The I-IEF pattern from the urine sample obtained immediately following the run detected multiple glycosylated proteins that looked nearly identical to that generated by small quantities of Epoetin-β (Roche), strongly suggesting the presence of this drug in the runner’s blood. This “positive” result was not seen in urine donated as little as one hour after the run. The fact that Epo has a plasma half-life of more than eight hours suggested that the first result was a false positive, perhaps related to the transient post-exercise proteinuria. The authors then retested the “positive” urine samples by a different technique — immunoblotting after protein denaturation and separation and size fractionation by electrophoresis — and found that the protein detected by the “Epo-specific” monoclonal antibody used in both assays was substantially larger than Epoetin-β, and so must be different. Their interpretation, based on the fact that this larger protein was only detected in the sample generated during a period of transient proteinuria, was that this larger protein was a normal serum protein component to which the anti-Epo monoclonal antibody was cross-reacting.
The results of these studies establish Abl as an important factor in the physiology of cardiomyocyte survival. However, it should be kept in mind that functional cardiac failure was reproduced only at an excessively high-dose level of imatinib in mice. In addition, it is of note that there have been recent studies that used lower dosages of imatinib and demonstrated favorable cardio-protective effects1.
From a clinical standpoint, the results of this study are especially relevant as they make physicians well aware of the unanticipated cardiotoxic side effects of imatinib. Currently, the clinical significance of cardiotoxicity arising during/after imatinib treatment remains to be assessed. Clinical cases of cardiac failure have rarely been reported. Future studies will need to sort out if and how frequently cardiotoxicity develops in patients on imatinib, and how it depends on dose level and cumulative dose of imatinib. Furthermore, it is unclear whether the cardiotoxicity might precipitate at greater frequencies in patients who receive additional cardiotoxic therapies (e.g., anthracyclines), or in those with preexisting heart disease. In spite of these open questions, physicians for the time being will need to closely monitor their patients on imatinib for symptoms of left ventricular dysfunction and intervene at the earliest signs of clinical cardiotoxicity.
Nancy Andrews, MD, PhD
Dr. Andrews indicated no relevant conflicts of interest.
Shi Q, Wilcox DA, Fahs SA, et al. Factor VIII ectopically targeted to platelets is therapeutic in hemophilia A with high-titer inhibitory antibodies. J Clin Invest 2006;116:1974-1982.
Hemophilia A remains a vexing clinical problem in spite of our detailed molecular understanding of its pathogenesis. Factor VIII (FVIII) infusion has been the mainstay of therapy for 40 years — initially with partially purified plasma proteins, and more recently with recombinant human FVIII produced with its carrier protein, vWF. Hemophilia has been a darling of experimental gene therapists impressed by the seriousness of the disease and the potential for cure. Relatively modest production of the missing factor provides clear clinical benefit, and there are appropriate animal models for pre-clinical studies. However, although recombinant FVIII protein is effective whether produced in cultured cells or in vivo, up to 30 percent of patients develop inhibitory antibodies that limit its therapeutic potential. In this report, Shi and colleagues report a clever new gene transfer approach, to deliver FVIII to sites of bleeding while keeping it carefully hidden from the immune system.
This paper builds on previous studies showing that hooking up a modified FVIII gene to the glycoprotein Ib alpha (GPIbα) promoter forces the expression of FVIII in platelets, where vWF is already produced1. That strategy corrected bleeding in transgenic mice that lacked FVIII. Shi and colleagues confirmed the earlier results and went several steps further. They showed that platelets engineered to express FVIII could correct hemophilia when transfused into FVIII-null mice. The platelet-produced FVIII even worked in mice that produced high-titer inhibitor antibodies against FVIII, indicating that storage of the factor in platelet alpha granules shields it from the immune system and allows it to be released upon platelet activation at the site of vascular injury. In fact, almost none of the transgenic FVIII protein could be detected in plasma, even though it was clearly effective at preventing bleeding. The authors speculate that this strategy might be used in hemophilia patients, by introducing the GPIbα-FVIII transgene into autologous hematopoietic stem cells, transplanting them back, and relying on expression in mature platelets. The proof-of-concept experiments in mice are very encouraging. But there are still substantial issues to be overcome in human subjects, as discussed in a commentary that accompanied the article2.
1. Yarovoi HV, Kufrin D, Eslin DE, et al. Factor VIII ectopically
expressed in platelets: efficacy in hemophilia A treatment.
Blood 2003;102:4006-4013.
