High-dose sequential chemotherapy with stem cell support for non-metastatic breast cancer

Bone Marrow Transplantation, (1997) 20, 199 203 1997 Stockton Press All rights reserved 0268 3369/97 $12.00 High-dose sequential chemotherapy with stem cell support for non-metastatic breast cancer P Viens 1, G Gravis 1, D Genre 1, F Bertucci 1, D Cowen 2, J Camerlo 1, M-A Cappiello 1, M Conte 1, M Finaud 1, C Chabannon 4, G Houvenaeghel 3 and D Maraninchi 1 Departments of 1 Medical Oncology, 2 Radiotherapy, 3 Surgery, 4 Gene Therapy, Institut Paoli-Calmettes, Marseille, France Summary: drugs used to treat breast cancer for which a dose-response relationship has been demonstrated. 5,6 Hematological toxicity, The importance of dose-intensity has been suggested in particularly for alkylating agents is a limiting factor breast cancer. The aim of this study was to evaluate the in the escalation of dose-intensity. Hematopoietic growth feasability of a high-dose intensity doxorubicin-cyclophosphamide factors have been successfully used to decrease myeloid regimen with supporting G-CSF and toxicity in standard regimens 7,8 and to increase dose and blood stem cells. Twenty-five patients with non-metastatic dose-intensity. 9,10 However, the repetition of a high-dose breast cancer received four cycles of doxorubicin combination regimen supported by GM-CSF or G-CSF (75 mg/m 2 ) and cyclophosphamide (3000 mg/m 2 )at3 generally leads to a decrease in platelet count which week intervals. Apheresis was performed after the first appears to be a limiting toxicity, independent of neutropenia, cycle and if necessary after the second cycle. Stem cells to high-dose intensity regimen. Peripheral blood procycle were reinfused after the third and fourth cycles. G-CSF genitors have the capacity to enhance hematological recovery was started on day 3 of each cycle (5 g/kg/day) and after myeloablative regimens, particularly for the was stopped the day before the last apheresis or when platelet lineage 11 and decrease myelotoxicity of repetitive absolute neutrophil count was above 0.5 10 9 /l. Median high-dose regimens. 12 15 received dose-intensity was respectively 25 mg/m 2 /week The aim of this study was to increase dose-intensity and (range 22 26) and 1000 mg/m 2 /week (range 904 1065) total cyclophosphamide dose 5-fold and by 1.25 dose-intenfor doxorubicin and cyclophosphamide. Grade IV sity and total dose of doxorubicin in the standard regimen thrombocytopenia occurred in 8% of cycles. Two associating cyclophosphamide (600 mg/m 2 4) and patients needed platelets and 12 red cell transfusion. Fif- doxorubicin (60 mg/m 2 4). 16 We report here tolerance of teen patients were readmitted for a median duration of 100 cycles of such a regimen in 25 patients with breast 4 days (range 1 7). We have established a safe, out- cancer and show that with the use of both G-CSF (filgrastim) and peripheral blood progenitors, a major dose- escalation can be routinely achieved in an outpatient setting. patient, high-dose intensity doxorubicin-cyclophosphamide regimen with supporting G-CSF and blood stem cells which can be submitted for comparison with the current standards. Keywords: stem cells; breast cancer; chemotherapy Patients and methods Patients with poor prognosis non-metastatic breast cancer were enrolled in this study. Patient eligibility requirements included women aged at least 18 years, non-menopausal, with non-metastatic histologically proven breast cancer and more than three involved homolateral axillary lymph nodes. WHO performance status had to be 0, 1 or 2, neutrophil count above 1.5 10 9 /l, platelet count above 100 10 9 /l, total bilirubin, ASAT, ALAT and serum creatinine less than 1.25 times the upper limit of normal range. Patients with inflammatory breast cancer, contralateral or supra-clavicu- lar involved lymph nodes were excluded from the study as were all cases of relapse of breast cancer. The study was approved by the local ethical committee and all patients were asked to give written informed consent. Pre-treatment evaluation included complete physical examination with medical history, CBC with differential, liver and renal function test, coagulation profile, EKG and The increase in dose and dose-intensity (amount of chemotherapy per unit of time) may increase response in various solid tumors including breast cancer. 