1. Resectable Pancreas Cancer
GAP Study: Phase II study of Gemcitabine and nab-Paclitaxel for
Resectable Pancreas Cancer
A. Barbour1, N. O'Rourke2, J.S. Samra3, K.S. Haghighi4, J. Kench5, J. Mitchell6, N. Pavlakis7, M.E. Burge2, J. Fawcett8, S. Gananadha9, M. Harris10, M. Aghmesheh11, Y.J. Chua12, W. L. Joubert14, M.M.K. Chan6, M.D. Chandrasegaram6, S. Yip6, J. Simes6, V. Gebski6, D. Goldstein14.
1. The University of Queensland, Brisbane, Australia; 2. Royal Brisbane and Women's Hospital, Herston, Australia; 3. Royal North Shore Hospital; University of Sydney, Sydney, Australia; 4. UNSW; Prince of Wales Hospital, Randwick, Sydney, Australia; 5. Department of Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital, Sydney, Australia; 6. NHMRC Clinical Trials Centre, Sydney, Australia; 7. Department of Medical Oncology, Royal North Shore Hospital, The University of Sydney, Sydney, Australia; 8. Queensland Liver Transplant Service, Brisbane, Australia; 9. ANU Medical School, The Canberra Hospital, Garran, Australia; 10. Monash Medical Centre, Melbourne, Australia; 11. Illawarra Cancer Care Centre, Wollongong, Australia; 12. The Canberra Hospital, Garran, Australia; 13. Princess Alexandra Hospital, Brisbane, Australia; 14. Prince of Wales Hospital, Sydney, Australia
Background
Methods and Study Schema
Surgical Complications (N=36)
Compliance with planned treatment
Local and distant recurrences are frequent problems following resection of pancreas adenocarcinoma with curative intent.
As a result, adjuvant and neoadjuvant protocols using radiotherapy (RT), chemotherapy (CT), or multimodality treatment have been investigated with adjuvant gemcitabine chemotherapy standard of care [1, 2, 3].
Unfortunately, up to 30% of patients will not be fit for post-operative therapy after pancreatic resection.
Up to 25% patients will have progressive disease during neoadjuvant therapy[1, 2, 4]. Thus, neoadjuvant therapy is able to identify patients with aggressive tumour biology who may be spared the morbidity of resection.
Resection margin (R) status is an important prognostic factor in the majority of analyses of resected pancreatic cancer.
Historically, an R0 resection was defined by a minimum 0 mm margin [5]. The R0 rate using 0 mm in 7 studies with 1926 patients, the mean R0 rate was 73% (95% CI % or range 48-83% ) [Manuscript in progress].
However, with the current definitions of a minimum 1 mm margin, the R0 rate in 14 studies with 2392 patients, the mean R0 rate was 38% (95% CI 30 – 47% or range 15% - 66%) [Manuscript in progress].
In the study by Evans et al, evaluating preoperative gemcitabine-based chemoradiotherapy (CRT) in resectable pancreas cancer, 73 of 86 (85%) had surgery with disease progression precluding 13 patients from surgery. Of these 64 had pancreaticoduodecectomy corresponding to a resectability rate of 64/86 (74%) [6].
In the study by Varadhachary et al, evaluating preoperative gemcitabine and cisplatin followed by gemcitabine-based radiotherapy, 79 (88%) of the 90 patients enrolled completed chemoradiation. 62 (78%) of the 79 patients had surgery and 52 (66%) underwent pancreaticoduodenectomy. [7].
In a more recent study by O’Reilly et al, evaluating gemcitabine and oxaliplatin in resectable pancreas cancer, 35 (92%) of 38 patients completed neoadjuvant treatment. 27 patients underwent resection corresponding to a resectability rate of 71% [8].
The administration of pre-operative chemotherapy or chemoradiotherapy has been postulated as a factor that may affect the prognostic significance and rate of R0 margin status [9].
Recently, nab-paclitaxel and gemcitabine chemotherapy has been shown to improve survival in advanced PDAC compared with gemcitabine alone with acceptable toxicity [10].
 We aimed to determine the feasibility and R0 resection rate with peri‐operative nab‐Paclitaxel and Gemcitabine for resectable PDAC .
 We had aimed to recruit 50 patients to produce an R0 rate of 85% using a 1mm margin assessing all margins.   
39/41 planned [95%] of patients completed induction chemo therapy  (although dose modifications and omissions were required by D15 of the treatment cycles.)
13/30 planned [XX%] completed 4 cycles post surgery chemo therapy.
