ANTICANCER RESEARCH 26: 3143-3150 (2006)

Vinorelbine and 5- Bolus and/or Continuous Venous Infusion Plus Levofolinic Acid as Second-line for Metastatic : an Analysis of Results in Clinical Practice of the Gruppo Oncologico Italia Meridionale (GOIM)

VITTORIO GEBBIA1,2, MICHELE CARUSO3, NICOLÒ BORSELLINO4, ROSANNA AJELLO3, MARIA LINA TIRRITO5, MAURIZIO CHIARENZA6, ROBERTO VALENZA6, FRANCESCO VERDERAME7, FRANCESCA VARVARA8, ANTONIO MARRAZZO1,9, EUGENIA BAJARDI2, FRANCESCO FERRAÙ10, ROBERTO BORDONARO11 and PAOLO TRALONGO12

1Department of Experimental Oncology and Clinical Application, University of Palermo; 2Medical Oncology Unit and 9Breast Cancer Unit, La Maddalena Clinic for Cancer, Palermo; 3Medical Oncology Unit, Centro Catanese di Oncologia, Catania; 4Division of Medical Oncology, Ospedale Buccheri La Ferla, Palermo; 5Division of Medical Oncology, Casa di Cura Torina, Palermo; 6Service of Medical Oncology, Ospedale Vittorio Emanuele, Gela; 7Service of Medical Oncology, Ospedale Civile, Sciacca; 8Service of Oncology, Casa di Cura Villa dei Gerani, Trapani; 10Division of Medical Oncology, Ospedale San Vincenzo, Taormina; 11Medical Oncology Unit, Ospedale Vittorio Emanuele, Catania; 12Medical Oncology Unit, Ospedale De Maria, Avola, Italy

Abstract. Background: This retrospective study evaluated the showed a partial response (40%, 95%CL 34%-46%), for an activity and toxicity profile of a regimen of and overall response rate of 44% (95CL 39%-50%). Sixty-one 5-fluorouracil with levofolinic acid, given to a large series of patients had stable disease (21%) and 98 patients progressive patients with recurrent or refractory after disease (34%). The tumor growth control rate was 63% (95%CL first-line chemotherapy. Patients and Methods: Overall, 286 60%-71%). Patients with soft tissue metastases as the dominant evaluable patients were included in the analysis. Two disease showed the highest response rate (56%), followed by chemotherapy schedules were reviewed: a) the bolus regimen viscera (48%) and bone (33%). The difference in response rate consisted of levofolinic acid 100 mg/m2 and 5-fluorouracil between patients with dominant visceral disease versus those with 375 mg/m2, both administered i.v. on days 1,2 and 3, plus dominant bone disease was statistically significant (p=0.038). vinorelbine 25 mg/m2 i.v. bolus on days 1 and 8 every 3 weeks; b) Patients treated with the bolus schedule achieved a 40% overall the infusional regimen of levofolinic acid 100 mg/m2 given as a response rate with a 5% complete response rate, while those who 2-hour infusion, followed by 5-fluorouracil 400 mg/m2 i.v. bolus received the infusional regimen had a 48% overall response rate and by 5-fluorouracil 600 mg/m2 administered as 22-hour with a 5% complete response rate. This difference was not continuous venous infusion (c.v.i) for 2 days, plus vinorelbine i.v. statistically significant (p=0.164). The overall median duration bolus on days 1 and 8. Results: Overall, twelve patients achieved of objective responses was 8.3 months (range 4-14 months), a complete response (4%; 95%CL 2%-7%) and 115 patients median time to progression of the all series was 6.1 months (range 2-24 months) and the median overall survival was 14.6 months (range 3-32). There was a statistically significant difference in survival among responder and non-responder Correspondence to: Vittorio Gebbia, MD, Ph.D., Medical Oncology patients (p=0.0009). Conclusion: The results of this large off- Unit, Oncological Department La Maddalena, Via San Lorenzo trial analysis confirmed the clinical activity and adverse-event Colli n. 312d, 90100 Palermo, Italy. Tel: ++39-091-6806710, Fax: ++39-091-6806906, e-mail: [email protected] profile reported in controlled clinical trials of the vinorelbine/ 5-fluorouracil with levofolinic acid regimen in clinical practice. Key Words: Breast cancer, metastases, second-line chemotherapy. This combination regimen was active with a low toxicity burden

