Schwager, Susan M.

Abstract Background: Allogeneic hematopoietic stem cell transplantation is potentially curative in agnogenic myeloid metaplasia (AMM) but is associated with substantial mortality and morbidity that necessitates accurate identification of patients in whom benefit outweighs risk. The current single institutional study investigates prognostic variables in transplant-eligible patients with AMM with the main objective of improved discrimination between intermediate- and high-risk patient categories. Methods: Patients diagnosed with AMM before the age of 60 years and seen at Mayo Clinic were identified and the diagnosis confirmed. Relevant demographic, clinical and laboratory characteristics were abstracted and the impact of various parameters on overall survival was evaluated with univariate and multivariate analysis. Results: A cohort of 159 patients (median age 52 years, range 18–60; 89 males) with AMM is described. Median follow-up from initial diagnosis was 63 months (range 0–300). During this period, 102 patients have died; overall median survival 79 months. Multivariate analysis of parameters measured in all study patients at diagnosis identified thrombocytopenia (platelet count < 100 x 109/L) as the strongest predictor of inferior survival (p=0.002). In addition, a hemoglobin level of <10 g/dL (p=0.003), white blood cell count of either <4 or >30 x 109/L (p=0.03), and older age (p=0.02) were also found to be independent indicators of poor prognosis. However, when the analysis included parameters that were measured in variable proportion of the study population, the independent prognostic factors for poor survival were thrombocytopenia (p=0.0001), anemia (p=0.01), and the presence of unfavorable cytogenetic abnormalities (0.001). Based on the above findings, we constructed a new complete blood count (CBC)-based prognostic scoring system (Figure 1) that performed better than the Dupriez scoring system in discriminating intermediate- from high-risk patient categories(Figure 2). Figure Figure Conclusions: Thrombocytopenia is a strong predictor of poor survival in transplant-eligible patients with AMM. The incorporation of platelet count into the Dupriez prognostic scoring system might allow the construction of an improved, CBC-based scoring system that can accurately identify high-risk as well as intermediate-risk patients with AMM.

Abstract Background: In essential thrombocythemia (ET), homozygosity for the JAK2 V617F mutant allele is rare despite the frequent (23–57%) occurrence of the heterozygous state. This is in direct contrast with the situation in polycythemia vera (PV) where homozygosity for the mutation is seen in 25–33% of the cases. The particular observation is consistent with what has been observed in vitro as well as in animal studies where JAK2 V617F has been associated with induction of erythropoietin hypersensitivity in cell lines and erythrocytosis in mice (James et al. Nature2005;434:1144). Therefore, regardless of whether or not JAK2 V617F represents a primary event in myeloproliferative disorders, it is probable that it promotes a PV phenotype. In the current study, we explore this hypothesis as well as investigate the long-term prognostic relevance of the mutation in a well defined cohort of patients with ET seen at a single institution. Methods: The diagnosis of ET was according to the WHO classification of tumors (Jaffe ES et al. Lyon, IARC press, 2001). All patients were seen at the Mayo Clinic and patient characteristics and subsequent clinical events were determined both retrospectively (chart review) and prospectively (by means of a questionnaire sent both to patients and their primary doctors). In addition date and cause of death were confirmed by probing both the social security index and the national death index. DNA for mutation screening was derived and sequenced from either archived unsorted cells from the bone marrow or prospectively collected peripheral blood granulocytes according to previously published methods (Levine et al. Cancer Cell, 2005;7:387). Results: The study cohort consisted of 150 consecutive patients with ET including 99 females and 51 males (median age 49.8 years; range 16.8–98). Overall, JAK2 V617F was detected in 73 patients (48.7%) but none were homozygous for the mutant allele. At diagnosis, ET patients with JAK2 V617F were significantly older (median age of 53.5 vs. 42.3; p=0.02) and also displayed higher readings of both hemoglobin level (p=0.0002) and leukocyte count (p=0.005). All study patients have now been followed for a median of 11.4 years from diagnosis (range 0–32.7) and follow-up period was similar between the two mutational categories (p=0.53). During this period, 32 patients have died and the median survival for the entire group was 19.3 years; 21.9 and 17.7 years in the absence and presence of the mutation, respectively (p=0.09). Disease transformation into acute leukemia, myelofibrosis, and PV occurred in 6 (4%), 15 (10%), and 8 (5.3%) patients, respectively, and the only significant difference between the two mutational categories in this regard was the higher incidence of transformation into PV in patients with the mutation (p=0.02). Multivariate analysis identified age, hemoglobin level, and thrombosis history at diagnosis but not JAK2 V617F mutational status as independent risk factors for survival. Conclusion: Although the presence of JAK2 V617F in ET appears to promote a PV phenotype, it might not carry treatment-relevant information.

