Skip to main content
. 2017 Mar 15;31(4):315–323. doi: 10.1007/s12149-017-1159-2

Table 3.

Prognostic value of metabolic parameters on PET and histopathological tumour response for survival

n (%) Univariate Cox regression test Univariate Cox regression test
OS HR 95% CI p DFS HR 95% CI p
Baseline SUVmax
 SUV1max ≥9.70 20 (57.1) 1.000 0.044 1.000 0.030
 SUV1max < 9.70 15 (42.9) 2.589 1.025–6.535 2.797 1.107–7.071
Post-treatment SUVmax
 SUV2max ≥3.75 13 (44.8) 1.000 0.376 1.000 0.352
 SUV2max < 3.75 16 (55.2) 1.627 0.554–4.780 1.658 0.572–4.812
Metabolic response based on ∆SUVmax
 ∆SUVmax ≥ 5.75 17 (58.6) 1.000 0.112 1.000 0.066
 ∆SUVmax < 5.75 12 (41.2) 2.223 0.829–6.013 2.563 0.941–6.981
Metabolic response based on ROC analysis of ∆%SUVmax
 ∆%SUVmax ≥70% 12 (41.2) 1.000 0.063 1.000 0.028
 ∆%SUVmax <70% 17 (58.6) 2.956 0.945–9.247 3.614 1.150–11.533
Baseline MTV (cm3)
 MTV1 ≥47.30 18 (51.4) 1.000 0.141 1.000 0.133
 MTV1 < 47.30 17 (48.6) 1.996 0.795–5.013 2.031 0.806–5.118
Post-treatment MTV (cm3)
 MTV2 ≥12.00 10 (34.5) 1.000 0.817 1.000 0.795
 MTV2 < 12.00 16 (55.2) 0.860 0.241–3.074 0.845 0.238–3.003
 Missing 3 (10.3)
Metabolic response based on ∆MTV
 ∆MTV ≥39.40 16 (55.2) 1.000 0.305 1.000 0. 362
 ∆MTV < 39.40 10 (34.5) 1.182 0.583–5.639 1.698 0.544–5.296
 Missing 3 (10.3)
Metabolic response based on ∆% MTV
 ∆%MTV ≥80% 10 (34.5) 1.000 0.707 1.000 0. 681
 ∆%MTV < 80% 16 (55.2) 0.792 0.235–2.672 0.775 0.231–2.605
 Missing 3 (10.3)
Metabolic response based on various cut-offs of ∆%SUVmax
 ∆%SUVmax = 100% 4 (13.8) 1.000 1.000
 ∆%SUVmax ≥70–99% 8 (27.6) 0.402 0.056–2.898 0.366 0.454 0.063–3.247 0.431
 ∆%SUVmax ≥35–69% 14 (48.3) 1.552 0.331–7.727 0.577 2.137 0.460–9.922 0.332
 ∆%SUVmax <35% 3 (10.3) 3.241 0.429–24.495 0.254 3.718 0.491–28.183 0.204
Histopathological tumour response
 TRG1a–1b 16 (45.7) 1.000 1.000
 TRG2–3 19 (54.3) 8.461 2.355–30.396 0.001 6.385 2.019–20.195 0.002

Statistical significant result is in bold

Post-treatment scans in five patients were excluded from analysis due to oesophageal stent insertion. One patient was excluded due to significant difference in uptake times between the two scans

PET positron emission tomography, OS overall survival, DFS Disease-free survival, HR Hazard ratio, CI confidence interval, n number of patients, SUV1 max baseline SUVmax, SUV2 max post-treatment SUVmax, ∆SUV max absolute reduction in SUVmax, ∆%SUV max relative reduction in SUVmax, MTV metabolic tumour volume, MTV1 Baseline MTV, MTV2 post-treatment MTV, ∆MTV absolute reduction in MTV, ∆%MTV relative reduction in MTV, TRG tumour regression grade