capecitabine monotherapy in patients with anthracycline-and taxane-pretreated metastatic breast...

IntroductionSystemic cytotoxic chemotherapy is the treatment

of choice in breast cancer patients who have hor-mone-refractory, hormone-receptor-negative, orrapidly growing visceral diseases (1,2). The mostactive cytotoxic agents in metastatic breast cancer

Capecitabine Monotherapy in Patients with Anthracycline-and Taxane-Pretreated Metastatic Breast Cancer

S.-H. Lee, J. Lee, J. Park, S. H. Park, K.-E. Lee, S. I. Lee, E. Nam, J. O. Park, K. Kim,C. W. Jung, Y. S. Park, S. S. Yoon, W. K. Kang, M. H. Lee, K. Park, and Y.-H. Im

Division of Hematology and Oncology, Department of Medicine, Samsung Medical Center,Sungkyunkwan University School of Medicine, Seoul, Korea

Abstract

The selection of chemotherapeutic regimens is challenging for metastatic breast cancer (MBC) patientswhose diseases have failed to respond to anthracyline and taxane. Capecitabine has advantages of oral admin-istration and favorable toxicity profiles. This study was conducted to evaluate the efficacy of capecitabine andto identify the subgroup of patients who would potentially have benefit from capecitabine monotherapy inpatients with anthracycline- and taxane-pretreated MBC. Female patients with MBC who had been previouslytreated with anthracycline and taxane received oral capecitabine 2500 mg/m2 divided in two doses daily for 2wk with 1-wk rest period. Between September, 1999, and December, 2002, a total of 38 patients were enrolled.Among the 36 evaluable patients, one patient achieved a complete response (CR), 9 patients had partialresponses (PRs), and 13 patients had stable diseases (SDs). Response rate was 26% [95% confidence interval(CI), 12–40%] and the tumor control rate (TCR, CR+PR+SD) was 61% (95% CI, 45–77%). The median fol-low-up duration was 27.8 mo. The median duration of response was 8.9 mo, the median time to progressionwas 4.6 mo, and the median overall survival was 18.1 mo. The major toxicities were hand–foot syndrome,diarrhea, and emesis. There was no treatment-related death. The predictors of better overall survival were pos-itivity of hormone receptor, disease-free survival longer than 1 yr, non-refractoriness to anthracycline, andfewer number (≤ 3) of involved organs. Capecitabine monotherapy is effective and well tolerated for MBCpatients who had previously been treated with anthracycline and taxane. The TCR could predict overall sur-vival as well as the objective response in this study, suggesting a possible role of TCR as a surrogate markerfor survival in MBC patients on salvage chemotherapy. The patients who have relatively slow growing tumorand less tumor burden could have benefit from capecitabine monotherapy following anthracycline- and tax-ane-based chemotherapy.

Key Words: Breast carcinoma; chemotherapy; capecitabine; monotherapy; tumor control rate.

223

Original Article

Received 02/05/04; Accepted 02/10/04.Young-Hyuck Im, MD, Division of Hematology and

Oncology, Department of Medicine, Samsung Medical Center,Sungkyunkwan University School of Medicine, 50 Ilwon-dong,Kangnam-gu, Seoul, 135-710, Korea. E-mail: [email protected].

Medical Oncology, vol. 21, no. 3, 223–231, 2004© Copyright 2004 by Humana Press Inc. All rights of any nature whatsoever reserved. 1357-0560/04/21:223–231/$25.00

(MBC) have been known to be the anthracyclinesand taxanes, whose response rate is 20–40% whenused as a monotherapy and 70–80% in combinationregimens (1). Although there are no definite datasupporting the superiority of any particular regimenas the first-line chemotherapy (3), some randomizedtrials of first-line regimens in MBC demonstratedthat the first-line anthracycline-containing regimenconferred a marginal survival benefit compared withnon-anthracycline regimens (4). Therefore, anthra-cycline was considered the standard first-line agentin the treatment of MBC between 1970s and mid-1990s. However, since 1990s, taxanes have beenshown to improve overall survival of patients withMBC in well-designed clinical trials (5). Single-agent taxanes showed at least equivalent efficacycompared with anthracycline, and the integration oftaxanes either alone or in combination with otheractive agents resulted in improved overall survivalcompared with regimens lacking taxanes (5,6).

