European Journal of Cancer (2014) 50, 1422–1429
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
The combination of a chemotherapy doublet (gemcitabine and capecitabine) with a biological doublet (bevacizumab and erlotinib) in patients with advanced pancreatic adenocarcinoma. The results of a phase I/II study q D.J. Watkins a, N. Starling a, D. Cunningham a,⇑, J. Thomas a, J. Webb a, G. Brown b, Y. Barbachano c, J. Oates a, I. Chau a a
Department of Medicine, Royal Marsden Hospital NHS Foundation Trust, London and Surrey, United Kingdom Department of Diagnostic Imaging, Royal Marsden Hospital NHS Foundation Trust, London and Surrey, United Kingdom c Computing and Statistics, Royal Marsden Hospital NHS Foundation Trust, London and Surrey, United Kingdom b
Available online 6 March 2014
KEYWORDS Anti-angiogenic therapy Chemotherapy Combination chemotherapy Epidermal growth factor receptor (EGFR) inhibition Molecularly targeted therapy Pancreatic cancer
q
Abstract Background: Preclinical data support the combined inhibition of vascular endothelial growth factor (VEGF) and epidermal growth factor receptor (EGFR) pathways in the treatment of pancreatic cancer. Following a dose finding phase I study the efficacy and toxicity of a four-drug regimen utilising the cytotoxic doublet of gemcitabine and capecitabine (GemCap), with the biological doublet of erlotinib and bevacizumab were further assessed in patients with advanced pancreatic cancer. Patients and methods: In a phase II expansion cohort, patients with chemonaive locally advanced or metastatic pancreatic cancer received gemcitabine (1000 mg/m2 D1, 8, 15), capecitabine (1400 mg/m2 D1–21), erlotinib (100 mg daily) and bevacizumab (5 mg/kg D1, 15) every 28 days. The primary endpoint was radiological response rate by response evaluation criteria in solid tumours (RECIST). Computed tomography (CT) assessment was performed every 8 weeks. Consolidation radiotherapy was considered in locally advanced patients following six cycles of treatment. Results: In total 44 patients (phases I & II) were recruited. The median cycles delivered were 6 (range 1–16). Confirmed radiological responses were seen in 23% (95% confidence interval (CI): 11–38%) of patients. The median progression-free and overall survival for the entire
Provisional analysis presented at ASCO 2010: Poster Discussion Abstract No: 4036. Citation: J Clin Oncol 28:15s, 2010 (suppl.; abstr. 4036).
⇑ Corresponding author at: Department of Medicine, Royal Marsden Hospital, Downs Road, Sutton, Surrey SM2 5PT, United Kingdom. Tel.:
+44 (0)208 661 3156; fax: +44 (0)208 643 9414. E-mail address: [email protected] (D. Cunningham). http://dx.doi.org/10.1016/j.ejca.2014.02.003 0959-8049/Ó 2014 Elsevier Ltd. All rights reserved.
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
1423
cohort was 8.4 and 12.6 months, respectively. In patients with metastatic disease the median overall survival was 10.1 months. Common grade 3/4 toxicities were; neutropenia 52%, lethargy 32%, diarrhoea 18% and hand-foot syndrome 18%. Conclusion: The combination of gemcitabine, capecitabine, erlotinib and bevacizumab was feasible with a manageable toxicity profile and demonstrated encouraging efficacy data in a good performance status population. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Pancreatic cancer is amongst the most intractable of all solid tumours with little improvement in the observed 5 year survival rate over recent decades [1]. In 2008 there were approximately 278,684 new cases of pancreatic cancer diagnosed worldwide with 266,669 deaths related to the disease [2], highlighting the dismal prognosis. Single agent gemcitabine has been a standard of care in the management of advanced pancreatic cancer since 1997, when it demonstrated a modest survival advantage (5.65 versus 4.41 months) and greater clinical benefit as compared to 5-fluorouracil [3]. Subsequent phase III studies and meta-analysis have demonstrated that additional modest gains in treatment efficacy can be achieved through the use of doublet regimens pairing gemcitabine with either capecitabine, erlotinib or platinum [4–6]. More recently the results achieved with the combination of oxaliplatin, irinotecan, leucovorin and fluorouracil (FOLFIRINOX) have established that significant survival gains are attainable with the use of a multi-drug regimen [7]. Compared to gemcitabine monotherapy FOLFIRINOX demonstrated superior response rate (31.6% versus 9.4%), progression-free survival (median 6.4 months versus 3.3 months) and overall survival (median 11.1 months versus 6.8 months) in a good performance status patient group [7]. In contrast to the advances attained with conventional cytotoxic therapies [7,8], the phase III results for molecularly targeted agents in pancreatic cancer have been disappointing [9–11], with erlotinib remaining the only licenced targeted agent in this disease. The dysregulation of multiple growth promoting signalling pathways is frequently observed in malignant disease and the existence of pathway cross-talk and redundancy has been seen to provide a mechanism of escape from the inhibitory effects of molecularly targeted therapies. Cross-talk between epidermal growth factor receptor (EGFR) and vascular endothelial growth factor (VEGF) pathways has been identified in pre-clinical models, with evidence for the involvement of EGFR signalling in the promotion of angiogenesis and the demonstration of the anti-angiogenic properties of EGFR inhibitors [12]. Furthermore the upregulation of VEGF has been implicated as a poten-
tial mediator of both innate and acquired resistance to EGFR inhibition [13,14]. These pre-clinical data provided the rationale for the evaluation of dual EGFR and VEGF blockade as part of a multi-drug regimen in the treatment of advanced pancreatic cancer. Following on from a phase I dose finding study [15], a phase II expansion aimed to evaluate the efficacy and tolerability of the four drug regimen containing the cytotoxic doublet of gemcitabine and capecitabine, in combination with the biological doublet of erlotinib and bevacizumab in patients with advanced pancreatic cancer. 2. Patients and methods 2.1. Patient selection Eligible patients were 18 years or older with histologically or cytologically proven locally advanced or metastatic pancreatic ductal adenocarcinoma or undifferentiated carcinoma. Patients were required to have adequate renal, bone marrow and liver function (creatinine clearance P 50 mL/min, platelets > 100 109/L, leucocytes > 3.0 109/L, neutrophil count > 1.5 109/ L and bilirubin < 2 the upper limit of normal), unidimensional measurable disease and have a WHO (World Health Organisation) performance status of 62. No prior chemotherapy or radiotherapy was allowed. Other key exclusion criteria included clinically significant (i.e. active) cardiovascular disease, ongoing treatment with aspirin (>325 mg/day) or other medications known to predispose to gastrointestinal ulceration, central nervous system metastases, uncontrolled hypertension, major surgery/traumatic injury within 28 days, non-healing wound/fracture, major bleeding diathesis or full-dose anticoagulation and uncontrolled comorbid conditions. When the study initially commenced patients with duodenal invasion were excluded, however these patients were later included following a protocol amendment. The study was approved by the local research ethics committee (CCR No. 2631) and conducted in compliance with International Committee on Harmonization Good Clinical Practice Guidelines. Written informed consent was obtained from all patients (ClinicalTrials.gov Identifier: NCT00260364).
