Leukemia & Lymphoma
ISSN: 1042-8194 (Print) 1029-2403 (Online) Journal homepage: http://www.tandfonline.com/loi/ilal20
Serum ferritin as prognostic marker in classical Hodgkin lymphoma treated with ABVD-based therapy Ruben Fernandez-Alvarez, Ana P. Gonzalez-Rodriguez, M. Esther Gonzalez, Arturo Rubio-Castro, Francisco Dominguez-Iglesias, Jackeline Solano, Eva Alonso-Nogues, Carmen Fernandez-Alvarez, Yahya Zanabili, Jose Manuel Alonso, Angel Ramirez Payer, Jose Maria Vicente, Jesus Medina & Juan M. Sancho To cite this article: Ruben Fernandez-Alvarez, Ana P. Gonzalez-Rodriguez, M. Esther Gonzalez, Arturo Rubio-Castro, Francisco Dominguez-Iglesias, Jackeline Solano, Eva Alonso-Nogues, Carmen Fernandez-Alvarez, Yahya Zanabili, Jose Manuel Alonso, Angel Ramirez Payer, Jose Maria Vicente, Jesus Medina & Juan M. Sancho (2015): Serum ferritin as prognostic marker in classical Hodgkin lymphoma treated with ABVD-based therapy, Leukemia & Lymphoma To link to this article: http://dx.doi.org/10.3109/10428194.2015.1038709
View supplementary material
Accepted author version posted online: 10 Apr 2015. Published online: 12 May 2015.
Submit your article to this journal
Article views: 37
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ilal20 Download by: [Chinese University of Hong Kong]
Date: 06 November 2015, At: 05:07
Leukemia & Lymphoma, 2015; Early Online: 1–7 © 2015 Informa UK, Ltd. ISSN: 1042-8194 print / 1029-2403 online DOI: 10.3109/10428194.2015.1038709
ORIGINAL ARTICLE: CLINICAL
Serum ferritin as prognostic marker in classical Hodgkin lymphoma treated with ABVD-based therapy
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
Ruben Fernandez-Alvarez1, Ana P. Gonzalez-Rodriguez2, M. Esther Gonzalez1, Arturo Rubio-Castro1, Francisco Dominguez-Iglesias1, Jackeline Solano2, Eva Alonso-Nogues2, Carmen Fernandez-Alvarez1, Yahya Zanabili3, Jose Manuel Alonso4, Angel Ramirez Payer2, Jose Maria Vicente5, Jesus Medina3 & Juan M. Sancho6 Hematology Departments of 1Hospital de Cabueñes, Gijon, 2Hospital Central de Asturias, Oviedo, 3Hospital San Agustin, Aviles, 4Hospital Valle del Nalon, Langreo, 5Hospital Alvarez Buylla, Mieres, and 6ICO-Hospital Germans Trias i Pujol, Jose Carreras Research Institute, Badalona, Universitat Autonoma de Barcelona, Spain
(IPS) [3]. However, this model is less suitable for patients with limited stage disease and it fails to identify a group of patients whose probabillity of cure is less than 50% [4,5]. In recent years, biological markers have gained interest to better risk-stratify patients at diagnosis. In cHL, the neoplastic Hodgkin and Reed-Sternberg cells represent only a minority of the total tumor mass and are outnumbered by a majority of inflammatory cells [6]. The composition of this microenvironment may predict clinical outcomes in cHL, as it has been demonstrated by recent gene-expression profiling studies [7–9]. The peripheral blood lymphocyte/monocyte ratio (LMR), a surrogate biomarker of the cellular content of the microenvironment, has been shown to be an useful prognostic parameter in cHL [10,11]. Nevertheless, these new biological markers still need clinical validation. Ferritin is the most widely used surrogate marker for systemic iron stores. However, it also operates as an acute-phase protein, and elevated levels of serum ferritin might be found in inflammatory conditions, infectious diseases and malignancies. Serum ferritin levels correlate with disease activity in patients with malignant lymphoma [12,13] and recent data suggest that elevated ferritin levels may be associated with poor outcomes in non-Hodgkin lymphoma (NHL) patients [14,15]. In cHL, alterations in iron metabolism occur as a consequence of upregulation of cytokines [16,17] and ferritin levels have been reported to be elevated in advanced stages and during disease progression [18,19]. However, no previous studies have reported a correlation between serum ferritin levels and prognosis in cHL in the modern treatment era. In this study, we analyzed the prognostic role of ferritin levels at diagnosis in cHL patients treated with ABVD-based chemotherapy.
Abstract Ferritin levels might correlate with disease activity in classical Hodgkin lymphoma (cHL). We analyzed the prognostic significance of the ferritin value at diagnosis in 173 cHL patients treated with ABVD between 2003 and 2013. The 5-year overall survival (OS) and progression-free survival (PFS) probabilities were 80% and 64%, respectively. Patients with ferritin 350 mg/l [high ferritin group (HF), n 62] were more likely to have advanced stage disease, B-symptoms and higher International Prognostic Score (IPS) compared with patients with ferritin 350 mg/l [low ferritin group (LF), n 111]. The complete remission (CR) rate and 5-year PFS and OS probabilities were lower in HF vs. LF patients (69% vs. 89%, p 0.025; 40% vs. 78%, p 0.001; 61% vs. 90%, p 0.001; respectively). Multivariate analysis revealed that advanced stage (p 0.001) and ferritin levels 350 mg/l (p 0.002) were independent predictors for PFS. In conclusion, the ferritin level at diagnosis is a useful prognostic marker for cHL. Keywords: Biomarker, ferritin, Hodgkin lymphoma, lymphoma, prognosis
Introduction Classical Hodgkin lymphoma (cHL) is a potentially curable disease, as therapy is successful in 80–90% of cases. However, up to 30% of patients will relapse after first-line therapy and another proportion may be overtreated [1]. Thus, it is essential to identify those patients with a high likelihood of treatment failure at the time of diagnosis. In clinical practice, stage of disease is the most important disease characteristic and is used to stratify treatment strategies [2]. Besides the stage, the most widely used stratification system is the International Prognostic Score
Correspondence: Dr Ruben Fernandez-Alvarez, MD, Department of Hematology, Hospital de Cabueñes, Los Prados 395, 33394, Gijon, Spain. Tel: 34 985 185005. E-mail: [email protected] Received 8 January 2015; revised 20 March 2015; accepted 1 April 2015
1
2 R. Fernandez-Alvarez et al.
Material and methods
participating center and the research was conducted in accordance with the Declaration of Helsinki.
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
Patients This was a retrospective study conducted in five centers from the Principality of Asturias (Spain), which has a total population of 1 080 138 inhabitants. Only confirmed cases of cHL, diagnosed between January 2003 and December 2013, with clinical, epidemiological and laboratory parameters available after a thorough chart review were included in this study. Histologic diagnoses were established according to the World Health Organization (WHO) classification [20]. We identified a total of 279 patients diagnosed with cHL during the study period. Serum ferritin levels were available at diagnosis for 196 cases. For the present analysis, only patients treated with ABVD plus/minus radiotherapy and with a negative human immunodeficiency virus (HIV) status were accepted. Twenty-three cases were excluded from the study: nine patients treated with BEACOPP as first-line therapy, seven cases treated with only radiation or palliative care, six patients positive for HIV infection and one patient with missing follow-up. Data for the remaining 173 patients, who form the patient population this report, were analyzed. Clinical and outcome data were retrospectively obtained from the files of the participant hospitals, and updated information was requested from local physicians. Approval for the retrospective review of these patients’ records was obtained from the ethics committee at each
Staging and laboratory evaluations Clinical staging was conducted according to the Ann Arbor system, using data from the following: history, physical examination, chest X-radiography, computed tomography of the chest, abdomen and pelvis, and bone marrow biopsy. For the definition of advanced stage we used the criteria from the German Hodgkin Study Group (GHSG): patients with Ann Arbor stage III, IV or stage IIB with either a large mediastinal mass or extranodal involvement. Laboratory evaluations included complete blood counts, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and serum albumin levels. Serum ferritin levels were measured in each participant hospital using a two-site chemiluminometric sandwich immunoassay (ADVIA Centaur Ferritin analyzer). The 95th percentile range of serum ferritin for healthy men is 22–322 mg/l and for healthy women is 10–291 mg/l. The LMR was estimated by dividing the absolute lymphocyte count by the absolute monocyte count from the complete blood count.