Abstract Background: One of the utilities of molecular markers in hematological malignancies is their potential to predict response to drug therapy. Accordingly, we inquired about the effect of both cytogenetic profile and JAK2V617F mutational status on drug therapy response in myelofibrosis with myeloid metaplasia (MMM). Methods: Mutation analysis for JAK2V617 was performed in DNA derived from peripheral blood mononuclear cells, granulocytes, or both. Genomic DNA was amplified by PCR and fluorescent dye chemistry sequencing was performed using the same primers used for amplification (Levin et al. Cancer Cell2005;7:387). Abnormal cytogenetic findings were considered to be either favorable (sole 13q- or 20q- abnormalities) or unfavorable (other abnormalities) (Tefferi et al. BJH2001;113:763). Results: i. Patients and treatment: A total of 69 patients with MMM (median age 62 years, range 38–75; 47 males) received the following drugs as first-line therapy for either anemia or symptomatic splenomegaly; erythropoietin (Epo) alone or in combination with hydroxyurea (n=25), hydroxyurea alone (HU; n=17), interferon alpha (IFN; n=11), combination of thalidomide and prednisone (ThalPred; n=7), androgen preparations (Andro; n=5), and etanercept (n=4). ii. Results of cytogenetic studies and JAK2 mutation screening: Cytogenetic findings were abnormal in 35 patients (51%); favorable in 9 and unfavorable in 26. JAK2V617 mutation analysis revealed wild-type allele in 31 patients (45%) and either heterozygous (n=31) or homozygous (n=7) mutation in the remaining 38 patients (55%). iii. Correlation of response to cytogenetic/molecular markers: Overall, 17 patients (25%) achieved either a major (n=8) or minor (n=9) response in anemia from treatment with one of the aforementioned drugs. Regardless of either JAK2V617 mutational status or cytogenetic profile, treatment-induced responses in anemia were poor for IFN (0%) and HU (12%). In contrast, the best anemia responses were documented for ThalPred (57%), Andro (40%), and Epo-based therapy (32%). Among the 25 patients that received Epo-based therapy, all 3 patients with favorable cytogenetic abnormalities achieved major responses in anemia whereas such responses were seen in none of the 9 patients with unfavorable cytogenetic abnormalities and only 1 of 13 patients with normal cytogenetics (p=0.004). Furthermore, 3 of the 4 major responders were heterozygous for JAK2V617 and one carried the wild-type allele. In contrast, none of the 3 JAK2V617 homozygotes showed any type of response but all 3 also displayed unfavorable cytogenetics. None of the 7 ThalPred-treated patients carried favorable cytogenetics and yet 3 achieved major responses including one with unfavorable cytogenetics. JAK2V617 mutational status was heterozygous in 1 and wild-type in the other two. There were no major responses among the 5 Andro-treated patients despite the presence of favorable cytogenetic abnormalities in 2 patients. In order to investigate the relationship between cytogenetic profile and JAK2V617 mutational status further, we referred to an expanded database of 116 patients and found the incidence of unfavorable cytogenetics to be higher in JAK2V617 homozygotes compared to non-homozygotes (57% vs. 31%; p=0.16) Conclusion: The current study suggests clustering of favorable cytogenetic abnormalities in MMM with anemia response to Epo therapy and unfavorable cytogenetic abnormalities with homozygosity for JAK2V617 mutation.

