Since the introduction of gemcitabine in the treatment of pancreatic cancer, progress in the use of combination chemotherapies has been very limited. Of the different novel options, anti-angiogenic treatment strategies are among those being intensively studied in preclinical and clini¬cal settings of adenocarcinoma of the pancreas. Phase I and limited-size phase II studies using drugs with antiangiogenic properties have re¬ported encouraging results. Overall, the results of phase III studies with some metalloprotease inhibitors and bevacizumab have so far failed to demonstrate a survival benefit for these drugs. Further investigations that will take into account the heterogeneity of pancreatic cancer are war¬ranted using these or other antiangiogenic active substances.
14.1 Introduction
Targeted therapies inhibiting specific pathways that facilitate the growth and progress of malig-nant tumors are under intensive investigation in oncology. So far, several agents developed to in-hibit tumor angiogenesis have been evaluated in clinical studies [11, 20]. In this context advanced adenocarcinoma of the pancreas represents a malignancy worth studying.
The highly aggressive nature of pancreatic cancer may be due to the expression of growth factors, resulting in high intrinsic tyrosine kinase activities that stimulate cell proliferation, dis-semination, and neoangiogenesis. A number of studies have indicated that angiogenesis involves several cytokines such as vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and angiopoietin 1 [1, 4, 21, 28]. VEGF and bFGF are clearly overexpressed in pancreatic cancer and result in accelerated tumor growth [29, 46]. The cellular effects of these fac-tors are mediated by specific cell surface recep-tors with intrinsic tyrosine kinase activities.
The results of qualitative and semiquantita-tive assessments of VEGF and VEGF receptor expressions in pancreatic cancer have been con¬firmed by quantitative methods, demonstrating that VEGF and its two principal receptors were expressed in adenocarcinomas of the pancreas [8, 9]. These expressions vary in different indi¬vidual tumors, but a significant association was found between expression of VEGF-R2 and poor prognosis, suggesting that VEGF-R2 expression is a marker of more aggressive disease [8]. Both VEGF and VEGF-receptors are overexpressed in pancreatic cancer, and the degree of expression correlates with microvessel density and poor prognosis [40]. This is true not only in advanced metastatic pancreatic adenocarcinoma but also for early recurrences after curatively intended resection [37].
In preclinical models, inhibition of VEGF decreased pancreatic cancer growth by inhibi¬tion of neovascularization and lymphangio-genesis [4, 9, 31]. In addition, VEGF inhibition decreases interstitial pressure within the tumor and thereby increases the delivery of chemo-therapeutic agents, suggesting an additive effect of antiangiogenic and conventional cytotoxic therapies [4, 50].
In addition to synergistic interactions with other treatment modalities, inhibition of tumor angiogenesis theoretically offers several benefits such as a lack of severe toxicities familiar to cy-totoxic therapies, a lack of tumor resistance, and thus a potent antitumor effect. It has been dem¬onstrated that a disruption of VEGF signaling, for example by the use of neutralizing antibod¬ies, slows tumor growth in preclinical models. It is important to realize that the antiangiogenetic suppression of VEGF signaling targets the nutri¬tional support of tumor cells by inhibiting blood vessel formation.
Treatment with inhibitors of angiogenesis most likely will prevent tumor progression, fa¬cilitate tumor cell killing by cytotoxic agents, radiation, or immunological strategies, and may induce tumor cell senescence and stimulate apoptosis. In addition to the antitumor efficacy of antiangiogenic agents demonstrated in pa¬tients with different tumor types and the antitu¬mor effect of cytotoxic drugs, the combination may result in a supra-additive effect.
While an increasing number of new drugs for antiangiogenetic strategies have been developed, it has become clear that the classical cytotoxic drug that is effective in the treatment of patients with pancreatic cancer, gemcitabine, may have antiangiogenetic effects too.
