Sequential Dose-Dense Adjuvant Therapy With Doxorubicin, Paclitaxel, and Cyclophosphamide

April 1, 1997

The recognition of paclitaxel's (Taxol's) activity and non-cross-resistance with doxorubicin (Adriamycin) in the treatment of metastatic breast cancer has motivated study of the agent in the adjuvant setting. However, the ideal

ABSTRACT: The recognition of paclitaxel's (Taxol's)activity and non-cross-resistance with doxorubicin (Adriamycin) in thetreatment of metastatic breast cancer has motivated study of the agentin the adjuvant setting. However, the ideal means of incorporating thisnew agent is not yet known. In stage IV disease, exciting results havebeen seen with combinations of doxorubicin plus paclitaxel, and this approachis being tested as adjuvant treatment. An alternative approach that hasproduced superior results with other non-cross-resistant regimens is sequentialadministration of the combination agents. Based on prior clinical trials,we tested sequential dose-dense therapy with high-dose doxorubicin, followedfirst by paclitaxel and then by cyclophosphamide (Cytoxan) for high-riskoperable breast cancer. This regimen was associated with marked toxicitybut was nonetheless tolerable and resulted in promising relapse-free survival.This regimen is now being compared to high-dose chemotherapy with autologousstem cell support for women with operable breast cancer, metastatic tofour to nine axillary lymph nodes. [ONCOLOGY 11(Suppl):15-18, 1997]


Despite the application of polychemotherapy, the majority of patientswith node-positive early stage breast cancer ultimately experience relapseand die from their disease.[1] Two efforts that aim to increase the curativeeffect of systemic chemotherapy include increased dose-intensity and inclusionof non-cross- resistant drugs or regimens. Because of paclitaxel's (Taxol)activity and clinical non-cross-resistance with doxorubicin (Adriamycin)in the treatment of metastatic breast cancer, its use in the adjuvant settingis an exciting possibility.[2-5] However, the optimal means of incorporatingthis drug while maximizing the benefit of the other active agents remainsunclear. One approach to applying combination chemotherapy is to simplycoadminister two or more active drugs. When synergy occurs, this is a desirablestrategy. Unfortunately, concurrent administration is more likely to beassociated with increased toxicity than with synergy. Further, the usualstrategy for coping with overlapping toxicities is to reduce the dose ofthe individual drugs. In the case of drugs with a steep dose-to-responserelationship, dose reductions are especially undesirable, as they may reducethe effectiveness of the regimen.

To avoid the problem of dose-limiting overlapping toxicities of agentsgiven in combination, non-cross-resistant drugs can be given separatelyrather than concurrently. In the absence of synergistic cell kill, thisis a reasonable approach. Various models have been used to plan the optimaldelivery of non-cross-resistant regimens; however, it is the clinical resultthat should determine the best model for use in subsequent studies.[6-8]


An elegant trial from Milan is especially informative, because it comparedthe use of an alternating regimen with sequential administration. Thisseminal trial, now reported with a 10-year follow-up, confirmed the prediction of superiority for sequential therapy.[9]In this study, women with metastatic breast cancer, including women withfour or more involved nodes, were treated with doxorubicin and cyclophosphamide/ methotrexate/5-fluorouracil(CMF) using one of two schedules: In the alternating-treatment arm, twocycles of CMF were delivered for every one of doxorubicin (symbolized as CCACCACCACCA). In the sequential-treatmentarm, all four doses of doxorubicin were given first, followed by all eightdoses of CMF (AAAACCCCCCCC).[10] Although it has been speculated that theearly use of doxorubicin was responsible for the improved results seenin the sequential-treatment arm, a simpler explanation is that the benefitsresulted from the greater dose intensity of doxorubicin achieved with thesequential-delivery approach.[7,11-13] To build on the exciting implicationsof this study, we performed a follow-up series of clinical trials (Table1).


Doxorubicin Followed by High-Dose Cyclophosphamide (A®C)

In an attempt to capitalize on the steep dose-to-response relationshipassociated with alkylating agents while incorporating the demonstratedbenefits of sequential therapy, our group evaluated a regimen comprisingdoxorubicin (Adriamycin) followed by high-dose cyclophosphamide (A®C).In this trial, cyclophosphamide 3,000 mg/m² was given intravenouslyfor three cycles at 14-day intervals, replacing the 24-week regimen ofCMF given every 21 days that was used in the earlier Milan model.

