Cancer Treatment Reviews

Transcription

1 Cancer Treatment Reviews xxx (2011) xxx xxx Contents lists available at SciVerse ScienceDirect Cancer Treatment Reviews journal homepage: Antitumour treatment Systemic treatment of advanced pancreatic cancer Volker Heinemann, Michael Haas, Stefan Boeck Klinikum Grosshadern, Department of Oncology and Comprehensive Cancer Center, Ludwig-Maximilians-Universität München, Germany article info abstract Article history: Received 18 October 2011 Received in revised form 9 December 2011 Accepted 12 December 2011 Available online xxxx Keywords: Pancreatic cancer Chemotherapy Supportive therapy Targeted therapy Pancreatic cancer belongs to the most malignant gastrointestinal cancers and, in its advanced stage, remains a deadly disease for nearly all affected patients. Treatment of metastatic adenocarcinoma of the pancreas not only involves chemotherapy and targeted therapy, but also requires attention to accompanying comorbidities as well as frequently intensive supportive treatment and psychosocial support. Gemcitabine-based combinations with fluoropyrimidines and platin analogs have essentially failed to provide a substantial prolongation of survival and may constitute a treatment option only in patients with a good performance status. Among targeted therapies, only the EGFR tyrosine kinase inhibitor erlotinib has shown activity which is marginal in the overall population, but clinically relevant in patients developing skin rash. New avenues of polychemotherapy are presently explored since the gemcitabinefree FOLFIRINOX-regimen (infusional 5-fluorouracil/folinic acid plus irinotecan and oxaliplatin) was shown to be markedly superior to gemcitabine in selected good-performance patients. Pancreatic cancer is notably characterized as a hypovascular tumor rich in desmoplastic stromal tissue. An innovative approach to treatment therefore focuses on peritumoral fibroblasts and aims to induce a depletion of the stroma either by inhibition of the hedgehog pathway or by targeting SPARC (secreted protein acidic and rich in cysteine) via application of albumin-bound paclitaxel. Ó 2011 Elsevier Ltd. All rights reserved. Introduction Only a minority of pancreatic cancer (PC) patients (15 20%) present with resectable disease at first diagnosis. Patients with locally advanced, non-metastatic PC (LAPC) represent 15 20% of patients and have a median survival of 9 11 months. Metastatic disease, by comparison, is documented in 60 70% of patients and is associated with a much shorter survival of only 6 8 months in most studies. This observation has meanwhile led to a broad consensus that LAPC and metastatic PC should be viewed as prognostically (and possibly also biologically) different disease entities which require different treatment algorithms and accordingly should be investigated in separate clinical studies. 1 The present review evaluates the available data on systemic treatment of advanced PC and specifically focuses on chemotherapy and biologically targeted therapy. In this context, it needs to be recognized that most of the randomized studies, even if they predominantly investigated metastatic PC, they also included varying, but smaller amounts of LAPC patients (Tables 1 5). Only some of the more recent trials were rightfully limited to metastatic PC. The available evidence therefore regards advanced pancreatic cancer as a composite group in most studies. Since data were mostly Corresponding author. Tel.: ; fax: address: [email protected] (V. Heinemann). derived from patients with metastatic PC, conclusions may have the greatest validity for this patient subgroup. Post hoc evaluations of treatment results obtained in LAPC were performed in some trials, but, due to the limited number of patients, did not allow formal conclusions. This smaller patient subgroup therefore needs to be addressed separately in properly designed trials. The present review evaluates the available armamentarium of anticancer agents and aims to clarify the present treatment options. In view of the dismal prognosis of advanced disease, it also emphasizes the importance of adequate supportive therapy and early provision of psychosocial help. Diagnosis Once standard imaging procedures (CT or MRT) led to the classification of the tumor as non-resectable, the patho-histological or -cytological verification of neoplastic disease needs to be performed before oncological treatment can start. The clinical constellation of a typical mass in the pancreas combined with an elevated tumor marker (CA 19-9 or CEA) alone is not sufficient. Specifically, the interpretation of elevated CA 19-9 can be confounded by any affection of the biliary system frequently accompanying pancreatic disease. Accordingly, a biopsy of the pancreatic mass or associated metastatic lesions is strongly recommended in all patients with advanced disease /$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi: /

2 2 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx Table 1 Gemcitabine-based polychemotherapy regimens. Reference Regimen Study design No of pts LAPC (%) ORR (%) PFS/TTP (mo) OS (mo) 1-Year survival (%) Reni et al. 19 PEFG Phase III na 38.5 GEM na 21.3 Wagner et al. 20 GEM/oxaliplatin/infusional 5-FU Phase II Endlicher et al. 21 GEM/irinotecan/5-FU Phase II Hess et al. 22 GEMOXEL phase I/II Phase I/II na GEM, gemcitabine; PEFG, cisplatin, epirubicin, gemcitabine, 5-FU; GEMOXEL, gemcitabine, oxaliplatin, capecitabine; LAPC, locally advanced pancreatic cancer; na, data not available. Table 2 Fluoropyrimidine-/oxaliplatin-based chemotherapy (selected studies). Reference Regimen Study design No of pts LAPC (%) ORR (%) PFS/TTP (mo) OS (mo) 1-Year survival (%) First-line therapy Ducreux et al. 26 CI 5-FU Randomized Phase II Oxaliplatin CI 5-FU + oxaliplatin >20 Ghosn et al. 27 FOLFOX6 Phase II Boeck et al. 28 Capecitabine + oxaliplatin Phase II Second-line therapy Gebbia et al. 29 FOLFOX4 Retrospective survey na Pelzer et al FU/FA + oxaliplatin Randomised phase II na 3.3 * 6.5 ** na 5-FU/FA na na Xiong et al. 31 Capecitabine + oxaliplatin Phase II LAPC, locally advanced pancreatic cancer; CI, continuous infusion; FOLFOX, combination of infusional 5-fluorouracil, folinic acid, and oxaliplatin; CAPOX, combination of capecitabine and oxaliplatin; 5-FU/FA, combination of 5-fluorouracil and folinic acid; na, data not available. * p = ** p = Table 3 Irinotecan-based chemotherapy (selected studies). Reference Regimen Study design No of pts LAPC (%) ORR (%) PFS/TTP (mo) OS (mo) 1-Year survival (%) First-line therapy Rocha-Lima et al. 32 GEM/irinotecan Randomised phase III GEM Stathopoulos et al. 33 GEM/irinotecan Randomised phase III GEM Taieb et al. 34 FOLFIRI.3 Phase II % Second-line therapy Yi et al. 35 Irinotecan Phase II na Gebbia et al. 36 FOLFIRI Retrospective survey na Cereda et al. 37 CAPIRI/FOLFIRI Observational study 34 na na Yoo et al. 38 FOLFIRI.3 Randomised phase II 31 na na FOLFOX 30 na na GEM, gemcitabine; FOLFIRI, combination of infusional 5-fluorouracil, folinic acid, and irinotecan; CAPIRI, combination of capecitabine and irinotecan; FOLFOX, combination of infusional 5-fluorouracil, folinic acid, and oxaliplatin; na, data not available. Table 4 Chemotherapy with the FOLFIRINOX regimen. Reference Regimen Study design No of pts LAPC (%) ORR (%) PFS/TTP (mo) OS (mo) 1-Year survival (%) Conroy et al. 39 FOLFIRINOX Phase II Conroy et al. 23 FOLFIRINOX Phase III * 6.4 * 11.1 * 48.4 GEM LAPC, locally advanced pancreatic cancer; FOLFIRINOX, infusional 5-FU/folinic acid plus irinotecan and oxaliplatin; na, data not available. * P < Psychosocial support The sudden confrontation with a limitation of life expectancy to several months may cause substantial distress and poses problems most patients have not learned to cope with. Clearly, reliable information on prognosis can only be provided to patients once the diagnostic work-up has been completed and once a final report on the stage of disease has been obtained. Only then, the oncologist may provide a realistic prognosis and may propose an adequate concept of cancer treatment. 2 Importantly, this information is intended to guide patient decisions in view of a shortened life span, it is not meant to take away all elements of hope. The oncologist therefore not only needs to clarify the technical aspects of treatment in a timely manner, he also has the task to understand the expectations and psychosocial needs of the patient. To improve patient care, a first step may be to implement routine

