Fax: (314) 454-5218
Phase I study of pegylated liposomal doxorubicin and gemcitabine in patients with advanced malignancies
Article first published online: 31 OCT 2002
Copyright © 2002 American Cancer Society
Volume 95, Issue 10, pages 2223–2229, 15 November 2002
How to Cite
Fracasso, P. M., Blum, K. A., Tan, B. R., Fears, C. L., Bartlett, N. L., Arquette, M. A. and Clark, R. S. (2002), Phase I study of pegylated liposomal doxorubicin and gemcitabine in patients with advanced malignancies. Cancer, 95: 2223–2229. doi: 10.1002/cncr.10937
- Issue published online: 31 OCT 2002
- Article first published online: 31 OCT 2002
- Manuscript Accepted: 18 JUN 2002
- Manuscript Revised: 10 JUN 2002
- Manuscript Received: 21 MAY 2002
- American Cancer Society. Grant Number: CDA-96-27
- Eli Lilly and Company
- palmar–plantar erythrodysesthesia;
- pegylated liposomal doxorubicin;
- Phase I
Pegylated liposomal doxorubicin (PEG-LD) and gemcitabine have single-agent activity in breast and ovarian carcinoma patients. We conducted a Phase I trial to evaluate the maximum tolerated dose (MTD) and toxicities of this combination in patients with advanced malignancies.
Twenty-six patients with refractory or recurrent malignancies were enrolled in this dose escalation trial. Dose escalation proceeded from a starting level of PEG-LD 20 mg/m2 and gemcitabine 1000 mg/m2 administered on Days 1 and 15 of a 28-day cycle.
The MTD was PEG-LD 20 mg/m2 and gemcitabine 2000 mg/m2 administered on Days 1 and 15 of a 28-day cycle. Dose-limiting toxicity, a Grade 3 rash, was observed in one patient during Cycle 1 and Grade 3 stomatitis and a rash were observed in a second patient during Cycle 2 after administration of PEG-LD 25 mg/m2 and gemcitabine 2000 mg/m2. Other side effects included palmar–plantar erythrodysesthesia, nausea, and fatigue. One complete and two partial responses were observed.
The recommended Phase II dose is PEG-LD 20 mg/m2 with gemcitabine 2000 mg/m2 on Days 1 and 15 of a 28-day cycle. A trial with this combination is currently ongoing at this institution comprising patients with refractory ovarian carcinoma. Cancer 2002;95:2223–29. © 2002 American Cancer Society.
Pegylated liposomal doxorubicin (PEG-LD; Doxil®) is a formulation of doxorubicin encapsulated in small, sterically stabilized liposomal vesicles. In Phase I trials of single-agent PEG-LD, the maximum tolerated dose (MTD) was 50 mg/m2 every 3 weeks or 60 mg/m2 every 4 weeks.1 This formulation enables PEG-LD to have a longer half-life, preferential concentration in tumors, and a more favorable toxicity profile than doxorubicin. The dose-limiting toxicities (DLT) are palmar–plantar erythrodysesthesia (PPE), a painful desquamating dermatitis of the hands and feet, and stomatitis. Congestive heart failure (CHF) did not occur among 14 patients receiving cumulative doses higher than 450 mg/m2,1 suggesting that liposomal preparations either limit exposure to nonmalignant tissue or PEG-LD actually has its own toxicity profile, independent of those seen with doxorubicin. Five trials demonstrate the safety of PEG-LD in combination with other chemotherapeutic agents, namely gemcitabine,2, 3 vinorelbine,4 paclitaxel,5 and docetaxel.6 However, a sixth trial combining topotecan with PEG-LD led to unacceptable Grade IV neutropenia and thrombocytopenia.7
Gemcitabine (2′, 2′-difluorodeoxycytidine, Gemzar®), an antimetabolite, slows DNA chain synthesis by competing with deoxycytidine for incorporation into DNA. In Phase I trials of single-agent gemcitabine, dose levels escalated as high as 2362 mg/m2 per week.8, 9 The most commonly used schedule is 1000 mg/m2 weekly for 3 weeks with the fourth week off. DLT consisted of thrombocytopenia and anemia.
