Introduction

The simultaneously accelerating pace of discovery in biomedical science and innovations in clinical practice threaten to create a widening schism between laboratory investigators and clinicians. This problem applies to all fields of medicine. However, nowhere is it more evident than in hematology, which has traditionally occupied the cutting edge of scientific research and patient care. Indeed, it is the very excitement and intellectual stimulation at this interface which attracted many of us into the field in the first place. We now need innovative mechanisms to bridge research and practice.

To this end, “Diffusion” will be a regular feature in The Hematologist. In each issue, members of our editorial board will identify the most important and provocative recent publications in the world literature that apply to the field of hematology. They will provide very concise abstracts of the reported findings, followed by interpretations of the potential impact and applications of the work. Basic research papers will be translated for practicing clinicians, highlighting their clinical implications. Clinical papers will be translated for laboratory investigators, along with ideas for future research directions. Since some of the most important articles will have been published in Blood, these will be included among the selected literature digests. Indeed, we hope to stimulate our readers who may not routinely peruse Blood to read the original articles. Your feedback and suggestions for the “Diffusion” column will be most welcome.

Andrew I. Schafer, M.D.
Editor-in-Chief, The Hematologist

Sanford J. Shattil, M.D.
Editor-in-Chief, Blood

Return to Top

FANCL - A New Letter in the Fanconi Anemia Alphabet

Meetei AR, deWinter JP, Medhurst AL, et al. A novel ubiquitin ligase is deficient in Fanconi anemia. Nature Genetics. 2003;35:165-170.

By Nancy Andrews, M.D., Ph.D.

Fanconi anemia is a rare genetic disease, known to pediatricians for its association with bone marrow failure and birth defects, and to internists for its predisposition to cancer. Based on the hypersensitivity of Fanconi anemia cells to DNA crosslinking agents (e.g., mitomycin C, diepoxybutane), it has been inferred that the disease results from failure of normal DNA repair functions. Insights into the molecular details of this disorder will not only teach us about the pathophysiology of Fanconi anemia, but will undoubtedly also help us understand cellular events leading to cancer.

Rapid progress has been made in identifying the multiple genes (at least 8) that can be mutated to cause Fanconi anemia. Until recently however, little was known about the normal functions of the proteins they encode. Six of the Fanconi anemia proteins (FANCs) assemble together to form a multiprotein “core” complex of uncertain function that resides in the cell nucleus (Figure 1; reviewed by D'Andrea and Grompe, Nat Rev Cancer 3: 23 -34, 2003). One of the proteins, FANCD1, is also known as BRCA2, owing to its role in susceptibility to breast cancer. Recently, Meetei and colleagues reported elegant studies of a new Fanconi anemia disease-causing gene, FANCL (also called PHF9). They identified FANCL by biochemical purification of the Fanconi core complex, and mass spectroscopy analysis of a previously unknown protein component. While the molecular functions of the other FANC proteins remain obscure, Meetei et al. showed that FANCL functions as an ubiquitin ligase.

Ubiquitin is a small protein tag that can be attached to other proteins, to target them within the cell to their site of use or degradation. In response to DNA damage, an ubiquitin tag is attached to FANCD2, causing it to relocate within the nucleus to cooperate with BRCA1 and other proteins that directly mediate DNA repair. FANCL appears to be the member of the core complex that enzymatically attaches the ubiquitin tag.

The discovery of FANCL gives new insight into how cells ensure the stability of their genomes, and how genomic instability predisposes to cancer. It is reasonable to speculate that mutations in FANCL also play an as-yet unrecognized role in the development of human tumors. Recent information suggests that identification of Fanconi anemia disease genes is not just an academic exercise. Shimamura and D'Andrea argue that it also provides prognostic information that can be used to modify clinical management. Patients with mutations in FANCA may have a milder clinical course, with later development of aplastic anemia, while those with mutations in FANCC and FANCG appear to have more aggressive disease. At present, clinical labs are not equipped to identify FANC mutations in individual patients, but new tests are likely to become available in the future.

Return to Top

Releasing the Brake on Stem Cell Cycling: HSC Expansion by Inhibiting p21

Stier S, Cheng T, Forkert R, Lutz C, Dombkowski DM, Zhang JL, and Scadden DT. Ex vivo targeting of p21 Cip1/Waf1 permits relative ex pansion of human hematopoietic stem cells, Blood, 2003; 102:1260 - 1266.