0250-7005/2006 $2.00+.40 3143 ANTICANCER RESEARCH 26: 3143-3150 (2006) and, therefore, represents a good therapeutic choice for patients all patients treated in twelve oncology units across Sicily in the last who require second-line chemotherapy. 5 years, from January 1999 until December 2003. Anonymous clinical data from the medical records were collected centrally at Despite the proven benefit of adjuvant systemic therapy in the Clinical Research Unit of the Medical Oncology Unit, La Maddalena Clinic for Cancer, Palermo, Italy. Adequately staged reducing the risk of recurrence (1, 2), a significant number patients were considered for the analysis. Patients lacking clinico- of patients with operable breast carcinoma will eventually radiological evidence of response and/or certain data for time- develop metastatic disease, and ultimately die of advanced related parameters were excluded. The clinical data for all patients metastatic breast carcinoma (MBC) (3). Although MBC is were submitted for external review. sensitive to systemic first-line chemotherapy (CT), with response rates ranging from 40-75%, it still remains an Eligibility criteria. Patients were eligible for the analysis if they incurable disease using current therapeutic approaches, fulfilled the following criteria: histologically confirmed MBC while the role played by systemic chemotherapy is still pretreated with one CT regimen for advanced disease, age >18 years, an ECOG performance status (PS) ≤2, adequate mainly palliative (3, 4). hematological (WBC ≥3,500/mmc, PLT ≥100,000/mmc), renal After first-line chemotherapy, many MBC patients have (BUN <50 mg%, serum creatinine <1,2 mg%) and liver (serum good performance status and may benefit from further bilirubin <1.2 mg%, serum transaminases within 2 times the medical treatment (5). In clinical practice, most patients normal values) functions. Previous hormonotherapy and adjuvant with MBC, refractory or recurrent after first-line CT with and/or were allowed. All patients chemotherapy, receive second- and even multiple-line were also required to have measurable disease according to World treatments (6, 7). Due to the palliative nature of these Health Organisation (30); absence of clinically detectable CNS metastatic involvement; no prior CT with VNR and/or 5FU for treatments, medical oncologists are inclined to use effective advanced disease; no severe and uncontrolled metabolic, infectious, but well-tolerated regimens, with the aim of controlling the cardiological or neurological diseases. Patients with osteoblastic disease while maintaining the patient’s quality of life. bone lesion or ascites as the sole site of disease were excluded. In the last decade, among the anti-neoplastic drugs Patients were not included if they had a past or current history of employed in clinical practice, the semi-synthetic neoplasm other than MBC, except for basal cell skin cancer or vinorelbine (VNR) has shown a good therapeutic curatively-treated carcinoma in situ of the cervix. index (8). Single-agent VNR was proven to be very active in As in previous studies, the patients were classified as -sensitive or -resistant/ refractory based on the MBC, yielding a response rate of approximately 15-20% following definitions: refractory (progressive disease during when given as second-line treatment (9, 10). Several trials neoadjuvant or adjuvant CT, or progressive disease as the best also established the efficacy, in the same clinical setting, of response to palliative CT); resistant (relapse within the 12 months 5-fluorouracil (5FU), administered both as bolus and following either adjuvant or neoadjuvant CT, or disease progression continuous venous infusion, with or without folinic acid (FA) on palliative CT after initial response); potentially sensitive (relapse (11-15). 5FU/FA has also been successfully combined with more than 12 months after adjuvant treatment or disease progression other agents such as anthracyclines, oxazaphosphorines and more than 4 weeks after the end of palliative CT (5). platinum (16, 17). Chemotherapy schedules. Two schedules of the VNR/5FU/FA Based on the above-reported data, several studies have regimen were reviewed. The bolus regimen consisted of levofolinic tested the combination of VNR and 5FU, with or without acid (lFA) 100 mg/m2 diluted in 100 cc of normal saline and 5FU FA as second-line regimen, for MBC and shown substantial 375 mg/m2 diluted in 250 cc of 5% dextrose, both administered i.v. activity and acceptable toxicity with the main side-effects on days 1, 2 and 3, plus VNR given as an i.v. bolus of 25 mg/m2 on being neutropenia and mucosal toxicity (18-27). This days 1 and 8. This regimen was recycled every 3 weeks if there had regimen has also been employed as front-line treatment for been no significant toxicity. The infusional regimen consisted of 2 MBC, alone or in combination with FA, with a 60% overall lFA 100 mg/m diluted in 500 cc of normal saline and given i.v. as a 2-hour infusion, followed by 5FU 400 mg/m2 i.v. bolus and by response rate (28, 29). 5FU 600 mg/m2 administered as a 22-hour continuous venous In this retrospective analysis, the clinical efficacy and infusion (c.v.i.) for 2 days; VNR was given as an i.v. bolus on days safety of the VNR plus 5FU/FA regimen, given in clinical 1 and 8. The data concerning hematological toxicity were inferred practice as second-line treatment in patients with MBC, who from hemochromocytometric analysis, routinly done before every had relapsed after anthracycline and/or -based first- CT administration. line CT, are reported. Patient evaluation. Objective response evaluation was carried out Patients and Methods according to the WHO criteria (30), so that complete response (CR) was defined as the disappearance of all clinical, radiological Study design. Patients with recurrent or refractory MBC after first- and laboratory evidence of disease for at least 4 weeks; partial line CT, and treated with the VNR and 5FU with FA regimen as response (PR) was defined as a ≥50% reduction in the sum of the second-line therapy, were retrospectically collected and analyzed product of the largest perpendicular diameters of indicator lesions for clinical efficacy and toxicity. This retrospective analysis involved for at least 4 weeks; stable disease (SD) was defined as a <50%

