Efficacy of lower dose capecitabine in patients with metastatic breast cancer and factors influencing therapeutic response and outcome.

Objective: Capecitabine exerts considerable therapeutic efficacy in metastatic breast cancer (MBC) patients previously treated with anthracyclines and taxanes.

Materials and Methods: In this study, the efficacy and safety of lower dose capecitabine (2000 mg/[m.sup.2]/d) in patients with anthracycline- and taxane-pretreated MBC were studied with a special emphasis on the potential predictors of time to tumor progression (TTP) and response to the capecitabine treatment.

Results: The overall response rate (ORR) was 17%. The median TTP was 5 months. Among various factors analyzed, univariate analysis showed that a performance status (PS) of 2 and the presence of visceral metastases were inversely correlated with TTP. Multivariate analysis showed that a poor PS score was associated with impaired TTP.

Conclusions: Our study indicates that lower dose capecitabine has substantial antitumor activity and a favorable safety profile in the treatment of anthracycline- and taxane-pretreated MBC. Also, only performance score was demonstrated to be a significant parameter affecting TTP.

Key Words: capecitabine, metastatic breast cancer

**********

Breast cancer is the most common solid tumor in women, and the main cause of mortality due to cancer. (1) A considerable proportion of patients with breast cancer have metastatic disease. Chemotherapy is the cornerstone of treatment in such cases, especially in patients with hormone-resistant, receptor negative and/or rapidly disseminating disease. Among various chemotherapeutic agents, anthracyclines and taxanes are extensively used in the treatment of breast cancer, either in the adjuvant or metastatic setting. First-line treatment of metastatic breast cancer (MBC) with taxane or an-thracycline monotherapy produces a response rate of 20 to 40%, and may reach up to 80% with combined use of the agents. (2) Although patients with MBC show high response rates to taxane and/or anthracycline combinations, it seems unlikely that these patients can achieve durable complete remission. Therefore, newer drugs with significant antitumoral activity and an acceptable safety profile are clearly needed.

Among various drugs used for the treatment of MBC, capecitabine has become one of the most important. The drug is converted into 5-fluorouracil (5-FU), particularly in the tumor tissue, mimicking the effect of sustained 5-FU infusion. (3,4) Thymidine phosphorylase (TP) enzyme is known to be involved in tumoral selectivity, and is found in higher levels in most solid tumors compared with normal tissues. (3) Studies have demonstrated the activity of capecitabine in MBC patients previously treated with anthracyclines and taxanes. (5-7) The response rates to capecitabine monotherapy for such patients reached 20 to 26% with a more than one year median survival time. (5,6) Capecitabine is commonly used at the daily dose of 2500 mg/[m.sup.2] for 14 days every 21 days in the treatment of MBC. However, toxicity is considerable, (5-8) and retrospective studies showed that lower doses (2000 mg/[m.sup.2]/d) of capecitabine exerted similar efficacy with a better toxicity profile compared with the commonly used dose level. (9-11)

Currently, there is no standard marker to predict the response to chemotherapy in MBC. However, growth factors and related receptors, oncogene expression and hormone receptor status could help to predict the response to chemotherapy or hormonal treatment in certain subgroups of patients. Overexpression of c-erbB-2, which is a growth factor receptor, has been demonstrated to be related to poor response to several chemotherapeutic drugs (5-FU, methotrexate, and cyclophosphamide) as well as tamoxifen. (12-16)

In the present study, the efficacy of lower dose capecitabine treatment in patients with anthracycline- and taxane-pretreated MBC was investigated and the potential factors affecting response and TTP were analyzed.