3 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx 3 Table 5 Patient characteristics in phase III trials. Burris et al. 14 Moore et al. 41 Van Cutsem et al. 46 Cunningham et al. 16 Conroy et al. 23 Median age (years) 61/ / Male gender (%) Performance status (%) 0 (KPS 100%) 30 (ECOG 0) 23 (KPS 100%) 23 (ECOG 0) 38 (ECOG 0) 31 (KPS 80 90%) 51 (ECOG 1) 64 (KPS 80 90%) 58 (ECOG 1) 62 (ECOG 1) 69 (KPS 50 70%) 19 (ECOG 2) 13 (KPS 60 70%) 19 (ECOG 2) 0 (ECOG 2) Site of primary tumor Head of pancreas (%) na na 54% 70% 38% Corpus/cauda (%) na na 46% 28% 59% Missing (%) 2% Multicentric (%) 3% ECOG, Eastern Cooperative Oncology Group performance status; KPS, Karnofsky performance status; na, data not available. screening for distress. 3 Psychosocial interventions may then be offered to selected patients requiring and requesting help. Another option may be to offer psychosocial support as a routine measure accompanying anticancer therapy. Expectedly this offer will be accepted only by a fraction of patients. Early access to palliative treatment Patients are usually referred to palliative care in a late stage of their disease. Improved results with regard to quality and delivery may, however, be obtained when palliative care is offered early on and is further applied during the continuum of treatment. In this context, palliative care means control of symptoms, psychological support and assistance in decision making. 4 That approach was previously shown to be beneficial in lung cancer, a comparably malignant and incurable illness. In a randomized trial, early palliative care contributed to a better quality of life, fewer depressive symptoms (16% vs. 38%, P = 0.01), and a longer median survival (11.6 vs. 8.9 months, P = 0.02). 4 Most importantly, survival was prolonged in patients receiving early palliative care despite the fact that aggressive end-of-life care was applied less frequently (33% vs. 54%, P = 0.05). This can either be explained by the positive effects of an intensive supportive and psychosocial care on symptoms control and psychosocial well being or as the benefit from earlier referral to a hospice program. In addition, it cannot be excluded that less intravenous chemotherapy near the end of life may also have an effect on survival. 2 These data clearly need to be confirmed prospectively in PC patients. Nevertheless, also at this state of knowledge it may be concluded that efforts to maintain physical and psychosocial well being should be given a high priority and should be integrated into a comprehensive treatment concept. Management of comorbidity Pancreatic cancer is characterized by a number of comorbidities which often require increased attention and a close follow-up by the treating physician. Many patients suffer from pain as a primary disease-related symptom which requires appropriate medication or application of a celiac blockade. Before antitumor treatment can be started in advanced disease, it is necessary to ensure optimal bile drainage and to apply biliary stents when hyperbilirubinemia due to bile duct obstruction is present. Patients with biliary stents in place have a significant risk of recurring stent occlusion and cholangitis. They need to be well informed about their risk situation and need to have rapid access to diagnostic facilities, medical treatment and antibiotic therapy if necessary. Patients frequently present with a marked loss of body weight and even kachexia at first diagnosis. Optimal nutritional support is necessary to stabilize body weight. In case of bowel obstruction, therapeutic measures such as parenteral feeding, surgical bypass or duodenal stenting need to be considered. When exocrine insufficiency of the pancreas is apparent accompanied by malabsorption, diarrhea, and fatty stools, oral enzyme replacement (lipase, amylase, proteases) may help to stabilize body weight. Many PC patients suffer from diabetes mellitus and its sequelae. This comorbidity may specifically gain importance when steroids are applied in the context of antiemetic medication during chemotherapy. Pre-existing sensory neuropathy may limit the application of platinum compounds in these patients. The risk of thromboembolic complications is markedly elevated in PC patients. 5 In most reports, the incidence ranges between 20% and 30% and has been associated with advanced disease and poor prognosis. 6 The unfavorable outcome may be a direct result of thromboembolic complications (e.g. pulmonary embolism) or may reflect a more aggressive biology of the disease. 7 The use of prophylactic anticoagulation remains controversial. A recent randomized study demonstrated that the prophylactic use of low-molecular weight heparin (LMWH) significantly decreased the incidence of symptomatic venous thromboembolism. This, however, had no impact on overall survival. 8 Another randomized study investigating the prophylactic use of nadoparin did also not show a survival benefit in pancreatic-, prostate- and lung-cancer patients. 9 Accordingly, LMWH are not recommended as prophylactic agents, but are applied as standard of care once venous thromboembolism has occurred. With regard to long-term use, there is evidence to support the application of LMWH over warfarin in cancer patients because of improved efficacy and outcome. 10,11 Patient selection Clearly, it needs to be asked to which extent clinical trials reflect patient reality since clinical trials tend to include younger patients and patients with a better socio-economic background. In a retrospective analysis, El-Rayes et al. demonstrated that patients treated within clinical trials at the Karmanos Cancer Institute had a median survival of 8.5 months as compared to patients treated at that institution outside clinical trials (5.0 months) or at non-institutional community centers (2.8 months). 12 Performance status belongs to the most relevant prognostic factors in PC. Other parameters recommended to characterize patient subgroups are weight loss and nutritional status. While these factors appear to be reasonable in view of the disease, it needs to be taken into account that specifically weight loss is not an objective parameter since documented analyses of normal weight do not exist in most patients. Weight loss before treatment start is therefore frequently documented only as an estimate based on assumptions of the patient. Chemotherapy is a standard of care in advanced pancreatic cancer In a meta-analytical evaluation of 7 randomized controlled trials involving 432 patients with advanced PC, Sultana and coworkers

4 4 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx demonstrated that chemotherapy improved survival compared to best supportive care with a hazard ratio (HR) of 0.64 (95% CI, ). However, it was pointed out that there was a significant heterogeneity between studies (P = ) and that the upper limit of the confidence interval (CI) was quite close to Activity of gemcitabine in 1st-line therapy Since more than a decade, the antimetabolite gemcitabine (GEM) has been established as a chemotherapeutic standard in the treatment of advanced PC. Registration of GEM was based on a randomized trial which demonstrated the superiority of GEM over bolus 5-fluorouracil (5-FU) with regard to clinical benefit response (23.8% vs. 4.8%, P = ), and median overall survival (5.65 vs months, P = ). 14 In addition, this study also indicated that GEM induced a marked improvement of 1-year survival (18% vs. 2%). A 1-year survival rate in the range of 18 28% was later confirmed by numerous trials testing single-agent GEM as a comparator treatment and thus established a clinically relevant mile stone in the treatment of PC. Gemcitabine-based chemotherapy doublets While the introduction of GEM clearly improved therapeutic efficacy and 1-year survival, its impact on median overall survival remained modest. More intensive combination chemotherapies involving fluoropyrimidines, platinum analogs and other cytotoxic agents have been investigated in numerous phase II and III trials. Most of these failed, however, to show a statistically significant survival benefit compared to GEM alone. 15 The true benefit gained from GEM-based combination chemotherapy therefore still remains a matter of debate. The problem can best be described based on a recent trial comparing GEM plus capecitabine (GEM + CAP) to GEM alone. 16 GEM-CAP significantly improved response rate (19.1% vs. 12.4%, P = 0.034) and progression-free survival (HR = 0.78, P = 0.004), however, only a strong trend toward improvement was reached with regard to overall survival (7.1 vs. 6.2 months; HR = 0.86, P = 0.08). The level of statistical significance could only be achieved by pooling the results of this trial with other two randomised trials reaching a total of 935 patients (HR, 0.86; 95% CI, , P = 0.02). 16 While the authors proclaimed GEM + CAP as one of the standard first-line options, this statement appears not sufficiently supported by the hazard ratio. Meta-analytical evaluation of gemcitabine-based doublets Since single studies were frequently criticized because of their underpowered statistical design, several meta-analyses were performed to allow more reliable conclusions based on larger patient numbers. 13 Heinemann and coworkers reported a meta-analysis of fifteen trials comparing GEM versus GEM plus cyototoxic agent (GEM + X) which revealed a significant survival benefit for GEM + X with a pooled hazard ratio (HR) of 0.91, P = 0.004). 15 An identical HR (0.91; 95% CI, ) was also published by Sultana and coworkers. 