1 Retrospective studies have pointed to the impact of dose-intensity, either on disease-free survival or survival, in adjuvant chemotherapy regimen. 2,3 A prospective study from CALGB 4 tested three levels of dose-intensity in an association of doxorubicincyclophosphamide-5 fluorouracil in node-positive breast cancer with a benefit in progression free survival and survival for patients receiving the two highest dose-intensity levels. In all these studies, dose-intensity ranged from 0.75 to 2 of standard levels and total doses were low. Doxorubicin and cyclophosphamide are two of the major Correspondence: Dr P Viens, Medical Oncology Department, Institut Paoli-Calmettes, 232 Boulevard de Sainte-Marguerite, 13273 Marseille Cedex 9, France Received 28 January 1997; accepted 22 April 1997 Patients eligibility and characteristics

200 radionuclide ventriculography or echocardiography, chest X-ray, liver ultrasonography and bone scan. Between September 1993 and September 1995, 25 nonmenopausal women with poor prognosis breast cancer were enrolled in the study. Their median age was 45 years (range 27 57). Twenty-three patients had a ductal invasive carcinoma and two a lobular. The median number of positive homolateral axillary lymph nodes was five (range 4 8). Initial surgery consisted of a modified mastectomy with axillary dissection for one patient and tumorectomy with axillary dissection for the other 24. All 25 patients received the four cycles of chemotherapy. Treatment three bags: two containing at least 2 10 6 CD34 + cells/kg and one for back-up. Blood stem cells were reinfused on day 3 of cycle 3 and cycle 4 (at least 2 10 6 CD34 + cells/kg each infusion). Apheresis was started when the absolute number of CD34 + cells in the peripheral blood rose to 20/ l. Apheresis was performed using the Cobe Spectra automated processor (Cobe, Denver, CO, USA), using an ACD/blood ratio of approximately 1/8 to 1/12. Two blood masses were treated during each 3 h session. Supportive care: Anti-emetic therapy consisted of ondansetron 8 mg and dexamethasone 20 mg i.v., infused 30 min before chemotherapy. Patients were discharged from hospital after chemotherapy. Blood stem cell collection and reinfusion were performed in a conventional hospital unit or in an outpatient clinic. No antibiotic prophylaxis was given but patients were hospitalized if they became febrile ( 38 C) to receive i.v. antibiotics. Red blood cells were transfused when hemoglobin was 8 g/dl or for anemia Chemotherapy (Figure 1): After initial surgery, patients were treated with four cycles of high-dose cyclophosphamide (3000 mg/m 2 ) and doxorubicin (75 mg/m 2 ). Chemotherapy was scheduled every 21 days and administered during a 24 h hospitalization. Doxorubicin was infused as a 15 min infusion and cyclophosphamide as a 1 h infusion, symptoms and platelets were transfused for hemorrhagic both on day 1. Patients were hydrated with 3 l/m 2 of a symptoms. All blood products, except blood stem cells, semi-saline solution. were irradiated at 25 Gy. Chemotherapy was administered if absolute neutrophil count (ANC) was 1.5 10 9 /l and platelet count Toxicity: Once per course (D1), physical examination was 100 10 9 /l. If the patient s neutrophil and platelet count carried out and WHO performance status, vital signs, EKG, did not meet these criteria on day 21 chemotherapy