The combination of doxorubicin and gemcitabine has been evaluated in patients with advanced breast carcinoma10 and transitional cell carcinoma of the bladder.11 In patients with advanced breast carcinoma, the combination led to a response rate of 55%.10 Maximum doses of doxorubicin 25 mg/m2 and gemcitabine 800 mg/m2 weekly were administered to patients with breast carcinoma.10 In patients with transitional cell carcinoma, a maximal dose of doxorubicin 50 mg/m2 and gemcitabine 2000 mg/m2 every 2 weeks was reached with toxicities consisting primarily of anemia and fatigue.11
With data supporting the safety and high response rate of doxorubicin and gemcitabine in patients with metastatic breast carcinoma plus existing data suggesting the safety of PEG-LD combination regimens, it is reasonable to explore the combination of PEG-LD and gemcitabine. This trial was designed as a Phase I trial to evaluate the safety and determine the MTD of PEG-LD in combination with gemcitabine in the treatment of patients with refractory solid tumors.
MATERIALS AND METHODS
Eligibility criteria included the presence of a pathologically confirmed malignancy refractory to standard treatment or for which no standard treatment exists, evaluable disease (by tumor markers, physical examination, or radiology), age 18 years or older, life expectancy of 3 months or more, and an Eastern Cooperative Oncology Group performance status of 0–2. Required laboratory parameters were a hemoglobin level greater than or equal to 10 g/dL, an absolute neutrophil count (ANC) greater than 1500 cells per mm3, a platelet count greater than or equal to 100,000 platelets per mm3, a creatinine level less than or equal to 2 mg/dL, a bilirubin level less than or equal to 1.5 times the institutional upper limit of normal (ULN), and aspartate transaminase and alanine transaminase levels less than or equal to two times the ULN. Patients with previous exposure to PEG-LD or gemcitabine were excluded and previous doxorubicin exposure was limited to less than or equal to 300 mg/m2. No treatment with chemotherapy, radiation therapy, or immunotherapy was allowed 3 weeks before enrollment. Other exclusion criteria included pregnancy or breast-feeding, concurrent malignancies (except nonmelanoma skin or cervical carcinoma), and CHF. All patients provided written informed consent.
Within 14 days before treatment initiation, all patients underwent history, a physical examination, performance status assessment, laboratory evaluation (complete blood count [CBC], chemistries, urinalysis, and tumor markers), imaging assessment, a resting ventriculogram, and an electrocardiogram. A CBC was checked weekly and a physical examination, performance status assessment, and medication assessment were recorded every 4 weeks. In addition, tumor measurements either by imaging studies and/or tumor markers were performed every 8 weeks. Adverse events were reported and recorded when they occurred.
PEG-DL (Doxil®, Alza Pharmaceuticals, formerly Sequus Pharmaceuticals, Mountain View, CA) was obtained commercially. Sequus Pharmaceuticals provided this agent to those patients for whom the drug was not covered by medical insurance. Eli Lilly and Company (Indianapolis, IN) provided patients on this study with gemcitabine hydrochloride (Gemzar®). Gemcitabine was administered intravenously over 30 minutes followed by the administration of PEG-LD intravenously over 60 minutes. A cycle was defined as treatment with both agents on Days 1 and 15 of a 28-day schedule.
Dose Escalation and Dose Modifications
Three to six patients were enrolled at each dose level. Toxicities were graded according to the National Cancer Institute common toxicity criteria. If no DLT was observed for the first three patients at a given dose level, escalation proceeded to the next dose level. Initially, only the first cycle was used to determine the DLT. Because of significant nonhematologic toxicity observed on the second cycle in the fourth dose escalation, the second cycle was also used to determine the DLT in Cohort 4 as well as in subsequent cohorts. After the first DLT, up to three more patients were enrolled at that same dose level. Escalation continued only if DLT was limited to one of six patients. Escalation halted if DLT occurred in two or more patients. DLT was defined as Grade 4 neutropenia or thrombocytopenia lasting for more than 4 days, a greater than 14-day delay in the start of the subsequent cycle due to an ANC less than 1500 cells per microliter or a platelet count less than 100,000 platelets per microliter, Grade 3 stomatitis/esophagitis lasting for more than 3 days, or Grade 3 or 4 nonhematologic toxicity (with the exception of alopecia, nausea, vomiting, fever, or anorexia). The MTD was defined as one dose level below that in which DLT was observed in greater than or equal to two to six patients. Six patients were accrued at the defined MTD to fully evaluate any toxicity encountered at this level. There was no intrapatient dose escalation.