Materials and Methods

Patient Population

A retrospective review of the medical charts of patients with MBC who had been treated in the Medical Oncology Clinics of Ege University School of Medicine and Uludag University School of Medicine between December 2000 and December 2003 was performed, and 76 patients who had been treated with capecitabine treatment were identified. Subjects with insufficient data (n = 7) were excluded, leaving 69 patients for the analysis. Histologic proof of breast cancer and radiographic evidence of metastatic disease with at least one measurable or assessable lesion were required. Except for two patients who had not received any anthracycline treatment for cardiac reasons, all patients had been pretreated with anthracycline- and taxane-containing chemotherapy, and disease progression was determined after chemotherapy.

Treatment

Patients were treated with oral capecitabine (2000 mg/[m.sup.2]/d, two weeks of treatment followed by one week of rest). Capecitabine was administered orally twice daily, within 30 minutes after a meal, approximately 12 hours apart. Patients achieving a complete or partial response or stable disease after 6 weeks of therapy continued on treatment until disease progression or development of unacceptable toxicity.

Study Assessments

Before the initiation of treatment, each patient had an ECG, chest x-ray, computed tomography scan of the thorax and abdomen. Basic medical history, physical examination. Eastern Cooperative Oncology Group (ECOG) performance status and laboratory tests (hematology, blood chemistry) were also obtained. Complete blood counts and serum biochemistry were performed at the beginning of each chemotherapy cycle. All toxicities were scored according to the World Health Organization scale. The appearance of grade 3 to 4 toxicity warranted dose interruption of capecitabine until recovery. Subsequent doses were applied as 75% and 50% of the initial dose for the first and second appearance of the toxicities, respectively.

Evaluation of Response

Radiological assessments were repeated every three chemotherapy cycles. Complete response was defined as disappearance of all clinical and radiological evidence of tumor. Partial response (PR) was defined as 50% or greater decrease in the sum of the products of diameters of all measurable lesions persisting for at least one cycle of therapy. For responding patients, duration of response was defined as the interval between the first recording of the response and the time of disease progression or death. Progressive disease (PD) was defined as either an increase of [greater than or equal to]25% in the sum of the diameters of any measurable lesions, or a [greater than or equal to]50% increase in the estimated size of nonmeasurable lesions, or the appearance of any new lesion. Stable disease (SD) was defined as a tumor reduction of <50% or an increase in a tumor size of <25% lasting for at least six weeks.

Immunohistochemical Studies

All tissues were fixed in 4% buffered formalin, processed, and embedded in paraffin. From each block, 5 [micro]m thick sections were cut on coated slides and dried overnight at 37[degrees]C. The sections were deparaffinized in xylene, rehydrated through graded concentrations of ethanol to distilled water, and boiled in citrate buffer (pH = 6.0) in a microwave oven for 20 minutes. Immunohistochemical stainings were performed by using a commercial ABC Kit (Lab Vision; Ultravision Large Volume Detection System Anti-Mouse, HRP, CA) directed against mouse IgG. Blocking serum was applied for 15 minutes followed by overnight incubation with the diluted monoclonal primary antibody c-erbB-2 (Dako, code #: A0485, Denmark), estrogen receptors (Dako, IDS code #: M7047, Denmark), progesterone receptors (Dako, PgR 636 code #: M3569, CA), and p53 (Dako, clone DO-7, code #: M7001, Denmark). The sections were then incubated with the biotinylated second antibody and the peroxidase-labeled ABC for 30 minutes each. All dilutions were made in PBS (pH = 7.2), and all incubations were performed in humid chambers at room temperature. Between each step in the staining procedures (except before incubation with the antibody), the slides were rinsed three times in PBS-bound peroxidase, and were then visualized on all slides with a 3-amino-9-ethylcarbazole solution (Sigma: 0.2 mg/mL in 0.05 mol/L acetate buffer containing 0.03% perhydrol, pH = 5.0) at room temperature for 15 minutes. Finally, the sections were lightly counterstained in Mayer's hematoxylin and mounted in Surgipath Micromount (Surgipath Medical Industries, Grayslake, IL).

