Docetaxel vs Doxorubicin in Metastatic Breast Cancer Resistant to Alkylating Chemotherapy

August 1, 1997

Single-agent docetaxel (Taxotere) has been shown to be highly active in metastatic breast cancer, with an overall response rate of 47%, median time to progression of 4 months, and survival of 10 months when administered as

ABSTRACT: Single-agent docetaxel (Taxotere)has been shown to be highly active in metastatic breast cancer, with anoverall response rate of 47%, median time to progression of 4 months, andsurvival of 10 months when administered as second-line therapy. These datacompare favorably with those reported for doxorubicin (Adriamycin), whichhas been considered the most active single agent in this setting. Thisnonblinded, multicenter, randomized phase III study compared the mediantime to progression, response rate, quality of life, toxicity, and survivalafter treatment with docetaxel or doxorubicin in patients with metastaticbreast cancer in whom previous alkylating chemotherapy failed. Patientswere randomized to receive an intravenous infusion of either docetaxel,100 mg/m², for 1 hour once every 3 weeks or doxorubicin, 75 mg/m²,for 15 to 20 minutes once every 3 weeks. This preliminary analysis presentsdata on 200 of 326 patients recruited. It was performed after the completionof patient accrual. The median time to progression was greater in the docetaxelgroup than in the doxorubicin group (29 vs 21 weeks, respectively; P =not significant). Overall response rates were higher with docetaxel (47%vs 27%), and fewer patients in the docetaxel group had progressive diseaseas their best overall response (10% vs 22%). Both regimens caused the sameincidence and severity of neutropenia, yet patients treated with doxorubicinhad a higher incidence of infection and febrile neutropenia. In addition,doxorubicin produced a higher incidence of grade 3 to 4 thrombocytopenia.Cardiac toxicity led to discontinuation in 7 patients and death in 2 patientsin the doxorubicin group; fluid retention led to discontinuation in 1 patientin the docetaxel group. Based on this preliminary analysis, docetaxel wasmore active and safer than doxorubicin in patients with metastatic breastcancer in whom previous alkylating chemotherapy failed. [ONCOLOGY 11(Suppl 8):19-24,1997]


Docetaxel (Taxotere) is reported to have excellent activity againstanthracycline-resistant metastatic breast cancer. In a series of 4 multicenterphase II trials, 100 mg/m² of docetaxel administered as a 1-hour intravenousinfusion once every 3 weeks produced an overall response rate of 47% (range:41% to 58%; 95% confidence interval), with a median time to progressionof 4 months and survival of 10 months.[1-5]

These results are higher than those reported with doxorubicin (Adriamycin)monotherapy, which is considered to be the most active single agent insecond-line therapy for patients with metastatic breast cancer. Responserates associated with doxorubicin in the second-line treatment of metastaticbreast cancer have ranged from 25% to 33%, with a median time to progressionof 3.6 months and a median survival of 8.9 months.[6-13]

Because docetaxel has shown significant second-line activity in phaseII trials, a phase III study was performed to compare the effects of docetaxeland doxorubicin in patients with metastatic breast cancer in whom prioralkylating chemotherapy failed. This preliminary analysis presents comparativedata on the median time to progression after treatment, response rates,and toxicity profiles following treatment with docetaxel and doxorubicin.Data are presented on 200 of the 326 recruited patients.



Women aged 18 to 75 years who had histologically or cytologically provenprogressive metastatic adenocarcinoma of the breast and measurable and/orevaluable disease were considered for study participation, provided theymet the following criteria: a performance status of at least 60% (Karnofskyindex); no previous therapy with anthracyclines, anthracenediones, or taxanes;and failure after alkylating chemotherapy. Prior treatment with hormonaltherapy for either advanced disease or in the adjuvant setting was permitted,as was radiation therapy.