was complete blood count (CBC) and differential, liver funcdelayed until adequate count recovery. No dose-reduction tion, creatinine and coagulation profile were obtained. Durwas planned. If, on day 35 of one cycle, chemotherapy was ing the treatment period, CBC and differential were not possible the patient was excluded from the study. obtained three times a week. At the end of chemotherapy For each drug, dose intensity was calculated as the total the pre-treatment evaluation, except radionuclide bone amount of chemotherapy given, divided by the body surface scan, was repeated with toxicities being assessed according area and the delay (in weeks) between day 1 of cycle 1 and to WHO criteria. 3 weeks after day 1 of cycle 4. Relative dose-intensity (RDI) was the result of dose intensity divided by the theoretical dose-intensity. G-CSF and apheresis (cell procedure): G-CSF (filgrastim; Amgen, USA) was administered subcutaneously at a daily dosage of 5 g/kg (maximum 300 g per day) from day 3 of each cycle until the day before the last apheresis or until ANC reached 0.5 10 9 /l on 3 consecutive days for cycles without apheresis. At the beginning of the study, apheresis was systemati- cally done after the first and the second cycle of chemotherapy, and was stopped as soon as the collection of CD 34 + cells exceeded 6 10 6 /kg. Cells were divided into Cycle (D1 = D21) G-CSF** 1 AC* 2 3 4 AC AC AC D5 D5 D3 D3 ( ) ( ) BSC BSC BSC BSC Results Stem cell collection For the first patients, stem cells were collected systematically with two sets of apheresis after the first cycle of chemotherapy and one or two sets of apheresis after the second. After six patients had been treated, apheresis was stopped as soon as 6 10 6 CD34 + cells/kg were collected. Two sets of apheresis were done for 18 patients and one set for seven after the first cycle of chemotherapy, and two sets for two patients and one for five patients after the second cycle. The median number of CD34 + cells collected per kg was 9.54 10 6 (range 4.26 24.61) after the first cycle and 1.88 10 6 (range 1.39 13.67) after the second cycle. The number of 6 10 6 CD34 + cells/kg was obtained for 24 out of 25 patients after the first cycle of chemotherapy. Hematological toxicity Figure 1 Treatment schedule *AC, doxorubicin 75 mg/m 2 + cyclophos- phamide 3000 mg/m 2. **Stop G-CSF, the day before the last apheresis or ANC 500/mm 3. D, day;, Apheresis;, Stem cell infusion; BSC, blood stem cells. Neutropenia (Table 1): G-CSF was administered for 11 days (range 8 14) at cycle 1, 11 days (range 8 12) at cycle 2, 11 days (range 8 15) at cycle 3 and 11 days (range 8 14) at cycle 4. All patients experienced grade 3 neutropenia

Table 1 Neutropenia and hospitalization 1065) and for doxorubicin 25 mg/m 2 /week (range 22 26). Relative dose-intensity for cyclophosphamide was superior Median duration Percentage of Median duration to 95% for 23 patients and between 90 and 95% for two. of neutropenia cycles with of hospitalization For doxorubicin RDI was superior to 95% for 24 patients grade IV febrile (days) and was 88% for one. For this patient (RDI of doxorubicin (days) neutropenia and cyclophosphamide were respectively of 88% and 90%) Cycle 1 2 (1; 4) 32 4 (1; 6) cycles 2 and 4 were delayed by 4 and 5 days for neutrop- Cycle 2 3 (1; 5) 20 3 (1; 6) enia (day 1 of cycle 2 ANC: 1.3 10 9 /l, day 1 of cycle 4 Cycle 3 3 (1; 5) 48 3 (1; 6) ANC: 1 10 9 /l). Cycle 4 1 (1; 5) 24 3 (1; 7) All 25 patients received local treatment after chemotherapy, which consisted of surgery followed by radiotherapy for nine patients and radiotherapy alone for 16 patients. The first treatment was performed in a median of and 23 out of 25 a grade 4. Grade 4 neutropenia occurred 42 days (range 27 71) after day 1 of cycle 4 in the case in 66% of the 100 cycles of chemotherapy administered. of