1424
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
2.2. Trial design and treatment This was an open-label, single-centre study evaluating the efficacy and tolerability of the gemcitabine, capecitabine, erlotinib and bevacizumab combination. An initial dose escalation study had been undertaken to identify the optimal dose of capecitabine to use in the phase II expansion. The design and results of the dose escalation study have been reported previously [15]. In the phase II expansion all patients received gemcitabine 1000 mg/m2 on days 1, 8 and 15 of a 28-day cycle, together with bevacizumab 5 mg/kg intravenously on days 1 and 15, oral capecitabine 1400 mg/m2 days 1–21 and erlotinib 100 mg orally throughout days 1 through 28. The dose of capecitabine used in this study equates to a 15% reduction from the reported gemcitabine and capecitabine (GemCap) doublet regimen [4]. Responding patients could continue to receive study therapy for six cycles, or longer in patients deriving ongoing clinical benefit. 2.3. Dose modifications Gemcitabine was dosed according to the haematological function on the day of treatment. If the neutrophil count was 0.5–1.0 109/L or the platelet count was 50– 100 109/L, gemcitabine was dose reduced by 25%. Gemcitabine was omitted for a week for a neutrophil count of less than 0.5 109/L or platelets less than 50 109/L. In the event of neutropenic sepsis or recurrent low neutrophil/platelet counts a maintained 25% dose reduction of gemcitabine was instituted. The use of granulocyte colony stimulating factor (GCSF) was at the discretion of the investigator. For capecitabine related non-haematological toxicity capecitabine therapy was interrupted until toxicity had resolved to 6grade 1 and then resumed at 75% of the intended dose, and on second occurrence, capecitabine was resumed at 50% of the intended dose. For gemcitabine related non-haematological toxicities Pgrade 2 the dose of gemcitabine was reduced by 25%, reduction/omission was considered for grade 3 non-haematological toxicity. Erlotinib dosing was interrupted for related grade 3 non-haematological toxicities and reduced to 50 mg once a day on resolution of toxicities to 6grade 1, or discontinued for grade 4 toxicities. There were no dose reductions for bevacizumab, standard guidelines were followed for proteinuria. Bevacizumab was discontinued for GI perforation, grade 3 or 4 haemorrhage, grade 4 hypertension, uncontrolled grade 3 hypertension, arterial thromboembolism or symptomatic grade 4 venous thromboembolism. In the event of a treatment delay of greater than 3 weeks study treatment was discontinued. 2.4. Patient evaluation At every treatment visit, toxicity and a standard laboratory panel were assessed. Adverse events were graded
according to National Cancer Institute Common Terminology Criteria for Adverse Events (version 3.0). Radiological response was evaluated by CT imaging of the chest/abdomen/pelvis at baseline (within 4 weeks of starting protocol therapy) and every 8 weeks thereafter. Responses were confirmed at least 4 weeks after a responding scan. Radiological response was assessed by central review and reported according to response evaluation criteria in solid tumours (RECIST) guidelines [16]. Serum CA 19-9 marker response was assessed at four weekly intervals during treatment. Consolidation radiotherapy was permissible in patients with locally advanced disease following six cycles of treatment. Following discontinuation of treatment, patients were seen every three months until death. 2.5. Statistical plan The primary objective was to assess the overall radiological response rate of the four drug regimen in patients with previously untreated locally advanced/metastatic pancreatic cancer. Secondary objectives included an assessment of toxicity, disease control rate, progression free survival (PFS) and overall survival (OS). The sample size was determined using an optimal Simon 2 stage design. Assuming a minimum worthwhile disease response rate of 8% (p0) and an expected disease response rate of 25% (p1), 44 patients were required to achieve 90% power with a one sided alpha of 0.05. The first stage of accrual required at least two out of 16 enrolled patients to achieve a response to allow continuation to the second stage. In the second stage at least seven out of 44 patients were required to achieve a response in order to demonstrate that the lower limit of the 90% confidence interval for response rate excludes 8% and that the treatment is acceptable. Patients recruited during the dose escalation phase of the study were incorporated into the Simon 2 stage design for efficacy. 3. Results 3.1. Patient characteristics Between December 2005 and March 2009 a total of 44 patients were enrolled in both the dose finding cohort (20 patients) and phase II expansion. During the dose finding portion of the study patients were treated at one of four dose levels as previously reported [15]. In the phase II expansion all patients received treatment at the recommended dose level (dose level 3) as detailed previously. Patient characteristics are summarised in Table 1. Of the 44 enrolled patients 30 received treatment at the recommended dose level as defined during the dose finding portion of the study. The median age was 63 years and
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
98% of patients were performance status 0 or 1. Fifty five percent of patients had metastatic disease.
1425
Enrolled patients n=44
3.2. Treatment delivered The median number of cycles delivered was 6 (range 1–16). Nine patients (20%) received less than four cycles and 28 patients (64%) completed P 6 cycles of study therapy including 16 of 24 patients (67%) with metastatic disease. The treatment delivered and reasons for early discontinuation of study therapy are summarised in Fig. 1. Fatigue was the most common adverse event resulting in the discontinuation of study treatment. Other adverse events resulting in early discontinuation included; abdominal pain, bowel obstruction and infection. For the 30 patients treated at the recommended dose level, dose reductions were required for capecitabine or erlotinib in 15 patients (50%) and 11 patients (37%), respectively. The mean relative dose intensity (accounting for delays and dose reductions) for each drug was; capecitabine 79%, gemcitabine 85%, erlotinib 79%,and bevacizumab 92%. Of the 20 patients with locally advanced disease, 12 (60%) received consolidation radiotherapy and two underwent pancreatectomy with curative intent. 16 of all enrolled patients (36%) received post-study chemotherapy, nine of 24 patients (37.5%) in the metastatic disease group. 3.3. Tumour response Forty-two patients were evaluable for radiological response. The overall response rate was 23% (95% confidence interval (CI): 11–38%) with confirmed partial responses observed in ten patients. No complete responses were seen (Table 2). Sixteen patients with locally advanced disease and 19 patients with metastatic disease had elevated CA19-9 levels at study entry. The trends in CA19-9 level during treatment for each of these patients are shown in suppleTable 1 Patient characteristics and proportion of patients treated at each dose level. Characteristics
Total n = 44 (%)
Median age (range) Male:Female ECOG PS 0:1:2 Locally advanced:metastatic Head of the pancreas Body/tail of the pancreas
63 (33–80) 27:17 4:39:1 20:24 27 (61) 17 (39)
Dose level received 1 2 3 (recommended phase II dose level) 4
8 (18) 3 (7) 30 (68) 3 (7)
5 Disease progression 4 Fatigue 5 Other adverse event 2 *Other
Discontinued prior to cycle 6 n=16
Completed ≥6 cycles n=28
16 (67%) Metastatic 12 (60%) Locally advanced
Fig. 1. Treatment delivered and reasons for early discontinuation. *One patient with locally advanced disease received consolidation radiotherapy after five cycles of treatment. A further patient withdrew from the study following one cycle of treatment due to a change in personal circumstances.
mentary data A1. An exploratory analysis was undertaken examining the proportion of patients achieving a reduction in baseline CA19-9 and the duration of this response. Twenty-five of the 35 patients (71%) with elevated baseline CA19-9 levels achieved a 50% reduction lasting P 8 weeks (Table 3). Twelve patients (four locally advanced and eight metastatic) achieved a >90% reduction from baseline CA19-9 levels. 3.4. Survival The dataset was locked and analysed in August 2011. At time of analysis three patients were known to remain Table 2 Radiological response rate. Response evaluation criteria in solid tumours (RECIST) response (confirmed)
Total (%)
Radiological response rate Complete response Partial response Stable disease Disease control rate Progressive disease Non evaluable
10 (23) (95% CI:11–38%) 0 10 (23) 28 (64) 38 (86) 4 (9) 2 (5)
n = 44
Table 3 Proportion of patients achieving a 50% reduction in CA19-9 and duration of CA-19-9 response. Patients with raised baseline CA19-9
Locally advanced n = 16 (%)
Metastatic n = 19 (%)
50% reduction CA19-9 for P8 weeks 50% reduction CA19-9 for P16 weeks Median duration of CA19-9 response
12 (75)
13 (68)
5 (31)
12 (63)
14 weeks (range 10–24)
20 weeks (range 13–48)
1426
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
alive, two of whom were progression-free. One patient is lost to follow-up. For all enrolled patients the median overall survival was 12.6 months with a 1 year survival rate of 61% (95% CI: 46–76%). The corresponding median progression-free survival was 8.4 months. The survival outcomes for the 30 patients treated at the recommended phase II dose level were comparable to the whole study population with a median and 1 year overall survival of 12.6 months and 63% (95% CI: 44–80%), respectively. The median progression-free survival was 7.9 months in this subset. Kaplan-Meier survival curves are shown in Fig. 2 and survival data summarised in Table 4. The severity of skin rash was not found to be predictive of treatment outcome with a non-statistically significant trend towards longer PFS and OS in patients with Pgrade 2 acneiform skin rash as compared to those with grade 0 or I skin reaction (Table 5). 3.5. Toxicity Adverse events for all patients and those treated at the recommended phase II dose level are summarised
in Table 6. The most frequently observed non-haematological grade 3/4 adverse event was lethargy which occurred in 32% of patients. Other frequent (>15%) grade 3/4 non-haematological toxicities were diarrhoea, hand-foot syndrome, stomatitis and acneiform rash. Although the rate of neutropenia was high, with grade 3/4 neutropenia observed in 52% of patients, no febrile neutropenia was observed. Granulocyte colony stimulating factor (GCSF) was only used in two of 30 patients (6.7%) treated at the recommended dose level.