Response after alkylating chemotherapy was defined as:

  • Primary resistant--patients who relapsed during adjuvant chemotherapyor disease progression as the best response to chemotherapy for metastaticbreast cancer
  • Secondary resistant--patients who relapsed within 12 months after adjuvantchemotherapy or disease progression on chemotherapy for metastatic breastcancer after an initial response
  • Not resistant--patients who relapsed at least 12 months after receivingadjuvant (first-line) chemotherapy or had disease progression at least30 days after chemotherapy for metastatic breast cancer.

Laboratory entry criteria included the following values: absolute neutrophilcount ³ 2.0 or greater × 109/L; a platelet count ³ 100.0 × 109/L; total bilirubin £ 1.25 or less × upper normal limit(UNL); aspartate aminotransferase (ASAT) or alanine aminotransferase (ALAT) £ 3 × UNL; alkaline phosphatase £ 6 × UNL; ASAT orALAT or both £ 1.5 × UNL associated with alkaline phosphatase £ 2.5 × UNL; serum creatinine £ 1.5 × UNL; and aresting left ventricular ejection fraction above the lower normal limitof the institution, as measured by echocardiography or radionuclide angiocardiography.

Specific criteria for exclusion were: local recurrence within partiallyresected breast or locally advanced inoperable breast cancer (stage IIIB)as the only manifestation of the disease; more than 1 line of chemotherapyfor advanced or metastatic disease; history or presence of brain or leptomeningealmetastases; prior or concurrent malignancies, with the exception of adequatelytreated in situ carcinoma of the uterine cervix and cured nonmelanoma skincancer, osteoblastic skeletal lesions, a single osteolytic lesion, lymphedema,pulmonary lymphangitic metastases, pleural effusion, or ascites as theonly site of disease; and symptomatic peripheral neuropathy of at leastgrade 2 according to National Cancer Institute (NCI) Common Toxicity Criteria.

Patients were recruited from 42 centers worldwide. Ethics committeeapproval and informed patient consent were obtained before the start ofthe trial.

Study Design and Treatment Plan

This was a nonblinded, randomized, multicenter phase III study. Therandomization was stratified by center, and patients were assigned randomlyto receive an intravenous infusion of either docetaxel, 100 mg/m²,for 1 hour once every 3 weeks, or doxorubicin, 75 mg/m², for 15 to20 minutes once every 3 weeks. Premedication was specified for patientsin the docetaxel group only, and was comprised of 8 mg of oral dexamethasone,given 13 hours, 7 hours, and 1 hour before docetaxel infusion, and foran additional 4 days at a dose of 8 mg twice daily, starting immediatelyafter docetaxel infusion.

The maximum duration of treatment was 7 cycles for both groups, unlessprogression or unacceptable toxicity occurred. If a patient failed to respondto the assigned treatment, further treatment was at the discretion of theinvestigator. Patients withdrawn from the study before disease progressioncould not receive other antitumor therapy until progression was documented,unless deemed necessary by the investigator. Patients were observed for1 month after their last study treatment infusion to document any lateadverse events, with a follow-up every 3 months until death, to documenttime to progression and survival.

Dose reductions were permitted for severe hematologic and nonhematologictoxicities other than alopecia and anemia, graded according to NCI CommonToxicity Criteria. A maximum of 2 dose reductions were allowed per patient--ie,from 100 to 75 mg/m² and from 75 to 55 mg/m² for docetaxel, andfrom 75 to 60 mg/m² and from 60 to 45 mg/m² for doxorubicin.

Concomitant bisphosphonate treatment was not allowed unless initiatedmore than 3 months before the start of the study.

Study Assessments

A complete tumor assessment was performed during the 3 weeks beforethe first infusion of study medication, and included chest x-ray, bonescintigraphy and bone radiological examination, abdominal computed tomographyscan, or ultrasound and physical examination. Bone scintigraphy could beperformed 4 weeks before the first infusion of study medication. All evaluableand nonevaluable lesions were to be assessed at least every 2 treatmentcycles.