14.2 Results in Preclinical Models or Phase I Studies in Patients
Small molecule tyrosine kinase inhibitors such as PD173074 block VEGF-R2 as well as bFGF-R1 signaling, thus showing antiangiogenetic and antimitogenic activities and abrogating two im-portant pathways in cancer growth and metas-tasis [9]. In human pancreatic cancer cell lines, PD173074 inhibited cell growth in correlation with the level of bFGF-R1 expression of the tu¬mor cells. This resulted in the inhibition of cell cycle progression at the G0/G1 transmission point with a consecutive increase in apoptosis [9, 11, 20]. In a model of xenografted human pancreatic cancer cells, inhibition of orthotopic tumor growth was achieved most likely by a combination of inhibited mitogenesis, in¬creased apoptosis, and reduced angiogenesis in PD173074-treated animals. This study showed that a high-dose single drug treatment may have clinically relevant therapeutic activity, but data in patients are not yet available.
The concept of creating synergistic antitumor effects by the combined inhibition of several pathways through the use of combination thera-pies [such as by inhibiting the epidermal growth factor receptor (EGFR) and the VEGF-receptor pathways using different drugs] represents an attractive approach. Indeed, clinical studies in patients with advanced colorectal cancer have shown an additive affect of such a combination with regard to response, progression-free sur¬vival, and overall survival using the anti-VEGF bevacizumab and the anti-EGFR cetuximab. This approach has been clinically tested in patients with pancreatic cancer too, but results are not yet available [33].
Using multikinase inhibitors such as sorafenib—thus targeting several pathways, in-cluding VEGF signaling, via a single molecule, and thus showing antiangiogenetic properties in preclinical models—is an alternative treat¬ment strategy to the combination of drugs. The treatment option with sorafenib has been dem-onstrated in preclinical models and in early clinical studies to have activity against several tumor types including adenocarcinoma of the pancreas.
Matrix metalloproteinases (MMP) are dif¬ferent proteolytic enzymes responsible for the breakdown of connective tissue proteins [12]. They play an important role in growth regula¬tion, differentiation, tumor cell spread, and tis¬sue repair. The activity of MMP is highly regu¬lated at different levels. Preclinical and clinical data show that MMP overexpression correlates with increased tumor cell growth and spread in several malignancies [26, 38, 41]. Inhibition of MMP and restoring the normal balance of pro-teolytic activity may prevent tumor growth and metastasis [13, 18, 49]. A low molecular weight MMPinhibitor, marimastat, has been demon-strated to inhibit tumor growth and spread in preclinical cancer models [48, 49] including hu-man pancreatic cancer.
The mammalian target of rapamycin (mTOR) is a serine-threonine kinase with important effects on the regulation of cell growth and proliferation [17]. In a preclinical model of human pancreatic cancer rapamycin alone and, more actively, its combination with an anti-VEGF antibody strongly inhibited primary and metastatic tumor growth [42]. In this in¬vestigation combination therapy improved the effect of single agent treatments. Rapamycin in combination with anti-VEGF antibody inhibited pancreatic tumor cell proliferation, induced apoptosis, and decreased tumor angiogenesis. A clinical trial with rapamycin in pancreatic cancer is ongoing.
14.3 Clinical Studies in Patients
with Pancreatic Adenocarcinoma
There are several different antiangiogenic treat-ment strategies that have been investigated in pa-tients with pancreatic cancer. These include the matrix metalloproteinase inhibitors, agents in-terfering with the VEGF signaling pathway, and several others such as thalidomide, cyclooxygen-ase II inhibitors, and EGFR inhibitors.
14.3.1 Matrix Metalloproteinase Inhibitors
A randomized dose-finding study [5] compar¬ing single agent marimastat, a low molecular weight, broad-spectrum MMP inhibitor, with gemcitabine demonstrated a clear dose-response effect for marimastat, with 1-year survival rates of marimastat at a dose of 25 mg b.i.d. similar to that of gemcitabine, whereas the lower doses (5 and 10 mg b.i.d.) were clearly less effective than gemcitabine.
