Introduction

Myelofibrosis (MF) is a myeloproliferative neoplasm (MPN) comprising primary myelofibrosis (PMF) and myelofibrosis developing from either polycythemia vera (post-PV MF) or essential thrombocythemia (post-ET MF).

Methods

Two hundred and sixty-seven consecutive patients with MF (PMF, N=206); post-PV MF, N=26; post-ET MF, N=35) diagnosed between January 1988 and December 2013 at seven regional hospitals in Hong Kong were prospectively followed up. Data on the clinicopathologic features and treatment outcome were collected.

Results

The median duration of follow-up was 45 (1 – 247) months. The median overall survival (OS) was 65 months (95% confidence interval, CI: 56.3–73.7). The 5-year and 10-year OS were 52.2% and 26.7% respectively. On univariate analysis, factors associated with an inferior OS included age > 65 years (hazard ratio ,HR, =1.81; 95% CI:1.33–2.48; P<0.001), the presence of constitutional symptoms (HR=1.52; 95% CI:1.11–2.08; P=0.009), hemoglobin < 10 g/dL (HR=1.71; 95% CI:1.25–2.35; P=0.001), circulating blasts ≥ 1% (HR=1.55; 95% CI:1.13–2.16; P=0.007), platelet < 100 x 109/L (HR=2.51; 95% CI:1.79–3.53; P<0.001), high risk IPSS (HR=3.09; 95% CI: 1.78–5.37; P<0.001), intermediate-2 risk DIPSS (HR=2.37; 95% CI: 1.38–4.06; P=0.002), high risk DIPSS (HR=2.92; 95% CI:1.51–5.64; P=0.001), transfusion dependence within the first year of diagnosis (HR=2.61; 95% CI:1.92–3.55; P<0.001), transfusion dependence after the first year of diagnosis (HR=2.42; 95% CI:1.61–3.54; P<0.001), palpable hepatomegaly at diagnosis (HR=1.44; 95% CI:1.06–1.96; P=0.02) and secondary AML (HR=1.75; 95% CI:1.24–2.47; P=0.001). On multivariate analysis, post-PV MF (P=0.03), platelet < 100 × 109/L (P=0.001), high risk IPSS (P=0.009), transfusion dependence within the first year (P=0.001), transfusion dependence after the first year (P=0.02), and transformation to secondary acute myeloid leukemia (AML) (P=0.007) were independent risks associated with inferior OS.

On univariate analysis, factors associated with increased risk of secondary AML include age ≤ 55 years (odds ratio [OR] = 2.61; 95% CI:1.33–5.12; P=0.005), circulating blasts ≥ 1% (OR=2.24; 95% CI:1.18–4.26; P=0.01), transfusion dependence within the first year (OR=5.57; 95%:2.16–14.88; P<0.001), transfusion dependence after the first year (OR=5.57; 95% CI:2.16–14.88; P<0.001), hepatomegaly (OR=4.05; 95% CI:2.03–8.10), splenomegaly (OR=3.71; 95% CI:1.09–9.26; P=0.04), abnormal karyotypes (OR=3.3; 95% CI:1.14–9.56; P=0.02), and the presence of unfavourable karyotypes (OR=4.9; 95% CI: 1.22–19.96). On multivariate analysis, transfusion dependence within the first year of diagnosis (P=0.04), transfusion dependence after the first year of diagnosis (P=0.003) and hepatomegaly (P=0.006) were independent risks associated with secondary AML.

The 5-year and 10-year risks of leukemic transformation were 17.1% and 29.7 respectively. Factors associated with inferior leukemia-free survival (LFS) on univariate analysis included post-ET MF (HR=2.15; 95% CI: 1.03–4.50; P=0.04), presence of constitutional symptoms (HR=1.85; 95% CI:1.04–3.31; P=0.04), circulating blasts ≥ 1% (HR=2.89; 95% CI:1.62–5.16; P<0.001), platelet count < 100 x 109/L (HR=2.56; 95% CI: 1.35–4.84; P=0.004), transfusion dependence within the first year of diagnosis (HR=3.05; 95% CI: 1.70–5.50; P<0.001), transfusion dependence after the first year of diagnosis (HR=6.49; 95% CI: 2.33–18.10; P<0.001), hepatomegaly (HR=3.21; 95% CI:1.69–6.08; P<0.001) and unfavorable karyotype (HR=3.01; 95% CI:1.08–8.35; P=0.03). On multivariate analysis, male gender (P=0.05), presence of constitutional symptoms (P=0.04) and unfavorable karyotypes (P=0.01) were independent risks associated with inferior LFS.

Eighteen patients underwent allogeneic haematopoietic stem cell transplantation (HSCT) (matched sibling, N=14; matched-unrelated, N=4), with 15 patients achieving complete remission. Seven patients relapsed with subsequent progression to secondary AML. Acute and chronic graft-versus-host disease occurred in seven (39.9%) and six (33.3%) patients respectively. Transplant-related mortality occurred in three patients. The 5-year and 10-year OS following HSCT was 51.5%.

Conclusion

Findings of this study complement current prognostic models in guiding treatment decisions at diagnosis and during the course of MF.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution