Phase I study of pegylated liposomal doxorubicin, paclitaxel, and cisplatin in patients with advanced solid tumors

Annuls of Oncology 12: 1743-1747, 2001. © 2001 Kluwer Academic Publishers Printed in the Netherlands.

Original article Phase I study of pegylated liposomal doxorubicin, paclitaxel, and cisplatin in patients with advanced solid tumors C. Eng,1 A. M. Mauer,1 G. F. Fleming,1 D. Bertucci,1 J. Rotmensch,2 R. H. Jacobs' & M.J. Ratain1'3 1

Department of Medicine, Section of Hematology Oncology, 2 Department of Obstetrics and Gynecology, 'Committee on Clinical Pharmacology. University of Chicago, Chicago, USA

Summary Background: The combination of doxorubicin, paclitaxel, and cisplatin has activity in gynecologic malignancies but requires colony stimulating factor (G-CSF) support. Moreover, there is concern about cardiotoxicity with doxorubicin/paclitaxel combinations. Pegylated liposomal doxorubicin may result in less myelosuppression and cardiac toxicity than free doxorubicin. The purpose of this study was to determine the maximal tolerated dose of pegylated liposomal doxorubicin with fixed doses of paclitaxel and cisplatin without using G-CSF support in advanced solid malignancies. Patients and methods: Twenty-three patients were enrolled; none of the patients had received prior doxorubicin. Patients received paclitaxel (90 mg/m 2 for dose level one. escalating to 135 mg/m 2 for all subsequent dose levels), with a fixed dose of cisplatin (60 mg/m 2 ), followed by escalating doses of pegylated liposomal doxorubicin every 21 days.

Introduction Standard first-line therapy for ovarian carcinoma currently consists of a platinum agent combined with a taxane. This combination has clearly demonstrated activity in endometrial cancer [1]. The Ovarian Cancer Meta-Analysis Project, published with data obtained in the pre-taxane era, suggested that the addition of an anthracycline to a platinum-based regimen provided a small but significant seven percent survival advantage [2]. Therefore, there has been interest in combining an anthracycline with a platinum/taxane regimen for use in ovarian cancer and other gynecologic malignancies. Pegylated liposomal doxorubicin has documented activity in ovarian cancer. Phase II studies in platinumrefractory patients using doses ranging from 40-50 mg/m 2 have reported response rates of 25%-59% [3, 4]. A preliminary report of a phase III randomized trial comparing pegylated liposomal doxorubicin (50 mg/m" every 28 days) to topotecan (1.5 mg/m 2 days 1-5 every 21 days) as first-line salvage therapy for 474 previously treated women reported overall response rates of 20%

Results: A total of 73 cycles was administered. Grade 4 neutropenia was seen after cycle one in two of eight patients receiving 30 mg/m 2 of pegylated liposomal doxorubicin and three of seven patients receiving 40 mg/m 2 of pegylated liposomal doxorubicin when combined with 135 mg/m 2 of paclitaxel and 60 mg/m 2 of cisplatin. Two additional patients at the 40 mg/m 2 dose level developed grade 4 neutropenia following cycles 2 and 5. The mean decline in left ventricular ejection fraction (LVEF) after 2 cycles was 5 percentage points (/> = 0.012). Conclusion: The combination of pegylated liposomal doxorubicin, paclitaxel and cisplatin is feasible without G-CSF support.

Key words: cardiotoxicity, paclitaxel, cisplatin, pegylated liposomal doxorubicin

vs. 17%, respectively. A subpopulation of platinum-refractory patients demonstrated overall response rates of 12% and 7% in the liposomal doxorubicin and topotecan arms, respectively, but this difference was not found to be statistically significant [5]. The primary toxicity of liposomal doxorubicin was grade 3-4 palmar-plantar erythrodysesthesia (PPE) seen in 23% of patients, with only 12% of patients experiencing neutropenia. The goal of liposomal encapsulation of doxorubicin is to enhance drug delivery while decreasing toxicity. In comparison to free doxorubicin, pegylated liposomal doxorubicin results in a greater plasma area under the concentration curve (AUC) [6]. The pegylated liposomal preparation is not taken up by the reticuloendothelial system, which may result in a longer circulation time. Moreover, the drug may be extravasated through the abnormal permeable vessels found to be characteristic of several tumors, possibly allowing higher drug delivery concentration to the tumor relative to normal tissue [7, 8]. We therefore conducted this phase 1 study to determine the maximal tolerated dose of pegylated liposomal

1744 doxorubicin that could be administered with fixed doses of cisplatin and paclitaxel without G-CSF support in patients with minimally pretreated solid tumors. It is likely that this triplet regimen has activity in many solid tumors, not just gynecologic malignancies.