Abstract Background Estrogen-based hormone treatment (EBHT), in the form of either oral contraceptives or hormone replacement therapy, is generally felt to be associated with an increased risk of thrombosis in women with essential thrombocythemia (ET). However, there is insufficient evidence to support or refute such a concern. Methods Data was abstracted from medical records of a consecutive cohort of women with ET seen at the Mayo Clinic. Thrombotic risk was determined by the occurrence of a major thrombotic event namely myocardial infarction, angina, cerebrovascular accident, transient ischemic attack, peripheral arterial thrombosis, pulmonary embolism, or deep venous thrombosis at or after diagnosis of ET. Results i. Information regarding EBHT at or after the diagnosis of ET The study cohort included 307 women with ET (median age, 55 years; range 16–88) with a median follow-up of 133 months from diagnosis (range, 0.2–397). EBHT at or after diagnosis was documented in 93 women (30%); 59 patients were on EBHT at time of diagnosis and treatment was initiated after diagnosis in 34 patients. Among the 59 patients who were receiving EBHT at the time of diagnosis, treatment was discontinued within one month of their ET diagnosis in 17 patients but continued in the other 42 patients for a median of 48 months (range, 2–168). In the 34 patients with EBHT that was initiated after the diagnosis of ET, median treatment period was 44 months (range, 1–144). ii. Correlation between EBHT and thrombosis at diagnosis At diagnosis, major thrombosis was documented in a total of 74 patients (24%) that included 11 of the 59 patients (19%) on EBHT and 63 of 248 patients (25%) that were not receiving such therapy at the time of diagnosis (p=0.28). iii. Correlation between EBHT and thrombosis after diagnosis A total of 95 patients (31%) experienced a subsequent (i.e. post-diagnosis) thrombotic episode during the study period. These occurred in 65 of 214 patients (30%) that were not exposed to EBHT at or after diagnosis, 5 of 17 patients (29%) in whom EBHT was discontinued with the diagnosis of ET, 13 of 42 patients (31%) in whom EBHT was continued despite the diagnosis of ET, and 12 of 34 patients (35%) in whom EBHT was started after the diagnosis of ET (p=0.95). Furthermore, duration of EBHT, using a cutoff value of 1 year, did not significantly affect the incidence of thrombotic events (p=0.67). The different comparative groups were similar in age distribution as well as frequency of other cardiovascular risk factors including smoking, diabetes mellitus, hypertension, and hyperlipidemia. Conclusion The current study addresses the important issue of safety pertaining to the use of estrogen-based hormones in women with ET and did not reveal a significant alteration of thrombosis risk as a result of hormone therapy.

Abstract Background: The occurrence of an activating JAK2 mutation (JAK2 V617F) has been reported in a variable proportion of patients with bcr/abl-negative myeloproliferative disorders including myelofibrosis with myeloid metaplasia (MMM). In the current study, we report mutational frequencies in each of the 3 MMM variants as well as examine the prognostic relevance of the mutation in a large group of patients from a single institution. Methods: Mutation analysis for JAK2 V617F was performed in DNA derived from peripheral blood mononuclear cells (PBMC) from all study patients (n=157). Concomitant analysis was also performed in DNA derived from purified granulocytes (n=57), CD34-positive cells (n=25), and T cells (n=19). Genomic DNA was amplified by PCR and fluorescent dye chemistry sequencing was performed using the same primers used for amplification. Results: The study cohort included 157 consecutive patients with MMM including 117 with agnogenic (AMM), 22 with post polycythemic (PPMM), and 18 with post thrombocythemic (PTMM) myeloid metaplasia. The overall detection rate for JAK2 V617F in PBMC was 51% (homozygous in 5.7%) and was significantly (p=0.0002) higher in PPMM (91%; homozygous in 18%) compared to either AMM (45.3%; homozygous in 2.6%) or PTMM (38.9%; homozygous in 11.1%). In AMM (n=117), the presence of JAK2 V617F was significantly associated with an older age at diagnosis (p=0.03), and a history of both thrombosis (p=0.01) and pruritus (p=0.005). No other statistical associations were evident and the presence of the mutation did not predict leukemic transformation. Furthermore, multivariate analysis for overall survival of patients with AMM identified only age and the Dupriez prognostic score as independent prognostic factors; JAK2 V617F had no prognostic significance. Finally, the mutation was not found in T cells and its detection rate in either granulocytes or CD34-positive cells was similar to that in PBMC. Conclusion: JAK2 V617F is a myeloid lineage-specific event, its incidence in MMM is significantly higher with an antecedent history of polycythemia vera, and its presence in AMM does not affect prognosis but is associated with PV-characteristic clinical features.