surgery and 36 days (range 24 50) after day 1 of cycle Incidence of grade 4 neutropenia from cycle 1 to 4 was 4 in the case of radiotherapy. respectively 72, 56, 72 and 64%. We did not see any increase in the incidence or duration of neutropenia from cycle 1 to cycle 4. Thirty-four cycles of chemotherapy were followed by hospitalization for treatment-related Discussion complications, which were, in 90% of cases, Twenty-five patients received 100 cycles of high-dose febrile neutropenia. Only one was documented with a cyclophosphamide and doxorubicin with supporting G-CSF corynebacterium. Ten patients did not need any hospitaliz- (filgrastim) and blood stem cells. This study was designed ation for toxicity during the four cycles. to evaluate the toxicity of an outpatient high-dose intensity doxorubicin-cyclophosphamide regimen which could be Thrombocytopenia (Table 2): Toxicity of the platelet lingiven as adjuvant treatment for breast cancer. The main eage was very mild. Thirteen patients had no grade III toxicity of the regimen was hematologic toxicity. As thrombocytopenia. No patient experienced grade IV thromexpected with the use of G-CSF (filgrastim) grade IV neubocytopenia in cycle 1 but the incidence of this event was tropenia was of short duration with a maximum of 5 days. 8% in cycle 2, 16% in cycle 3 and 8% in cycle 4. There Neutropenia delayed chemotherapy for only one patient. was no increase of duration of extreme thrombocytopenia Febrile neutropenia occurred in 30% of cycles and led to from cycle 2 to cycle 4. brief hospitalization. This problem was always managed in a standard oncology medical unit and bacterial docu- Additional toxicity: Other grade 2 toxicities were essenmentation occurred only once. There was little need for tially nausea and vomiting (36% of cycles), asthenia (14% platelet transfusion (2/25 patients) but requirement for red of cycles), mucositis (10% of cycles) and bone pain. No blood cells was relatively high (12/25 patients). When hemliver toxicity grade 2, acute cardiac toxicity or decrease atopoietic growth factors are used to increase dose intensity in left ventricular ejection fraction occurred. of conventional chemotherapy regimens, thrombocytopenia and the requirement for platelet transfusion is generally Treatment administration cumulative. 8,10,17,18 Recently, a dose-finding study of very high-dose CHOP for non-hodgkin s lymphomas with G- Twenty-five patients received 100 cycles of chemotherapy. CSF but without peripheral blood progenitors clearly ident- No dose-reduction, either of cyclophosphamide or doxorubified thrombocytopenia as the dose-limiting toxicity. 19 In icin, occurred. The median received dose-intensity for the study carried out by Shipp et al, thrombocytopenia cyclophosphamide was 1000 mg/m 2 /week (range 904 increased with doses and, in each step, with cycles of chemotherapy establishing the maximum tolerable doses of Table 2 Thrombocytopenia and transfusions cyclophosphamide and doxorubicin at respectively 4000 mg/m 2 and 70 mg/m 2 every 21 days. In our study, where Grade IV Platelet Red cell doses of chemotherapy are at a similar level, there was no thrombocytopenia transfusion transfusion cumulative thrombocytopenia. As shown in transplan- No. patients No. patients No. patients tation, 11 infusion of peripheral blood progenitors probably (duration) (No. of packed red cells) allows such low thrombocytopenia compared to other stud- ies using G-CSF or GM-CSF alone. However, one of the Cycle 1 0 0 2 risks in using peripheral blood progenitors is the mobiliz- (1, 2) ation, collection and reinfusion of the tumor cells. 20 Many Cycle 2 2 1 4 studies are actually carried out with the aim of identifying (1, 2) 2 (1,3) tumor cells in the apheresis product and removing them, Cycle 3 4 0 11 2 (1, 3) 2 (1, 2) especially by positive selection of CD34 + cells. On this Cycle 4 2 1 5 point, the use of peripheral blood progenitors could be crit- (3, 3) 2 (2, 2) icized in our study where patients are non-metastatic and collection is done after the first cycle of chemotherapy. 201