Another prospective randomized study [6] comparing marimastat (10 mg b.i.d.) in com-bination with gemcitabine to gemcitabine plus placebo in 269 patients with advanced pancreatic cancer failed to demonstrate a significant survival benefit by the addition of marimastat (median survival 165.5 vs 164 days; 1-year survival 18% vs 17%). Grade 3 or 4 musculoskeletal toxicity, a well-known adverse event of marimastat therapy, was reported in less than 5% of the patients.
Due to this study, the further development of marimastat in this indication was stopped.
BAY 12-9566, another MMP inhibitor with
antiangiogenetic properties, is well tolerated and showed antitumor activity in a phase I study in pancreatic adenocarcinoma [19, 22, 24, 25]. A phase III study [36] comparing head-to-head BAY 12-9566 (800 mg b.i.d.) with gemcitabine enrolled 277 patients. In this study the median survival of the MMP inhibitor was demonstrated to be significantly inferior to gemcitabine (3.74 vs 6.59 months).
14.3.2 VEGF Pathway Inhibition
Bevacizumab, a recombinant anti-VEGF mono-clonal antibody, first demonstrated significant improvement in response, progression-free sur-vival, and overall-survival in a phase III random-ized trial in patients with advanced colorectal cancer, when combined with cytotoxic drugs (be-vacizumab 5 mg/kg every second week, together with folinic acid, 5-fluorouracil and irinotecan) [27]. Based on these findings and some evidence of clinical efficacy in pancreatic cancer but with¬out a cancer-specific dose-finding investigation, a phase II study [32] in 52 patients with advanced pancreatic adenocarcinoma combining bevaci-zumab (10 mg/kg every second week) with gem¬citabine as the first-line treatment was initiated. Of the patients, 21% achieved a partial response and another 46% of patients had stable disease resulting in a median progression-free survival of 5.4 months and a median overall survival of 8.8 months—results clearly better than those usually reported for single-agent gemcitabine. Grade 3 and 4 toxicities included hypertension and thrombosis in 19% and 13% of the patients, respectively. Intestinal perforations occured in 8% and bleeding in 2%. There was no correlation observed between pretreatment VEGF plamsa levels and response to anti-VEGF therapy.
Due to these results, a phase III trial [34] of gemcitabine plus bevacizumab as compared to gemcitabine plus placebo was initiated by the cancer and leukemia group B (CALGB) but stopped recruitment after an interim analysis showing that there was no chance of reaching the primary study endpoint of a significant pro-longation of overall survival due to the addition of bevacizumab. Accordingly, the first study re-sults presented at the American Society of Clini-cal Oncology (ASCO)-GI 2007 symposium do not support the superiority of bevacizumab plus gemcitabine: median survival of the 302 patients on bevacizumab and gemcitabine was 5.7 months as compared to 6.0 months for the 300 patients on placebo and gemcitabine.
There is some discussion with regard to the 10 mg/kg of bevacizumab given every 2 weeks used in this trial, as studies in colorectal cancer had demonstrated clinical activity with lower doses (5 mg/kg every 2 weeks) of bevacizumab. On the other hand (interestingly too) there was no increase in gastrointestinal perforations (0% vs 0%) or grade 3/4 bleedings (3% vs 2%) re¬ported for the experimental arm.
Despite the fact that single agent gemcitabine is standard therapy for patients with advanced pancreatic cancer [10], there are several hints in favor of combination chemotherapy in this in-dication too. This is true especially for the com-bination of gemcitabine with platin or fluoropy-rimidine derivates. Premature results of ongoing multicenter phase II studies reported response rates of 11%-22% and median overall survival times of 7.5 to 8.9 months when bevacizumab was added to gemcitabine/oxaliplatin, gem-citabine/cisplatin, or gemcitabine/capecitabine chemotherapy [30, 35, 44]. In these studies se¬vere toxicities with regard to perforations, bleed¬ing, and thromboembolism have been reported. Final study results will be reported soon and may warrant new phase III studies.