13 When different combination partners were evaluated separately, the analysis of platinum-based combinations indicated a HR of 0.85 (95% CI: , P = 0.010), while for fluoropyrimidine-based combinations a HR of 0.90 (95% CI: , P = 0.030) was reported. No risk reduction was observed in the group of trials combining GEM with irinotecan, exatecan or pemetrexed (HR = 0.99). 15 In a more recent evaluation, Vaccaro et al. analyzed seven randomised trials including 2422 patients in which single-agent GEM was compared to combinations of GEM with cisplatin, oxaliplatin or capecitabine. 17 This analysis included the more recent trial by Colucci and coworkers published in and showed a HR of 0.94 (P = 0.61) for GEM plus cisplatin. For the combination of GEM plus oxaliplatin a HR of 0.86 (P = 0.04), and for GEM plus capecitabine a HR of 0.86 (P = 0.04) were reported. 17 In addition, the authors stated that power calculations reliably (80% power, two-tailed alpha of 0.05) rule out the possibility that GEM-based doublets could improve 1-year survival by more than 5%. As a consequence and in view of the promising FOLFIRINOX data, they concluded that clinical research should move away from the investigation of GEM-based combinations and should seek for more effective strategies. 17 Gemcitabine-based polychemotherapy regimens Several studies investigated GEM-based polychemotherapy regimens involving 3 4 cytotoxic agents (Table 1). Reni and coworkers performed a small randomised trial (n = 99) testing the PEFG-regimen (cisplatin, epirubicin, fluorouracil, and GEM) versus GEM alone. 19 Progression-free survival at 4 months was evaluated as a primary endpoint and was superior in the PEFGarm (60% vs. 28%, HR = 0.46). While median overall survival was nearly identical in both treatment arms, the proportion of patients surviving at 1 year was greater in the PEFG-group (38.5% vs. 21.3%). Further three trials were performed to evaluate the efficacy of GEM combined with a fluoropyrimidine plus oxaliplatin or irinotecan They induced median overall survival times in the range of 7.5 to 8.3 months and reported 1-year survival rates of 25% and 33%. The efficacy parameters favor polychemotherapy compared to GEM alone. However, with regard to survival, the benefit appears to be less clear as observed in the FOLFIRINOX regimen. None of the indicated studies reported inclusion criteria as rigorous as reported for the Conroy study Fluorouracil-based regimen Since Burris and coworkers had demonstrated that 5-fluorouracil (5-FU) was significantly inferior to GEM, this agent appeared to have lost clinical relevance in the treatment of PC. 14 It needs, however, to be mentioned that application of 5-FU as a bolus-regimen most probably represents the least effective way, this agent can be applied. Subsequently performed studies provide more insight into the topic of infusional 5-FU regimens. Ducreux and coworkers performed a randomized trial comparing bolus 5-FU (500 mg/m 2 /day for 5 days) to an infusional 5-FU regimen (1000 mg/m 2 /day for 5 days) plus cisplatin 100 mg/m 2 on day 1 or day 2) in 207 patients with advanced PC. Median overall survival was short and comparable in both treatment arms (102 vs. 112 days). 24 The combination of 5-FU plus mitomycin was investigated by Maisey and coworkers in a randomized phase III study using 5-FU in the control arm. In both treatment arms, 5-FU was applied as a protracted venous infusion (300 mg/m 2 /day for a maximum of 24 weeks). Also in this trial, the combination arm failed to induce a significant survival benefit compared to 5-FU alone (5.1 months vs. 6.5 months, P = 034). 25 These studies lead to the conclusion that neither infusional 5-FU alone nor its combination with cisplatin or mitomycin C induced a major improvement of survival in PC. In this context, a three-armed randomized phase II study is important which compared the efficacy of the single agents 5-FU (1000 mg/m 2 /day continuous infusion days 1 4) and oxaliplatin (130 mg/m 2 day 1) to the combined application of both drugs (OXFU). 26 Median overall survival was markedly higher in the OXFU-combination arm (9.0 months) than with 5-FU (2.4 months) or oxaliplatin (3.4 months) alone. This trial, for the first time, demonstrated in PC that clinically relevant efficacy can be achieved by

5 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx 5 the combination of 5-FU plus oxaliplatin, while the respective single agents were essentially ineffective. A critical view of this study is, however, recommended not only because of the very low number of patients included (n = 63), but also because of a major imbalance of tumor stage in the three treatment groups (Table 2). The clinical efficacy of infusional 5-FU/leucovorin combined with oxaliplatin (OFF-regimen) was later confirmed in the 2nd-line treatment of PC patients. 30 Capecitabine plus oxaliplatin (CapOx) The first randomized study to evaluate the oral fluoropyrimidine capecitabine in combination with oxaliplatin (CapOx) as a palliative first-line treatment was a German AIO trial (Table 2). CapOx was compared to the combination of capecitabine plus GEM (Cap/ Gem) or the combination of GEM plus oxaliplatin (mgemox). Similar clinical efficacy was observed for the three drug combinations with regard to PFS (4.2, 5.7, 3.9 months) and overall survival (8.1, 9.0, 6.9 months). Expectedly, significant differences were observed in toxicity profiles, but side-effects were manageable. 28 In the setting of second-line treatment, the clinical activity of CapOx was moderate (PFS = 2.3 months, OS = 5.4 months) 31 (Table 2). Irinotecan-based chemotherapy The evidence on the activity of irinotecan in PC is limited (Table 3). Among others, this may be due to two randomized trials demonstrating that the addition of irinotecan to GEM did not increase treatment efficacy compared to GEM alone. 32,33 Due to the remarkable clinical activity of the FOLFIRINOX regimen the question arises to which extent irinotecan is responsible for this effect. Some evidence may come from a phase II study by Taieb and coworkers 34 who investigated a modified FOLFIRI regimen (FOLFIRI.3) where irinotecan (90 mg/m 2 ) was applied on days 1 and 3 of a 2-week schedule. This regimen showed high efficacy (ORR 37.5%, PFS 5.6 months, OS 12.1 months) in the first-line treatment of advanced PC which needs, however, to be verified by controlled trials. In a randomized study performed in the setting of second-line treatment, a dose-reduced FOLFIRI.3 regimen showed only moderate activity which was comparable to that of a modified FOLFOX regimen. 38 FOLFIRINOX a new treatment standard It is the merit of Conroy and coworkers to have introduced the FOLFIRINOX regimen, a combination of infusional 5-FU/folinic acid, irinotecan, and oxaliplatin, into the treatment of metastatic PC. 23 In a randomized phase III trial, FOLFIRINOX was compared to singleagent GEM and demonstrated the clear superiority of FOLFIRINOX with regard to objective response rate (31.6% vs. 9.4%, P < 0.001), progression-free survival (6.4 months vs. 3.3 months; HR 0.47, P < 0.001) and overall survival (11.1 months vs. 6.8 months; HR 0.57, P < 0.001). Also 1-year survival was markedly greater in the FOLFIRINOX- compared to the GEM arm (48.4% vs. 20.6%) (Table 4). While, the therapeutic activity achieved by FOLFIRINOX is unprecedented in randomized studies investigating metastatic PC, it is supported by a previous phase II study from the same group which provided nearly identical results (ORR = 26%, time to progression 8.2 months, overall survival 10.2 months). 39 In addition, the results are credible because the efficacy parameters obtained in the GEM arm fit very well to previously published data 23 (Table 4). Expectedly, the toxicity associated with the FOLFIRINOX regimen was greater than with GEM alone. Specifically, grade 3 4 neutropenia (45.7% vs 21%, P < 0.001) and febrile neutropenia (5.4% vs. 1.2%, P = 0.03) were more frequent in the combination arm and required the application of granulocyte colony stimulating factor (G-CSF) in 42.5% vs. 5.3% (P < 0.001). Also, the incidence of grade 3 4 thrombocytopenia, diarrhea, and sensory neuropathy was significantly greater in the FOLFIRINOX group. The treatment-associated mortality was, however, low with only one toxic death reported in each arm. The expectedly higher toxicity of the FOLFIRINOX regimen did not appear to have a significant impact on global health status and quality of life. Only diarrhea had higher scores during the first eight cycles of combination treatment. In addition, time to definitive degradation of global health status/quality of life was significantly longer in the FOLFIRINOX-arm. Certainly, it cannot be excluded that this observation represents, at least in part, a carry-over effect of the improved progression-free survival. However, it also may support the conclusion that, specifically in a highly malignant disease like PC, treatment efficacy is a more important determinant of quality of life than treatment-associated toxicity. Reviewing