Pancreatic cancer remains a major public health issue and one of the greatest therapeutic challenges for the clinical oncologist. A recently published report estimated that 28,300 patients would be diagnosed in the United States in 2000 with pancreatic cancer and that 28,200 would die from the disease.1 Pancreatic cancer is the tenth most common malignancy in the U. S. but the fourth most common cause of death.1 Until recently, the mainstay of treatment was 5-fluorouracil. In 1998, the Food and Drug Administration approved gemcitabine for use in the palliative treatment of patients with pancreatic carcinoma. Gemcitabine is a nucleoside (deoxycitidine) analog with at least two mechanisms of action. It is phosphorylated intracellularly to both diphosphate and triphosphate derivatives by deoxycitidine kinase. The diphosphate form inhibits DNA synthesis dependent ribonucleotide reductase and the triphosphate form competes with deoxycitidine triphosphate as a fraudulent base in DNA.2 Gemcitabine is also self-potentiating, resulting in greater intracellular concentrations than other nucleoside analogs such as cytarabine.3 The introduction of gemcitabine has sparked a multitude of exploratory trials examining combination therapy with many other agents. The current review will describe studies evaluating the toxicity and efficacy of gemcitabine combinations with established and novel agents.
Since its approval in 1998 by the Food and Drug Administration, gemcitabine has rapidly become accepted as a standard part of palliative therapy for patients with advanced pancreatic cancer. It has modest activity when used as a single agent, and this has led to numerous studies combining it with other active agents, including cytotoxic as well as biologic and targeted therapies. Numerous Phase II studies suggest that the combinations may have greater activity than single agent therapy. However, randomized trials concerning this are still in progress. In the current article, a variety of gemcitabine-based novel combinations showing promising activity are reviewed. Cancer 2002;95:923–7. © 2002 American Cancer Society.
Gemcitabine and Docetaxel
In 1997 we began a Phase I/II, single institution clinical trial at Virginia Mason Medical Center of the combination gemcitabine and docetaxel in patients with unresectable pancreatic carcinoma. We have previously reported early results of this trial.4, 5 This study has been completed and an update of the data follows. The rationale behind combining these two agents was based on the results of a Phase III study comparing 5-fluorouracil to single agent gemcitabine. Burris et al.6 reported that gemcitabine had superior activity in patients with advanced pancreatic cancer. Two separate Phase II trials of single agent docetaxel in patients with advanced pancreatic cancer also showed activity. A study by Rougier et al.7 reported on 24 patients, of whom 17 were evaluable for response; 5 (29%) of these patients experienced a partial response lasting one to five months, and 5 patients had stable disease. A second report from Abbruzzese et al.8 at the University of Texas M.D. Anderson Cancer Center reported the results of treatment in 16 patients, of whom 10 were evaluable for response. Two achieved a partial response (20%). Based on these preliminary results, their nonoverlapping toxicities, and different mechanisms of action, we decided to combine docetaxel and gemcitabine in a Phase I/II trial for the first-line treatment of patients with advanced pancreatic carcinoma.
Eligible patients had to have biopsy-proven, unresectable pancreatic carcinoma with bidimensionally measurable disease. A total of 34 patients were enrolled over 24 months; 33 were evaluable for response. Eighteen were male and 16 were female. The median age was 60 years (range, 39–78 years). The median Karnofsky score was 90 (range, 70–100). The first 18 patients (Group A) received the doses and schedule described in Table 1, and the remaining 16 patients received the doses and schedule described in Table 2.
|Gemcitabine 800 mg/m2||X||X||X|
|Docetaxel 75 mg/m2||X|
|Gemcitabine 1000 mg/m2||X||X|
|Docetaxel 40 mg/m2||X||X|
The schedule was changed after an analysis of toxicity in the first 18 patients showed excessive hematologic toxicity, requiring a dose reduction in 13 patients. The toxicities are shown in Table 3. As can be seen, the change in the schedule and dose resulted in far less hematologic toxicity. In Group A, one patient had Grade 3 nausea and vomiting and another had Grade 3 fatigue. Five patients experienced Grade 2 and eight patients experienced Grade 3 hematologic toxicity. In Group B, two patients had Grade 2 hematologic toxicity. There were no episodes of neutropenic fever. Two patients developed pulmonary toxicity requiring discontinuation of treatment. In one of these patients respiratory failure proved fatal. Severe pulmonary toxicity has been reported in association with gemcitabine therapy.9–11 There have been no other significant toxicities in Group B. In general, the second schedule was well tolerated.
|Group A||Group B|
|No. of patients||18||16|
Overall, 10 patients (29%) achieved a partial response and 10 (29%) either a minor response or stable disease. The sites of response were pancreas, peritoneum, lungs, skin, and liver. The median time to progression was four months. The median survival was 10.5 months.