Due to the risk of visceral perforation, bleed-ing, and thrombosis, most likely due to the co-medication with bevacizumab, patients with lo¬cally advanced pancreatic cancer with duodenal involvement are excluded in most phase II study of bevacizumab plus radiation therapy in locally advanced nonmetastasized pancreatic cancer. A dose-finding study [14] for bevacizumab to¬gether with capecitabine (825 mg/m2 b.i.d. 5 days a week) and concomitant to radiation therapy with 50 Gy was initiated in patients with locally advanced pancreatic adenocarcinoma. The addi-tion of bevacizumab did not increase the acute toxicity of this radiochemotherapy as compared to historical controls. An overall response rate of 19% was obtained with 6 out of 12 patients, dem-onstrating a partial response at the 5 mg/kg level of bevacizumab.
14.3.3 Other Antiangiogenic Strategies
14.3.3.1 Thalidomide
Thalidomide has shown antitumor activities in several malignancies, especially hematological neoplasms such as multiple myeloma [45]. In ad-dition, in renal cell carcinoma, prostate cancer, and hepatocellular cancer, antitumor effects have been reported [43]. The antitumor effect of tha-lidomide is not clearly understood [16]. In addi-tion to inhibition of tumor necrosis factor alpha, clinical efficacy may be mediated, at least in part, by anti-VEGF effects. A phase I/II clinical study [15] of combination therapy with gemcitabine, celecoxib (400 mg b.i.d.), and thalidomide (200-300 mg o.d.) resulted in no tumor regression, but a mean overall survival of 10 months was reported. Treatment was clinically tolerable with skin rash in 25% of the patients.
Another 50 patients with advanced pancreatic adenocarcinoma who had a weight loss of at least 10% were randomized to receive thalidomide (200 mg daily) or placebo in a single-center double-blind randomized controlled trial [23]. At 4 weeks, 17 patients on thalidomide gained an average of 0.37 kg compared to a loss of 2.21 kg in 16 controls. At 8 week, 12 patients on thalidomide lost 0.06 kg vs 3.62 kg in 8 controls. In this study, data on tumor response are not available.
14.3.3.2 Inhibition of Epidermal
Growth Factor Receptor
EGFR is a well-recognized target of anticancer therapy [2]. Monoclonal antibodies and small molecule tyrosine kinase inhibitors have been shown to be active in different cancers and have obtained approval. Results obtained with this treatment strategy are reported elsewhere in this issue of Recent Results in Cancer Research. Furthermore, the first results of a randomized phase II study [33] comparing gemcitabine/be-vacizumab together with erlotinib or cetuximab suggest an increase in response rate (ca. 20%) by the use of triple drug therapies.
If this holds true will soon become evident, as the results of a major phase III study, the com-bination of gemcitabine/erlotinib together with bevacizumab or placebo, will be reported.
14.3.3.3 Cyclooxygenase 2 Inhibition
Cyclooxygenase (Cox 2) stimulates tumor growth by its proangiogenic and apoptosis-inhibiting ef-fects. Celecoxib, a specific Cox 2 inhibitor, has antitumor activity against several human cancers in preclinical models, including pancreatic can¬cer xenograft [3, 39, 47, 53].
There were 28 patients included in a phase II study [52] evaluating the role of celecoxib in addition to gemcitabine in advanced pancre¬atic cancer. The results of 20 patients were pre¬sented on the annual meeting of the ASCO in 2006 and showed a median overall survival of 6.2 months—not dramatically different from re¬sults expected for gemcitabine alone. Grade 3 or 4 nonhematological toxicities included nausea, vomiting, supraventricular arrhythmias, dys¬pnea, pleural effusions, and hyponatremia, as well as gastrointestinal bleeding in one patient.
Further evaluation of this agent has been stopped due to cardiac toxicity possibly caused by celecoxib [7].
