Skip to main content
. 2021 Aug 26;13(17):4291. doi: 10.3390/cancers13174291

Table 2.

Uni- and multivariable analyses of time to biochemical recurrence including Gleason grade groups assessed using routine Gleason scores.

Variable Group Univariable Analysis Multivariable Analysis *
HR (95% CI) p Value HR (95% CI) p Value
(A) Standard clinicopathologic parameters
Ploidy and PTEN status <0.0001 0.017
Diploid and PTEN-high ref. ref.
Non-diploid or PTEN-low 1.94 (1.15–3.30) 1.11 (0.62–1.99)
Non-diploid and PTEN-low 4.63 (2.50–8.57) 2.82 (1.34–5.94)
Age at surgery 10-year increment 1.51 (1.00–2.26) 0.048 0.98 (0.94–1.03) 0.51
Preoperative PSA log2(1 + ng/mL) increment 2.39 (1.74–3.28) <0.0001 2.44 (1.52–3.91) <0.0001
Gleason grade group <0.0001 0.017
1 (GS 6) ref. ref.
2 (GS 3 + 4) 8.27 (2.52–27.17) 3.97 (1.17–13.48)
3 (GS 4 + 3) 10.25 (3.02–34.79) 3.86 (1.07–13.88)
4 (GS 8) 22.91 (6.38–82.24) 6.51 (1.65–25.72)
5 (GS 9–10) 57.17 (15.64–208.92) 10.08 (2.43–41.79)
Extracapsular extension Present vs. Absent 3.97 (2.45–6.44) <0.0001 1.52 (0.84–2.75) 0.17
Surgical margins Positive vs. Negative 2.90 (1.80–4.67) <0.0001 2.09 (1.21–3.59) 0.008
Seminal vesicle invasion Present vs. Absent 5.39 (3.22–9.02) <0.0001 1.64 (0.88–3.06) 0.12
Lymph node involvement Present vs. Absent 6.25 (2.67–14.60) <0.0001 2.33 (0.89–6.14) 0.09
(B) CAPRA-S risk groups
Ploidy and PTEN status <0.0001 0.011
Diploid and PTEN-high ref. ref.
Non-diploid or PTEN-low 1.94 (1.15–3.30) 1.19 (0.68–2.07)
Non-diploid and PTEN-low 4.63 (2.50–8.57) 2.69 (1.39–5.21)
CAPRA-S risk group
Low (score 0–2) ref <0.0001 ref. <0.0001
Intermediate (score 3–5) 4.52 (1.85–11.01) 4.63 (1.89–11.31)
High (score ≥ 6) 17.44 (7.36–41.35) 14.68 (6.07–35.52)

CAPRA-S = Cancer of the Prostate Risk Assessment Postsurgical; CI = confidence interval; GS = Gleason score; HR = hazard ratio; PSA = prostate-specific antigen; PTEN = phosphatase and tensin homolog. * Of the 259 patients included in univariable analyses (71 with event and 188 without event), 251 (69 with event and 182 without event) had complete data and were included in the multivariable analysis.