Published online May 21, 2020. doi: 10.3748/wjg.v26.i19.2427
Peer-review started: January 21, 2020
First decision: March 6, 2020
Revised: April 23, 2020
Accepted: April 28, 2020
Article in press: April 28, 2020
Published online: May 21, 2020
Survival benefit of neoadjuvant chemotherapy (NAC) for advanced gastric cancer (AGC) is a debatable issue. Studies have shown that the survival benefit of NAC is dependent on the pathological response to chemotherapy drugs. For those who achieve pathological complete response (pCR), NAC significantly prolonged prolapsed-free survival and overall survival. For those with poor response, NAC yielded no survival benefit, only toxicity and increased risk for tumor progression during chemotherapy, which may hinder surgical resection. Thus, predicting pCR to NAC is of great clinical significance and can help achieve individualized treatment in AGC patients.
To establish a nomogram for predicting pCR to NAC for AGC patients.
Two-hundred and eight patients diagnosed with AGC who received NAC followed by resection surgery from March 2012 to July 2019 were enrolled in this study. Their clinical data were retrospectively analyzed by logistic regression analysis to determine the possible predictors for pCR. Based on these predictors, a nomogram model was developed and internally validated using the bootstrap method.
pCR was confirmed in 27 patients (27/208, 13.0%). Multivariate logistic regression analysis showed that higher carcinoembryonic antigen level, lymphocyte ratio, lower monocyte count and tumor differentiation grade were associated with higher pCR. Concordance statistic of the established nomogram was 0.767.
A nomogram predicting pCR to NAC was established. Since this nomogram exhibited satisfactory predictive power despite utilizing easily available pretreatment parameters, it can be inferred that this nomogram is practical for the development of personalized treatment strategy for AGC patients.
Core tip: Pathological complete response is an important prognosis factor for advanced gastric cancer patients who underwent neoadjuvant chemotherapy and tumor resection. In our study, we built a nomogram that predicted pathological complete response to neoadjuvant chemotherapy utilizing only easily available pretreatment parameters such as carcinoembryonic antigen level, lymphocyte ratio, monocyte count and tumor differentiation grade. It showed satisfactory predictive power with an area under the receiver operating characteristic curve of 0.823 and a concordance statistic of 0.767. It can be inferred that this nomogram is practical for the development of personalized treatment strategy for advanced gastric cancer patients.