Phase I Study of Doxil-Cisplatin Combination Chemotherapy in Patients with Advanced Malignancies1

3040 Vol. 7, 3040–3046, October 2001 Clinical Cancer Research

Olga Lyass, Ayala Hubert, and cancer patients. In four of seven responders, the time to 2 Alberto A. Gabizon disease progression exceeded 1 year. Doxil 1-h (Cmax equiv- Department of Oncology, Hadassah-Hebrew University Medical alent) levels were assessed in 20 patients. The mean Doxil Center, Jerusalem 91120, Israel Cmax (mg/l plasma) increased gradually with dose escalation ,from 14.7 ؎ 1.9 for 40 mg/m2, to 17.3 ؎ 3.0 for 50 mg/m2 2 2 ؎ and 23.3 5.5 for 60 mg/m . The 60 mg/m Cmax was similar ABSTRACT to data obtained in parallel clinical studies at our institution Purpose: Our first objective was to evaluate the feasi- with single-agent Doxil at 60 mg/m2. However, the 7-day bility of administering a combination of Doxil, a pegylated Doxil postinfusion levels were significantly lower in patients liposome formulation of doxorubicin, and cisplatin and to receiving the Doxil-cisplatin combination than in those re- determine the maximum tolerated dose of the combination. ceiving single-agent Doxil. A secondary objective was to examine Doxil peak and 7-day Conclusion: Doxil can be administered at full maximum postinjection plasma levels at the various dose levels tested. tolerated dose (50 mg/m2 every 4 weeks) in combination with Methods: Patients with advanced solid tumors were 60 mg/m2 cisplatin, with no evidence of major overlapping treated every 4 weeks with cisplatin on day 1 and Doxil on toxicities. Palmar-plantar erythrodysesthesia incidence and day 2. In the first three dose levels, the dose of Doxil was severity appears to be diminished, in comparison with data fixed at 40 mg/m2, whereas the dose of cisplatin was esca- available for single-agent Doxil. Plasma concentration data lated from 40 to 50 and 60 mg/m2. At the fourth and fifth point to an accelerated clearance of Doxil when adminis- dose levels, the dose of cisplatin was fixed at 60 mg/m2, tered after cisplatin. whereas the dose of Doxil was escalated to 50 and to 60 mg/m2. Plasma Doxil (doxorubicin-equivalent) levels were measured by a high-performance liquid chromatography INTRODUCTION assay with fluorescence detection at 1 h and 7 days after The anthracycline antibiotic doxorubicin is one of the most infusion of Doxil. widely used antineoplastic agents and the mainstay of chemo- Results: Twenty-six patients entered the study. Twenty- therapy for lymphomas and a variety of solid tumors. Unfortu- four patients completed a minimum of 2 courses and were nately, toxicity limits the therapeutic activity of doxorubicin and fully assessable for toxicity and efficacy. Eighteen patients may preclude adequate dosing. Development of doxorubicin had received prior chemotherapy, 11 of them with anthra- encapsulation in liposomes is one of the attempts to improve the cycline-containing regimens. A total of 177 courses were therapeutic index of the drug. administered within the study. In 12 patients, cisplatin was Pegylated liposomal doxorubicin (Doxil, also known as discontinued after 1 to 13 courses, and Doxil was continued Caelyx) is a novel formulation of doxorubicin in long-circulat- alone for 1–22 courses. All other patients received both ing (Stealth) liposomes that drastically changes the drug phar- drugs until discontinuation of therapy. The dose-limiting macokinetics and biodistribution (1). The formulation markedly toxicities were neutropenia and mucositis. Grade 4 neutro- lengthens liposome circulation time by retarding reticulo-endo- penia was seen in 3 patients (one with neutropenic fever) at thelial system clearance. Eventually, large numbers of lipo- dose levels 4 and 5. Grade 3 mucositis was observed in 4 somes are extravasated through the abnormally permeable ves- patients at dose levels 3, 4, and 5. In contrast, the most sels characteristic of many tumors, resulting in enhanced severe palmar-plantar erythrodysesthesia manifestation was liposome deposition in tumors (2, 3). The toxicity profile of grade 2 seen in 1 patient only. Tumor responses included Doxil is characterized by dose-limiting mucocutaneous toxici- seven partial responses, of which three were in ovarian ties, mild myelosuppression, minimal alopecia, and no apparent cardiac toxicity (4). Doxil appears to be the most active agent available for Kaposi’s sarcoma (5, 6) and has shown significant activity against a variety of solid tumors (reviewed in Ref. 7), Received 3/30/01; revised 6/8/01; accepted 7/9/01. specifically in recurrent epithelial ovarian carcinoma (8, 9) and The costs of publication of this article were defrayed in part by the metastatic breast cancer (10, 11). payment of page charges. This article must therefore be hereby marked Cisplatin is one of the most potent chemotherapeutic agents advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. available and forms the basis of many standard combination 1 Supported by ALZA Corporation. A preliminary report on this study chemotherapy regimens. Cisplatin is used in combination with was presented in poster form at the 1999 European Conference on doxorubicin in the treatment of a variety of solid tumors. These Clinical Oncology (Lyass, O., Hubert, A., Tzemach, D., et al., Eur. J. two agents have different mechanisms of action, show no cross- Cancer, 35: S287, 1999). 2 To whom requests for reprints should be addressed, at Department of resistance, and their toxicities do not overlap. Because Doxil Oncology, Hadassah Medical Center, Kiryat Hadassah, Jerusalem appears to be a promising form of delivering doxorubicin with 91120, Israel. Fax: 972-2-643-0622; E-mail: [email protected]. decrease of some of the most problematic toxicities (cardiac

