schemes. In fact, if new biological agents proved to be effective in the treatment of ABC, therapeutic strategies in the elderly could include a very useful tool, considered their excellent toxicity profile, very suitable for this peculiar patient population.

In conclusion, chemotherapy is feasible and active in ABC elderly patients resistant to hormonal therapy or with visceral metastases. Considering that the most part of these patients need to be treated with chemotherapy, large randomised phase III trials, including quality of life evaluation, are warranted.

From right Paolo Maione, Alida Barbato,Dario Nicolella, Giuseppe Airoma,Giuseppe Colantuoni, Filomena Del Gaizo, Rosa Bruno, Antonio Rossi, Carmine Ferrara.