Table I. Dose levels. Dose Level

Doxil (mg/m 2 )

Paclitaxel (mg/m 2 )

Cisplatin (mg/m 2 )

1

20 20 30 40

90 135 135 135

60 60 60 60

2

Patients and methods

3 4

Eligibility criteria Patients were required to have a histogically proven diagnosis of an advanced solid tumor that had progressed despite standard treatment or for which there was no standard therapy. Measurable or evaluable disease was required. Measurable disease was defined as clearly measurable in two dimensions by clinical exam. X-ray, computed tomography (CT) scan, or magnetic resonance imaging (MRI). Only one or fewer prior chemotherapy regimens were permitted. The prior cumulative dose of doxorubicin was not to exceed 300 mg/m 2 . and at least 4 weeks were to have elapsed since previous chemotherapy or radiation therapy. Prior radiation therapy to the entire pelvis or lumbo-sacral spine was not allowed. Other eligibility criteria included a Karnofsky performance status of > 70%, age ^ 18 years, absolute neutrophil count (ANC) >l500/ul and platelets > 100.000/ul. total bihrubin within normal limits: aspartate aminotransferase less than three times upper limits of normal, serum creatinine < 1.5 mg/dl (or calculated creatinine clearance > 35 ml/min/m 2 using the Cockcroft and Gault formula) [9]. Patients were required to have a left ventricular ejection fraction by a multigated angiogram (MUGA) of greater than or equal to 50% and no prior history of congestive heart failure. Patients with uncontrolled angina, history of complete heart block without placement of a cardiac pacemaker, myocardial infarct in the past six months, or serious uncontrolled cardiac arrythmias were ineligible Patients with ^ grade 2 peripheral neuropathy were excluded. Women of childbearing potential were required to have a negative pregnancy test and agree to utilize birth control while on study. All subjects gave written informed consent in accordance with Federal and institutional guidelines

Drug administration Recommended hydration included 500 ml of 0 9% normal saline over 1-2 hours prior to treatment and 500 ml of 0.9% normal saline after chemotherapy administration. Paclitaxel (Taxol, Bristol-Myers Squibb Oncology, Princeton. New Jersey) was commercially obtained. Paclilaxel was diluted in 250 ml of 0.9%) sodium chloride injection, or 5%> dextrose injection (D 5 W), and infused over 60 minutes on day I of each cycle. Paclitaxel premedications included 20 mg of oral dexamethasone 14 and 7 hours prior to treatment, along with 20 mg famotidine and 25-50 mg diphenhydramine intravenously 30 minutes prior to treatment. Cisplatin was obtained commercially and diluted in 500 ml of 0.9% sodium chloride with 12 5 g of manmtol and infused over one hour immediately following paclitaxel. The anti-emetic regimen chosen prior to the infusion of cisplatin was left to the discretion of the treating physician. Pegylated liposomal doxorubicin (Doxil®, Alza Pharmaceuticals, Inc.) was supplied in sterile vials each containing 20 mg of doxorubicin hydrochloride at a concentration of 2.0 mg/ml. Doxil was diluted in 250-500 ml D5W and infused over 60 minutes following infusion of cisplatin. Patients received pyridoxine 50 mg by mouth three times daily for days 2-21 of each cycle in an attempt to prevent palmar-plantar erythodysesthesia [10]. The treatment regimen was repeated every 21 days Treatment was continued for a maximum of six cycles in patients who achieved objective responses or stable disease after two successive cycles. G-CSF (5 ug/kg) could be administered subculaneously for neutropenic fevers at the discretion of the treating physician.