Table 4 Progression-free survival and overall survival for all patients and in the locally advanced and metastatic subgroups.
Median progression free survival (months) Median overall survival (months) 1 year overall survival % (95% confidence interval)
Progression -free Survival
Metastatic
8.4
10.1
7.6
12.6
13.9
10.1
61 (46–76)
80 (62–98)
46 (26–66)
n = 24
100 % Overall Survival
% Progression free survival
Locally advanced n = 20
Overall Survival
100 80 60 40 20
80 60 40 20
0
0 0
6
12
18
0
24
Time since starting chemotherapy in months
6 12 18 24 30 Time since starting chemotherapy in months
36
Overall Survival
Progression-free Survival 100
100
Locally Advanced Metastatic
Locally Advanced Metastatic
80
80 % Overall Survival
% Progression free Survival
All patients n = 44
60
40
60 40 20
20
0
0 0
6
12
18
Time since starting chemotherapy in months
24
0
6
12
18
24
30
36
Time since starting chemotherapy in months
Fig. 2. Kaplan–Meier survival plots for progression-free survival and overall survival in all patients (upper panel) and in locally advanced (n = 20) and metastatic (n = 24) subgroups (lower panel).
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
setting. On this background we assessed the efficacy and tolerability of a four drug combination utilising the effective cytotoxic doublet of gemcitabine and capecitabine [4] with the rational biological doublet of erlotinib and bevacizumab. When considered against historical trials’ data the efficacy findings for this four drug combination are encouraging (Table 7). The possible influence of case selection bias must always be considered when interpreting efficacy data from single-arm phase II studies. In this study the inclusion of both locally advanced and metastatic patient populations must be taken into account when considering the PFS and OS data. For the overall study population the confirmed radiological response rate was 23% with a disease control rate of 86% and median PFS and OS of 8.4 months and 12.6 months, respectively. When evaluating the outcomes of only those patients with metastatic disease the efficacy findings remain encouraging with a median PFS of 7.6 months and an overall survival of 10.1 months. The validity of the PFS data in the metastatic patient group is supported by the high proportion of patients who achieved a sustained CA19-9 marker response (68% with >50% reduction for P8 weeks) with a median duration of CA19-9 response of 20 weeks (median time to response of 4 weeks, range 2–10 weeks). The proportion of patients
Table 5 Treatment outcomes according to severity of acneiform skin rash. Acneiform rash
All patients n = 44
Grade 0/1 n = 29
Grade P2 n = 15
Radiological response rate Median progression free survival (months) Median overall survival (months) 1 year overall survival % (95% confidence interval)
10 (23) 8.4
9 (31) 7.8
1 (7) 9.3
12.6 61 (46–76)
12.1 52 (33–70)
13.1 80 (59–100)
1427
The most common adverse events attributable to bevacizumab were epistaxis, proteinuria and hypertension. Venous thromboembolism was seen in five patients (11%) and two patients had grade 3/4 gastrointestinal bleeds which may have been related to bevacizumab treatment. Other serious bevacizumab related toxicity was infrequent. Sixty day all-cause mortality was 4.5%. 4. Discussion In the 15 years since gemcitabine was defined as a standard of care, progress in the treatment of advanced pancreatic cancer has been modest with numerous novel agents failing to demonstrate benefit in the phase III
Table 6 Common adverse events and bevacizumab specific adverse events. Toxicity
Dose level 3 (n = 30) All grades (%)
Grade 3/4 (%)
All grades (%)
Grade 3/4 (%)
Diarrhoea Hand-foot syndrome Stomatitis Acneiform rash Nausea/vomiting Lethargy Febrile neutropenia Neutropenia Platelets Venous thromboembolic events Arterial thromboembolic events GI bleed Epistaxis Proteinuria Hypertension Wound healing GI perforation
24 (80) 20 (67) 17 (57) 15 (50) 23 (77) 30 (100) – 26 (87) 25 (83) 4 (13) 0 5 (16) 16 (53) 17 (57) 7 (23) 3 (10) 0
6 (20) 5 (17) 5 (17) 6 (20) 2 (7) 8 (27) 0 (0) 17 (57) 6 (20) 4 (13) 0 2 (7) 0 0 0 0 0
34 (77) 29 (66) 29 (66) 31 (70) 35 (80) 44 (100) – 39 (89) 36 (82) 5 (11) 0 5 (11) 25 (57) 21(48) 10 (23) 4 (9) 0
8 (18) 8 (18) 7 (16) 7 (16) 3 (7) 14 (32) 0 (0) 23 (52) * 6 (14) 5 (11) 0 2 (5) 1 (2) 1 (2) 1 (2) 0 0
*
All patients (n = 44)
All grade III events.
Table 7 Summary data for published trials in metastatic pancreatic cancer. Trial/arm
Progression free survival
Overall survival
Response rate (%)
AViTA – Gemcitabine + Erlotinib (n = 301) [19] AViTA – Gemcitabine + Erlotinib + Bevacizumab (n = 306) [19] FOLFIRINOX (n = 171) [7] MPACT – nab-Paclitaxel + Gemcitabine (n = 431) [8]
3.6 4.6 6.4 5.5
6.0 7.1 11.1 8.5
8.6 13.5 32 23
1428
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
that achieved a P90% reduction in CA19-9 level was 42% in the metastatic subgroup; which compares favourably to the results seen with nab-paclitaxel/gemcitabine (31%) and single agent gemcitabine (14%) in the MPACT study [8]. Prior clinical studies have shown a >90% reduction in CA19-9 levels to be associated with improved survival [8,17]. The four drug combination utilised in this study was associated with significant levels of grade 3/4 toxicity with neutropenia observed in 52%, lethargy in 32% and diarrhoea in 18%. In this good performance status patient population the toxicity profile was manageable through the use of appropriate dose reductions with minimal use of GCSF. However it is noted that four patients discontinued treatment prematurely due to persisting fatigue, suggesting that it would be difficult to apply this regimen to patients of poorer performance status. Although based on small numbers, it is notable that the outcomes in the dose escalation cohort were comparable to those seen at the recommended dose level [15]. The rate of grade 3/4 neutropenia (36%) and diarrhoea (9%) was lower in the dose escalation cohort (dose level I & II; n = 11) suggesting a more favourable toxicity/response profile might be achievable at a lower dose of capecitabine. The improved survival outcomes attained with FOLFIRINOX are also at the expense of toxicity with rates of grade 3/4 neutropenia and fatigue of 45.7% and 23.6%, respectively. Clinical experience with FOLFIRINOX has highlighted the importance of appropriate patient selection, and in routine practice initial dose modifications are frequently instigated due to concerns over tolerability [18]. Of relevance to the current study are the results of the phase III AViTA study that examined the potential benefit of bevacizumab in combination with gemcitabine and erlotinib. In this study the dual inhibition of EGFR and VEGF signalling resulted in significant prolongation of progression-free survival, median PFS 4.6 months versus 3.6 months (Hazard ratio (HR) = 0.73; 95% CI, 0.61–0.86; P = 0.0002), but no significant improvement in either response rate (13.5% for gemcitabine, erlotinib, bevacizumab versus 8.6% for gemcitabine, erlotinib) or overall survival [19]. Subgroup analysis indicated a survival advantage for bevacizumab in patients with raised baseline lactate dehydrogenase (LDH). This observation is of interest as an association between raised LDH and benefit from anti-angiogenic therapy has been noted in colorectal cancer for bevacizumab [20] and other investigational anti-angiogenic agents [21]. Retrospective molecular analysis of the AViTA study has identified further potential biomarkers, specifically; high baseline levels of plasma VEGFA [22] and polymorphisms of the VEGFR1 tyrosine-kinase domain [23]. Currently these biomarkers have not been validated in pancreatic cancer