3144 Gebbia et al: Second-line Chemotherapy for Metastatic Breast Cancer decrease or a <25% increase in the size of tumor deposits; and disease. One hundred and ninety-seven patients (69%) had progressive disease (PD) was defined as a >25% increase in the dominant visceral metastatic disease, 73 (25%) patients had size of indicator lesions or the appearance of any new lesions. dominant bone disease and 16 patients (6%) dominant soft- Patients were first evaluated after 3 to 4 cycles of CT. However, tissue disease. There were also three patients affected by patients with clinically PD or who had died before radiological evaluation were considered PD according to an intent-to-treat inflammatory breast cancer. The majority of patients (69%) analysis. CR plus PR were also reported as the overall response had multiple sites of disease. The median number of organs rate (ORR), while major responses plus SD were reported as involved was two. The clinical characteristics of the patients tumor growth control rate (TGCR). The response duration was according to the two CT schedules employed are also depicted calculated from the start of CT until the date when progression was in Table I. The baseline characteristics of the patients evidenced, or last follow-up evaluation or death. Time to appeared to be well balanced in the two series. progression (TTP) was calculated from the date of the first CT cycle until disease progression. Overall survival (OS) was calculated Clinical efficacy. The objective response rates, TTP and OS from the first day of CT to the time of death or last follow-up evaluation. Dose intensity was calculated by dividing the amount are reported in Table II. Five treatment drops-outs occurred of drug (mg/m2) by the time expressed in weeks between the first before re-evaluation of disease because of prolonged life- and the last day of CT administration (31). threatening hematological toxicity after two cycles (two patients), CT-related arrhythmia (one patient) and toxicity- Statistical methods. Objective responses were evaluated as relative unrelated refusal after the first cycle (two patients). These rates with their 95% confidence limits (95% CL). A two-sided Chi- cases were considered as treatment failures on intent-to- square analysis was applied for comparison of the rates. All percentages were adjusted to the nearest unit. A univariate analysis treat analysis. Overall, twelve patients achieved a CR (4%; of survival data according to the product-limit estimate (Kaplan- 95%CL 2%-7%), 115 patients showed a PR (40%, 95%CL Meier) was performed. The analysis of survival distribution was 34%-46%) for an ORR of 44% (95CL 39%-50%). done using the log-rank test. Moreover, 61 patients were classed as SD (21%) and 98 patients as PD (34%). The TGCR was 63% (95%CL 60%- Results 71%). When the responses were analyzed according to the dominant site of disease, soft tissue metastases showed the Patient population. Between January 1998 and December highest response rate (56%), followed by viscera (48%), 2003, 357 patients were enrolled in twelve institutions. while lower results were observed in patients with bone as However, 286 out of the 357 patients (80%) were the dominant site of disease (33%). The difference in considered evaluable for the analysis and 89 patients (20%) response rate between patients with dominant visceral were excluded from the analysis because of incomplete data disease versus those with dominant bone disease was on objective response and survival for reasons other than statistically significant (p=0.038). Although this was not a CT-related toxicity or disease-related events. The baseline randomized schedule comparative study, the response rates characteristics of the 286 treated patients are shown in achieved with different schedules are summarized in Table Table I. The median age was 56 years, 81% of patients had II. Briefly, patients treated with the bolus schedule achieved a PS of 1 and 220 patients (77%) were postmenopausal. The a 40% ORR with a 5% CR rate, while those who received hormonal receptor status was positive in 166 patients (58%), the infusional regimen had a 48% ORR with a 5% CR rate. negative in 97 patients (34%) and unknown in the Although the analysis showed an advantage of the infusional remaining 23 patients (8%). Pretreatments had included regimen over the bolus regimen, this difference was not surgery in 97% of patients, adjuvant radiotherapy in 63% of statistically significant (p=0.164). The median duration of cases, adjuvant CT in 84% of patients, of whom 53% had objective responses was 8.3 months (range 4-14 months). anthracycline-based regimens, and adjuvant hormonal When objective responses were evaluated according to therapy in 66% of patients. Moreover, 103 patients (36%) the previous treatments, patients not pretreated with an had had hormonal therapy for advanced disease and 34 anthracycline-containing regimen showed a statistically patients (12%) also had radiotherapy. All patients had had significant higher response rate as compared to those CT for MBC: taxane plus anthracyclines (41%), taxanes plus pretreated with anthracycline (56% versus 40%; p=0.027). (15%), taxanes plus herceptin (9%), or other No statistically significant difference in ORR was seen anthracycline-based regimens (35%). between the anthracycline-resistant patients and the All patients had hormone-insensitive or -refractory MBC anthracycline-refractory ones (p=0.124). and 216 patients (76%) had received prior treatment. The median TTP of the all patient series was 6.1 months According to the definitions for anthracycline (range 2-24 months). When only the anthracycline- sensitivity/resistance, 90 patients (29%) had anthracycline- resistant/refractory patients were taken into account, the refractory MBC, 126 (39%) patients had anthracycline- median TTP was 5.9 months. The median OS was 14.6 resistant disease and 70 had potentially anthracycline-sensitive months for the whole group of patients (range 3-32), while

3145 ANTICANCER RESEARCH 26: 3143-3150 (2006)

Table I. Patient characteristics (*).