Treatment modality Treatment strategy differed according to the clinical stage. All patients were treated with the ABVD (doxorubicin 25 mg/m2/ day, days 1 and 15; bleomycin 10 mg/m2/day, days 1 and 15;
Table I. Baseline characteristics based on the serum ferritin value at diagnosis. Characteristic Median age (range), years Male, n (%) Histology, n (%) Nodular sclerosis Mixed cellularity Lymphocyte-rich Lymphocyte-depleted Unclassified Ann Arbor stage, n (%) I II III IV B symptoms, n (%) ESR 50 mm/h, n (%) Mediastinal bulky mass, n (%) Bone marrow involvement, n (%) Extranodal disease, n (%) Stage, n (%) Localized Advanced IPS risk factors, n (%) Male Albumin 40 g/l Age 45 Hemoglobin 10.5 g/dl Stage IV ALC 0.6 109/l WBC 15 109/l IPS 3, n (%) Treatment, n (%) ABVD ABVD radiotherapy
All cases (n 173)
Ferritin 350 mg/l (HF, n 62)
Ferritin 350 mg/l (LF, n 111)
p-value
38 (15–77) 116 (67)
48 (19–77) 52 (84)
33 (15–71) 64 (58)
0.001 0.001
90 (52) 62 (35) 7 (4) 3 (2) 11 (6)
34 (55) 19 (31) 1 (2) 3 (5) 5 (85)
56 (51) 43 (39) 6 (5) 0 (0) 6 (5)
0.360
26 (15) 54 (31) 57 (33) 36 (21) 91 (53) 66 (38) 25 (14) 10 (6) 39 (22)
1 (2) 15 (24) 22 (35) 24 (39) 52 (84) 37 (59) 8 (13) 8 (13) 24 (39)
25 (22) 39 (35) 35 (32) 12 (11) 39 (35) 29 (26) 17 (15) 2 (2) 15 (13)
0.001
72 (42) 101 (58)
13 (21) 49 (79)
59 (53) 52 (47)
0.001
116 (67) 116 (67) 62 (36) 40 (23) 36 (21) 34 (20) 21 (12) 71 (41)
52 (84) 50 (81) 35 (56) 32 (52) 24 (39) 20 (33) 10 (16) 46 (74)
64 (58) 66 (62) 27 (24) 8 (7) 12 (11) 14 (13) 11 (10) 25 (22)
0.001 0.009 0.001 0.001 0.001 0.002 0.001 0.001
88 (51) 85 (49)
41 (66) 21 (34)
47 (42) 64 (58)
0.001
0.001 0.001 0.424 0.005 0.001
F, high ferritin group; LF, low ferritin group; ESR, erythrocyte sedimentation rate; IPS, International Prognostic Score; ALC, absolute lymphocyte count; WBC, white H blood cell count.
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
Ferritin in classical Hodgkin lymphoma 3
Figure 2. (A) Progression-free survival (PFS) and (B) overall survival (OS) of the overall series according to the serum ferritin value at diagnosis. HF, high ferritin group (ferritin 350 mg/l); LF, low ferritin group (ferritin 350 mg/l).
Figure 1. Receiver operating characteristic curve (ROC) and area under the curve (AUC) of the association between serum ferritin levels at diagnosis and (A) progression-free survival (PFS) and (B) overall survival (OS). 95% CI, 95% confidence interval.
vinblastine 6 mg/m2/day, days 1 and 15; and dacarbazine 375 mg/m2/day, days 1 and 15) regimen. In addition 85 patients (49%) received radiotherapy, 70 of them had early stage and received involved-field radiotherapy, while 15 patients with advanced stage disease received radiotherapy due to initial bulky disease or residual mass after chemotherapy. Tumor responses were classified as complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD) according to the International Workshop Criteria [21].
Statistical analysis Comparison of baseline patient characteristics according to ferritin level was performed by Pearson’s Chi-square, Fisher’s exact, Student’s t-test or Mann-Whitney U test, as appropriate. Correlation coefficients (r) between serum ferritin levels and other continuous variables were determined using the Spearman rank correlation test. The choice
of the best cut-off value for the ferritin to assess survival was based on its utility as a marker for the clinically relevant binary outcome of death/survival using the receiver operating characteristic curve (ROC) and area under the curve (AUC) analysis. Definitions for progression-free survival (PFS) and overall survival (OS) were based on the guidelines from the International Harmonization Project Lymphoma [21]. PFS was defined as the time from diagnosis to the time of progression, relapse from complete remission, death as a result of any cause or last follow-up. OS was defined as the time from diagnosis to death as a result of any cause or last
Table II. Correlations between serum ferritin levels and parameters of inflammation and disease activity in classical Hodgkin lymphoma (cHL) patients. Ferritin Parameter
r-value
p-value
Hemoglobin Serum albumin ESR CRP WBC ALC LMR
0.376 0.395 0.254 0.469 0.029 0.135 0.062
0.001 0.001 0.001 0.004 0.695 0.077 0.432
pearman rank correlation coefficients (r) were calculated for each pair of S variables. ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; WBC, white blood cell count; ALC, absolute lymphocyte count; LMR, lymphocyte/monocyte ratio.
4 R. Fernandez-Alvarez et al. Table III. Univariate and multivariate analysis for the progression-free survival (PFS). Univariate analysis Variable Advanced stage B symptoms IPS 3 Ferritin 350 mg/l LMR 1.1 ESR 50 mm/h Age 45 Male sex Albumin 40 g/l Hemoglobin 10.5 g/dl ALC 0.6 109/l WBC 15 109/l
Multivariate analysis
HR (95% CI)
p-value
HR (95% CI)
p-value
0.187 (0.087–0.401) 0.436 (0.238–0.783) 0.374 (0.179–0.781) 0.252 (0.142–0.447) 0.689 (0.362–1.213) 0.805 (0.451–1.436) 0.363 (0.207–0.636) 0.789 (0.425–1.462) 0.443 (0.280––0.944) 0.368 (0.208–0.649) 0.605 (0.324–1.127) 0.773 (0.363–1.647)
0.001 0.006 0.009 0.001 0.118 0.462 0.001 0.452 0.035 0.001 0.113 0.504
0.266 (0.120–0.591) 0.656 (0.271–1.261) 0.741 (0.382–1.613) 0.386 (0.212–0.704)
0.001 0.182 0.407 0.002
1.591 (0.961–2.586)
0.059
0.923 (0.405–2.091) 1.346 (0.782–2.215)
0.817 0.123
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
FS, progression-free survival; HR, hazard ratio; 95% CI, 95% confidence interval; IPS, International Prognostic Score; LMR, P lymphocyte/monocyte ratio; ESR, erythrocyte sedimentation rate; RT, radiation; ALC, absolute lymphocyte count; WBC, white blood cell count.
follow-up. The estimation of survival was performed by the Kaplan-Meier method and were compared by the log-rank test [22,23]. Univariate and multivariate analyses for PFS and OS were performed using the Cox proportional hazard regression model [24]. Only significant statistical variables in univariate analysis were included in the multivariate Cox regression model. Two-sided p-values less than 0.05 were considered as statistically significant. The statistical package SPSS, version 15.0 (SPSS Inc., Chicago, IL, USA) was used for all analyses.
Results
(51%) patients were treated only with ABVD, while 85 (49%) received ABVD plus radiotherapy. Overall, 147 patients (85%) achieved CR and 15 (9%) achieved PR; 11 patients (6%) had progressive disease. Relapse occurred in 26 of the 147 patients (18%) who had achieved CR. With a median follow-up of the series of 41.4 months (range: 1–147 months), the 5-year OS and PFS probabilities for the entire cohort were 80% and 64%, respectively. In total, 25 patients died: 18 patients died due to relapse or progression of lymphoma and seven patients died of unrelated lymphoma causes.
Ferritin levels in cHL patients
Atotal of 173 newly diagnosed cHL patients were included in this study. Median age at diagnosis was 38 years (ranging from 15–77) and 116 (67%) were male. The clinical characteristics of these patients are summarized in Table I. The main histologic subtypes were nodular sclerosis (52%) and mixed cellularity (35%). According to the Ann Arbor staging system, 80 patients (46%) had stage I/II disease, while 93 patients (54%) had stage III/IV disease. A total of 91 patients (53%) presented B symptoms and 25 (14%) had bulky mediastinal mass. Bone marrow involvement was detected in 10 cases (6%). Altogether, 101 patients (58%) presented with advanced stage according to GHSG criteria and 71 (41%) were classified as having high risk IPS (3–7). A total of 88
The median serum ferritin level at the time of diagnosis was 241 mg/l (range: 8–3969 mg/l). Anemia, defined as Hb concentration 12 g/dl, was present in 85 (49%) patients. Looking at the mechanism of anemia, we found an inverse correlation between ferritin and Hb levels (r 0.38, p 0.001). No patients displayed characteristics of iron-deficiency anemia, as defined by Hb levels 12 g/dl, microcytosis and ferritin levels 15 mg/l. Ferritin levels correlated with several laboratory parameters, as listed in Table II. We examined the relationship between ferritin and other proteins of the acutephase reaction. We found statistically significant correlations with serum albumin (r 0.39, p 0.001), ESR (r 0.25; p 0.001) and CRP (r 0.469; p 0.004) (Table II).