Treatment was delayed on a weekly basis for an ANC less than 1500 cells per microliter and a platelet count of less than 100,000 platelets per microliter. A 25% dose reduction of PEG-LD and gemcitabine was administered in subsequent cycles, if during the preceding cycle, the patient's nadir ANC was less than or equal to 500 cells per microliter for 7 days or failed to recover by Day 1 of the next cycle or the patient experienced febrile neutropenia. PEG-LD alone was reduced by 25% or 50% for Grade 3 or 4 PPE, respectively. Grade 3–4 nonhematologic toxicities (with the exception of alopecia, nausea, vomiting, fever, and anorexia) led to a 25% dose reduction of both drugs. Therapy was discontinued for a performance status of 3 or higher that persisted for more than 3 weeks, voluntary withdrawal, or disease progression.
A complete response (CR) was defined as the disappearance of all measurable disease or normalization of tumor markers for a minimum of 4 weeks. A partial response (PR) was defined as a 50% or greater decrease in the sum of the product of the diameters of all measured lesions, with the appearance of no new lesions, for at least 4 weeks. Stable disease (SD) was defined as a less than 50% decrease and a less than 25% increase in the sum of the diameter products with no new lesions. Progressive disease (PD) was defined as a 25% or greater increase in the total area of any bidimensionally measurable lesion compared with best response or as the appearance of any new lesion.
Twenty-six patients were enrolled between January 1998 and February 1999. The characteristics of these patients are listed in Table 1 and the dose levels studied are summarized in Table 2. Dose escalation proceeded without any DLT in Cohorts 1–3. However, an additional patient was added to each of these cohorts because one patient in each cohort did not complete the first cycle (secondary to PD, an allergic reaction to PEG-LD, and prolonged treatment delay attributed to preexisting myelodysplastic syndrome). In Cohort 4, one DLT (Grade 3 rash) was observed. As a result, three more patients were planned for this cohort. However, the next patient enrolled experienced Grade 3 stomatitis and rash on Day 4 of Cycle 2. Subsequent dose escalation was halted due to the early onset of these toxicities and the DLT was expanded to use Cycles 1 and 2. Three additional patients were planned for Cohort 3. Four patients were treated because one patient discontinued treatment before completion of the first cycle due to Grade 2 fatigue and rash. There was no DLT in this cohort. Because the majority of side effects (rash and stomatitis) in Cohort 4 were believed to be PEG-LD related, a dose level of PEG-LD 20 mg/m2 and gemcitabine 2000 mg/m2 (Cohort 5) was then added to continue gemcitabine escalation without further PEG-LD escalation. There was no DLT in this cohort.
|Characteristics||No. of patients|
|ECOG performance status|
|Head and neck||3|
|Mean no. of regimens (range)||2.2 (1–8)|
|Previous doxorubicin treatment||8|
|Mean dose (mg/m2)||197|
|Previous mitoxantrone treatment||1|
|Total dose (mg/m2)||30|
|Cohort||PEG-LD (mg/m2)a||Gemcitabine (mg/m2)a||No. of patients||No. of cycles administered||No. of dose reductions (no. of patients)||No. of cycles delayed (no. of patients)|
|PEG-LD + gemcitabine||PEG-LD alone|
|1||20||1000||4||5.5b||0 (0)||0 (0)||0 (0)|
|2||20||1500||4||6.0b||0 (0)||0 (0)||0 (0)|
|3||25||1500||8||39.0b||3 (2)||4 (3)||23 (6)|
|4||25||2000||4||10.5b||2 (2)||1 (1)||3 (2)|
|5||20||2000||6||20.0||4 (2)||0 (0)||9 (2)|
Ten dose reductions secondary to stomatitis, rash, PPE, neutropenia, and thrombocytopenia occurred in Cohorts 3 and 4 (Table 2). In Cohort 5, there were no dose reductions for PPE or stomatitis, although two of six patients underwent dose reductions for neutropenia (Table 2). Of 12 patients in Cohorts 3 and 4, 8 experienced treatment delays (total of 26) compared with two patients in Cycle 5.