Table 1 Dose levels clide ventriculography (MUGA scan) or echocardiography. No. of Cisplatin, Doxil, Prior treatment with anthracyclines was allowed only if the 2 Dose level patients mg/m2, day 1a mg/m2, day 2a cumulative dose of doxorubicin was Ͻ400 mg/m . The study 1 5 40 40 was approved by the Hadassah institutional review board, and 2 4 50 40 patients were required to provide signed, witnessed informed 3 4 60 40 consent. 4 5 60 50 Pretreatment evaluation included a complete history and 5 8 60 60 physical examination and documentation of performance status. a Cycles every 4 weeks. Laboratory evaluation included a complete blood count, chemistry panel with liver and renal function tests, prothrombin time, and serum tumor marker levels if relevant to tumor condition. toxicity and myelosuppression), a combination with cisplatin is All patients had a pretreatment 12-lead electrocardiogram and likely to be tolerated without enhanced toxicity, which may cardiac ejection fraction evaluation. Radiographic evaluation enable us to deliver to patients optimal doses of both agents and included a chest radiograph and any other imaging examinations achieve significant advantages over standard cisplatin-doxoru- (computed tomography scan and ultrasonography) indicated by bicin combination. the type and site of the malignancy. Before each new treatment The principal objectives of the present study were to eval- cycle, vital signs, weight, symptom-oriented physical check, uate the feasibility of administering a combination of Doxil and Karnofsky status assessment, complete blood count, chemistry cisplatin and to determine the MTD3 of the combination. In panel with liver and renal function tests, and report on adverse addition, Doxil plasma levels were assessed in subgroup of events occurring after the previous dose were obtained. During patients and compared with our records for patients receiving the first cycle, a complete blood count was repeated on a weekly Doxil as single agent (11, 12). basis. From the second cycle and onward, a complete blood count was performed before treatment and 2 weeks after each 2 PATIENTS AND METHODS dose. After a 300 mg/m cumulative dose of Doxil, cardiac ejection fraction was reevaluated by MUGA scan or echocar- Study Design. The study reported herein was a single- diography. institution Phase I trial at the Hadassah Medical Center. Be- Treatment was administered until disease progression or tween February 1997 and October 1998, 26 patients were en- dose-limiting toxicity. All patients who received one or more rolled. courses of cisplatin and Doxil were evaluable for toxicity. The starting dose of Doxil was 40 mg/m2 combined with Toxicity was graded according to the National Cancer Institute 40 mg/m2 cisplatin every 4 weeks. Doses of Doxil and cisplatin Common Toxicity Criteria. For the sake of clarity, PPE grades were escalated according to a planned sequence (Table 1), 3–4 always entailed a situation in which simple physical activ- covering a range of 40–60 mg/m2 for both drugs. A minimum ities (writing, clenching fists, and walking with closed shoes) of 4 patients receiving at least 2 courses of treatment were were highly painful or unfeasible. enrolled per dose level. At each dose level, the study progressed For stomatitis and PPE, dose-limiting toxicity was defined to the next level if no patient experienced nonhematological as any grade 3 or 4. For patients with grade 3 or 4 toxicity, the toxicity of grade 3 or hematological toxicity of grade 4 during dose of subsequent courses was reduced by 25%. Patients were the two first dosing cycles. If only one patient experienced grade retreated only after the toxicity grade returned to 0 to 1. If 3 or 4 toxicity, up to 4 subsequent patients were to be accrued necessary, the dosing intervals could be increased by 1 or 2 at the same dose level until the dose tolerance could be accu- weeks to enable resolution of toxicity. rately assessed. If two or more patients experienced grade 3 or For myelosuppression, dose-limiting toxicity was defined 4 toxicity per dose level, then this level was declared as the as grade 4. In case of grade 3 neutropenia or thrombocytopenia, initial toxic dose and the preceding level was considered as the study regime was withheld until toxicity resolved to grade the MTD. 0–1 (neutrophil count Ն1,500; platelets Ն75,000), after which Eligibility Criteria and Follow-Up. Patients with histo- study treatment was resumed with no dose modification. If logically documented advanced solid malignancies refractory to grade 4 neutropenia (neutrophil count Ͻ500) or thrombocyto- conventional therapy or for whom no effective therapy exists penia (platelet count Ͻ25,000) developed, therapy was withheld were candidates for this study. Eligibility criteria included: age until resolved to grade 0–1 and resumed at a 25% dose reduc- Ն18 years; Karnofsky performance status score Ն60%; no tion. Granulocyte-colony stimulating factor or granulocyte/ prior chemotherapy or radiotherapy within 4 weeks of entering macrophage-colony stimulating factor were allowed under spe- the study; WBC count Ն3,500/mm3, neutrophil count Ն1,500/ cific circumstances in this study, in patients who showed mm3, platelet count Ն75,000/mm3, and hemoglobin Ն10 g% prolonged neutropenia or in the occurrence of febrile neutrope- (with or without transfusion); total bilirubin Յ25 ␮l; creatinine nia in a prior cycle of treatment. Յ200 ␮l; and left ventricle ejection fraction Ն50% by radionu- Antitumor responses were evaluated after the first 2 courses and thereafter every 3 courses by physical examination, imaging radiograms, and, when indicated, tumor markers. Pa- Ն 3 The abbreviations used are: MTD, maximum tolerated dose; MUGA, tients receiving 2 courses were considered assessable for multigated radionuclide ventriculography; PPE, palmar-plantar erythro- antitumor response. The criteria for evaluating responses were dysesthesia. defined as reported previously (4) and included the following