14.4 Conclusions
Adenocarcinoma of the pancreas remains a major cause of cancer morbidity and mortal¬ity worldwide. Recent results confirm that the dismal prognosis of patients can be improved by small steps. Still, there is an urgent need for novel treatment strategies. In contrast to anti-EGFR targeted, small-molecule therapy, the first large clinical trials failed to prove the effectivity of other antiangiogenic strategies in combina¬tion with gemcitabine in this indication. Never¬theless, stimulated by the results with erlotinib, antiangiogenic treatment options continue to be among those most likely to further improve the prognosis in this tumor entity.
14.1 Introduction
Targeted therapies inhibiting specific pathways that facilitate the growth and progress of malig-nant tumors are under intensive investigation in oncology. So far, several agents developed to in-hibit tumor angiogenesis have been evaluated in clinical studies [11, 20]. In this context advanced adenocarcinoma of the pancreas represents a malignancy worth studying.
The highly aggressive nature of pancreatic cancer may be due to the expression of growth factors, resulting in high intrinsic tyrosine kinase activities that stimulate cell proliferation, dis-semination, and neoangiogenesis. A number of studies have indicated that angiogenesis involves several cytokines such as vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and angiopoietin 1 [1, 4, 21, 28]. VEGF and bFGF are clearly overexpressed in pancreatic cancer and result in accelerated tumor growth [29, 46]. The cellular effects of these fac-tors are mediated by specific cell surface recep-tors with intrinsic tyrosine kinase activities.
The results of qualitative and semiquantita-tive assessments of VEGF and VEGF receptor expressions in pancreatic cancer have been con¬firmed by quantitative methods, demonstrating that VEGF and its two principal receptors were expressed in adenocarcinomas of the pancreas [8, 9]. These expressions vary in different indi¬vidual tumors, but a significant association was found between expression of VEGF-R2 and poor prognosis, suggesting that VEGF-R2 expression is a marker of more aggressive disease [8]. Both VEGF and VEGF-receptors are overexpressed in pancreatic cancer, and the degree of expression correlates with microvessel density and poor prognosis [40]. This is true not only in advanced metastatic pancreatic adenocarcinoma but also for early recurrences after curatively intended resection [37].
In preclinical models, inhibition of VEGF decreased pancreatic cancer growth by inhibi¬tion of neovascularization and lymphangio-genesis [4, 9, 31]. In addition, VEGF inhibition decreases interstitial pressure within the tumor and thereby increases the delivery of chemo-therapeutic agents, suggesting an additive effect of antiangiogenic and conventional cytotoxic therapies [4, 50].
In addition to synergistic interactions with other treatment modalities, inhibition of tumor angiogenesis theoretically offers several benefits such as a lack of severe toxicities familiar to cy-totoxic therapies, a lack of tumor resistance, and thus a potent antitumor effect. It has been dem¬onstrated that a disruption of VEGF signaling, for example by the use of neutralizing antibod¬ies, slows tumor growth in preclinical models. It is important to realize that the antiangiogenetic suppression of VEGF signaling targets the nutri¬tional support of tumor cells by inhibiting blood vessel formation.
Treatment with inhibitors of angiogenesis most likely will prevent tumor progression, fa¬cilitate tumor cell killing by cytotoxic agents, radiation, or immunological strategies, and may induce tumor cell senescence and stimulate apoptosis. In addition to the antitumor efficacy of antiangiogenic agents demonstrated in pa¬tients with different tumor types and the antitu¬mor effect of cytotoxic drugs, the combination may result in a supra-additive effect.
While an increasing number of new drugs for antiangiogenetic strategies have been developed, it has become clear that the classical cytotoxic drug that is effective in the treatment of patients with pancreatic cancer, gemcitabine, may have antiangiogenetic effects too.
14.2 Results in Preclinical Models or Phase I Studies in Patients
Small molecule tyrosine kinase inhibitors such as PD173074 block VEGF-R2 as well as bFGF-R1 signaling, thus showing antiangiogenetic and antimitogenic activities and abrogating two im-portant pathways in cancer growth and metas-tasis [9]. In human pancreatic cancer cell lines, PD173074 inhibited cell growth in correlation with the level of bFGF-R1 expression of the tu¬mor cells. This resulted in the inhibition of cell cycle progression at the G0/G1 transmission point with a consecutive increase in apoptosis [9, 11, 20]. In a model of xenografted human pancreatic cancer cells, inhibition of orthotopic tumor growth was achieved most likely by a combination of inhibited mitogenesis, in¬creased apoptosis, and reduced angiogenesis in PD173074-treated animals. This study showed that a high-dose single drug treatment may have clinically relevant therapeutic activity, but data in patients are not yet available.
