Table 1.
Multivariate Cox-regression analysis reveal PRL-3 as an independent prognostic marker
| Cox-regression analysis of AML patient survival | ||||
|---|---|---|---|---|
| Univariate analysis | Multivariate analysis | |||
| Clinicopathological variables | Hazard ratio (95% CI) | p-Value | Hazard ratio (95% CI) | p-Value |
| Age (n = 221) | 1.034 (1.022–1.046) | <0.0001 | 1.032 (1.019–1.045) | <0.0001 |
| Sex | ||||
| Female (n = 112) | 1 | Reference | ||
| Male (n = 109) | 0.951 (0.681–1.326) | 0.765 | ||
| Cytogenetic risk group | ||||
| Favourable (n = 63) | 1 | Reference | 1 | Reference |
| Intermediate (n = 142) | 3.131 (1.950–5.026) | <0.0001 | 2.488 (1.458–4.244) | 0.001 |
| Adverse (n = 16) | 5.528 (2.817–10.847) | <0.0001 | 4.431 (2.067–9.499) | <0.0001 |
| Karyotype | ||||
| Normal (n = 101) | 1 | Reference | ||
| Others (n = 120) | 0.651 (0.465–0.911) | 0.012 | ||
| FLT3 mutation status | ||||
| Normal (n = 141) | 1 | Reference | 1 | Reference |
| TKD (n = 29) | 0.644 (0.358–1.156) | 0.141 | 0.665 (0.355–1.243) | 0.201 |
| ITD (n = 50) | 1.920 (1.322–2.789) | 0.001 | 1.697 (1.303–2.896) | 0.001 |
| ITD/TKD (n = 1) | 1.963 (0.272–14.155) | 0.504 | 1.004 (0.133–7.590) | 0.997 |
| NPM mutation status | ||||
| Normal (n = 163) | 1 | Reference | ||
| Mutant (n = 58) | 1.319 (0.917–1.897) | 0.136 | ||
| PRL-3 mRNA expression | 1.306 (1.041–1.639) | 0.021 | 1.577 (1.199–2.073) | 0.001 |
| FAB group (missing; n = 10) | ||||
| FAB group 1 (n = 50) | 1 | Reference | ||
| FAB group 2 (n = 64) | 0.736 (0.457–1.184) | 0.206 | ||
| FAB group 3 (n = 12) | 0.592 (0.231–1.519) | 0.276 | ||
| FAB group 4 (n = 44) | 0.662 (0.392–1.116) | 0.122 | ||
| FAB group 5 (n = 35) | 1.073 (0.640–1.799) | 0.789 | ||
| FAB group 6 (n = 4) | 1.504 (0.461–4.906) | 0.499 | ||
| FAB group 7 (n = 2) | 6.913 (1.598–29.903) | 0.010 | ||
PRL-3 acts as a novel prognostic marker in AML. By multivariate Cox-regression analysis using backward conditional stepwise method with a removal limit of p > 0.05, PRL-3 constituted one of the key independent predictors for poorer patient survival in our Cohort 1 (n = 221).