1IDBBC, Brussels, Belgium; 2Weston Park Hospital, Sheffield, UK; 3Fondation Bergonié, Bordeaux, France; 4Institut Jules Bordet, Brussels, Belgium; 5Academisch Ziekenhuis Leiden, Leiden, The Netherlands; 6Universiteit Ziekenhuis Nijmegen, Nijmegen, The Netherlands; 7Antoni van Leeuwenhoekhuis, Amsterdam, The Netherlands
Received 26 November 2001; accepted 19 December 2001.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We performed a phase I study of polyethylene glycol (pegylated, Stealth®) liposomal doxorubicin (Caelyx®, Doxil®) using a prolonged (6-week) dose interval to reduce the incidence of skin toxicity that was dose-limiting at more conventional dose intervals, and which appeared to be schedule dependent.
Patients and methods
Eligible for the study were metastatic breast cancer patients who had received a maximum of one prior therapy for metastatic disease. The defined dose levels were 60, 70, 80 and 90 mg/m2.
Results
Twenty patients were assessed at starting doses of 60 mg/m2 (n = 9) or 70 mg/m2 (n = 11). The dose-limiting toxicity was mucositis. Severe skin toxicity was not observed at the 60 mg/m2 dose level, and occurred in only one patient treated at 70 mg/m2. Significant neutropenia, alopecia, and nausea and vomiting were rare events. No clinical cardiac events occurred, despite a median cumulative doxorubicin dose of 323 mg/m2 (range 5630 mg/m2). Partial responses were documented in five patients. Pharmacokinetics were assessed in 15 patients, and confirmed the long terminal half-life of the agent (median 77 h) demonstrated in earlier studies.
Conclusions
The recommended dose of Caelyx®/Doxil® using this schedule is 60 mg/m2 every 6 weeks. This is a safe and effective regimen that permits prolonged administration of anthracycline to patients with metastatic breast cancer.
Key words: breast cancer, liposomal doxorubicin, schedule
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The activity of pegylated liposomal doxorubicin as a single agent has been demonstrated in Kaposis sarcoma (20 mg/m2 every 3 weeks) [8], in ovarian cancer (4050 mg/m2 every 3 weeks) [11, 12] and in breast cancer (45 mg/m2 every 4 weeks) [7]. The absolute doses of doxorubicin that can be delivered per cycle in the liposomal formulation are significantly lower than those of the free drug. Despite this, liposomal encapsulation does not appear to impact adversely on the anti-tumour activity of doxorubicin, probably due to the described advantages in pharmacokinetics and tissue distribution. In breast cancer, the potential benefits of liposomal doxorubicin over free doxorubicin are not insignificant. Alopecia is a major psychosocial issue for many women receiving chemotherapy. Patients receiving anthracyclines for metastatic disease often cease therapy while still in response because of a reduction in left ventricular ejection fraction, or concerns relating to cardiac toxicity at cumulative doses >450 mg/m2 [13], and elderly patients are rarely offered anthracyclines at all, again because of concerns regarding cardiac toxicity. Pegylated liposomal doxorubicin rarely causes alopecia, and cumulative doses of >500 mg/m2 have been administered safely [10]. This formulation therefore has the potential to offer effective, ambulatory therapy to women of all ages without hair loss or other significant toxicity, if the schedule-dependent skin toxicity and the mucositis are controlled.
