Serum ferritin (SF) is a muti-functional protein, which is involved in cell proliferation, angiogenesis, immunosuppression and iron delivery.1 The elevation of SF is associated with iron overload and inflammation.2 Therefore, patients who undergo hematopoietic SCT (HSCT), a curative treatment for malignant and nonmalignant hematologic diseases, are at risk for excess accumulation of SF due to red blood cell transfusions and recent infection.3,4 Although emerging evidence suggests that there is a strong link between hyperferritinemia (HF) and inferior OS in patients receiving HSCT, the precise magnitude of the association remains uncertain, and it is not clear whether the HF is as an independent prognostic factor or simply reflects iron overload or acute-phase reactants. To comprehensively assess the prognostic significance of pretransplantation HF, we performed herein a meta-analysis.
All studies that investigated the association between pretransplantation HF and clinical outcomes in adults undergoing allogeneic HSCT published to date and included in the PubMed, EMBASE and Web of SCIENCE were considered. Information retrieved from the researches included author, publication year, study location, number of subjects, underlying disease, median follow-up, age range, conditioning regimen (myeloablative or reduced-intensity), cutoff value of HF, detection time before HSCT, covariates adjusted for HF and hazard ratio (HR) with 95% confidence intervals (CIs) for clinical outcomes (non-relapse mortality (NRM) and OS). To quantify the predictive ability of pretransplantation HF on clinical outcomes after HSCT, the HRs and their 95% CIs of NRM and OS were extracted from original articles. Heterogeneity test was assessed with I2 statistics and Q test.5 When the heterogeneity was high (P<0.05 or I2>50%), a random-effects model was used; otherwise, a fixed-effects model was applied. By convention, HR>1 indicates that the patients with pretransplantation HF are associated with a worse outcome or a higher risk as compared to those without the HF. Only when the 95% CIs did not overlap with 1, were the results statistically significant. The statistical calculations for this meta-analysis were performed with Stata Statistical Software Version 12.0 (StataCorp LP, College Station, TX, USA), with P-values of <0.05 considered statistically significant.
Eighteen articles were identified to assess the association between pretransplantation HF and OS or NRM (see Supplementary Table I). The total number of subjects was 1734 and the incidence of HF varied from 33.8% to 71.5%. There were four studies in which HR was adjusted for potential confounders including inflammatory markers such as C-reactive protein. Eight studies were adjusted for disease status. Studies that provided data on OS and NRM were used to calculate summary HRs. We did not find any evidence of publication bias for OS.
The summary HR for OS was 1.96 (95% CI: 1.73–2.23; P<0.001) for the overall number of patients undergoing HSCT (Supplementary Figure 1): 2.11 (95% CI: 1.73–2.58; P<0.001) in univariate analysis, 1.90 (95% CI: 1.54–2.36; P<0.001) in analysis not adjusted for inflammatory markers and 1.82 (95% CI: 1.42–2.34; P<0.001) in analysis adjusted for inflammatory marker by fixed-effects models; moreover, the meta-analysis of other subgroups showed similar results (see Table 1), indicating that the presence of pretransplantation HF is a deleterious and independent prognostic factor for OS. The test for heterogeneity in the overall analysis showed no significant heterogeneity (Q=8.10, df=17, I2=0%, P=0.964).
There were five studies that provided multivariate analysis of NRM data in patients undergoing allogeneic HSCT (see Supplementary Table I). The test of homogeneity showed that heterogeneity existed among the studies included (Q=10.05, df=4, I2=60.2%, P=0.04); hence a random-effects model was used. The result of the meta-analysis revealed a combined estimate HR of 1.59 (95% CI: 1.17–2.16, P<0.01), indicating that the pretransplantation HF was associated with a higher incidence of NRM (see Supplementary Figure 1B).
Meta-analysis is a useful statistical method for integrating the results from independent studies for special outcomes. However, as is often the case with meta-analyses, a substantial effect of heterogeneity needs to be taken into account. Although we did not find any significant heterogeneity in the OS analysis, the sample size, underlying diseases, stem cell source, type of HSCT, HLA matching status, conditioning regimen, definition of HF, detection time of SF and follow-up time were varied, which might contribute in part to the observed heterogeneity of NRM analysis due to the limited data. Therefore, it would be worthwhile further investigating these factors in order to arrive at a more accurate estimation of the prognostic significance of pretransplantation HF.