The clinicians are frequently faced with treatingpatients with MBC that is resistant to these highlyactive agents, because there is a tendency to useanthracyclines and taxanes earlier in the course ofdisease. However, MBC is basically an incurabledisease, and nearly all patients who have beentreated with anthracycline and taxane eventuallyprogress. In this context, the aims of chemotherapyare amelioration of tumor-related symptoms,improvement of quality of life, and finally prolonga-tion of survival. The most important prerequisite ofan ideal cytotoxic chemotherapy in this setting is areasonable efficacy with minimal toxicity of the reg-imen. Several agents, including vinorelbine, gemc-itabine, and capecitabine, have been evaluated in thissetting, either alone or in combination with othercytotoxic agents (7,8). When balancing the benefitand toxicity, there is no standard regimen for anthra-cycline- and taxane-pretreated patients (7,8).

Capecitabine (Xeloda®; Hoffmann–La Roche,Nutley, NJ), and oral fluoropyrimidine carbamate,was designed to generate 5-fluorouracil preferentiallyin tumor tissue and to mimic continuous infusion of 5-fluorouracil. It can be administered orally in outpa-tient clinic, and has favorable toxicity profiles. Inthree multicenter phase II clinical trials, capecitabinewas proven to be active and safe regimen for taxane-pretreated breast cancer patients (9–11). Capecitabine

produced objective response rates of 15–26% andmedian overall survivals of 10.1–12.2 mo with afavorable safety profile in these phase II trials. Thisstudy was designed to evaluate the efficacy ofcapecitabine in patients with both anthracycline- andtaxane-pretreated MBC and to identify the subgroupof patients who would potentially have benefit fromcapecitabine monotherapy in this setting.

Materials and MethodsThis study was an open-label, single center, phase

II trial of capecitabine in MBC patients who hadbeen previously treated with anthracycline and tax-anes. Patients with MBC were enrolled at SamsungMedical Center between September, 1999, andDecember, 2002. The primary end point was todetermine the response rate of capecitabine in thetreatment of patients with MBC. Secondary endpoints were to evaluate the duration of response,time to progression, overall survival, and the toxicityof capecitabine, and to identify the subgroup ofpatients who would potentially benefit fromcapecitabine monotherapy in this setting. Informedconsent was obtained from all patients before enroll-ment in this study.

Patient EligibilityWomen with histologically confirmed MBC were

eligible for this study. All patients were required to beat least 18 yr old, have performance status (PS) of 0 to2 on an Eastern Cooperative Oncology Group(ECOG) scale, and have a life expectancy at least 3mo. They all had at least one measurable lesion bidi-mensionally, the lesion capable of two-dimensionalmeasurement by computed tomography scan, chestX-ray, or physical examination above 2 cm in largestdiameter. Eligible patients must have received priortherapy with anthracycline- and taxane(s)-containingregimen. Refractoriness to anthracycline or taxanewas defined as disease recurrence within 6 mo ofcompleting adjuvant therapy, objective response fol-lowed by progression within 6 wk of the last dose, orprogression while receiving therapy without improve-ment. Adequate hematologic function [absolute neu-trophil count (ANC) ≥ 1,500/mm3, platelet count ≥100,000/mm3], hepatic function [asparate aminotrans-ferase/alanine aminotransferase (AST/ALT) ≤ 3.0

224 Lee et al.

Medical Oncology Volume 21, 2004

times the upper normal limit (UNL), bilirubin ≤ 1.25times the UNL], and renal function (serum creatinine≤ 1.5 times the UNL) were required. Patients who hadpreviously experienced a severe and unexpected reac-tion to fluoropyrimidine therapy or who had knownhypersensitivity to 5-FU were excluded. Prior historyof another malignancy within 5 yr of study entry, apartfrom basal cell carcinoma of the skin or carcinoma insitu of the uterine cervix, precluded participation inthe current trial. Patients with clinically significantcardiac disease as defined by symptomatic ventriculararrhythmias, history of congestive heart failure, or his-tory of previous myocardial infarction within 12 moof study entry were also excluded.