2. High KA. The leak stops here: platelets as delivery vehicles for
coagulation factors. J Clin Invest 2006;116:1840-1842.
Return to Top
Screening Patients with Idiopathic Budd Chiari Syndrome for JAK2V617F
Kenneth Kaushansky, MD
Dr. Kaushansky indicated no relevant conflicts of interest.
Patel RK, Lea NC, Heneghan MA, et al. Prevalence of the activating JAK2 tyrosine kinase mutation V617F in the Budd-Chiari syndrome. Gastroenterology 2006;130:2031-2038.
Over the past year our understanding of the etiology of polycythemia vera (PV), essential thrombocytosis (ET), and idiopathic myelofibrosis (IMF) has taken a quantal leap with the identification of an acquired, activating mutation (V617F) of the hematopoietic growth factor receptor signaling kinase JAK2. In this paper, Patel and colleagues studied 41 patients with idiopathic Budd Chiari Syndrome (BCS) with a JAK2V617F-allele specific, reverse transcriptase polymerase chain reaction (RT-PCR) assay, and found that nearly 60 percent were positive for one or more alleles of the mutant kinase. Of note, clinical and conventional laboratory analysis was felt to rule out a diagnosis of a myeloproliferative disease. Since the JAK2V617F mutation has been found only in patients with myeloproliferative diseases, or very rarely in patients with myelodysplastic or acute leukemic conditions, these results add to our diagnostic tool box for BCS. They also provide an important insight into the capacity of the mutant JAK2 kinase to trigger a hypercoagulable state, and raise a question as to whether other patients with pathological coagulopathies should be screened for JAK2V617F.
The term BCS describes a group of disorders in which hepatic vein thrombosis results in abdominal pain and swelling, ascites, biochemical deterioration of the liver, and the sequelae of acute and/or chronic liver failure. Although mechanical obstruction to blood flow (trauma, tumors) can cause a barrier to hepatic venous outflow and secondary thrombosis, the cause of BCS is traced to a hypercoagulable condition in from one-third to two-thirds of patients. Myeloproliferative disorders are the most commonly recognized hypercoagulable state in published series, followed by exogenous estrogen therapy, paroxysmal nocturnal hemoglobinuria, and disorders of coagulation proteins. However, in most series up to half of the patients do not have a diagnosis established at presentation. JAK2V617F has been identified in virtually all patients with PV, and approximately half of those with ET and IMF, and is felt to contribute to the uncontrolled myeloproliferation characteristic of these disorders. Some have speculated that the mutation might also contribute to the hypercoagulable state that characterizes PV and ET. The work by Patel and colleagues appears to validate this speculation; despite normal levels of all blood cell types, 26 patients in their series developed BCS for no apparent reason other than JAK2V617F. The work also prompts the question: should patients with other types of pathological thrombosis be tested for JAK2V617F?
Return to Top
Immunogenic Cell Death in Cancer: Potential Synergism Between Chemotherapy and Immune Responses
Peter Lee, MD
Dr. Lee indicated no relevant conflicts of interest.
Casares N, Pequignot MO, Tesniere A, et al. Caspase-dependent immunogenicity of doxorubicin-induced tumor cell death. J Exp Med 2005;202:1691-701.
Throughout life, millions of cells die each day in an orderly process (apoptosis) and are replaced by new cells to maintain homeostasis. From an immunologic perspective, apoptosis does not elicit an inflammatory response and is generally thought to be immunologically silent — this is important as otherwise autoimmunity could ensue. Cancer cells overall have a lower propensity to undergo apoptosis. It is now recognized that a dominant mode of cell death caused by chemotherapy is apoptosis. Since chemotherapy also suppresses the immune response, and apoptosis is thought to be immunologically silent, it is generally assumed that chemotherapy-induced cancer cell death does not trigger an immune response.