or other solid tumours. The absence of correlative serum/tissue biomarker analysis is a weakness of the current study. The recently reported MPACT study has demonstrated improved efficacy for the combination of gemcitabine and nab-paclitaxel over single agent gemcitabine, representing a further therapeutic option in the first-line setting [8]. Although the MPACT study enrolled patients with a Karnofsky performance status (KPS) of 70, the eligible criteria remained restrictive [24]. Consequently patients with KPS 70 constituted only 7% of randomised patients. Subgroup analysis indicated poor survival outcomes in this patient group, with a median overall survival of 3.9 and 2.8 months (HR 0.99, P = 0.9632) for the combination and single agent gemcitabine arms respectively. The corresponding figures for the KPS 80 patient group were 8.1 and 5.6 months (HR 0.55, P < 0.0001) [25]. These data suggest that the benefit of nab-paclitaxel is limited to patients with good performance (KPS 80-100) and highlight the significant impact of performance status on overall survival in pancreatic cancer. The presence of constitutional symptoms (fatigue, weight loss and cachexia) is frequently observed in patients with pancreatic cancer and results in compromised QOL, performance status and treatment delivery/tolerability. Consequently the recent advances in systemic therapies are not applicable to a significant proportion of patients. The mechanisms of cachexia and fatigue are multifactorial and novel strategies are required to specifically address these aspects, in conjunction with active cytotoxic therapies, in order to improve both the tolerability and applicability of current therapies [26,27]. In conclusion, this study demonstrated that the combination of gemcitabine, capecitabine, erlotinib and bevacizumab was feasible in a good performance status population. The identification of predictive markers of response, to guide patient selection, would aid the further evaluation of this specific regimen [28]. Conflict of interest statement D. Cunningham has received research funding from Hoffmann-La Roche, D. Cunningham and I. Chau have attended advisory boards for Hoffman-La Roche and received honoraria. All remaining authors have declared no conflicts of interest. Acknowledgements and Funding This work was supported by The Royal Marsden Hospital National Health Service (NHS) Foundation Trust and endorsed by Cancer Research United Kingdom (UK). We acknowledge NHS funding from the
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
National Institute for Health Research (NIHR) Biomedical Research Centre. Hoffmann-La Roche provided free capecitabine, bevacizumab and erlotinib and an educational grant to help support the study. Roche have not had any involvement in the production of the manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/ 10.1016/j.ejca.2014.02.003.
[14]
[15]
[16]
References [1] CRUK. 2012. Available from infocancerresearchuk.org/cancerstats/types/pancreas/mortality/#Trends. [accessed 21.1.14]. [2] Ferlay J, Parkin DM, Steliarova-Foucher E. Estimates of cancer incidence and mortality in Europe in 2008. Eur J Cancer 2010;46(4):765–81. [3] Burris III HA, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano MR, 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(6):2403–13. [4] 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(33):5513–8. [5] Heinemann V, Labianca R, Hinke A, Louvet C. Increased survival using platinum analog combined with gemcitabine as compared to single-agent gemcitabine in advanced pancreatic cancer: pooled analysis of two randomized trials, the GERCOR/ GISCAD intergroup study and a German multicenter study. Ann Oncol 2007;18(10):1652–9. [6] 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(15):1960–6. [7] 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(19):1817–25. [8] Von Hoff DD, Ervin T, Arena FP, Chiorean EG, Infante J, Moore M, et al. Increased survival in pancreatic cancer with nabpaclitaxel plus gemcitabine. N Engl J Med 2013;369(18):1691–703. [9] Amgen. 2012. Available from http://www.amgen.com/media/ media_pr_detail.jsp?releaseID=1723925. [accessed 20.1.14]. [10] 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(3):256–62. [11] 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 Group-Directed Intergroup Trial S0205. J Clin Oncol 2010;28(22):3605–10. [12] Ellis LM. Epidermal growth factor receptor in tumor angiogenesis. Hematol Oncol Clin North Am 2004;18(5):1007–21, viii. [13] Naumov GN, Nilsson MB, Cascone T, Briggs A, Straume O, Akslen LA, et al. Combined vascular endothelial growth factor
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
1429
receptor and epidermal growth factor receptor (EGFR) blockade inhibits tumor growth in xenograft models of EGFR inhibitor resistance. Clin Cancer Res 2009;15(10):3484–94. Viloria-Petit A, Crombet T, Jothy S, Hicklin D, Bohlen P, Schlaeppi JM, et al. Acquired resistance to the antitumor effect of epidermal growth factor receptor-blocking antibodies in vivo: a role for altered tumor angiogenesis. Cancer Res 2001;61(13):5090–101. Starling N, Watkins D, Cunningham D, Thomas J, Webb J, Brown G, et al. Dose finding and early efficacy study of gemcitabine plus capecitabine in combination with bevacizumab plus erlotinib in advanced pancreatic cancer. J Clin Oncol 2009;27(33):5499–505. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92(3):205–16. Reni M, Cereda S, Balzano G, Passoni P, Rognone A, Fugazza C, et al. Carbohydrate antigen 19-9 change during chemotherapy for advanced pancreatic adenocarcinoma. Cancer 2009;115(12):2630–9. Verslype C, Van Cutsem E, Dicato M, Arber N, Berlin JD, Bu¨chler MW, et al. The management of metastatic pancreatic cancer: expert discussion and recommendations from the 14th ESMO/World Congress on Gastrointestinal Cancer, Barcelona, 2012. Ann Oncol 2013;24(Suppl. 4):iv5–iv10. 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(13):2231–7. Scartozzi M, Giampieri R, Maccaroni E, Del Prete M, Faloppi L, Bianconi M, et al. Pre-treatment lactate dehydrogenase levels as predictor of efficacy of first-line bevacizumab-based therapy in metastatic colorectal cancer patients. Br J Cancer 2012;106(5):799–804. Sobrero AF, Bruzzi P. Vatalanib in advanced colorectal cancer: two studies with identical results. J Clin Oncol 2011;29(15):1938–40. Jayson GC, de Haas S, Delmar P, Miles DW, Shah MA, Van Cutsem E, et al. 804 ORAL evaluation of plasma VEGFA as a potential predictive pan-tumour biomarker for bevacizumab. Eur J Cancer 2011;47:S96. Lambrechts D, Claes B, Delmar P, Reumers J, Mazzone M, Yesilyurt BT, et al. VEGF pathway genetic variants as biomarkers of treatment outcome with bevacizumab: an analysis of data from the AViTA and AVOREN randomised trials. Lancet Oncol 2012;13(7):724–33. Clinicaltrials.gov. Available from http://clinicaltrials.gov/ct2/ show/study/NCT00844649?term=abraxane+pancreatic&rank=16#locn [accessed 20.7.13]. Tabernero J et al.. Phase III trial (MPACT) of weekly nabpaclitaxel plus gemcitabine in metastatic pancreatic cancer: Influence of prognostic factors on survival. Ann Oncol 2013;24(Suppl. 4) [abstract O-0001]. Ryan JL, Carroll JK, Ryan EP, Mustian KM, Fiscella K, Morrow GR. Mechanisms of cancer-related fatigue. Oncologist 2007;12(Suppl. 1):22–34. Fearon KCH. Cancer cachexia: developing multimodal therapy for a multidimensional problem. Eur J Cancer 2008;44(8):1124–32. Lambrechts D, Lenz H-J, de Haas S, Carmeliet P, Scherer SJ. Markers of response for the antiangiogenic agent bevacizumab. J Clin Oncol 2013;31(9):1219–30.