We were not convinced that the optimal schedule for pegylated liposomal doxorubicin had yet been established. Indirect comparison of the toxicity profiles from the published schedules suggested that a longer dose interval might produce less skin toxicity without compromising activity. The selection of the 6-week dose interval, however, was largely pragmatic. It was longer than the 4-week interval that had already been studied, but would combine easily with drugs that required 3-weekly dosing. The potential benefits of the 6-week interval over the 4-week interval included reduced skin and mucosal toxicity, the financial and social benefits of fewer hospital visits, and the possibility of prolonging the duration of effective therapy without cardiac sequelae, particularly in elderly patients and in the metastatic setting. The potential disadvantage was that efficacy might be compromised.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Eligibility criteria
Eligibility criteria included patients 18 years of age with a histological or cytological diagnosis of locally advanced (stage IIIB) or metastatic breast cancer, with measurable disease, defined as bidimensionally measurable lesions with clearly defined margins. This was determined by: (i) chest X-ray with at least one diameter
1 cm; (ii) CT scan, MRI or other imaging scan with both diameters
2 cm (at least the distance between cuts of the imaging study); or (iii) palpation with both diameters
1 cm. Bone lesions and malignant effusions were considered non-measurable. Patients >70 years of age were required to be anthracycline-naïve and to have received a maximum of one prior chemotherapy for metastatic disease. Patients <70 years of age were not permitted to have received anthracyclines for metastatic disease. Those who had received adjuvant anthracyclines were permitted a maximum of 360 mg/m2 of prior doxorubicin (prior epirubicin dose was converted to an approximate prior doxorubicin dose by multiplication by a factor of 0.6), had an anthracycline-free interval of
12 months and had received a maximum of one prior chemotherapy for metastatic disease. Those who were anthracycline-naïve were required to have received one prior chemotherapy for metastatic disease (patients who were <70 years of age, anthracycline-naïve and had not received prior chemotherapy for metastases were preferentially offered a trial of an anthracyclinetaxane combination). In addition, patients had a performance status of 02 and adequate organ function [platelets
100 x 109/l, absolute neutrophil count
2.0 x 109/l, haemoglobin
10 g/dl, aspartate aminotransferase/alanine aminotransferase <3x normal, bilirubin <1.5x normal, serum creatinine within normal limits, left ventricular ejection fraction (LVEF) within normal limits]. Patients with any New York Heart Association grade of cardiac disease, or who were requiring medication for cardiac failure were excluded, as were those with non-malignant liver disease or a history of anthracycline hypersensitivity. Also excluded were patients with known brain metastases, uncontrolled infection, impaired mental status, or a history of another cancer other than adequately treated non-melanomatous skin cancer or carcinoma of the cervix in situ, and those who were pregnant or breast feeding. All patients provided written informed consent to participate in the study.
Therapeutic plan and toxicity evaluation
At all dose levels, patients were treated once every 6 weeks. Pegylated liposomal doxorubicin was administered on an out-patient basis as a 1 h intravenous infusion. Patients were permitted to continue therapy until disease progression or unacceptable toxicity, regardless of cumulative anthracycline dose, but six cycles were recommended.
No premedication was specified by the protocol; dexamethasone and 5-HT antagonist were used by some investigators, but these were not routinely administered and many patients received a dopamine antagonist as their only antiemetic therapy. To minimise skin toxicity, patients were instructed to wear loose clothing, to avoid activities that cause increased perfusion of the skin or mild trauma to skin surfaces, and to check skin folds, pressure points, and hands and feet daily for any localised redness or oedema. Initially, instruction in mouth care was left to the individual physician, but the protocol was amended to standardise mouth care when mucositis was observed at the 70 mg/m2 dose level; all patients then received a mouthwash of sodium bicarbonate 1/6 M 500 ml, xylocaine 2% 40 ml and nystatin 4.8 million IU (or institutional equivalent), and were instructed to administer it six times a day throughout each treatment cycle.
Eight evaluable patients were included at each dose level to permit a thorough evaluation of toxicity, an adequate sample for pharmacokinetic evaluation and a limited assessment of anti-tumour activity. If less than four patients developed a dose-limiting toxicity (DLT), the study proceeded to the next dose level. If four or more patients experienced a DLT, the maximum tolerated dose was reached and the previous dose level was defined as the recommended phase II dose (RPTD). The defined dose levels were 60, 70, 80 and 90 mg/m2.