Recent work by Casares et al. challenges this somewhat simplistic view. In this paper, the authors show that even in the absence of any adjuvant, tumor cells dying in response to doxorubicin (DX) could elicit an effective antitumor immune response that suppressed the growth of inoculated tumor cells and even led to regression of established tumors in immunocompetent mice. In contrast, tumor cells killed by treatment with mitomycin C (MC) were immunologically silent and did not elicit an antitumor immune response. This surprising and important difference would not be expected a priori as both agents induce apoptosis in cancer cell lines. Upon detailed inspection, the authors showed that while both DX and MC induced caspase activation and chromatin condensation, only MC led to detectable DNA fragmentation. To test the immunogenicity of these cells, chemotherapy-treated cancer cells (and necrotic cells induced by freeze/thaw [F/T] as controls) were injected into one flank while live, untreated cells were injected into the opposite flank. Only DX-treated, but not MC- or F/T-treated, cells reduced tumor formation by live, untreated cells by ~ 45 percent on day 30 and ~ 30 percent on day 120. Furthermore, protected animals in the DX-treated group were immune to a second challenge with the same live, untreated cancer cells, but not to an unrelated cancer cell line, and splenic T cells from protected animals could transfer protection to naïve animals. Vaccination with DX-treated apoptotic cells was ~ 80 percent protective after one prophylactic injection prior to challenge with live, untreated cells — this represents a higher degree of protection than MC- or F/T-treated, or γ irradiated cells. Other anthracyclines including daunorubicin and idarubicin conferred similar effects. Taken together, these results indicate that anthracyclines induce a form of apoptosis which is immunogenic and can lead to protective antitumor immune responses. While the mechanisms involved remain unclear, the role of caspase in the immunogenicity of apoptotic cells induced by DX was shown using a broad spectrum caspase inhibitor Z-VAD-fmk. When combined with DX (together referred to as DXZ), DXZ-treated cells exhibited delayed death and were significantly less immunogenic than DX-treated cells. Depletion of dendritic cells or CD8 T cells abolished the immune response against DX-treated tumor cells in vivo. These results suggest that full caspase activation is a necessary step in the immunogenicity of DX-induced apoptosis, and that the response is mediated by dendritic and CD8 T cells.
Together, these results show that not all apoptosis is the same. Importantly, at least one form — that induced by DX and other anthracyclines — may be immunogenic and could lead to protective antitumor immune responses. This may have significant implications in cancer therapy. DX (and other anthracylines) is a common agent in the treatment of acute leukemia. It may be that long-term remission, perhaps cure, in some patients may arise through a synergistic effect of, initially, cytoreduction by chemotherapy, followed by an effective antileukemia immune response that eliminates residual tumor cells, perhaps including cancer stem cells and escape variants. If true, it could be worthwhile to revise anthracycline-containing chemotherapy regimens to minimize their immunosuppressive activity, and potentially couple these with active immunotherapeutic (vaccination) approaches. On the flip side, these data also raise caution that normal cells induced into apoptosis via a DX-like mechanism may lead to autoimmunity. While autoimmunity after anthracycline treatment is rare, other drugs or processes may induce death of normal tissues via an analogous mechanism. Hence, a better understanding into the mechanism underlying the immunogenicity of DX-induced apoptosis could lead to better cancer treatment strategies as well as, possibly, ways to prevent certain autoimmune processes.
Return to Top
Osteoclasts Chew Out Stem Cells
Robert Lowsky, MD
Dr. Lowsky indicated no relevant conflicts of interest.
Kollet O, Dar A, Shivtiel S, Kalinkovich A, et al. Osteoclasts degrade endosteal components and promote mobilization of hematopoietic progenitor cells. Nat Med 2006;12:657-64.
Harvesting adequate numbers of mobilized hematopoietic progenitor cells from the blood of patients or donors is the start point for successful hematopoietic cell transplantation. Strategies that increase the cell yield have significant appeal for transplantation. Osteoclasts, themselves derived from hematopoietic stem cells, carve out the marrow space on the endosteal bone region, the niche where hematopoietic stem cells reside. They have long been considered bystanders, not active participants in hematopoietic progenitor cell circulation. In this paper, Kollet et al. link bone remolding with regulation of hematopoiesis and provide evidence that osteoclasts can mediate hematopoietic progenitor cell recruitment into the circulation.
Following stress-induced situations of bleeding and treatment with lipopolysacchride (LPS), Kollet and colleagues report an increase in osteoclast number and development in endosteal niches in association with mobilization of hematopoietic progenitor cells. The in vivo administration of RANK ligand (RANKL), the osteoclast-differentiating cytokine critical for osteoclastogenesis produced by osteoblasts and thought to link bone production with destruction for balanced skeletal remodeling, was accompanied by increased levels of circulating hematopoietic progenitor cells without affecting neutrophil number (in contrast to mobilization with granulocyte colony-stimulating factor [G-CSF] that induces a significant rise in neutrophil number). Inhibition of osteoclasts with calcitonin reduced progenitor egress during homoeostasis, G-CSF mobilization, and stress situations. The administration of RANKL up-regulated osteoclast expression of the major bone resorbing proteinases MMP-9 and Cathepsin K (CTK) that also were shown to cleave SDF-1 and membrane-bound kit ligand important for stem cell anchorage and quiescence. To further strengthen the relation between osteoclast activation and hematopoietic cell mobilization experiments using mice deficient in the protein tyrosine phosphatase-epsilon (PTPε) that have dysfunctional osteoclasts with no identified hematopoietic disorders, treatment with either RANKL or G-CSF did not result in mobilization of progenitor cells.