Cells were considered positive for p53 and hormone receptor (HR) when distinct nuclear staining was identified. For depiction of the material and phenotype analysis, a 10% cutpoint for low and high expression for p53 and HR was chosen. (17,18)

C-erbB-2 overexpression was scored in a semiquantitave fashion using a 0 to 3+ scale (0: no staining, 1+: weak membrane staining, 2+: moderate membrane staining observed in more than 10% of the tumor cells, 3+: strong circumferential membrane staining creating a fishnet pattern). Cytoplasmic staining was considered nonspecific, and was not included in the final scoring. For the tumors having moderate (2+) expression by immunohistochemical (IHC) staining, fluorescence in situ hybridization (FISH) analysis was performed with the HER-2/neu DNA Probe Kit (Vysis Inc., Downers Grove, IL). Tumors were classified as c-erbB-2 positive if the primary lesion had either strong overexpression (3+) for c-erbB-2 by immunohistochemistry, or if the tumor showed HER-2/neu gene amplification by FISH for the tumors having 2+ c-erbB-2 by IHC. (19) Tumors were considered c-erbB-2 negative if they had either 0/1 + for c-erbB-2 by IHC, or if they did not show HER-2/neu gene amplification by FISH for the tumors having 2+ c-erbB-2 by IHC.

Statistical Analysis

TTP was calculated from the date of the start of the first cycle to the date of the event occurrence. Median TTP was estimated by the Kaplan-Meier method, and differences between the groups were tested by the log-rank test. The Fisher exact test was used to investigate the differences between categorical variables. Only those variables that achieved statistical significance in the univariate analysis were subsequently evaluated in the multivariate analysis using a stepwise forward Cox regression analysis. SPSS (SPSS Inc., Chicago, IL) version 11.5 was used for the statistical analysis. For all statistical tests, a p value of [less than or equal to]0.05 was considered statistically significant.

Results

Patient Characteristics

A total of 69 patients were evaluated, and all patients were assessed for response, toxicity and median TTP. The clinicopathologic characteristics and the previous treatments given to the patients are summarized in Table 1.

Toxicity and Response

Patients received a total of 524 cycles of capecitabine with a median of 7.6 (range 1-26) cycles. Three patients demonstrated clinical and/or radiological progression after the first cycle. Capecitabine was administered as first-line therapy to three patients, second-line to 20 patients, third- or more lines to 46 patients. Twenty patients (29%) experienced grade 3 to 4 toxicity that can be listed as follows: hand-foot syndrome (6 patients), fatigue (3 patients), anorexia (2 patients), dyspepsia (2 patients), nausea (1 patient), diarrhea (1 patient), conjunctivitis (1 patient), infection (1 patient), anemia (1 patient), increased bilirubin (1 patient), and toxic hepatitis (1 patient). While there were no treatment-related deaths, four patients could not continue the treatment due to unacceptable toxicities (conjunctivitis, infection, toxic hepatitis and gastrointestinal intolerance in one patient each). The frequency of grade 3 to 4 toxicity was not found to be associated with the treatment line (P = 0.5, Fisher exact test).

Twelve patients (17%) showed PR with a median duration of 15.4 months (95% CI 11.9-18.2). Twenty-one patients (31%) had PD and 36 patients (52%) had SD. The median duration of disease stabilization was 6 months (95% CI 3.28-8.72). After a median follow-up of 18 months (range 2-26), the median TTP was calculated as 5 months (95% CI 3.2-6.8). Response evaluations were also carried out in patient subgroups that were formed by the use of the main clinical and histopathological characteristics (patients' age, PS, presence of visceral metastasis, number of metastatic sites, treatment line, HR status, c-erbB-2, and p53 expressions), which are expressed in Table 2. None of these parameters had any significant impact on the treatment response.