Table IV. Univariate and multivariate analysis for the overall survival (OS). Univariate analysis Variable Advanced stage B symptoms IPS 3 Ferritin 350 mg/l LMR 1.1 ESR 50 mm/h Age 45 Male sex Albumin 40 g/l Hemoglobin 10.5 g/dl ALC 0.6 109/l WBC 15 109/l
Multivariate analysis
HR (95% CI)
p-value
HR (95% CI)
p-value
0.133 (0.040–0.439) 0.388 (0.172–0.876) 0.172 (0.052–0.573) 0.222 (0.103–0.480) 0.350 (0.132–0.809) 1.181 (0.562–2.478) 0.172 (0.076–0.391) 0.944 (0.429–2.075) 0.518 (0.197–1.360) 0.294 (0.142–0.610) 0.361 (0.168–0.770) 0.884 (0.307–2.545)
0.001 0.023 0.004 0.001 0.017 0.661 0.001 0.866 0.182 0.001 0.008 0.822
0.168 (1.050–3.560) 1.251 (0.741–2.156) 0.791 (0.468–1.412) 0.512 (0.219–1.515) 1.051 (0.957–2.143)
0.004 0.861 0.324 0.129 0.781
0.212 (1.093–4.483)
0.001
0.670 (0.324–1.967) 0.814 (0.609–1.546)
0.221 0.586
S, overall survival; HR, hazard ratio; 95% CI, 95% confidence interval; IPS, International Prognostic Score; LMR, lymphocyte/ O monocyte ratio; ESR, erythrocyte sedimentation rate; RT, radiation; ALC, absolute lymphocyte count; WBC, white blood cell count.
Ferritin in classical Hodgkin lymphoma 5
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
111 patients (64%) were classified into the low ferritin group (LF, ferritin 350 mg/l). Patients in the HF group were more likely to present with advanced stage (p 0.001), B symptoms (p 0.001), extranodal involvement (p 0.001), including bone marrow involvement (p 0.005), and an IPS score 3 (p 0.001), than those in the LF group. Furthermore, a higher frequency of each factor comprising the IPS was observed in the HF vs. LF group (Table I). The CR rate for patients in the HF group was significantly lower than in the LF group (70% vs. 90%, respectively; p 0.025). Similarly, patients in the HF group experienced inferior PFS (Figure 2A) and OS (Figure 2B) compared with those in the LF group. The 5-year PFS and OS rates for patients in the HF vs. LF group were 40% vs. 78% (p 0.001) and 61% vs. 90% (p 0.001), respectively.
Univariate and multivariate analyses
Figure 3. Comparison of progression-free survival (PFS) based on the ferritin level at diagnosis according to localized (A) or advanced (B) stage. HF, high ferritin group (ferritin 350 mg/l); LF, low ferritin group (ferritin 350 mg/l).
The ROC curve analysis showed the appropriate cut-off value of serum ferritin level to be 350 mg/l for predicting PFS [AUC, 0.712; 95% confidence interval (CI) of 0.623–0.801, p 0.001; Figure 1A] and OS (AUC, 0.700; 95% CI of 0.587– 0.811, p 0.001; Figure 1B). When female patients were evaluated separately from males, as the normal range of serum ferritin is lower, there were no differences with males, possibly because other prognostic features were differently distributed between males and females. Based on this result, we selected a serum ferritin of 350 mg/l as the optimal cut-off point for survival analysis in the entire cohort. Two groups of patients were considered according to the serum ferritin level (Table I). A total of 62 patients (36%) were categorized as having high ferritin levels (HF, ferritin 350 mg/l), whereas
In the univariate analysis, the clinical factors that had a negative impact on both PFS and OS were: advanced stage, age 45 years, the presence of B symptoms, hemoglobin level 105 g/l, IPS 3, serum ferritin levels 350 mg/l and treatment with chemotherapy alone (Tables III and IV). Multivariate analysis revealed that advance stage [hazard ratio (HR) 0.266, 95% CI of 0.120–0.591, p 0.001] and serum ferritin levels 350 mg/l (HR 0.386, 95% CI of 0.212–0.704, p 0.002) were independent prognostic factors for PFS. Regarding OS, only age 45 years (HR 0.212, 95% CI of 0.093–0.483, p0.001) and advanced stage (HR 0.168, 95% CI of 0.050–0.560, p 0.004) were associated with a significant adverse influence. However, considering patients younger than 65 years old, the only factor that remained significant for the OS was ferritin levels 350 mg/l (HR 0.143, 95% CI of 0.044–0.469, p 0.001), while no other risk factor showed a significant association (data not shown). As stage and ferritin levels were the two significant independent predictors of survival, we analyzed the prognostic impact of the ferritin levels for patients with limited and advanced stage patients separately (Figure 3). For advanced stage patients the ferritin level was able to discriminate clinical outcomes: the median progression-free survival for patients with ferritin 350 mg/l was 2.2 years compared with 6.1 years for patients with ferritin 350 mg/l; the 5-year PFS rates were 26% (95% CI, 17–40%) vs. 63% (95% CI, 50–80%) respectively, p 0.002.
Discussion cHL is characterized by the presence of a minority of neoplastic Hodgkin-Reed Sternberg cells, surrounded by a majority of reactive inflammatory cells. This composition is dependant on the release of soluble factors, such as cytokines and chemokines, which are responsible for the development of systemic symptoms and laboratory abnormalities [25]. In our study, we found that ferritin levels strongly correlate with laboratory changes that are typical for inflammation and reflect disease activity in cHL, such as anemia, hypoalbuminemia, and elevated ESR and CRP levels. Therefore, serum ferritin levels may reflect the inflammatory activity of the cHL microenvironment. The striking inverse correlation
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
6 R. Fernandez-Alvarez et al. with the hemoglobin level observed in this study illustrates the inflammatory mechanism of anemia in cHL. This is in accordance with previous reports of a similar association for hepcidin [26]. Hepcidin levels are significantly elevated in cHL patients, as a result of inflammatory activity; in particular, interleukin-6 upregulates hepcidin. Elevated hepcidin levels block the release of iron from the reticuloendothelial system that results in anemia of chronic disease, with typical changes in iron metabolism, such as elevated ferritin [16,17]. In addition, we found that an elevated serum ferritin level ( 350 mg/l) was significantly associated with: advanced stage disease, the presence of B symptoms, bone marrow involvement and extranodal disease. These are traditional prognostic factors in cHL, which indicate high tumor burden and disease activity. There was also a significant association with higher risk IPS and with each individual prognostic factor comprising this score. This association between elevated ferritin levels and unfavourable prognostic factors suggests that ferritin may have a role as a biomarker, and ferritin levels may affect both the CR rate and PFS. Furthermore, we observed that a serum ferritin level 350 mg/l is an independent predictor for poor PFS in patients with newly diagnosed cHL. In the multivariate analysis, the only two independent prognostic factors for poor PFS were elevated levels of serum ferritin and the presence of advanced stage disease. The prognostic influence of advanced stage is in line with previous reports and with more recent data where advanced stage stands out as the main prognostic clinical factor [27]. However, serum ferritin levels have not been documented as a prognostic marker in cHL. When patients with limited and advanced stage were analyzed separately, we found that the ferritin level was able to discriminate clinical outcomes only in patients with advanced stage disease, while its impact in limited stage disease did not reach statistical significance. This fact may be explained by the small number of patients with high ferritin levels in this group (only 13 patients). A remarkable finding of our study is that the combination of advanced stage and elevated ferritin levels allows to identify a subgroup of patients (accounting for 28% of the overall series) with very poor outcome (5-year PFS of only 26%). Thus, there is a distinct group of patients with advanced stage cHL that might be identified as being at very high risk on the basis of simple clinical features. In our analysis, ferritin stands out as the most important parameter for acute-phase reaction to predict cHL outcomes. In contrast, other inflammatory markers, such as hypoalbuminemia, elevated ESR, anemia or white-cell counts abnormalities (leukocytosis, lymphocytopenia, altered LMR) did not predict patient outcomes in the multivariate analysis. As stated before, ferritin is linked to the inflammatory activity of the cHL microenvironment, and this might explain its big prognostic impact. Previous studies have demonstrated that ferritin inhibits immune responses in vivo, although the mechanisms and the signaling pathways that mediate this immunomodulatory activity are unknown [28,29]. In line with these observa-
