Open Access Highly Accessed Open Badges Research article

Possibility of multivariate function composed of plasma amino acid profiles as a novel screening index for non-small cell lung cancer: a case control study

Jun Maeda1, Masahiko Higashiyama1, Akira Imaizumi2*, Tomio Nakayama3, Hiroshi Yamamoto24, Takashi Daimon5, Minoru Yamakado6, Fumio Imamura3 and Ken Kodama1

Author Affiliations

1 Department of Thoracic Surgery, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan

2 Institute for Innovation, Ajinomoto, CO., Inc., Kawasaki, Japan

3 Department of Pulmonary Oncology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan

4 HI Department, Ajinomoto, CO., Inc., Tokyo, Japan

5 Department of Biostatistics, Hyogo College of Medicine, Nishinomiya, Japan

6 Center for Multiphasic Health Testing & Services, Mitsui Memorial Hospital, Tokyo, Japan

For all author emails, please log on.

BMC Cancer 2010, 10:690  doi:10.1186/1471-2407-10-690

Published: 22 December 2010



The amino-acid balance in cancer patients often differs from that in healthy individuals, because of metabolic changes. This study investigated the use of plasma amino-acid profiles as a novel marker for screening non-small-cell lung cancer (NSCLC) patients.


The amino-acid concentrations in venous blood samples from pre-treatment NSCLC patients (n = 141), and age-matched, gender-matched, and smoking status-matched controls (n = 423), were measured using liquid chromatography and mass spectrometry. The resultant study data set was subjected to multiple logistic regression analysis to identify amino acids related with NSCLC and construct the criteria for discriminating NSCLC patients from controls. A test data set derived from 162 patients and 3,917 controls was used to validate the stability of the constructed criteria.


The plasma amino-acid profiles significantly differed between the NSCLC patients and the controls. The obtained model (including alanine, valine, isoleucine, histidine, tryptophan and ornithine concentrations) performed well, with an area under the curve of the receiver-operator characteristic curve (ROC_AUC) of >0.8, and allowed NSCLC patients and controls to be discriminated regardless of disease stage or histological type.


This study shows that plasma amino acid profiling will be a potential screening tool for NSCLC.