the Conroy study the question arises, how these results can be explained in view of the numerous studies on combination chemotherapy reported in the literature. The clinical relevance of the Conroy study can probably be best understood by an evaluation of the inclusion criteria. In contrast to the majority of published studies, the Conroy trial is limited to patients with metastatic PC. In fact, GEM-based combinations such as the GemOX regimen did not induce a clear benefit in locally advanced compared to metastatic PC. 40 The expected intensity of the polychemotherapy regimen required strict observation of rigorous selection criteria. Specifically, hyperbilirubinemia needed to be excluded to avoid irinotecanassociated toxicity. The results of the Conroy study are therefore valid only for previously untreated patients aged years with a good performance status (ECOG status score of 0 or 1), nearly normal bilirubin (<1.5 upper limit of normal) and without a history of ischemic cardiac disease. The exclusion of hyperbilirubinemia caused an important shift in patient selection in that the rate of patients with tumors of the pancreatic head (38%) and those with biliary stents was relatively low (14.3%). By comparison, other studies have reported a rate of pancreatic head tumors in the range of 52% to 70% (Table 5). This stringent patient selection may also be the reason why, despite a markedly increased rate of neutropenia in the FLOFIRINOX arm, no cholangitis was observed throughout the study. In addition, education, monitoring and active management of study centers may have contributed to a remarkably low mortality rate (2/342). In conclusion, the Conroy study reflects only a limited spectrum of the clinical reality. Through its selection criteria the results of the study are therefore applicable only for selected patients. In addition, the clinical use of FOLFIRINOX requires specialized centers experienced in the supportive care of patients with an elevated risk of neutropenic infections. The Conroy study also entrails a perspective regarding the development of PC treatment. For the first time, a GEM-free combination chemotherapy has successfully been used in a randomized phase III trial. This study therefore paved the way for future regimens not incorporating GEM. In addition, the high efficacy of FOLFIRINOX in the metastatic setting, for the first time, provides a sufficiently strong rationale to investigate intensive polychemotherapy in the context of multidisciplinary perioperative treatment. Inhibitors of the epidermal growth factor receptor (EGFR) Inhibition of the EGFR has become an established treatment strategy in several solid tumors such as colorectal- or lung cancer. In PC both, the anti-egfr directed antibody cetuximab and the oral

6 6 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx EGFR tyrosine kinase inhibitor erlotinib were tested in several randomised trials. Only the combination of GEM with erlotinib proved to be effective and was registered for treatment of metastatic PC. Gemcitabine plus erlotinib Moore and coworkers performed a randomized trial to compare GEM plus erlotinib versus GEM alone. 41 The addition of erlotinib to GEM induced a statistically significant improvement of progression-free- (HR = 0.77, P = 0.004) and overall survival (HR = 0.82, P = 0.038), while objective response rate was not significantly different between treatment arms. Since overall survival of LAPC patients was only marginally affected (HR = 0.94), registration of GEM plus erlotinib was limited to metastatic disease in Europe. Despite the statistical superiority of the combination, its effect on median survival time was rather low (6.24 months vs months; D = 0.33 months) which remains the source of controversial debates. In this context, it was of interest to note that patients with a rash grade P2 derived the greatest benefit from the addition of erlotinib, while survival remained short if no rash was evident (10.5 vs. 5.3 months). Rash of all grades was observed in 70% of patients and usually developed during the first 2 4 weeks of treatment. In the PA.3 study the dose of erlotinib was limited to 100 mg/day since at the dose of 150 mg/day 11 of 23 patients required dose reductions because of toxicity. A later study by Boeck and coworkers showed the tolerability of GEM plus erlotinib given at a dose of 150 mg/day. 42 Treatment strategy: sequential application of drugs The German AIO PK0104 trial compared first-line treatment with GEM plus erlotinib to capecitabine plus erlotinib. 43 At disease progression, patients were scheduled to cross over to the respectively other chemotherapy, while treatment with erlotinib was stopped. This study indicated that both treatment strategies induced nearly identical survival times (6.6 months vs. 6.9 months). First-line treatment with GEM/erlotinib was, however, more effective that capecitabine/erlotinib when first time to treatment failure (TTF1) was analyzed (HR 0.69, P = ). Second-line treatment was given in 51% of eligible patients and also in this setting the superiority of GEM over capecitabine was shown with regard to time to treatment failure for the cross-over 2nd-line chemotherapy (HR 1.74, P = ). In view of nearly identical survival times observed in the AIO- and the PA.3 study, there was no signal that the higher erlotinib dose in the AIO study (150 mg/day vs. 100 mg/day) would relate to a relevant gain in treatment efficacy. 41,43 Gemcitabine plus cetuximab The clinical efficacy of extracellular EGFR inhibition was tested in a randomized study performed in patients with advanced PC. 44 When GEM plus cetuximab was compared to GEM, no significant difference was observed between the two arms of the study with regard to overall survival (6.3 vs. 5.9 months, HR = 1.06). Also progression-free survival (3.4 vs. 3.0 months, HR = 1.07) and overall response rates (12% vs. 14%) were similar in the GEM plus cetuximab- and the GEM arm, respectively. While median time to treatment failure was longer in the combination arm (2.3 vs. 1.8 months; HR = 1.21, P = 0.06), the median difference was only two weeks. EGFR expression was analyzed in 595 of 702 eligible patients, but had no impact on survival. According to the authors, this supports the findings also obtained in other tumors that indicate a lack of correlation between EGFR expression and outcome. 44 The findings by Philip et al. 44 compare well to another randomized phase II study where the addition of cetuximab to the combination of GEM plus cisplatin did not increase response or survival in advanced PC. 45 Inhibition of angiogenesis The combination of angiogenesis inhibitors such as bevacizumab or axitinib with GEM-based regimens has essentially failed to improve survival in advanced PC patients (Table 6). Again, it became clear that promising data obtained at the phase II level did not necessarily translate into a statistically and clinically relevant survival benefit when tested in phase III studies. 50,51 It has been argued that PC in most cases represents a hypovascularized tumor where inhibition of angiogenesis as a modulator of tumor growth is a priori ineffective. On the other hand, it cannot be excluded that GEM is an inappropriate combination partner for angiogenesis inhibitors and that a fluoropyrimidine-based chemotherapy backbone may lead to a different outcome. Last but not least, it may be hypothesized that the lack of predictive factors did not allow a proper selection of those patients who would have responded to angiogenesis inhibitors. This topic was addressed in a randomized study comparing GEM/erlotinib plus bevacizumab to GEM/erlotinib plus placebo. 46 The addition of bevacizumab had no significant effect on overall survival when the whole study population was analyzed (HR 0.89, P = 0.21). However, bevacizumab significantly prolonged survival in patients whose tumors were located in the tail of the pancreas (HR 0.54, P = ) or those who presented with elevated levels of baseline CRP (HR 0.65, P = ) or lactate dehydrogenase (HR 0.59, P = ). From this observation the authors concluded that perhaps patients with more aggressive disease, as indicated by elevated CRP or LDH, might benefit to a greater extent from bevacizumab than others. Targeting the stroma Pancreatic cancer is characterized by hypovascularity and desmoplastic stroma which both may contribute to impaired drug delivery and subsequent resistance to chemotherapy. 52 An innova- Table 6 Inhibition of angiogenesis. Reference Regimen Study design No of pts LAPC (%) ORR (%) PFS/TTP (mo) OS (mo) 1-Year survival (%) Van Cutsem et al. 46 GEM + erlotinib + Bev Phase III na GEM + erlotinib + placebo na Javle et al. 47 GEM + capecitabine + Bev Phase II Fogelman et al. 48 GEM + oxaliplatin + Bev Phase II Ko et al. 49 Cetuximab + Bev + GEM Phase II Cetuximab + Bev Spano et al. 50 GEM + axitinib Phase II GEM Kindler et al. 51 GEM + axitinib Phase III na GEM + placebo na GEM, gemcitabine; Bev, bevacizumab; LAPC, locally advanced pancreatic cancer.