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
The combination of a chemotherapy doublet (gemcitabine and capecitabine) with a biological doublet (bevacizumab and erlotinib) in patients with advanced pancreatic adenocarcinoma. The results of a phase I/II study q D.J. Watkins a, N. Starling a, D. Cunningham a,⇑, J. Thomas a, J. Webb a, G. Brown b, Y. Barbachano c, J. Oates a, I. Chau a a
Department of Medicine, Royal Marsden Hospital NHS Foundation Trust, London and Surrey, United Kingdom Department of Diagnostic Imaging, Royal Marsden Hospital NHS Foundation Trust, London and Surrey, United Kingdom c Computing and Statistics, Royal Marsden Hospital NHS Foundation Trust, London and Surrey, United Kingdom b
Available online 6 March 2014
KEYWORDS Anti-angiogenic therapy Chemotherapy Combination chemotherapy Epidermal growth factor receptor (EGFR) inhibition Molecularly targeted therapy Pancreatic cancer
q
Abstract Background: Preclinical data support the combined inhibition of vascular endothelial growth factor (VEGF) and epidermal growth factor receptor (EGFR) pathways in the treatment of pancreatic cancer. Following a dose finding phase I study the efficacy and toxicity of a four-drug regimen utilising the cytotoxic doublet of gemcitabine and capecitabine (GemCap), with the biological doublet of erlotinib and bevacizumab were further assessed in patients with advanced pancreatic cancer. Patients and methods: In a phase II expansion cohort, patients with chemonaive locally advanced or metastatic pancreatic cancer received gemcitabine (1000 mg/m2 D1, 8, 15), capecitabine (1400 mg/m2 D1–21), erlotinib (100 mg daily) and bevacizumab (5 mg/kg D1, 15) every 28 days. The primary endpoint was radiological response rate by response evaluation criteria in solid tumours (RECIST). Computed tomography (CT) assessment was performed every 8 weeks. Consolidation radiotherapy was considered in locally advanced patients following six cycles of treatment. Results: In total 44 patients (phases I & II) were recruited. The median cycles delivered were 6 (range 1–16). Confirmed radiological responses were seen in 23% (95% confidence interval (CI): 11–38%) of patients. The median progression-free and overall survival for the entire
Provisional analysis presented at ASCO 2010: Poster Discussion Abstract No: 4036. Citation: J Clin Oncol 28:15s, 2010 (suppl.; abstr. 4036).
⇑ Corresponding author at: Department of Medicine, Royal Marsden Hospital, Downs Road, Sutton, Surrey SM2 5PT, United Kingdom. Tel.:
+44 (0)208 661 3156; fax: +44 (0)208 643 9414. E-mail address: [email protected] (D. Cunningham). http://dx.doi.org/10.1016/j.ejca.2014.02.003 0959-8049/Ó 2014 Elsevier Ltd. All rights reserved.
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
1423
cohort was 8.4 and 12.6 months, respectively. In patients with metastatic disease the median overall survival was 10.1 months. Common grade 3/4 toxicities were; neutropenia 52%, lethargy 32%, diarrhoea 18% and hand-foot syndrome 18%. Conclusion: The combination of gemcitabine, capecitabine, erlotinib and bevacizumab was feasible with a manageable toxicity profile and demonstrated encouraging efficacy data in a good performance status population. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Pancreatic cancer is amongst the most intractable of all solid tumours with little improvement in the observed 5 year survival rate over recent decades [1]. In 2008 there were approximately 278,684 new cases of pancreatic cancer diagnosed worldwide with 266,669 deaths related to the disease [2], highlighting the dismal prognosis. Single agent gemcitabine has been a standard of care in the management of advanced pancreatic cancer since 1997, when it demonstrated a modest survival advantage (5.65 versus 4.41 months) and greater clinical benefit as compared to 5-fluorouracil [3]. Subsequent phase III studies and meta-analysis have demonstrated that additional modest gains in treatment efficacy can be achieved through the use of doublet regimens pairing gemcitabine with either capecitabine, erlotinib or platinum [4–6]. More recently the results achieved with the combination of oxaliplatin, irinotecan, leucovorin and fluorouracil (FOLFIRINOX) have established that significant survival gains are attainable with the use of a multi-drug regimen [7]. Compared to gemcitabine monotherapy FOLFIRINOX demonstrated superior response rate (31.6% versus 9.4%), progression-free survival (median 6.4 months versus 3.3 months) and overall survival (median 11.1 months versus 6.8 months) in a good performance status patient group [7]. In contrast to the advances attained with conventional cytotoxic therapies [7,8], the phase III results for molecularly targeted agents in pancreatic cancer have been disappointing [9–11], with erlotinib remaining the only licenced targeted agent in this disease. The dysregulation of multiple growth promoting signalling pathways is frequently observed in malignant disease and the existence of pathway cross-talk and redundancy has been seen to provide a mechanism of escape from the inhibitory effects of molecularly targeted therapies. Cross-talk between epidermal growth factor receptor (EGFR) and vascular endothelial growth factor (VEGF) pathways has been identified in pre-clinical models, with evidence for the involvement of EGFR signalling in the promotion of angiogenesis and the demonstration of the anti-angiogenic properties of EGFR inhibitors [12]. Furthermore the upregulation of VEGF has been implicated as a poten-
tial mediator of both innate and acquired resistance to EGFR inhibition [13,14]. These pre-clinical data provided the rationale for the evaluation of dual EGFR and VEGF blockade as part of a multi-drug regimen in the treatment of advanced pancreatic cancer. Following on from a phase I dose finding study [15], a phase II expansion aimed to evaluate the efficacy and tolerability of the four drug regimen containing the cytotoxic doublet of gemcitabine and capecitabine, in combination with the biological doublet of erlotinib and bevacizumab in patients with advanced pancreatic cancer. 2. Patients and methods 2.1. Patient selection Eligible patients were 18 years or older with histologically or cytologically proven locally advanced or metastatic pancreatic ductal adenocarcinoma or undifferentiated carcinoma. Patients were required to have adequate renal, bone marrow and liver function (creatinine clearance P 50 mL/min, platelets > 100 109/L, leucocytes > 3.0 109/L, neutrophil count > 1.5 109/ L and bilirubin < 2 the upper limit of normal), unidimensional measurable disease and have a WHO (World Health Organisation) performance status of 62. No prior chemotherapy or radiotherapy was allowed. Other key exclusion criteria included clinically significant (i.e. active) cardiovascular disease, ongoing treatment with aspirin (>325 mg/day) or other medications known to predispose to gastrointestinal ulceration, central nervous system metastases, uncontrolled hypertension, major surgery/traumatic injury within 28 days, non-healing wound/fracture, major bleeding diathesis or full-dose anticoagulation and uncontrolled comorbid conditions. When the study initially commenced patients with duodenal invasion were excluded, however these patients were later included following a protocol amendment. The study was approved by the local research ethics committee (CCR No. 2631) and conducted in compliance with International Committee on Harmonization Good Clinical Practice Guidelines. Written informed consent was obtained from all patients (ClinicalTrials.gov Identifier: NCT00260364).