Study design Dose levels are shown in Table 1. A minimum of three patients were evaluated at each dose level at which there was at most grade 0-2 toxicity. If a dose-limiting toxicity occurred in one of the first three patients at a dose level, then up to three additional patients were treated at that dose level to determine if the maximal tolerated dose (MTD) was exceeded. Only toxicities reported on cycle I were considered dose limiting (DLT) and prevented further dose escalation. Hematologic events that were considered dose limiting included ANC < 500 for > 4 days duration, neutropenic fever requiring hospitalization, and grade 4 thrombocytopenia. Non-hematologic DLT's were defined as events 5= grade 3 and grade 2 toxicity lasting > 20 days other than alopecia, weight loss, allergic reaction, or fatigue.

Patient assessment Toxicities were assessed according to the NCI Common Toxicity Criteria version 1.0. Pretreatment laboratory testing were obtained including a complete blood count, and metabolic panel including sodium, potassium, blood urea nitrogen, creatinine. total bilirubin, aspartate transaminase. alanine transaminase, and serum magnesium. Baseline radiologic studies and MUGA scan were to be obtained within 30 days of initiating therapy. Patients had a weekly complete blood count and metabolic panel during the first cycle. History, physical exam, complete blood count, and metabolic panel were required prior to each cycle. Radiologic studies to evaluate disease progression and MUGA scans were repeated every two cycles. Although response was not an endpoint of this study, and measurable disease was not required, responses were noted. A complete response (CR) was defined as disappearance of all clinical, biochemical, symptomatic, and radiographic evidence of tumor. A partial response (PR) was defined as a 5= 50%o decrease in the sum of the products of the two greatest perpendicular diameters of all measured lesions. A simultaneous increase in size of any lesion or the appearance of a new lesion was not allowed. Progressive disease (PD) was defined as a ^ 2 5 % increase in the sum of the products of the longest perpendicular diameters of all measured indicated lesions compared to the smallest previous measurement or the appearance of a new lesion Stable disease (SD) was defined as fitting none of the above criteria Patients with responding or stable disease and no DLT were allowed to continue on treatment at the same dose level. Patients developing a DLT could be retreated at the next lowest dose level, if they had evidence of tumor response or had only one cycle of chemotherapy. Patients with progressive disease were withdrawn from the study

Results Twenty-three patients were entered on this trial between May 1997 and March 2000. None had received prior doxorubicin chemotherapy. Patient characteristics are shown in Table 2. A total of 73 cycles was administered. Three patients completed the full six cycles. Ten patients received four or more cycles. Eighteen patients had

1745 Table 2 Patient characteristics.

Table 3. Neutrophil toxicity following cycle 1.

Characteristics 2~_

Study dose level

1 16 64 37-82

4

10 13

Table 4 Worst neutrophil toxicity in any cycle.

Total number of patients enrolled Sex Male Female Median age (years) Range Karnofsky performance status 90-100 70-80 Prior treatment None XRT Chemotherapy Other (immunotherapy) Number of cycles Range Median Number of patients treated for > 2 cycles Tumor type Cervical Esophageal Melanoma Endometrial Ovarian Unknown primary adenocarcinoma Bladder Urethral Gastric Carcinoid Lung

Starting dose level

8 9 5 2

Neutropenia

7

n

Grade 1

Grade 2

Grade 3

Grade 4

1

2

1

3

Grade 3

Grade 4

Neutropenia Grade 1

Grade 2

0-6 3 18 6 3 3 2 2 2

Hematologic toxicity

received two or more cycles and hence were considered evaluable for response. Four patients were removed from study after cycle 1; one was non-compliant, one developed an elevated creatinine, and two developed an allergic reaction to paclitaxel despite standard premedications and were not rechallenged. The two patients who did not tolerate paclitaxel are not considered further in this analysis. One patient died from pneumonia resulting in sepsis, ten days following cycle 2. She was not neutropenic during her hospitalization; her death was not believed to be treatment-related.