Toxicity was assessed according to the NCIC Common Toxicity Criteria at the end of each cycle. Isotopic LVEF assessments were performed after every second cycle after a cumulative anthracycline dose of 360 mg/m2, and every cycle after 540 mg/m2. Dose-limiting toxicities were defined as any of the following occurring during the first two cycles of therapy: grade 4 neutropenia lasting >7 days or accompanied by fever 38°C, grade 4 thrombocytopenia, grade 3 or 4 non-haematological toxicity other than alopecia, nausea or vomiting, or grade 2 plantar palmar erythrodysesthesia (PPE) requiring a 2-week dose delay.
Mucosal, skin, neutrophil and platelet toxicities were required to be recovered to grade 0 or 1 before retreatment. Dose reduction by 10 mg/m2 was defined for those patients who had experienced a DLT or who were delayed by >1 week. Those patients who experienced a DLT and were delayed by >1 week were either dose reduced by 20 mg/m2 or withdrawn from study at the investigators discretion. Those who could not be retreated after a 2-week delay were withdrawn from study. Patients whose bilirubin was 1.52x normal on day 1 of a cycle were dose reduced by 10 mg/m2, and those whose bilirubin was >2x normal were withdrawn from study.
Patients were withdrawn from study if they experienced a relative fall in LVEF of 10% to an abnormal value (i.e. LVEF fell to <90% of their baseline, and to a value below the normal limit), or a fall of any magnitude to a value
5% below the normal limit (i.e. LVEF fell to <95% of the normal limit, regardless of the baseline), or congestive cardiac failure. In the absence of toxicity or progression, a total of six cycles over 9 months was planned, but continuation beyond six cycles was permitted.
Response evaluation
All patients were staged within 4 weeks of commencing therapy with chest X-ray or CT scan and liver CT scan (liver ultrasound was accepted only if negative). Bone scan was optional. Therapeutic response was assessed by UICC criteria. Disease evaluation was performed after every second cycle (i.e. every 3 months), and all responses were confirmed at least 4 weeks after the initial observation. All responses were reviewed and confirmed by the assembled study investigators and an independent radiologist. Time to progression was calculated from the first day of treatment to the date when suspicion of disease progression was first documented.
Pharmacokinetic study
Blood for pharmacokinetic analysis was collected in EDTA-containing tubes during the first cycle of therapy. Samples were taken prior to therapy (time 0), at the end of the infusion (+1 h), and at times +2 h, +5 h, +7 h, +25 h, +49 h and weekly thereafter until completion of the cycle. They were stored on ice until centrifugation, within 1 h of collection, at 3000 r.p.m. for 10 min. Plasma was divided into three aliquots and stored at 20°C until transport on dry ice to the analytical laboratory (PHARMout, Sunnyvale, CA, USA). Samples were analysed for total plasma doxorubicin using reverse phase high performance liquid chromatography with fluorescence detection. The linear range of the assay for doxorubicin was established between 0.01 and 4.0 mg/l. Based on the quality control samples (at 0.025, 0.5 and 5.0 mg/l), accuracy was <4%, while between-day precision was <4.73%.
Total plasma doxorubicin concentrationtime data were modelled with the ADAPT II program [14] using weighted, iterative, non-linear least squares regression, with weights related to assay error. Models investigated were one-, two- and three-compartment linear models, and a one-compartment model with saturable elimination. Goodness of fit was evaluated using Akaikes Information Criterion (a smaller number indicated superiority of fit) [15]. For each patient and each model tested, numerical parameter estimation was repeated 10 times with initial estimates chosen randomly and independently, in order to ascertain stability of resulting parameter estimates. Population parameters were obtained using both Bayesian iterative two-stage analysis (IT2S) [16] and non-linear mixed effects methods [17].
Methods used to detect correlations between pharmacokinetic and pharmacodynamic end points included Spearmans correlation coefficient (correcting for ties) for graded toxicities, Pearsons correlation coefficients for all toxicities, and the Hill Emax effect model for haematological effects [14, 20].