By Stephen G. Emerson, M.D., Ph.D.

Can hematopoietic stem cells be grown in the laboratory, without damaging them or altering their properties after re-transplantation? In this paper, Stier and colleagues show that blocking the function of p21 allows the expansion of the hematopoietic stem cell (HSC) population without inhibiting the ability of HSCs to produce differentiated blood cells. p21, initially discovered as a downstream target of the tumor suppressor gene p53, is a cyclin-dependent kinase inhibitor (CKI) that acts as a tonic break to downmodulate progression through the cell cycle. Previous experiments have shown that mice lacking p21 have increased numbers of HSCs, but that these p21-deficient HSCs are less resistant to normal stress in vivo ( Rodrigues et al. Science 287:1804-8, 2000). If one were to target the p21 pathway for HSC expansion for clinical application, one would ideally like to block p21 expression for a period of time, then restore its normal activity once stem cells were expanded.

In this paper, Stier et al. use lentiviral based retroviral gene transfer with anti-sense p21 to block p21 expression in HSC. This approach, which permanently infects nearly one-half of the HSCs, increases the number of HSCs that actively cycle, and increases HSC numbers three-to-four fold. The infected cells are still able to produce normal ratios of myeloid and lymphoid cells. The success of this approach suggests that temporarily blocking p21 function in HSCs should similarly increase HSC numbers, without having to resort to permanent alteration of the HSC genome imposed by retroviral gene transfer techniques. Any protein approach, from traditional small molecule drug inhibitors to large dominant-interfering proteins linked to nuclear transport inducing HIV TAT peptides, could be attractive.

Hematopoietic stem cells (HSC) are the paradigm of cells that retain the ability to maintain their numbers (self-renewal) while simultaneously producing billions of mature daughter cells. One would think that this ability to self-replicate should be easy to amplify experimentally (and therefore clinically) for blood cell therapies, simply by identifying the cytokines or other external stimuli that induce HSC to normally replicate at slow rates within the bone marrow. However, nearly two decades of such efforts have not been terribly successful, either because environmental stimuli don't induce significant stem cell expansion or because we have not yet identified the proper, normal environmental cytokines. This new paper exemplifies an alternative strategy, now pursued in several laboratories around the world, which attempts to target specific intracellular gene programs that have been found to be important for stem cell proliferation and differentiation. This next step, to use short-acting, reversible modifiers of cell cycle proteins and other stem cell self-renewal regulators (transcription factors, histone modification enzymes) (Sauvageau et al. Cell. 109:39-45, 2002) should prove to be very exciting and may provide the techniques to safely and effectively expand HSCs. Doing so would open many of the opportunities for blood banking, transfusion and transplantation medicine, and stem cell gene therapy that are so eagerly anticipated.

Return to Top

Platelets in Atherosclerosis

Huo Y, Schober A, Forlow SB, Smith DF, Hyman MC, Jung S, Littman DR, Weber C, Ley K. Circulating activated platelets exacerbate atherosclerosis in mice deficient in apolipoprotein E. Nat Med. 2003;9:61-67.

Burger PC, Wagner DD. Platelet P-selectin facilitates atherosclerotic lesion development. Blood. 2003;101:2661-2666.

By Kenneth Kaushansky, M.D.

While it is certain that platelets play a critical role in acute thrombotic complications of vascular disease, whether or not they contribute to the pathogenesis of atherosclerosis has been in dispute for many years. In two recent papers in the journals Nature Medicine and Blood, important roles for platelet P-selectin in the progression of atherosclerotic lesions have been established. Huo and colleagues ( 1 ) studied whether circulating activated platelets might impact the atherosclerotic lesions that develop in the apolipoprotein-E (ApoE) deficient murine model of atherosclerosis. These investigators periodically injected thrombin-activated platelets and tracked their fate. They found the injected platelets quickly aggregated with monocytes and that the aggregates soon thereafter localized to developing atherosclerotic lesions. In so doing, the platelets helped deliver chemokines, such as RANTES and platelet factor 4, to the endothelium of the atherosclerotic arteries. The monocytes in the aggregates also were encouraged to bind to vascular cell adhesion molecule (VCAM)-1 on the surface of the diseased endothelium, where they adhered and migrated into the lesion. After twelve weeks, there was a significant increase in the degree of atherosclerosis present in the ApoE -/- mice. Of considerable interest, injection of activated P-selectin deficient platelets failed to recruit monocytes to atherosclerotic lesions and atherosclerosis did not progress any faster in ApoE -/- mice treated with activated P-selectin deficient platelets than in control ApoE -/- mice. In a subsequent study that complements these findings, published by Burger and Wagner ( 2 ), a stem cell transplantation strategy was employed to examine the effects of P-selectin on the development of atherosclerosis in the same murine model. These investigators transplanted stem cells from ApoE deficient or ApoE deficient P-selectin deficient mice into either ApoE -/- or ApoE -/- // P-selectin -/- recipients, thereby generating four groups of ApoE -/- mice bearing P-selectin on both platelets and endothelial cells, on platelets only, on endothelial cells only, or on neither. Over the ensuing seven months, they monitored the development of atherosclerosis, and found that endothelial cell and platelet P-selectin both contribute to the development of atherosclerosis.

