Pancreatic cancer within a UK cancer network with special emphasis on locally advanced non-metastatic pancreatic cancer

Aims The outcome of patients with pancreatic cancer from an unselected population within a UK region has not previously been reported. We undertook a review of pancreatic cancer in southeast Wales, with an emphasis on locally advanced non-metastatic pancreatic cancer (LANPC) in an attempt to define a subgroup of patients who would probably benefit from multi-modality treatment. Materials and methods Case notes of patients referred to Velindre Hospital between 1 January 2002 and 31 December 2005 were reviewed. Data on patient demographics, tumour characteristics, treatment, treatment response and overall survival were collected. The Log-rank test was used to compare survival between groups and Cox regression was used to evaluate whether age, gender, tumour site and treatment response correlated with overall survival in LANPC. Results Of the 354 referrals (complete data on 315 patients), 93% were inoperable and 51% of inoperable patients received active treatment (149/294). One hundred and fourteen patients out of 315 (36%) had LANPC and 72/114 (64%) were fit for active treatment, including chemotherapy (n = 66) and chemoradiotherapy (CRT) (n = 6). The median survival of patients with LANPC was 7.4 months (95% confidence interval 6.4–8.5). Survival for patients receiving chemotherapy, CRT and no treatment was 9.2 (7.5–10.7), 12.6 (6.1–19.1) and 4.5 (3.7–5.3) months, respectively. Overall survival of patients who had non-progressive disease after initial chemotherapy was significantly better than those who progressed (11.8 vs 6.6 months, P = 0.01). Of the 180/315 (57%) patients presenting with metastatic disease, 43% received active treatment. Overall survival of metastatic patients was 2.8 months (2.3–3.2 months); for those receiving active treatment, this was 5.6 months (5.1–6.1 months) and for those receiving active supportive care 1.8 months (1.6–2.0 months). Conclusions In this UK network, about half of the patients received active treatment. Although the overall outcome was poor, that of treated patients was comparable with published studies. For patients with LANPC, the initial response or disease stabilisation on chemotherapy defined a subset of patients who had better outcome. The role of CRT over and above chemotherapy needs to be defined through trials that should include a neoadjuvant ‘chemotherapy-only’ phase to select out patients who benefit from multi-modality treatment