| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL PAPER |
Division of Hematology and Oncology, St. Luke’s-Roosevelt Hospital Center, New York, New York, USA
Correspondence: Peter S. Kozuch, M.D., Division of Hematology and Oncology, St. Luke’s-Roosevelt Hospital Center, 1000 10th Avenue, Suite 11 G, New York, New York 10019, USA. Telephone: 212-523-6769; Fax: 212-523-2004; e-mail: Pkozuch{at}slrhc.org
 |
ABSTRACT |
|---|
 Top Abstract IntroductionSymptoms and Palliative...ChemotherapyConclusionReferences |
|---|
Key Words. Pancreatic cancer • Chemotherapy • Palliative care • Pain control
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionSymptoms and Palliative...ChemotherapyConclusionReferences |
|---|
Despite the introduction of gemcitabine and attempts at developing combination chemotherapy regimens, pancreatic cancer remains a chemoresistant tumor. All-stage 5-year survival is less than 5% [2]. Unfortunately, the goal of therapy for most patients is largely palliative, and optimization of supportive care is critical. The future of pancreatic cancer therapy may reside in the new fields of targeted and molecular therapies. This article reviews supportive care strategies, emerging palliative chemotherapy options, developing targeted therapies, and ongoing clinical research for the care of patients with metastatic pancreatic cancer.
|
SYMPTOMS AND PALLIATIVE TREATMENT |
|---|
|
TopAbstractIntroductionSymptoms and Palliative...ChemotherapyConclusionReferences |
|---|
Frequently present in the early stages of the disease, pain is reported by 75%-80% of patients at their initial evaluation [5, 8]. Fortunately, only 10% have severe pain [9]. Some patients complain of vague, intermittent epigastric pain. The etiology of this discomfort is less clear, but may be secondary to pancreatic duct obstruction and resultant pancreatic insufficiency (obstructive and inflammatory types) [3]. Optimal pain management consists of a combination of antitumor therapy, analgesic drug therapy, anesthetic blocks, and behavioral approaches [7, 10, 11]. Initially, pain associated with unresectable pancreatic cancer can be controlled with nonsteroidal anti-inflammatory drugs (NSAIDs) or oral or transdermal narcotic analgesics. However, upon disease progression, opioids alone may not be sufficient.
The second major modality for pain control is anesthetic block of the celiac plexus, using injection of a solution of 50%-100% alcohol or a phenol solution. This offers fast and effective pain control for a duration of 3 to 4 months in 80%-90% of patients [12]. Celiac block was described more than 80 years ago by Kappis and can be performed percutaneously (under radiologic guidance) or intraoperatively [5, 8, 10, 12–14].
A double-blinded, randomized trial comparing chemical neurolysis with placebo injection analyzed pain control and survival outcomes of 139 patients with pancreatic cancer undergoing surgical exploration. Neurolysis was performed by injecting 20 ml of 50% alcohol solution on each side of the aorta at the level of the celiac axis at the time of initial surgical exploration. An interesting observation came from the analysis of the subgroup of 34 patients with significant epigastric pain prior to initial surgical exploration: 20 patients were injected with alcohol and 14 were injected with placebo; median survival favored the experimental arm (6 months versus 2 months) [12]. With regard to pain control, all patients in that study required subsequent percutaneous celiac axis block, but the average time interval to celiac axis block again favored the experimental arm (11.8 ± 3.2 versus 4.0 ± 1.1 months, respectively). The effect of celiac plexus neurolysis is unfortunately not permanent, and severe pain can return after a variable period. Chemical splanchnicectomy or celiac axis block can then be repeated using numerous techniques (computerized tomography guided, magnetic resonance imaging guided, endoscopic ultrasound guided, or laparoscopically) [13, 15–18].
External beam radiation therapy with or without chemotherapy may be used to palliate pain associated with pancreatic cancer. Unfortunately, pain control may take several weeks to be achieved with radiation [1, 5]. A feasibility study of high-dose conformal radiotherapy for patients with locally advanced pancreatic carcinoma using 70-72 Gy over a 7-week period in 44 patients showed pain relief in 68% of patients during or after treatment. The median pain-progression-free interval was 6 months, and it was difficult to distinguish between pain secondary to primary tumor and metastatic progression [19]. Another study, in 104 patients treated with precision high-dose radiation therapy (50 Gy with a cone down to the tumor of 65 Gy), demonstrated that improvement of pain symptoms can be achieved within 8 to 12 hours, but did not report the duration of pain relief [20].
Chemotherapy also can achieve pain control in pancreatic cancer patients, as shown with the use of gemcitabine. Almost 24% of patients treated with gemcitabine experienced a clinical benefit with regard to improved pain and/or fatigue [21–23].
For obstructive pain, pancreatic duct stenting or pancreatic enzyme replacement may be effective. In a study reporting successful and uncomplicated pancreatic duct stenting in 10 patients, 70% had significant improvement in their pain symptoms, and 50% no longer required analgesia [11].
Obstructive Jaundice
Seventy percent of pancreatic cancers are located in the head of the pancreas, and obstructive jaundice is often the first symptom leading to diagnosis [3]. Obstructive jaundice manifests clinically by moderate to severe pruritus, jaundiced sclerae and skin, discolored urine, clay-colored stool, cholangitis, and ultimately, hepatic failure [5]. Since biliary obstruction relief reduces symptoms and improves quality of life, it is imperative to attempt a durable means of decompression.