Atherosclerosis is the leading cause of morbidity and mortality in North America and Europe, and its untoward effects are rapidly increasing in the remainder of the world. Thirty years ago, Russell Ross and colleagues enunciated the “response to injury” hypothesis for atherosclerosis, in which endothelial cell damage initiated the recruitment of platelets at sites destined to become atheroscleorotic lesions. In this construct, platelets played a key role in the pathogenesis of atherosclerosis, contributing growth factors (e.g. platelet derived growth factor) that trigger smooth muscle cell hypertrophy and the resulting luminal narrowing. Since then, much work has focused on lipids and the immune system; it is clear that hypercholesterolemia is an absolute prerequisite for the development of atherosclerosis, and that a vigorous innate immune response to the presence of oxidized low density lipoprotein (LDL) particles in the vascular wall results in a complex lesion composed of foam cells, monocytes and lymphocytes, and the growth factors and chemokines that fuel this process. As these later pathogenic processes have shed important new insights into our understanding of the development of atherosclerotic lesions, they had displaced platelets from their previous level of importance in the process. Nevertheless, new evidence is now beginning to emerge that platelets play a vital role in the progression of atheroscleorotic lesions.

It is clear from these two studies that platelets remain an important component of the ongoing development of atheroscleorotic lesions. Although it remains controversial, a number of studies have demonstrated that platelet count and platelet activation state correlate with the morbidity from cardiovascular disease. Common medical practice supports this notion; anti-platelet agents have been used for decades in patients with cardiovascular disease, and have repeatedly been shown to reduce the morbidity and mortality in such patients. With the impact of chronic anti-platelet therapy, it is unlikely that the sole benefit of these agents is preventing the acute thrombotic effects that occur upon plaque rupture. These two new studies further our understanding of the role played by platelets in the atheroscleorotic process and point to a new avenue for developing additional approaches to manipulating platelet function for therapeutic benefit.

Return to Top

A Fine Balance Between Immunity and Cancer?

Dhodapkar MV, Geller MD, Chang DH, Shimizu K, Fujii S, Dhodapkar KM, Krasovsky J. A reversible defect in natural killer T cell function characterizes the progression of premalignant to malignant multiple myeloma. J Exp Med. 2003; 197(12):1667-76.

By Peter Lee, M.D.

Dogma in immunology states that the immune system responds to foreign antigens and spares self. Since cancer cells are of ‘self' origin, they are often thought to be ignored by the immune system. As such, cancer progression is viewed as solely a function of proliferation of cancer cells. An alternative view may be that cancer represents a dynamic balance between cancer cell proliferation and the host immune response. There is now mounting evidence that an anti-cancer immune response (involving both T cells and B cells) develops in many patients – even with advanced disease. However, such responses are non-protective in these patient. What about in patients with earlier disease? Could a host immune response modulate disease progression?