We concluded from this study that the chemotherapy schedule used in Group B was well tolerated and that this combination of gemcitabine and docetaxel had encouraging activity in patients with advanced pancreatic cancer. Several other studies have been reported using this combination in different doses and schedules. They are summarized in Table 4.
|Author/reference||Gemcitabine||Docetaxel||No. of patients||Responses||Survival|
|Lueck et al.19||800 mg/m2 weekly 3 weeks rest after Cycles 6 and 9||25–45 mg/m2 with gemcitabine||20 enrolled 12 evaluated||Not given||Not given|
|Lueck et al.20||800–1000 mg/m2 as above 1000 mg/m2 recommended||25–45 mg/m2 as above 35 mg/m2 recommended||25 enrolled 13 evaluated||23% PR||Not given|
|Kakolyris et al.21||1000 mg/m2 day 1 +8 q 21 days||100 mg/m2 day 8 q 21 days + GCSF||38 enrolled 27 evaluated||7.4%||Median 7 months 22% alive at 1 year|
|Ridwelski et al.22||1000 mg/m2 day 1,8,15 q 28 days||35 mg/m2 day 1,8,15 q 28 days||43 evaluated||3 CR 5 PR 19% overall||Median 9 months|
|Shepard et al.23||2000 mg/m2 every other week × 4||75 mg/m2 every other week × 4||16 evaluated||2 CR 1 PR 19% overall||Median 4.7 months|
|Petrovic24||800 mg/m2 day 1,8,15 q 28 days||75 mg/m2 day 1 q 28 days||14 enrolled||1 PR||Median 6.1 months|
|Clark et al.25||600 mg/m2 day 1,8,15 q 28 days||60 mg/m2 day 1 q 28 days||34 enrolled 24 evaluated||1 CR 1 PR 8% overall||Not given|
|Gonzalez-Cao et al.26||800 mg/m2 days 1,14||60 mg/m2 day 1||24 evaluated||38% overall||Median 6 months One year survival 2.5%|
Based on these promising early data with the gemcitabine-docetaxel combination, the Cancer and Leukemia Group B (CALGB) is conducting a Phase II randomized trial (CALGB 89504) in patients with advanced pancreatic cancer comparing four different gemcitabine based combinations, one of which is gemcitabine 1000 mg/m2 and docetaxel 40 mg/m2, given together intravenously on Days 1 and 8 of a 21 day cycle.
Gemcitabine and Trastuzumab
Following the recently reported success of trastuzumab (Herceptin) in patients with HER-2 overexpressing metastatic breast cancer, investigators have started to explore the efficacy of this agent in other solid tumors. Safran et al.12 recently reported the preliminary results of the combination of trastuzumab and gemcitabine in HER-2 overexpressing patients with advanced pancreatic carcinoma. The study required that there be at least 2 or 3+ overexpression by the HercepTest™ (DAKO Corp., Carpinteria, CA) immunocytochemistry assay. One hundred and fifty one patients were studied, and 21 (14%) met this criterion. Thirteen (62%) had 2+ overexpression, 4 (19%) had 2–3+ overexpression, and 4 (19%) showed 3+ overexpression. The schedule of chemotherapy is shown in Table 5.
|Gemcitabine||1000 mg/m2 weekly × seven weeks|
|Trastuzumab||4 mg/m2 week 1 then 2 mg/kg/week weekly × seven weeks|
Nineteen patients were evaluable for toxicity. The following Grade 3/4 toxicities were seen: three patients with neutropenia, one with thrombocytopenia, and one with transaminase elevation. One patient had a Grade 3 decrease in left ventricular ejection fraction. Eighteen patients were evaluable for response; four (22%) achieved a partial response measured radiographically and nine (50%) had a > 50% decrease in CA19-9. The authors concluded that the combination of Herceptin and gemcitabine in this population of patients had promising activity.
Gemcitabine and Cetuximab (IMC-C225)
IMC-C225 is a chimeric monoclonal antibody directed against the epidermal growth factor receptor (EGFR). The anti-tumor effect of IMC-C225 has been studied in vitro on human pancreatic carcinoma cells, both as a single agent and in combination with gemcitabine.13 Abbruzzese et al.14 have reported the preliminary results of a Phase II study combining gemcitabine and IMC-225 in patients with advanced pancreatic cancer. These patients were previously untreated and had bidimensionally measurable disease. The doses and schedule of treatment are described in Table 6.
|Gemcitabine||1000 mg/m2 weekly × seven weeks|
|IMC-C225||400 mg/m2 week 1|
|250 mg/m2 weekly × six weeks|
Eighty nine patients were screened for this study, with EGFR expression as a criterion for eligibility. Forty one patients were enrolled, and after two courses of therapy 5 patients (12%) achieved a partial response and 16 (39%) had either a minor response or stable disease. The median time to progression was 16 weeks. The most common toxicities were an acneiform rash in 38% of patients, folliculitis in 16%, fatigue in 41%, and Grades 1 and 2 fever in 38%. The authors concluded that this combination was well tolerated and showed promising activity in this population of patients with advanced pancreatic cancer.