Survival Analysis

Parameters such as presence of visceral metastasis, metastatic site number, treatment line, patient age, PS, HR status, c-erbB-2 and p53 expression were used for the distribution of median TTP analysis for patient subgroups. In the univariate analysis, poor PS and presence of visceral metastasis were associated with worse outcome (P [less than or equal to] 0.05) (Table 3). Although the difference is not significant, HR negative and c-erbB-2 negative patients showed a trend for TTP benefit when compared with other patient subgroups (P = 0.09). The other parameters did not have any impact on TTP.

Cox regression analysis was used to examine the independent prognostic value of significant variables in univariate analysis. In multivariate analysis, poor PS was found to be associated with shorter TTP (Table 4) (Fig. 1).

Discussion

With the introduction of capecitabine, an effective and safe option for the treatment of metastatic breast carcinoma has become available. We observed that lower dose capecitabine was well tolerated and considered to be an effective treatment with an overall response rate of 17% and disease stabilization rate of 52% in patients with MBC, although it was mostly used as third or more lines of therapy. In one study, survival of patients demonstrating stable disease were found to be similar to those patients who responded to capecitabine, and were longer than those who were unresponsive to capecitabine treatment. (5) Our study supports the idea that the disease stabilization rate obtained for more than half of the patients has a clinical significance. Likewise, the median time of 5 months for TTP despite poor prognostic signs supports this finding.

[FIGURE OMITTED]

Intensive research has been carried out to determine the optimal dose of capecitabine. The main studies concerning the efficacy of capecitabine in pretreated MBC patients are summarized in Table 5. The FDA approved 2500 mg/[m.sup.2]/d of capecitabine given orally in two divided doses for 14 days, followed by 7 days of rest. However, a considerable proportion of patients who received the drug at this dose level suffered from drug-induced toxicities. These side effects may preclude the effective administration of chemotherapy, and may require a dose reduction. (5-7) Clinical studies demonstrated that a lower dose of capecitabine (2000 mg/[m.sup.2]/d) was better tolerated than the commonly administered dose of 2500 mg/[m.sup.2]/d with preserved efficacy. (9-11) A lower dose of capecitabine has been suggested to be the standard treatment approach in elderly patients with MBC because of its efficacy and lower toxicity. (11) In a recent study, the efficacy and tolerability of the approved and lowered doses of capecitabine were retrospectively compared and an initial dose level of 2000 mg/[m.sup.2]/d was found to have a more favorable therapeutic index in MBC. (9) The considerable response and disease stabilization rates with acceptable grade 3 to 4 toxicities (29%) obtained in our study support these findings. Based on the overall results, we think that capecitabine may be given at the dose of 2000 mg/[m.sup.2]/d without any detrimental effect on its efficacy in heavily treated MBC patients.

In the previous studies, some factors affecting response to chemotherapeutic agents were demonstrated. As one of the most attractive therapeutical targets in breast cancer treatment, c-erbB-2 overexpression was demonstrated to be related to poor response to certain chemotherapeutic drugs. (12,13) Although we could not identify any clinical or histopathological parameters affecting response to capecitabine in our study population, it should be noted that the number of cases were too limited to make definitive conclusions, and further trials with larger cohorts are clearly needed. Not surprisingly, our data suggest a trend to lower response rates and impaired TTP in c-erbB-2 overexpressed MBC patients treated with capecitabine alone. In such patients, it is reasonable to consider the combination of capecitabine and trastuzumab after progression on trastuzumab combinations, including taxanes and vinorelbine. (20) We believe that further comprehensive research on the subject will clarify the rational use and effectiveness of capecitabine as well as the addition of monoclonal antibodies such as trastuzumab to capecitabine in patients with heavily pretreated MBC.

In this study, we have shown that poor performance status was the most important factor affecting TTP. Similar observations have been found in previous studies. Pierga et al (21) demonstrated that performance status was the most important factor affecting prognosis of patients with MBC who were treated with capecitabine. In that study, as well as in our study, poor performance status was observed among the patients with visceral metastasis, and those heavily pretreated with plenty of chemotherapy schedules. It might be expected that these negative factors influenced the TTP of patients with poor performance status who were administered capecitabine.