This paper provides support – in humans – that the immune response may indeed modulate cancer progression. This study focuses on the transition of multiple myeloma (MM) from MGUS, and also compared against patients with non-progression myeloma (NPM). The investigators first looked for T cells reactive against autologous tumor (via secretion of IFN- g ) from the blood or bone marrow but found no difference amongst these three patient groups (MM vs. NPM vs. MGUS). They then went on to look at function of natural killer T (NKT) cells. NKT cells express a restricted T cell receptor (V a 24-V b 11 in human) which recognizes glycolipid rather than peptide ligands. They have been shown to be important in surveillance against spontaneous and carcinogen-induced tumors in mouse models. In this study, NKT cells could be detected in the blood and bone marrow at similar levels in MM, NPM, and MGUS patients. However, these cells are functional in NPM and MGUS patients, but strikingly non-functional (with regards to IFN- g production) in MM patients. When cultured, NKT cells from MM patients could regain their function and respond to autologous tumor cells, suggesting that this in vivo functional defect is reversible. Taken together, these data suggest that functional NKT cells may help to maintain myeloma cells at a steady state in MGUS and NPM patients; loss of NKT cell function may led to progression of disease.

These findings support the intriguing notion that cancer may represent a fine balance between cancer cell proliferation and the host immune response. Acquisition of the ability to modulate the host immune response may be an important factor in cancer progression, perhaps as important as an increase in proliferative potential. This would be further supported by accurate measurements of the in vivo proliferation rates of myeloma cells from MGUS and MM, which is now feasible using a novel deuterium labeling method in humans. If confirmed, these results suggest new therapeutic strategies directed not just at the tumor cells, but also their interactions with the immune system. A therapy which blocks or reverses the functional defect in NKT cells in MM patients may enable the immune response to regain control of myeloma cells and perhaps bring patients back into the NPM state for prolonged periods.

Return to Top

Teasing Apart GVH from GVT Reactions with Regulatory T Cells

Edinger M, Hoffmann P, Ermann J, et al. CD4+ CD25+ regulatory T cells preserve graft-versus-tumor activity while inhibiting graft-versus-host disease after bone marrow transplantation. Nat Med. 2003;9:1144-1150.

Robert Lowsky, M.D.

Mature donor T cells cause graft-vs-host disease (GVHD) but are also responsible for the beneficial graft-vs-tumor (GVT) effect. In this paper, Edinger et al., using a murine model of transplantation, showed that by exploiting the immunomodulating capabilities of regulatory T (T reg ) cells, immune responses of conventional T (T conv ) cells can be shaped and modified so as to inhibit GVHD yet preserve GVT activity. The investigators transplanted lethally irradiated mice with T cell depleted marrow from mismatched donors. Fatal GVHD was induced by the addition of donor derived splenic T conv cells. In contrast, animals receiving donor regulatory CD4 + CD25 + T (T reg ) cells, given at a ratio of 1:1 with T conv cells, were protected from GVHD. The T reg cells inhibited early expansion and proliferation of alloreactive T conv cells in GVHD target tissues. As a result of this decrease in numbers of total T conv cells, there was a significant reduction in serum IFN- g and TNF- a levels, two proinflammatory cytokines thought important in mediating GVHD. Of significant interest, addition of T reg cells did not interfere with GVT activity as T conv cells still eradicated syngeneic leukemia or lymphoma cells inoculated at the time of bone marrow transplantation. The authors demonstrated that direct cell-to-cell contact, perforin and Fas-Fas-ligand killing pathways mediate GVT reactions. Thus, it seems that T reg cells have dual abilities. First, they impair T conv cell proliferation in target tissues and consequently reduce levels of proinflammatory cytokines fueling GVHD. Second, they do not impair perforin and Fas-Fas-ligand killing pathways of T conv cells thereby maintaining GVT activity.

The limiting undesirable complication of allogeneic hematopoietic cell transplantation (HCT) has been life threatening acute GVHD and the opportunistic infections that often accompany it. It has been frustrating that strategies to eliminate GVHD have been accompanied by loss of the antitumor effect. Thus, one of the elusive goals in allogeneic HCT has remained the elimination of GVHD while maintaining a robust GVT effect.

The paper by Edinger et al., suggests that co-transplantation of CD4 + CD25 + regulatory T cells isolated from donor peripheral blood or marrow may improve outcomes in clinical allogeneic HCT by preventing GVHD without eliminating GVT activity. The challenge is to now translate these findings to human transplantation. CD4 + CD25 + T cells represent only a minority of the cells within donor grafts; thus conditions for ex vivo expansion need to be determined, as does optimal dosing and timing for their injection. Will post-transplant immunosuppressive therapy (i.e. cyclosporin, FK-506, methotrexate) affect the activity of regulatory T cells? Do shared HLA antigens between donor, recipient and regulatory T cells alter the efficacy of regulatory T cells (the murine experiments involved class I and class II mismatched animals)? Although important questions remain unanswered, this exciting study points to a new avenue for developing approaches to manipulating T cell subsets to benefit human blood and marrow transplantation.