Gemcitabine and Camptothecan Derivatives
Irinotecan is a topoisomerase inhibitor that has been evaluated in a Phase II trial in combination with gemcitabine in patients with advanced pancreatic carcinoma.15 The gemcitabine dose used in the study was 1000 mg/m2 and the irinotecan dose was 100 mg/m2; the two were given together on Days 1 and 8 of a 21 day cycle. The protocol design allowed for a dose escalation of irinotecan to 150 mg/m2 in later cycles, if there were no hematologic toxicities greater than Grade 2. Forty five patients were treated, 27 male and 18 female. The median age was 60 years (range, 31–89 years). The authors reported that nine patients (20%) had an objective response. Thirteen patients (32.5%) had a greater than 50% reduction in CA19-9 tumor marker. The median time to treatment failure was 2.9 months, and the median survival was 6 months. The authors considered the toxicity to be mild, with 4.2% of patients experiencing Grade 3/4 vomiting, 6.7% Grade 3/4 diarrhea, and 15% Grade 3/4 neutropenia. There were no episodes of neutropenic fever. These investigators are proceeding with a Phase III trial comparing this regimen to single agent gemcitabine.
Rubitecan is an oral topoisomerase I inhibitor which has shown activity in Phase II trials of patients with advanced pancreatic cancer. A randomized trial has recently been completed comparing rubitecan to gemcitabine in patients with untreated pancreatic cancer; however, the results have not yet been reported. Szarka et al.16 have recently published results from a Phase I study combining rubitecan with gemcitabine in patients with unresectable pancreatic carcinoma. Previous treatment with gemcitabine was an exclusion criterion. The rubitecan was administered as a single oral dose for five consecutive days each week. Gemcitabine was given intravenously over 30 minutes on Days 5, 12, and 19 of a 28 day cycle. The methodology used to determine the dose levels of rubitecan and gemcitabine for each patient was based on a Bayesian dose escalation schema called escalation with overdose control. Doses were escalated within single patient cohorts. Patients were followed for toxicity, pharmacokinetics, pain, performance status, CA19-9 levels, and radiographic disease response. Five different dose levels were reported, as shown in Table 7.
|Rubitecan/gemcitabine mg/m2||No. of patients|
The Grade 3 toxicities seen included anorexia, nausea, vomiting, fatigue, and liver function abnormalities. Grade 4 myelosuppression was also observed. One patient who received 1.5 mg/m2 of rubitecan together with 800 mg/m2 of gemcitabine achieved a partial response. One patient at the first dose level had stable disease for eight months. At the time of reporting, the trial remained open to accrual.
Gemcitabine and Flutamide
Mayer et al.17 have studied the combination of gemcitabine and flutamide. The choice of flutamide, which is an anti-androgen, was based on a report by Greenway et al.18 suggesting that treatment with flutamide improved the median survival of patients with advanced pancreatic cancer. These authors chose to combine flutamide with gemcitabine in a Phase II study. All patients had biopsy proven unresectable pancreatic cancer with measurable disease. The schedule is shown in Table 8.
|Gematabine 1000 mg/m2||X||X||X|
|Flutamide 250 mg three times daily||Continuous||→|
Twenty seven patients were entered into the study, 17 male and 10 female. The median age was 58 years (range, 36–73). The authors reported four (15%) partial responses, three by radiologic measurement and one by clinical measurement. Twelve patients (44%) had stable disease, and 10 patients (37%) progressed. The median survival had not yet been reached at 5.7 months of followup. Seven patients experienced Grade 3 toxicity (nausea/vomiting, neutropenia, anemia, erythema, and liver toxicity). The authors concluded that this combination was promising and had acceptable toxicity.
Despite modest activity as a single agent in the treatment of patients with advanced pancreatic cancer, gemcitabine has become a firmly established component of combination therapy in this disease. Gemcitabine and docetaxel have shown anti-tumor activity and favorable toxicity in a number of Phase II trials of patients with advanced pancreatic cancer. They have different mechanisms of action and therefore make an attractive combination for further study in randomized trials, such as that of the CALGB. Because of the favorable toxicity profile, the study of three-drug regimens would also be feasible. One such combination under study is gemcitabine, docetaxel, and irinotecan. As investigators gain further insights into the basic biology of pancreatic carcinoma, this will inevitably lead to the identification of molecular targets, which should also provide a rational basis for further study of combinations of gemcitabine with novel targeted therapies. A promising start has already been made with trastuzumab and IMC-C225.