Abstract Background: Angiogenesis has been found to be an important regulator in the disease progression of both solid tumors and hematologic malignancies. In solid tumors, the balance of pro and anti-angiogenic factors appears to be a critical factor in tumorigenesis, with disturbance in favor of angiogenesis promoting tumor growth and metastasis. This observation led to the concept of an “angiogenic switch” where pro-angiogenic influences outweigh anti-angiogenic factors. Patients with CLL have detectable levels of plasma and cellular pro and anti-angiogenic cytokines as well as abnormal neovascularization in the marrow and lymph nodes. Limited studies of pro-angiogenic cytokines have suggested that inter-patient variation in serum/plasma and cellular levels of these markers may have prognostic implications. We evaluated pro- and anti-angiogenic cytokines in a large sample of patients with CLL to evaluate the presence of angiogenic switching and its implications for disease progression. Methods: In a prospective, longitudinal study, we analyzed the serum/plasma, intracellular(determined by flow cytometry), and CLL B cell secreted (24 hour medium values) levels of pro-angiogenic cytokines (VEGF, bFGF) and an anti-angiogenic cytokine (Thrombospondin [TSP]) in 311 patients with previously untreated CLL. VEGF:TSP and bFGF:TSP ratios were calculated to evaluate for evidence of angiogenic switching (higher ratio suggests a pro-angiogenic phenotype). The relationship of angiogenic cytokines to Rai stage, IgVH gene mutation status, % CD38 protein expression, and time to treatment (TTT) from the date of sample were evaluated (median follow-up=11 months). Sequential samples were available on >90% of patients and were evaluated to assess if changes in levels of angiogenic cytokines occurred in patients requiring treatment. Results: Serum, cellular, and secreted levels of VEGF, bFGF, and TSP showed no correlation with Rai stage, IgVH gene mutation (continuous variable), or % expression of CD38 (continuous variable). However, among patients with Rai stage 0-II disease, individuals with lower TSP levels (<75 percentile, p=0.007) or higher bFGF:TSP ratios (continuous variable; p=0.003) had shorter TTT (Figure 1). Similar relationships for a shorter TTT were also observed for secreted TSP levels (p=0.007) and the cellular bFGF:TSP ratio (p=0.009). On analysis of sequential samples, no clear difference in serum, cellular, or secreted levels of VEGF, bFGF, TSP, VEGF:TSP, or bFGF:TSP occurred among those who required treatment. Conclusions: Angiogenic cytokines appear to have prognostic significance in patients with previously untreated early stage CLL. Importantly, we report the original observation that the ratio of bFGF (pro-angiogenic) to TSP (anti-angiogenic) correlates with time to treatment as a continuous variable. This result provides further evidence that angiogenic switching relates to clinical progression in CLL. Longer follow-up of this cohort is needed to fully define the prognostic implications of angiogenic cytokines in CLL and determine if integration of these markers with other prognostic parameters improves risk stratification for early stage patients. Survival According to Serum Thrombospondin Level Survival According to Serum Thrombospondin Level

Abstract Background: Microcytosis is a characteristic laboratory feature for both iron deficiency anemia and thalassemia. Microcytosis is also infrequently seen with “anemia of chronic disease”. A partial list of the latter include rheumatoid arthritis, polymyalgia rheumatica, diabetes mellitus, connective tissue disease, chronic infection, Hodgkin’s lymphoma, Castleman’s disease, and renal cell carcinoma. In the current study, we show that microcytosis is a frequent laboratory feature in myelofibrosis with myeloid metaplasia (MMM) and investigate its clinical relevance in the particular setting. Methods: In order to minimize the confounding effects of nutritional deficiencies, red blood cell transfusion, and drug therapy on mean corpuscular volume (MCV), the current study was limited to young patients (age < 60 years) that had not received either red blood cell transfusion or drug therapy for MMM during the time of their initial laboratory investigation. Results: The study population consisted of 102 consecutive patients with MMM (median age 51 years, range 18–60) including 60 males. At diagnosis, the median (range) values for hemoglobin, leukocyte count, and platelet count were 11 gm/dL (7.4–14.9), 8.3 x 109/L (1–40) and 357 x 109/L (14–1512), respectively. Spleen measurements