Return to Top

Aspirin, Thrombosis, and New Challenges for Scientists in Polycythemia Vera

Landolfi R, Marchioli R, Kutti J, Gisslinger H, Tognoni G, Patrono C, Barbui T, for the European Collaborationon Low-Dose Aspirin in Polycythemia Vera Investigators.  Efficacy and safety of low-dose aspirin in polycythemia vera.  N Engl J Med 2004; 350:114-124.

By Josef T. Prchal, M.D.

Several presentations at the latest ASH meeting, and in press in several journals including the New England Journal of Medicine, report the results of a large multicenter prospective study, The European Collaboration on Low-dose Aspirin in PV (ECLAP), which included over 1600 polycythemia vera (PV) patients. Low-dose aspirin (100 mg/day) lowered the risk of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, pulmonary embolism, and major vein thrombisis (relative risk 0.40 {95 percent Confidence Interval 0.18 to 0.91}, P=0.03), with only minor risk of bleeding.

PV, one of four recognized myeloproliferative disorders, is an acquired disorder of the pluripotent stem cell that results in clonal expansion of myeloid progenitors (red cells, granulolocytes, and platelets). Since the molecular basis of this disorder is unknown, there is considerable disagreement about its diagnosis and treatment. In the 1960s, the Polycythemia Vera Study Group established that thrombosis and hemorrhage are major causes of morbidity and mortality in this disorder and showed that myelosuppressive therapy decreases thrombosis risk. Attempts to reduce thrombotic complications with full dose aspirin were associated with a higher risk of bleeding and no significant reduction in thrombosis. However, it is now known that lower doses of aspirin are sufficient to fully inhibit synthesis of the platelet aggregation product thromboxane and platelet function in normal and in PV subjects.

Since most of the thrombotic complications in the present study were not prevented, the ECLAP study suggests that only a minor fraction of thromboses are attributable to platelets and raises the question of what is the major mechanism of thrombosis in PV. Are the white cells, which are often elevated, a contributing factor to thrombotic complications? Multivariate analysis of a study of hydroxyurea in sickle cell disease revealed a benefit to decreasing neutrophils in the prevention of sickle cell vascular events. The possibility that cytoreductive agents directly affect platelets is another poorly explored area.

One of the ECLAP results presented at the ASH meeting analyzed additional risk factors in the ECLAP population, including the relationship of hematocrit to thrombosis. Studies by Pearson (Pearson et al. Lancet. 1978;2:1219-1221) which suggested that PV patients with high hematocrits had proportionally greater risk of strokes compared to those with lower hematocrits led to the widely accepted recommendation that PV patients should be phlebotomized to keep their hematocrit below 45 percent (Shibata et al. Blood. 2003 101: 4416-4422). Yet multivariate anayses of the ECLAP data revealed no statistical relationship between hematocrit in the range of 41 to 50 percent and thrombosis (Marchioli, Abstract 100, and Barbui, Education Program; ASH). Further, transgenic mice with extreme polycythemai due to constitutive overexpression of erythropoietin do not suffer from thrombotic complications (Spivak, Blood. 2002;100:4272-4290). In addition, thrombotic complications in other forms of polycythemia (e.g. high altitude, cyanotic congenital heart disease, Chuvash polycythemia) do not occur or occur at much lower frequency than in PV at comparable hematocrits. These observations suggest that the thrombotic complications of PV may be caused by other features of PV, such as quantitative and non-thromboxane-dependent qualitative defects of platelets and neutrophils. It might even be speculated that PV is also a disease of endothelial cells derived from the hemangioblast, but this has not yet been studied. Similarly, macrophage-derived, tissue factor-bearing microparticles (macrophages are also derived from the PV clone) were shown to be a major contributor to arterial thromboses. These microparticles may be yet another unexplored mechanism for thrombotic complications in PV.

