200P - Diagnostic accuracy of Raman spectroscopy for malignant and benign colorectal lesions: A meta-analysis

Date 17 December 2016
Event ESMO Asia 2016 Congress
Session Poster lunch
Topics Colon and Rectal Cancer
Rectal Cancer
Staging Procedures (clinical staging)
Presenter Yan Tie
Citation Annals of Oncology (2016) 27 (suppl_9): ix53-ix67. 10.1093/annonc/mdw581
Authors Y. Tie
  • Department Of Oncology, West China Hospital, Huaxi, Sichuan University, 610041 - Chengdu/CN



Raman spectroscopy (RS) is fast, non-invasive, relatively specific, widely available and convenient, which can provide important biochemical macromolecules information including nucleic acids, proteins and lipids based on the inelastic light scattering. RS has been a potential method for colorectal cancer early diagnosis. This work aims to determine, in a meta-analysis for the first time, the diagnostic performance of RS contrast with biopsy in patients with colorectal lesions.


Relevant studies were identified through a comprehensive search of PubMed, Medline, Embase and the Cochrane Library. Studies evaluating the diagnostic performance of RS in patients suspected of having colorectal cancer who underwent RS and biopsy were reviewed. We included studies using biopsy as a reference standard. The pooled weighted sensitivity and specificity were calculated and a summary receiver operating characteristic curve (SROC) was rendered by using Meta-Disc Version 1.4 and STATA 12.0. Further subgroup analysis was conducted in this work. We conducted Deeks ' funnel plot asymmetry test to investigate publication bias. The quality of the studies was rated with the QUADAS 2.


The search strategy produced 113 hits after duplicates removed. Finally, 14 studies were included in this meta-analysis. A total of 1274 patients and 1660 lesions were assessed. Pooled weighted estimates of sensitivity and specificity were 0.87 (95% confidence interval: 0.86 - 0.89) and 0.89 (95% CI: 0.88 - 0.90), respectively. The pooled diagnostic odds ratio of RS in the diagnosis of colorectal cancer was 66.42 (95% CI, 32.90 - 134.08).Symmetric SROC curves showed an overall area under the curve of 0.9578. There was no significant publication bias all around the world (P = 0.34).


RS has considerable sensitivity and specificity in the evaluation of colorectal lesions. RS is a promising, reliable method for differential diagnosis of benign and malignant colorectal lesions.

Clinical trial indentification

Legal entity responsible for the study

Yan Tie


Sichuan University, Chengdu, China


All authors have declared no conflicts of interest.