Treatment ScheduleCapecitabine was administered at a dose of 1250

mg/m2 orally twice daily for 14 d followed by 1-wkrest period, repeated every 3 wk. Pyridoxine of 100mg was administered daily during the period ofcapecitabine for 14 d to ameliorate the hand–footsyndrome. Patients were discontinued from studywhen there was evidence of disease progression,unacceptable toxicity, request of the patients for dis-continuation, or the decision for withdrawal by theinvestigators.

Dose adjustment criteria for capecitabine werebased on the toxicities evaluated on d 1. Toxicity wasevaluated before each treatment cycle according toNational Cancer Institute Common Toxicity Criteria(NCI-CTC) version 2.0 (12). Chemotherapy waswithheld if the ANC was < 1000/mm3 or the plateletcount was < 75,000/mm3 on d 1. In this case, the com-plete blood count was repeated at least weekly andchemotherapy was restarted as soon as ANC reached≥ 1000/mm3 and platelet count ≥ 75,000/mm3. If theANC was 1000–1499/mm3 and/or the platelet countwas 75,000–99,000/mm3 on d 1, the dose was reducedby 25%. The drug was modified according to non-hematologic toxicities. Treatment was continued atthe same dose if toxicities were considered by theinvestigators as unlikely to become serious or life-threatening. Treatment was interrupted and/or reducedin dose from the next cycle if the patients experiencedany other adverse event. At the first occurrence ofgrade 2 toxicity, treatment was interrupted and thenresumed after resolution to grade 0 or 1 at the samedose, and the original dose was restarted at the next

cycle. At the first occurrence of grade 3 or the subse-quent event of grade 2 toxicity, treatment was inter-rupted until resolved to grade 0 or 1 and followed bya 25% dose reduction, and 25% of original dose wasreduced at the next cycle. At the second appearance ofa given grade 3 or the first event of a given grade 4toxicity, treatment was interrupted until resolved tograde 0 or 1 and followed by a 50% dose reduction,and 50% of original dose was administered at the nextcycle. At the third occurrence of a given grade 3 toxi-city or the second appearance of a given grade 4 toxi-city, the treatment was discontinued and the patientwas withdrawn from study.

Pretreatment and Follow-up EvaluationAll patients underwent baseline assessment

including a complete medical history and physicalexamination, complete blood cell count (CBC),chemistry, chest X-ray, and proper imaging studies(computed tomography, bone scan, and ultrasonog-raphy). Complete medical history and physicalexamination, CBC, chemistry, and chest X-ray (ifmeasurable lesion was present) were performed on d1 of every cycle. Computed tomography scans wereperformed to evaluate the treatment response everythree cycles. The compliance for capecitabine wasmonitored by questioning patients and counting theirremaining pills at each outpatient visit.

Efficacy and Toxicity EvaluationResponse was assessed using standard WHO cri-

teria. Complete response (CR) was defined as thecomplete disappearance of all clinically detectabledisease for at least 4 wk. Partial response (PR) wasdefined as a more than 50% decrease in the sum ofthe products of the two longest perpendicular diam-eters of all measurable lesions for at least 4 wk withno increase in size of any area of known malignantdisease and no appearance of new areas of malignantdisease. Progressive disease (PD) was defined as agreater than 25% increase in the sum of the productsof the perpendicular diameters of all measurablelesions or the appearance of any new lesion. Allother outcomes were scored as stable disease (SD).

Response rate was calculated as the ratio of numberof patients who achieved CR or PR to number ofintent-to-treat (ITT) patients, and tumor control rate(TCR) was calculated as the ratio of number of

Capecitabine for Anthracycline- and Taxane-Pretreated MBC 225


patients who achieved CR, PR, or SD to number ofITT patients. Duration of response was calculatedfrom the date of its initial documentation to the date onwhich progressive disease was first observed or of thelast follow-up, with the group of responding patients.Time to progression (TTP) was calculated from thefirst day of treatment to the date on which progressivedisease was first observed or of the last follow-up.Overall survival (OS) was calculated from the first dayof treatment to the date of death or last follow-up.