7 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx 7 tive approach to deplete stromal tissue has been introduced by clinical application of nab-paclitaxel. 53 This albumin-bound drug formulation was initially developed to avoid the toxicity associated with the polyethylated castor oil, used as a solvent for the drug. 54 Preclinical evidence supports the assumption that intratumoral uptake of nab-paclitaxel is facilitated through binding of albumin to SPARC (secreted protein acidic and rich in cysteine). 55 This extracellular matrix glycoprotein is expressed in the peritumoral stroma and at the invasion front of tumors and is involved in cell migration, proliferation, angiogenesis and tissue remodeling. 56 The rationale to use nab-paclitaxel in PC is based on molecular analyses demonstrating overexpression of SPARC in pancreatic tumors. Previous work by Infante and coworkers had shown that the expression of SPARC in peritumoral fibroblasts was a negative prognostic factor in patients with resectable PC, while SPARC expression in tumor cells did not appear to correlate with survival. 56 The clinical efficacy of nab-paclitaxel in metastatic PC was recently investigated by von Hoff and coworkers in a phase I/II trial. At the maximal tolerated dose (MTD) (GEM 1000 mg/m 2 plus nabpaclitaxel 125 mg/m 2 once a week for three weeks every 28 days) a response rate of 48% was achieved which was accompanied by a PFS of 7.9 months, an OS of 12.2 months and a 1-year survival rate of 48%. 57 A phase III trial is presently ongoing to confirm these promising results. In human PC xenograft models treated with GEM plus nab-paclitaxel, depletion of desmoplastic stroma was associated by increased drug delivery leading to a 2.8-fold increase of intratumoral GEM concentrations. This treatment approach is in line with another stroma-directed strategy that aims to facilitate drug delivery by improved tumor perfusion. Paracrine hedgehog signaling from pancreatic tumor cells notably induces stromal cells to form desmoplastic tissue. Preclinical data suggest that inhibition of hedgehog signaling causes stromal depletion and subsequent stimulation of angiogenesis. As a result, increased vascular density and improved tumor perfusion may augment delivery and efficacy of chemotherapeutic agents. 58 Several clinical studies are presently ongoing to test this hypothesis in PC patients (Table 7). Second-line treatment An evaluation of 2nd-line therapy in randomised trials indicates that 16 57% of PC patients did receive salvage chemotherapy after failure of 1st-line GEM. 59 Median survival in GEM-resistant patients receiving best supportive care was 2.3 months in a small randomised trial. 60 Meanwhile, several clinical studies (mainly phase II) support the clinical efficacy of 2nd-line treatment with overall survival times of 3 9 months (calculated from the start of second-line therapy) and PFST/TTP durations of 2 5 months. 61 GEM-resistant patients in good performance status and with high motivation should therefore be offered 2nd-line regimens preferably consisting of fluoropyrimidines plus oxaliplatin. 59 Patients not tolerating 2nd-line combination therapy may receive fluoropyrimidines alone. Only limited data are available on how to treat patients after progression on FOLFIRINOX therapy. The rate of 2nd-line treatment in the phase-iii trial by Conroy and coworkers was 47% in the FOLFIRINOX-arm and 50% in the GEM arm. 23 The most common second-line regimens after FOLFIRINOX were GEM (82.5%) or GEM-based combinations (12.5%). Median survival from the introduction of second-line therapy was identical (4.4 months) and independent of first-line therapy with FOLFIRINOX or GEM. Based on this observation it may be speculated that neither intensity nor efficacy of first-line therapy are critically important for the implementation and outcome of second-line therapy. Prognostic and predictive factors Prognostic factors are patient- and tumor related factors that predict patient outcome (mostly survival) independent of treatment. By contrast, predictive factors predict response of the tumor to treatment (measured in terms of tumor size or survival). 62 Optimal management of metastatic PC includes the evaluation of these parameters for optimal guidance of therapy. The following analyses differentiate parameters determined before the start of treatment (baseline parameters) from those obtained during treatment (dynamic parameters). Baseline parameters of the patient Among baseline parameters, specifically performance status has gained some attention and has been discussed as one of the most prominent prognostic factors in advanced PC. 63 In fact, there is some evidence indicating that only patients with a good performance status (ECOG 0 1) derive a significant benefit from combination chemotherapy (HR = 0.76, p < ), while patients with a poor performance status do not (HR = 1.08, p = 0.40). 15 This conclusion is supported by most studies performing a separate analysis of good- and bad performance patients. It was, however, contradicted by a recent Italian study where the combination of GEM plus cisplatin was not superior to GEM alone independent of the performance status. 18 Further baseline parameters such as CA 19-9, LDH and CRP have been identified to correlate with survival, but are not ready to determine the choice of treatment. 45,64,65 In the FOLFIRINOX-trial Table 7 Anti-EGFR-induced rash: a predictor of overall survival. Studies Design of study Regimen No of pts Overall survival (months) Level of significance All pts No rash Rash-positive PA.3 study 41 Phase III GEM + erlotinib (rash grade p < P2) PanTar study 76 Phase II GEM + erlotinib (rash grade 10.2 (rash grade P = <2) P2) AVITA study 77 Phase III GEM + erlotinib + placebo (rash grade P2) HR 0.54, P < Cross-over AIO Phase III GEM + erlotinib vs (rash grade P1) HR 0.54, P < study 43 Capecitabine + erlotinib 6.9 Intergroup trial Phase III GEM + cetuximab (grade 0 1) 7.8 (grade P2) HR 0.64, P < S GEMOXCET trial 79 Phase II GEM + oxaliplatin + cetuximab (rash grade P1) HR 0.42, P = GEM, gemcitabine; PA.3 study, gemcitabine/erlotinib vs. gemcitabine; AVITA study, gemcitabine/erlotinib ± bevacizumab; Cross-over AIO study, gemcitabine/erlotinib vs. capecitabine/erlotinib; Intergroup trial S0205, gemcitabine plus cetuximab vs. gemcitabine.

8 8 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx by Conroy and coworkers, synchronous metastases, low baseline albumin level (<3.5 g per deciliter), hepatic metastasis, and age >65 years were reported as independent adverse prognostic factors for overall survival. 23 Molecular parameters of the tumor Specifically, when targeted therapies are employed, the question arises to which extent the expression or amplification of tumor-related targets can be used to predict outcome. So far, only few reports are available focussing on this topic. A first data set has been generated from the PA.3 study which compared GEM plus erlotinib to GEM plus placebo. 41 This analysis was compromised by a low retrieval rate of tumor probes allowing molecular analyses only in small subsets of patients. Among evaluable patients, EGFR amplification or high polysomy were observed in 47%, and KRAS mutations in 79% of tumors. Neither EGFR-IHC, nor EGFR-FISH or KRAS mutation status were significantly related to outcome. 41,66 The retrospective nature of this investigation, the low percentage of KRAS wild-type tumors combined with the low availability of tumor probes are relevant limitations to these analyses. Molecular parameters of the EGFR pathway were also analyzed in the AIO cross-over trial which compared GEM/erlotinib to capecitabine/erlotinib. Again, this study did not show a significant correlation between EGFR-IHC or EGFR-FISH and survival. KRAS mutations (all in codon 12) were observed in 70% of tumors. In univariate biomarker analyses, KRAS mutation status was significantly associated with overall survival favoring KRAS wild-type patients (HR 0.60 P = 0.005). 67 Since erlotinib was used in both study arms, the prognostic and predictive role of KRAS could not be differentiated. More evidence regarding this question comes from a recent publication by Kim and coworkers. 68 In a retrospective analysis of 136 patients, this group demonstrated that KRAS wildtype was associated with a survival benefit (9.7 vs. 5.2 months, P = 0.002) only in patients receiving GEM/erlotinib, while KRAS mutation status had no effect on outcome in patients treated without erlotinib (7.0 vs 7.0 months, P = 0.121). This observation supports the role of KRAS mutation as a predictor of response to erlotinib, but needs to be verified in a controlled prospective study. An important step into the molecular classification of PC has recently been described by Collisson and coworkers. 69 This group performed a molecular analysis of pancreatic tumors and identified three subgroups characterized by distinct gene signatures: (1) the classical type with a high expression of adhesion-associated and epithelial genes, (2) the quasi-mesenchymal type showing high expression of mesenchyme-associated genes, and (3) the exocrine-like subtype expressing tumor cell-derived digestive enzyme genes. These subgroups were not only different with regard to clinical outcome; first analyses performed in cell lines also suggest different response to treatment. 69 Given that these data are verified, a new door is opened for stratified treatment of PC patients according to the molecular signature of their tumors. Pharmacodynamic parameters GEM is a nucleoside analog which requires specific membrane transporter proteins for active uptake into the tumor cell. Among these, the human equilibrative nucleoside transporter 1 (hent1) and, to a lesser degree, the human concentrative nucleoside transporter 3 (hcnt3) have been identified as important determinants of sensitivity to GEM. 70 Accordingly, data obtained in vitro showed resistance to GEM in tumor cells lacking hent1 expression. These results were supported by clinical analyses indicating a significantly less favorable outcome in tumors without hent1 expression. 71 Evaluation of dynamic parameters during therapy Several parameters may help to guide treatment decisions during therapy. Among others, CA 19-9 kinetics, performance status, body weight and pain have been reported as useful markers to define treatment efficacy in combination with imaging. 14,72,73 While an elevated pretreatment level of CA 19-9 has been identified as an independent negative prognostic factor in several studies, the clinical relevance of CA 19-9 kinetics during therapy is more controversial. 65,74 Several authors reported prolonged survival in CA 19-9 responders defined e.g. by a P20% reduction of CA 19-9 baseline levels within the first 8 weeks of treatment. 74 This was, however, not confirmed by others. 75 Erlotinib-induced skin toxicity An important dynamic parameter evolving during therapy is erlotinib-induced skin toxicity. It typically develops during the initial 4 to 8 weeks of anti-egfr directed treatment (Table 8). In the PA.3 study, 36% of patients developed skin toxicity grade P2. These patients had a median survival which was 5 months longer than that of patients without skin toxicity (10.5 months vs. 5.3 months, p < 0.001). Also the rate of 1-year survival was nearly 3-fold greater in patients with skin toxicity grade P2 (43% vs. 16%, p < 0.001). 41 Patients with a good performance status (P = 0.03) and those younger than 65 years (P = 0.01) were more likely to develop skin rash in the PA.3 trial. A Spanish single-arm phase II study (n = 156) indicated that patients with skin toxicity grade 1 lived markedly longer than patients without any skin toxicity (6.8 months vs. 3.8 months, P < 0.001). Also this study could show that skin toxicity P2 was associated with prolonged survival when compared to skin toxicity grade <2 (10.2 months vs. 5.0 months, P = 0.001). 76 Since grading of skin toxicity is, to some extent, subjective and may also depend on Table 8 Ongoing phase II/III studies targeting the tumor stroma in pancreatic cancer patients (ClinicalTrials.gov). Clinical trial identifier NCT NCT NCT NCT Phase Clinical setting Regimen Number of patients Phase III Phase II Phase II Phase II Metastatic disease GEM ± nab-paclitaxel 842 OS Primary endpoint Metastatic disease GEM + nabpaclitaxel + GDC PFS Preoperative setting GDC Effect of treatment on stromal cell and tumor cell hedgehog signaling Recurrent or metastatic GEM ± GDC PFS disease Nab-paclitaxel, nano-particle albumin-bound paclitaxel; GDC-0049, oral small-molecule inhibitor of the hedgehog pathway; OS, overall survival; PFS, progression-free survival.