The primary efficacy variable was time to progression, calculated fromthe date of randomization to the first progression. Response rate, definedas the percentage of patients in the group who achieved a complete or partialresponse, was a secondary efficacy variable. Patients with disease progressionbefore the end of the second treatment cycle were considered to have earlyprogression, whereas patients who received at least 2 cycles of therapyhad their response to treatment classified as follows: complete response,partial response, stable disease, or progressive disease, according tothe World Health Organization response criteria.

Weekly blood counts were performed. An initial assessment of left ventricularejection fraction was made during the 2 weeks before study entry, usinga multiple-gated acquisition scan or echocardiography; left ventricularejection fraction was reassessed after a 400 mg/m² cumulative dosein the doxorubicin group, and at the end of the study in both treatmentgroups. Two types of analysis were performed on left ventricular ejectionfraction: relative decrease in left ventricular ejection fraction frombaseline according to NCI Common Toxicity Criteria, and absolute decreasein left ventricular ejection fraction from baseline according to the Schwartzcriteria-- ie, a decrease in left ventricular ejection fraction of at least10 absolute percentage points and below the lower normal limit.[14]

Statistical Analyses

A two-tailed log-rank test was used to compare differences in the mediantime to progression between treatment groups. A significance level of 0.001,according to Peto sequential procedure,[15] was used for this preliminaryevaluation, which was performed after patient accrual was completed. Patientswere required to have received at least 2 cycles of treatment and to havehad at least 1 follow-up tumor assessment to be evaluable for efficacy,unless disease progression occurred.

Two subpopulations were analyzed for efficacy: a second-line populationcomprised of patients who relapsed during treatment or within 12 monthsof the end of adjuvant chemotherapy containing an alkylating agent, orpatients who had received 1 previous alkylating chemotherapy regimen foradvanced or metastatic disease; and a first-line population, comprisedof patients who relapsed more than 12 months after the end of adjuvantchemotherapy containing an alkylating agent and who had not received chemotherapyfor advanced disease. All patients who received at least 1 infusion ofstudy medication were evaluable for safety. Times-to-event variables wereanalyzed by the Kaplan-Meier method. All analyses were performed at leaston the randomized (intent- to-treat) population.

Preliminary Results

Patient Demographics

This preliminary analysis reports on data from 200 patients randomizedto receive either docetaxel (N = 102) or doxorubicin (N = 98). No differencein patient and tumor characteristics was found in the treatment groups(Table 1). With regard to metastaticsites, patients in this study had large tumor burdens, as indicated byliver involvement in approximately 39% of patients and by the 39% of patientswith 3 or more involved sites. In addition, measurable disease was notedin approximately 83% of patients.

Treatment Administration

A median of 7 cycles (range: 1 to 10) of docetaxel was administeredto 101 patients with a median cumulative dose of 620 mg/m² (range:5 to 992 mg/m²). For doxorubicin, a median of 5 cycles (range: 1 to7) was administered with a median cumulative dose of 353 mg/m² (range:73 to 540 mg/m²). The median relative dose intensity was 0.98 (range:0.5 to 1.07) for docetaxel and 0.95 (range: 0.49 to 1.05) for doxorubicin,with almost all patients in both treatment groups receiving a relativedose intensity of more than 70% of the planned dose.

Response Rate and Median Time to Progression

Using an intent-to-treat analysis, the overall response rate for patientsreceiving docetaxel was 47% (37% to 57%; 95% confidence interval) comparedwith 27% (18% to 36%; 95% confidence interval) for the doxorubicin-treatedpatients (Table 2). In the docetaxelgroup, 10% of patients achieved a complete response compared with 3% ofpatients in the doxorubicin group. More patients in the doxorubicin group(22%) experienced disease progression compared with docetaxel-treated patients(10%).

Preliminary analysis indicated that patients who received docetaxelhad a median time to progression of 29 weeks compared with 21 weeks forthe doxorubicin-treated patients. The response rates reported here arepreliminary and need to be confirmed based on data derived from the entirepatient population.