Toxicity was assessed according to NCI-CTC ver-sion 2.0 (12). The severity of any toxicities notdefined in the NCI-CTC were graded as 1=mild,2=moderate, 3=severe, or 4=very severe.

Statistical AnalysisDescriptive statistics are reported as proportions

and medians. Kaplan–Meier estimates were used inthe analysis of all time-to-event variables, and the95% confidence interval (CI) for the median time toevent was computed. The dose intensity (DI) wascalculated as the ratio of the total dose (expressed inmilligrams) per meter squared of the patient, dividedby the total treatment duration expressed in days. Inthis calculation, the end of treatment was consideredto be 21 d after d 1 of the last cycle of chemotherapy.The relative DI was calculated as the ratio of the DIactually delivered to the DI planned by the protocol.

A Cox proportional hazards regression model wasused to assess the effect of each potential prognosticvariable on TTP and OS. All potential prognosticvariables were included in multivariate Cox propor-tional hazards regression model. Variables were thenremoved from this model one at a time in a backwardselection process using the likelihood ratio test andsignificance level of 0.05. We calculated the HR andits associated 95% CI for each variable that remainedin the model.

According to Simon’s two-stage optimal design,a sample size of 34 was required to accept thehypothesis that the true response rate is greater than35% with a 80% power and to reject the hypothesisthat the response rate is less than 15% with a 5%significance. At the first stage, if there were noresponses out of the initial nine patients, the studywould terminate.

SPSS for Windows (SPSS Inc., Chicago, IL)packages were used for statistical analysis.

ResultsPatient Characteristics

From September, 1999, to December, 2002, 38patients were enrolled. The clinical characteristics ofthe enrolled patients are provided in Table 1. Themedian age was 48 yr (range, 31–66). The averagenumber of involved organs was 2.5 (range, 1–5). Allpatients were previously treated with taxane as pallia-tive aims and anthracycline therapy as adjuvant and/orpalliative aims. Eleven patients (29%) had anthracy-cline- and taxane-refractory disease, and the diseaseof fifteen patients (39%) was refractory to taxane butnot refractory to anthracycline. Twenty-six patients(68%) had received ≥ 3 previous chemotherapeuticregimens, and most patients (95%) had received ≥ 5cytotoxic agents. Nineteen patients (50%) had beenpretreated with both paclitaxel and docetaxel. Thirty-

226 Lee et al.


Table 1Patient Characteristics

Characteristics No.

AgeMedian (range, yr) 48 (31–66)< 35 yr 5 (13%)

Performance statusECOG 0–1 21 (55%)ECOG 2 17 (45%)

Predominant site of metastatic diseaseVisceral 22 (58%)Soft tissue 13 (34%)Bone 3 (8%)

Number of involved organs (mean: 2.5)1–3 31 (82%)4–5 7 (18%)

Disease-free survival0–1 mo 4 (11%)1–12 mo 5 (13%)Longer than 12 mo 29 (76%)

Use of anthracyclineAdjuvant only 20 (53%)Palliative only 15 (39%)Both 3 (8%)

Average number of prior systemic therapies (range)Chemotherapeutic regimens 2.9 (2–5)Chemotherapeutic agents 6.1 (3–9)Taxanes 1.5 (1–2)Hormonal therapies 0.5 (1–3)

four patients (89%) had undergone surgery with cura-tive intent, with median disease-free survival (DFS) of2.0 yr (range, 0.4–10.5 yr). Thirteen patients (34%)had received prior hormonal therapy.

Administration of ChemotherapyA total of 218 cycles were administered, and the

patients received a median of six cycles (range, 1–15cycles). The average of relative dose intensity was0.85 (range, 0.55–1.0). Dose reduction was done in61 cycles (28%) and cycle delay was done in 27cycles (12%).

EfficacyAmong 38 patients, 36 were evaluable for

response after excluding two patients: one droppedout due to patient’s refusal after one cycle and onelost to follow-up after two cycles. CR was observedin 1 patient (3%), 9 patients (24%) showed PRs, and13 patients (34%) showed SDs (Table 2). The objec-tive response rate was 26% (95% CI, 12–40%), andthe tumor control rate (TCR, CR+PR+SD) was 61%(95% CI, 45–77%). The median duration of responsewas 8.9 mo (95% CI, 6.5–11.4 mo) for all respon-ders. The median follow-up period was 27.8 mo(range, 7.1–46.7 mo), the TTP was 4.6 mo (95% CI,2.4–6.8 mo), and the median OS was 18.1 mo (95%CI, 5.6–30.6 mo) for the ITT population (Fig. 1).

