Findings from a study presented at the virtual 22nd ESMO World Congress on Gastrointestinal Cancer posit that a nomogram based on 5 independent prognostic factors can conveniently predict survival in patients with unresectable pancreatic cancer.

“Although the American Joint Committee on Cancer (AJCC) TNM classification is used as a traditional staging system for cancer, the survival of patients with unresectable pancreatic cancer varies even in patients with the same AJCC stage,” wrote Taro Shibuki, MD, National Cancer Center Hospital East, Kashiwa, Japan, and colleagues, who sought to create and validate a prognostic nomogram for use before administration of first-line gemcitabine plus nab-paclitaxel or combination fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) in this patient population.

The nomogram study by Dr Shibuki et al was based on an existing cohort of patients involved in the multi-center, retrospective NAPOLEON study, in which patients with unresectable pancreatic cancer given first-line gemcitabine plus nab-paclitaxel or FOLFIRINOX between December 2013 and March 2017 were assessed.

Overall, 318 patients with unresectable, locally advanced or metastatic pancreatic cancer were included in the analysis.

The investigators identified unique prognostic variables for overall survival using the Cox proportional hazards model, and developed a prognostic nomogram to predict 6-, 12- and 18-month survival probability that was then validated via concordance index (C-index) and calibration plot by the bootstrapping method.

In addition, to assess the discrimination ability for survival prediction, Dr Shibuki and colleagues compared the final model with the AJCC TNM staging system.

“We attempted to develop a risk stratification for survival according to the nomogram total points; lowest, middle and highest tertile were respectively defined as low-, moderate- and high-risk groups. We also estimated the nomogram-predicted survival probability and the hazard ratio (HR) with the 95% CI regarding the risk groups,” they explained.

Ultimately, the variables used in the final model to generate a nomogram were performance status, LDH, CRP, CA19-9, and liver metastasis, which had acceptable discriminations according to C-indexes (bootstrapping 95% CI) of 0.77 (0.73-0.81), 0.72 (0.67-0.76) and 0.70 (0.65-0.75) with 6-, 12- and 18-months survival, respectively.

These indexes were significantly higher at all points statistically (all P values <.01) with the nomogram versus the AJCC TNM staging system. There was agreement between prediction and actual observation at each point according to calibration plots.

A clear split in survival curves was achieved with the risk stratification using nomogram total points. Patients in the low-, moderate- and high-risk groups had median survival times of 15.8 months, 12.8 months (HR, 1.44; 95% CI, 1.03-2.01; P = .03), and 7.8 months (HR, 3.34; 95% CI, 2.40-4.64; P <.01), respectively.

“This study provided a nomogram based on 5 independent prognostic factors to predict survival in patients with unresectable pancreatic cancer. The nomogram can be a convenient way to estimate individual prognosis more accurately,” Dr Shibuki and co-investigators concluded.—Hina M. Porcelli

Shibuki T, Mizuta T, Shimokawa M, et al. Prognostic nomogram to predict overall survival in patients with unresectable pancreatic cancer treated with gemcitabine plus nab-paclitaxel or FOLFIRINOX: Real-world results from the multicenter retrospective study (NAPOLEON study). Presented at: the 22nd ESMO World Congress on Gastrointestinal Cancer; July 1-4, 2020; virtual. Abstract SO-1.