In conclusion, lower-dose capecitabine has substantial antitumor efficacy with low toxicity in some subgroups of patients with MBC who were previously treated with anthra-cyclines and taxanes. In addition, prospective trials with larger cohorts are needed to examine the potential role of certain biologic factors, especially c-erbB-2, on the efficacy of capecitabine in heavily pretreated patients with metastatic breast cancer.

Acknowledgments

The authors thank Timur Kose for his contributions to the statistical analysis of the study and Levent Okyay for editing the manuscript.

References

1. Parkin DM, Pisani P, Ferlay J. Global cancer statistics. CA Cancer J Clin 1999;49:33-64.

2. Harris J, Morrow M, Norton L. Malign tumors of the breast. In: DeVita Jr, VT Hellman Rosenberg SA, eds. Cancer: Principles and Practices of Oncology. 5th edition. Philadelphia, PA, Lippincott-Raven, 1997, pp 1557-1606.

3. Miwa M, Ura M, Nishida M, et al. Design of a novel oral fluoropyrimidine carbamate, capecitabine, which generates 5-fluorouracil selectively in tumours by enzymes concentrated in human liver and cancer tissue. Eur J Cancer 1998;34:1274-1281.

4. Schuller J, Cassidy J, Dumont E, et al. Preferential activation of capecitabine in tumor following oral administration in colorectal cancer patients. Cancer Chemother Pharmacol 2000;45:291-297.

5. Blum JL, Jones SE, Buzdar AU, et al. Multicenter phase II study of capecitabine in paclitaxel-refractory metastatic breast cancer. J Clin Oncol 1999;17:485-493.

6. Blum JL, Dieras V, Lo Russo PM, et al. Multicenter, phase II study of capecitabine in taxane-pretreated metastatic breast carcinoma patients. Cancer 2001;92:1759-1768.

7. Reichardt P, von Minckwitz G, Thuss-Patience PC, et al. Multicenter phase II study of oral capecitabine (Xeloda) in patients with metastatic breast cancer relapsing after treatment with a taxane-containing therapy. Ann Oncol 2003;14:1227-1233.

8. Fumoleau P, Largillier R, Clippe C, et al. Multicentre, phase II study evaluating capecitabine monotherapy in patients with anthracycline- and taxane-pretreated metastatic breast cancer. Eur J Cancer 2004;40:536-542.

9. Hennessy BT, Gauthier AM, Michaud LB, et al. Lower dose capecitabine has a more favorable therapeutic index in metastatic breast cancer: retrospective analysis of patients treated at M. D. Anderson Cancer Center and a review of capecitabine toxicity in the literature. Ann Oncol 2005;16:1289-1296.

10. O'Shaughnessy J, Blum J. A retrospective evaluation of the impact of dose reduction in patients treated with Xeloda (capecitabine). Proc Am Soc Clin Oncol 2000;19:400.

11. Bajetta E, Procopio G, Celio G, et al. Safety and efficacy of two different doses of capecitabine in the treatment of advanced breast cancer in older women. J Clin Oncol 2005;23:2155-2161.

12. Muss HB, Thor AD, Berry DA, et al. C-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266.

13. Berns EM, Foekens J A, van Staveren IL, et al. Oncogene amplification and prognosis in breast cancer: relationship with systemic treatment. Gene 1995;159:11-18.

14. Benz C, Scott GK, Sarup JC, et al. Estrogen-dependent, tamoxifenresistant tumorigenic growth of MCF-7 cells transfected with Her2/neu. Breast Cancer Res Treat 1992;24:85-95.

15. Sunderland MC, McGuire WL. Oncogenes as clinical prognostic indicators. Cancer Treat Res 1991;53:3-22.

16. Borg A, Baldetorp B, Ferno M, et al. ERBB2 amplification is associated with tamoxifen resistance in steroid-receptor positive breast cancer. Cancer Lett 1994;81:137-144.

17. Sjostrom J, Blomqvist C, Heikkila P, et al. Predictive value of p53, mdm-2, p21, and mib-1 for chemotherapy response in advanced breast cancer. Clin Cancer Res 2000;6:3103-3110.

18. Middleton LP, Palacios DM, Bryant BR, et al. Pleomorphic lobular carcinoma: morphology, immunohistochemistry, and molecular analysis. Am J Surg Pathol 2000;24:1650-1656.

19. Ridolfi RL, Jamehdor MR, Arber JM. HER-2/neu testing in breast carcinoma: a combined immunohistochemical and fluorescence in situ hybridization approach. Mod Pathol 2000;13:866-873.

20. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER-2. N Engl J Med 2001;344:783-792.

21. Pierga JY, Fumoleau P, Brewer Y, et al. Efficacy and safety of single agent capecitabine in pretreated metastatic breast cancer patients from the French Compassionate Use Program. Breast Cancer Res Treat 2004;88:117-129.

Canfeza Sezgin, MD, Ender Kurt, Turkan Evrensel, Necomettin Ozdemir, Osman Manavoglu, and Erdem Goker

From the Division of Medical Oncology and the Department of Pathology, Ege University School of Medicine, Izmir, Turkey, and the Division of Medical Oncology, Uludag University School of Medicine, Bursa, Turkey.

Reprint requests to Canfeza Sezgin, MD, Ege University School of Medicine, Department of Internal Medicine, 5th Floor, Division of Medical Oncology, 35100 Bornova, Izmir, Turkey. Email: [e-mail]

The first two authors took an equal share in the realization and drafting of the manuscript.

Accepted June 23, 2006.

RELATED ARTICLE: Key Points

* The therapy of metastatic breast cancer is palliative and systemic treatment options mainly consist of chemotherapy, hormonal therapy, as well as the targeted therapies.

* Capecitabine is active against anthracycline- and taxane-pretreated metastatic breast cancer. The use of capecitabine at the FDA-approved dose (2500 mg/[m.sup.2]/day) leads to unacceptable toxicity in many patients. Anecdotal reports have suggested that dose reductions improve tolerability without compromising efficacy.

* Factors influencing therapeutic response to capecitabine have not been clearly demonstrated.

* Our study showed that a lower starting dose (2000 mg/[m.sup.2]/day) of capecitabine improves tolerability without compromising efficacy in pretreated metastatic breast cancer. However, a phase III, randomized, multicenter study to establish the safety and efficacy of different doses of capecitabine is needed.

Table 1. Patient and tumor characteristics

Characteristics                                   n (%)

Number of patients                                69 (100)
Median age                                        52 (range: 29-76)
Performance status (ECOG)
  0                                                6 (8)
  1                                               50 (73)
  2                                               13 (19)
Hormone receptor status
  Positive                                        34 (49)
  Negative                                        30 (44)
  Unknown                                          5 (7)
c-erbB-2
  Positive
    3+                                            23 (34)
    IHC 2+, FISH+                                  5 (7)
  Negative
    IHC 0/1+                                      27 (39)
    IHC 2+, FISH-                                 10 (14)
  Unknown                                          4 (6)
p53
  High expression ([greater than or equal to]10)   8 (12)
  Low expression (<10)                            41 (59)
  Unknown                                         20 (29)
Visceral metastasis
  Present                                         52 (75)
  Absent                                          17 (25)
Prior adjuvant treatments
  Chemotherapy                                    52 (75)
  Hormonotherapy                                  31 (45)
  Radiotherapy                                    34 (49)

ECOG, Eastern Cooperative Oncology Group; IHC, immunohistochemistry;
FISH, fluorescence in situ hybridization.