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
Toxicity
Table 3 describes the observed toxicity by allocated dose level. The dose-limiting toxicity was mucositis, experienced by 9/11 patients at the 70 mg/m2 dose level (eight at grade 3 and one at grade 4). Of those nine, one patient also experienced grade 3 vulvovaginitis, one had grade 3 fatigue and one had grade 3 plantar-palmar erythrodysesthesia. An additional patient had neutropenia for >7 days, but did not experience dose-limiting mucositis. In the initial cohort of eight patients treated at 70 mg/m2, patients who had received rigorous prophylactic mouth washes (n = 5) appeared to have experienced less severe mucositis than those who had not. Three additional patients were therefore treated with strict adherence to the mouth-wash schedule, but two of them experienced dose-limiting mucositis despite these measures.
|
In marked contrast, the 60 mg/m2 dose level was well tolerated. Only one patient experienced dose-limiting toxicity (grade 3 cellulitis in cycle 1 and grade 3 mucositis in cycle 2). Of the eight patients whose dose reduced from 70 mg/m2 in the first cycle to <60 mg/m2 in the second cycle, however, two experienced a recurrence of grade 3 mucositis in the second cycle.
Of note, alopecia, nausea and vomiting were rare. Neutropenia was also uncommon. Overall, only 16% of patients experienced grade 4 neutropenia, and only one haematological DLT was documented.
Table 4 describes serious adverse events according to the administered dose. The 60 mg/m2 dose level was generally extremely well tolerated over a full course of six cycles, without evidence of cumulative toxicity.
|
Efficacy
Table 5 describes the anti-tumour efficacy of pegylated liposomal doxorubicin using the 6-week schedule. It is important to note that the schedule of disease assessment in this study (every 2 cycles or 3 months) did not adjust for the long cycle duration. Thus, patients were required to have stable or responding disease for a minimum of 46 months to qualify as stable or responding.
|
Pharmacokinetics
Individual doxorubicin pharmacokinetic profiles were obtained for 15 patients during the first course of therapy (seven re-ceiving 60 mg/m2, eight receiving 70 mg/m2, for a total of 172 observations).
(i) Individual modelling
The one-compartment model with saturable (MichaelisMenten) elimination was unstable and did not fit the data well. The Akaikes Information Criterion value for the three-compartment linear model was smaller than that of the one- and two-compartment linear models for only two patients, but in both cases the three-compartment parameter estimates were extremely unstable in the sense that estimates resulting from different choices of initial parameter values were highly variable. Data for all patients were therefore fitted by either one-compartment (n = 9) or two-compartment linear (n = 6) models.
(ii) Population modelling
The population pharmacokinetic parameters for the one-compartment linear model, estimated by iterative two-stage method, were similar for the two subgroups (60 and 70 mg/m2) (Table 6). Therefore, the data from the two dose levels were pooled in a single population analysis. The use of Bayesian priors in the iterative two-stage method allowed us to perform population two-compartment linear modelling on the whole population (n = 15), despite the unstability of two-compartment model parameters for some individual profiles. Due to numerical difficulties, Nonmem population analysis was performed using the one-compartment linear model.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In the absence of the schedule-dependent skin toxicity, mucositis emerged as the dose-limiting toxicity. The incidence of mucositis seen with this 6-week schedule (53%) was comparable to that seen with the previously published 60 mg/m2 every 3 weeks schedule (54%). Of the patients who received 45 mg/m2 in that study, 1931% had dose-limiting mucositis. In our study, only one patient who commenced at 60 mg/m2 every 6 weeks (12.5%) experienced dose-limiting mucositis. A total of five cycles (8.3%) at this dose and schedule were associated with severe mucositis, of which two were in patients who had previously experienced this toxicity at the higher dose level. When Tables 3 and 4 are compared, it is immediately obvious that very few serious adverse events occurred after the first two cycles, indicating that dose reduction from 70 to 60 mg/m2 was generally effective in preventing further severe toxicity.