Table 2 Patient characteristics the most frequent being soft tissue sarcoma, non-small cell lung Characteristic No. of patients cancer, ovary, mesothelioma, and prostate. The majority of the patients (18 of 26) had been exposed to chemotherapy before No. entered 26 Male:Female ratio 16:10 entering the study, with 11 of them having been treated with Age, yr [median (range)] 58 (21–73) anthracycline-containing regimens. The total number of courses Performance status [median (range)] 80 (60–90) was 177. The median number of courses per patient was 5 Primary tumor site (range, 1–22). In 12 patients, cisplatin was discontinued after Soft tissue sarcoma 6 1–13 courses, and Doxil was continued alone for 1–22 courses. Non-small cell lung cancer 5 Ovarian cancer 4 One patient received only 1 course of therapy, discontinued the Pleural mesothelioma 3 treatment because of social problems, and was not assessable for Prostate cancer 3 skin toxicity (which, in most instances, requires at least two a Miscellaneous 5 cycles of Doxil to manifest) nor for antitumor response. Another Prior chemotherapy 18 Prior anthracyclines 11 patient received 70 mg of free doxorubicin in addition to his No. of courses scheduled therapy because of an error of the nursing team. Total 177 Because of this protocol violation, he was not available for Median per patient (range) 5 (1–27) evaluation of toxicity and antitumor response. Thus, 24 fully a Colon cancer, (n ϭ 1); hepatoma, (n ϭ 1); neuroendocrine evaluable patients remained. pancreatic tumor, (n ϭ 1); cholangiocarcinoma, (n ϭ 1); adrenocortical Treatment-related Toxicities. Neutropenia was the ϭ carcinoma, (n 1). most prominent and dose-limiting hematological side effect at the dose levels tested. On the basis of data from the first two cycles, grade 3 to 4 neutropenia was observed in 7 patients, with 1 case at dose level 2 and 2 cases at each of dose levels 3, 4, and categories: complete response, partial response, stable (includ- 5. It was complicated by neutropenic fever in only 1 instance. ing improvement), and progressive disease. Time to disease Median nadir counts of granulocytes for all dose levels tested progression and survival were defined as number of months were in the range of 2,600–4,500/␮l for the first cycle and from start of therapy until documentation of disease progression 1,100 to 4,000/␮l for the second cycle, suggesting a slight trend or death, respectively. to lower nadirs after the second cycle (Table 3). However, the Administration of Treatment. All patients were hospi- range was broad, and in some patients, no significant neutrope- talized for 24 h for treatment. Doxil was obtained from ALZA nia was detected. Also, in most instances of granulocytopenia, Corp. (Mountain View, CA) and cisplatin from Teva Pharma- there was quick recovery of the cell counts within Յ7 days. ceuticals (Petach Tikva, Israel). Patients were hydrated with Nadir counts were observed either on day 14 or day 21 after physiological saline and hypertonic mannitol or, when indi- chemotherapy. cated, furosemide. Immediately prior to cisplatin infusion, an- Thrombocytopenia was mild, reaching grade 3 in only 2% tiemetic premedication including dexamethasone (8–20 mg) of the cycles (Table 3). No grade 4 thrombocytopenia was and ondansetron (8–16 mg) or granisetron (3 mg) was admin- observed. Treatment-related anemia was also mild. Grade 3 istered. Cisplatin was given in 150 ml of 0.9% sodium chloride anemia was documented in 3.5% of the cycles. In one patient, solution, followed by adequate infusion of saline. The next grade 4 anemia that required blood transfusion was reported, morning, 12–18 h after cisplatin infusion, patients received and it was attributable to massive tumor-related bleeding. Me- Doxil diluted in 250 ml of 5% dextrose (Baxter, Deerfield, IL) dian nadir values of hemoglobin were Ͼ8.7 g% at all dose without further antiemetic therapy. Doxil infusion was given levels. over 1–2 h, beginning with a slow drip rate of 0.1–0.2 mg/min Of note, one patient with recurrent ovarian carcinoma, and reaching an average rate of 1 mg/min. while on partial remission of her underlying disease, died of Analysis of Plasma Doxil Levels. Blood samples (4 ml) acute myelogenous leukemia that developed 25 months after the were collected into EDTA-containing tubes 1 h after completion start of treatment. This patient had received carboplatin, cyclo- of Doxil infusion and 7 days later. Blood was immediately phosphamide, and paclitaxel in the past. The respective contri- stored at 5°C. Plasma was separated by centrifugation within bution of the various drugs received by the patient to the Ϫ 24 h and stored at 20°C. Extraction of doxorubicin from development of acute leukemia is not known. plasma and determination by an isocratic high-performance The major nonhematological adverse effect was mucositis liquid chromatography system with fluorescence detection was in the form of stomatitis-pharyngitis (Table 4). Mucositis was as described in detail previously (13). On the basis of prior the second dose-limiting toxicity of this combination regimen, experience (11), the 1-h postinfusion sample was considered reaching grade 3 in 4 of 17 patients treated at the highest three roughly indicative of the peak plasma level or Cmax of Doxil. dose levels. Also (not shown in Table 4), one patient at dose level 5 developed grade 4 mucositis after the fourth cycle. In RESULTS contrast, only 1 case of grade 1 mucositis was observed at the 2 Patient Characteristics. Twenty-six patients, whose lowest dose levels. Thus, severity and incidence of mucositis characteristics are listed in Table 2, entered this study. There increased clearly at higher dose levels (Table 4). were 10 female and 16 male patients with a median age of 56 Regarding PPE, no grade 3 or 4 was reported in the first years (range, 21–73 years) and a median performance status of three courses at any dose levels. Grade 2 PPE was recorded in 80% (range, 60–90%). There were 10 different types of tumors, only 1 of 8 patients at dose level 5. Grade 1 PPE was seen in 3

