New applications of gemcitabine and future directions in the management of pancreatic cancer


  • James L. Abbruzzese M.D.

    Corresponding author
    1. Department of Gastrointestinal Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    • Department of Gastrointestinal Medical Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030-4009
    Search for more papers by this author
    • Fax: (713) 745-1163


Gemcitabine has been reported to be an active agent in pancreatic cancer. Recent applications have included the use of a fixed-dose rate regimen in patients with advanced pancreatic cancer based on the observation that deoxycytidine kinase is saturated at the plasma gemcitabine concentrations achieved with standard infusion, thereby limiting the accumulation of intracellular gemcitabine triphosphate. In a Phase II study, this regimen was associated with survival rates better than those typically observed in patients with advanced disease. Gemcitabine also has been assessed as a radiosensitizer in locally advanced cancer and although toxicity was significant, objective responses were observed and included tumor response, which permitted curative resection. Future directions in therapy for pancreatic cancer include the development of agents targeting signal transduction pathways and nuclear transcription factors based on the continually improving understanding of the role of molecular events in carcinogenesis. Cancer 2002;95:941–5. © 2002 American Cancer Society.

DOI 10.1002/cncr.10753

Carcinoma of the pancreas remains the fifth leading cause of cancer-related death in the U.S.1 New agents such as gemcitabine, a pyrimidine antimetabolite with good single-agent activity, have been used with success in patients with pancreatic cancer.2, 3 In recent studies, promising results have been achieved with this agent when given in a fixed-dose rate regimen in patients with advanced pancreatic cancer4 and when given concurrently with radiation therapy in patients with locally advanced pancreatic cancer.5 Further significant progress in the treatment or prevention of pancreatic cancer may result from ongoing improvement in the understanding of the molecular events specific to pancreatic carcinogenesis and the development of agents that target specific pathways and factors mediating these processes.

New Applications for Gemcitabine in Pancreatic Cancer

The activity of gemcitabine in a variety of solid tumors differentiates the agent from such other cytidine analogues as cytarabine. One potential explanation for the difference in antitumor effects observed with these two nucleoside analogs is that gemcitabine results in “masked” DNA chain termination,6 whereby the incorporation of gemcitabine triphosphate as a fraudulent base in the growing DNA chain is disguised by the incorporation of one additional normal nucleotide base; although chain termination occurs immediately with the addition of the cytarabine nucleotide, this substitution is more vulnerable to excision and repair by DNA polymerases than is the masked substitution of the gemcitabine nucleotide. In addition, gemcitabine exhibits self-potentiation, which results in the achievement of higher intracellular concentrations.7 Self-potentiating mechanisms include the inhibition of ribonucleotide reductase (responsible for the formation of deoxynucleotide triphosphates) by gemcitabine diphosphate and direct inhibition of deoxycytidine monophosphate deaminase by gemcitabine triphosphate (Fig. 1).

Figure 1.

Three mechanisms of gemcitabine self-potentiation. CDP: cytidine diphosphate; dCMP: deoxycytidine monophosphate; dFdUMP: difluorodeoxyuridine monophosphate; dCTP: deoxycytidine triphosphate.

Studies of gemcitabine kinetics have indicated that saturation of deoxycytidine kinase, the rate-limiting enzyme in the conversion of gemcitabine to its active triphosphate form, occurs at plasma gemcitabine concentrations of approximately 20–30 μM, effectively limiting the intracellular accumulation of gemcitabine triphosphate. It was hypothesized that the administration of gemcitabine at lower doses over a prolonged period would avoid enzyme saturation and permit greater intracellular accumulation.

To assess this hypothesis, we performed a Phase II trial of dose-intense gemcitabine comparing a regimen of 2200 mg/m2 via a standard 30-minute infusion with a fixed-dose rate regimen of 1500 mg/m2 over 150 minutes given weekly for 3 of 4 weeks in patients with metastatic pancreatic cancer.4 Pharmacokinetics studies in a subgroup of patients that assessed levels of gemcitabine triphosphate in circulating mononuclear cells as a surrogate for tumor cell levels demonstrated much higher intracellular levels in the group of patients receiving the fixed-dose rate compared with the group of patients receiving the standard infusion, with no apparent peak in intracellular accumulation during the fixed-dose rate dosing period (Fig. 2).

Figure 2.

Median concentrations of (A) gemcitabine (difluorodeoxycytidine [dFdC]) in plasma and (B) gemcitabine triphosphate (difluorodeoxycytidine triphosphate [dFdCTP]) in circulating mononuclear cells in patients receiving gemcitabine at a dose of 2200 mg/m2 via a 30-minute infusion (squares) or 1500 mg/m2/150 minutes at a dose of 10 mg/m2/minute (circles).

Preliminary findings in the first 67 patients enrolled in the current study indicated objective response rates of 16.6% in the fixed-dose rate group (including 2 complete responses) versus an objective response rate of 2.7% in the standard infusion group; the median survival rates were 6.1 months versus 4.7 months and the 1-year survival rates were 23% versus 8%. Updated survival data in 83 evaluable patients indicated median survival durations of 7.8 months in the fixed-dose rate group versus 5.0 months in the standard infusion group and 1-year survival rates of 24% versus 7%, and 2-year survival rates of 18% versus 2%. Among the 92 patients assessed for safety, WHO Grade 3 or Grade 4 leukopenia (17% vs. 11%), granulocytopenia (20% vs. 13%), and thrombocytopenia (16% vs. 3%) were more common in the fixed-dose rate group, with the frequency of anemia being similar (10% vs. 9%). This fixed-dose rate regimen also was associated with a greater frequency of improvement in performance status (25.7% vs. 9.1%) and an improvement in pain (27.3% vs. 15.6%); improved or stable weight was somewhat more common in the patients treated with a standard infusion (71.7% vs. 62.5%). Although these survival differences are not statistically significant, they do suggest a benefit for fixed-dose rate infusion. Future planned studies include the assessment of fixed-dose rate gemcitabine in synergistic combinations (e.g., with cisplatin) in patients with pancreatic cancer and in patients with other tumors reported to be responsive to gemcitabine.