1424
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
2.2. Trial design and treatment This was an open-label, single-centre study evaluating the efficacy and tolerability of the gemcitabine, capecitabine, erlotinib and bevacizumab combination. An initial dose escalation study had been undertaken to identify the optimal dose of capecitabine to use in the phase II expansion. The design and results of the dose escalation study have been reported previously [15]. In the phase II expansion all patients received gemcitabine 1000 mg/m2 on days 1, 8 and 15 of a 28-day cycle, together with bevacizumab 5 mg/kg intravenously on days 1 and 15, oral capecitabine 1400 mg/m2 days 1–21 and erlotinib 100 mg orally throughout days 1 through 28. The dose of capecitabine used in this study equates to a 15% reduction from the reported gemcitabine and capecitabine (GemCap) doublet regimen [4]. Responding patients could continue to receive study therapy for six cycles, or longer in patients deriving ongoing clinical benefit. 2.3. Dose modifications Gemcitabine was dosed according to the haematological function on the day of treatment. If the neutrophil count was 0.5–1.0 109/L or the platelet count was 50– 100 109/L, gemcitabine was dose reduced by 25%. Gemcitabine was omitted for a week for a neutrophil count of less than 0.5 109/L or platelets less than 50 109/L. In the event of neutropenic sepsis or recurrent low neutrophil/platelet counts a maintained 25% dose reduction of gemcitabine was instituted. The use of granulocyte colony stimulating factor (GCSF) was at the discretion of the investigator. For capecitabine related non-haematological toxicity capecitabine therapy was interrupted until toxicity had resolved to 6grade 1 and then resumed at 75% of the intended dose, and on second occurrence, capecitabine was resumed at 50% of the intended dose. For gemcitabine related non-haematological toxicities Pgrade 2 the dose of gemcitabine was reduced by 25%, reduction/omission was considered for grade 3 non-haematological toxicity. Erlotinib dosing was interrupted for related grade 3 non-haematological toxicities and reduced to 50 mg once a day on resolution of toxicities to 6grade 1, or discontinued for grade 4 toxicities. There were no dose reductions for bevacizumab, standard guidelines were followed for proteinuria. Bevacizumab was discontinued for GI perforation, grade 3 or 4 haemorrhage, grade 4 hypertension, uncontrolled grade 3 hypertension, arterial thromboembolism or symptomatic grade 4 venous thromboembolism. In the event of a treatment delay of greater than 3 weeks study treatment was discontinued. 2.4. Patient evaluation At every treatment visit, toxicity and a standard laboratory panel were assessed. Adverse events were graded
according to National Cancer Institute Common Terminology Criteria for Adverse Events (version 3.0). Radiological response was evaluated by CT imaging of the chest/abdomen/pelvis at baseline (within 4 weeks of starting protocol therapy) and every 8 weeks thereafter. Responses were confirmed at least 4 weeks after a responding scan. Radiological response was assessed by central review and reported according to response evaluation criteria in solid tumours (RECIST) guidelines [16]. Serum CA 19-9 marker response was assessed at four weekly intervals during treatment. Consolidation radiotherapy was permissible in patients with locally advanced disease following six cycles of treatment. Following discontinuation of treatment, patients were seen every three months until death. 2.5. Statistical plan The primary objective was to assess the overall radiological response rate of the four drug regimen in patients with previously untreated locally advanced/metastatic pancreatic cancer. Secondary objectives included an assessment of toxicity, disease control rate, progression free survival (PFS) and overall survival (OS). The sample size was determined using an optimal Simon 2 stage design. Assuming a minimum worthwhile disease response rate of 8% (p0) and an expected disease response rate of 25% (p1), 44 patients were required to achieve 90% power with a one sided alpha of 0.05. The first stage of accrual required at least two out of 16 enrolled patients to achieve a response to allow continuation to the second stage. In the second stage at least seven out of 44 patients were required to achieve a response in order to demonstrate that the lower limit of the 90% confidence interval for response rate excludes 8% and that the treatment is acceptable. Patients recruited during the dose escalation phase of the study were incorporated into the Simon 2 stage design for efficacy. 3. Results 3.1. Patient characteristics Between December 2005 and March 2009 a total of 44 patients were enrolled in both the dose finding cohort (20 patients) and phase II expansion. During the dose finding portion of the study patients were treated at one of four dose levels as previously reported [15]. In the phase II expansion all patients received treatment at the recommended dose level (dose level 3) as detailed previously. Patient characteristics are summarised in Table 1. Of the 44 enrolled patients 30 received treatment at the recommended dose level as defined during the dose finding portion of the study. The median age was 63 years and
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
98% of patients were performance status 0 or 1. Fifty five percent of patients had metastatic disease.
1425
Enrolled patients n=44
3.2. Treatment delivered The median number of cycles delivered was 6 (range 1–16). Nine patients (20%) received less than four cycles and 28 patients (64%) completed P 6 cycles of study therapy including 16 of 24 patients (67%) with metastatic disease. The treatment delivered and reasons for early discontinuation of study therapy are summarised in Fig. 1. Fatigue was the most common adverse event resulting in the discontinuation of study treatment. Other adverse events resulting in early discontinuation included; abdominal pain, bowel obstruction and infection. For the 30 patients treated at the recommended dose level, dose reductions were required for capecitabine or erlotinib in 15 patients (50%) and 11 patients (37%), respectively. The mean relative dose intensity (accounting for delays and dose reductions) for each drug was; capecitabine 79%, gemcitabine 85%, erlotinib 79%,and bevacizumab 92%. Of the 20 patients with locally advanced disease, 12 (60%) received consolidation radiotherapy and two underwent pancreatectomy with curative intent. 16 of all enrolled patients (36%) received post-study chemotherapy, nine of 24 patients (37.5%) in the metastatic disease group. 3.3. Tumour response Forty-two patients were evaluable for radiological response. The overall response rate was 23% (95% confidence interval (CI): 11–38%) with confirmed partial responses observed in ten patients. No complete responses were seen (Table 2). Sixteen patients with locally advanced disease and 19 patients with metastatic disease had elevated CA19-9 levels at study entry. The trends in CA19-9 level during treatment for each of these patients are shown in suppleTable 1 Patient characteristics and proportion of patients treated at each dose level. Characteristics
Total n = 44 (%)
Median age (range) Male:Female ECOG PS 0:1:2 Locally advanced:metastatic Head of the pancreas Body/tail of the pancreas
63 (33–80) 27:17 4:39:1 20:24 27 (61) 17 (39)
Dose level received 1 2 3 (recommended phase II dose level) 4
8 (18) 3 (7) 30 (68) 3 (7)
5 Disease progression 4 Fatigue 5 Other adverse event 2 *Other
Discontinued prior to cycle 6 n=16
Completed ≥6 cycles n=28
16 (67%) Metastatic 12 (60%) Locally advanced
Fig. 1. Treatment delivered and reasons for early discontinuation. *One patient with locally advanced disease received consolidation radiotherapy after five cycles of treatment. A further patient withdrew from the study following one cycle of treatment due to a change in personal circumstances.
mentary data A1. An exploratory analysis was undertaken examining the proportion of patients achieving a reduction in baseline CA19-9 and the duration of this response. Twenty-five of the 35 patients (71%) with elevated baseline CA19-9 levels achieved a 50% reduction lasting P 8 weeks (Table 3). Twelve patients (four locally advanced and eight metastatic) achieved a >90% reduction from baseline CA19-9 levels. 3.4. Survival The dataset was locked and analysed in August 2011. At time of analysis three patients were known to remain Table 2 Radiological response rate. Response evaluation criteria in solid tumours (RECIST) response (confirmed)
Total (%)
Radiological response rate Complete response Partial response Stable disease Disease control rate Progressive disease Non evaluable
10 (23) (95% CI:11–38%) 0 10 (23) 28 (64) 38 (86) 4 (9) 2 (5)
n = 44
Table 3 Proportion of patients achieving a 50% reduction in CA19-9 and duration of CA-19-9 response. Patients with raised baseline CA19-9
Locally advanced n = 16 (%)
Metastatic n = 19 (%)
50% reduction CA19-9 for P8 weeks 50% reduction CA19-9 for P16 weeks Median duration of CA19-9 response
12 (75)
13 (68)
5 (31)
12 (63)
14 weeks (range 10–24)
20 weeks (range 13–48)
1426
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
alive, two of whom were progression-free. One patient is lost to follow-up. For all enrolled patients the median overall survival was 12.6 months with a 1 year survival rate of 61% (95% CI: 46–76%). The corresponding median progression-free survival was 8.4 months. The survival outcomes for the 30 patients treated at the recommended phase II dose level were comparable to the whole study population with a median and 1 year overall survival of 12.6 months and 63% (95% CI: 44–80%), respectively. The median progression-free survival was 7.9 months in this subset. Kaplan-Meier survival curves are shown in Fig. 2 and survival data summarised in Table 4. The severity of skin rash was not found to be predictive of treatment outcome with a non-statistically significant trend towards longer PFS and OS in patients with Pgrade 2 acneiform skin rash as compared to those with grade 0 or I skin reaction (Table 5). 3.5. Toxicity Adverse events for all patients and those treated at the recommended phase II dose level are summarised
in Table 6. The most frequently observed non-haematological grade 3/4 adverse event was lethargy which occurred in 32% of patients. Other frequent (>15%) grade 3/4 non-haematological toxicities were diarrhoea, hand-foot syndrome, stomatitis and acneiform rash. Although the rate of neutropenia was high, with grade 3/4 neutropenia observed in 52% of patients, no febrile neutropenia was observed. Granulocyte colony stimulating factor (GCSF) was only used in two of 30 patients (6.7%) treated at the recommended dose level.