Other acute toxicities classically associated with anthracycline therapy (nausea, vomiting, alopecia) were observed far less frequently than would be expected with free doxorubicin. Some perspective on the haematological toxicity can be gleaned from comparison with other single-agent anthracycline studies performed by the EORTC. In EORTC 10923 [18], single-agent doxorubicin, at a dose of 75 mg/m2, was administered as first or second line therapy for metastatic breast cancer in anthracycline-naïve patients. In the first cycle of first-line therapy, 53% of patients experienced grade 4 neutropenia, and in the first cycle of second-line therapy the corresponding figure was 44%. Overall, 20% of patients ex-perienced febrile neutropenia while receiving free doxorubicin in that study. In contrast, no patient experienced grade 4 neutropenia in the first cycle of pegylated liposomal doxorubicin in this study; only 16% of patients reported this toxicity at any time during therapy, and febrile neutropenia did not occur, although undocumented haematological toxicity may have contributed to the toxic death.
This study was too small to contribute significantly to the published data regarding the cardiac safety of pegylated liposomal doxorubicin; however, no clinical cardiac events occurred, and two patients safely received cumulative doxorubicin doses of >500 mg/m2.
As only one patient received more than two cycles at 70 mg/m2, the efficacy results reflect almost universal therapy at 60 mg/m2. At first glance, the response rate of 25% is somewhat lower than would normally be expected with an anthracycline, however, this figure should be interpreted in the light of a number of observations. First, the sample size is small, and the confidence intervals are therefore wide. Secondly, this was not a classical anthracycline study population; patients were elderly (median age 67 years) and all but one had visceral disease. Thirdly, a large number (25%) of patients were not evaluable for response, but have been included in the denominator according to the intention-to-treat principle. Finally, we chose to assess response after every second cycle without adjustment for the long dose interval, so that a minimum of 4 months was required for a response to be confirmed. The median time-to-progression is less affected by these factors, and is a very respectable 33 weeks.
Consistent with previous population pharmacokinetic studies of pegylated liposomal doxorubicin in AIDS-related Kaposis sarcoma patients [2, 3], the majority of individual plasma doxorubicin concentrationtime data and the pooled population data were best described by a two-compartment linear model.
When compared with previous iterative-two-stage pharmacokinetic results in patients with prostate cancer [19] and breast cancer [20], terminal half-life is comparable (7699 h in the present study versus 7090 h in [19] and 6286 h in [20]), while clearance (Cl) and steady-state volume of distribution (Vss) appear to be 45% lower in the present study (Cl 15 ml/h/m2 here versus 4445 ml/h in [19] and 3243 ml/h in [20]; Vss
1.651.8 l/m2 here versus 5 l in [19] and 3.54.0 l in [20]). The predicted peak plasma concentrations obtained in this study were consistently 80% higher than in previous publications. Some of these differences may be ascribed to the different populations involved, differences in analytical assays and methodological differences (e.g. variable infusion rates and durations, modelling techniques [20]).
No correlation between pharmacokinetic and pharmacodynamic end points was observed. In particular, possibly due to the very low incidence of PPE, our study was unable to address a previous report [20] correlating half-life with this toxicity.
The recommended dose level of pegylated liposomal doxorubicin using a 6-week schedule is 60 mg/m2, as this dose has been demonstrated to be very well tolerated over multiple cycles in an elderly population. Significant activity has been demonstrated in a population with visceral metastatic breast cancer. This schedule may have pharmacoeconomic advantages over shorter dose intervals, and this should be assessed in future studies.
![]() |
Conclusions |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Acknowledgements |
---|
![]() |
Footnotes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2.
Northfelt DW, Martin FJ, Working P et al. Doxorubicin encapsulated in liposomes containing surface-bound polyethylene glycol: pharmacokinetics, tumor localization, and safety in patients with AIDS-related Kaposis sarcoma. J Clin Pharmacol 1996; 36: 5563.