The combination of pegylated liposomal doxorubin, paclitaxel, and cisplatin was well tolerated for dose levels 1-3 (Tables 3 and 4). Five patients on dose level 4 (40 mg/m 2 of pegylated liposomal doxorubicin) experienced grade 4 neutropenia at some point (cycle 1: three patients, cycles 2 and 5: one patient each, respectively) during their chemotherapy: one of them developed febrile neutropenia that was treated as an outpatient with empiric antibiotics. All five were retreated at 30 mg/m 2 of pegylated liposomal doxorubicin; four of the five required a second reduction. Eight patients were treated at dose level 3 and received a total of 23 cycles of chemotherapy; four patients received ^ 3 cycles. Only two patients developed grade 4 neutropenia at any point in their therapy. No grade 1 or greater thrombocytopenia was seen at any dose level. Non-hematologic toxicity Non-hematologic dose limiting toxicities were mild (Table 5). Grade 1 and 2 fatigue, anorexia, and nausea were reported frequently. No episodes of dose-limiting stomatitis or palmar-plantar erythrodysesthesia (PPE)

Table 5 Worst non-hematologic toxicity of any cycle Dose level

1 2 3 4

n

3 3 8 7

Fatigue

PPE

Anorexia

Renal insufficiency

Gr 1/2

Gr3

Gr4

Gr 1/2

Gr3

Gr4

Gr 1/2

Gr3

Gr4

Gr 1/2

Gr3

Gr4

2 3 5 4

1 0 1 1

0 0 0 0

1 1 5 3

1 0 0 0

0 0 0 0

0 1 1 2

0 0 0 0

0 0 0 0

1 1 2 1

0 0 0 0

0 0 0 0

Abbreviations. Gr - grade; PPE - palmar-plantar erythrodysesthesia.

1746 Tahle 6 Mean change in left ventricular ejection fraction from baseline. Dose level

After 2 cycles No. observed Mean change After 4 cycles No. observed Mean change After 6 cycles No. observed Mean change

20/90 mg/m 2

20/135 mg/m 2

30/135 mg/m 2

40/135 mg/m 2

3 S.3%

3 -1.3%

6 -8.2%

5 -2.4%

2 2%

2 -5.5%

3 -3.7%

4 -2.5%,

Cycle 2

Baseline

40

50

60

70

80

Figure 1. LVEF at baseline and following cycle 2.

0

1 1%

1 -2%

1 5%,

were reported. No patient had PPE after the first two cycles. Grade 1/2 PPE was seen in 4 of 13 patients resulting in no treatment-related interruptions or dosereductions; one patient each at 20 mg/m 2 and 30 mg/m 2 after the completion of cycle 3; two patients at 40 mg/m 2 (one patient following cycle 3, a second patient following cycle 4). Cardiac toxicity Symptomatic congestive heart failure was not observed. All 23 patients had a baseline MUGA scan prior to initiating study. All subsequent MUGA scans were obtained at the same institution as the baseline study. Eighteen of these patients had at least one subsequent MUGA scan for comparison; in seventeen patients this was following cycle 2 (Figure 1, Table 6). One patient did not receive a MUGA scan following cycle 2 but received a MUGA following both cycles 4 and 6. The median LVEFat baseline was 61% (range: 52%-78%). Following two cycles of chemotherapy, the median LVEF decreased to 57% (range: 33%—70%). Overall, the mean decline in left ventricular ejection fraction after receiving two cycles of chemotherapy was 5% (P = 0.012). Eight patients developed grade 1 or 2 decreases in LVEF. One 49-year-old woman at dose level 4 with a history of hypertension, hypercholesterolemia, and non-insulin dependent diabetes developed supraventricular tachycardia following cycle 2 requiring hospitalization for observation. She proceeded to receive a total of six cycles of chemotherapy; subsequent MUGA scans did not demonstrate a decline in ejection fraction after the completion of her treatment. Specific cardiac risk factors were not identified in these patients with large decreases in LVEF. At dose level 1, a 54-year-old man with a previously untreated unknown primary adenocarcinoma and no prior comorbidities was noted to have an asymptomatic decline in his LVEF from 78% to 54% following cycle 2 resulting in grade 2 toxicity. He developed progressive disease following cycle 2 and was not evaluated further. Two patients were removed from study for declines in LVEF. One 60-year-old woman with previously untreated adenocarcinoma of unknown primary at dose level 3 had a decline in her ejection fraction from 52%) to 33% after

completing two cycles. A second 41 year-old man with widely metastatic melanoma at dose level 4 had a decrease in his LVEF from 60% to 45% following 4 cycles of chemotherapy. He had been treated in the past with cisplatin, DTIC, BCNU, Dacarbazine, IL-2, and alphaINF. None of these patients had received prior radiation to the chest wall or mediastinum. Response Response was assessed in the 18 patients who completed ^ 2 cycles. Two patients had a CR, one with cervical cancer, and one with endometrial cancer. The endometrial cancer patient had presented with a central recurrence in a previously irradiated field, and remains in CR approximately two years after completing five cycles of protocol chemotherapy. Four patients (urethral, cervical, ovarian, and an unknown primary adenocarcinoma) achieved a PR.