VNR+5FU/FA

All patients Bolus Infusional

No. of patients 286 (100%) 92 (100%) 194 (100%)

Median age (years; range) 56 (28-78) 57 (35-72) 56 (28-78)

Median ECOG performance status (range) 1 (0-2) 1 (0-2) 1 (0-2)

Histology Ductal infiltrating 267 (93%) 86 (94%) 181 (93%) Lobular 11 (4%) 4 (4%) 7 (4%) Nos 8 (3%) 2 (2%) 6 (3%)

Menopausal status Premenopausal 66 (23%) 20 (22%) 46 (24%) Postmenopausal 220 (77%) 72 (78%) 154 (76%)

Receptor status Positive 166 (58%) 57 (61%) 107 (56%) Negative 97 (34%) 29 (32%) 68 (35%) Unknown 23 (8%) 6 (7%) 17 (9%)

Previous treatments Surgery 274 (96%) 88 (96%) 186 (96%) Adjuvant radiotherapy 179 (63%) 61 (66%) 118 (61%) Adjuvant hormonotherapy 189 (66%) 64 (70%) 125 (64%) Adjuvant chemotherapy 240 (84%) 75 (82%) 165 (85%) None 46 (16%) 17 (18%) 29 (15%) CMF 86 (30%) 25 (27%) 61 (32%) Antracycline-based 140 (49%) 47 (51%) 93 (48%) Taxane-based 14 (5%) 3 (3%) 11 (6%) Radiotherapy for advanced disease 34 (12%) 9 (10%) 25 (13%) Hormonotherapy for advanced disease 103 (36%) 28 (30%) 75 (37%) Chemotherapy for advanced disease 286 (100%) 92 (100%) 194 (100%) Taxane + anthracycline 116 (41%) 37 (40%) 79 (41%) Taxane + gemcitabine 42 (15%) 8 (9%) 36 (18%) Taxane + herceptin 28 (9%) 10 (11%) 18 (9%) Anthracycline-based(**) 100 (35%) 27 (29%) 73 (38%) Anthracycline-refractory 90 (29%) 30 (33%) 50 (28%) Anthracycline-resistant 126 (39%) 34 (37%) 102 (52%)

Sites of disease Bone 100 (35%) 32 (35%) 68 (35%) Lung 83 (29%) 25 (27%) 58 (30%) 80 (28%) 27 (29%) 53 (27%) Node 63 (22%) 20 (22%) 41 (21%) Pleura 26 (9%) 3 (3%) 23 (12%) Skin 37 (13%) 9 (10%) 28 (14%) Breast 34 (12%) 10 (11%) 24 (12%)

Dominant site of disease viscera 197 (69%) 63 (68%) 134 (69%) bone 73 (25%) 24 (26%) 49 (25%) soft tissue 16 (6%) 7 (8%) 9 (5%)

Number of involved sites single 87 (31%) 26 (28%) 61 (31%) multiple 199 (69%) 65 (71%) 134 (69%)

*All percentages refer to total number of patients. **Regimens other than taxanes/anthracycline.

3146 Gebbia et al: Second-line Chemotherapy for Metastatic Breast Cancer

Table II. Objective response rates according to chemotherapy schedule, site of dominant disease and previous anthracycline treatment.

Type of objective response Patient subgroups CR PR ORR SD PD TGCR

All patients 12 (4%) 115 (40%) 127 (44%) 61 (21%) 98 (34%) 188 (63%)

Bolus FA/5FU+VNR 3 (3%) 32 (37%) 35 (40%) 16 (17%) 41 (45%) 51 (55%) (no.=92)

Infusional FA/5FU+VNR 9 (5%) 83 (43%) 91 (48%) 35 (23%) 68 (35%) 126 (65%) (no.=194)

Visceral dominant disease (**) 9 (5%) 85 (43%) 94 (48%) 44(22%) 59 (30%) 138 (70%) (no.=197)

Bone dominant disease (*) 0 24 (33%) 24 (33%) 14 (20%) 35 (48%) 38 (52%) (no.=73)

Soft-tissue dominant disease (*) 3 (19%) 6 (37%) 9 (56%) 3 (19%) 4 (25%) 12 (75%) (no.=16)

Non-anthracycline pretreated (*) 6 (9%) 33 (47%) 39 (56%) 11 (16%) 20 (29%) 50 (71%) (no.=70)

Anthracycline pretreated 6 (3%) 82 (38%) 87 (40%) 38 (18%) 91 (42%) 125 (58%) (no.=216)