Table 3 Myelosuppressiona Nadir count, blood Median (range) Granulocytes (ϫ 103/␮l) Platelets (ϫ 103/␮l) Hemoglobin (g/dl) Dose level First course Second course First course Second course First course Second course 1 4.5 (3.7–6.8) 3.8 (3.2–4.6) 367 (228–504) 290 (212–346) 11.6 (8.8–12.8) 11.9 (11.7–12.4) 2 2.6 (0.9–8.7) 2.2 (1.2–6.0) 137 (105–335) 208 (86–319) 11.7 (10.5–13.8) 11.6 (8.8–13.0) 3 2.6 (1.4–3.7) 1.1 (0.7–7.4) 95 (70–146) 107 (42–183) 8.8 (5.0–11.6) 9.5 (8.2–10.9) 4 3.2 (0.2–4.4) 1.7 (1.5–2.4) 286 (30–476) 265 (155–413) 9.0 (7.5–11.7) 8.7 (8.6–11.0) 5 4.4 (0.3–5.6) 4.0 (0.4–4.2) 231 (70–448) 154 (34–505) 11.0 (8.7–12.7) 11.1 (8.9–11.9) a The patient numbers for each dose level are as shown in Table 1.

Table 4 Mucositisa and peripheral neurosensory toxicity in 1 heavily cisplatin- No. of events/No. of patients pretreated patient after 13 courses. Because of difficulties in compliance with repeated treatment, there were another eight Dose level Grade 1 Grade 2 Grade 3 Grade 4 patients in whom cisplatin was discontinued after five or more 1 1/4 0/4 0/4 0/4 cycles, and Doxil was continued alone as maintenance therapy, 2 0/4 0/4 0/4 0/4 thus avoiding the need for hydration and hospitalization. 3 0/4 1/4 1/4 0/4 4 0/5 2/5 1/5 0/5 Monitoring of cardiac function was performed by MUGA 5 1/8 0/8 2/8 0/8 scan or echocardiography at baseline and serially in patients Ͼ 2 a Based on the first 2 cycles, except for 1 patient who received only receiving cumulative doses 300 mg/m . A total of 8 patients 2 1 course at dose level 1 and discontinued therapy thereafter. received a cumulative dose of Doxil Ն450 mg/m (including a patient who reached a cumulative dose of 1160 mg/m2), and remained with normal cardiac ejection fractions. Clinical con- Table 5 Antitumor response gestive heart failure did not occur. Time to Survival Dose-limiting Toxicities and MTD. The dose-limiting Response and no. of patients progression (mo) (mo) toxicities observed were: 1 patient with grade 3 mucositis at Partial response (n ϭ 7) dose level 3; 1 patient with grade 3 mucositis and grade 4 Ovarian cancer (n ϭ 3) 11, 23, 32 16, 25, 37ϩ neutropenia at dose level 4; and 3 patients with grade 3 mucosi- Mesothelioma 17 24 tis (2 patients) and grade 4 neutropenia (2 patients, 1 of them Hepatoma 6 7 with neutropenic fever) at dose level 5. On the basis of the Neuroendocrine cancer 17 33 Prostate cancera 617protocol requirement of two events of dose-limiting toxicity per Stable (n ϭ 3) dose level in different patients during the first two cycles, dose Cholangiocarcinoma 6 13 level 5 was considered unacceptably toxic, and the MTD was Prostate cancer 12 28 established at dose level 4, i.e., 50 mg/m2 Doxil in combination Colon cancer 8 13 with 60 mg/m2 cisplatin every 4 weeks. As to cisplatin dose, it a Response based on Ͼ50% drop of the prostate-specific antigen is still possible that it can be escalated slightly further, especially level from baseline on at least two consecutive measurements 1 month with the help of hematopoietic colony-stimulating factors. How- apart. ever, because this study was designed to examine toxicity within a maximal target dose of 60 mg/m2, we did not proceed with further escalation. patients, at dose levels 2, 4, and 5 (1 each). After Ն4 courses of Antitumor Activity. The antitumor responses obtained therapy, we observed two more events of PPE, grade 2 and in 24 assessable patients receiving at least 2 cycles of treat- grade 3, at dose levels 1 and 5, respectively. Thus, PPE was not ment are listed in Table 5. Seven patients achieved a partial an important side effect for this cisplatin-Doxil combination, at response. In all but 1 patient, these responses were based on least on a short-term (3 cycles) treatment basis. disease measurable by computed tomography scan. In 1 pa- Other nonhematological toxicities observed included the tient with prostate cancer, the response was based on a Ͼ50% following: mild to moderate alopecia in 2 patients; skin pigmen- decrease of the prostate-specific antigen level. Three of the tation spots, which were usually reversible, in 2 patients; pro- responders were patients with recurrent epithelial ovarian tracted nausea in 1 patient; and grade 2–3 fatigue in 3 patients cancer. Some of these responses were long-lasting as indi- with a large bulk of tumor during the first cycle of treatment, a cated by the long time to disease progression in responding symptom that did not reappear in subsequent cycles. In addition, patients (median, 17 months; range, 5–32 months). Addition- there were 4 patients who experienced cisplatin-related toxici- ally, 3 patients remained with stable disease for 6–12 months ties, in whom cisplatin was discontinued: grade 1 nephrotoxicity after the start of combination therapy. Responses were seen at (shown by a rise in creatinine to 150 ␮M) in 1 patient at dose all dose levels tested. level 4 after one course; ototoxicity (hearing loss) in 2 patients Doxil Plasma Levels. The 1-h postinfusion plasma lev- at dose levels 2 and 5, after 4 and 5 courses of cisplatin; els of Doxil showed a gradual increase with dose escalation