Table 3 displays the mean nadir of the ANC, hemoglobin, and platelets by cohort. In all cohorts, profound cytopenias were rare. Grade 3 febrile neutropenia occurred in Cohorts 3 and 5. In Cohort 3, one patient, already dose reduced to PEG-LD 20 mg/m2 and gemcitabine 1500 mg/m2 secondary to stomatitis, presented with fever to 38.8 °C and an ANC of 300 cells per mm3 on Day 41 of Cycle 7. The patient improved with empiric antibiotics and all cultures remained negative. Another patient in Cohort 3 was admitted on Day 21 of Cycle 1 with a temperature of 37.3 °C, an ANC of 500 cells per mm3, and a dental abscess which cultured group A beta hemolytic streptococci. The patient recovered with intravenous antibiotics and the ANC increased to 1500 cells per mm3 within 4 days of treatment. A patient in Cohort 5 presented with an ANC of 700 cells per mm3 and a computed tomographic (CT) scan showed evidence of sinusitis on Day 7 of Cycle 2. This patient was treated for infection and continued treatment following a dose reduction. Of the 26 patients, Grade 3/4 neutropenia was observed in 11, Grade 3/4 anemia in 3, and Grade 3/4 thrombocytopenia in 2 (Table 4). Of seven patients who were heavily pretreated (three or more previous regimens), four patients had a Grade 3/4 toxicity consisting of neutropenia, thrombocytopenia, or anemia.
|Cohort||Dose (mg/m2)||Mean, nadir, counts (range)|
|ANC (× 103/μL)||Hemoglobin (g/dL)||Platelets (× 103/μL)|
|PEG-LD||Gemcitabine||Cycle 1||All other cycles||Cycle 1||All other cycles||Cycle 1||All other cycles|
|1||15||1000||3.06 (1.21–5.98)||3.09 (1.21–5.98)||10 (8.2–12.4)||10.2 (8.2–12.4)||233 (175–341)||249 (175–298)|
|2||20||1500||1.91 (0.84–4.66)||1.55 (0.36–4.67)||11.8 (9.5–13.5)||11.1 (8.8–13.5)||141 (93–190)||162 (93–199)|
|3||25||1500||2.48 (0.52–7.21)||1.95 (0.25–7.21)||10.3 (7.3–12.4)||10.8 (7.3–14.2)||157 (51–274)||175 (7–600)|
|4||25||2000||1.93 (1.51–2.30)||1.97 (0.85–2.71)||10.7 (9.0–12.4)||10.9 (8.6–12.4)||314 (158–597)||296 (158–597)|
|5||20||2000||1.87 (0.42–3.77)||1.09 (0.14–3.77)||11.3 (7.9–13.3)||11.6 (7.9–12.7)||146 (65–245)||146 (65–245)|
|Toxicity||Grade 1/2 No. of patients (%)||Grade 3/4 No. of patients (%)|
|Reaction to PEG-LD||2 (7.7)||2 (7.7)|
|Anemia||18 (69.2)||3 (11.5)|
|Neutropenia||13 (50.0)||11 (42.3)|
|Thrombocytopenia||10 (38.5)||2 (7.7)|
|Fatigue/malaise||9 (34.6)||1 (3.8)|
|Fever w/o neutropenia||2 (7.7)||0|
|Rash||11 (42.3)||3 (11.5)|
|PPE||9 (34.6)||3 (11.5)|
|Alteration in taste||1 (3.8)||0|
|Esophageal reflux||1 (3.8)||0|
|Stomatitis||12 (46.2)||2 (7.7)|
|Infection w/o neutropenia||2 (7.7)||0|
|Infection with neutropenia||0||1 (3.8)|
|Fever with neutropenia||0||3 (11.5)|
Tables 4 and 5 list the nonhematologic toxicities encountered in this dose escalation trial, including stomatitis, PPE, rash, fatigue, and nausea. Very few Grade 3 and no Grade 4 nonhematologic toxicities occurred. Skin toxicity and stomatitis, the most frequent Grade 3 events, improved with a treatment delay and subsequent dose reduction. These toxicities were much less common at PEG-LD doses of 20 mg/m2 or less. Of the 14 patients who developed stomatitis, 12 developed skin toxicity as well, either as a generalized rash or in the form of PPE, generally no more than one cycle apart. Only four patients developed skin toxicity without coincident stomatitis. Rashes were seen in all cycles, whereas PPE was generally seen after Cycle 1. Grade 3 stomatitis and rash/PPE occurred exclusively in Cohorts 3 and 4, those patients receiving PEG-LD 25 mg/m2.