Thus it appears that hematocrit is not the sole, and likely not even the major, thrombotic risk factor in PV. Clearly, the clinical observations raised by the ECLAP studies need to be followed by rigorous laboratory studies and should open a multitude of avenues in basic research. Fortunately, the lack of research in this area was recognized by the U.S. State Appropriations Committee, which stated in its report for FY '03 budget that NCI should expand research in myeloproliferative diseases (United States Senate Committee on Appropriations Senate Report Ho 107-2216, page 100), and by NIH (NHLBI, NCI State of the Science Implementation Working Group Meeting on Myeloproliferative, Myelodysplastics, and Marrow Failure Status, March 3, 2003). Only delineation of the molecular defect(s) of PV and elucidation of its pathophysiology will have a major impact on improved understanding and ultimately better prognosis of this enigmatic acquired blood disorder.

Return to Top

Better Weapons for Cancer-Associated Thrombembolism

Lee AYY, Levine MN, Baker RI, Bowden C, et al. Low-Molecular-Weight Heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003;349:146-153.

By J. Douglas Rizzo, M.D.

Patients with cancer experience elevated risk of thromboembolism. Treatment has traditionally been heparin followed by prolonged warfarin, though clinicians have suspected higher recurrence rates in cancer patients treated with vitamin K antagonists. Cancer patients may also have higher risk of bleeding complications during treatment because of thrombocytopenia, tumor invasiveness, or excessive anticoagulation with warfarin due to malnutrition or drug interactions. Whether low-molecular-weight heparin (LMWH) reduces risk of recurrent venous thromboembolism in cancer patients compared to warfarin is addressed by a well-conducted randomized trial published by Lee et al.

Adults with active cancer and symptomatic deep-vein thrombosis (DVT) or pulmonary embolism (PE) were randomly assigned to treatment with subcutaneous dalteparin or oral anticoagulation. All patients began therapy with dalteparin, 200 IU/kg/day. For those randomized to the oral anticoagulant group, warfarin or acenocoumarol was given within 24 hours of randomization, and dalteparin was discontinued after a minimum of five days once the INR was above 2.0 for two consecutive days. Warfarin was continued for six months. The patients assigned to dalteparin received 200 IU/kg/d for one month, then about 150 IU/kg/d for a total of 6 months. 338 patients were assigned to the treatment arms, which were well balanced. Over the study period, patients treated with dalteparin had substantially less risk of recurrent thromboembolism than the oral anticoagulant group (HR=0.48, 95% CI: 0.3 to 0.7). The authors estimated that patients in the oral anticoagulant group were below the targeted INR range about 30% of the time, and nearly 40% of thrombotic events occurred when the INR was less than 2.0. Patients in the dalteparin group experienced a similar rate of major bleeding compared to oral anticoagulants (6% vs. 4%). The rates of fatal/critical bleeding and overall mortality were the same in both groups.

The results of this study hold promise for improving treatment of cancer-associated thromboembolism. However, it also serves to remind of many unanswered challenges. More work is needed to define the pathogenetic mechanisms that increase the risk of thrombosis for cancer patients. These include activation of the coagulation system, platelet activation, endothelial damage, indwelling venous access devices, direct effects of chemotherapy/hormonal therapy, and host inflammatory responses. Pathophysiologic mechanisms are also specific malignancy-dependent. Cost and complications (primarily bleeding) make prophylaxis of all cancer patients difficult to justify. Precise understanding of mechanisms will help identify those cancer patients with highest risk of thromboembolism, who can be reasonably targeted for prophylactic intervention. Similarly, definition of molecular mechanisms for thrombogenesis may stimulate development of new tests to identify early hemostatic activation, prior to clot development. Improved diagnostic testing would further facilitate risk-adjusted prophylactic anticoagulation that would limit exposure to the accompanying risk of bleeding. Furthermore, better identification of mechanisms at the molecular level may lead to development of anticoagulants that interrupt the pathways of thromboembolism unique to malignancies.

Treatment (and by extension, prophylaxis) with heparins appears to be better than vitamin K antagonists, however we need to learn more about the reason for this difference. Is increased risk with warfarin due to inadequate INR maintenance (as shown for 30% of patients in this study), or less effective interruption of the cancer-disrupted coagulation cascade? Are drugs that directly affect thrombin inherently more effective? If so, one can speculate whether the new generation of oral direct thrombin inhibitors (e.g. ximelagatran) will perform at least as well as LMWH and with greater convenience.

Return to Top

Seasonal Variation of Fibrinogen Levels

Hermida RC, Calvo C, Ayala DE, et al. Seasonal variation of fibrinogen in dipper and nondipper hypertensive patients. Circulation 2003; 108:1101-1106.

By Andrew Schafer, M.D.