The clinical applicability of this study comes quickly to mind — evaluate in animal models of transplantation whether mobilized progenitor cells harvested following osteoclast activation with RANKL restore hematopoiesis in lethally irradiated hosts. If successful, consider osteoclast stimulation using RANKL with or without other mobilizing agents to harvest progenitor cells for use in clinical autologous transplantation protocols, especially in patients deemed “poor mobilizers.” Can osteoclast stimulation with RANKL or its analogue be safely administered to healthy individuals? Does RANKL effect T cell alloimmune reactions? Does RANKL affect cancer stem cells? Concern remains. Some cancers express RANK receptors and have been shown to metastasize to bone marrow niches when RANKL is in large supply. In the long-term understanding, the dynamic between niche and stem cells extend beyond the need for better mobilization strategies for use in hematopoietic cell transplantation. Rather, knowledge about the mechanism(s) that governs stem cell localization and circulation underlies the processes of tissue development, maintenance, regeneration, and repair. Currently, however, we will have to be content with knowing that when osteoclasts get activated, they can chew out stem cells.
Return to Top
DIgging Up the Details of the Mechanism of Action of IVIg in ITP
Lilli Petruzzelli, MD, PhD
Dr. Petruzzelli indicated no relevant conflicts of interest.
Siragam V, Crow AR, Brinc D, et al. Intravenous immunoglobulin ameliorates ITP via activating Fcgamma receptors on dendritic cells. Nat Med 2006;12:688-692.
IVIg has been used across a spectrum of autoimmune disease with success in idiopathic thrombocytopenic purpura (ITP) but has been of limited use in other diseases such as systemic lupus erythematosus (SLE) or rheumatoid arthritis. In this manuscript, the authors probe the mechanism of IVIg’s efficacy in ITP and hypothesize, based upon their findings presented here and those in an earlier manuscript, a model to refine therapy in ITP and explain why IVIg may not work in those diseases where there is evidence of significant circulating immune complexes. Here, the investigators use a mouse model and generate ITP by the injection of an antibody to platelets and take advantage of mouse strains where components thought to be important in modulating ITP have been deleted. Previous work by others demonstrated that the inhibitory Fcγ receptor, FcγRIIB is critical to the action of ITP and that increased expression of this receptor on macrophages is associated with reduced platelet clearance. In this work, the investigators demonstrate that cells of the innate immune system, but not lymphocytes, are central for mediating the effect of IVIg on ITP. Expression of FcγRIIB in the recipient is important for treating ITP but its expression is not necessary on the leukocytes. Activating signals from other Fc receptor family members are shown to be involved in IVIg action and an antibody that directly activates FcγIIIA on leukocytes treats ITP. Crosslinking of this receptor diminishes ITP even in those mice where FcRγIIB is not expressed. Furthermore, direct activation of the Fcγ chain ameliorated ITP. Selection of splenic leukocytes with antibodies established that only CD11c + dendritic cells ameliorated ITP and other leukocytes or isolated macrophages did not.
The overall model established from this manuscript is that IVIg works by interacting with Fcγ receptors on dendritic cells and, more specifically, with the activating FcRIIIA. The work points to a functional role for the inhibitory receptor in their mouse model, but this may occur as a downstream event in IVIg action. Of note is that IVIg does not work in diseases where there is a high level of circulating immune complexes such as SLE. One hypothesis to explain this is that high levels of circulating antibodies may result in down modulation of either expression or activating signals of the Fcγ receptors and there is little benefit to giving IVIg in these cases. Although this is a plausible explanation, direct proof of this has yet to be obtained.