202 However, the impact of potential reinfusion of tumor cells intensity of adjuvant chemotherapy for stage II, node-positive is not yet clearly established but on the other hand, it breast carcinoma. New Engl J Med 1994; 330: 1253 1259. probably would have been difficult to deliver as safely the 5 Jones RB, Holland JF, Bhardwaj S et al. A phase I II study same dose and dose-intensity without peripheral blood of intensive-dose adriamycin for advanced breast cancer. J Clin Oncol 1987; 5: 172 177. progenitors. 6 Frei E III, Teicher BA, Holden SA et al. Preclinical studies Overall, the median dose intensity was 1000 mg/m 2 /week and clinical correlation of the effect of alkylating dose. Cancer for cyclophosphamide and 25 mg/m 2 /week for doxorub- Res 1988; 48: 6417 6423. icin. All the patients, except one, received more than 90% 7 Gabrilove JL, Jakubowski A, Scher H et al. Effect of granuloof the planned dose-intensity of cyclophosphamide and cyte colony-stimulating factor on neutropenia and associated doxorubicin. Increasing dose and/or dose-intensity is curof morbidity due to chemotherapy for transitional-cell carcinoma rently of major concern in breast cancer chemotherapy. One the urothelium. New Engl J Med 1988; 318: 1414 1422. method is to give, after four to six cycles of standard 8 Trillet-Lenoir V, Green J, Manegold C et al. Recombinant chemotherapy, one course of high-dose alkylating agents in granulocyte colony stimulating factor reduces the infectious combination with bone marrow support and more recently complications of cytotoxic chemotherapy. Eur J Cancer 1993; 29A: 319 324. peripheral blood progenitors. Results of pilot studies have 9 Sternberg CN, de Mulder PHM, van Oosterom AT et al. Escabeen encouraging 21,22 and several controlled studies are lated M-VAC chemotherapy and recombinant human granuloongoing which compare this approach to standard treat- cyte macrophage colony stimulating factor (rhgm-csf) in ment. However, the drawbacks of such an approach are the patients with advanced urothelial tract tumors. Ann Oncol necessity of hospitalization in a bone marrow transplant 1993; 4: 403 407. unit and possibly the potential risk of emergence of resistvents 10 Chevallier B, Chollet P, Merrouche Y et al. Lenograstim pre- ant cells during initiation of the treatment. More recently, morbidity from intensive induction chemotherapy in the several investigators have suggested the use of sequential treatment of inflammatory breast cancer. J Clin Oncol 1995; high-dose chemotherapy with hematopoietic growth factors 13: 1564 1571. and peripheral blood progenitors. 12 15,23 Most of these profilgrastim-mobilised peripheral blood progenitor cell trans- 11 Schmitz N, Linch DC, Dreger P et al. Randomised trial of grams use alkylating agents in combination or alternately, plantation versus autologous bone-marrow transplantation in and toxicity is often similar to one of the first approaches. lymphoma patients. Lancet 1996; 347: 353 357. Other groups have increased the dose-intensity of conven- 12 Pettengell R, Woll PJ, Thatcher N et al. Multicyclic, dosetional regimen, with supporting G-CSF or rh-gmcsf by intensive chemotherapy supported by sequential reinfusion of increasing the dose or reducing delay. 