Plasma fibrinogen levels in a group of 1006 hypertensive subjects exhibited a highly significant seasonal variation. Highest levels were recorded in February and lowest values in September-October. This predictable and stable “circannual” variation of fibrinogen was established for all subgroups studied. The elevated plasma fibrinogen levels correlated directly with risk of vascular events, which are more prominent during the late winter months (see figure).

Increased fibrinogen levels have been strongly associated with risk of acute myocardial infarction and stroke, and also possibly venous thromboembolism. These same vascular and thrombotic disorders have also exhibited winter peaks and summer troughs in incidence and mortality. Future research should be directed at proving a causal relationship between the seasonal variation in fibrinogen levels and the almost superimposable seasonal variation in arterial and venous vascular events. Other studies have demonstrated winter nadirs in levels of free protein S and fibrinolytic activity, supporting the concept that hemostatic balance is shifted toward hypercoagulability in winter months. Further work on assessing seasonal variability in markers of activated coagulation in humans, as well as animal models of thrombosis, including transgenic mice, will be needed to establish a clear cause-and-effect relationship between seasonal variation in fibrinogen levels and risk of vascular events. In the Northwick Park Heart Study, an increase in fibrinogen above one standard deviation (by 59 mg/dL) increased the risk of death from ischemic heart disease by 67% during five years of follow-up. In the current study, the circannual amplitude of fibrinogen levels from February peak to September-October trough was 40 mg/dL. This would extrapolate to a 45% increase in risk during the winter, suggesting the potential for this observation to have significant clinical relevance.

The fascinating findings in this study should also stimulate further research on the reasons for the marked seasonal variation in plasma fibrinogen levels. Seasonal change in environmental temperature may be a factor: but in this study from Spain, seasonal temperature variation was quite small (11 o C). Other potential causes of winter elevation in fibrinogen levels might include increased respiratory infections (which activate acute phase responses), physiologic perturbations by duration of daily light exposure, and reduced physical activity in the winter. Understanding the chronobiology of hemostasis and coagulation, including previously studied circadian and the currently studied circannual (seasonal) variability may lead to the development of preventive and therapeutic anticoagulation strategies directed by thrombosis susceptibility rhythms.

Return to Top

Decreased Mortality of Ischemic Heart Disease Among Carriers of Hemophilia

Sramek A, Kriek M, Rosendaal FR. Decreased mortality of ischaemic heart disease among carriers of haemophilia. Lancet. 2003 Aug 2;362(9381):351-4.

By Samuel M. Silver, M.D., Ph.D.

The pathogenesis of ischemic heart disease is complex. The role of platelets and fibrinogen is discussed in other Diffusion articles in this issue. In a previous study of Dutch patients with hemophilia A or B, there was an 80% reduction in the risk of fatal ischemic heart disease (Triemstra et al. Ann Int Med 123: 823-827, 1995). An interesting paper by Sramek and colleagues now examines the role of clotting factor VIII or IX in the mortality of ischemic heart disease in mothers of patients with hemophilia. Since hemophilia is a genetic recessive X-linked trait, these carriers would be expected to have about 50% of the normal amounts of factor VIII or IX, and their bleeding risk is small. In the present study, the authors made use of the comprehensive Dutch population registry to assess death rates of mothers of hemophiliac patients between 1861 and 2001 and cause of death from 1950 onward. Vital statistics were available for 97% of 1012 women whose sons had hemophilia A or B as identified in a nation wide survey of hemophilia patients in the Netherlands between 1985-1992. The cause of death in 98% of the women who died between 1950-2002 could be determined. Overall mortality during follow-up was reduced by 22% in the hemophilia carriers (261 women died where 333 were expected based upon population figures). Mortality did not differ by the type or the severity of their sons' hemophilia. The cause-specific mortality compared to the general female population differed only in ischemic heart disease, where it was reduced by 36%, and was present in all age groups. Although hemophilia carriers had an increase in mortality from extracranial hemorrhage, the absolute numbers were very small.