The meta-analysis reported here clearly identified an association between the pretransplantation HF and poor outcomes in patients with hematologic malignancies undergoing HSCT, which was independent of infection and disease status. To further stratify patients in clinical studies, a scoring system that consists of SF and disease status at transplantation might be valuable to predict the outcomes.6,7 From this study, we cannot deduce that the adverse prognostic impact of pretransplantation HF is associated with iron overload. To provide definitive answers regarding the role of iron overload in HSCT, we should in future, as Dr Armand suggested, perform studies that incorporate direct measurements of parenchymal and labile iron.8 Besides, there is another way to indirectly explore the relationship between HF and iron overload: by investigating the therapeutic effect of removing excess iron in patients with pretransplantation HF undergoing HSCT. Two such studies showed that the SF levels could be significantly reduced and that the OS and DFS could be significantly improved after the treatment of phlebotomy or iron chelation.9,10 As the sample sizes of the above two studies are small, large-scale randomized and controlled trials are needed in future. Although there are some limitations to this meta-analysis, based on the current findings, assessing the levels of pretransplantation SF could provide better prognostic information for adults with hematologic malignancies undergoing HSCT. Further large-scale cohort studies are needed to validate our results.
Alkhateeb AA, Connor JR The significance of ferritin in cancer: Anti-oxidation, inflammation and tumorigenesis. Biochim Biophys Acta 2013; 1836: 245–254.
Tachibana T, Tanaka M, Numata A, Takasaki H, Ito S, Ohshima R et al. Pretransplant serum ferritin has a prognostic influence on allogeneic transplant regardless of disease risk. Leuk Lymphoma 2012; 53: 456–461.
Storey JA, Connor RF, Lewis ZT, Hurd D, Pomper G, Keung YK et al. The transplant iron score as a predictor of stem cell transplant survival. J Hematol Oncol 2009; 2: 44.
Ozyilmaz E, Aydogdu M, Sucak G, Aki SZ, Ozkurt ZN, Yegin ZA et al. Risk factors for fungal pulmonary infections in hematopoietic stem cell transplantation recipients: the role of iron overload. Bone Marrow Transplant 2010; 45: 1528–1533.
Yusuf S, Peto R, Lewis J, Collins R, Sleight P Beta blockade during and after myocardial infarction: an overview of the randomized trials. Prog Cardiovasc Dis 1985; 27: 335–371.
Tanaka M, Tachibana T, Numata A, Takasaki H, Matsumoto K, Maruta A et al. A prognostic score with pretransplant serum ferritin and disease status predicts outcome following reduced-intensity SCT. Bone Marrow Transplant 2012; 47: 596–597.
Tachibana T, Takasaki H, Tanaka M, Maruta A, Hyo R, Ishigatsubo Y et al. Serum ferritin and disease status at transplantation predict the outcome of allo-SCT in patients with AML or myelodysplastic syndrome. Bone Marrow Transplant 2011; 46: 150–151.
Armand P Hyperferritinemia in stem cell transplantation. Biol Blood Marrow Transplant 2013; 19: 335–337.
Sivgin S, Baldane S, Akyol G, Keklik M, Kaynar L, Kurnaz F et al. The oral iron chelator deferasirox might improve survival in allogeneic hematopoietic cell transplant (alloHSCT) recipients with transfusional iron overload. Transfusion Apheresis Sci 2013; 49: 295–301.
Lee JW, Kang HJ, Kim EK, Kim H, Shin HY, Ahn HS Effect of iron overload and iron-chelating therapy on allogeneic hematopoietic SCT in children. Bone Marrow Transplant 2009; 44: 793–797.
This study was supported in part by Grants from the National Natural Science Foundation of China (No.: 30971283, No.: 81170502), the Zhejiang Provincial Natural Science Foundation of China (No.: LZ12H08001), the Creative Team Project of Zhejiang Provincial Bureau of Science and Technology (No.: 2011R50015) and Zhejiang Provincial Fund of Health Bureau (No.: 2007B122).
The authors declare no conflict of interest.
Supplementary Information accompanies this paper on Bone Marrow Transplantation website
About this article
Improved survival outcomes and restoration of graft-vs-leukemia effect by deferasirox after allogeneic stem cell transplantation in acute myeloid leukemia
Cancer Medicine (2019)
Blood Reviews (2019)
ATG vs thiotepa with busulfan and cyclophosphamide in matched-related bone marrow transplantation for thalassemia
Blood Advances (2017)
Potential anti-leukemic activity of iron chelation after allogeneic hematopoietic stem cell transplantation in patients with acute myeloid leukemia
Leukemia & Lymphoma (2017)