Biliary decompression can be accomplished by surgical bypass (including cholecystojejunostomy, choledochojejunostomy, or hepaticojejunostomy) or endobiliary stenting. Each procedure has advantages and disadvantages. The choice of surgery versus stenting should be made in the context of the patient’s overall treatment plan. A surgical bypass procedure provides durable relief from obstructive jaundice. However, because of operative morbidity, surgical biliary decompression should be reserved for patients undergoing laparotomy for potential curative resection and for patients in whom stenting cannot be accomplished [24–27].
Because of the lower cost, hospitalization rate, and periprocedure morbidity, as well as the shorter length of hospital stay compared with surgical bypass, most patients are best palliated with stent placement [28–30]. Endoscopic stent placement has a higher success rate and lower 30-day mortality rate than percutaneous stenting and represents the procedure of choice [31]. The decision to use an expandable metal or Teflon stent versus a plastic stent is similarly practical. While plastic stents are fraught with relatively more complications, such as migration and occlusion, they have the distinct advantage of being removable. This advantage is of particular relevance for patients undergoing neoadjuvant chemoradiation. Subsequent pancreaticoduodenectomy is technically easier when the extent of bile duct resection does not have to accommodate an embedded metallic stent. For this reason, plastic stents should be used for decompression in patients with potentially resectable pancreatic cancer [29]. Otherwise, because of more durable patency, expandable metal stents should be offered in patients with metastatic or unresectable disease.
Data do not support prophylactic bypass procedures in patients who do not otherwise require surgery. Bypass procedures can be reserved for patients in whom stenting fails [32–36]. A single-institution study of 155 laparoscopically staged patients, 40 with locally advanced disease and 115 with metastatic disease, assessed the need for a surgical bypass procedure. Ninety-eight percent of those patients did not require an open surgical procedure to treat biliary or gastric obstruction [35].
Weight Loss or Cachexia
Cachexia in the advanced pancreatic cancer patient results not only from loss of appetite and malnutrition but also from hypercatabolism of lean tissue and the anorexia/cachexia syndrome of patients with advanced cancer [37]. Cachexia often is associated with weakness, fatigue, and poor quality of life [38]. Weight loss also contributes to depression and is predictive of poor outcome and greater morbidity [38–40]. Management of pancreatic exocrine insufficiency is especially important and is detailed in the next section.
Cachexia is thought to be mediated by the release of cytokines and other factors secreted by the tumor [39, 41]. The incriminating factors include tumor necrosis factor-alpha (TNF-
), interleukin-1ß, interleukin-6, ciliary neurotropic factor, and the proteolysis-inducing factor, initially isolated in murine models of cancers and recently isolated in the urine of approximately 80% of patients with pancreatic cancer and severe cachexia [39, 42].
Initial management is usually supportive nutrition: caloric supplementation and hydration, preferably orally [39, 40]. Enteral or parenteral nutrition is not justified in most cases of advanced cancer anorexia/cachexia [38]. Pharmaceutical therapeutic intervention may be helpful in many cases. Anorexia is well treated with appetite stimulants, such as progestational agents and corticosteroids [37–40, 43]. These agents are potent antiemetics, act rapidly, and have been proven to increase nonfluid body weight in advanced cancer patients with anorexia. The most studied and used agents are megesterol acetate, medroxyprogesterone acetate, and dexamethasone.
Dexamethasone can be administered at doses ranging from 3-8 mg/day and megesterol acetate doses range from 400-800 mg/day. They are comparable in efficacy and produce an improvement in appetite with subsequent weight gain in approximately 15% of patients. Dexamethasone has a short-lived duration of action (4 weeks) and a significantly worse side-effect profile, including myopathy, mood changes, and hyperglycemia [39, 44].
Other classes of drugs used are orexigenic agents such as the prokinetic medication metoclopramide and the cannabinoid dronabinol [37]. Dronabinol has been studied at doses of 2.5-7.5 mg/day and has been shown to improve chemotherapy-induced nausea and vomiting in 65% of cancer patients. Dronabinol has also been shown to improve HIV-associated cachexia in up to 69% of patients.
Additional agents that may reverse cancer-associated cachexia include eicosapentaenoic acid (EPA), thalidomide, adenosine triphosphate, and ibuprofen. EPA (found in fish oil pills) at doses ranging from 1-6 g/day, results in weight stabilization after 3 to 4 weeks and 1-2 kg weight gain after 7 weeks. Animal studies suggest EPA works by suppressing the effect of proteolysis-inducing factor [45–48]. Thalidomide, 100-200 mg/day at bedtime, provides some weight stabilization but no weight gain. It appears to shorten the half-life of the mRNA of the TNF- [37].
Ibuprofen, an NSAID, has a role in abrogating the catabolic process and may lead to modest increases in weight [49]. The combination of ibuprofen with megestrol acetate has been shown to reverse weight loss and improve quality of life [50, 51].
Pancreatic Insufficiency
Pancreatic exocrine insufficiency is common but usually moderate in patients with pancreatic cancer. Approximately 65% of patients will have some degree of fat malabsorption and 50% will have some degree of protein malabsorption. This may contribute to weight loss, epigastric discomfort/ pain, and malabsorption symptoms such as excessive flatus, bloating, diarrhea, and steatorrhea [52–54]. Pancreatic enzyme-replacement therapy, therefore, has a uniquely important role in the supportive care of these patients. Pancreatic enzyme-replacement has been shown to improve malabsorption, bloating, and diarrhea and helps prevent further weight loss in pancreatic cancer patients [53–56]. Pancreatic enzyme therapy may sometimes be associated with gastrointestinal side effects such as nausea, vomiting, cramps, constipation, or diarrhea that can be ameliorated by adjusting doses or brands. Pancreatic duct stenting may also palliate obstructive symptoms and improve nutritional status [57]. Empiric pancreolipase replacement should be considered for most patients. A placebo-controlled trial randomized 21 patients with unresectable pancreatic cancer of the pancreatic head and suspected pancreatic duct obstruction to 8 weeks of oral high-dose enteric-coated pancreatic enzyme versus placebo, prior to stenting. There was a 1.2% increase in body weight in the group of patients assigned to receive the pancreatic enzymes versus 3.7% decrease in body weight for the placebo group [55].
Gastric Outlet Obstruction (GOO)
GOO is a late complication of advanced pancreatic cancer affecting approximately 10%-20% of patients who survive more than 15 months [1, 5, 25, 26, 30, 35, 57]. However, fewer than 3% of patients developing GOO require surgical bypass [35, 58, 59]. Both prophylactic and palliative strategies are described in the literature for the management of this problem. Data to support each approach are available [1, 5, 25, 26, 57]. Laparoscopic and endoscopic management of GOO have been described.
Importantly, 60% of patients with advanced pancreatic cancer have no evidence of gastric or duodenal invasion but nevertheless have abnormal gastric motility with delayed emptying. Also, nausea and vomiting may develop secondary to tumor infiltration of the nerve plexi of the stomach and duodenum [1]. After gastric outlet and small bowel obstructions have been ruled out, empiric prokinetic drugs, such as metoclopramide, can be used successfully to manage these symptoms [5].
Depression and Fatigue
Depression has been demonstrated in 47%-71% of patients with pancreatic cancer [9, 60]. A significant correlation has been demonstrated among increasing pain, depressive symptoms, and impaired quality of life [9]. Fatigue, either caused by the illness itself, side effects of therapy, or depression, is another symptom affecting quality of life [61].
Treatment usually includes tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), or a combination of both [7]. SSRIs have fewer side effects than tricyclics, in particular, the anticholinergic symptoms (dry mouth) and somnolence, but lack the analgesic efficacy of the tricyclics. Psychostimulants such as methylphenidate can be used to improve fatigue [7].
Anemia can also contribute to fatigue and may be present prior to treatment or result from treatment. Anemia should be corrected once diagnosed. Recombinant human erythropoietin can be used to maintain an adequate hemoglobin level.
|
CHEMOTHERAPY |
|---|
|
TopAbstractIntroductionSymptoms and Palliative...ChemotherapyConclusionReferences |
|---|
summarize many randomized trials and multiagent clinical trials. Well-tolerated combinations may emerge as a more effective standard with which targeted therapies may be incorporated.
|
|
|
Consistent with the emphasis on maintaining quality of life and palliating symptoms in this patient population, a composite end point, termed clinical benefit response (CBR), was developed to quantify the impact of gemcitabine on symptoms typically associated with progressive pancreatic cancer. If patients had improvement in any one of three parameters (improvement in either pain intensity, analgesic consumption, or Karnofsky performance status) for at least 4 weeks with none of the other parameters worsening within 12 weeks, they were considered a clinical benefit responder [63, 64]. A weight gain of 7% above pretreatment baseline was a secondary end point and would serve to classify patients as clinical benefit responders if the primary end points all remained stable.
In the successful registration trial, 126 patients with previously untreated advanced symptomatic pancreatic cancer were randomized to receive either gemcitabine 1,000 mg/m2 weekly (63 patients) or 5-FU 600 mg/m2 once weekly (63 patients) [22]. CBR was experienced by 23.8% of gemcitabine-treated patients compared with 4.8% of 5-FU-treated patients. Median survival duration was 5.65 months versus 4.41 months favoring gemcitabine treatment (p = 0.0025). The survival rate at 12 months was 18% for gemcitabine patients and 2% for 5-FU patients. Notably, patients who attained a CBR with either treatment had a longer median survival than patients who did not attain a CBR (10.7 months versus 4.8 months).
In an effort to improve the activity of gemcitabine, an adjustment to the infusion rate also has been evaluated in a phase II study [65]. Gemcitabine at 2,200 mg/m2 administered over 30 minutes was compared with a 10-mg/m2/min infusion, weekly for 3 weeks, every 4 weeks (gemcitabine 1,500 mg was administered over 150 minutes). This fixed-dose-rate infusion schedule was designed to avoid exceeding the kinetics of activation of deoxycytidine kinase, the rate-limiting enzyme leading to gemcitabine triphosphate, the active metabolite. An objective response of 16.6% versus 2.7%, longer median survival of 6.1 months versus 4.7 months, and 1-year survival of 23% versus 0% favored the fixed-rate infusion schedule. The fixed-rate infusion schedule also was associated with significantly higher median gemcitabine triphosphate levels in peripheral circulating mononuclear cells.