This intriguing study raises the question whether reduced, but not dangerously low, concentrations of clotting factors VIII or IX have a protective effect in ischemic heart disease. Life-threateningly low levels of these factors appear to have a protective effect, but can a moderate reduction be helpful, and what can this tell us about the pathogenesis of ischemic heart disease? There appears to be a dose-dependent relationship between factor VIII levels and venous thrombosis on the high end of the scale (levels >100), although the data for arterial thrombosis is less clear (for review see Kamphuisen et al. Arterioscler Thromb Vasc Biol 21: 731-738, 2001). Hemophilia does not appear to affect atherogenesis (Sramek et al. Circulation 104: 762-767, 2001). Therefore, the protective effect might be modulation of intracoronary thrombosis over plaque disruption, a major mechanism in the final stage of ischemic heart disease. This epidemiological study can serve as a model for laboratory-based experiments to elucidate the role of coagulation factors in ischemic heart disease. Heterozygous factor VIII and IX knockout mice can be crossed with murine models of atherosclerosis such as the apolipoprotein-E deficient mice to see if there is inhibition in the progression of the coronary lesions. These mouse models may not reflect what goes on in humans. In fact, the protective effect of a 50% level of factor VIII on a woman's coronary system might not reflect what goes on in men. But the interesting findings in this paper may help us understand the complexities of coagulation in the pathogenesis of ischemic heart disease.

Return to Top

Biologic and Clinical Subsets of Lymphoma: Better Diagnosis, Better Treatment?

Hans CP et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood 2003; prepublished online September 22.

By Michael E. Williams, M.D.

Three molecularly and prognostically distinct subtypes of diffuse large B-cell lymphoma (DLBCL) have been defined by cDNA microarray analysis. This sophisticated molecular analysis, however, is currently impractical for routine clinical use. In an effort to bridge this subtyping from the laboratory to the clinic, Hans and colleagues utilized a panel of six antibodies to assess the immunophenotypic expression of genes that are highly associated with germinal center B-cell-like (GCB) versus activated B-cell-like or “type 3” DLBCL (non-GCB). Representative small cores were obtained from paraffin blocks of 152 cases, most previously classified by cDNA microarray analysis, and placed into a “tissue array” for staining. Samples were considered positive if greater than 30% of the tumor cells expressed the antigen, as determined independently by two pathologists.

GCB vs. non-GCB DLBCL subtype could be accurately predicted with only three antibodies, utilizing cDNA microarray as the gold standard:

Classification by tissue array analysis showed a five year overall survival of 76% for the GCB versus 34% for non-GCB, very similar to the findings with cDNA microarray. Analysis of bcl-2 and cyclin D2 expression provided additional prognostic information in the non-GCB group, with poorer event-free and overall survival when either of these proteins was expressed. These subsets of DLBCL biologic markers appeared prognostically independent of, but complementary to, the clinical IPI score (International Prognostic Index); for example, non-GCB DLBCL with high IPI scores of 3-5 had a median survival of only one year.

DLBCL, the most common non-Hodgkin's lymphoma subtype, has long been recognized as clinically heterogeneous. Approximately 40-50% of patients are cured by primary chemotherapy, another 40% respond but later relapse, while the remaining patients are refractory to initial therapy. Efforts to identify patients at risk of treatment failure have included the IPI score as well as tumor cell expression of certain immunophenotypic markers, such as those for proliferation (e.g., Ki-67 [MIB-1]) or apoptosis (e.g., bcl-2).

In contrast, gene-expression profiling utilizing cDNA microarray technology simultaneously assesses the expression of thousands of genes. Using this powerful technique, clinically relevant subsets of leukemias and lymphomas have been identified (reviewed by L. Staudt, N Engl J Med 2003; 348:1777). Such subtyping identifies patients at increased risk of failure with standard therapies and may enhance risk-adapted therapeutic approaches, including the identification of patients potentially likely to benefit from dose intensive therapy and stem cell transplantation. It will be important for ongoing and future lymphoma clinical trials to incorporate these analyses, to verify their reproducibility, and to standardize methodologies and data interpretation.

cDNA microarray analysis also provides a tool for gene discovery and identification of novel therapeutic targets. For example, expression of protein kinase C-beta (PKC- b ) has been identified in fatal/refractory, but not cured, DLBCL (M.A. Shipp et al, Nature Medicine 2002; 8:68). Inhibitors of this enzyme are now being tested for a potential therapeutic role in these aggressive lymphomas. The fascinating findings of cDNA microarray analyses, and the report from Hans and colleagues, are important steps in moving from the era of empirically-applied lymphoma therapies to those in which immunologic and molecular subtyping will allow treatment to be targeted and modified for individual patients.