Table 4 Progression-free survival and overall survival for all patients and in the locally advanced and metastatic subgroups.
Median progression free survival (months) Median overall survival (months) 1 year overall survival % (95% confidence interval)
Progression -free Survival
Metastatic
8.4
10.1
7.6
12.6
13.9
10.1
61 (46–76)
80 (62–98)
46 (26–66)
n = 24
100 % Overall Survival
% Progression free survival
Locally advanced n = 20
Overall Survival
100 80 60 40 20
80 60 40 20
0
0 0
6
12
18
0
24
Time since starting chemotherapy in months
6 12 18 24 30 Time since starting chemotherapy in months
36
Overall Survival
Progression-free Survival 100
100
Locally Advanced Metastatic
Locally Advanced Metastatic
80
80 % Overall Survival
% Progression free Survival
All patients n = 44
60
40
60 40 20
20
0
0 0
6
12
18
Time since starting chemotherapy in months
24
0
6
12
18
24
30
36
Time since starting chemotherapy in months
Fig. 2. Kaplan–Meier survival plots for progression-free survival and overall survival in all patients (upper panel) and in locally advanced (n = 20) and metastatic (n = 24) subgroups (lower panel).
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
setting. On this background we assessed the efficacy and tolerability of a four drug combination utilising the effective cytotoxic doublet of gemcitabine and capecitabine [4] with the rational biological doublet of erlotinib and bevacizumab. When considered against historical trials’ data the efficacy findings for this four drug combination are encouraging (Table 7). The possible influence of case selection bias must always be considered when interpreting efficacy data from single-arm phase II studies. In this study the inclusion of both locally advanced and metastatic patient populations must be taken into account when considering the PFS and OS data. For the overall study population the confirmed radiological response rate was 23% with a disease control rate of 86% and median PFS and OS of 8.4 months and 12.6 months, respectively. When evaluating the outcomes of only those patients with metastatic disease the efficacy findings remain encouraging with a median PFS of 7.6 months and an overall survival of 10.1 months. The validity of the PFS data in the metastatic patient group is supported by the high proportion of patients who achieved a sustained CA19-9 marker response (68% with >50% reduction for P8 weeks) with a median duration of CA19-9 response of 20 weeks (median time to response of 4 weeks, range 2–10 weeks). The proportion of patients
Table 5 Treatment outcomes according to severity of acneiform skin rash. Acneiform rash
All patients n = 44
Grade 0/1 n = 29
Grade P2 n = 15
Radiological response rate Median progression free survival (months) Median overall survival (months) 1 year overall survival % (95% confidence interval)
10 (23) 8.4
9 (31) 7.8
1 (7) 9.3
12.6 61 (46–76)
12.1 52 (33–70)
13.1 80 (59–100)
1427
The most common adverse events attributable to bevacizumab were epistaxis, proteinuria and hypertension. Venous thromboembolism was seen in five patients (11%) and two patients had grade 3/4 gastrointestinal bleeds which may have been related to bevacizumab treatment. Other serious bevacizumab related toxicity was infrequent. Sixty day all-cause mortality was 4.5%. 4. Discussion In the 15 years since gemcitabine was defined as a standard of care, progress in the treatment of advanced pancreatic cancer has been modest with numerous novel agents failing to demonstrate benefit in the phase III
Table 6 Common adverse events and bevacizumab specific adverse events. Toxicity
Dose level 3 (n = 30) All grades (%)
Grade 3/4 (%)
All grades (%)
Grade 3/4 (%)
Diarrhoea Hand-foot syndrome Stomatitis Acneiform rash Nausea/vomiting Lethargy Febrile neutropenia Neutropenia Platelets Venous thromboembolic events Arterial thromboembolic events GI bleed Epistaxis Proteinuria Hypertension Wound healing GI perforation
24 (80) 20 (67) 17 (57) 15 (50) 23 (77) 30 (100) – 26 (87) 25 (83) 4 (13) 0 5 (16) 16 (53) 17 (57) 7 (23) 3 (10) 0
6 (20) 5 (17) 5 (17) 6 (20) 2 (7) 8 (27) 0 (0) 17 (57) 6 (20) 4 (13) 0 2 (7) 0 0 0 0 0
34 (77) 29 (66) 29 (66) 31 (70) 35 (80) 44 (100) – 39 (89) 36 (82) 5 (11) 0 5 (11) 25 (57) 21(48) 10 (23) 4 (9) 0
8 (18) 8 (18) 7 (16) 7 (16) 3 (7) 14 (32) 0 (0) 23 (52) * 6 (14) 5 (11) 0 2 (5) 1 (2) 1 (2) 1 (2) 0 0
*
All patients (n = 44)
All grade III events.
Table 7 Summary data for published trials in metastatic pancreatic cancer. Trial/arm
Progression free survival
Overall survival
Response rate (%)
AViTA – Gemcitabine + Erlotinib (n = 301) [19] AViTA – Gemcitabine + Erlotinib + Bevacizumab (n = 306) [19] FOLFIRINOX (n = 171) [7] MPACT – nab-Paclitaxel + Gemcitabine (n = 431) [8]
3.6 4.6 6.4 5.5
6.0 7.1 11.1 8.5
8.6 13.5 32 23
1428
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
that achieved a P90% reduction in CA19-9 level was 42% in the metastatic subgroup; which compares favourably to the results seen with nab-paclitaxel/gemcitabine (31%) and single agent gemcitabine (14%) in the MPACT study [8]. Prior clinical studies have shown a >90% reduction in CA19-9 levels to be associated with improved survival [8,17]. The four drug combination utilised in this study was associated with significant levels of grade 3/4 toxicity with neutropenia observed in 52%, lethargy in 32% and diarrhoea in 18%. In this good performance status patient population the toxicity profile was manageable through the use of appropriate dose reductions with minimal use of GCSF. However it is noted that four patients discontinued treatment prematurely due to persisting fatigue, suggesting that it would be difficult to apply this regimen to patients of poorer performance status. Although based on small numbers, it is notable that the outcomes in the dose escalation cohort were comparable to those seen at the recommended dose level [15]. The rate of grade 3/4 neutropenia (36%) and diarrhoea (9%) was lower in the dose escalation cohort (dose level I & II; n = 11) suggesting a more favourable toxicity/response profile might be achievable at a lower dose of capecitabine. The improved survival outcomes attained with FOLFIRINOX are also at the expense of toxicity with rates of grade 3/4 neutropenia and fatigue of 45.7% and 23.6%, respectively. Clinical experience with FOLFIRINOX has highlighted the importance of appropriate patient selection, and in routine practice initial dose modifications are frequently instigated due to concerns over tolerability [18]. Of relevance to the current study are the results of the phase III AViTA study that examined the potential benefit of bevacizumab in combination with gemcitabine and erlotinib. In this study the dual inhibition of EGFR and VEGF signalling resulted in significant prolongation of progression-free survival, median PFS 4.6 months versus 3.6 months (Hazard ratio (HR) = 0.73; 95% CI, 0.61–0.86; P = 0.0002), but no significant improvement in either response rate (13.5% for gemcitabine, erlotinib, bevacizumab versus 8.6% for gemcitabine, erlotinib) or overall survival [19]. Subgroup analysis indicated a survival advantage for bevacizumab in patients with raised baseline lactate dehydrogenase (LDH). This observation is of interest as an association between raised LDH and benefit from anti-angiogenic therapy has been noted in colorectal cancer for bevacizumab [20] and other investigational anti-angiogenic agents [21]. Retrospective molecular analysis of the AViTA study has identified further potential biomarkers, specifically; high baseline levels of plasma VEGFA [22] and polymorphisms of the VEGFR1 tyrosine-kinase domain [23]. Currently these biomarkers have not been validated in pancreatic cancer