Anthracycline-refractory (*) 2 (2%) 29 (34%) 31 (34%) 16 (18%) 43 (48%) 47 (52%) (no.=90)

Anthracycline-resistant (*) 4 (3%) 53 (42%) 57 (45%) 28 (22%) 41 (33%) 85 (67%) (no.126)

CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; TGCR, tumor growth control rate. (**) All series of patients.

it was 19 months for PR patients. There was a statistically Table III. Main grade 3-4 side-effects. significant difference in survival among the responder and non-responder patients (p=0.0009). NCI-CTC Grade (Number of patients and percentage)

Safety. Of the 286 patients analyzed, 103 patients withdrew Toxicity Bolus schedule Infusional schedule from treatment before completing six cycles of CT. The majority of patients (34%) stopped CT because of PD or Grade 1-2 Grade 3-4 Grade 1-2 Grade 3-4 cancer-related death. Less than 2% of patients (n=5) were withdrawn for severe toxicity, but all received only VNR or 16 (17%) 5 (6%) 30 (15%) 3 (2%) FA/5FU depending on each physician’s decision. Asthenia 14 (15%) 3 (3%) 19 (10%) 4 (3%) (*) 15 (16%) 24 (26%) 57 (29%) 4 (3%) Toxic effects, recorded according to the National Cancer Neutropenia (*) 27 (29%) 32 (35%) 74 (38%) 10 (5%) Institute - Common Toxicity Criteria, are listed in Table III. Neuritis (peripheral) 1 (1%) – 3 (7%) – In total, 1323 cycles were administered with a mean of 4.6 Paraesthesia 3 (3%) – 10 (5%) – cycles/patient (range 2-8). Toxicity-related dose reductions 16 (15%) 2 (2%) 21 (11%) 1 (1%) were made in 12% of cycles. The median interval between Stomatitis (*) 18 (20%) 32 (35%) 35 (18%) 14 (7%) courses was 23.6 days. Overall, delays longer than 1 week Thrombocytopenia 15 (14%) 6 (7%) 35 (18%) 7 (4%) occurred in 25% of cycles in the bolus group and in 9% of Vomiting 29 (31%) 4 (4%) 36 (18%) 3 (2%) cycles in the infusional group. The planned dose-intensity of VNR was 16.7 mg/m2/week, (*) The differences in reported rates between the bolus and infusional while that of bolus and infusional 5FU were 267 and 400 schedules are statistically significant (p<0.0001).

3147 ANTICANCER RESEARCH 26: 3143-3150 (2006)

Table IV. Main data from medical literature concerning vinorelbine/5-fluorouracil regimens in pretreated patients.

Authors No. of pts Schedule ORR CRR Duration TTP OS (reference) (months)

Zambetti et al. (18) 28 5FU 700 mg/m2 d 1→5 61% 14% 8 – 12 VNR 20 mg/m2 d 1,5 q28d

Froudarakis et al. (19) 20 5FU 300 mg/m2 CVI d 1→3 43% 6% – 8 12 VNR 30 mg/m2 d 1

Kornek et al. (20) 16 5FU 400 mg/m2 d1→5 19% 0 – 7 14 VNR 40 mg/m2 d 1,14 q28d (*)

Berruti A et al. (21) 83 5FU 200 mg/m2 PIF 61% 7% – 8.8 22.3 VNR 30 mg/m2 d 1,15 q28d

Lombardi et al. (22) 83 5FU 200 mg/m2 PIF 45% 15% – 9 20 VNR 20 mg/m2 d 1,8 q28d

Zambelli et al. (23) 48 5FU 1000 mg/m2 CVI 4 d 50% 8% – 9 16 VNR 20 mg/m2 d 1,5 q21d

Zambetti et al. (24) 56 5FU 700 mg/m2 CVI 5 d 48% 12% 8 – 17 VNR 20 mg/m2 d 1,6 q21d

Pienkowski et al. (25) 65 5FU 700 mg/m2 d 1->5 47% 2% – 4 10.3 VNR 25 mg/m2 d 1,6 q28d

Boneterre et al. (26) 90 5FU 750 mg/m2 CVI 5 d 39% 4% – 5.1 15 VNR 25 mg/m2 d 1,5 q21d

Razis et al. (27) 35 5FU 600 mg/m2 CVI d1,15 23% 0 – – 12.3 VNR 25 mg/m2 d 1,15

ORR, overall response rate; CRR, complete response rate; TTP, time to progression; OS, overall survival; CVI, continuous venous infusion; PIF, prolonged infusion. (*) plus LV 100 mg/m2 d 1 → 5 and G-CSF .