An earlier pilot study conducted in patients with advanced disease[14]prompted use of this dose and schedule of cyclophosphamide, given withG-CSF administered between cycles of cyclophosphamide to facilitate rapidrecovery of peripheral blood counts and rapid retreatment. Radiation therapyand tamoxifen were given after chemotherapy in appropriate cases.

In 74 patients with breast cancer and a median of nine involved nodes(range, four to 49), our pilot study demonstrated that such a regimen wasnot only feasible but also had a promising effect on relapse-free survival.[15]A prospective randomized Intergroup trial led by the Southwest OncologyGroup is currently testing this treatment approach in patients with zeroto three involved lymph nodes.[16]

Sequential Therapy With Doxorubicin/Paclitaxel/ Cyclophosphamide(A®T®C)

To maximize the potential benefit of paclitaxel as adjuvant therapy,we chose to incorporate the agent into our sequential dose-dense treatmentplan with doxorubicin and cyclophosphamide. Given that every model of chemotherapythat assumes fractional cell kill predicts increased effectiveness withmore frequent delivery of chemotherapy, we wanted both to shorten the intertreatmentintervals between dosing of the three agents[17,18] and to exploit thedose-response relationship demonstrated for doxorubicin.[12] We thereforedesigned a regimen consisting of doxorubicin (Adriamycin) 90 mg/m²,given every 14 days for three cycles, followed first by paclitaxel (Taxol)250 mg/m² over 24 hours every 14 days for three cycles and then bycyclophosphamide 3,000 mg/m² every 14 days for three cycles (A®T®C). All nine cycles of chemotherapy were supported by G-CSF, withradiation therapy and tamoxifen as appropriate.

The 42 patients treated had resectable breast cancer with a median ofeight involved lymph nodes (range, four to 25). One patient developed diseaseprogression while receiving doxorubicin and was taken off study beforereceiving paclitaxel and cyclophosphamide. Overall, 69% of the patientsrequired hospitalization (usually for neutropenic fever), and 67% requiredtransfusion of packed red blood cells. Platelet transfusion was requiredin 10%. Dose reductions of doxorubicin and paclitaxel were required in10% and 20% of patients, respectively. Despite the extent of toxicity,relapse-free survival was impressive, with only three patients relapsingafter just over 1 year of follow-up. We concluded that this regimen wasfeasible and worth further study, particularly in light of very promisingrelapse-free survival in patients who had undergone definitive local controlsurgery.[19]

Previous successful delivery of two cycles of cyclophosphamide 3,000mg/m² combined with paclitaxel 250 to 300 mg/m², given over 24hours, suggested that combination of these drugs might be just as feasibleas giving them purely by sequential administration.[20,21] Reasoning thatwe might then be able to treat patients in the adjuvant setting with allthree agents over a shorter time period, while maintaining the dose-intensityof the sequential plan, we conducted a randomized phase II trial to prospectivelycompare the toxicity of this shorter regimen with sequential A®T®C. As before, treatment began with doxorubicin given every 14 days forthree cycles, with G-CSF support. However, because the earlier trial requireddose reductions of both doxorubicin and paclitaxel, we started with a lowerdoxorubicin dose of 80 mg/m². Then, by random assignment, patientswere assigned to one of two treatment groups:

  • Group A received paclitaxel 200 mg/m² over 24 hours for threecycles, followed by cyclophosphamide 3,000 mg/m² every 14 days forthree cycles, again supported by G-CSF.
  • Group B received three courses of combination treatment with both agentsat the same dose levels.

Thus, half of the patients had nine cycles of chemotherapy (doxorubicinfollowed by paclitaxel followed by cyclophosphamide), and half had onlysix (doxorubicin followed by combination paclitaxel/cyclophosphamide),but all received the same total dose of the three study agents. Althoughfollow-up time is insufficient for comparison with previous studies, resultsof this study showed that the feasibility of the shorter six-cycle regimenwas reduced in comparison to the longer, sequential-therapy regimen. Patientshad significantly more dose delays, dose reductions, blood transfusions,and hospitalizations.[22] Based on results of this trial, we concludedthat the nine-cycle, sequential regimen was preferable for further study.