Another recent study has assessed the utility of gemcitabine as a radiosensitizer in patients with locally advanced pancreatic cancer; prior studies of gemcitabine as a radiosensitizer in other types of tumors have demonstrated some promise.8, 9 In a Phase I study by Wolff et al.,5 18 patients with locally advanced adenocarcinoma of the pancreatic head were given 7 weekly doses of gemcitabine of 350 mg/m2 (n = 6 patients), 400 mg/m2 (n = 9 patients), or 500 mg/m2 (n = 3 patients), with 3000 centigrays of external beam radiation therapy being delivered during the first 2 weeks. Grade 3/4 hematologic toxicities occurred in > 50% of the patients and nonhematologic toxicities were significant, with 44% of patients, including all 3 patients who received the 500 mg/m2 dose, requiring hospital admission for nausea/emesis and dehydration. Of the 17 evaluable patients, 8 (47%) had evidence of tumor response, including 4 patients (24%) with a partial response. The median survival for all patients was 6 months; the 1-year survival rate was 66% for patients who achieved an objective response and 75% for those with a partial response. Two patients demonstrated sufficient improvement in the relation between the tumor and the superior mesenteric vasculature to warrant surgical exploration with intention to resect the tumor (Fig. 3).

Figure 3.

Computed tomography scans in a 53-year-old female patient with locally advanced disease demonstrating (left) a heterogeneous mass abutting the portal vein prior to treatment and (right) a significant reduction in the mass 12 weeks after treatment. The patient underwent pancreaticoduodenectomy with curative intent. Reproduced with permission from Wolff RA, Evans DB, Gravel DM, et al. Phase I trial of gemcitabine combined with radiation for the treatment of locally advanced pancreatic adenocarcinoma. Clin Cancer Res. 2001;7:2246–2253.

The use of gemcitabine with concurrent radiation therapy, which McGinn et al. review elsewhere in this issue,10 might improve the outcome of patients with pancreatic cancer, and several Phase I studies have attempted to define a tolerable regimen. McGinn et al.11 previously investigated once-weekly gemcitabine during a course of conventional radiation therapy (50.4 grays [Gy]) to maximize radiosensitization in patients with localized, unresectable pancreatic cancer. Blackstock et al.12 used a regimen of twice-weekly gemcitabine concurrent with 50.4 Gy of radiation therapy to treat patients with advanced pancreatic cancer. Based on these data, the Cancer and Leukemia Group B (CALGB) studied a twice-weekly dose (Monday and Thursday) of gemcitabine (40 mg/m2) concurrent with 50.4 Gy of radiation in its Phase II trial (CALGB 89805). Preliminary data indicated that the median overall survival for 38 patients with locoregional adenocarcinoma of the pancreas who were treated with this regimen was 7.9 months.13

Future Approaches to Pancreatic Cancer

In broad terms, the molecular events leading to pancreatic cancer, as with other malignancies, begin with the accumulation of DNA adducts (with the majority of cases most likely resulting from environmental exposure to such factors as cigarette smoke, organic chlorines, and other chemicals) and DNA mutation, leading to oncogene activation and tumor suppressor inactivation. A working hypothesis in pancreatic cancer is that signal transduction events mediate the biology of carcinogenesis, with mutation in K-ras (a commonly activated gene in pancreatic cancer) most likely playing a major role. Figure 4 shows some of the molecular abnormalities and affected signaling pathways in pancreatic cancer. In addition to K-ras activation and overexpression and epidermal growth factor receptor (EGFR) overexpression, including HER-2/neu overexpression, other events identified in pancreatic cancer include tumor suppressor mutations in p53, p16, and SMAD4 (DPC4) and abnormalities in nuclear transcription factors, including NFκB. Extensive activation of nuclear transcription factors leads to numerous abnormalities that are associated with alterations in vascular endothelial growth factor, cadherins, and matrix metalloproteinases, all of which interact to promote tumor growth.14–17 Currently, we are particularly interested in targeting nuclear transcription factors to impact beneficially downstream effector genes believed to be involved in resistance to chemotherapy and radiation therapy.

Figure 4.

Signal transduction events may mediate the biology of pancreatic cancer. EGFR: epidermal growth factor receptor; IGF-1: insulin-like growth factor-1; FGFR-1: fibroblast growth factor receptor-1; chemo/XRT: chemotherapy/radiation therapy.

On the basis of what already has been learned regarding the molecular biology of pancreatic carcinogenesis, a variety of agents are being developed for use in the treatment of pancreatic cancer that target signal transduction pathways or nuclear transcription factors. In addition to monoclonal antibodies or other inhibitors of EGFR or EGFR tyrosine kinase (e.g., trastuzumab and IMC-C225), antiangiogenic agents, and matrix metalloproteinase inhibitors, these include farnesyl transferase inhibitors that target the ras oncoprotein, Raf-1 inhibitors, NF-κB inhibitors, and Sp-1 inhibitors. Additional progress in understanding the nature and sequence of molecular events in the development of carcinoma of the pancreas ultimately will enable us to achieve early diagnosis and may permit the development of an array of therapies targeting specific pathways and events.