|Toxicity||Cohort 1||Cohort 2||Cohort 3||Cohort 4||Cohort 5||Total events|
|Cycle 1||All other cycles||Cycle 1||All other cycles||Cycle 1||All other cycles||Cycle 1||All other cycles||Cycle 1||All other cycles|
Allergic reactions to PEG-LD occurred in all cohorts. Two patients experienced Grade 3 hypersensitivity to PEG-LD. One patient, in Cohort 4, became unresponsive at the onset of the PEG-LD infusion. Although probably a syncopal event, seizure could not be ruled out as the patient was incontinent of urine. The patient, never hypotensive, was hospitalized for 24 hours without further event. CT scan and magnetic resonance imaging of the brain revealed no pathology. On Day 15, treatment resumed without further complications. The second patient, in Cohort 6, experienced severe back and leg pain until the infusion was discontinued. The Cycle 1, Day 15 dose was administered without difficulty. Both patients with Grade 3 reactions tolerated rechallenge with PEG-LD without recurrent events. However, one patient, in Cohort 2, with a Grade 2 allergic reaction experienced recurrent symptoms upon rechallenge and was withdrawn from the study.
Of 26 patients, 22 were evaluable for response. Responses could not be ascertained for four patients who did not complete two cycles of treatment. There was one CR lasting 12 months in a patient with ovarian carcinoma who had disease recurrence after paclitaxel/carboplatin and topotecan. This patient received nine cycles of treatment in Cohort 3. Two PRs were observed in patients with sarcoma (Cohort 4) and adrenocortical carcinoma (Cohort 5). These responses persisted 2 months and 23 months, respectively. One patient with refractory Hodgkin disease, who had failed to respond to eight previous regimens, had a near PR with a 48% decrease in tumor burden. One other patient (breast carcinoma) had evidence of SD and the remaining 17 progressed while on the protocol.
Doxorubicin is used to treat patients with sarcoma, lymphoma, and breast and ovarian carcinoma.12 Unfortunately, its use is limited by potential cardiac toxicity, particularly at cumulative doses greater than 400 mg/m2.12 Liposomal formulations of doxorubicin may improve its toxicity profile.2, 13, 14 Gemcitabine, like PEG-LD, is used to treat patients with breast carcinoma,15 Hodgkin lymphoma,16 soft tissue sarcoma,17 and ovarian carcinoma.18 This Phase I trial evaluated the safety profile and MTD of the combination of PEG-LD and gemcitabine in patients with advanced malignancies.