17,18,24 Generally, hematopoietic progenitors in whole blood. J Clin Oncol 1995; dose-escalation is limited by hematological toxicity, parti- 13: 148 156. cularly thrombocytopenia, and dose-intensity is lower than 13 Shea TC, Mason JR, Storniolo AM et al. Sequential cycles of the level reached in our study. high-dose carboplatin administered with recombinant human As the combination of doxorubicin-cyclophosphamide is granulocyte macrophage colony-stimulating factor and one of the standard adjuvant regimens in breast cancer 13 repeated infusions of autologous peripheral-blood progenitor cells: a novel and effective method for delivering multiple we established, in this study, a safe outpatient high-dose courses of dose-intensive therapy. J Clin Oncol 1992; 10: intensity combination of the two drugs. The dose-intensity 464 473. achieved, the low rates of complications and hospitalization 14 Tepler I, Cannistra SA, Frei E III et al. Use of peripheraland the acceptable delay necessary to perform further local blood progenitor cells abrogates the myelotoxicity of repetitreatment allow us to put this regimen forward for compari- tive outpatient high-dose carboplatin and cyclophosphamide son with the current standard methods. chemotherapy. J Clin Oncol 1993; 11: 1583 1591. 15 Patrone F, Ballestrero A, Ferrando F et al. Four-step highdose sequential chemotherapy with double hematopoietic progenitor-cell Acknowledgements rescue for metastatic breast cancer. J Clin Oncol 1995; 13: 840 846. 16 Fisher B, Brown AM, Dimitrov NV et al. Two months of This work was granted in part by Amgen France, Association pour doxorubicin-cyclophosphamide with and without interval reinla Recherche contre le Cancer (Commission Grand Sud, January duction therapy compared with 6 months of cyclophospham- 95) and Fondation de France. ide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifen-nonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project References B-15. J Clin Oncol 1990; 8: 1483 1496. 17 Scinto AF, Ferraresi V, Campioni N et al. Accelerated chemotherapy with high-dose epirubicin and cyclophosphamide plus 1 Hryniuk W, Bush H. The importance of dose intensity in r-met-hug-csf in locally advanced and metastatic breast chemotherapy of metastatic breast cancer. J Clin Oncol 1984; cancer. Ann Oncol 1995; 6: 665 671. 2: 1281 1288. 18 Honkoop AH, Hoekman K, Wagstaff J et al. Dose-intensive 2 Hryniuk W, Levine MN. Analysis of dose intensity for adju- chemotherapy with doxorubicin, cyclophosphamide and GMvant chemotherapy trials in stage II breast cancer. J Clin CSF fails to improve survival of metastatic breast cancer Oncol 1986; 4: 1162 1170. patients. Ann Oncol 1996; 7: 35 39. 3 Bonadonna G, Valagussa P. Dose-response effect of adjuvant 19 Shipp MA, Neuberg D, Janicek M et al. High-dose CHOP as chemotherapy in breast cancer. New Engl J Med 1981; 304: initial therapy for patients with poor-prognosis aggressive 10 15. non-hodgkin s lymphoma: a dose-finding pilot study. J Clin 4 Wood WC, Budman DR, Korzun AH et al. Dose and dose Oncol 1995; 13: 2916 2923.

20 Brugger W, Bross KJ, Glatt M et al. Mobilization of tumor 23 Vahdat L, Raptis G, Fennelly D et al. Rapidly cycled courses cells and hematopoietic progenitor cells into peripheral blood of high-dose alkylating agents supported by filgrastim and perof patients with solid tumors. Blood 1994; 83: 636 640. ipheral blood progenitor cells in patients with metastatic breast 21 Williams SF, Gilewski T, Mick R, Bitran JD. High-dose con- cancer. Clin Cancer Res 1995; 1: 1267 1273. solidation therapy with autologous stem-cell rescue in stage 24 Piccart MJ, Bruning P, Wildiers J et al. An EORTC pilot study IV breast cancer: follow-up report. J Clin Oncol 1992; 10: of filgrastim (recombinant human granulocyte colony stimul- 1743 1774. ating factor) as support to a high dose-intensive epiadria- 22 Peters WP, Ross M, Vredenburgh JJ et al. High-dose chemo- mycin-cyclophosphamide regimen in chemotherapy-naive therapy and autologous bone marrow support as consolidation patients with locally advanced or metastatic breast cancer. Ann after standard-dose adjuvant therapy for high-risk primary bre- Oncol 1995; 6: 673 677. ast cancer. J Clin Oncol 1993; 11: 1132 1143. 203