In the subpopulation analysis, overall response rates favored docetaxelboth as first- and second-line subgroups (Table3). As first-line therapy, the response rate was 56% in the docetaxelgroup and 46% in the doxorubicin group. In addition, more patients achieveda complete response (15% vs 4%) and fewer patients had progression as thebest response (5% vs 13%) in the docetaxel group compared with the doxorubicingroup.

As second-line therapy, a response rate of 41% (range: 29% to 54%; 95%confidence interval) was noted in patients receiving docetaxel comparedwith 20% (range: 12% to 31%; 95% confidence interval) in doxorubicin-treatedpatients. As in the first-line subanalysis, fewer patients had progressionas best response on docetaxel therapy (13%) as compared with doxorubicintherapy (26%).


The most common reason for treatment discontinuation among the 200 patientswas disease progression (docetaxel: N = 30; doxorubicin: N = 39). The twogroups differed with regard to the adverse events that led to discontinuationof treatment. For example, cardiotoxicity was exclusively found in thedoxorubicin arm (N = 7) and was responsible for 2 deaths. The cumulativedose of doxorubicin for those 2 patients was 350 and 450 mg/m². Onepatient was a 44-year-old woman with a history of diabetes mellitus andthe second, a 73-year-old women, who had a history of arterial hypertension.Neither patient had received prior chest-wall radiotherapy. In contrast,neuropathy (3 patients) and fluid retention (1 patient) accounted for discontinuationin 4 patients who were receiving docetaxel.

The primary hematologic toxicity noted during the study was neutropenia,which was comparable between the 2 groups (Table4). However, febrile neutropenia was more prevalent in the doxorubicin-treatedpatients (11.3%) compared with the docetaxel group (5.9%). Grade 3 to 4infections due to neutropenia were also more common in the doxorubicingroup (6.2%) compared with the docetaxel group (3.0%). The largest differencewas seen with the incidence of grade 3 to 4 thrombocytopenia, with incidencesof 1.0% in the docetaxel group vs 16.9% in the doxorubicin group.

With regard to acute nonhematologic toxicities, patients receiving doxorubicintended to experience more severe nausea (17.5% vs 4%), vomiting (14.4%vs 3%), and stomatitis (14.4% vs 5%) compared with docetaxel-treated patients(Table 5). On the other hand, diarrheawas more common in the docetaxel group (9.9%) compared with the doxorubicingroup (1.0%). The chronic nonhematologic toxicities for both treatmentgroups have been noted in previous trials (Table5).


The preliminary analysis of thislarge, multicenter phase III trial suggeststhat 100 mg/m² of docetaxel administered as a 1-hour infusion onceevery 3 weeks is more active than 75 mg/m² of doxorubicin administeredonce every 3 weeks in patients with metastatic breast cancer in whom previousalkylating chemotherapy failed. The complete response (10% vs 3%) and overallresponse (47% vs 27%) rates achieved with docetaxel were greater than thoseachieved by doxorubicin.

The response rates reported in this preliminary analysis are consistentwith those from phase II trials.[7,16,17] In addition, there were fewerpatients in the docetaxel group who had progressive disease as the bestresponse (10% vs 22%), and there was a trend for longer median time toprogression (29 vs 21 weeks) compared with the doxorubicin group.

Although both regimens caused the same incidence and severity of neutropenia,patients treated with doxorubicin had a higher incidence of infection andfebrile neutropenia. In addition, doxorubicin produced a higher incidenceof grade 3 to 4 thrombocytopenia. Cardiotoxicity associated with doxorubicinat doses of 450 mg/m²or less led to the death of 2 patients. By contrast,the fluid retention associated with docetaxel is reversible, manageable,and not life-threatening, as evidenced by the fact that it was responsiblefor the discontinuation of treatment in only 1 patient and no fatalities.

In conclusion, the preliminary results of this multicenter, randomizedphase III study indicate that docetaxel is more active and appears to besafer than doxorubicin in patients with metastatic breast cancer in whomprevious alkylating chemotherapy failed. Final analysis is awaited to confirmthese preliminary results.