Gemcitabine is a well-tolerated drug, ideal for palliation of symptomatic cancer. At the commonly used weekly schedule with doses ranging from 800-1,500 mg/m2, myelosuppression is the main side effect. Other major adverse effects are uncommon and usually easy to manage. They include fever (7.3%), pain (6.8%), asthenia (6.0%), abdominal pain (5.5%), dyspnea (5.0%), vomiting (3.9%), anorexia (3.6%), and deep venous thrombosis (3.2%) [66]. The hemolytic uremic syndrome and an acute respiratory distress syndrome are two very rare but potentially fatal side effects [67, 68].
5-Fluorouracil
Until the introduction of gemcitabine, 5-FU was the cornerstone of chemotherapy for pancreatic cancer even though there was no clear evidence of its benefit in patients with advanced disease. The optimal dose and schedule of 5-FU as a single agent in pancreatic cancer has not been defined. Recent data from studies comparing 5-FU with gemcitabine or gemcitabine with or without 5-FU suggest that bolus 5-FU administration may be of little or no benefit [22, 69]. 5-FU together with radiation, however, was claimed to improve results in locally advanced pancreatic cancer in postoperative treatment.
Results from trials of bolus 5-FU have been associated with disappointing median survival outcomes of less than 5 months. Biomodulation of 5-FU with leucovorin, N-(phosphonacetyl)-L-aspartic acid (PALA), or -2b interferon has not improved upon these outcomes [70–74]. However, reviewing the survival outcomes associated with varying 5-FU doses and schedules, more prolonged infusion schedules are associated with median overall survival times of 6-8 months. For example, an 8.8 month median survival has been attained with 5-FU as a continuous infusion (2,600 mg/m2 over 24 hours, once weekly x 6 weeks, every 8 weeks) [75]. Capecitabine is an oral fluoropyrimidine designed to generate 5-FU as a metabolite predominantly within tumor cells at levels of 5-FU approximating those associated with continuous infusion 5-FU. A phase II study evaluated 42 patients with pancreatic cancer treated with 1,250 mg/m2 of capecitabine twice daily for 2 weeks followed by 1 week off. Results showed a 7.3% objective response rate, 24% CBR, and a median survival of 6 months [76].
Topoisomerase I Inhibitors
Several topoisomerase I inhibitors, including irinotecan, exatecan, and the oral drug rubitecan, are undergoing clinical development as part of novel gemcitabine-based combinations.
Irinotecan has a broad-spectrum activity against colorectal, esophageal, gastric, pancreatic, and lung cancer. As a single agent, irinotecan 350 mg/m2 every 3 weeks demonstrated a response rate of 9% and a median survival of 5.2 months in patients with metastatic pancreatic cancer [77]. In another phase II study, irinotecan was given at 100 mg/m2 weekly or 150 mg/m2 every 2 weeks, with an 11% response rate and acceptable toxicity profile [78].
Exatecan, another intravenous topoisomerase I inhibitor, also has been evaluated in a phase II study that included 39 patients, 20 refractory to gemcitabine and 19 previously untreated. The overall response rate was 8%, but 39% had stable disease. The median survival was 5.5 months for all 39 patients and 10.3 months for the 19 chemotherapy-naïve patients. The 1-year survival rate was 35% [79]. This encouraging activity has prompted a phase III evaluation of exatecan plus gemcitabine versus gemcitabine alone.
9-nitrocamptothecin (rubitecan, 9NC) is an oral camptothecin analogue with powerful topoisomerase I inhibition. In a phase II trial of both untreated and gemcitabine-treated patients, median survival times of 7.3 months and 4.7 months, respectively, were observed [80]. Rubitecan is currently being tested in three phase III trials: rubitecan versus gemcitabine, rubitecan versus 5-FU (in gemcitabine-refractory patients), and rubitecan versus the investigator’s best chemotherapy (in gemcitabine-refractory patients).
Taxanes
Docetaxel has demonstrated variable activity in pancreatic cancer with response rates ranging from 0%-28% and reported median survivals of >6 months [81–84]. Docetaxel has been combined with gemcitabine in multiple studies. Gemcitabine was given weekly at the standard dose of 1,000 mg/m2, and docetaxel doses ranged from 75-100 mg/m2 every 3 weeks. Response rates ranged from 7.4%-33%, and the best median survival was 7 months [85, 86]. Paclitaxel also has been evaluated as a single agent in metastatic pancreatic cancer in a phase II trial; the response rate was 8% and median survival was 5 months [87].
Epirubicin
Epirubicin, an anthracycline, was evaluated in two single-agent phase II studies in metastatic and locally advanced pancreatic cancer. The first study evaluated epirubicin at 90 mg/m2 every 4 weeks and showed an overall response rate of 24%, a response duration of 7 months, and a time to progression of 3 months. The median survival was 9 months for responders and 5 months for all patients [88].
In the second, more recent study, investigators used epirubicin at doses ranging from 90-135 mg/m2 every 3 weeks, with G-CSF support and dexverapamil 1,000-1,200 mg/day on days 1 to 3 to counter multidrug resistance. Response rates were modestly better (32%) and responses were attained at doses ranging from 105-120 mg/m2 [89]. Survival outcomes were not available at the time of this report. Nevertheless, there is little rationale for the use of dose-escalated therapy that requires cytokine support in the palliative setting.
Cisplatin and Oxaliplatin
Cisplatin and oxaliplatin have shown minimal single-agent activity in metastatic pancreatic carcinoma [90, 91], but seem to have an additive effect in several phase II trials with infusional 5-FU (Table 3
). Response rates and median survival times for platinum/5-FU combinations ranged from 11%-26.5% and from 4 to 11.5 months, respectively. In these small trials, prolonged (at least 24 hours) infusion 5-FU was associated with better response and survival outcomes than bolus 5-FU, again suggesting that 5-FU activity in pancreatic cancer is schedule dependent. Grade 3/4 toxicities included neutropenia (12%-23%), nausea and vomiting (12%-17%), and mucositis (up to 14%) [92–97]. Combinations of platinum compounds with gemcitabine also have demonstrated encouraging outcomes with response rates ranging from 11%-31% and median survival times of 6-8.5 months (Table 2). Grade 3/4 toxicities included neutropenia (11%-26%) and nausea (10%-50%) [98–102].
Irofulven
Irofulven (6-hydroxymethylacylfulvene [HMAF], MGI 114) is a novel alkylating agent that has demonstrated impressive antitumor activity in a broad range of human tumors in vitro and in vivo, including pancreatic adenocarcinoma [103–105]. Irofulven binds to multiple intracellular targets, inducing S-phase blockade and cell death via caspase-induced apoptosis [103–104, 106].
A phase II trial of irofulven as a single agent in patients with gemcitabine-refractory pancreatic cancer showed modest activity. Given at 11 mg/m2 daily for 5 days every 28 days to 42 patients, irovulfen attained one partial and one complete response [105, 107]. Survival outcomes were not defined at a median follow-up of 6 months. Unfortunately, this phase II activity could not be confirmed in a phase III trial comparing irofulven with 5-FU in patients with gemcitabine-refractory metastatic pancreatic cancer. That phase III trial was stopped in April 2002 when a planned preliminary analysis indicated that the 5-FU comparator arm demonstrated better-than-expected survival benefit. Ongoing development of irofulven as a single agent in combination with other chemotherapies continues in other solid tumor types.
Ukrain (NSC-631570)
Ukrain is a semisynthetic compound of thiophosphoric acid and the alkaloid chelidonine derived from the plant Chelidonium majus, a common weed in Europe and western Asia. For many centuries, the plant has been used in the therapy of warts and skin cancers. Ukrain has been used in alternative herbal medicine for more than 20 years without knowledge of its mechanism of action [108]. It showed activity against a broad spectrum of tumors and is included in multiple trials in Europe, including pancreatic cancer trials [109, 110]. Ukrain inhibits pancreatic cell cycle progression in the M phase by stabilizing monomeric tubulin [111].
A single-institution phase II trial randomizing 42 patients to receive 10 mg ukrain every other day for 20 days or placebo demonstrated a longer median survival of 17.17 months in the ukrain group, compared with 6.97 months in the placebo arm [110].
5-FU-Based Combinations
Phase II trials of FAM (5-FU with doxorubicin and mitomycin C), FAM-S (with streptozocin), and sequential 5-FU, cyclophosphamide, methotrexate, vincristine, and mitomycin C (the Mallinson regimen) demonstrated response rates ranging from 20%-40% and median survival times of up to 10 months [112, 113]. However, the encouraging activity of these combinations did not translate into survival advantages upon subsequent testing against single-agent 5-FU. A randomized phase III trial comparing 5-FU alone (500 mg/m2/day x 5 consecutive days every 5 weeks) with the Mallinson regimen demonstrated equivalent survival outcomes, 3.5 months versus 4.5 months, respectively, p > 0.48. FAP (5-FU, doxorubicin, and cisplatin) was the third arm of that trial and also demonstrated no survival advantage over 5-FU alone [113].
Similarly, FAM and FAM-S compared with 5-FU alone in a randomized trial of the North Central Cancer Treatment Group showed no difference in terms of survival [114–117]. Therefore, the above 5-FU combinations have not shown any advantage over 5-FU (or best supportive care) and are not recommended as standard practice.
A phase II study of 5-FU, leucovorin, methotrexate, and cisplatin (M-FLP) by the Italian Oncology Group for Clinical Research, showed a 12% response rate and 20-week median survival in 35 patients [118]. Of interest, 5-FU/leucovorin combined with etoposide demonstrated a quality-of-life improvement comparable with gemcitabine. In that trial, 90 patients with pancreatic or biliary cancer were randomized to either chemotherapy in addition to best supportive care or best supportive care. Thirty-six percent of patients in the chemotherapy group attained an improved or prolonged high quality of life for a minimum period of 4 months compared with 10% of those in the best supportive care group. Median survivals were 6.0 months versus 2.5 months, respectively [119].
5-FU also has been combined with epirubicin in two phase II trials. Response rates were 21% and 19%, and median survival times were 6 months and 5.2 months, respectively [120, 121].
The combination of 5-FU with oxaliplatin has demonstrated a response rate of 11% and a median survival of 8.5 months. Toxicities included grade 3 nausea/vomiting (17%), grade 2/3 mucositis (17%), grade 2/3 diarrhea (13%), and grade 2/3 sensory neuropathy (17%) [91].