ToxicityAmong the 218 cycles of chemotherapy, hemato-

logic toxicities were minimal (Table 3). No patientexperienced a neutropenic fever event. There was nohospitalized patient for the treatment of infection orother toxicities. The non-hematologic toxicitiesgreater than grade 2 were hand–foot syndrome,

stomatitis, and diarrhea (Table 4). Grade 4 non-hematologic toxicity was not observed. There wasno discontinuation due to toxicity of chemotherapyand treatment-related mortality.



Table 2Responses to Treatment

Response No. %

Tumor control 23 61Complete response 1 3Partial response 9 24Stable disease 13 34

Progressive disease 13 34Not evaluable 2 5

Table 3Hematologic Toxicity

Patient No. (%)

Hematologic toxicity Grade 2 Grade 3 Grade 4

Leukopenia 5 (13) 1 (3) —Granulocytopenia 10 (26) 2 (5) 2 (5)Anemia 8 (21) — —Thrombocytopenia 3 (8) — —

Fig. 1. Time to progression (TTP) and overall survival(OS) curves.

Table 4Non-hematologic Toxicitya

Patient No. (%)

Toxicity Grade 1 Grade 2 Grade 3

Hand–foot syndrome 6 (16) 8 (21) 2 (5)Stomatitis 2 (5) 2 (5) 1 (3)Nausea 3 (8) 7 (18) —Vomiting 4 (11) 3 (8) —Diarrhea 3 (8) — 1 (3)

a Grade 4 toxicity was not observed.

Prognostic Factor Analysis

Various factors were analyzed as shown in Table5. The performance status, pathology, refractorinessto taxane, and the number of involved organs weresignificantly correlated with tumor control rate.Disease-free survival, refractoriness to anthracy-cline, and the number of involved organs were inde-pendent prognostic factors for TTP. Status ofhormone receptor, disease-free survival, refractori-ness to anthracycline, and the number of involvedorgans were independent prognostic factors for OS(Table 6).

DiscussionThe anthracyclines and taxanes are the most

active agents in the treatment of MBC, and thesedrugs are commonly used in the initial treatment ofMBC patients. Clinicians are frequently faced withMBC that is resistant to these highly active agents,but the optimal treatment regimen and the method ofcombination (concurrent versus sequential) is notdetermined for these patients. Considering that theaims of chemotherapy for MBC are the ameliorationof tumor-related symptoms, the improvement ofquality of life, and the prolongation of survival, an

228 Lee et al.


Table 5Potential Prognostic Variables

Demographic Pathologic Clinicala

Age Pathology DFS (≥ 1 year versus < 1 year)Performance status Hormone receptor Refractoriness to anthracycline

c-erbB2 Refractoriness to taxanep53 Number of regimens used

Number of agents usedNumber of taxanes usedResponse to the previous regimenHistory of 5-FU useInvolved organs (≤ 3 versus > 3)Presence of visceral metastasis

a 5-FU, 5-fluorouracil; DFS, disease-free survival.

Table 6Prognostic Factors for Overall Survival

Univariate Multivariate

Median OS Prognostic factors (mo) p value p value

Status of hormone receptors Positive NR 0.041 0.015Negative 10.7

Disease-free survival ≥ 1 year 23.4 0.027 0.018< 1 year 11.9

Refractoriness to anthracycline Not refractory 22.2 0.020 0.022Refractory 5.3

Number of involved organs 1–3 23.4 0.004 0.0014–5 8.4

NR, not reached; OS, overall survival

ideal cytotoxic chemotherapy in this setting shouldhave a reasonable efficacy with minimal toxicity. Inthis setting, most commonly used chemotherapeuticregimens are single agents such as vinorelbine, gem-citabine, and capecitabine. Currently, no prospectiverandomized studies have compared any of theseagents with one another.

