Human epidermal growth factor receptor 2 (HER2) status remains an outcome-associated biomarker independent of known prognostic factors for metastatic gastric cancer (MGC). This study aimed to develop nomograms that combine HER2 status and other prognostic factors to predict the survival outcome of individual patients with MGC starting first-line treatment.
We retrospectively analyzed 838 consecutive patients with MGC starting first-line chemotherapy at a single institution between January 2005 and December 2012. We established the nomograms that calculated the predicted probability of survival at different time points: 1 and 2 years for overall survival (OS). To determine which prognostic factors were included in the nomograms, the HER2 status and the other relevant covariates were analyzed in the Cox proportional hazards models. Missing covariate data were estimated using multiple imputation methods. The discriminatory ability and accuracy of the models were assessed using Harrell's c-index. IHC3+ or IHC2 + /ISH+ tumors were defined as HER2 positive.
At a median follow-up of 12.3 months (range, 0.2–92.5 months), 757 (90.3%) patients had died. The median OS was 12.5 months [95% confidence interval (CI), 11.8–13.2 months]. HER2 positivity showed a trend toward improvement in OS [hazard ratio 0.87 (95% CI, 0.70–1.08)]. Selected five independent prognostic factors of OS, such as Eastern Cooperative Oncology Group performance status (0 vs. 1 vs. >2), history of gastrectomy (no vs. yes), serum lactic acid dehydrogenase (≥380 vs. <380 IU/L), serum alkaline phosphatase level (≥240 vs. <240 IU/L) and HER2 status (negative vs. positive), were used to develop the nomograms. The developed nomograms were capable of predicting OS with a c-index of 0.68. The 2-year overall survival for patients with total predicting score > 100 was significantly worse than those with < 55 (2-year survival rate, 7.0% vs. 37.9%).
These nomograms may provide objective and approximate prediction of the OS for individual MGC patients in clinical settings. A validation study is warranted to evaluate the effectiveness of these nomograms.
Clinical trial identification
All authors have declared no conflicts of interest.