were documented in 78 patients (median 6 cm below the left costal margin, range 0–24) and the incidence of hypercatabolic symptoms was 18% (12 of 67 evaluable patients). The Dupriez prognostic score distribution included low-risk (n=73), intermediate-risk (n=19), and high-risk (n=10) disease categories. The overall incidence of microcytosis (MCV < 80 fL) was 28%; median MCV was 84.3 fL (range, 60.6–99.1). Among the 28 patients with microcytosis, MCV was less than 70 fL in 4 patients. Concomitant serum ferritin levels were available in 44 patients (median, 92 μg/L; range, 4–6440) and the levels were below 30 μg/L in 7 patients including 4 with microcytosis. Bone marrow iron stains were sporadically performed and showed normal iron stores in the majority of evaluable cases. The study cohort was followed for a median of 70 months (range 0–300). Multivariate analysis of variable that were measured in all study patients identified anemia, thrombocytopenia, and age but not MCV as independent prognostic indicators of inferior survival. The addition of variables that were measured in a variable proportion of the study population disclosed additional prognostic value for hypercatabolic symptoms and abnormal cytogenetics. Interestingly, microcytosis was significantly associated with the absence of cytogenetic abnormalities (p=0.04); among the 10 patients with abnormal cytogenetic findings, 9 had an MCV of ≥ 80 fL. Microcytosis did not correlate with several other clinical and laboratory features including gender, presence of hypercatabolic symptoms, spleen size, hemoglobin level, platelet count, leukocyte count, serum lactate dehydrogenase level, and the Dupriez prognostic score. Conclusion: Microcytosis with a normal or elevated serum ferritin level is a frequent laboratory feature in MMM but does not appear to influence survival. The observation regarding the low incidence of cytogenetic abnormalities in MMM patients with microcytosis is intriguing but requires validation by a prospective study. Whether or not MMM-associated microcytosis represents an acquired defect of globin synthesis, as might be the case in some patients with myelodysplastic syndrome, remains to be elucidated.

Abstract BACK GROUND Primary myelofibrosis (PMF) leads to weight loss, splenomegaly and constitutional symptoms (Cancer2007;109:68–76) and hypocholesterolemia associated with decreased survival (Blood2007;110:a2548). Initial trials of JAK2 inhibitors have demonstrated improvements in these (JCO2008;26:a7004), yet little objective data exists regarding the dynamic changes in weight and spleen size during the course of PMF. METHODS We analyzed the Mayo PMF database for patients with information on disease prognosis, presentation, therapies, height and weight at diagnosis, and outcomes. Additionally, when available, we analyzed additional weights during the clinical course, the body mass index (BMI- (weight/(height*height)), spleen size, and peripheral blood studies including lipids. Results: 370 patients with PMF were identified with 193 (52%) having a weight at the time of diagnosis, the remainder had a weight obtained a median of 10.0 months (range 1.8–159) after diagnosis. Additionally, 210 patients (57%) had a weight value available from 1–3 additional time points during the course of their disease. At PMF diagnosis patients had a median BMI of 25.3 (range 17.6 to 38.4), with 53% considered overweight (USA Norms (NHANES 2004): 66% and 32.2% are considered overweight or obese). Although weight lost prior to diagnosis unclear, only 2% of patients would be deemed “underweight” at diagnosis. Serial analysis demonstrated 67% experienced a decrease in weight over time, with 23% and 6% experiencing a >10% or 20% decrease, respectively. Overall during the course of illness 65% of patients remained in the same BMI category with 27% dropping to a lower category and 8% gaining in a category (Table 1). Table 1. Distribution of Body Mass Index and Spleen Size in 370 PMF patients followed serially DX of PMF Weight 2 Weight 3 Weight 4 Underweight BMI<18.5 2% 1% 2% 2% Normal BMI=18.5–24.9 45% 41% 41% 54% Overweight BMI=25–29.9 33% 42% 41% 35% Obese BMI>30 20% 16% 16% 10% Spleen Size (cm BLCM) (median: range) 5.0 (0–30) 6 (0–33) 7 (0–33) 7.5 (0–28) Time from PMF DX (median(range) months) 0 10.3 (1.8–159) 28.3 (6–191) 57 (15–258) Major weight increases were rare <1%, and might include buildup of fluid or splenic mass. Indeed spleen sizes generally increased during the course of the illness, with 51% having “massive” splenomegaly (either >10 (45%) or >20 (20%) cm