The present study showed the results ofcapecitabine monotherapy following progression ofdisease after anthracycline- and taxane-basedchemotherapy for MBC patients. The response rateof 26% in this study was similar to those reported inother studies of capecitabine in anthracycline- andtaxane-pretreated breast carcinoma patients (9–11).Blum et al. reported a response rate of 20% in tax-ane-refractory breast carcinoma patients (9) and thatof 26% in taxane-pretreated MBC patients (10).More than 80% of patients in two studies weretreated previously with an anthracycline-containingregimen (9,10). Reichardt et al. (11) reported aresponse rate of 15%. The response rate of 15–30%would be reliably reproducible in the salvagechemotherapy with capecitabine in anthracycline-and taxane-pretreated MBC patients. In addition, adurable duration of response of 8.9 mo in this studywas also reproduced and comparable to thosereported in the previous studies (8.0 mo, 8.3 mo, and7.5 mo, respectively) (9–11). Compared to otherstudies with various monotherapy agents in anthra-cycline- and taxane-pretreated MBC patients, theresult of this study are promising. No objectiveresponse was obtained with vinorelbine alone instandard dose (13), and 25% response rate with high-dose vinorelbine with growth factor support (14) orweekly administration of vinorelbine (15). However,a considerable incidence of grade 34 neutropeniaand neuropathy was reported in the vinorelbine stud-ies (14,15). In the studies of gemcitabine monother-apy as a salvage regimen, in which accurate profilesof previous regimens were unknown, the objectiveresponse was obtained between 0% and 30%(16–18). Because no standard regimen has beenestablished in the patients previously treated withanthracycline and taxane, a future study comparingcapecitabine with other monotherapy is worthwhileinvestigating. An additional important finding in thisstudy was the tumor control rate (as defined by CR +PR + SD), which was 61%. The predictability of

TCR for overall survival was comparable to that ofthe objective response rate as shown in Fig. 2, sug-gesting a possible role of TCR as a surrogate markerfor survival in MBC patients on the third-line sal-vage chemotherapy.

As mentioned above, a third-line chemotherapyshould be considered in the aspects of both efficacyand the toxicity. This regimen showed very favorabletoxicity profiles. No neutropenic fever, no grade 4non-hematologic toxicities, and no treatment-relatedmortality were observed. Oral application, whichenables treatment on an outpatient basis, was anotherimportant advantage of capecitabine monotherapy,



Fig. 2. (A) Overall survival curves according toresponse. (B) Overall survival curves according to tumorcontrol.

considering the quality of life and the preference oforal chemotherapeutic agents (19).

Another objective of this study was to identify thesubgroup of patients who would achieve relativelylong survival after capecitabine monotherapy. It isextremely valuable that we could find out the subgroupof patients with MBC who could achieve benefit fromcapecitabine monotherapy with minimal toxicity. Inpresent study, the multivariate analysis for OS showedthat a positivity of hormone receptor, non-refractori-ness to anthracycline, DFS longer than 1 yr, and thefewer number (≤ 3) of involved organs were favorableprognostic factors. The median OS of the patients whohad 0–1 risk factor was 28.9 mo and that of the patientswho had 3–4 risk factors was 5.2 mo (p = < 0.0001,Fig. 3). Patients with good prognostic factors (risk fac-tor 0–1) could achieve relatively long survival aftermonotherapy of capecitabine following anthracyclineand taxane chemotherapy. The patients who have thelonger survival have indicators of relatively slow grow-ing disease or less tumor burden. Whether their longersurvival is the mere reflection of disease biology or dueto the benefit from capecitabine treatment is unclear.Anyway, these patients could achieve the ultimate goalfrom capecitabine monotherapy without unnecessarytoxicity of combination chemotherapy.

In conclusion, capecitabine monotherapy follow-ing anthracycline- and taxane-based treatment iseffective and well tolerated for MBC patients. TheTCR could predict overall survival as well as theobjective response in this study, suggesting a possi-ble role of TCR as a surrogate marker for survival inMBC patients on the third-line salvage chemother-apy. The patients who have prognostic factors suchas positive hormone receptor, non-refractoriness toanthracycline, disease-free survival longer than 1 yr,and the fewer number (≤ 3) of involved organs couldhave benefit from capecitabine monotherapy follow-ing anthracycline- and taxane-based chemotherapy.