In this trial, only one DLT was observed during the first cycle in Cohort 4 (PEG-LD 25 mg/m2 and gemcitabine 2000 mg/m2 on Days 1 and 15). However, a Grade 3 rash and stomatitis in a second patient at the onset of Cycle 2 prompted closure of this cohort. Given the potential activity of this combination in patients with metastatic breast and ovarian carcinoma who are likely to receive multiple cycles of therapy, this early toxicity suggested that few patients would be able to tolerate this regimen with subsequent cycles. Therefore, Cohort 3 (PEG-LD 25 mg/m2 and gemcitabine 1500 mg/m2 on Days 1 and 15) was determined to be the MTD. This cohort still had frequent adverse events in later cycles, namely, skin toxicity and stomatitis believed to be secondary to PEG-LD. As a result, another cohort (Cohort 5, PEG-LD 20 mg/m2 and gemcitabine 2000 mg/m2 on Days 1 and 15) was added to permit further gemcitabine escalation and limit skin toxicity and stomatitis. Toxicities in Cohort 5 were minimal with one patient developing febrile neutropenia in Cycle 2. Three of the six patients in Cohort 5 completed more than one cycle (one patient completing 11 cycles), allowing observation of toxicity in later cycles. As a result, the Phase II recommended starting dose is PEG-LD 20 mg/m2 and gemcitabine 2000 mg/m2 on Days 1 and 15 of a 28-day schedule.
Two other Phase I trials have evaluated the combination of PEG-LD and gemcitabine in patients with recurrent breast and ovarian carcinoma. In patients with metastatic breast carcinoma, the MTD of the combination was PEG-LD 24 mg/m2 on Day 1 and gemcitabine 800 mg/m2 on Days 1 and 8.2 Treatment was initially planned with PEG-LD on Day 1 combined with gemcitabine on Days 1, 8, and 15 of a 28-day cycle. However, Grade 3/4 myelosuppression on Day 15 required omission of gemcitabine on Day 15. As a result, gemcitabine was eliminated on Day 15 of the schedule and the cycle length was shortened to 21 days. Despite this dose and schedule change, myelosuppression remained the most frequent adverse event in the 153 cycles of this combination administered to 27 patients. Even in a minimally pretreated population, further escalation was not possible due to hematologic toxicities. Of 153 cycles, 33 were delayed mainly due to neutropenia and thrombocytopenia. Other toxicities included PPE, stomatitis, and fatigue. The response rate in this population was 33% with two CRs, seven PRs, two minor responses, and eight SD.
In their study of patients with recurrent ovarian carcinoma, Tobias et al.3 reported that the MTD of the combination was PEG-LD 25 mg/m2 on Day 1 and gemcitabine 650 mg/m2 on Days 1 and 8 of a 28-day schedule. In this dose escalation study,3 77 cycles of this combination were administered to 17 patients. Tobias et al. observed three episodes of Grade 3–4 myelosuppression and two episodes of Grade 3 stomatitis. Only one episode of Grade 4 thrombocytopenia was reported at the MTD. The response rate in this population was 43% with five CRs, one PR, and five SD.
In our trial, the every other week (Days 1 and 15) schedule of PEG-LD and gemcitabine allowed for a higher dose intensity of these agents. This schedule ameliorated the hematologic complications, which prevented further dose escalation of either drug in the previous two trials. This schedule delivered doses closer to the single-agent schedules of both drugs. Whether a higher cumulative dose translates into higher response rates will need to be further evaluated.
Although not a primary objective in this trial, responses were observed, including one CR and two PRs for an overall response rate of 12%. One of two patients with ovarian carcinoma had a CR; the other had PD. The only patient with breast carcinoma had SD during 14 cycles of treatment. These results coupled with others2, 3 warrant further Phase II studies of this combination in patients with breast and ovarian carcinoma.
In conclusion, the combination of PEG-LD 20 mg/m2 and gemcitabine 2000 mg/m2 on a Day 1, 15 schedule is a well tolerated regimen that minimizes skin and hematologic toxicities. At our institution, a Phase II study of this combination is ongoing in patients with recurrent ovarian carcinoma.
The authors thank the nurses and data managers at the Siteman Cancer Center for caring for the patients on this trial. They also thank Dr. Teresa J. Vietti for her editorial assistance.
- 3A phase I trial of gemcitabine and Doxil for recurrent epithelial ovarian cancer [Abstract]. Proc Am Soc Clin Oncol. 2000; 19: 392a., , , et al.
- 5Doxil/Caelyx plus paclitaxel in patients with metastatic breast cancer: update of a phase II efficacy and safety study [Abstract]. Proc Am Soc Clin Oncol. 2001; 20: 67b., , , et al.