Discussion We defined a regimen (30 mg/m 2 of pegylated liposomal doxorubicin, 60 mg/m 2 of cisplatin, 135 mg/m 2 of paclitaxel over three hours) that produced tolerable myelosuppression without requiring the administration of colony-stimulating factors. This suggested dose of paclitaxel is lower than that conventionally used to treat ovarian cancer, but is clearly active. There is no evidence of a steep dose-response curve with paclitaxel [11]. One of the rationales for developing this regimen was to decrease the amount of cardiac toxicity that has been reported in prior clinical trials. Doxorubicin can cause cardiotoxity presenting as congestive heart failure. This may be more frequent when paclitaxel is administered with doxorubicin. An early study of the paclitaxel/doxorubicin combination in metastatic breast cancer reported that 21% of patients developed clinical congestive heart failure at a median doxorubicin dose of 480 mg/m 2 [12]. Subsequent trials suggested the increased risk was more modest. In one overview, Gianni et al., retrospectively reviewed multiple clinical trials evaluating the combination of doxorubicin and paclitaxel in patients with advanced breast cancer to determine the incidence of cardiac toxicity [13]. In the 389 patients who received doxorubicin (50-60 mg/m 2 ) followed by paclitaxel (125-200 mg/m 2 ) for 4-8 cycles, the overall incidence of congestive heart failure was 4%. A recent Gynecological Oncology Group (GOG) phase I trial evaluated escalating doses of paclitaxel and free doxorubicin combined with fixed doses of cisplatin

1747 (60 mg/m 2 ) in patients with advanced gynecologic malignancies [14]. Ten of eighty patients (13%) were removed from study for declines in left ventricular ejection fraction (decline to < 45% or decline of > 20% of baseline value), although no symptomatic congestive heart failure was observed. Pegylated liposomal doxorubicin may be less cardiotoxic than bolus injection of free drug. Safra et al. noted that none of 41 patients treated with a cumulative dose of pegylated liposomal doxorubicin of ^ 500 mg/m 2 exhibited symptomatic congestive heart failure at a median follow-up of 1.2 years [15]. Median change in LVEF was - 1 % (range -15% to 9%). No correlation was found between the cumulative dose and the change in LVEF. Moreover, other studies examining cardiac biopsies to assess cardiac toxicity have suggested that Kaposi's sarcoma patients treated with pegylated liposomal doxorubicin demonstrate less cardiac toxicity than historical controls who received free doxorubicin [16]. Despite the use of modest doses of pegylated liposomal doxorubicin in our study, two patients (11%) demonstrated ^ 2 0 % decline in the baseline value of their LVEF on subsequent MUGA scans. These numbers are exceedingly small, but they are comparable to the changes observed in the recent GOG phase I trial of paclitaxel/cisplatin/free doxorubicin. Sporadic early cardiac toxicity was also observed in that study, and the LVEF was reported to have decreased by 43% of baseline value in at least one patient following only two cycles of that regimen [14]. No particular predisposing factors were identified in our patients to account for the greatest declines in LVEF. A larger study would be needed for a precise comparison of the cardiac toxicity of the two regimens, but we believe that the early declines seen in a few of our patients are similar to what is occasionally seen with a free doxorubicin/paclitaxel regimen. A number of other taxane/platinum/anthracycline combinations have also been evaluated and have shown activity in patients with gynecologic malignancies. However, the paclitaxel/ pegylated liposomal doxorubicin/ cisplatin regimen produced major responses, does not require G-CSF support, and is suitable for further testing where multiagent therapy is desired. We did not have many patients who received six cycles of therapy, a larger phase II trial would be suited to assess cumulative tolerance in specific groups of patients. It is possible sequential combination of two agents (doublets) or sequential single agents may be a less toxic way to deliver multiple drugs in many situations. Randomized trials will be required to assess efficacy of sequential vs. simultaneous administration of chemotherapy drugs. References 1. Zanotti KM, Belinson JL, Kennedy AW et al.The use of paclitaxel and platinum-based chemotherapy in uterine papillary serous carcinoma. Gynecol Oncol 1999; 74: 272-7.