tions, it may be speculated that ferritin may exert adverse effects on the immune response in the cHL microenvironment, which enable the tumor cells to evade the immune system. In the present series the prognostic effect of IPS could not be reproduced. In the multivariate analysis, a high IPS (over 2 points) was not significant for outcome. A possible explanation is that our study included patients with localized stage disease, where IPS is less suitable for prognostic purposes [4]. However, recent studies suggest that the predictive range of the IPS has narrowed due to improved outcomes of patients with current therapies [5]. Furthermore, in our study, multivariate analysis did not confirm any prognostic impact of lymphocytopenia or the LMR. This finding differs from previous reports and possibly reflects differences in the population studied [10,11]. The retrospective nature of the present study constitutes its main limitation. The decision to determine ferritin levels depended solely on the criteria of the attending physician, therefore it cannot be ruled out selection bias. It can be hypothesized that the presence of some features at the time of diagnosis, like anemia, B symptoms or older age, could influence this decision. However, in the present series the proportion of patients with anemia, defined as Hb levels 12 g/dl (n 85, 49%) and B symptoms (n 91, 53%) is comparable to the original Hasenclever series [3]. Our population-based cohort shows a higher age (median age 38 years, 22% over 65 years) when compared to other cohorts of patients with cHL. This fact may reflect demographic factors that differ from different studied populations [30]. To minimize the inherent biases of a retrospective study, we included patients homogeneously treated (ABVD plus/minus RT) and excluded patients treated up-front with BEACOPP, palliative care or radiation therapy alone. Patients who were positive for HIV were also excluded as HIV infection directly influences ferritin value. Despite the mentioned limitations, our results are concordant with those from previous cohorts of cHL patients investigating prognostic factors in the modern treatment era [5,27]. A further limitation of our study is that analysis of cytokine levels were not performed. As ferritin is a non-specific biological variable, it is important to obtain sufficient data on more specific features, including serum cytokine levels and microenvironment data. Future research should correlate ferritin levels, serum cytokine levels and microenvironment data with outcomes. In conclusion, our study suggests that high ferritin levels are associated with clinical features of aggressive disease and with a poor prognosis in patients with cHL. We recommend considering serum ferritin levels before treatment initiation, as an inexpensive, widely available and easily reproducible biomarker. However, validation of this finding in a larger number of cases is needed. Potential conflict of interest: Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.
Ferritin in classical Hodgkin lymphoma 7 This work was supported in part by Spanish grant of Fondo de Investigaciones Sanitarias (Instituto Carlos III), PS09/00420 and PI12/01280.
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
References [1] Kuruvilla J, Keating A, Crump M. How I treat relapsed and refractory Hodgkin lymphoma. Blood 2011;117:4208–4217. [2] Follows GA, Ardeshna KM, Barrington SF, et al. Guidelines for the first line management of classical Hodgkin lymphoma. Br J Haematol 2014;166:34–49. [3] Hasenclever D, Diehl V. A prognostic score for advanced Hodgkin’s disease: International Prognostic Factors Project on Advanced Hodgkin’s Disease. N Engl J Med 1998;339:1506–1514. [4] Franklin J, Paulus U, Lieerz D, et al. Is the international prognostic score for advanced stage Hodgkin’s disease applicable to early stage patients? Annals Oncol 2000;11:617–623. [5] Moccia AA, Donaldson J, Chhanabhai M, et al. International Prognostic Score (IPS) in advanced-stage Hodgkin’s lymphoma; altered utility in the modern era. J Clin Oncol 2012;30:3383–3388. [6] Harris NL, Jaffe ES, Stein H, et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994;84:1361–1392. [7] Steidl C, Lee T, Shah SP, et al. Tumor-associated macrophages and survival in classic Hodgkin’s lymphoma. N Engl J Med 2010;362: 875–885. [8] Steidl C, Connors JM, Gascoyne RD. Molecular pathogenesis of Hodgkin’s lymphoma: increasing evidence of the importance of the microenvironment. J Clin Oncol 2011;29:1812–1826. [9] Scott DW, Chan FC, Hong F, et al. Gene expression-based model using formalin-fixed paraffin-embedded biopsies predicts overall survival in advanced-stage classical Hodgkin lymphoma. J Clin Oncol 2013;31:692–700. [10] Porrata LF, Ristow K, Colgan JP, et al. Peripheral blood lymphocyte/monocyte ratio at diagnosis and survival in classical Hodgkin’s lymphoma. Haematologica 2011;97:262–269. [11] Koh YW, Kang HJ, Park C, et al. The ratio of the absolute lymphocyte count to the absolute monocyte count is associated with prognosis in Hodgkin’s lymphoma: correlation with tumor-associated macrophages. Oncologist 2012;17:871–880. [12] Matzner Y, Konijn AM, Hershko C. Serum ferritin in hematologic malignancies. Am J Hematol 1980;9:13–22. [13] Zhang XZ, Su AL, Hu MQ, et al. Elevated serum ferritin levels in patients with hematological malignancies. Asian Pac J Cancer Prev 2014;15:6099–6101.
[14] Yamazaki E, Tomita N, Koyama S, et al. Serum ferritin level is prognostic of patient outcome in extranodal NK/T cell lymphoma, nasal type. Med Oncol 2014;31:149. [15] Yoh KA, Lee HS, Park LC, et al. The prognostic significance of elevated levels of serum ferritin before chemotherapy in patients with non-Hodgkin lymphoma. Clin Lymphoma Myeloma Leuk 2014; 14:43–49. [16] Hohaus S, Massini G, Giachelia M, et al. Anemia in Hodgkin’s lymphoma: the role of interleukin-6 and hepcidin. J Clin Oncol 2010;28:2538–2543. [17] Hohaus S, Giachelia M, Cuccaro A, et al. Iron in Hodgkin’s lymphoma. Crit Rev Oncog 2013;18:463–469. [18] Dörner MH, Abel U, Fritze D, et al. Serum ferritin in relation to the course of Hodgkin’s disease. Cancer 1983;52:2308–2312. [19] Bezwoda WR, Derman DP, Bothwell TH, et al. Serum ferritin and Hodgkin’s disease. Scand J Haematol 1985;35:505–510. [20] Swerdlow SH, Campo E, Harris N, et al. WHO Classification of tumours of haematopoietic and lymphoid Tissues, 4th ed. Lyon, France: IARC; 2008. [21] Cheson BD, Pfistner B, Juweis ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol 2007;25:579–586. [22] Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Statist Assoc 1958;53:457–481. [23] Peto R, Pike MC. Conservatism of the approximation Σ(O-E)2/E in the log-rank test for survival data or tumor incidence data. Biometrics 1973;29:579–584. [24] Cox DR. Regression models and life-tables. J Roy Stat Soc Series B 1972;34:187–202. [25] Skinnider BF, Mak TW. The role of cytokines in classical Hodgkin lymphoma. Blood 2002;99:4283–4297. [26] Ganz T. Hepcidin, a key regulator of iron metabolism and mediator of anemia of inflammation. Blood 2003;102:783–788. [27] Guisado-Vasco P, Arranz-Saez R, Canales M, et al. Stage IV and age over 45 years are the only prognostic factors of the International Prognostic Score for the outcome of advanced Hodgkin lymphoma in the Spanish Hodgkin Lymphoma Study Group series. Leuk Lymphoma 2012;53:812–819. [28] Recalcati S, Invernizzi P, Arosio P, Cairo G. New functions for an iron storage protein: the role of ferritin in immunity and autoimmunity. J Autoimmun 2008;30:84–89. [29] Wang W, Knovich MA, Coffman LG, et al. Serum ferritin: Past, present and future. Biochim Biophys Acta 2010;1800:760–769. [30] Diepstra A, van Imhoff GW, Schaapveld M, et al. Latent EpsteinBarr virus infection of tumor cells in classical Hodgkin’s lymphoma predicts adverse outcome in older adult patients. J Clin Oncol 2009;27:3815–3821.