Paclitaxel Plus Doxorubicin

Two European phase II trials[23,24] suggested that paclitaxel plus doxorubicinmight provide additive, if not synergistic, activity. A prospective randomizedtrial comparing the combination versus either drug alone has been completedbut not yet published and promises to clarify the relative advantages andtoxicities of the combination versus single-agent therapy with either drug.[25]

To assess the value of this doxorubicin/paclitaxel couplet, and basedon the exciting results of their phase II study, the group from Milan hasbegun a randomized trial to evaluate this combination as adjuvant or neoadjuvanttherapy. Patients are randomly assigned to receive high-dose doxorubicinfollowed sequentially by CMF, similar to the regimen used in the earliertrial from Milan[9], or to receive lower-dose doxorubicin coupled witha 3-hour infusion of paclitaxel, a regimen that appears to be superior in phase II studies of patients with advanced disease.In this trial, patients also have been randomized for the timing of surgeryand chemotherapy.

Doxorubicin/Cyclophosphamide With or Without Paclitaxel

An alternative means of adding paclitaxel that may avoid the potentialfor increased toxicity seen in some studies of concurrent dosing is deliveryof paclitaxel following administration of the other agents. Two multicentertrials currently under way in the United States take this approach: Thefirst study, led by the Cancer and Leukemia Group B but available throughthe Intergroup, randomizes patients to receive four courses of A®C usingstandard, moderately increased, or markedly increased doses of doxorubicin.Subsequently, half of the patients receive paclitaxel and half do not.A second trial of similar design is being performed by the National SurgicalAdjuvant Breast and Bowel Project. In this trial, four courses of standardA®C are followed by four courses of paclitaxel, compared with four coursesof standard A®C without paclitaxel in women with resected node-positive breast cancer. Taken together, thesetwo trials can be expected to determine the value of adding paclitaxelto A®C as adjuvant chemotherapy for node-positive breast cancer.[16]


Currently, two prospective randomized trials are under way to test thebenefits of high-dose autologous stem-cell_supported chemotherapy for patientswith breast cancer involving 10 or more axillary lymph nodes.[16] To comparethe efficacy and toxicity of sequential dose-dense therapy with A®T®C versus the more conventional dose-escalated approach, the Intergrouphas begun a trial led by the Southwest Oncology Group (SWOG) in women withbreast cancer and four to nine involved nodes. Eligible postoperative patientsare randomized to treatment with sequential A®T®C as described aboveor to four cycles of A®C followed by high-dose, autologous stem-cell-supportedchemotherapy. Follow-up tamoxifen and radiotherapy are given as appropriateon both arms. Together with the two trials evaluating the simple additionof paclitaxel and the two trials evaluating high-dose chemotherapy, thisstudy should help to further define the role of single-agent paclitaxelin the adjuvant setting and to clarify the relative benefits and toxicitiesof the dose-dense approach.


1. Anonymous: Systemic treatment of early breast cancer by hormonal,cytotoxic, or immune therapy: 33 randomised trials involving 31,000 recurrencesand 24,000 deaths among 75,000 women. Lancet 339:1-15,71-85, 1992.

2. Holmes F, Walters R, Theriault R, et al: Phase II trial oftaxol, an active drug in the treatment of metastatic breast cancer. JNatl Cancer Inst 83:1797-1805, 1991.

3. Reichman BS, Seidman AD, Crown JP, et al: Paclitaxel and recombinanthuman granulocyte colony-stimulating factor as initial chemotherapy formetastatic breast cancer. J Clin Oncol 11:1943-1951, 1993.

4. Seidman A, Reichman B, Crown J, et al: Paclitaxel as secondand subsequent therapy for metastatic breast cancer: Activity independentof prior anthracycline response. J Clin Oncol 13:1152-1159, 1995.

5. Gianni L, Munzone E, Capri G, et al: Paclitaxel in metastatic breastcancer: A trial of two doses by a 3-hour infusion in patients with diseaserecurrence after prior therapy with anthracyclines. J Natl Cancer Inst87:1169-1175, 1995.

6. Goldie JH, Coldman AJ, Gudauskas GA: Rationale for the use of alternatingnon_cross-resistant chemotherapy. Cancer Treat Rep 66:439-449, 1982.