or other solid tumours. The absence of correlative serum/tissue biomarker analysis is a weakness of the current study. The recently reported MPACT study has demonstrated improved efficacy for the combination of gemcitabine and nab-paclitaxel over single agent gemcitabine, representing a further therapeutic option in the first-line setting [8]. Although the MPACT study enrolled patients with a Karnofsky performance status (KPS) of 70, the eligible criteria remained restrictive [24]. Consequently patients with KPS 70 constituted only 7% of randomised patients. Subgroup analysis indicated poor survival outcomes in this patient group, with a median overall survival of 3.9 and 2.8 months (HR 0.99, P = 0.9632) for the combination and single agent gemcitabine arms respectively. The corresponding figures for the KPS 80 patient group were 8.1 and 5.6 months (HR 0.55, P < 0.0001) [25]. These data suggest that the benefit of nab-paclitaxel is limited to patients with good performance (KPS 80-100) and highlight the significant impact of performance status on overall survival in pancreatic cancer. The presence of constitutional symptoms (fatigue, weight loss and cachexia) is frequently observed in patients with pancreatic cancer and results in compromised QOL, performance status and treatment delivery/tolerability. Consequently the recent advances in systemic therapies are not applicable to a significant proportion of patients. The mechanisms of cachexia and fatigue are multifactorial and novel strategies are required to specifically address these aspects, in conjunction with active cytotoxic therapies, in order to improve both the tolerability and applicability of current therapies [26,27]. In conclusion, this study demonstrated that the combination of gemcitabine, capecitabine, erlotinib and bevacizumab was feasible in a good performance status population. The identification of predictive markers of response, to guide patient selection, would aid the further evaluation of this specific regimen [28]. Conflict of interest statement D. Cunningham has received research funding from Hoffmann-La Roche, D. Cunningham and I. Chau have attended advisory boards for Hoffman-La Roche and received honoraria. All remaining authors have declared no conflicts of interest. Acknowledgements and Funding This work was supported by The Royal Marsden Hospital National Health Service (NHS) Foundation Trust and endorsed by Cancer Research United Kingdom (UK). We acknowledge NHS funding from the
D.J. Watkins et al. / European Journal of Cancer 50 (2014) 1422–1429
National Institute for Health Research (NIHR) Biomedical Research Centre. Hoffmann-La Roche provided free capecitabine, bevacizumab and erlotinib and an educational grant to help support the study. Roche have not had any involvement in the production of the manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/ 10.1016/j.ejca.2014.02.003.
[14]
[15]
[16]
References [1] CRUK. 2012. Available from infocancerresearchuk.org/cancerstats/types/pancreas/mortality/#Trends. [accessed 21.1.14]. [2] Ferlay J, Parkin DM, Steliarova-Foucher E. Estimates of cancer incidence and mortality in Europe in 2008. Eur J Cancer 2010;46(4):765–81. [3] Burris III HA, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano MR, 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(6):2403–13. [4] 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(33):5513–8. [5] Heinemann V, Labianca R, Hinke A, Louvet C. Increased survival using platinum analog combined with gemcitabine as compared to single-agent gemcitabine in advanced pancreatic cancer: pooled analysis of two randomized trials, the GERCOR/ GISCAD intergroup study and a German multicenter study. Ann Oncol 2007;18(10):1652–9. [6] 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(15):1960–6. [7] 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(19):1817–25. [8] Von Hoff DD, Ervin T, Arena FP, Chiorean EG, Infante J, Moore M, et al. Increased survival in pancreatic cancer with nabpaclitaxel plus gemcitabine. N Engl J Med 2013;369(18):1691–703. [9] Amgen. 2012. Available from http://www.amgen.com/media/ media_pr_detail.jsp?releaseID=1723925. [accessed 20.1.14]. [10] 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(3):256–62. [11] 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 Group-Directed Intergroup Trial S0205. J Clin Oncol 2010;28(22):3605–10. [12] Ellis LM. Epidermal growth factor receptor in tumor angiogenesis. Hematol Oncol Clin North Am 2004;18(5):1007–21, viii. [13] Naumov GN, Nilsson MB, Cascone T, Briggs A, Straume O, Akslen LA, et al. Combined vascular endothelial growth factor
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
1429
receptor and epidermal growth factor receptor (EGFR) blockade inhibits tumor growth in xenograft models of EGFR inhibitor resistance. Clin Cancer Res 2009;15(10):3484–94. Viloria-Petit A, Crombet T, Jothy S, Hicklin D, Bohlen P, Schlaeppi JM, et al. Acquired resistance to the antitumor effect of epidermal growth factor receptor-blocking antibodies in vivo: a role for altered tumor angiogenesis. Cancer Res 2001;61(13):5090–101. Starling N, Watkins D, Cunningham D, Thomas J, Webb J, Brown G, et al. Dose finding and early efficacy study of gemcitabine plus capecitabine in combination with bevacizumab plus erlotinib in advanced pancreatic cancer. J Clin Oncol 2009;27(33):5499–505. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92(3):205–16. Reni M, Cereda S, Balzano G, Passoni P, Rognone A, Fugazza C, et al. Carbohydrate antigen 19-9 change during chemotherapy for advanced pancreatic adenocarcinoma. Cancer 2009;115(12):2630–9. Verslype C, Van Cutsem E, Dicato M, Arber N, Berlin JD, Bu¨chler MW, et al. The management of metastatic pancreatic cancer: expert discussion and recommendations from the 14th ESMO/World Congress on Gastrointestinal Cancer, Barcelona, 2012. Ann Oncol 2013;24(Suppl. 4):iv5–iv10. 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(13):2231–7. Scartozzi M, Giampieri R, Maccaroni E, Del Prete M, Faloppi L, Bianconi M, et al. Pre-treatment lactate dehydrogenase levels as predictor of efficacy of first-line bevacizumab-based therapy in metastatic colorectal cancer patients. Br J Cancer 2012;106(5):799–804. Sobrero AF, Bruzzi P. Vatalanib in advanced colorectal cancer: two studies with identical results. J Clin Oncol 2011;29(15):1938–40. Jayson GC, de Haas S, Delmar P, Miles DW, Shah MA, Van Cutsem E, et al. 804 ORAL evaluation of plasma VEGFA as a potential predictive pan-tumour biomarker for bevacizumab. Eur J Cancer 2011;47:S96. Lambrechts D, Claes B, Delmar P, Reumers J, Mazzone M, Yesilyurt BT, et al. VEGF pathway genetic variants as biomarkers of treatment outcome with bevacizumab: an analysis of data from the AViTA and AVOREN randomised trials. Lancet Oncol 2012;13(7):724–33. Clinicaltrials.gov. Available from http://clinicaltrials.gov/ct2/ show/study/NCT00844649?term=abraxane+pancreatic&rank=16#locn [accessed 20.7.13]. Tabernero J et al.. Phase III trial (MPACT) of weekly nabpaclitaxel plus gemcitabine in metastatic pancreatic cancer: Influence of prognostic factors on survival. Ann Oncol 2013;24(Suppl. 4) [abstract O-0001]. Ryan JL, Carroll JK, Ryan EP, Mustian KM, Fiscella K, Morrow GR. Mechanisms of cancer-related fatigue. Oncologist 2007;12(Suppl. 1):22–34. Fearon KCH. Cancer cachexia: developing multimodal therapy for a multidimensional problem. Eur J Cancer 2008;44(8):1124–32. Lambrechts D, Lenz H-J, de Haas S, Carmeliet P, Scherer SJ. Markers of response for the antiangiogenic agent bevacizumab. J Clin Oncol 2013;31(9):1219–30.