ISSN: 1042-8194 (Print) 1029-2403 (Online) Journal homepage: http://www.tandfonline.com/loi/ilal20
Serum ferritin as prognostic marker in classical Hodgkin lymphoma treated with ABVD-based therapy Ruben Fernandez-Alvarez, Ana P. Gonzalez-Rodriguez, M. Esther Gonzalez, Arturo Rubio-Castro, Francisco Dominguez-Iglesias, Jackeline Solano, Eva Alonso-Nogues, Carmen Fernandez-Alvarez, Yahya Zanabili, Jose Manuel Alonso, Angel Ramirez Payer, Jose Maria Vicente, Jesus Medina & Juan M. Sancho To cite this article: Ruben Fernandez-Alvarez, Ana P. Gonzalez-Rodriguez, M. Esther Gonzalez, Arturo Rubio-Castro, Francisco Dominguez-Iglesias, Jackeline Solano, Eva Alonso-Nogues, Carmen Fernandez-Alvarez, Yahya Zanabili, Jose Manuel Alonso, Angel Ramirez Payer, Jose Maria Vicente, Jesus Medina & Juan M. Sancho (2015): Serum ferritin as prognostic marker in classical Hodgkin lymphoma treated with ABVD-based therapy, Leukemia & Lymphoma To link to this article: http://dx.doi.org/10.3109/10428194.2015.1038709
View supplementary material
Accepted author version posted online: 10 Apr 2015. Published online: 12 May 2015.
Submit your article to this journal
Article views: 37
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ilal20 Download by: [Chinese University of Hong Kong]
Date: 06 November 2015, At: 05:07
Leukemia & Lymphoma, 2015; Early Online: 1–7 © 2015 Informa UK, Ltd. ISSN: 1042-8194 print / 1029-2403 online DOI: 10.3109/10428194.2015.1038709
ORIGINAL ARTICLE: CLINICAL
Serum ferritin as prognostic marker in classical Hodgkin lymphoma treated with ABVD-based therapy
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
Ruben Fernandez-Alvarez1, Ana P. Gonzalez-Rodriguez2, M. Esther Gonzalez1, Arturo Rubio-Castro1, Francisco Dominguez-Iglesias1, Jackeline Solano2, Eva Alonso-Nogues2, Carmen Fernandez-Alvarez1, Yahya Zanabili3, Jose Manuel Alonso4, Angel Ramirez Payer2, Jose Maria Vicente5, Jesus Medina3 & Juan M. Sancho6 Hematology Departments of 1Hospital de Cabueñes, Gijon, 2Hospital Central de Asturias, Oviedo, 3Hospital San Agustin, Aviles, 4Hospital Valle del Nalon, Langreo, 5Hospital Alvarez Buylla, Mieres, and 6ICO-Hospital Germans Trias i Pujol, Jose Carreras Research Institute, Badalona, Universitat Autonoma de Barcelona, Spain
(IPS) [3]. However, this model is less suitable for patients with limited stage disease and it fails to identify a group of patients whose probabillity of cure is less than 50% [4,5]. In recent years, biological markers have gained interest to better risk-stratify patients at diagnosis. In cHL, the neoplastic Hodgkin and Reed-Sternberg cells represent only a minority of the total tumor mass and are outnumbered by a majority of inflammatory cells [6]. The composition of this microenvironment may predict clinical outcomes in cHL, as it has been demonstrated by recent gene-expression profiling studies [7–9]. The peripheral blood lymphocyte/monocyte ratio (LMR), a surrogate biomarker of the cellular content of the microenvironment, has been shown to be an useful prognostic parameter in cHL [10,11]. Nevertheless, these new biological markers still need clinical validation. Ferritin is the most widely used surrogate marker for systemic iron stores. However, it also operates as an acute-phase protein, and elevated levels of serum ferritin might be found in inflammatory conditions, infectious diseases and malignancies. Serum ferritin levels correlate with disease activity in patients with malignant lymphoma [12,13] and recent data suggest that elevated ferritin levels may be associated with poor outcomes in non-Hodgkin lymphoma (NHL) patients [14,15]. In cHL, alterations in iron metabolism occur as a consequence of upregulation of cytokines [16,17] and ferritin levels have been reported to be elevated in advanced stages and during disease progression [18,19]. However, no previous studies have reported a correlation between serum ferritin levels and prognosis in cHL in the modern treatment era. In this study, we analyzed the prognostic role of ferritin levels at diagnosis in cHL patients treated with ABVD-based chemotherapy.
Abstract Ferritin levels might correlate with disease activity in classical Hodgkin lymphoma (cHL). We analyzed the prognostic significance of the ferritin value at diagnosis in 173 cHL patients treated with ABVD between 2003 and 2013. The 5-year overall survival (OS) and progression-free survival (PFS) probabilities were 80% and 64%, respectively. Patients with ferritin 350 mg/l [high ferritin group (HF), n 62] were more likely to have advanced stage disease, B-symptoms and higher International Prognostic Score (IPS) compared with patients with ferritin 350 mg/l [low ferritin group (LF), n 111]. The complete remission (CR) rate and 5-year PFS and OS probabilities were lower in HF vs. LF patients (69% vs. 89%, p 0.025; 40% vs. 78%, p 0.001; 61% vs. 90%, p 0.001; respectively). Multivariate analysis revealed that advanced stage (p 0.001) and ferritin levels 350 mg/l (p 0.002) were independent predictors for PFS. In conclusion, the ferritin level at diagnosis is a useful prognostic marker for cHL. Keywords: Biomarker, ferritin, Hodgkin lymphoma, lymphoma, prognosis
Introduction Classical Hodgkin lymphoma (cHL) is a potentially curable disease, as therapy is successful in 80–90% of cases. However, up to 30% of patients will relapse after first-line therapy and another proportion may be overtreated [1]. Thus, it is essential to identify those patients with a high likelihood of treatment failure at the time of diagnosis. In clinical practice, stage of disease is the most important disease characteristic and is used to stratify treatment strategies [2]. Besides the stage, the most widely used stratification system is the International Prognostic Score
Correspondence: Dr Ruben Fernandez-Alvarez, MD, Department of Hematology, Hospital de Cabueñes, Los Prados 395, 33394, Gijon, Spain. Tel: 34 985 185005. E-mail: [email protected] Received 8 January 2015; revised 20 March 2015; accepted 1 April 2015
1
2 R. Fernandez-Alvarez et al.
Material and methods
participating center and the research was conducted in accordance with the Declaration of Helsinki.
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
Patients This was a retrospective study conducted in five centers from the Principality of Asturias (Spain), which has a total population of 1 080 138 inhabitants. Only confirmed cases of cHL, diagnosed between January 2003 and December 2013, with clinical, epidemiological and laboratory parameters available after a thorough chart review were included in this study. Histologic diagnoses were established according to the World Health Organization (WHO) classification [20]. We identified a total of 279 patients diagnosed with cHL during the study period. Serum ferritin levels were available at diagnosis for 196 cases. For the present analysis, only patients treated with ABVD plus/minus radiotherapy and with a negative human immunodeficiency virus (HIV) status were accepted. Twenty-three cases were excluded from the study: nine patients treated with BEACOPP as first-line therapy, seven cases treated with only radiation or palliative care, six patients positive for HIV infection and one patient with missing follow-up. Data for the remaining 173 patients, who form the patient population this report, were analyzed. Clinical and outcome data were retrospectively obtained from the files of the participant hospitals, and updated information was requested from local physicians. Approval for the retrospective review of these patients’ records was obtained from the ethics committee at each
Staging and laboratory evaluations Clinical staging was conducted according to the Ann Arbor system, using data from the following: history, physical examination, chest X-radiography, computed tomography of the chest, abdomen and pelvis, and bone marrow biopsy. For the definition of advanced stage we used the criteria from the German Hodgkin Study Group (GHSG): patients with Ann Arbor stage III, IV or stage IIB with either a large mediastinal mass or extranodal involvement. Laboratory evaluations included complete blood counts, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and serum albumin levels. Serum ferritin levels were measured in each participant hospital using a two-site chemiluminometric sandwich immunoassay (ADVIA Centaur Ferritin analyzer). The 95th percentile range of serum ferritin for healthy men is 22–322 mg/l and for healthy women is 10–291 mg/l. The LMR was estimated by dividing the absolute lymphocyte count by the absolute monocyte count from the complete blood count.