7. Day RS: Treatment sequencing, asymmetry, and uncertainty: Protocolstrategies for combination chemotherapy. Cancer Res 46:3876-3885,1986.

8. Norton L, Simon R: Tumor size, sensitivity to therapy, and designof treatment schedules. Cancer Treat Rep 61:1307-1315, 1977.

9. Norton L, Simon R: The Norton-Simon hypothesis revisited. CancerTreat Rep 70:163-169, 1986.

10. Bonadonna G, Zambette M, Valagussa P: Sequential or alternatingdoxorubicin and CMF regimens in breast cancer with more than three positivenodes. JAMA 273: 542-547, 1995.

11. Jones RB, Holland JF, Bhardwaj S, et al: A phase I-II study of intensive-doseAdriamycin for advanced breast cancer. J Clin Oncol 5:172-177, 1987.

12. Bronchud MH, Howell A, Crowther D, et al: The use of granulocytecolony-stimulating factor to increase the intensity of treatment with doxorubicin in patients with advanced breast and ovarian cancer. BrJ Cancer 60:121-125, 1989.

13. Wood W, Budman D, Korzun A, et al: Dose and dose intensitytrial of adjuvant chemotherapy for stage II, node-positive breast carcinoma.N Engl J Med 330:1253-1259, 1994.

14. Lichtman S, Ratain M, Van Echo D, et al: Phase I trial of granulocyte-macrophage colony-stimulating factor (GM-CSF)plus high dose biweekly cyclophosphamide: A CALGB study. J Natl CancerInst 85:1319-1326, 1993.

15. Hudis C, Lebwohl D, Crown J, et al: Adjuvant Therapy of CancerIV, pp 214-219. Philadelphia: Lippincott, 1993.

16. Hudis C, Norton L: Adjuvant drug therapy for operable breast cancer.Semin Oncol 23:475-493, 1996.

17. Skipper HE: Kinetics of mammary tumor cell growth and implicationsfor therapy. Cancer 28:1479-1499, 1971.

18. Norton L, Simon R, Brereton J, et al: Predicting the courseof Gompertzian growth. Nature 264:542-545, 1976.

19. Hudis C, Seidman A, Baselga J, et al: Sequential adjuvanttherapy with doxorubicin/ paclitaxel/cyclophosphamide for resectable breastcancer involving four or more axillary nodes. Semin Oncol 22(suppl15):18-23, 1995.

20. Fennelly D, Schneider J, Spriggs D, et al: Dose escalationof paclitaxel with high-dose cyclophosphamide, with analysis of progenitor-cellmobilization and hematologic support of advanced ovarian cancer patientsreceiving rapidly sequenced high-dose carboplatin/cyclophosphamide courses.J Clin Oncol 13:1160-1166, 1995.

21. Raptis G, Tiersten A, Fennelly D, et al: The addition of paclitaxel to high-dose cyclophosphamide and filgrastim markedlyenhances the mobilization of CD 34+ peripheral blood progenitor cells inpatients with responding metastatic breast cancer undergoing high-dosechemotherapy [Abstract]. Proc Am Soc Clin Oncol 14:313, 1995.

22. Hudis C, Seidman A, Raptis G, et al: Sequential dose-densedoxorubicin®paclitaxel®cyclophosphamide is less toxic than doxorubicin®concurrent paclitaxel+ cyclophosphamide as adjuvant therapy in resected node positive breastcancer [Abstract]. Proc Am Soc Clin Oncol 15:119, 1996.

23. Gianni L, Munzone E, Capri G, et al: Paclitaxel by 3-hourinfusion in combination with bolus doxorubicin in women with untreatedmetastatic breast cancer: High antitumor efficacy and cardiac effects ina dose-finding and sequence-finding study. J Clin Oncol 13:2688-2699,1995.

24. Dombernowsky P, Gehl J, Boesgaard M, et al: Treatment ofmetastatic breast cancer with paclitaxel and doxorubicin. Semin Oncol22:13-17, 1995.

25. Sledge GJ: Doxorubicin/paclitaxel combination chemotherapy for metastaticbreast cancer: The Eastern Cooperative Oncology Group experience. SeminOncol 22(suppl 12):123-129, 1995.