Table A2.

Multivariate analyses of the associations between individual metabolites and 5-year RFS by sex, using patients with complete information of KRAS mutation, BRAF mutation, and MSI status (N = 136, females = 67, males = 69).

Metabolite Name	Females			Males			Int. Sex p Value b
	HR	95% CI	p Value a	HR	95% CI	p Value a
Arginine	3.05	1.33–7.02	0.009	1.88	0.64–5.52	0.252	0.218
Asparagine	2.60	1.06–6.37	0.036	1.02	0.56–1.85	0.943	0.058
Glutathione disulfide	0.67	0.47–0.94	0.022	0.84	0.62–1.15	0.274	0.264
Hypoxanthine	6.02	0.55–65.93	0.142	0.37	0.17–0.80	0.011	0.023
LysoPC(16:0)	1.94	1.11–3.37	0.020	0.69	0.39–1.22	0.202	0.025
LysoPE(18:2)	1.75	0.79–3.89	0.167	0.40	0.17–0.94	0.035	0.059
Methionine	3.40	1.10–10.53	0.034	1.12	0.43–2.89	0.814	0.095
N1-acetylspermine	0.95	0.66–1.36	0.781	3.20	1.32–7.76	0.010	0.016
Succinate	0.87	0.25–3.11	0.834	0.07	0.01–0.92	0.043	0.143

^a p values for the association between individual metabolites and 5-year RFS calculated by multivariate Cox PH regression models, adjusted for anatomic location, clinical stages, chemotherapy history, age, KRAS mutation, BRAF mutation, and MSI status (before FDR adjustment). The abundance of each metabolite was treated as a continuous variable and was log₂ transformed. A metabolite with HR < 1 was associated with better RFS; a metabolite with HR > 1 was associated with worse RFS. ^b p value of the interaction between the metabolite and sex. Int. Sex p value < 0.05 indicate a significant sex interaction.