187P - Optimal treatment strategy of first-line oxaliplatin (Oxa)-containing therapy in metastatic colorectal cancer (mCRC): A trial-level meta-analysis

Date 17 December 2016
Event ESMO Asia 2016 Congress
Session Poster lunch
Topics Anticancer Agents
Colon and Rectal Cancer
Rectal Cancer
Presenter Toshikazu Moriwaki
Citation Annals of Oncology (2016) 27 (suppl_9): ix53-ix67. 10.1093/annonc/mdw581
Authors T. Moriwaki1, M. Gosho2, A. Sugaya1, T. Yamada1, Y. Yamamoto1, I. Hyodo1
  • 1Faculty Of Medicine, Division Of Gastroenterology, University of Tsukuba, 305-8575 - Tsukuba/JP
  • 2Faculty Of Medicine, Department Of Clinical Trial And Clinical Epidermiology, University of Tsukuba, 305-8575 - Tsukuba/JP



The treatment strategy of Oxa-containing therapy, which is comprised of induction therapy (IT) followed by maintenance therapy without Oxa, and then considering Oxa reintroduction (ReOxa), is recognized as an option in mCRC patients because of particularly Oxa induced chronic peripheral neuropathy. However, the optimal duration of IT and number (No.) of maintenance drugs (MDs) and the need for ReOxa are unclear.


We conducted a literature search of phase II and III trials of first-line Oxa-containing chemotherapy including maintenance therapy for mCRC that were presented up to March 2016, and investigated relationships between median overall survival (OS) and duration of IT, No. of MDs, and ReOxa rate using correlation analysis and weighted multivariate regression analysis.


Twenty-seven treatment arms were identified among 18 trials and a sample size of 3,109 patients was analyzed. IT was 11 chemotherapy alones and 16 chemotherapy + targeted agents. Median duration of IT was 18 weeks (range, 12 to 27). No. of MDs was 0 in 8 treatment arms, 1 (FU or targeted agent) in 11 treatment arms, and 2 (FU + targeted agent) in 9 treatment arms. Median ReOxa rate was 30.9% (range, 0 to 100). Median OS was 23.2 months (range, 15.8 to 31.0). Duration of IT was weakly correlated with OS (Spearman’s partial correlation coefficient, r = 0.14). No. of MDs was moderately correlation with OS (r = 0.60). ReOxa rate was weakly correlation with OS (r = 0.07). In multivariate regression analysis, No. of MDs was significantly correlated with OS (P = 0.004). There was a significant interaction between duration of IT and No. of MDs (P = 0.014). The longest predicted OS was the treatment strategy comprised of IT for 12 weeks and 2 MDs when IT for 12–24 weeks and 0–2 MDs were applied to the model.


The strongest correlation with improvement of OS was the maintenance therapy using FU + targeted agent. The optimal treatment strategy of Oxa-containing therapy may be comprised of IT for 12 weeks followed by two MDs, not considering ReOxa in first-line setting for mCRC.

Clinical trial indentification

Legal entity responsible for the study

National university corporation


Division of Gastroenterology


All authors have declared no conflicts of interest.