erated in patients receiving a cisplatin preinfusion. It should be noted that none of the control group (breast and prostate cancer) patients had received treatment with cisplatin or carboplatin, although all breast cancer patients had been pretreated with other forms of chemotherapy prior to entering the Doxil study.

DISCUSSION The anthracycline antibiotic doxorubicin and cisplatin are potent anticancer compounds widely used in a broad spectrum of malignancies. Because of their different and nonoverlapping toxic effects, there has been considerable interest in combining them. Cisplatin and doxorubicin are widely used as the main

Fig. 1 Cmax (plasma concentration at 1-h postinfusion) of Doxil in components of many chemotherapy regimes in the treatment of patients receiving Doxil-cisplatin (Doxil dose, 40, 50, or 60 mg/m2)or a variety of advanced cancers. Doxil only (Doxil dose, 60 mg/m2; from Refs. 11, 12). Both of the 2 However, cumulative dose-dependent myocardial damage groups receiving 60 mg/m were well matched in terms of sex distri- and high degree of myelosuppression associated with doxoru- bution (male:female ratio, 5:3 for Doxil-cisplatin and 7:6 for Doxil only) and median age [57 years (range, 21–73 years) for Doxil-cisplatin and bicin may result in life-threatening toxicity and entail often a 60 years (range, 32–75 years) for Doxil only]. Bars, SD. significant dose compromise and even treatment interruption in responding patients. This is not the case with regard to Doxil, which has a markedly different toxicity profile. Recently reported data in a group of 40 patients receiving between 500 and 1500 mg/m2 (median, 635 mg/m2) suggest that Doxil can be administered at high cumulative doses with none of the patients developing clinical congestive heart failure or ventriculographic signs of drug-induced damage (14). This observation is con- cordant with results of animal studies pointing to reduced cardiotoxicity of Doxil as compared with free doxorubicin (15) and with clinical data reporting low endomyocardial biopsy scores in 10 Doxil-treated patients with AIDS-related Kaposi’s sarcoma relative to controls receiving free doxorubicin (16). In addition, Doxil causes mild myelosuppression (7), in contrast to profound neutropenia associated with bolus administration of doxorubicin at the standard 3-week schedule without growth 2 Fig. 2 Fraction of Cmax remaining in plasma at 7 days postinfusion of factor support. Unfortunately, the MTD of Doxil (50 mg/m Doxil in patients receiving Doxil-cisplatin (Doxil dose, 60 mg/m2)or every 4 weeks) is actually lower than that of conventionally 2 Doxil only (Doxil dose, 60 mg/m ; from Refs. 11, 12). The difference administered doxorubicin (60 mg/m2 every 3 weeks) because of between the two groups was statistically significant (P ϭ 0.0173, t test). Both groups were well-matched in terms of sex distribution (male: dose-limiting mucocutaneous toxicities, resulting from the ac- female ratio, 5:3 for Doxil-cisplatin and 7:6 for Doxil only) and median cumulation of large amounts of Stealth liposomes in skin and age [57 years (range, 21–73 years) for Doxil-cisplatin and 60 years probably also mucosas (7). (range, 32–75 years) for Doxil only]. Bars, SD. Besides toxicity considerations, a unique feature of Doxil is its high tumor accumulation (17–19), attributable to the ability of Stealth liposomes to extravasate across the leaky microvas- culature of tumors (20). This feature can make certain tumors from 14.7 Ϯ 1.9 for 40 mg/m2, to 17.3 Ϯ 3.0 for 50 mg/m2, and more susceptible to the antitumor effects of Doxil relative to 23.3 Ϯ 5.5 for 60 mg/m2 (mean Ϯ SD, in mg/l) . As seen in Fig. free doxorubicin. In the specific case of epithelial ovarian can- 1, the drug levels at 60 mg/m2 were similar to those obtained in cer, Doxil is a recognized option in the treatment of recurrences a group of patients with breast and prostate cancer receiving (8, 9), whereas the role of doxorubicin is controversial (21). single-agent Doxil also at 60 mg/m2 (23.9 Ϯ 4.7 mg/l) within Because Doxil appears to be a promising form of deliver- the frame of parallel clinical studies (11, 12), indicating that the ing doxorubicin with potential enhancement of antitumor activ-