below left costal margin or requiring splenectomy (19%)). Massive splenomegaly was not associated with weight loss or weight gain, suggesting increasing splenic mass came at the expense of alternate tissue mass (muscle and or fat). Median survival for the entire cohort was 77 months and PMF prognostic scores being valid (Lille PMF score p<0.01; IWG-MRT PMF Score - Submitted separately Cervantes et. al. p<0.01), however BMI number and category was not independently predictive of survival. Additionally, neither percentage of weight loss nor changing BMI category were associated with decreased survival. Although splenomegaly was not independently associated with decreased survival, the need for splenectomy clearly was (p=0.02). Consistent with our prior observation hypocholesterolemia (total <150 (median of 151 (range 12–366) or a HDL cholesterol of <60 (median 77 (range 25–248)) were associated with decreased survival (p<0.01 and =0.05, respectively). Decreasing weight in PMF patients was most highly correlated with decreases in HDL (p=0.03). Hypocholesterolemia was highly associated with splenomegaly (p<0.001 for spleen size >10 cm at any point during disease course). Conclusions: Weight loss is common in most PMF patients, despite the confounders of edema and increasing splenomegaly which could mask the degree of loss of lean body mass. Although few PMF patients are considered “underweight” by BMI standards, the distribution compared to USA norms suggest a lower weight distribution. Additionally, patients during the course of their illness are most likely to fall from a BMI overweight category to a “normal” category which may underestimate the true degree of catabolism of muscle mass occurring. Although not independently associated with decreased survival, weight loss remains common and a sign of advancing disease in this terminal illness. More precise methodology to estimate true lean muscle mass lost or gained may yield a better understanding of the hypercatabolic state seen in PMF. Efficacious therapy (such as potentially a JAK2 inhibitor) will likely reverse the trend of weight loss in these patients.

Abstract Background: Leukocytosis has recently been implicated as an adverse prognostic feature for thrombosis in both ET and PV. Such an association would be therapeutically most relevant in the context of “low-risk” disease. In the current study, we sought to clarify the relationship between leukocytosis at diagnosis and the subsequent occurrence of either arterial or venous thrombosis, in “low-risk” patients with ET or PV. Methods: Data was abstracted from the medical records of a consecutive cohort of patients with WHO-defined ET or PV seen at the Mayo Clinic. Low-risk disease was defined by the absence of both thrombosis history and age 3 60 years. Cox proportional hazards model was utilized to determine the impact of clinical and laboratory variables on thrombosis-free survival (TFS). Arterial- or venous-specific TFS curves were constructed by Kaplan-Meier method. Results: i) Patient characteristics and outcome A total of 407 “low-risk” patients were studied; 153 had PV (median age 48 years; females 43%) and 254 ET (median age 42 years; females 71%). A total of 46 thrombotic events (22 arterial and 24 venous) were recorded in 41 (27%) patients with PV during a median follow up of 130 months (range 2–562 months). The corresponding figures in ET were 54 total thrombotic events (41 arterial and 14 venous) in 47 (19%) patients at a median follow up of 104 months (range 0.25–424 months). Cytoreductive therapy was avoided in the presence of < 1000 x 109/L platelet count but, at the discretion of the treating physician, some patients with higher platelet counts received prophylactic cytoreductive therapy. ii) Correlation between leukocytosis and thrombosis A leukocyte count of 3 15 x 109/L at diagnosis was documented in 42 (27%) patients with PV and 21 (8%) patients with ET; 102 patients (40%) with ET had > 9.4 x 109/L leukocyte count. Leukocyte count considered as either a continuous or categorical variable (using cutoff levels of 15 x 109/L for PV and either 15 x 109/L or 9.4 x 109/L for ET) was not significantly associated with either arterial or venous thrombosis (Figure 1). iii) Correlation between other risk factors and thrombosis In univariate analysis, presence of the JAK2V617F was significantly associated with arterial thrombosis in ET (p=0.049) but significance was lost during multivariable analysis that included age as a covariate. Only advanced age was found to be significantly associated with arterial thrombosis in PV (p=0.04) and higher hemoglobin level with venous thrombosis in ET (p<0.0001). Conclusion: The current study does not identify leukocyte count at diagnosis as a marker of increased thrombosis risk in low-risk patients with either ET or PV. Instead, the study found an association between arterial thrombosis and advanced age in PV and venous thrombosis and higher hemoglobin level in ET. Figure Figure