The concept of creating synergistic antitumor effects by the combined inhibition of several pathways through the use of combination thera-pies [such as by inhibiting the epidermal growth factor receptor (EGFR) and the VEGF-receptor pathways using different drugs] represents an attractive approach. Indeed, clinical studies in patients with advanced colorectal cancer have shown an additive affect of such a combination with regard to response, progression-free sur¬vival, and overall survival using the anti-VEGF bevacizumab and the anti-EGFR cetuximab. This approach has been clinically tested in patients with pancreatic cancer too, but results are not yet available [33].
Using multikinase inhibitors such as sorafenib—thus targeting several pathways, in-cluding VEGF signaling, via a single molecule, and thus showing antiangiogenetic properties in preclinical models—is an alternative treat¬ment strategy to the combination of drugs. The treatment option with sorafenib has been dem-onstrated in preclinical models and in early clinical studies to have activity against several tumor types including adenocarcinoma of the pancreas.
Matrix metalloproteinases (MMP) are dif¬ferent proteolytic enzymes responsible for the breakdown of connective tissue proteins [12]. They play an important role in growth regula¬tion, differentiation, tumor cell spread, and tis¬sue repair. The activity of MMP is highly regu¬lated at different levels. Preclinical and clinical data show that MMP overexpression correlates with increased tumor cell growth and spread in several malignancies [26, 38, 41]. Inhibition of MMP and restoring the normal balance of pro-teolytic activity may prevent tumor growth and metastasis [13, 18, 49]. A low molecular weight MMPinhibitor, marimastat, has been demon-strated to inhibit tumor growth and spread in preclinical cancer models [48, 49] including hu-man pancreatic cancer.
The mammalian target of rapamycin (mTOR) is a serine-threonine kinase with important effects on the regulation of cell growth and proliferation [17]. In a preclinical model of human pancreatic cancer rapamycin alone and, more actively, its combination with an anti-VEGF antibody strongly inhibited primary and metastatic tumor growth [42]. In this in¬vestigation combination therapy improved the effect of single agent treatments. Rapamycin in combination with anti-VEGF antibody inhibited pancreatic tumor cell proliferation, induced apoptosis, and decreased tumor angiogenesis. A clinical trial with rapamycin in pancreatic cancer is ongoing.
14.3 Clinical Studies in Patients
with Pancreatic Adenocarcinoma
There are several different antiangiogenic treat-ment strategies that have been investigated in pa-tients with pancreatic cancer. These include the matrix metalloproteinase inhibitors, agents in-terfering with the VEGF signaling pathway, and several others such as thalidomide, cyclooxygen-ase II inhibitors, and EGFR inhibitors.
14.3.1 Matrix Metalloproteinase Inhibitors
A randomized dose-finding study [5] compar¬ing single agent marimastat, a low molecular weight, broad-spectrum MMP inhibitor, with gemcitabine demonstrated a clear dose-response effect for marimastat, with 1-year survival rates of marimastat at a dose of 25 mg b.i.d. similar to that of gemcitabine, whereas the lower doses (5 and 10 mg b.i.d.) were clearly less effective than gemcitabine.
Another prospective randomized study [6] comparing marimastat (10 mg b.i.d.) in com-bination with gemcitabine to gemcitabine plus placebo in 269 patients with advanced pancreatic cancer failed to demonstrate a significant survival benefit by the addition of marimastat (median survival 165.5 vs 164 days; 1-year survival 18% vs 17%). Grade 3 or 4 musculoskeletal toxicity, a well-known adverse event of marimastat therapy, was reported in less than 5% of the patients.