Treatment modality Treatment strategy differed according to the clinical stage. All patients were treated with the ABVD (doxorubicin 25 mg/m2/ day, days 1 and 15; bleomycin 10 mg/m2/day, days 1 and 15;
Table I. Baseline characteristics based on the serum ferritin value at diagnosis. Characteristic Median age (range), years Male, n (%) Histology, n (%) Nodular sclerosis Mixed cellularity Lymphocyte-rich Lymphocyte-depleted Unclassified Ann Arbor stage, n (%) I II III IV B symptoms, n (%) ESR 50 mm/h, n (%) Mediastinal bulky mass, n (%) Bone marrow involvement, n (%) Extranodal disease, n (%) Stage, n (%) Localized Advanced IPS risk factors, n (%) Male Albumin 40 g/l Age 45 Hemoglobin 10.5 g/dl Stage IV ALC 0.6 109/l WBC 15 109/l IPS 3, n (%) Treatment, n (%) ABVD ABVD radiotherapy
All cases (n 173)
Ferritin 350 mg/l (HF, n 62)
Ferritin 350 mg/l (LF, n 111)
p-value
38 (15–77) 116 (67)
48 (19–77) 52 (84)
33 (15–71) 64 (58)
0.001 0.001
90 (52) 62 (35) 7 (4) 3 (2) 11 (6)
34 (55) 19 (31) 1 (2) 3 (5) 5 (85)
56 (51) 43 (39) 6 (5) 0 (0) 6 (5)
0.360
26 (15) 54 (31) 57 (33) 36 (21) 91 (53) 66 (38) 25 (14) 10 (6) 39 (22)
1 (2) 15 (24) 22 (35) 24 (39) 52 (84) 37 (59) 8 (13) 8 (13) 24 (39)
25 (22) 39 (35) 35 (32) 12 (11) 39 (35) 29 (26) 17 (15) 2 (2) 15 (13)
0.001
72 (42) 101 (58)
13 (21) 49 (79)
59 (53) 52 (47)
0.001
116 (67) 116 (67) 62 (36) 40 (23) 36 (21) 34 (20) 21 (12) 71 (41)
52 (84) 50 (81) 35 (56) 32 (52) 24 (39) 20 (33) 10 (16) 46 (74)
64 (58) 66 (62) 27 (24) 8 (7) 12 (11) 14 (13) 11 (10) 25 (22)
0.001 0.009 0.001 0.001 0.001 0.002 0.001 0.001
88 (51) 85 (49)
41 (66) 21 (34)
47 (42) 64 (58)
0.001
0.001 0.001 0.424 0.005 0.001
F, high ferritin group; LF, low ferritin group; ESR, erythrocyte sedimentation rate; IPS, International Prognostic Score; ALC, absolute lymphocyte count; WBC, white H blood cell count.
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
Ferritin in classical Hodgkin lymphoma 3
Figure 2. (A) Progression-free survival (PFS) and (B) overall survival (OS) of the overall series according to the serum ferritin value at diagnosis. HF, high ferritin group (ferritin 350 mg/l); LF, low ferritin group (ferritin 350 mg/l).
Figure 1. Receiver operating characteristic curve (ROC) and area under the curve (AUC) of the association between serum ferritin levels at diagnosis and (A) progression-free survival (PFS) and (B) overall survival (OS). 95% CI, 95% confidence interval.
vinblastine 6 mg/m2/day, days 1 and 15; and dacarbazine 375 mg/m2/day, days 1 and 15) regimen. In addition 85 patients (49%) received radiotherapy, 70 of them had early stage and received involved-field radiotherapy, while 15 patients with advanced stage disease received radiotherapy due to initial bulky disease or residual mass after chemotherapy. Tumor responses were classified as complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD) according to the International Workshop Criteria [21].
Statistical analysis Comparison of baseline patient characteristics according to ferritin level was performed by Pearson’s Chi-square, Fisher’s exact, Student’s t-test or Mann-Whitney U test, as appropriate. Correlation coefficients (r) between serum ferritin levels and other continuous variables were determined using the Spearman rank correlation test. The choice
of the best cut-off value for the ferritin to assess survival was based on its utility as a marker for the clinically relevant binary outcome of death/survival using the receiver operating characteristic curve (ROC) and area under the curve (AUC) analysis. Definitions for progression-free survival (PFS) and overall survival (OS) were based on the guidelines from the International Harmonization Project Lymphoma [21]. PFS was defined as the time from diagnosis to the time of progression, relapse from complete remission, death as a result of any cause or last follow-up. OS was defined as the time from diagnosis to death as a result of any cause or last
Table II. Correlations between serum ferritin levels and parameters of inflammation and disease activity in classical Hodgkin lymphoma (cHL) patients. Ferritin Parameter
r-value
p-value
Hemoglobin Serum albumin ESR CRP WBC ALC LMR
0.376 0.395 0.254 0.469 0.029 0.135 0.062
0.001 0.001 0.001 0.004 0.695 0.077 0.432
pearman rank correlation coefficients (r) were calculated for each pair of S variables. ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; WBC, white blood cell count; ALC, absolute lymphocyte count; LMR, lymphocyte/monocyte ratio.
4 R. Fernandez-Alvarez et al. Table III. Univariate and multivariate analysis for the progression-free survival (PFS). Univariate analysis Variable Advanced stage B symptoms IPS 3 Ferritin 350 mg/l LMR 1.1 ESR 50 mm/h Age 45 Male sex Albumin 40 g/l Hemoglobin 10.5 g/dl ALC 0.6 109/l WBC 15 109/l
Multivariate analysis
HR (95% CI)
p-value
HR (95% CI)
p-value
0.187 (0.087–0.401) 0.436 (0.238–0.783) 0.374 (0.179–0.781) 0.252 (0.142–0.447) 0.689 (0.362–1.213) 0.805 (0.451–1.436) 0.363 (0.207–0.636) 0.789 (0.425–1.462) 0.443 (0.280––0.944) 0.368 (0.208–0.649) 0.605 (0.324–1.127) 0.773 (0.363–1.647)
0.001 0.006 0.009 0.001 0.118 0.462 0.001 0.452 0.035 0.001 0.113 0.504
0.266 (0.120–0.591) 0.656 (0.271–1.261) 0.741 (0.382–1.613) 0.386 (0.212–0.704)
0.001 0.182 0.407 0.002
1.591 (0.961–2.586)
0.059
0.923 (0.405–2.091) 1.346 (0.782–2.215)
0.817 0.123
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
FS, progression-free survival; HR, hazard ratio; 95% CI, 95% confidence interval; IPS, International Prognostic Score; LMR, P lymphocyte/monocyte ratio; ESR, erythrocyte sedimentation rate; RT, radiation; ALC, absolute lymphocyte count; WBC, white blood cell count.
follow-up. The estimation of survival was performed by the Kaplan-Meier method and were compared by the log-rank test [22,23]. Univariate and multivariate analyses for PFS and OS were performed using the Cox proportional hazard regression model [24]. Only significant statistical variables in univariate analysis were included in the multivariate Cox regression model. Two-sided p-values less than 0.05 were considered as statistically significant. The statistical package SPSS, version 15.0 (SPSS Inc., Chicago, IL, USA) was used for all analyses.
Results
(51%) patients were treated only with ABVD, while 85 (49%) received ABVD plus radiotherapy. Overall, 147 patients (85%) achieved CR and 15 (9%) achieved PR; 11 patients (6%) had progressive disease. Relapse occurred in 26 of the 147 patients (18%) who had achieved CR. With a median follow-up of the series of 41.4 months (range: 1–147 months), the 5-year OS and PFS probabilities for the entire cohort were 80% and 64%, respectively. In total, 25 patients died: 18 patients died due to relapse or progression of lymphoma and seven patients died of unrelated lymphoma causes.
Ferritin levels in cHL patients
Atotal of 173 newly diagnosed cHL patients were included in this study. Median age at diagnosis was 38 years (ranging from 15–77) and 116 (67%) were male. The clinical characteristics of these patients are summarized in Table I. The main histologic subtypes were nodular sclerosis (52%) and mixed cellularity (35%). According to the Ann Arbor staging system, 80 patients (46%) had stage I/II disease, while 93 patients (54%) had stage III/IV disease. A total of 91 patients (53%) presented B symptoms and 25 (14%) had bulky mediastinal mass. Bone marrow involvement was detected in 10 cases (6%). Altogether, 101 patients (58%) presented with advanced stage according to GHSG criteria and 71 (41%) were classified as having high risk IPS (3–7). A total of 88
The median serum ferritin level at the time of diagnosis was 241 mg/l (range: 8–3969 mg/l). Anemia, defined as Hb concentration 12 g/dl, was present in 85 (49%) patients. Looking at the mechanism of anemia, we found an inverse correlation between ferritin and Hb levels (r 0.38, p 0.001). No patients displayed characteristics of iron-deficiency anemia, as defined by Hb levels 12 g/dl, microcytosis and ferritin levels 15 mg/l. Ferritin levels correlated with several laboratory parameters, as listed in Table II. We examined the relationship between ferritin and other proteins of the acutephase reaction. We found statistically significant correlations with serum albumin (r 0.39, p 0.001), ESR (r 0.25; p 0.001) and CRP (r 0.469; p 0.004) (Table II).