3. Amantea MA, Forrest A, Northfelt DW et al. Population pharmacokinetics and pharmacodynamics of pegylated-liposomal doxorubicin in patients with AIDS-related Kaposis sarcoma. Clin Pharmacol Ther 1997; 61: 301311.[ISI][Medline]
4. Gabizon A, Catane R, Uziely B et al. Prolonged circulation time and enhanced accumulation in malignant exudates of doxorubicin encapsulated in polyethylene-glycol coated liposomes. Cancer Res 1994; 54: 987992.[Abstract]
5. Vaage J, Barbera-Guillem E, Abra R et al. Tissue distribution and therapeutic effect of intravenous free or encapsulated liposomal doxorubicin on human prostate carcinoma xenografts. Cancer 1994; 73: 14781484.[ISI][Medline]
6. Koukourakis MI, Koukouraki S, Giatromanolaki A et al. High intratumoral accumulation of stealth liposomal doxorubicin in sarcomasrationale for combination with radiotherapy. Acta Oncol 2000; 39: 207211.[ISI][Medline]
7. Ranson MR, Carmichael J, OByrne K et al. Treatment of advanced breast cancer with sterically stabilized liposomal doxorubicin: results of a multicenter phase II trial. J Clin Oncol 1997; 15: 31853191.[Abstract]
8. Northfelt DW, Dezube BJ, Thommes JA et al. Efficacy of pegylated-liposomal doxorubicin in the treatment of AIDS-related Kaposis sarcoma after failure of standard chemotherapy. J Clin Oncol 1997; 15: 653659.[Abstract]
9. Berry G, Billingham M, Alderman E et al. The use of cardiac biopsy to demonstrate reduced cardiotoxicity in AIDS Kaposis sarcoma patients treated with pegylated liposomal doxorubicin. Ann Oncol 1998; 9: 711716.[Abstract]
10. Safra T, Muggia F, Jeffers S et al. Pegylated liposomal doxorubicin (doxil): reduced clinical cardiotoxicity in patients reaching or exceeding cumulative doses of 500 mg/m2. Ann Oncol 2000; 11: 10291033.[Abstract]
11. Markman M, Kennedy A, Webster K et al. Phase 2 trial of liposomal doxorubicin (40 mg/m2) in platinum/paclitaxel-refractory ovarian and fallopian tube cancers and primary carcinoma of the peritoneum. Gynecol Oncol 2000; 78: 369372.[ISI][Medline]
12. Israel VP, Garcia AA, Roman L et al. Phase II study of liposomal doxorubicin in advanced gynecologic cancers. Gynecol Oncol 2000; 78: 143147.[ISI][Medline]
13. Allen A. The cardiotoxicity of chemotherapeutic drugs. Semin Oncol 1992; 19: 529542.[ISI][Medline]
14. DArgenio DZ, Schumitzky A. Adapt II users guide: pharmacokinetic/pharmacodynamic systems analysis software. Los Angeles, 1997.
15. Yamaoka K, Nakagawa T, Uno T. Application of Akaikes information criterion (AIC) in the evaluation of linear pharmacokinetic equations. J Pharmacokin Biopharm 1978; 6: 165175.[ISI][Medline]
16. Steimer JL, Mallet A, Golmard JL et al. Alternative approaches to estimation of population pharmacokinetic parameters: comparison with the nonlinear mixed-effect model. Drug Metab Rev 1984; 15: 265292.[ISI][Medline]
17. Sheiner LB, Boeckman AJ. NONMEN. Users guide part I to VI: University of California at San Francisco, 19881992.
18.
Paridaens R, Biganzoli L, Bruning P et al. Paclitaxel versus doxorubicin as first-line single-agent chemotherapy for metastatic breast cancer: a European Organization for Research and Treatment of Cancer Randomized Study with cross-over. J Clin Oncol 2000; 18: 724733.
19. Hubert A, Lyass O, Pode D et al. Doxil (Caelyx): an exploratory study with pharmacokinetics in patients with hormone-refractory prostate cancer. Anti-Cancer Drugs 2000; 11: 123127.[ISI][Medline]
20. Lyass O, Uziely B, Ben-Yosef R et al. Correlation of toxicity with pharmacokinetics of pegylated liposomal doxorubicin (Doxil) in metastatic breast carcinoma. Cancer 2000; 89: 10371047.[ISI][Medline]