Abstract BACKGROUND: In myelodysplastic syndromes (MDS) without excess blasts, red blood cell (RBC) transfusion requirement has been associated with poorer overall (OS) and leukemia-free (LFS) survival, suggesting that transfusion dependency is a marker of more severe disease (JCO2005;23:7594). We recently reported similar results in refractory anemia with ring sideroblasts (RARS): RBC transfusion requirement was an IPSS-independent adverse prognostic factor, but neither serum ferritin nor transfusion burden had prognostic value (AJH2008;83:611). Here, we examine the prognostic relevance of serum ferritin, need for RBC transfusions at time of diagnosis, and total transfusion burden in primary myelofibrosis (PMF). METHODS: We reviewed medical records to ascertain the clinical and laboratory features of 185 consecutive patients diagnosed with PMF according to the 2001 World Health Organization (WHO) criteria. Patients were excluded if ferritin measurements at time of diagnosis were unavailable. OS and LFS curves were constructed by Kaplan-Meier method, taking the interval from the date of diagnosis to death, leukemic transformation, or last contact. Log-rank test was used to test the homogeneity of survival curves over different groups. Cox proportional hazards model was utilized to determine the impact of various clinical and laboratory variables on OS and LFS. RESULTS: Clinical characteristics at diagnosis: Median age was 58 years (range 15–81; 110 males). Median serum ferritin level at diagnosis was 164 ng/mL (range 1–3903) and was ≥1000 ng/mL in 22 patients (12%). 101 (55%), 65 (35%) and 19 (10%) patients were assigned low, intermediate- and high-risk disease category (Blood1996;88:1013). In addition, 32 (17%) patients required RBC transfusions, 33% had constitutional symptoms and 39% displayed ≥ 1% circulating blasts. Where evaluated, 40% of the patients had cytogenetic abnormalities and 56% JAK2V617F. Events during the disease course: At a median follow-up of 28 months (range 0.5–231), 79 (43%) deaths and 15 (8%) leukemic transformations were documented. During this period, 126 (68%) patients required some form of therapy other than transfusion, including splenectomy in 33 (18%); only 4 underwent allogeneic stem cell transplantation. Causes of death were documented in 33 instances and none were attributed to iron overload. Median peak serum ferritin level during the disease course was 231ng/mL (range 9–13,080); 144 (78%) patients had peak levels <1000 ng/mL, 28 (15%) between 1000 and 3000 ng/mL, and 13 (7%) above 3000 ng/mL. Number of total RBC transfusions ranged from none to 121. Iron chelation therapy was reported in 62 patients (34%). Prognostic factors for OS and LFS: Kaplan-Meier projected median survival was 72 months. By univariate analysis, increased serum ferritin level at diagnosis considered as either a continuous (p<0.0001) or categorical (≥ 1000 ng/mL) variable (p<0.0001), RBC transfusion requirement at diagnosis (p<0.0001) and higher number of total RBC units transfused during the course of the disease (p=0.004) were all associated with inferior survival. However, only RBC transfusion requirement at diagnosis sustained its prognostic significance when age and conventional prognostic risk scores were added to the multivariable model as covariates. Similarly, a peak serum ferritin level of > 3000 ng/mL documented during the disease course was not detrimental to survival. History of iron chelation therapy was associated with shortened survival (p=0.003). Multivariable analysis that included previously described risk factors for LFS also identified RBC transfusion requirement at diagnosis as an additional and independent risk factor. CONCLUSIONS: Serum ferritin level at time of diagnosis or during the disease course of PMF lacks independent prognostic value for either OS or LFS. The same is true for total transfusion burden. However, although hemoglobin <10 g/dL is a component of all currently utilized prognostic scoring systems for PMF, the presence of a more severe erythropoietic defect as indicated by RBC transfusion need at time of diagnosis has an additional adverse prognostic value for both OS and LFS.