mg/m2/week, respectively. The given relative median dose- than the bolus schedule. These differences were statistically intensity of VNR was 0.89 and that of 5FU 0.90 for both significant (p<0.0001). Only one case of grade 1 cardiac bolus and infusional administrations. Statistical analysis to toxicity was recorded: it occurred in a patient with correlate the response rate with the CT dose-intensity was inflammatory breast cancer who underwent, after four cycles not significant. of chemotherapy, left radical mastectomy plus radiotherapy No CT-related deaths were observed. Treatment on the chest wall. Hand-foot syndrome was recorded in four discontinuation due to toxicity was less than 2%. The main patients who received the infusional schedule, but this side- side-effects are reported in Table III, according to grade effect was generally mild. Vascular peripheral toxicity was and CT schedules. In the whole series with regard to not evaluated, since it strongly depended on the lack of use hematological toxicity, grade 3-4 neutropenia was recorded of a central vein access. in 15%, grade 3-4 thrombocytopenia in 7%, and grade 3-4 anemia in 3% of cases. Eighteen patients were hospitalized Discussion for febrile neutropenia (6%) and 4% required packed red blood cell transfusions. Second-line CT for patients with MBC has a mainly The most frequently reported non-hematological grade palliative role (1, 4). Therefore, the search and/or 3-4 side-effects were as follows: stomatitis in 42 patients optimization of active chemotherapeutic regimens with a (16%), vomiting in seven (2%) and diarrhea in 28 cases good tolerability is a major end-point of clinical research in (10%). The infusional arm was less frequently associated medical oncology. The combination of two drugs, such as with severe grade 3-4 stomatitis, diarrhea and neutropenia vinorelbine and 5-FU, without overlapping toxicity profiles,

3148 Gebbia et al: Second-line Chemotherapy for Metastatic Breast Cancer seems particularly attractive in this clinical setting. The intensity. Although there was no statistically significant purpose of this retrospective multicentric analysis was to difference in ORR between the two schedules, the report the efficacy and toxicity profile, in daily clinical tolerability of the infusional 5FU schedule was better than practice, of the combination of VNR and bolus and/or that observed with the bolus regimen, especially regarding infusional 5FU with FA in a series of 286 heavily-pretreated mucosal and hematological side-effects. In total, the toxicity patients with MBC who had failed or had progressed after reported in this analysis with the infusional 5FU schedule previous first-line CT. seemed to be less than that reported in clinical trials with In the whole series of 286 patients, the combination of other schedules of the same combination regimen. These VNR and 5FU plus FA yielded a 44% ORR (95% CL 39%- observations are in line with data reported by other authors 50%) with a TGGR of 63%. The ORR was higher in (21, 22). Whatever the case, it is mandatory to carefully patients previously untreated with anthracycline-based evaluate the possible benefits and expected toxicity before regimens than in those pretreated with anthracyclines (56% giving a second-line CT to any patient with progressive versus 40%). This difference was statistically significant MBC after first-line treatment. In this evaluation, it should (p=0.027). Among patients pretreated with anthracycline- be considered that achievement of an objective response is based regimens, the ORR was 34% in anthracycline- considered by many oncologists to be a good surrogate refractory patients and 45% in anthracycline-resistant ones. marker of quality of life, despite acute and chronic This difference was not statistically significant. The treatment-related side-effects (32-34). A relationship seems infusional 5FU regimen yielded a higher ORR than the to exist between the achievement of a major objective bolus schedule, however this difference did not reach response and symptom improvement, even if the tumor statistical significance (40% versus 48%). Therefore, the response is more likely to positively affect the quality of life VNR/5FU/FA regimen also showed good activity in terms in a symptomatic patient (34). of response rate in this "real" population of patients with In conclusion, the results of this large off-trial analysis MBC. confirmed the clinical activity and adverse-event profile The efficacy of VNR plus 5FU as second-line treatment reported in controlled clinical trials of the VNR/5FU (+/– for MBC patients has been reported by several authors in FA) regimen in clinical practice. This combination regimen the medical literature in the last decade. A summary of the is active with a low toxicity burden and, therefore, main trials dealing with VNR/5FU with or without FA as represents a good therapeutic choice for patients who would second-line treatment is depicted in Table IV. Overall, all benefit from a second-line CT. these trials reported a median ORR of 46% (range 19% - 61%) with a median CR rate of 7% (range 0% - 15%). The References results reported in this large clinical practice analysis are consistent with the above reported figures, while the good 1 Harris JR, Morrow M and Bonadonna G: Cancer of the breast. activity shown by this combination regimen is further In: Cancer. Principles and Practice of Oncology, Fourth edition. supported by some phase II trials carried out by other De Vita VT Jr, Hellman S, Rosenberg SA (eds.). Philadelphia, PA: Lippincott Company, pp. 1264-1332, 1993. authors, with the same regimen as the first-line CT for the 2 Sledge GW: Breast cancer as a world challenge. Clin Breast MBC patients (28, 29). However, the toxicity profile seemed Cancer 5: 11, 2004. to vary significantly according to the schedule employed and 3 Early Breast Cancer Trialist’s Collaborative Group: Systemic patient selection. In some trials on pre-treated MBC treatment of early breast cancer by hormonal, cytotoxic, or patients, the VNR plus 5FU regimen exhibited substantial immune therapy. Lancet 339: 1-15, 1992. activity and acceptable tolerability, with the main toxic 4 Sledge GW: First-line chemotherapy for advanced breast effect being severe grade 3-4 neutropenia and mucositis in cancer. Cancer Control 6(Suppl 2): 17-21, 1999. less than 20% of patients (20, 22, 24, 25), or even less than 5 Nabholtz JM, Senn HJ, Bezwoda WR, Melnychuk D, Deschenes L, Douma J, Vandenberg TA, Rapoport B, Rosso 10% (21, 27). In other phase II studies, the schedule of R, Trillet-Lenoir V, Drbal J, Molino A, Nortier JW, Richel DJ, 2 2 VNR 30 mg/m on days 1 and 5 plus 5FU 750 mg/m /day Nagykalnai T, Siedlecki P, Wilking N, Genot JY, Hupperets PS, c.v.i. for 5 days as first-line treatment was associated with Pannuti F, Skarlos D, Tomiak EM, Murawsky M, Alakl M, Riva severe grade 3-4 neutropenia in 67%-90% of patients, grade A and Aapro M: Prospective randomized trial of 3-4 mucositis in 37% - 40%, infection in 13% and versus mitomycin plus in patients with metastatic constipation in 9% (26, 28). breast cancer progressing despite previous anthracycline- In our experience, the toxicity profile seems to be containing chemotherapy. 304 Study Group. J Clin Oncol 17: 1413-1424, 1999. particularly attractive, since the rate of severe grade 3-4 6 Cardoso F, Di Leo A, Lohrisch C, Bernard C, Ferreira F and toxicity was quite low in the whole series. Side-effects Piccart MJ: Second and subsequent lines of chemotherapy for usually related to 5FU, such as mucositis and hand-foot metastatic breast cancer: what did we learn in the last two syndrome, were rarely recorded and were generally mild in decades? Ann Oncol 13: 197-207, 2002.