Gemcitabine-Based Combinations
Over 50 two-, three-, and four-drug gemcitabine-based combinations have been reported. Most of these are single-institution phase II trials. Those regimens with relatively encouraging activity are summarized in Tables 2 and 3.
Gemcitabine + 5-FU
Several phase II studies combining gemcitabine with 5-FU have demonstrated the tolerability of the doublet, despite some greater toxicity in comparison with single-agent gemcitabine [122–124]. Those studies were followed by the Eastern Cooperative Oncology Group (ECOG) randomized phase III trial E2297, which concluded that the combination of gemcitabine with bolus 5-FU had no advantage over single-agent gemcitabine [69]. In that trial, 327 patients were randomized to receive gemcitabine 1,000 mg/m2 either with or without 5-FU 600 mg/m2 as a brief i.v. infusion, both given weekly, 3 of 4 weeks. Grade 3/4 leukopenia (29% versus 16%) and diarrhea (10% versus 4%) were observed more frequently in the combination arm. The median survival times were 5.4 months and 6.7 months for the gemcitabine and gemcitabine/5-FU arms, respectively (p = 0.11)
While this trial maintains gemcitabine as the current standard of care, it does not signal the end of 5-FU in the development of gemcitabine-based combination therapies. The activity of 5-FU against pancreatic cancer may be schedule dependent. Phase II studies using high-dose infusion schedules of 5-FU in combination with standard infusion gemcitabine have reported good tolerability and suggest improved clinical benefit, overall response rates, and median survivals compared with therapy with gemcitabine alone [125–127].
Gemcitabine + Platinum Analogues
Outcomes associated with phase II single-arm trials of cisplatin/gemcitabine combinations have demonstrated modest additional benefits compared with outcomes for gemcitabine as a single agent [99, 100]. Heinemann et al. administered gemcitabine, 1,000 mg/m2 weekly for 4 weeks, in combination with cisplatin, 50 mg/m2 on days 1 and 15, to 41 patients [99]. Therapy was tolerated without major compromises in quality of life. An overall response rate of 11.5%, median survival of 8.3 months, and a 1-year survival rate of 28% were demonstrated in the 35 evaluable patients.
A phase II trial of a biweekly gemcitabine 1,000 mg/m2 (day 1) and oxaliplatin 100 mg/m2 (day 2) doublet has demonstrated encouraging activity [101]. A 31% response rate, 18-week progression-free survival, and 62% 6-month survival rate were seen in 32 evaluable patients.
The best response outcomes for the gemcitabine/cisplatin doublet are those of a phase III trial randomizing 107 patients with unresectable or metastatic pancreatic cancer to standard infusion gemcitabine with or without weekly cisplatin 25 mg/m2. Response rates (26.4% versus 9.2%) and a longer median survival (30 weeks versus 20 weeks) favored the gemcitabine/cisplatin doublet. However, clinical benefit was similar (52.6% versus 42.6%). Notably, the clinical benefit response in both arms was higher than the 23.8% from the gemcitabine registration trial [102].
Gemcitabine + Topoisomerase I Inhibitors
Topoisomerase I inhibitors in combination with gemcitabine have demonstrated modest additional benefit over gemcitabine alone based on phase II trials. Irinotecan 100 mg/m2 combined with gemcitabine 1,000 mg/m2 over 90 minutes on days 1 and 8 repeated every 21 days, has been shown to be well tolerated and active. In a multicenter phase II trial of that regimen, the response rate was 20%, the median time to progression was 2.8 months, the median survival was 5.7 months, and the 1-year survival rate was 27%. Toxicity was moderate and limited to grade 4 neutropenia (2%), grade 4 vomiting (2%), and grade 3 diarrhea (7%) [128]. Another phase II trial evaluated the same combination with a different dose and schedule: gemcitabine, 900 mg/m2 on days 1 and 8, and irinotecan, 300 mg/m2 on day 8, with cycles repeated every 21 days. The response rate was lower (15%), and toxicity was more severe than the previously mentioned study. Median survival was not reported [129].
Gemcitabine + Docetaxel
Multiple trials have evaluated this combination, and most used weekly doses of gemcitabine ranging from 600-1,000 mg/m2 and docetaxel ranging from 25-45 mg/m2 [130–133]. Toxicity has been moderate and acceptable. The most complete data come from a phase II trial with chemotherapy-naïve patients. The study was conducted by Kakolyris et al. Gemcitabine 1,000 mg/m2 was administered on days 1 and 8 every 21 days, and docetaxel 100 mg/m2 was given on day 8 every 21 days. G-CSF support was given from days 9 to 15. The response rate was only 7.4%, but a CBR of 43.7% was attained. The time to progression was 4 months, and median survival was 7 months, with 22% of patients alive at 1 year [133].
The Cancer and Leukemia Group B (CALGB) has opened a four-arm randomized phase II trial (CALGB 89904) to better understand the activity of these various doublets compared with fixed-rate infusion 10 mg/m2/min gemcitabine monotherapy (Fig. 1).
|
Gemcitabine + Ukrain
Ukrain, gemcitabine, and a combination of both drugs were compared in a phase II randomized trial. Gemcitabine administration was per standard dose/schedule of 1,000 mg/m2 weekly for 3 weeks and was associated with an expected median survival of 5.2 months. Ukrain was administered similarly in both the single-agent and combination arms: 20 mg/m2 daily for 5 days. The three arms were well balanced and had 30 patients each. Survival outcomes associated with ukrain were promising, with median survival times of 7.9 and 10.4 months for ukrain alone and the combination arm, respectively. Toxicities were moderate and acceptable [111].