Due to this study, the further development of marimastat in this indication was stopped.
BAY 12-9566, another MMP inhibitor with
antiangiogenetic properties, is well tolerated and showed antitumor activity in a phase I study in pancreatic adenocarcinoma [19, 22, 24, 25]. A phase III study [36] comparing head-to-head BAY 12-9566 (800 mg b.i.d.) with gemcitabine enrolled 277 patients. In this study the median survival of the MMP inhibitor was demonstrated to be significantly inferior to gemcitabine (3.74 vs 6.59 months).
14.3.2 VEGF Pathway Inhibition
Bevacizumab, a recombinant anti-VEGF mono-clonal antibody, first demonstrated significant improvement in response, progression-free sur-vival, and overall-survival in a phase III random-ized trial in patients with advanced colorectal cancer, when combined with cytotoxic drugs (be-vacizumab 5 mg/kg every second week, together with folinic acid, 5-fluorouracil and irinotecan) [27]. Based on these findings and some evidence of clinical efficacy in pancreatic cancer but with¬out a cancer-specific dose-finding investigation, a phase II study [32] in 52 patients with advanced pancreatic adenocarcinoma combining bevaci-zumab (10 mg/kg every second week) with gem¬citabine as the first-line treatment was initiated. Of the patients, 21% achieved a partial response and another 46% of patients had stable disease resulting in a median progression-free survival of 5.4 months and a median overall survival of 8.8 months—results clearly better than those usually reported for single-agent gemcitabine. Grade 3 and 4 toxicities included hypertension and thrombosis in 19% and 13% of the patients, respectively. Intestinal perforations occured in 8% and bleeding in 2%. There was no correlation observed between pretreatment VEGF plamsa levels and response to anti-VEGF therapy.
Due to these results, a phase III trial [34] of gemcitabine plus bevacizumab as compared to gemcitabine plus placebo was initiated by the cancer and leukemia group B (CALGB) but stopped recruitment after an interim analysis showing that there was no chance of reaching the primary study endpoint of a significant pro-longation of overall survival due to the addition of bevacizumab. Accordingly, the first study re-sults presented at the American Society of Clini-cal Oncology (ASCO)-GI 2007 symposium do not support the superiority of bevacizumab plus gemcitabine: median survival of the 302 patients on bevacizumab and gemcitabine was 5.7 months as compared to 6.0 months for the 300 patients on placebo and gemcitabine.
There is some discussion with regard to the 10 mg/kg of bevacizumab given every 2 weeks used in this trial, as studies in colorectal cancer had demonstrated clinical activity with lower doses (5 mg/kg every 2 weeks) of bevacizumab. On the other hand (interestingly too) there was no increase in gastrointestinal perforations (0% vs 0%) or grade 3/4 bleedings (3% vs 2%) re¬ported for the experimental arm.
Despite the fact that single agent gemcitabine is standard therapy for patients with advanced pancreatic cancer [10], there are several hints in favor of combination chemotherapy in this in-dication too. This is true especially for the com-bination of gemcitabine with platin or fluoropy-rimidine derivates. Premature results of ongoing multicenter phase II studies reported response rates of 11%-22% and median overall survival times of 7.5 to 8.9 months when bevacizumab was added to gemcitabine/oxaliplatin, gem-citabine/cisplatin, or gemcitabine/capecitabine chemotherapy [30, 35, 44]. In these studies se¬vere toxicities with regard to perforations, bleed¬ing, and thromboembolism have been reported. Final study results will be reported soon and may warrant new phase III studies.
Due to the risk of visceral perforation, bleed-ing, and thrombosis, most likely due to the co-medication with bevacizumab, patients with lo¬cally advanced pancreatic cancer with duodenal involvement are excluded in most phase II study of bevacizumab plus radiation therapy in locally advanced nonmetastasized pancreatic cancer. A dose-finding study [14] for bevacizumab to¬gether with capecitabine (825 mg/m2 b.i.d. 5 days a week) and concomitant to radiation therapy with 50 Gy was initiated in patients with locally advanced pancreatic adenocarcinoma. The addi-tion of bevacizumab did not increase the acute toxicity of this radiochemotherapy as compared to historical controls. An overall response rate of 19% was obtained with 6 out of 12 patients, dem-onstrating a partial response at the 5 mg/kg level of bevacizumab.