3149 ANTICANCER RESEARCH 26: 3143-3150 (2006)

7 O’Brien MER, Leonard RC, Barrett-Lee PJ, Eggleton SPH and 21 Berruti A, Sperone P, Bottini A, Gorzegno G, Lorusso V, Bizzari JP: Docetaxel in the community setting: an analysis of 377 Brunelli A, Botta M, Tampellini M, Donadio M, Bancarella S, De breast cancer patients treated with docetaxel (Taxotere) in the lena M, Alquanti P and Dogliotti L: Phase II study of vinorelbine UK. Ann Oncol 10: 205-210, 1999. with protracted fluorouracil infusion as a second- or third-line 8 Gebbia V and Puozzo C: Oral versus intravenous vinorelbine: approach for advanced breast carcinoma patients previously clinical safety profile. Exp Opin Drug Safety 4: 915-928, 2005. treated with anthracyclines. J Clin Oncol 18: 3370-3377, 2000. 9 Degardin M, Bonneterre J, Hecquet B, Pion JM, Adenis A, 22 Lombardi D, Magri MD, Crivellari D, Spazzapan S, Paolello C, Horner D and Demaille A: Vinorelbine (navelbine) as a De Cicco M, Di Lauro V, Scuderi C and Veronesi A: salvage treatment for advanced breast cancer. Ann Oncol 5: Combination chemotherapy with navelbine and continuous 423-426, 1994. infusion of 5-fluorouracil in metastatic, chemotherapy refractory 10 Gasparini G, Caffo O, Barni S, Frontini L, Testolin A, breast cancer. Ann Oncol 11: 1041-1043, 2000. Guglielmi RB and Ambrosiani G: Vinorelbine is an active 23 Zambelli A, Robustelli della Cuna FS, Ponchio L, Ucci G, da antiproliferative agent in pretreated advanced breast cancer Prada O and Robustelli della Cuna G: Four-day infusion of patients: a phase II study. J Clin Oncol 12: 2094-2101, 1994. fluorouracil plus vinorelbine as salvage treatment of heavily 11 Margolin KA, Doroshow JH, Akman SA et al: Effective initial pretreated metastatic breast cancer. Breast Cancer Res Treat 61: therapy of advanced breast cancer with fluorouracil and high 241-247, 2000. dose continuous infusion calcium leucovorin. J Clin Oncol 10: 24 Zambetti M, Mariani G, Demicheli R, Verderio P, Potepan P and 1278-1283, 1992. Garbagnati F: Five-day infusion fluorouracil plus vinorelbine in 12 Swain SM, Lippman ME, Egan EF et al: Fluorouracil and high- women with breast cancer previously treated with anthracycline dose leucovorin in previously treated patients with metastatic and . Breast Cancer Res Treat 62: 135-139, 2000. breast cancer. J Clin Oncol 7: 890-899, 1989. 25 Pienkowski T and Jagiello-Gruszfeld A: Five-day infusion of 13 Doroshow JH, Leong L, Margolin K et al: Refractory metastatic fluorouracil and vinorelbine for advanced breast cancer patients breast cancer: salvage therapy with fluorouracil and high-dose treated previously with anthracyclines. Int J Clin Pharmacol Res continuous-infusion leucovorin calcium. J Clin Oncol 7: 439- 21: 111-117, 2001. 444, 1989. 26 Boneterre J, Rochè H, Monnier A, Guastalla JP, Namer M, 14 Marini G, Simoncini E, Zaniboni A et al: 5-Fluorouracil and Fargeot P and Assadourian S: Docetaxel vs. 5-fluorouracil plus high-dose folinic acid as salvage treatment of advanced breast vinorelbine in metastatic breast cancer after anthracycline therapy cancer: an update. Oncology 44: 336-340, 1987. failure. Br J Cancer 87: 1210-1215, 2002. 