Using Molecular Markers to Help Predict Who Will Fail after Radical Prostatectomy

Table 1

Ki-67 and outcomes after radical prostatectomy. The table indicates whether Ki-67 was positive on univariate or multivariate analysis for predicting failure. The failure of the entire cohort is given and then the outcomes for patients where Ki-67 was elevated versus not elevated.

Study	#pts	Med (mean) months f/u	Include LN+ (#)	Include Adj RX (#)	Definition of failure^@	Univariate positive	Multivariate positive	Group overall failure	Marker elevated outcome	Marker not elevated outcome
Khatami et al. [6]	50	(63)	No	NR	PSA > 0.2 × 2	Yes	NR	18%	NR	NR
Bubendorf et al. [7]	137	(64)	Yes (34)	Yes (60)	PSA, PAP, or ALP elevated*	Yes	Yes	29%	65% 5-yr dfs	78% 5-yr dfs
May et al. [8]	528	46 (49)	Yes (38)	No	PSA > 0.2	Yes	Yes	19%; 5-yr dfs 78%	70% 5-yr dfs	88% 5-yr dfs
Miyake et al. [9]	193	63	Yes (13)	No	PSA > 0.2	Yes	Yes	21%; 5-yr dfs 79%	A: 79% recur; B: 20% recur	C: 4% recur
Rubio et al. [10]	91	46.5	NR	No	PSA ≥0.2	Yes	Yes	32%	42% 5-yr dfs	84% 5-yr dfs
Laitinen et al. [11]	229	66 (62)	Yes (NR)	Yes (4)	PSA ≥ 0.5 × 2	Yes	Yes	63% 5-yr dfs; 10-yr 41%	5/10-yr dfs 2–15% : 62%/38% 16+% : 42%/27%	5/10-yr dfs; 0-1% : 92%/65%
Moul et al. [4]	162	(54)	Yes (1)	NR	PSA > 0.2 × 2	Yes	No	38%	31% 6-yr dfs	72% 6-yr dfs
Bettencort et al. [12] (same patients as [4] above)	180	(53)	Yes (1)	NR	PSA > 0.2 × 2	Yes	No	60% 5-yr dfs	5-yr dfs; 1+ 69%; 2–4+ 44%	5-yr dfs; 83%
Vis et al. [15]	112	113	Yes (6)	No	Clinical only^@	Yes for clinical recurrence	No		Clinical dfs 5-yr 52% 10-yr 42%	Clinical dfs; 5-yr 75%; 10-yr 75%

NR: not reported.
^@Most studies include biopsy-proven local recurrence and radiographic distant metastasis as failure in addition to PSA.
*Three factors: Ki-67, SV+, margin+; A = 2-3 factors, B = one factor, C = all 3 negative.