Prognostic Scale Developed for Overall Survival in Locally Advanced Pancreatic Cancer

Four independent prognostic factors for improved overall survival in patients with locally advanced pancreatic cancer emerged from an analysis of the LAP 07 phase III trial.

Four independent prognostic factors for improved overall survival (OS) in patients with locally advanced pancreatic cancer (LAPC) emerged from an analysis of the LAP 07 phase III trial presented at the ESMO 16th World Congress on Gastrointestinal Cancer. Based on these factors, a risk scale was developed that could be useful for stratifying and treating patients.

Lead author Dewi Vernerey, MSc, Methodological and Quality of Life in Oncology Unit, University Hospital, Besançon, France, discussed the grim reality of pancreatic cancer, noting that 30% of patients present with disease that is locally advanced. He added that pancreatic cancer disease mortality in Europe increased from 73,439 deaths in 2009 to 82,300 deaths in 2014, an increase of approximately 12%.

Vernerey presented results from a data analysis from LAP 07, an international multicenter, randomized phase III trial. In LAP 07, investigators evaluated whether continued chemoradiotherapy improved OS in patients with inoperable LAPC whose tumors were determined to be controlled following 4 months of induction gemcitabine-based chemotherapy. LAP 07 found no advantage for continued chemotherapy versus no treatment in OS (P = .09) or progression-free survival (P = .216).

In Vernerey et al’s analysis, thirty-five baseline characteristics from demographics, radiological, disease history, and biological parameters of 370 patients from LAP 07 comprised the variables evaluated by univariate and multivariate analysis to identify independent factors prognostic for OS.

Univariate analysis narrowed the characteristics to WHO tumor status 2, age at diagnosis, blood pressure, ASAT, albumin, the presence of pain, tumor size, and poorly differentiated histology grade as potential prognostic variables (all P values <.1).

Upon multivariate analysis, age (HR = 1.01; 95% CI, 1.00-1.03; P = .0418), tumor size (HR = 1.01; 95% CI, 1.00-1.02; P = .0033), increased albumin (HR = 0.96; 95% CI, 0.94-0.98; P = .0001), and the presence of pain (HR = 1.36; 95% CI, 1.08-1.71; P = .0094) emerged as independent prognostic factors of OS.

Using these four variables, a prognostic score scale for risk of death in patients with LAPC was devised, with scale scores ranging from 0 to 4. Based on the scale, three risk groups for death were identified: lower risk (0, 1), intermediate risk (1, 2), and higher risk (2, 3).

Patients were stratified using the scale, with 17, 166, and 187 patients identified for the lower, intermediate, and higher risk groups, respectively. By global rank test, the three risk groups were associated with a median OS of 18.8, 13.4, and 11.8 months, respectively (P = .0101).

“Nomogram development and this simple score should allow patient stratification that can direct treatment management and design of future clinical trials,” said Vernerey. He added that external validation of this score using a cohort from the large ARCAD (Aide et Recherche en Cancérologie Digestive) database is ongoing.

No external funding was reported.

Vernerey D, et al. Prognosis model for overall survival in locally advanced pancreatic cancer (LAPC): an ancillary study of the LAP 07 trial. Presented at: the ESMO 16th World Congress on Gastrointestinal Cancer; June 25-28, 2014; Barcelona, Spain. Abstract O-0002.


View more from the 2014 World GI Congress