15 Gebbia V, Mauceri G, Testa A, Tirrito ML, Varvara F, 27 Razis E, Kosmidis P, Aravantinos G, Bakoyiannis C, Janinis J, Cucchiara A, Borsellino N, Girlando A, Ferrara M and Caruso Christodoulou C and Foutzilas G: Second line chemotherapy with F: Treatment of refractory metastatic breast cancer with 5- 5-fluorouracil and vinorelbine in anthracycline and taxane fluorouracil and levofolinic acid as 48 hours continuous venous pretreated patients with metastatic breast cancer. Cancer Invest infusion. Anticancer Res 19: 2289-2292, 1999. 22: 10-15, 2004. 16 Colucci G, Romito S, Gebbia V, Pacilio G, Giotta F, Testa A, 28 Dieras V, Extra JM, Bellissant E, Espie M, Morvan F, Pierga JY, Pezzella G, Durini E, Agostara B, Cariello S, Palazzo S and Mignot L, Tresca P and Marty M: Efficacy and tolerance of Gebbia N: Levofolinic acid, 5-fluorouracil, vinorelbine and fluorouracil combination as first-line and escalating doses of with granulocyte colony- chemotherapy of advanced breast cancer: results of a phase II stimulating factor support in locally advanced and/or study using a sequential group method. J Clin Oncol 14: 3097- metastatic breast carcinoma: a phase I-II study of the 3104, 1996. Southern Italy Oncology Group (GOIM). Br J Cancer 72: 29 Nolè F, deBraud F, Aapro M, Minchella I, De pas M, Zampino 1245-1250, 1995. MG, Monti S, Androni G and Goldhirsch A: Phase I-II study of 17 Ray-Coquard I, Biron P, Bachelot T, Guastalla JP, Catimel G, vinorelbine in combination with 5-fluorouracil and folinic acid as Merrouche Y, Droz JP, Chauvin F and Blay JY: Vinorelbine first-line chemotherapy in metastatic breast cancer : a regimen and (CIVIC regimen) for the treatment of metastatic with a low subjective toxic burden. Ann Oncol 8: 865-870, 1997. breast carcinoma after failure of anthracycline- and/or 30 Miller AB, Hoogstraten B, Staquet M and Winkler A: Reporting paclitaxel-containing regimens. Cancer 82: 134-140, 1998. results of cancer treatment. Cancer 47: 207-214, 1981. 18 Zambetti M, Demicheli R, De candis D, Antonelli G, 31 Hryniuk WM: Average relative dose intensity and the impact on Giacobone A, Terenziani M, Laffranchi A, Garbagnati F, Biasi design of clinical trials. Sem Oncol 14: 65-74, 1987. S and Bonadonna G: Five-day infusion fluorouracil plus 32 Carlson RW: Quality of life issues in the treatment of metastatic vinorelbine iv in metastatic pretreated breast cancer patients. breast cancer. Oncology 12(Suppl 4): 27-31, 1998. Breast Cancer Res Treat 44: 255-260, 1997. 33 ASCO: Outcomes of cancer treatment for technology assessment 19 Froudarakis ME, Catimel G, Guastalla JP, Rebattu P and and cancer treatment guidelines. American Society of Clinical Clavel M: Phase II trial of navelbine and fluorouracil as second- Oncology. J Clin Oncol 14: 671-679, 1996. line chemotherapy in metastatic breast carcinoma. Oncology 55: 34 Geels P, Eisenhaur E, Bezjak A et al: Palliative effect of 87-88, 1998. chemotherapy: objective tumor response is associated with 20 Kornek GV, Haider K, Kwasny W, Lang F, Krauss G, Hejna M, symptom improvement in patients with metastatic breast cancer. Radiz M, Weinlander G, Depisch D and Scheithauer W: J Clin Oncol 18: 2395-2405, 2000. Effective treatment of advanced breast cancer with vinorelbine, 5-fluorouracil and l-leucovorin plus human granulocyte colony Received March 20, 2006 stimulating factor. Br J Cancer 78: 673-678, 1998. Accepted April 27, 2006

3150