Table 2. Univariate and Multivariate Cox Regression analysis of the development of metastatic PC from the time of surgery and CRPC, prostate cancer specific death after starting androgen deprivation therapy.

	No. of patients with event	Univariate analysis		Multivariate analysis
		Relative Risk (95% CI)	P value	Relative Risk (95% CI)	P value
Progression to metastatic PC (n = 87 a )	27
HIF1α status
HIF1α negative	0	1.0‡		1.0‡
HIF1α positive	27	10.7 (4.5–∞)	0.011	9.8 (3.9–∞)	0.017
Gleason score †
Gleason ≤7	1	1.0‡		1.0‡
Gleason >7	26	6.9 (5.8–∞)	0.002	10.7 (9.7–∞)	0.001
Age		1.1(1.0–1.1)	0.013	1.0 (1.0–1.1)	0.17
Pre-interventional PSA*		1.0 (0.99–1.00)		1.0 (1.0–1.0)	0.66
Development of CRPC in patients on ADT (n = 43)	23
HIF1α status †
HIF1α negative	0	1.0‡		1.0‡
HIF1α positive	23	8.3 (3.1–∞)	0.03	10.0 (4.0–∞)	0.021
Age		1.0 (1.0–1.1)	0.404	1.0 (0.9–1.0)	0.448
Pre-interventional PSA*		1.0 (1.0–1.0)	0.349	1.0 (1.0–1.0)	0.836
PC specific deaths in patients on ADT (n = 43)	13
HIF1α status †
HIF1α negative	0	1.0‡		1.0‡
HIF1α positive	13	5.7 (2.1–∞)	0.111	3.07 (1.5–∞)	0.36
Age		1.1 (1.0–1.1)	0.014	1.1(1.0–1.2)	0.38
Pre-interventional PSA*		1.0 (0.9–1.0)	0.270	1.0 (1.0–1.0)	0.78

^a13 patients excluded due to incomplete metastasis related data.

^†Cox regression with Firth’s penalized maximum likelihood method. CI denotes confidence interval.

^‡This group served as the reference group in the Cox regression analysis.

^*Pre-interventional PSA was defined as PSA immediately prior to obtaining the tissue sample.