References1. Winer EP, Morrow M, Osborne CK, Harris JR. Malignant

tumors of the breast. In: DeVita VT Jr., Hellman S,Rosenberg SA (eds). Cancer: Principles and Practice ofOncology. 6th ed. Lippincott Williams and Wilkins:Philadelphia, 2001, pp. 1651–1717.

2. Norton L. Salvage chemotherapy of breast cancer. SeminOncol 1994; 21:19–24.

3. Fossati R, et al. Cytotoxic and hormonal treatment formetastatic breast cancer: a systematic review of publishedrandomized trials involving 31,510 women. J Clin Oncol1998; 16:3439–3460.

4. A’Hern RP, Smith IE, Ebbs SR. Chemotherapy and survivalin advanced breast cancer: the inclusion of doxorubicin inCooper type regimens. Br J Cancer 1993; 67:801–805.

5. Nabholtz JM, et al. Prospective randomized trial of doc-etaxel versus mitomycin plus vinblastine in patients withmetastatic breast cancer progressing despite previousanthracycline-containing chemotherapy: 304 Study Group.J Clin Oncol 1999; 17:1413–1424.

6. Chan S, et al. Prospective randomized trial of docetaxel ver-sus doxorubicin in patients with metastatic breast cancer.The 303 Study Group. J Clin Oncol 1999, 17:2341–2354.

7. Seidman AD. Monotherapy options in the management ofmetastatic breast cancer. Semin Oncol 2003; 30:6–10.

8. Crown J, et al. Chemotherapy for metastatic breast cancer-report of a European expert panel. Lancet Oncol 2002;3:719–727.

9. Blum JL, et al. Multicenter phase II study of capecitabine inpaclitaxel-refractory metastatic breast cancer. J Clin Oncol1999; 17:485–493.

10. Blum JL, et al. Multicenter, phase II study of capecitabinein taxane-pretreated metastatic breast carcinoma patients.Cancer 2001; 92:1759–1768.

11. Reichardt P, et al. Multicenter phase II study of oralcapecitabine (Xeloda®) in patients with metastatic breastcancer relapsing after treatment with a taxane-containingtherapy. Ann Oncol 2003; 14:1227–1233.

12. NCI/CTC Web site: http://ctep.info.nih.gov.13. Fazeny B, et al. Vinorelbine-induced neurotoxicity in patients

with advanced breast cancer pretreated with paclitaxel—a

230 Lee et al.


Fig. 3. Overall survival curves according to the num-ber of risk factors. Risk factors include negative hormonereceptor, refractoriness to anthracycline, disease-free sur-vival shorter than 1 year, and more than three involvedorgans.

phase II study. Cancer Chemother Pharmacol 1996; 39:150–156.

14. Livingston RB, et al. Dose-intensive vinorelbine with con-current granulocyte colony-stimulating factor support inpaclitaxel-refractory metastatic breast cancer. J Clin Oncol1997; 15:1395–1400.

15. Zelek L, et al. Weekly vinorelbine is an effective palliative reg-imen after failure with anthracyclines and taxanes in metasta-tic breast cancer carcinoma. Cancer 2001; 92:2267–2272.

16. Smorenburg C, et al. Phase II study of weekly gemcitabinein patients with metastatic breast cancer relapsing or failing

both an anthracycline and a taxane. Breast Cancer Res Treat2001; 66:83–87.

17. Valerio MR, et al. Gemcitabine in pretreated breast cancer.Proc Am Soc Clin Oncol 2001; 20: 51b (Abstr 1953).

18. Rha SY, et al. Efficacy of gemcitabine as a salvage treatmentin breast cancer patients refractory to anthracycline andpaclitaxel based regimen. Proc Am Soc Clin Oncol 2002;21:57b (Abstr 2038).

19. Liu G, et al. Patient preferences for oral versus intravenouspalliative chemotherapy. J Clin Oncol 1997; 15:110–115.



capecitabine monotherapy in patients with anthracycline-and taxane-pretreated metastatic breast...

Documents