peak plasma levels (Cmax) of Doxil are not affected by cisplatin ity and a decrease of some of the undesirable toxicities, this drug preinfusion. However, when the 7-day postinfusion levels were is a candidate to replace doxorubicin in cisplatin-doxorubicin examined at the 60 mg/m2 dose level, patients receiving Doxil regimes and perhaps other combination chemotherapy regimes. after cisplatin had significantly lower doxorubicin levels (3.3 Ϯ The rationale for a cisplatin-Doxil combination is also based on 1.0 mg/l; n ϭ 8) than those receiving single-agent Doxil (5.3 Ϯ the fact that cisplatin and Doxil toxicities do not overlap: on the 2.1 mg/l; n ϭ 13) with P ϭ 0.0195 (t test). This observation one hand, the nephrotoxicity, neurotoxicity, and ototoxicity of remains valid when the 7-day postinfusion Doxil levels are cisplatin are totally absent from Doxil; on the other hand, the expressed as a percentage of the 1-h postinfusion levels (Fig. 2) mucositis and skin toxicity of Doxil are totally absent from and suggests that the first week of clearance of Doxil is accel- cisplatin. Therefore, the Doxil-cisplatin combination is likely to

be tolerated without enhanced toxicity. The choice of cisplatin level data, may explain the low incidence and severity of PPE over carboplatin was based mainly on the fact that carboplatin is observed in this study. Cisplatin has been shown to cause in more myelosuppressive than cisplatin, and, therefore, the risk of vitro and in vivo murine macrophage activation (24–26). If these overlapping myelosuppression with Doxil is greater. results are extrapolated to human macrophages, one possible All of this led us to undertake a Phase I trial with Doxil and explanation is that accelerated clearance of liposomes after cisplatin combination chemotherapy. We chose the 4-week cisplatin treatment is related to transient macrophage activation schedule for the combination because of prior observations induced by cisplatin. indicating that dose, schedule, and intensity of Doxil are impor- Antitumor responses to the combination of Doxil and cis- tant determinants of PPE. Thus, doses of 60, 50, and even 45 platin were observed in a variety of tumor types, some of them mg/m2 of single-agent Doxil are intolerable with regard to skin with remarkably long times to disease progression exceeding 1 toxicity on a 3-week schedule and have to be delivered on a year in several cases. Of note, 3 of 4 ovarian cancer patients 4-week or longer schedule (4, 8, 10). The dose-limiting toxici- responded with objective partial response. In contrast, none of 6 ties in this study were a combination of mucositis and neutro- patients with soft tissue sarcoma responded. Both groups of penia. Mucositis was more frequent and severe in patients patients had been pretreated with chemotherapy. Lack of re- receiving higher dose levels. It reached grade 3 in 4 of 17 sponse of sarcomas to Doxil has been reported (27). patients treated at dose levels 3, 4, and 5. The incidence and The results of our study established the MTD at 50 mg/m2 severity of mucositis reported in this study are roughly similar to of Doxil in combination with 60 mg/m2 of cisplatin every 4 those in patients receiving single-agent Doxil at similar doses (4, weeks. This indicates that Doxil can be given at full MTD and 2 8, 10, 11, 13). This is consistent with the finding that Doxil Cmax recommended dose intensity (12.5 mg/m /week) when com- values are not affected by cisplatin (Fig. 1) and with the reported bined with cisplatin. Thus far, in other combinations of Doxil observation of a direct correlation among dose, Cmax, and mu- with different anticancer drugs (taxanes, vinorelbine, and gem- cositis (11). citabine), there has been a need to trim Doxil dose intensity to Surprisingly, neutropenia was found to be a major dose- Յ10 mg/m2/week (28–33). In conclusion, the present study has limiting side-effect of the Doxil-cisplatin combination. Thus, 3 identified the Doxil-cisplatin combination as a promising regi- cases of grade 4 neutropenia (1 with neutropenic fever) were men for further evaluation in the treatment of malignant tumors, detected among 13 patients treated at dose levels 4 and 5. It is exhibiting a different toxicity profile from single-agent Doxil. noteworthy that recovery from severe neutropenia was relatively quick and, in all cases, within 7 days. Yet, the fact that neutro- REFERENCES penia was one of the dose-limiting toxicities was quite unex- 1. Martin, F. M. Clinical pharmacology and anti-tumor efficacy of pected. On the one hand, neutropenia is not a prominent feature Doxil (pegylated liposomal doxorubicin). In: D. D. Lasic and D. Papa- of cisplatin as a single agent. On the other hand, Doxil as a hadjopoulos (eds.)