Three- and Four-Drug Gemcitabine Combination Regimens
Three- and four-drug gemcitabine-based combinations, in single-institution phase II trials, have reported modest improvements in response and survival outcomes compared with reports of earlier single-agent or combination regimens (Table 3).
A phase II study of gemcitabine, 5-FU/leucovorin, and oxaliplatin (FOLFU-GEMOX) showed a response rate of 29%, (11% complete remissions), a CBR of 39%, and a median overall survival of 8 months [134]. Gemcitabine also has been combined with epirubicin, cisplatin, and 5-FU (PEF-G) in a phase II study. Fourty-three patients with metastatic pancreatic carcinoma were treated; the objective response rate was 58%, CBR rate was 78%, and median and 1-year survivals were 11 months and 39%, respectively [135].
A combination of gemcitabine, 5-FU, and cisplatin (GFP) was retrospectively reviewed in a pretreated patient population with metastatic pancreatic cancer. The median survival was 10.6 months [136]. A gemcitabine-based four-drug combination, incorporating irinotecan in the previously described GFP, designated G-FLIP, was administered over 2 days in a biweekly fashion and evaluated in 34 patients at our institution. The overall response rate was 24%, and the median survival was 10.3 months [137]. Most of these patients had progressed on the GFP regimen, and yet, a significant response rate (24%) and stabilization rate (24%) were obtained after adding irinotecan. We concluded that adding a single new agent to a first-line chemotherapy combination upon disease progression may be an important alternative to switching to a different drug class. The same regimen (G-FLIP) was evaluated as first-line treatment in metastatic pancreatic cancer and showed a response rate of 33% and a median time to progression of 40 weeks in a small retrospective study [138]. While these three- and four-drug regimens appear to be well tolerated and active, confirmatory phase II and phase III testing is needed to define their role in the treatment of patients with metastatic pancreatic cancer.
Targeted Therapy
The recent identification of dominant mechanisms of uncontrolled cell division, metastasis, invasion, and tumor/host microenvironment has translated into the introduction of targeted antitumor therapies. Among these therapeutic agents, those targeting signal transduction via the endothelial growth factor receptor family (EGF-R and ErbB2) and those targeting downstream signaling pathway elements, such as the K-ras oncoprotein and farnesyl transferase, are of potential interest in pancreatic cancer.
Signal Transduction Inhibitors
Several potential points of intervention have been identified along signal transduction cascades, and several classes of targeted therapies are in clinical development. Beginning with the extracellular domain of the HER family of receptor tyrosine kinases, agents targeting HER-2 and EGF-R are being clinically developed in pancreatic cancer. Two monoclonal antibodies have been explored, trastuzumab, directed against the HER-2 receptor, and cetuximab (IM-C225), targeting the EGF-receptor [139].
Approximately 20% of 151 evaluated pancreatic tumors overexpressed the HER-2/neu receptor. A phase II clinical trial of trastuzumab in combination with gemcitabine was conducted in patients with metastatic pancreatic cancer; gemcitabine was given at 1,000 mg/m2 weekly, and trastuzumab was given at a loading dose of 4 mg/kg for the first week and then 2 mg/kg weekly to 31 patients found to have tumors overexpressing HER-2/neu. The overall response rate was 22%, median overall survival was 7.5 months, and 24% of patients were alive at 1 year. Grade 3 hematologic toxicities occurred in 22% of treated patients. There was one case of cardiotoxicity with decline in the left ventricular ejection fraction [140].
IM-C225 also has been evaluated in combination with gemcitabine. Forty-one chemotherapy-naïve patients with metastatic pancreatic cancer were evaluated and treated in a phase II trial. EGF-R was overexpressed in 89% of tumors. Gemcitabine was administered at 1,000 mg/m2 weekly for 7 weeks then 1 week rest followed by weekly infusion for 3 weeks and then 1 week off, and IM-C225 was given at a loading dose of 400 mg/m2 then 250 mg/m2 weekly. Partial responses occurred in 12.5% of patients, and 52% of patients had disease stabilization. The median overall survival was 202 days, and 32% of patients were alive at 1 year. The main adverse effects reported were dermatological: acneiform rash in 38% and folliculitis in 16% [141].
ZD1839 and OSI-774, both orally available synthetic quinazolines that inhibit the EGF-receptor tyrosine kinase, are undergoing phase III trials. OSI-774 combined with gemcitabine is being compared with gemcitabine alone as initial treatment for patients with metastatic pancreatic cancer in an ongoing international phase III study [142–144].
Targeting Downstream Signaling Pathways: Ras Inhibition
Blocking Ras signaling seems an ideal target for therapy since the ras mutation is expressed in approximately 90% of pancreatic tumors [145]. Ras inhibition may be achieved by blocking the posttranslational farnesylation of K-ras [143]. Several farnesyl transferase inhibitors are under study: SCH66336, R115777, and BMS 214662 [146, 147].