4/14 received chemo radiation prior to receiving post op chemotherapy
Grade I
(n)
Grade II
Grade III (n)
Grade IV (n)
Cardiac Complications 1 3
Thromboembolic event 4
Gastroparesis
Pancreatic leak
Perioperative haemorrhage 2
Ascites
Abdominal Infection
Pleural Infusion
Sepsis
Wound complication
Acute Kidney Injury
Other 10
Baseline CT or MRI and PET scan. Ca 19.9, CEA
Primary endpoint (N=29)
R0 resection rate
R1 Resection Outcome – Chemotherapy as per R0 outcome or Chemoradiation
45Gy in 25 fractions with 5-FU (CI)
Gemcitabine 1000 mg/m2 and nab-Paclitaxel 125 mg/m2 D1, D8, D15 every 28 days x 4 (16 weeks)
Radiation at discretion of treating clinician
R0 resection outcome – Chemotherapy
Gemcitabine 1000 mg/m2 and nab-Paclitaxel 125 mg/m2
D1, D8, D15 every 28 days x 4 (16 weeks)
Patients with resectable pancreatic cancer; FNA +/- Core biopsy
(N= 42)
D1, D8, D15 every 28 days x2 (8 weeks) (N=41)
Day 15 FDG-PET scan plus Ca 19.9, CEA
Restaging with CT or MRI at completion of therapy prior to surgery
Surgery (N=36)
Minimum follow-up for 12 months
N=1 not eligible
N=5 did not go to surgery
N=1 not eligible on central review
N=6 unresectable disease
Chemotherapy omissions
Day 1 cycles omitted in 0% of pre op cycles and % of post op patients  
Day 8 cycles omitted in 5% and 25% pre-op and 7-27% of post op patients
Day 15 cycles omitted in  51% and 29% pre op and % post op
R0 definition and rates
Stage
Reason for omission
% of cycles omitted
% patients with an omitted dose*
Pre-operative
Haematological
36.3
53.7
Non- haematological
3.8
7.3
Both
1.3
2.4
Other
5.0
9.8
Post-operative
30.8
66.7
13.8
33.3
0.0
12.3
22.2
Chemotherapy (CT) Adverse Events (Grade III/IV)
Resectability post-neoadjuvant treatment
Most common symptoms/adverse events:
Grade III/IV n (%)
Neutrophil count decreased
21 (51%)
Alanine aminotransferase (ALT) increased
7 (17%)
Biliary tract infection
5 (12%)
Febrile neutropenia
4 (10%)
Aspartate aminotransferase (AST) increased
3 (7%)
Fatigue
* Denominator only includes patients who received at least one cycle of chemotherapy
Outcomes (n=41)
Resection rates and Primary Outcome (R0)
R0 resections
R1 resections
R0 (R0 / [R0+R1])
95% CI for R0 rates
N of patients, 1 mm margin 15 14
52% % N %
Underwent surgery 36
88%
Pancreatic resection
Evaluable cancer
No cancer on central review * 30 29 1
73%
Unresectable disease at surgery 6
15%
Did not go to surgery
Disease progression
Refused surgery
Cholangitis 5 2
12%
Study aim
Pathological tumour characteristics (n=29)
N, % (Central review)
Grade 0-2 tumour regression 15
52%
T Stage T1 2 7% T2 1 3% T3 26
90% T4 0%
N Stage N0 11
38%
*1 patient without cancer on central review excluded from tumour related outcomes but included in safety and surgical outcomes.
Methods
Baseline Characteristics (n=42)
42 patients were enrolled, 1 was not eligible.
A multi‐centre, single arm, phase 2 trial.
Histologically or cytologically confirmed resectable adenocarcinoma of the pancreas with radiological criteria of:
(1) no evidence of extra-pancreatic disease;
(2) no evidence of tumour abutment of the circumference of the superior mesenteric artery (SMA) or coeliac axis (T4); and
(3) no evidence of portal vein infiltration of more than 180°of the circumference or occlusion of the superior mesenteric vein (SMV) or SMV–portal vein (PTV) confluence.
R0 resection defined as all margins microscopically clear (minimum distance from tumour to resection margin ≥ 1.0 mm.
Conclusion
Pre-op treatment with GEM and nab‐Paclitaxel chemotherapy was delivered safely and associated with an R0 resection rate of 52% which is comparable or higher than surgical series using a minimum 1 mm margin definition
After a review by IDSMC, although there was no safety concerns it became apparent that the primary endpoint of the trial could not be met and recruitment was stopped at 42 patients.
The pancreatic resectability rate of 73% is comparable to recent neoadjuvant trials in resectable pancreas cancer
Pre-op GEM and nab‐Paclitaxel chemotherapy was delivered to 95% of patients but haematological toxicities required dose omissions in 36% of cycles
In contrast post op GEM and nab‐Paclitaxel chemotherapy was less tolerable with only 13 /30 resected patients completed the protocol planned chemotherapy.
Pre operative biliary tract infections were infrequent with a liberal use of metal biliary stents.
Tumour regression was seen in 52% of patients.
Uptake of post op chemo-radiation XXXXXXXXXXXXXXXXXXXXXXXXX
In conclusion pre-op GEM and nab-Paclitaxel chemotherapy is safe, and feasible. Post operatively there was no deaths however post operative chemotherapy was less deliverable with only 13/30 resected patients completing post op therapy. The value of this regimen will also depend on longer term outcomes related to recurrence and survival which will be reported in future analysis.
Characteristics
Patients, n 42
Age (years; median, range)
65 (43-79)
Gender (male)
17 (41%)
Ca 19.9 (U/mL; median, range)
81 (1 – 7305)
Tumour size (median, range; on final path assessment)
27.5 (25 – 35)
References
1 Neoptolemos JP et al. JAMA 2010; 2 Evans DB et al. J Clin Oncol 2008; 3 Varadhachary GR et al. J Clin Oncol 2008; 4 Neoptolemos JP et al. Lancet 2001; 5 Sobin L. International Union against cancer: TNM Classification of malignant tumours. New York: Wiley-Liss; 1997; 6 Evans DB et al. J Clin Oncol 2008; 7 Varadhachary et al. J Clin Oncol 2008; 8. O’reilly EM et al. Ann Surg 2014; 9. Palmer DH et al. Ann Surg Oncol 2007; Von Hoff DD et al. NEJM 2013
Clinical Trial Registry: ACTRN The GAP study is an investigator initiated study, sponsored by the AGITG. Specialised Therapeutics Australia supplied nab-paclitaxel, Gemcitabine available on PBS