9 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx 9 Treatment of Metastatic Pancreatic Cancer Karnofsky Performance Status 60% Karnofsky Performance Status 70%-80% Karnofsky Performance Status 90%-100% 1 st -Line Treatment No chemotherapy GEM (+ erlotinib) GEM (+ erlotinib) FOLFIRINOX* 2 nd -Line Treatment 5-FU/FA / Cape OFF/ FOLFOX Gemcitabin GEM Fig. 1. Treatment of metastatic pancreatic cancer. GEM, gemcitabine; 5-FU/FA, 5-fluorouracil + folinic acid; Cape, capecitabine; OFF, oxaliplatin + 5-fluorouracil + folinic acid; FOLFOX, 5-fluorouracil, folinic acid, oxaliplatin; FOLFIRINOX, 5-fluorouracil + folinic acid + oxaliplatin + irinotecan. Only in patients who meet the inclusion criteria of the Conroy study: metastatic disease, age < 75 years, ECOG 0 1, bilirubin <1.5 UNL. topical or systemic treatment of skin eruptions, other groups have compared any skin toxicity grade 1 4 (occurring in about 70% of patients) to no skin toxicity. In the cross-over AIO study, patients with skin toxicity had a markedly better outcome than those without (8.7 months vs. 3.7 months, HR 0.54, P < 0.001). 43 Also in the AVITA trial, erlotinib-induced skin toxicity grade P2 related to an overall survival time of 8.3 months as opposed to 4.3 months without skin toxicity. 77 Comparable data were reported for the correlation of cetuximab-induced skin toxicity and survival. 78,79 In conclusion, skin toxicity is clearly the most powerful indicator of prolonged overall survival in patients receiving erlotinibbased therapy. However, at present time data are not sufficient to discriminate between the prognostic and predictive potential of this parameter. The available data also demonstrate that the survival times (4 5 months) of non-rash patients are lower than expected for an unselected patient population treated with single-agent GEM. 80 This observation entrails several considerations: first, non-rash patients represent a poor-prognosis subgroup; second, a negative interaction of GEM and erlotinib cannot be excluded and requires alternative regimens to be explored; third, due to the short survival of this poor prognosis subgroup, treatment decisions should be made early to allow an appropriate benefit. 80 Perspective for the design of future clinical trials According to a consensus report of the National Cancer Institute of the United States, harmonization of clinical trials with respect to patient populations is sought to reduce heterogeneity between studies. 81 Comparability between studies may markedly be improved given that the following recommendations are realized: 1) Patients with locally advanced and metastatic disease should be investigated in separate trials. 2) Patients with unfavorable ECOG performance status (ECOG PS P 2) should be studied in separate and appropriately designed clinical trials. 3) Additional patient characteristics such as weight loss or nutritional status should be given more consideration in the selection strategy; 4) Early withdrawal from the study without receiving meaningful treatment should be considered in separately performed subgroup analyses. 5) Harmonization of eligibility criteria should be performed across trials. 81 6) With the availability of 2nd- and 3rd-line treatment options post-study treatment should be documented and reported in studies investigating overall survival as a primary endpoint. 7) Future trials need to incorporate translational research as a major driving force of clinical research. 82 This requires not only the prospective collection of tumor- and blood samples, but also upfront introduction of translational endpoints into the trial design. A major task of the future consists in the identification and validation of prognostic and predictive biomarkers which may help to define clinically relevant subgroups of PC. How to treat advanced pancreatic cancer in daily clinical practice (Fig. 1)? At present time, only GEM and erlotinib are registered for treatment of advanced pancreatic cancer. According to the European registration, the use of erlotinib is limited to patients with metastatic PC. Given that erlotinib is not available in all European countries, single-agent GEM remains the standard reference treatment of advanced PC. A treatment algorithm may be based on a primary evaluation of the performance status: (1) We recommend not to apply chemotherapy or at least to question its use in patients with a Karnofsky performance status (KPS) 6 60%. (2) In patients with a KPS 70%-80%, GEM can be applied as a single agent or in combination with erlotinib. Treatment with erlotinib should be re-evaluated according to the development of rash after 4 8 weeks and should be stopped if no rash occurred. (3) Patients with a very good KPS of % and high motivation may be offered intensive polychemotherapy with the FOLFIRINOX regimen if the inclusion criteria defined in the Conroy study are met, otherwise they should receive treatment with GEM (plus erlotinib). At disease progression, all patients should be offered secondline therapy which can be performed with single agents or combination chemotherapy according to performance status, overlapping side effects from previous therapy and patient wish (Fig. 1). As a general measure we recommend early use of supportive and psycho-social care to control pain, to stabilize body weight and to improve the psycho-social condition of the patient. References 1. Hidalgo M. Pancreatic cancer. N Engl J Med 2010;362: Temel JS, Greer JA, Admane S, et al. Longitudinal perceptions of prognosis and goals of therapy in patients with metastatic non-small-cell lung cancer: results of a randomized study of early palliative care. J Clin Oncol 2011;29:

10 10 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx 3. Carlson LE, Groff SL, Maciejewski O, Bultz BD. Screening for distress in lung and breast cancer outpatients: a randomized controlled trial. J Clin Oncol 2010;28: Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with metastatic non-small-cell lung cancer. New Engl J Med 2010;363: Shah MM, Saif MW. Pancreatic Cancer and Thrombosis. JOP 2010;11: Mandala M, Reni M, Cascinu S, Barni S, Floriani I, Cereda S, et al. Venous thromboembolism predicts poor prognosis in irresectable pancreatic cancer patients. Ann Oncol 2007;18: Mandala M, Tondini C. The impact of thromboprophylaxis on cancer survival: focus on pancreatic cancer. Expert Rev Anticancer Ther 2011;11: Riess H, Pelzer U, Opitz B, et al. A prospective, randomized trial of simultaneous pancreatic cancer treatment with enoxaparin and chemotherapy: Final results of the CONKO-004 trial. J Clin Oncol 2010; 28 (15s): abstract Van Doormaal FF, Di Nisio M, Otten HM, Richel DJ, Prins M, Buller HR. Randomized trial of the effect of the low molecular weight heparin nadroparin on survival in patients with cancer. J Clin Oncol 2011;29: Lee AY, Levine MN, Baker RI, et al. Low-molecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003;349: Lee AY, Rickles FR, Julian JA, et al. Randomized comparison of low molecular weight heparin and coumarin derivatives on the survival of patients with cancer and venous thromboembolism. J Clin Oncol 2005;23: El-Rayes BF, Jasti P, Severson RK, Almhanna K, Philip PA, Shields A, et al. Impact of race, age, and socioeconomic status on participation in pancreatic cancer clinical trials. Pancreas 2010;39: Sultana A, Smith CT, Cunningham D, Starling N, Neoptolemos JP, Ghaneh P. Meta-analyses of chemotherapy for locally advanced and metastatic pancreatic cancer. J Clin Oncol 2007;25: Burris H, Moore M, Anderson J, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol 1997;15: Heinemann V, Boeck S, Hinke A, Labianca R, Louvet C. Meta-analysis of randomized trials: evaluation of benefit from gemcitabine-based combination chemotherapy applied in advanced pancreatic cancer. BMC Cancer 2008;8: Cunningham D, Chau I, Stocken DD, Valle JW, Smith D, Steward W, et al. Phase III randomized comparison of gemcitabine versus gemcitabine plus capecitabine in patients with advanced pancreatic cancer. J Clin Oncol 2009;27: Vaccaro V, Sperduti I, Milella M. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med 2011;365:768-9; author reply Colucci G, Labianca R, Di Costanzo F, Gebbia V, Cartenì G, Massidda B, et al. Randomized phase III trial of gemcitabine plus cisplatin compared with singleagent gemcitabine as first-line treatment of patients with advanced pancreatic cancer: the GIP-1 Study. J Clin Oncol 2010;28: Reni M, Cordio S, Milandri C, Passoni P, Bonetto E, Oliani C, et al. Gemcitabine versus cisplatin, epirubicin, fluorouracil, and gemcitabine in advanced pancreatic cancer: a randomised controlled multicentre phase III trial. Lancet Oncol 2005;6: Wagner AD, Buechner-Steudel P, Wein A, Schmalenberg H, Lindig U, Moehler M, et al. Gemcitabine, oxaliplatin and weekly high-dose 5-FU as 24-h infusion in chemonaive patients with advanced or metastatic pancreatic adenocarcinoma: a multicenter phase II trial of the Arbeitsgemeinschaft Internistische Onkologie (AIO). Ann Oncol 2007;18: Endlicher E, Troppmann M, Kullmann A, Golder S, Herold T, Herfarth H, et al. Irinotecan plus gemcitabine and 5-fluorouracil in advanced pancreatic cancer: a phase II study. Oncology 2007;72: Hess V, Pratsch S, Potthast S, Lee L, Winterhalder R, Widmer L, et al. Combining gemcitabine, oxaliplatin and capecitabine (GEMOXEL) for patients with advanced pancreatic carcinoma (APC): a phase I/II trial. Ann Oncol 2010;21: Conroy T, Desseigne F, Ychou M, Bouche O, Guimbaud R, Becouarn Y, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med 2011;364: Ducreux M, Rougier P, Pignon JP, Douillard JY, Seitz JF, Bugat R, et al. A randomised trial comparing 5-FU with 5-FU plus cisplatin in advanced pancreatic carcinoma. Ann Oncol 2002;13: Maisey N, Chau I, Cunningham D, Norman A, Seymour M, Hickish T, et al. Multicenter randomized phase III trial comparing protracted venous infusion (PVI) fluorouracil (5-FU) with PVI 5-FU plus mitomycin in inoperable pancreatic cancer. J Clin Oncol 2002;20: Ducreux M, Mitry E, Ould-Kaci M, Boige V, Seitz JF, Bugat R, et al. Randomized phase II study evaluating oxaliplatin alone, oxaliplatin combined with infusional 5-FU, and infusional 5-FU alone in advanced pancreatic carcinoma patients. Ann Oncol 2004;15: Ghosn M, Farhat F, Kattan J, Younes F, Moukadem W, Nasr F, et al. FOLFOX-6 combination as the first-line treatment of locally advanced and/or metastatic pancreatic cancer. Am J Clin Oncol 2007;30: Boeck S, Hoehler T, Seipelt G, Mahlberg R, Wein A, Hochhaus A, et al. Capecitabine plus oxaliplatin (CapOx) versus capecitabine plus gemcitabine (CapGem) versus gemcitabine plus oxaliplatin (mgemox): final results of a multicenter randomized phase II trial in advanced pancreatic cancer. Ann Oncol 2008;19: Gebbia V, Maiello E, Giuliani F, Borsellino N, Caruso M, Di Maggio G, et al. Second-line chemotherapy in advanced pancreatic carcinoma: a multicenter survey of the Gruppo Oncologico Italia Meridionale on the activity and safety of the FOLFOX4 regimen in clinical practice. Ann Oncol 2007;18(Suppl 6):vi Pelzer U, Kubica K, Stieler J, Schwaner I, Heil G, Görner M, et al. A randomized trial in patients with gemcitabine refractory pancreatic cancer. Final results of the CONKO 003 study. J Clin Oncol 2008;26(Suppl): abstract Xiong HQ, Varadhachary GR, Blais JC, Hess KR, Abbruzzese JL, Wolff RA. Phase 2 trial of oxaliplatin plus capecitabine (XELOX) as second-line therapy for patients with advanced pancreatic cancer. Cancer 2008;113: Rocha Lima CM, Green MR, Rotche R, Miller Jr WH, Jeffrey GM, Cisar LA, et al. Irinotecan plus gemcitabine results in no survival advantage compared with gemcitabine monotherapy in patients with locally advanced or metastatic pancreatic cancer despite increased tumor response rate. J Clin Oncol 2004;22: Stathopoulos GP, Syrigos K, Aravantinos G, Polyzos A, Papakotoulas P, Fountzilas G, et al. A multicenter phase III trial comparing irinotecangemcitabine (IG) with gemcitabine (G) monotherapy as first-line treatment in patients with locally advanced or metastatic pancreatic cancer. Br J Cancer 2006;95: Taieb J, Lecomte T, Aparicio T, Asnacios A, Mansourbakht T, Artru P, et al. FOLFIRI.3, a new regimen combining 5-fluorouracil, folinic acid and irinotecan, for advanced pancreatic cancer: results of an Association des Gastro- Enterologues Oncologues (Gastroenterologist Oncologist Association) multicenter phase II study. Ann Oncol 2007;18: Yi SY, Park YS, Kim HS, Jun HJ, Kim KH, Chang MH, et al. Irinotecan monotherapy as second-line treatment in advanced pancreatic cancer. Cancer Chemother Pharmacol 2009;63: Gebbia V, Maiello E, Giuliani F, Borsellino N, Arcara C, Colucci G. Irinotecan plus bolus/infusional 5-Fluorouracil and leucovorin in patients with pretreated advanced pancreatic carcinoma: a multicenter experience of the Gruppo Oncologico Italia Meridionale. Am J Clin Oncol 2010;33: Cereda S, Reni M, Rognone A, Ghidini M, Belli C, Longoni S, et al. XELIRI or FOLFIRI as salvage therapy in advanced pancreatic cancer. Anticancer Res 2010;30: Yoo C, Hwang JY, Kim JE, Kim TW, Lee JS, Park DH, et al. A randomised phase II study of modified FOLFIRI.3 vs modified FOLFOX as second-line therapy in patients with gemcitabine-refractory advanced pancreatic cancer. Br J Cancer 2009;101: Conroy T, Paillot B, Francois E, Bugat R, Jacob JH, Stein U, et al. Irinotecan plus oxaliplatin and leucovorin-modulated fluorouracil in advanced pancreatic cancer a Groupe Tumeurs Digestives of the Federation Nationale des Centres de Lutte Contre le Cancer study. J Clin Oncol 2005;23: Louvet C, Labianca R, Hammel P, Lledo G, Zampino MG, Andre T, et al. Gemcitabine in combination with oxaliplatin compared with gemcitabine alone in locally advanced or metastatic pancreatic cancer: results of a GERCOR and GISCAD phase III trial. J Clin Oncol 2005;23: Moore MJ, Goldstein D, Hamm J, Figer A, Hecht JR, Gallinger S, et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007;25: Boeck S, Vehling-Kaiser U, Waldschmidt D, Kettner E, Märten A, Winkelmann C, et al. Erlotinib 150 mg daily plus chemotherapy in advanced pancreatic cancer: an interim safety analysis of a multicenter, randomized, cross-over phase III trial of the Arbeitsgemeinschaft Internistische Onkologie. Anticancer Drugs 2010;21: Boeck SH, Vehling-Kaiser U, Waldschmidt D, et al. Gemcitabine plus erlotinib (GE) followed by capecitabine (C) versus capecitabine plus erlotinib (CE) followed by gemcitabine (G) in advanced pancreatic cancer (APC): A randomized, cross-over phase III trial of the Arbeitsgemeinschaft Internistische Onkologie (AIO). J Clin Oncol 2010;28 (18s) LBA Philip PA, Benedetti J, Corless CL, Wong R, O Reilly EM, Flynn PJ, et al. Phase III study comparing gemcitabine plus cetuximab versus gemcitabine in patients with advanced pancreatic adenocarcinoma: Southwest Oncology Groupdirected intergroup trial S0205. J Clin Oncol 2010;28: Cascinu S, Berardi R, Labianca R, Siena S, Falcone A, Aitini E, et al. Cetuximab plus gemcitabine and cisplatin compared with gemcitabine and cisplatin alone in patients with advanced pancreatic cancer: a randomized, multicentre, phase II trial. Lancet Oncol 2008;9: Van Cutsem E, Vervenne WL, Bennouna J, Humblet Y, Gill S, Van Laethem JL, et al. Phase III trial of bevacizumab in combination with gemcitabine and erlotinib in patients with metastatic pancreatic cancer. J Clin Oncol 2009;27: Javle M, Yu J, Garrett C, Pande A, Kuvshinoff B, Litwin A, et al. Bevacizumab combined with gemcitabine and capecitabine for advanced pancreatic cancer: a phase II study. Br J Cancer 2009;100: Fogelman D, Jafari M, Varadhachary GR, Xiong H, Bullock S, Ozer H, et al. Bevacizumab plus gemcitabine and oxaliplatin as first-line therapy for metastatic or locally advanced pancreatic cancer: a phase II trial. Cancer Chemother Pharmacol 2011;68: Ko AH, Youssoufian H, Gurtler J, Dicke K, Kayaleh O, Lenz HJ, et al. A phase II randomized study of cetuximab and bevacizumab alone or in combination with gemcitabine as first-line therapy for metastatic pancreatic adenocarcinoma. Invest New Drugs 2011; ahead of print. 50. Spano JP, Chodkiewicz C, Maurel J, Wong R, Wasan H, Barone C, et al. Efficacy of gemcitabine plus axitinib compared with gemcitabine alone in patients with advanced pancreatic cancer: an open-label randomised phase II study. Lancet 2008;371:

11 V. Heinemann et al. / Cancer Treatment Reviews xxx (2011) xxx xxx Kindler HL, Ioka T, Richel DJ, Bennouna J, Letourneau R, Okusaka T, et al. Axitinib plus gemcitabine versus placebo plus gemcitabine in patients with advanced pancreatic adenocarcinoma: a double-blind randomised phase 3 study. Lancet Oncol 2011;12: Neesse A, Michl P, Frese KK, Feig C, Cook N, Jacobetz MA, et al. Stromal biology and therapy in pancreatic cancer. Gut 2011;60: Hawkins MJ, Soon-Shiong P, Desai N. Protein nanoparticles as drug carriers in clinical medicine. Advanced Drug Delivery Rev 2008;60: Gradishar WJ, Tjulandin S, Davidson N, Shaw H, Desai N, Bhar P, et al. Phase III trial of nanoparticle albumin-bound paclitaxel compared with polyethylated castor oil-based paclitaxel in women with breast cancer. J Clin Oncol 2005;23: Edmonds C, Cengel KA. Tumor-stroma interactions in pancreatic cancer. Will this SPARC prove a raging fire? Cancer Biol Ther 2008;7: Infante JR, Matsubayashi H, Sato N, Tonascia J, Klein AP, Riall TA, et al. Peritumoral fibroblast SPARC expression and patient outcome with resectable pancreatic adenocarcinoma. J Clin Oncol 2007;25: Von Hoff DD, Ramanathan RK, Borad MJ, Laheru DA, Smith LS, Wood TE. Et l. Gemcitabine plus nab-paclitaxel is an active regimen in patients with advanced pancreatic cancer: A phase I/II trial. J Clin Oncol 2011;29: Olive KP, Jacobetz MA, Davidson CJ, Gopinathan A, McIntyre D, Honess D, et al. Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science 2009;324: Boeck S, Heinemann V. Second-line therapy in gemcitabine-pretreated patients with advanced pancreatic cancer. J Clin Oncol 2008;26: Pelzer U, Schwaner I, Stieler J, Adler M, Seraphin J, Dorken B, et al. Best supportive care (BSC) versus oxaliplatin, folinic acid and 5-fluorouracil (OFF) plus BSC in patients for second-line advanced pancreatic cancer: a phase IIIstudy from the German CONKO-study group. Eur J Cancer 2011;47: Custodio A, Puente J, Sastre J, Diaz-Rubio E. Second-line therapy for advanced pancreatic cancer: a review of the literature and future directions. Cancer Treat Rev 2009;35: Shepherd FA, Tsao M-S. Unraveling the mystery of prognostic and predictive factors in epidermal grwoth factor receptor therapy. J Clin Oncol 2006;24: Boeck S, Hinke A, Wilkowski R, Heinemann V. Importance of performance status for treatment outcome in advanced pancreatic cancer. World J Gastroenterol 2007;13: Wong D, Ko AH, Hwang J, et al. Serum CA19-9 decline compared to radiographic response as a surrogate for clinical outcomes in patients with metastatic pancreatic cancer receiving chemotherapy. Pancreas 2008;37: Wasan HS, Springett GM, Chodkiewicz C, Wong R, Maurel J, Barone C, et al. CA 19-9 as a biomarker in advanced pancreatic cancer patients randomised to gemcitabine plus axitinib or gemcitabine alone. Br J Cancer 2009;101: Da Cunha Santos G, Dhani N, Tu D, Chin K, Ludkovski O, Kamel-Reid S, et al. Molecular predictors of outcome in a phase 3 study of gemcitabine and erlotinib therapy in patients with advanced pancreatic cancer: National Cancer Institute of Canada Clinical Trials Group Study PA.3. Cancer 2010;116: Boeck S, Jung A, Laubender RP, Neumann J, Egg R, Goritschan C, et al. Molecular markers of the EGFR pathway in erlotinib-treated patients with advanced pancreatic cancer (APC): Translational analyses of a randomized, cross-over AIO phase III trial. J Clin Oncol 2011;29: abstract Kim ST, Lim do DH, Jang KT, Lim T, Lee J, Choi YL, et al. Impact of KRAS mutations on Clinical Outcomes in Pancreatic Cancer Patients Treated with First-line Gemcitabine-based Chemotherapy. Mol Cancer Ther 2011;10: Collisson EA, Sadanandam A, Olson P, Gibb WJ, Truitt M, Gu S, et al. Subtypes of pancreatic ductal adenocarcinoma and their differing responses to therapy. Nat Med 2011;17: Marechal R, Mackey JR, Lai R, Demetter P, Peeters M, Polus M, et al. Human equilibrative nucleoside transporter 1 and human concentrative nucleoside transporter 3 predict survival after adjuvant gemcitabine therapy in resected pancreatic adenocarcinoma. Clin Cancer Res 2009;15: Farrell JJ, Elsaleh H, Garcia M, Lai R, Ammar A, Regine WF, et al. Human equilibrative nucleoside transporter 1 levels predict response to gemcitabine in patients with pancreatic cancer. Gastroenterology 2009;136: Boeck S, Haas M, Laubender RP, Kullmann F, Klose C, Bruns CJ, et al. Application of a time-varying covariate model to the analysis of CA 19-9 as serum biomarker in patients with advanced pancreatic cancer. Clin Cancer Res 2010;16: Moinpour CM, Vaught NL, Goldman B, Redman MW, Philip PA, Millwood B, et al. Pain and emotional well-being outcomes in Southwest Oncology Groupdirected intergroup trial S0205: a phase III study comparing gemcitabine plus cetuximab versus gemcitabine as first-line therapy in patients with advanced pancreas cancer. J Clin Oncol 2010;28: Boeck S, Stieber P, Holdenrieder S, Wilkowski R, Heinemann V. Prognostic and therapeutic significance of carbohydrate antigen 19-9 as tumor marker in patients with pancreatic cancer. Oncology 2006;70: Hess V, Glimelius B, Grawe P, Dietrich D, Bodoky G, Ruhstaller T, et al. CA 19-9 tumour-marker response to chemotherapy in patients with advanced pancreatic cancer enrolled in a randomised controlled trial. Lancet Oncol 2008;9: Manzano J. Rivera F, Galan M, Valladares M, Pericay C, Mendez MJ, et al. A phase II, open label study to evaluate the relationship between skin rash and survival in patients with unresectable and/or metastatic pancreatic cancer treated with erlotinib combined with gemcitabine. J Clin Oncol 2010;28(15s): abstract Verslype C, Verwvenne W, Bennouna J, Humblet Y, Cosaert J, van Cutsem E. Rash as a marker for the efficacy of gemcitabine plus erlotinib-based therapy in pancreatic cancer: Results from the AViTA Study. J Clin Oncol 2009;27(15s): abstract Philip P. Reply to S. Boeck et al. J Clin Oncol 2011;29:e Kullmann F, Hartmann A, Stöhr R, Messmann H, Dollinger MM, Trojan J, et al. KRAS mutation in metastatic pancreatic ductal adenocarcinoma: results of a multicenter phase II study evaluating efficacy of cetuximab plus gemcitabine/ oxaliplatin (GEMOXCET) in first-line therapy. Oncology 2011;81: Boeck S, Heinemann V. Anti-epidermal growth factor receptor treatment strategies in advanced pancreatic cancer: success or failure? J Clin Oncol 2011;29:e70 1; author reply e Philip PA, Mooney M, Jaffe D, Eckhardt G, Moore M, Meropol N, et al. Consensus report of the national cancer institute clinical trials planning meeting on pancreas cancer treatment. J Clin Oncol 2009;27: Laethem J-L, Verslype C, Iovanna JL, Michl P, Conroy T, Louvet C, et al. New strategies and designs in pancreatic cancer research: consensus guidelines report from a European expert panel. Ann Oncol 2011; ahead of print.