1. Van Oosterom AT, Schrijvers D: Docetaxel (Taxotere), a review ofpreclinical and clinical experience. Part II: clinical experience. AntiCancer Drugs 6:356-368, 1995.

2. ten Bokkel Huinink WW, Prove AM, Piccart M, et al: A phase II trialof docetaxel (Taxotere) in second line treatment with chemotherapy foradvanced breast cancer: A study of the EORTC Early Clinical Trials Group.Ann Oncol 5:527-532, 1994.

3. Radvin P, Burris HA, Cook G, et al: Phase II trial of docetaxel inadvanced anthracycline-resistant or anthracenedione-resistant breast cancer.J Clin Oncol 13:2879-2885, 1995.

4. Valero V, Holmes FA, Walters RS, et al: Phase II trial of docetaxel,a new, highly antineoplastic agent in the management of patients with anthracycline-resistantmetastatic breast cancer. J Clin Oncol 13:2886-2894, 1995.

5. Guastalla JP, Bonneterre J, Fumoleau P, et al: A phase II trial ofdocetaxel in patients with anthracycline-resistant metastatic breast cancer(MBC) (abstract 348). Eur J Cancer 31A:S75-S76, 1995.

6. Tormey DC: Adriamycin (NSC-123127) in breast cancer: An overviewof studies. Cancer Chemother Rep Part 3;6:19-327, 1975.

7. Henderson IC, Allegra JC, Woodcock T, et al: Randomized clinicaltrial comparing mitoxantrone with doxorubicin in previously treated patientswith metastatic breast cancer. J Clin Oncol 7:560-571, 1989.

8. Ingle JN, Maillard JA, Schaid DJ, et al: Randomized trial of doxorubicinalone or combined with vincristine and mitomycin C in women with metastaticbreast cancer. Am J Clin Oncol 12:474-480, 1989.

9. Neidhart JA, Gochnour D, Roach R, et al: A comparison of mitoxantroneand doxorubicin in breast cancer. J Clin Oncol 4:672-677, 1986.

10. Hoogstraten B, George SL, Samal B, et al: Combination chemotherapyand adriamycin in patients with advanced breast cancer. Cancer 38:13-20,1976.

11. Vaughn CB, Green SJ, O'Bryan R, et al: VP-16 + adriamycin vs. adriamycinalone in advanced adenocarcinoma of the breast, phase II, a randomizedtrial: A Southwest Oncology Group Study. Med Pediatr Oncol 16:312-319,1988.

12. Frederiksen FL, Joergensen ST, Roesdahl K, et al: Activity of adriamycinin metastatic breast cancer resistant to combination regimen with cyclophosphamide,methotrexate, 5-fluorouracil, vincristine and prednisone. Cancer TreatRep 62:449-450, 1978.

13. Brambilla C, Rossi A, Bonfante V, et al: Phase II study of doxorubicinversus epirubicin in advanced breast cancer. Cancer Treat Rep 70:261-266,1986.

14. Schwartz RG, McKenzie WB, Alexander J, et al: Congestive heart failureand left ventricular dysfunction complicating doxorubicin therapy. Seven-yearexperience using serial radionuclide angiocardiography. Am J Med 82:1109-1118,1987.

15. Peto R, Pike MC, Armitage P, et al: Design and analysis of randomizedclinical trials requiring prolonged observation of each patient. I. Introductionand design. Br J Cancer 34:585-612, 1976.

16. Paridaens R, Bruning P, Klijn J, et al: An EORTC crossover trialcomparing single-agent Taxol and doxorubicin as first- and second-linechemotherapy in metastatic breast cancer. Proc Am Soc Clin Oncol 16:154a(abstract 539), 1997.

17. Sledge GW, Neuberg D, Ingle J, et al: Phase III trial of doxorubicinvs. paclitaxel vs. doxorubicin + paclitaxel as first-line therapy for metastaticbreast cancer: An intergroup trial (abstract 2). Proc Am Soc Clin Oncol16:1a, 1997.