Table IV. Univariate and multivariate analysis for the overall survival (OS). Univariate analysis Variable Advanced stage B symptoms IPS 3 Ferritin 350 mg/l LMR 1.1 ESR 50 mm/h Age 45 Male sex Albumin 40 g/l Hemoglobin 10.5 g/dl ALC 0.6 109/l WBC 15 109/l
Multivariate analysis
HR (95% CI)
p-value
HR (95% CI)
p-value
0.133 (0.040–0.439) 0.388 (0.172–0.876) 0.172 (0.052–0.573) 0.222 (0.103–0.480) 0.350 (0.132–0.809) 1.181 (0.562–2.478) 0.172 (0.076–0.391) 0.944 (0.429–2.075) 0.518 (0.197–1.360) 0.294 (0.142–0.610) 0.361 (0.168–0.770) 0.884 (0.307–2.545)
0.001 0.023 0.004 0.001 0.017 0.661 0.001 0.866 0.182 0.001 0.008 0.822
0.168 (1.050–3.560) 1.251 (0.741–2.156) 0.791 (0.468–1.412) 0.512 (0.219–1.515) 1.051 (0.957–2.143)
0.004 0.861 0.324 0.129 0.781
0.212 (1.093–4.483)
0.001
0.670 (0.324–1.967) 0.814 (0.609–1.546)
0.221 0.586
S, overall survival; HR, hazard ratio; 95% CI, 95% confidence interval; IPS, International Prognostic Score; LMR, lymphocyte/ O monocyte ratio; ESR, erythrocyte sedimentation rate; RT, radiation; ALC, absolute lymphocyte count; WBC, white blood cell count.
Ferritin in classical Hodgkin lymphoma 5
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
111 patients (64%) were classified into the low ferritin group (LF, ferritin 350 mg/l). Patients in the HF group were more likely to present with advanced stage (p 0.001), B symptoms (p 0.001), extranodal involvement (p 0.001), including bone marrow involvement (p 0.005), and an IPS score 3 (p 0.001), than those in the LF group. Furthermore, a higher frequency of each factor comprising the IPS was observed in the HF vs. LF group (Table I). The CR rate for patients in the HF group was significantly lower than in the LF group (70% vs. 90%, respectively; p 0.025). Similarly, patients in the HF group experienced inferior PFS (Figure 2A) and OS (Figure 2B) compared with those in the LF group. The 5-year PFS and OS rates for patients in the HF vs. LF group were 40% vs. 78% (p 0.001) and 61% vs. 90% (p 0.001), respectively.
Univariate and multivariate analyses
Figure 3. Comparison of progression-free survival (PFS) based on the ferritin level at diagnosis according to localized (A) or advanced (B) stage. HF, high ferritin group (ferritin 350 mg/l); LF, low ferritin group (ferritin 350 mg/l).
The ROC curve analysis showed the appropriate cut-off value of serum ferritin level to be 350 mg/l for predicting PFS [AUC, 0.712; 95% confidence interval (CI) of 0.623–0.801, p 0.001; Figure 1A] and OS (AUC, 0.700; 95% CI of 0.587– 0.811, p 0.001; Figure 1B). When female patients were evaluated separately from males, as the normal range of serum ferritin is lower, there were no differences with males, possibly because other prognostic features were differently distributed between males and females. Based on this result, we selected a serum ferritin of 350 mg/l as the optimal cut-off point for survival analysis in the entire cohort. Two groups of patients were considered according to the serum ferritin level (Table I). A total of 62 patients (36%) were categorized as having high ferritin levels (HF, ferritin 350 mg/l), whereas
In the univariate analysis, the clinical factors that had a negative impact on both PFS and OS were: advanced stage, age 45 years, the presence of B symptoms, hemoglobin level 105 g/l, IPS 3, serum ferritin levels 350 mg/l and treatment with chemotherapy alone (Tables III and IV). Multivariate analysis revealed that advance stage [hazard ratio (HR) 0.266, 95% CI of 0.120–0.591, p 0.001] and serum ferritin levels 350 mg/l (HR 0.386, 95% CI of 0.212–0.704, p 0.002) were independent prognostic factors for PFS. Regarding OS, only age 45 years (HR 0.212, 95% CI of 0.093–0.483, p0.001) and advanced stage (HR 0.168, 95% CI of 0.050–0.560, p 0.004) were associated with a significant adverse influence. However, considering patients younger than 65 years old, the only factor that remained significant for the OS was ferritin levels 350 mg/l (HR 0.143, 95% CI of 0.044–0.469, p 0.001), while no other risk factor showed a significant association (data not shown). As stage and ferritin levels were the two significant independent predictors of survival, we analyzed the prognostic impact of the ferritin levels for patients with limited and advanced stage patients separately (Figure 3). For advanced stage patients the ferritin level was able to discriminate clinical outcomes: the median progression-free survival for patients with ferritin 350 mg/l was 2.2 years compared with 6.1 years for patients with ferritin 350 mg/l; the 5-year PFS rates were 26% (95% CI, 17–40%) vs. 63% (95% CI, 50–80%) respectively, p 0.002.
Discussion cHL is characterized by the presence of a minority of neoplastic Hodgkin-Reed Sternberg cells, surrounded by a majority of reactive inflammatory cells. This composition is dependant on the release of soluble factors, such as cytokines and chemokines, which are responsible for the development of systemic symptoms and laboratory abnormalities [25]. In our study, we found that ferritin levels strongly correlate with laboratory changes that are typical for inflammation and reflect disease activity in cHL, such as anemia, hypoalbuminemia, and elevated ESR and CRP levels. Therefore, serum ferritin levels may reflect the inflammatory activity of the cHL microenvironment. The striking inverse correlation
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
6 R. Fernandez-Alvarez et al. with the hemoglobin level observed in this study illustrates the inflammatory mechanism of anemia in cHL. This is in accordance with previous reports of a similar association for hepcidin [26]. Hepcidin levels are significantly elevated in cHL patients, as a result of inflammatory activity; in particular, interleukin-6 upregulates hepcidin. Elevated hepcidin levels block the release of iron from the reticuloendothelial system that results in anemia of chronic disease, with typical changes in iron metabolism, such as elevated ferritin [16,17]. In addition, we found that an elevated serum ferritin level ( 350 mg/l) was significantly associated with: advanced stage disease, the presence of B symptoms, bone marrow involvement and extranodal disease. These are traditional prognostic factors in cHL, which indicate high tumor burden and disease activity. There was also a significant association with higher risk IPS and with each individual prognostic factor comprising this score. This association between elevated ferritin levels and unfavourable prognostic factors suggests that ferritin may have a role as a biomarker, and ferritin levels may affect both the CR rate and PFS. Furthermore, we observed that a serum ferritin level 350 mg/l is an independent predictor for poor PFS in patients with newly diagnosed cHL. In the multivariate analysis, the only two independent prognostic factors for poor PFS were elevated levels of serum ferritin and the presence of advanced stage disease. The prognostic influence of advanced stage is in line with previous reports and with more recent data where advanced stage stands out as the main prognostic clinical factor [27]. However, serum ferritin levels have not been documented as a prognostic marker in cHL. When patients with limited and advanced stage were analyzed separately, we found that the ferritin level was able to discriminate clinical outcomes only in patients with advanced stage disease, while its impact in limited stage disease did not reach statistical significance. This fact may be explained by the small number of patients with high ferritin levels in this group (only 13 patients). A remarkable finding of our study is that the combination of advanced stage and elevated ferritin levels allows to identify a subgroup of patients (accounting for 28% of the overall series) with very poor outcome (5-year PFS of only 26%). Thus, there is a distinct group of patients with advanced stage cHL that might be identified as being at very high risk on the basis of simple clinical features. In our analysis, ferritin stands out as the most important parameter for acute-phase reaction to predict cHL outcomes. In contrast, other inflammatory markers, such as hypoalbuminemia, elevated ESR, anemia or white-cell counts abnormalities (leukocytosis, lymphocytopenia, altered LMR) did not predict patient outcomes in the multivariate analysis. As stated before, ferritin is linked to the inflammatory activity of the cHL microenvironment, and this might explain its big prognostic impact. Previous studies have demonstrated that ferritin inhibits immune responses in vivo, although the mechanisms and the signaling pathways that mediate this immunomodulatory activity are unknown [28,29]. In line with these observa-