Table 2. Correlation of treatment response with clinicopathological
features

                                                 Responded
                                                 n (%)      P (a)

Patient age                                                 0.61
  [less than or equal to]50 (n = 29)              5 (17)
  >50 (n = 40)                                    7 (18)
Performance status                                          1.00
  0-1 (n = 56)                                   10 (18)
  2(n = 13)                                       2 (15)
Hormone receptor                                            0.56
  Positive (n = 34)                               6 (18)
  Negative (n = 30)                               6 (20)
c-erbB-2                                                    0.14
  Positive (n = 28)                               3 (11)
  Negative (n = 37)                               9 (24)
p53                                                         0.84
  High expression, [greater than or equal to]10   1 (13)
    (n = 8)
  Low expressions, <10 (n = 41)                   8 (20)
Visceral metastasis                                         0.47
  Present (n = 52)                                8 (15)
  Absent (n = 17)                                 4 (24)
Metastatic site number                                      0.22
  [less than or equal to]2 (n = 56)               8 (14)
  [greater than or equal to]3 (n = 13)            4 (31)
Capecitabine treatment                                      0.19
  [less than or equal to]2 line (n = 23)          6 (26)
  [greater than or equal to]3 line (n = 46)       6 (13)

(a) Fisher exact test
PR, partial response; SD, stable disease; PD, progressive disease.

Table 3. Results of univariate analysis of TTP

                                     TTP (months)  CI 95%      P (a)

Patient age                                                    0.73
  [less than or equal to]50           5            [2.7-7.3]
  >50                                 5            [2.6-7.4]
Performance status                                             0.0004
  0-1                                 6            [3.0-9.0]
  2                                   2            [0.9-3.1]
Hormone receptor                                               0.09
  Positive                            4            [2.6-5.4]
  Negative                            6            [2.8-9.2]
c-erbB-2                                                       0.09
  Positive                            4            [2.7-5.3]
  Negative                            8            [3.5-12.5]
p53                                                            0.4
  [greater than or equal to]10        8            [5.2-10.8]
  <10                                 4            [1.5-6.5]
Visceral metastasis                                            0.05
  Present                             4            [2.3-5.7]
  Absent                             10            [4.4-15.6]
Metastatic site number                                         0.3
  [less than or equal to]2            5            [3.4-6.6]
  [greater than or equal to]3         8            [2.3-13.7]
Capecitabine treatment                                         0.28
  [less than or equal to]2 lines      6            [3.4-8.6]
  [greater than or equal to]3 lines   4            [1.5-6.5]

(a) Log-rank test
TTP, time to tumor progression; CI, confidence interval.

Table 4. Results of Cox regression analysis

Variable                 Hazard ratio  95% CI     P value

Poor performance status  2.90          1.49-5.60  0.004

CI, confidence interval.

Table 5. Summary of studies concerning the efficacy and toxicity of
capecitabine in pretreated metastatic breast cancer

                     No. of    Dose            ORR  Stable
                     patients  mg/[m.sup.2]/d  (%)  disease (%)

Blum et al (5)       162       2500            20   40
Blum et al (6)        74       2510            26   31
Reichardt et al (7)  136       2500            15   46
Fumoleau et al (8)   126       2500            28   35
Hennessy et al (9)    51       2500            18   35
Hennessy et al (9)    45       2000            24   37
Bajetta et al (11)    30       2500            37   33
Bajetta et al (11)    43       2000            35   46

                     Median        Grade 3-4     Dose
                     TTP (months)  toxicity (%)  reduction (%)

Blum et al (5)       3             57            NA
Blum et al (6)       3.2           50            50
Reichardt et al (7)  3.5           30            39
Fumoleau et al (8)   4.9           45            37
Hennessy et al (9)   2.8           67            41
Hennessy et al (9)   3.5           43            28
Bajetta et al (11)   4             40            30
Bajetta et al (11)   4             30             5

NA, not available; ORR, overall response rate; TTP, time to tumor
progression.