2. Ovarian Cancer Meta-Analysis Project Cyclophosphamide plus cisplatin vs. cyclophosphamide. doxorubicin, and cisplatin chemotherapy of ovarian carcinoma: A meta-analysis. J Clin Oncol 1991:9: 1668-74 3. Muggia FM. Hainsworth JD. Jeffers S el al. Phase II study of liposomal doxorubicin in refractory ovarian cancer: Antitumor activity and toxicity modification by liposomal encapsulation. J Clin Oncol 1997: 15:987-93. 4 Israel VP, Garcia AA, Roman L et al. Phase II study of liposomal doxorubicin in advanced gynecological cancers. Gynecol Oncol 2000; 78: 143-7. 5. Gordon AN. Granai CO. Rose PG el al. Phase II study of liposomal doxorubicin in platinum- and pachlaxel -refractory epithelial ovarian cancer. J Clin Oncol 2000; 18: 3093-100. 6. Rahman A. Treat J. Roh JK et al. A phase I clinical trial and pharmacokinetic evaluation of liposome-encapsulated doxorubicin. J Clin Oncol 1990; 8: 1093-100. 7. Allen TM. Hansen C, Martin F et al. Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) show prolonged circulation half-lives in vivo. Biochim Biophys Acta 1991; 1066. 29-36. 8 Wu NZ. Da D. Rudoll TL et al. Increased microvascular permeability contributes to preferential accumulation of Stealth liposomes in tumor tissue. Cancer Res 1993; 53: 3765-70. 9. Cockcroft DW. Gault MH. Prediction of creatininc clearance from serum creatinine. Nephron 1976: 16 (1)- 31-41. 10. Vail DM. Chun R. Thamm DH et al. Efficacy of pyridoxinc to ameliorate the cutaneous toxicity associated with doxorubicin containing pegylated (Stealth) liposomes: A randomized, doubleblind clinical trial using a canine model. Clin Cancer Res 1998: 4: 1567-71. 11. Eisenhauer EA. ten Bokkel Huinink WW. Svvenerton KD et al European-Canadian randomized trial of paclitaxel in relapsed ovarian cancer: High-dose vs. low-dose and long vs. short infusion J Clin Oncol 1994; 12: 2654-66 12. Gianni L, Munzone E, Capri G el al. Paclitaxel by three-hour infusion in combination with bolus doxorubicin in women wilh untreated metastatic breast cancer: High antitumor efficacy and cardiac effects in a dose-finding and sequence-finding study. J Clin Oncol 1995; 13. 2688-99. 13. Gianni L. Dombernowsky P. Sledge G et al. Cardiac function following combination therapy withTaxol (T) and Doxorubin (A) for advanced breast cancer (ABC) Proc Am Soc Clin Oncol 1998; 17: A444. 14. Fleming GF. Fowler JM. Waggoner SE et al. Phase I trial of escalating doses of paclitaxel combined with fixed doses of cisplatin and doxorubicin in advanced gynecologic malignancies. J Clin Oncol 2001; 19: 1021-9. 15. Safra T. Muggia F, Jeffers S et al. Pegylated liposomal doxorubicin (doxil): Reduced clinical cardioloxicity in patients reaching or exceeding cumulative doses of 500 mg/m 2 . Ann Oncol 2000: II: 1029-33 16. Berry G, Billingham M. Alderman E et al. The use of cardiac biopsy to demonstrate reduced cardiotoxicity in AIDS Kaposi's sarcoma patients treated with pegylated liposomal doxorubicin. Ann Oncol 1998:9: 711-6. Received 23 February 2001; accepted 3 July 2001.

Correspondence to G. Fleming, MD University of Chicago Medical Center Section of Hematology/Oncology 5841 S. Maryland Ave. MC2115 Chicago, 1L 60637 USA E-mail. [email protected]