The 55 STAR study: Prognostic and predictive value of the 55-gene classifier (55GC) in stage III colon cancer

• Published in: Journal of Clinical Oncology Vol. 37; no. 15_suppl; p. 3597
• Main Authors: Ishikawa, Toshiaki, Oki, Eiji, Shinto, Eiji, Shimokawa, Mototsugu, Yamaguchi, Shigeki, Ishiguro, Megumi, Hasegawa, Seiji, Takii, Yasumasa, Ishida, Hideyuki, Kusumoto, Tetsuya, Morita, Masaru, Tomita, Naohiro, Shiozawa, Manabu, Tanaka, Masafumi, Ozawa, Heita, Hashiguchi, Yojiro, Ohnuma, Shinobu, Tada, Sachiyo, Matsushima, Tomoko, Hase, Kazuo
• Format: Journal Article
• Language: English; Japanese
• Published: American Society of Clinical Oncology (ASCO) 20.05.2019
• DOI: 10.1200/jco.2019.37.15_suppl.3597
• ISSN: 0732-183X

Abstract only 3597 Background: Patient prognosis can be predicted based on cancer subtypes classified according to DNA microarray results. The most robust classification system involves the consensus molecular subtypes, which uses over 600 genes for classification. To simplify this classification, we recently constructed a 55-gene classifier (55GC) to classify colon cancer (CC) into three subtypes with different recurrence rates: “microsatellite instability (MSI)-like,” “chromosomal instability (CIN)-like,” and “stromal” colon cancers. The 55GC has been reported to be a useful and reproducible grading system for stage II CC recurrence risk stratification. This study aimed to explore the usefulness of 55GC for classifying stage III CC patients. Methods: We retrospectively identified stage III CC patients aged 20-79 years who underwent curative surgery and received adjuvant chemotherapy with or without oxaliplatin (OX) between 2009 and 2012 from 15 institutions. Propensity score matching was used to adjust for the number of lymph node metastases, tumor location, sex, and age. Results: Among 938 eligible patients, 203 and 201 cases involving adjuvant chemotherapy with and without OX were selected, respectively, using propensity score matching. Ninety-five cases each from groups were analyzed after exclusion of cases involving low-quality specimens and those involving chemotherapy for < 3 months. The 5-year relapse-free survival (RFS) in patients receiving adjuvant chemotherapy with and without OX were 77.1% and 73.7%, respectively. Classification of the stage III CC according to 55GC and related 5-year RFS rates were as follows: stromal (N = 60), 66.6%; CIN-like (N = 78), 80.5%; and MSI-like (N = 52), 78.4%. The HRs for 5-year RFS for adjuvant chemotherapy with and without OX in each subtype were as follows: stromal, HR = 0.791 (95% CI = 0.329-1.901); CIN-like, HR = 1.241 (95% CI = 0.465-3.308); and MSI-like, HR = 0.495 (95% CI = 0.145-1.692). Conclusions: The stromal subtype showed poor prognosis in stage III as well as stage II patients. Oxaliplatin had a good additive effect in adjuvant chemotherapy for MSI-like subtype. The 55GC is useful for predicting the efficacy of adjuvant chemotherapy for CC.