SCH66336, R1115777, and BMS 214662 are available in oral and i.v. forms. All have completed phase I testing and are presently being studied in phase II trials in chemotherapy-refractory solid tumors [148–151]. SCH66336 was evaluated in a randomized phase II study comparing it with gemcitabine in patients with metastatic pancreatic cancer, either pretreated or chemotherapy naïve. Sixty-three patients were randomized (SCH66336, n = 33, gemcitabine, n = 30) to receive treatment with SCH66336 daily at a dose of 200 mg orally twice a day or gemcitabine 1,000 mg/m2 weekly for 7 weeks followed by 1 week of rest. Subsequent cycles of gemcitabine consisted of three weekly treatments and then 1 week of rest. A 23% response rate was attained in the SCH66336 arm and the median survival time was 3.3 months. Toxicities were acceptable [152].
Two phase II studies of R115777 administered at 300 mg twice a day for 21 days of each 28-day cycle recently have been reported and concluded that this agent has no significant therapeutic benefit as a single-agent treatment for patients with chemotherapy-naïve locally advanced or metastatic pancreatic cancer. In one multi-institutional phase II trial of 20 patients, the median time to progression was 4.9 weeks and median survival was 19.9 weeks. The estimated 6-month survival rate was 28.3%, with no patients progression free at 6 months [153]. In the second study, by the Southwest Oncology Group, 47 patients were evaluable. The median survival was 2.7 months, 6-month survival rate was 17%, and the median time to treatment failure was 1.3 months [154].
Unfortunately, R115777 has not demonstrated additive or synergistic therapeutic benefit when combined with gemcitabine. A phase III randomized, double-blind trial compared gemcitabine with either R115777 200 mg twice a day given 21 of every 28 days or placebo. Six hundred eighty-eight patients, 76% with metastatic disease, were enrolled in 133 centers. No statistically significant differences in survival parameters were observed. The median overall survival for gemcitabine plus R115777 was 193 days, versus 182 days for the control arm. The 6-month and 1-year survival rates for gemcitabine plus R115777 were 53% and 27%, versus 49% and 24% for the control arm. The investigators concluded that, while the gemcitabine plus R115777 combination had an acceptable toxicity profile, it did not prolong survival outcomes compared with single-agent gemcitabine [155].
Targeting the Host Microenvironment
The complex mechanisms by which angiogenesis occurs and tumors or metastases grow also represent targets of molecular therapy. In pancreatic cancer, the role of matrix degradation is well known as an essential step and a promoter of tumor growth, in both the primary tumor and metastases. Matrix metalloproteinases have a central role in this process [156, 157].
A number of matrix metalloproteinase inhibitors (MMPIs) have been developed, including marimastat, BAY 12-9566, and BMS-275291. Marimastat as a single agent has been evaluated in a phase III randomized study comparing it at varying doses with gemcitabine (four-arm study) [158–160]. Three arms assessed marimastat at 5, 10, or 25 mg orally twice a day, and the fourth arm consisted of gemcitabine 1,000 mg/m2 weekly. Median survival favored the gemcitabine control arm (167 days compared with 111, 105, and 125 days for the marimastat arms). The 1-year survival rates were comparable for the gemcitabine arm and the 25 mg twice a day arm of marimastat [161].
Concurrent administration of marimastat and gemcitabine in unresectable pancreatic cancer was evaluated in a phase IB study without increase in the toxicity profiles or the incidence of side effects [162]. The combination is presently undergoing phase II evaluation.
Antiangiogenesis therapies also can be directed against the microenvironment. Vascular endothelial growth factor (VEGF) is a growth factor secreted by tumor cells that binds the VEGF receptor, activating its associated tyrosine kinase. SU5416 and SU6668 are potent inhibitors of the VEGF receptor and the VEGF receptor tyrosine kinase and are being evaluated in clinical studies [163–165]. Other antiangiogenic molecules such as endostatin, angiostatin, and TNP-470, a fumigallin analogue, are undergoing phase I trials, and there are no reports yet of their use in pancreatic cancer.
Of interest, gastrin has been identified as a growth peptide in human pancreatic cancer, and coexpression of gastrin and its receptor in human pancreatic adenocarcinomas has been demonstrated [166, 167]. G17DT is an antigastrin immunogen consisting of a nine-amino-acid fragment of the amino terminal of human gastrin-17 conjugated with diphtheria toxoid and formulated as a water-in-oil emulsion suitable for intramuscular injection. It has been evaluated as a treatment for metastatic pancreatic cancer in a phase II trial, comparing it with best supportive care. Median survival was significantly better with G17DT than with best supportive care (402 days versus 111.5 days). A phase III trial is ongoing [168].
|
CONCLUSION |
|---|
|
TopAbstractIntroductionSymptoms and Palliative...ChemotherapyConclusionReferences |
|---|
If a patient does not choose entry into a clinical trial, single-agent gemcitabine is the accepted standard as palliative chemotherapy. Administering gemcitabine at a fixed-dose rate of 10 mg/m2/min may be of additional benefit compared with a standard 30-minute infusion. Although response rates are poor, the clinical benefit associated with gemcitabine is significant.
Patient accrual to clinical trials incorporating novel chemotherapy combinations or targeted therapy is encouraged. Over the next several years, we will learn much more about the use of targeted therapies and new chemotherapies in this disease.
|
REFERENCES |
|---|
|
TopAbstractIntroductionSymptoms and Palliative...ChemotherapyConclusionReferences |
|---|