14.3.3 Other Antiangiogenic Strategies
14.3.3.1 Thalidomide
Thalidomide has shown antitumor activities in several malignancies, especially hematological neoplasms such as multiple myeloma [45]. In ad-dition, in renal cell carcinoma, prostate cancer, and hepatocellular cancer, antitumor effects have been reported [43]. The antitumor effect of tha-lidomide is not clearly understood [16]. In addi-tion to inhibition of tumor necrosis factor alpha, clinical efficacy may be mediated, at least in part, by anti-VEGF effects. A phase I/II clinical study [15] of combination therapy with gemcitabine, celecoxib (400 mg b.i.d.), and thalidomide (200-300 mg o.d.) resulted in no tumor regression, but a mean overall survival of 10 months was reported. Treatment was clinically tolerable with skin rash in 25% of the patients.
Another 50 patients with advanced pancreatic adenocarcinoma who had a weight loss of at least 10% were randomized to receive thalidomide (200 mg daily) or placebo in a single-center double-blind randomized controlled trial [23]. At 4 weeks, 17 patients on thalidomide gained an average of 0.37 kg compared to a loss of 2.21 kg in 16 controls. At 8 week, 12 patients on thalidomide lost 0.06 kg vs 3.62 kg in 8 controls. In this study, data on tumor response are not available.
14.3.3.2 Inhibition of Epidermal
Growth Factor Receptor
EGFR is a well-recognized target of anticancer therapy [2]. Monoclonal antibodies and small molecule tyrosine kinase inhibitors have been shown to be active in different cancers and have obtained approval. Results obtained with this treatment strategy are reported elsewhere in this issue of Recent Results in Cancer Research. Furthermore, the first results of a randomized phase II study [33] comparing gemcitabine/be-vacizumab together with erlotinib or cetuximab suggest an increase in response rate (ca. 20%) by the use of triple drug therapies.
If this holds true will soon become evident, as the results of a major phase III study, the com-bination of gemcitabine/erlotinib together with bevacizumab or placebo, will be reported.
14.3.3.3 Cyclooxygenase 2 Inhibition
Cyclooxygenase (Cox 2) stimulates tumor growth by its proangiogenic and apoptosis-inhibiting ef-fects. Celecoxib, a specific Cox 2 inhibitor, has antitumor activity against several human cancers in preclinical models, including pancreatic can¬cer xenograft [3, 39, 47, 53].
There were 28 patients included in a phase II study [52] evaluating the role of celecoxib in addition to gemcitabine in advanced pancre¬atic cancer. The results of 20 patients were pre¬sented on the annual meeting of the ASCO in 2006 and showed a median overall survival of 6.2 months—not dramatically different from re¬sults expected for gemcitabine alone. Grade 3 or 4 nonhematological toxicities included nausea, vomiting, supraventricular arrhythmias, dys¬pnea, pleural effusions, and hyponatremia, as well as gastrointestinal bleeding in one patient.
Further evaluation of this agent has been stopped due to cardiac toxicity possibly caused by celecoxib [7].
14.4 Conclusions
Adenocarcinoma of the pancreas remains a major cause of cancer morbidity and mortal¬ity worldwide. Recent results confirm that the dismal prognosis of patients can be improved by small steps. Still, there is an urgent need for novel treatment strategies. In contrast to anti-EGFR targeted, small-molecule therapy, the first large clinical trials failed to prove the effectivity of other antiangiogenic strategies in combina¬tion with gemcitabine in this indication. Never¬theless, stimulated by the results with erlotinib, antiangiogenic treatment options continue to be among those most likely to further improve the prognosis in this tumor entity.
Комментариев нет:
Отправить комментарий