tions, it may be speculated that ferritin may exert adverse effects on the immune response in the cHL microenvironment, which enable the tumor cells to evade the immune system. In the present series the prognostic effect of IPS could not be reproduced. In the multivariate analysis, a high IPS (over 2 points) was not significant for outcome. A possible explanation is that our study included patients with localized stage disease, where IPS is less suitable for prognostic purposes [4]. However, recent studies suggest that the predictive range of the IPS has narrowed due to improved outcomes of patients with current therapies [5]. Furthermore, in our study, multivariate analysis did not confirm any prognostic impact of lymphocytopenia or the LMR. This finding differs from previous reports and possibly reflects differences in the population studied [10,11]. The retrospective nature of the present study constitutes its main limitation. The decision to determine ferritin levels depended solely on the criteria of the attending physician, therefore it cannot be ruled out selection bias. It can be hypothesized that the presence of some features at the time of diagnosis, like anemia, B symptoms or older age, could influence this decision. However, in the present series the proportion of patients with anemia, defined as Hb levels 12 g/dl (n 85, 49%) and B symptoms (n 91, 53%) is comparable to the original Hasenclever series [3]. Our population-based cohort shows a higher age (median age 38 years, 22% over 65 years) when compared to other cohorts of patients with cHL. This fact may reflect demographic factors that differ from different studied populations [30]. To minimize the inherent biases of a retrospective study, we included patients homogeneously treated (ABVD plus/minus RT) and excluded patients treated up-front with BEACOPP, palliative care or radiation therapy alone. Patients who were positive for HIV were also excluded as HIV infection directly influences ferritin value. Despite the mentioned limitations, our results are concordant with those from previous cohorts of cHL patients investigating prognostic factors in the modern treatment era [5,27]. A further limitation of our study is that analysis of cytokine levels were not performed. As ferritin is a non-specific biological variable, it is important to obtain sufficient data on more specific features, including serum cytokine levels and microenvironment data. Future research should correlate ferritin levels, serum cytokine levels and microenvironment data with outcomes. In conclusion, our study suggests that high ferritin levels are associated with clinical features of aggressive disease and with a poor prognosis in patients with cHL. We recommend considering serum ferritin levels before treatment initiation, as an inexpensive, widely available and easily reproducible biomarker. However, validation of this finding in a larger number of cases is needed. Potential conflict of interest: Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.
Ferritin in classical Hodgkin lymphoma 7 This work was supported in part by Spanish grant of Fondo de Investigaciones Sanitarias (Instituto Carlos III), PS09/00420 and PI12/01280.
Downloaded by [Chinese University of Hong Kong] at 05:08 06 November 2015
References [1] Kuruvilla J, Keating A, Crump M. How I treat relapsed and refractory Hodgkin lymphoma. Blood 2011;117:4208–4217. [2] Follows GA, Ardeshna KM, Barrington SF, et al. Guidelines for the first line management of classical Hodgkin lymphoma. Br J Haematol 2014;166:34–49. [3] Hasenclever D, Diehl V. A prognostic score for advanced Hodgkin’s disease: International Prognostic Factors Project on Advanced Hodgkin’s Disease. N Engl J Med 1998;339:1506–1514. [4] Franklin J, Paulus U, Lieerz D, et al. Is the international prognostic score for advanced stage Hodgkin’s disease applicable to early stage patients? Annals Oncol 2000;11:617–623. [5] Moccia AA, Donaldson J, Chhanabhai M, et al. International Prognostic Score (IPS) in advanced-stage Hodgkin’s lymphoma; altered utility in the modern era. J Clin Oncol 2012;30:3383–3388. [6] Harris NL, Jaffe ES, Stein H, et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994;84:1361–1392. [7] Steidl C, Lee T, Shah SP, et al. Tumor-associated macrophages and survival in classic Hodgkin’s lymphoma. N Engl J Med 2010;362: 875–885. [8] Steidl C, Connors JM, Gascoyne RD. Molecular pathogenesis of Hodgkin’s lymphoma: increasing evidence of the importance of the microenvironment. J Clin Oncol 2011;29:1812–1826. [9] Scott DW, Chan FC, Hong F, et al. Gene expression-based model using formalin-fixed paraffin-embedded biopsies predicts overall survival in advanced-stage classical Hodgkin lymphoma. J Clin Oncol 2013;31:692–700. [10] Porrata LF, Ristow K, Colgan JP, et al. Peripheral blood lymphocyte/monocyte ratio at diagnosis and survival in classical Hodgkin’s lymphoma. Haematologica 2011;97:262–269. [11] Koh YW, Kang HJ, Park C, et al. The ratio of the absolute lymphocyte count to the absolute monocyte count is associated with prognosis in Hodgkin’s lymphoma: correlation with tumor-associated macrophages. Oncologist 2012;17:871–880. [12] Matzner Y, Konijn AM, Hershko C. Serum ferritin in hematologic malignancies. Am J Hematol 1980;9:13–22. [13] Zhang XZ, Su AL, Hu MQ, et al. Elevated serum ferritin levels in patients with hematological malignancies. Asian Pac J Cancer Prev 2014;15:6099–6101.
[14] Yamazaki E, Tomita N, Koyama S, et al. Serum ferritin level is prognostic of patient outcome in extranodal NK/T cell lymphoma, nasal type. Med Oncol 2014;31:149. [15] Yoh KA, Lee HS, Park LC, et al. The prognostic significance of elevated levels of serum ferritin before chemotherapy in patients with non-Hodgkin lymphoma. Clin Lymphoma Myeloma Leuk 2014; 14:43–49. [16] Hohaus S, Massini G, Giachelia M, et al. Anemia in Hodgkin’s lymphoma: the role of interleukin-6 and hepcidin. J Clin Oncol 2010;28:2538–2543. [17] Hohaus S, Giachelia M, Cuccaro A, et al. Iron in Hodgkin’s lymphoma. Crit Rev Oncog 2013;18:463–469. [18] Dörner MH, Abel U, Fritze D, et al. Serum ferritin in relation to the course of Hodgkin’s disease. Cancer 1983;52:2308–2312. [19] Bezwoda WR, Derman DP, Bothwell TH, et al. Serum ferritin and Hodgkin’s disease. Scand J Haematol 1985;35:505–510. [20] Swerdlow SH, Campo E, Harris N, et al. WHO Classification of tumours of haematopoietic and lymphoid Tissues, 4th ed. Lyon, France: IARC; 2008. [21] Cheson BD, Pfistner B, Juweis ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol 2007;25:579–586. [22] Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Statist Assoc 1958;53:457–481. [23] Peto R, Pike MC. Conservatism of the approximation Σ(O-E)2/E in the log-rank test for survival data or tumor incidence data. Biometrics 1973;29:579–584. [24] Cox DR. Regression models and life-tables. J Roy Stat Soc Series B 1972;34:187–202. [25] Skinnider BF, Mak TW. The role of cytokines in classical Hodgkin lymphoma. Blood 2002;99:4283–4297. [26] Ganz T. Hepcidin, a key regulator of iron metabolism and mediator of anemia of inflammation. Blood 2003;102:783–788. [27] Guisado-Vasco P, Arranz-Saez R, Canales M, et al. Stage IV and age over 45 years are the only prognostic factors of the International Prognostic Score for the outcome of advanced Hodgkin lymphoma in the Spanish Hodgkin Lymphoma Study Group series. Leuk Lymphoma 2012;53:812–819. [28] Recalcati S, Invernizzi P, Arosio P, Cairo G. New functions for an iron storage protein: the role of ferritin in immunity and autoimmunity. J Autoimmun 2008;30:84–89. [29] Wang W, Knovich MA, Coffman LG, et al. Serum ferritin: Past, present and future. Biochim Biophys Acta 2010;1800:760–769. [30] Diepstra A, van Imhoff GW, Schaapveld M, et al. Latent EpsteinBarr virus infection of tumor cells in classical Hodgkin’s lymphoma predicts adverse outcome in older adult patients. J Clin Oncol 2009;27:3815–3821.