., Medical Applications of Liposomes, pp. 635–688. Amsterdam: Elsevier, 1998. single agent is generally associated with mild neutropenia. For 2. Wu, N. Z., Da, D., Rudoll, T. L., Needham, D., Whorton, A. R., and instance, Gordon et al. (9) reported grade 4 neutropenia in 4% 2 Dewhirst, M. W. Increased microvascular permeability contributes to of 89 patients receiving Doxil 50 mg/m with only 1 case of preferential accumulation of Stealth liposomes in tumor tissue. Cancer neutropenic fever. Therefore, attention should be paid to the Res., 53: 3765–3770, 1993. possibility of a synergistic myelosuppressive effect when cis- 3. Yuan, F., Leunig, M., Huang, S. K., Berk, D. A., Papahadjopoulos, platin and Doxil are combined. D., and Jain, R. K. Microvascular permeability and intestinal penetration An important and encouraging observation of the present of sterically stabilized (stealth) liposomes in a human tumor xenograft. Cancer Res., 54: 3352–3356, 1994. study was the low incidence and severity of PPE in relation to 4. Uziely, B., Jeffers, S., Isacson, R., Kutsch, K., Wei-Tsao, D., data available for single-agent Doxil (4, 8, 10, 11, 12). A similar Yehoshua, Z., Libson, E., Muggia, F. M., and Gabizon, A. Liposomal finding has been reported in abstract form using a slightly doxorubicin: antitumor activity and unique toxicities during two compli- different regime of cisplatin and Doxil (22). The most severe mentary Phase I studies. J. Clin. Oncol., 13: 1777–1785, 1995. PPE manifestation reported in our study during the first 3 cycles 5. Stewart, S., Jablonowski, H., Goebel, F. D., Arasteh, K., Spittle, M., was grade 2, which occurred in only 1 patient at dose level 5. Rios, A., Abolafia, D., Galleshaw, J., and Dezube, B. J. Randomized comparative trial of pegylated liposomal doxorubicin versus bleomycin For comparison, grades 2 and 3 PPE have been reported in 34% and vincristine in the treatment of AIDS-related Kaposi’s sarcoma. of 32 breast cancer patients treated with single-agent Doxil at 45 J. Clin. Oncol., 16: 683–691, 1998. 2 mg/m every 4 weeks (10) and in 31% of 89 ovarian cancer 6. Northfelt, D. W., Dezube, B. J., Thommes, J. A., Miller, B. J., patients treated with treated with single-agent Doxil at 50 Fishl, M. A., Friedman-Kien, A., Kaplan, L. D., DuMond, C., Ma- mg/m2 every 4 weeks (9). Our limited pharmacokinetic data melok, R. D., and Henry, D. H. Pegylated liposomal doxorubicin versus may explain these results. A positive correlation between risk of doxorubicin, bleomycin and vincristine in the treatment of AIDS-related Kaposi’s sarcoma: results of a randomized Phase III clinical trial. PPE and plasma half-life of Doxil has been found previously J. Clin. Oncol., 16: 2445–2451, 1998. 2 (11). Patients receiving 60 mg/m of Doxil after cisplatin had 7. Gabizon, A. A. Pegylated liposomal doxorubicin: metamorphosis of significantly lower 7-day postinfusion levels than those receiv- an old drug into a new form of chemotherapy. Cancer Investig., 19: ing single-agent Doxil (Fig. 2). This change cannot be attributed 424–436, 2001. to the corticosteroid premedication, because no pharmacokinetic 8. Muggia, F. M., Hainsworth, J. D., Jeffers, S., Miller, P., Groshen, S., changes were observed when we used a similar premedication Tan, M., Roman, L., Uziely, B., Muderspach, L., Garcia, L., Burnett, A., Greco, F. A., Morrow, C. P., Paradiso, L., and Liang, L. J. Phase II regime in a recent study combining amifostine and Doxil (23). study of liposomal doxorubicin in refractory ovarian cancer: antitumor An accelerated clearance of Doxil in patients receiving the activity and toxicity modification by liposomal encapsulation. J. Clin. cisplatin-Doxil combination, as suggested by the 7-day plasma Oncol., 15: 987–993, 1997.

9. Gordon, A. N., Granai, C. O., Rose, P. G., Heinsworth, J., Lopez, A., 21. Omura, G. A., Bundy, B. N., Berek, J. S., Curry, S., Delgado, G., Weissman, C., Rosales, R., and Sharpington, T. Phase II study of and Mortel, R. Randomized trial of cyclophosphamide plus cisplatin liposomal doxorubicin in platinum- and paclitaxel-refractory epithelial with or without doxorubicin in ovarian carcinoma: a Gynecologic ovarian cancer. J. Clin. Oncol., 18: 3093–3100, 2000. Oncology Group study. J. Clin. Oncol., 7: 457–465, 1989. 10. Ranson, M. R., Carmichael, J., O’Byrne, K., Stewart, S., Smith, D., 22. Klein, P., Wasserheit, C., Hochster, H., Chachoua, A., Speyer, J. L., and Howell, A. Treatment of advanced breast cancer with sterically- Oratz, R., Downey, A., Sorich, J., Kotcher, L., Eisenberg, T., and stabilized liposomal doxorubicin: results of a multicenter Phase II trial. Muggia, F. Apparent protection of Doxil skin and oral toxicities when J. Clin. Oncol., 15: 3185–3191, 1997. combined with cisplatin: results of a Phase I study. Proc. Am. Soc. Clin. 11. Lyass, O., Uziely, B., Ben-Yosef, R., Tzemach, D., Heshing, N. I., Oncol., 18: 217a, 1999. Lotem M., Brufman, G., and Gabizon, A. Correlation of toxicity with 23. Lyass, O., Lotem, M., Edelmann, D., Hubert, A., and Gabizon, A. pharmacokinetics of pegylated liposomal doxorubicin (Doxil) in meta- Protective effect of amifostine on Doxil/Caelyx-induced palmar-plantar static breast carcinoma. Cancer (Phila.), 89: 1037–1047, 2000. erythrodysesthesia in patients with advanced cancer. Proc. Am. Soc. 12. Hubert, A., Lyass, O., Pode, D., and Gabizon, A. Doxil (Caelyx): an Clin. Oncol., 20: 99b, 2001. exploratory study with pharmacokinetics in patients with hormone- 24. Palma, J. P., Aggarwal, S. K., and Jiwa, A. Murine macrophage refractory prostate cancer. Anticancer Drugs, 11: 123–127, 2000. activation after cisplatin or carboplatin treatment. Anticancer Drugs, 3: 13. Gabizon, A., Chisin, R., Amselem, S., Druckmann, S., Cohen, R., 665–676, 1992. Goren, D., Fromer, I., Peretz, T., Sulkes, A., and Barenholz, Y. Phar- 25. Muenchen, H. J., and Aggarwal, S. K. Immune system activation by macokinetic and imaging studies in patients receiving a formulation of cisplatin and its analog poly-plat: an in vitro and in vivo study. Anti- liposome-associated Adriamycin. Br. J. Cancer, 64: 1125–1132, 1991. cancer Drugs, 9: 93–99, 1998. 14. Safra, T., Muggia, F., Jeffers, S., Tsao-Wei, D., Groshen, S., Lyass, 26. Burkitt, K., and Aggarwal, S. K. Immune system activation by O., Henderson, R., Berry, G., and Gabizon, A. Pegylated liposomal CDDP and poly-plat. Anticancer Res., 20: 2729–2737, 2000. doxorubicin (Doxil): reduced clinical cardiotoxicity in patients reaching or exceeding cumulative doses of 500 mg/m2. Ann. Oncol., 11: 1029– 27. Garcia, A. A., Kempf, R. A., Rogers, M., and Muggia, F. M. A 1033, 2000. Phase II study of Doxil (liposomal doxorubicin): lack of activity in poor 15. Working, P. K., Newman, M. S., Sullivan, T., and Yarrington, J. prognosis soft tissue sarcomas. Ann. Oncol., 9: 1131–1133, 1998. Reduction of the cardiotoxicity of doxorubicin in rabbits and dogs by 28. Burstein, H. J., Ramirez, M. J., Petros, W. P., Clarke, K. D., encapsulation in long-circulating, pegylated liposomes. J. Pharmacol. Warmth, M. A., Marcom, P. K., Matulonis, U. A., Parker, L. M., Har- Exp. Ther., 289: 1128–1133, 1999. ris, L. N., and Winer, E. P. Phase I study of Doxil and vinorelbine in 16. Berry, G., Billingham, M., Alderman, E., Richardson, P., metastatic breast cancer. Ann. Oncol., 10: 1113–1116, 1999. Torti, F., Lum, B., Patek, A., and Martin, F. J. The use of cardiac 29. Schwonzen, M., Kurbacher, C. M., and Mallmann, P. Liposomal biopsy to determine reduced cardiotoxicity in AIDS Kaposi’s sar- doxorubicin and weekly paclitaxel in the treatment of metastatic breast coma patients treated with pegylated liposomal doxorubicin. Ann. cancer. Anticancer Drugs, 11: 681–685, 2000. Oncol., 9: 711–716, 1998. 30. Campos, S., Penson, R. T., Matulonis, U. A., Berkowitz, R. S., 17. Koukourakis, M. I., Koukouraki, S., Giatromanolaki, A., Archiman- Duska, L. R., Fuller, A. F., Goodman, A. K., Nikrui, N., Ellen, S. E., dritis, S. C., Skarlatos, J., Beroukas, K., Bizakis, J. G., Retalis, G., Muto, M., McNeill, K. M., and Genovese, C. A Phase II and phar- Karkavitsas, N., and Helidonis, E. S. Liposomal doxorubicin and con- macokinetic/dynamic study of Doxil and weekly paclitaxel chemo- ventionally fractionated radiotherapy in the treatment of locally ad- therapy for recurrent mullerian tumors. Proc. Am. Soc. Clin. Oncol., vanced non-small-cell lung cancer and head and neck cancer. J. Clin. 19: 410a, 2000. Oncol., 17: 3512–3521, 1999. 31. Chodkiewiz, C., Wasserheit, C., Downey, A., Sorich, J., Liebes, L., 18. Symon, Z., Peyser, A., Tzemach, D., Lyass, O., Sucher, E., Phillips, Z., Hochster, H., Oratz, R., Pavlick, A., Speyer, J., Fried- Shezen, E., and Gabizon, A. Selective delivery of doxorubicin to pa- man, M., and Muggia, F. Doxil and docetaxel: pharmacokinetic results tients with breast carcinoma metastases by Stealth liposomes. Cancer at the recommended Phase II dose. Proc. Am. Soc. Clin. Oncol., 19: (Phila.), 86: 72–78, 1999. 212a, 2000. 19. Harrington, K. J., Mohammadtaghi, S., Uster, P. S., Glass, D., 32. Tauer, K., Hussein, A., Birch, R., Dickson, L., Thompson, D., Peters, A. M., Vile, R. G., and Stewart, J. S. Effective targeting of Schwartzberg, L., and Nelson-Hill, S. A Phase I study of Taxotere and solid tumors in patients with locally advanced cancers by radiola- Doxil in refractory cancer. Proc. Am. Soc. Clin. Oncol., 19: 230a, 2000. beled Pegylated liposomes. Clin. Cancer Res., 7: 243–254, 2001. 33. Tobias, D. H., Runowicz, C., Mandeli, J., Muggia, F., Astrow, A., 20. Gabizon, A. A. Stealth liposomes and tumor targeting: one step Cohen, C., and Argenta, A. A Phase I trial of gemcitabine and Doxil for further in the quest for the magic bullet. Clin. Cancer Res., 7: recurrent epithelial ovarian cancer. Proc. Am. Soc. Clin. Oncol., 19: 223–225, 2001. 392a, 2000.

Downloaded from clincancerres.aacrjournals.org on September 25, 2021. © 2001 American Association for Cancer Research. Phase I Study of Doxil-Cisplatin Combination Chemotherapy in Patients with Advanced Malignancies

Olga Lyass, Ayala Hubert and Alberto A. Gabizon

Clin Cancer Res 2001;7:3040-3046.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/7/10/3040

Cited articles This article cites 32 articles, 13 of which you can access for free at: http://clincancerres.aacrjournals.org/content/7/10/3040.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/7/10/3040.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/7/10/3040. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.