The work presented here focuses on a mouse model of ITP; nonetheless, the findings are of interest because not only do they suggest a model whereby IVIg works, but they also shed light on our understanding of why this therapy gives such variable effects in immune-mediated processes. Furthermore, the work sets forth potential areas to develop more refined therapeutic agents. It may be feasible to directly activate FcRIIIA by a specific antibody. Although direct activation of dendritic cells holds appeal, the isolation, treatment, and reinfusion may be less attractive than targeting the FcγIIIA receptor. The next step will involve applying these findings to human diseases, assessing whether treatment of human leukocytes with FCRIIIA antibodies results in activation of signals or whether the gamma chain needs to be directly targeted. Moving away from the complex issues associated with IVIg to a more specific and targeted therapy is an attractive goal.
Return to Top
The Diverse Functions of Lipocalin — a Recently Recognized Mediator of Transferrin-Independent Iron Transport, Innate Immunity, and Cancer Signaling
Emanuel Necas, MD, PhD, and Josef Prchal, MD
Drs. Necas and Prchal indicated no relevant conflicts of interest.
Devireddy LR, Gazin C, Zhu X, et al. A cell-surface receptor for lipocalin 24p3 selectively mediates apoptosis and iron uptake. Cell 2005;123:1293-305.
Berger T, Togawa A, Duncan GS, et al. Lipocalin 2-deficient mice exhibit increased sensitivity to Escherichia coli infection but not to ischemia-reperfusion injury. Proc Natl Acad Sci USA 2006;103:1834-9.
Until several years ago, iron transport was thought to be relatively simple and straightforward. Until the discovery of lipocalin, iron delivery to tissues was considered to be exclusively mediated by transferrin and its receptor. Lipocalin, a neutrophil-secreted protein, was found to be an iron-binding protein able to remove iron from bacterial siderophores1. Siderophores are used by some microorganisms to steal iron from the transferrin of their mammalian hosts, and lipocalin was found to be an iron-binding protein able to bind bacterial siderophores1, thus denying essential iron to bacteria. In addition, lipocalin provided an answer to the mystery of the non-lethality of atransferrinemia, since lipocalin was shown to transport iron from mammalian epithelial and ureteric bud cells2 and deliver it to a different cellular endosome than that of transferrin and its receptor cellular pathway3. In the first manuscript, Devireddy and colleagues show that lipocalin is even more multifaceted, as it appears to be involved in the apoptotic pathway. The authors show that at least some of the anti-apoptotic properties of bcr/abl are mediated by lipocalin. In the second manuscript, Berger and colleagues, using a lipocalin knockout mouse (Lcn-/-), further specify lipocalin’s role in response to various pro-apoptotic stimuli. They reproduced the previously reported role of lipocalin in innate immune protection against E.coli4, but they show that lipocalin’s role in apoptosis does not extend to other pro-apoptotic stimuli, as lipocalin did not prevent tissue ischemic injury.
Clearly, more remains to be learned about lipocalin’s role in innate immunity, but these new discoveries of lipocalin’s function raise important concerns for everyday medical practices. The injudicious use of iron for any undiagnosed anemia may worsen the outcome of bacterial and mycobacterial infections4,5. In contrast, the iron chelator desferroxamine can be used as a source of iron by fungus from the Rhizopis family with at times disastrous clinical outcomes5. In addition to the well-known Warburg effect mediated by HIF-1α modulation by iron containing proline hydroxylases and the effect of iron chelators on cancer progression6, lipocalin’s effect on cancerogenesis reveals yet another mechanism of cancer outcome influenced by iron manipulation. A new role for iron in cancerogenesis has emerged, namely the lipocalin-iron pathway that directly regulates apoptosis of certain cancers.
1. Goetz DH, Holmes MA, Borregaard N, et al. The neutrophil lipocalin NGAL is a bacterio-
static agent that interferes with siderophore-mediated iron acquisition. Mol Cell
2002;10:1033-43.
2. Yang J, Goetz D, Li JY, et al. An iron delivery pathway mediated by a lipocalin. Mol Cell
2002;10:1045-56.
3. Kaplan J. Mechanisms of cellular iron acquisition: another iron in the fire. Cell 2002;111:603-6. Review.
4. Flo TH, Smith KD, Sato S, et al. Lipocalin 2 mediates an innate immune response to
bacterial infection by sequestrating iron. Nature 2004;432:917-21.
5. Barash J, Mori K. Iron Thievery. Nature 2004;432:811-812.
6. Richardson DR. Molecular mechanisms of iron uptake by cells and the use of iron chelators for the treatment of cancer. Curr Med Chem 2005;12:2711-29. Review.
Return to Top
Coordination of the Systemic Inflammatory Response by Brain-Spleen Communication
Roy Silverstein, MD
Dr. Silverstein indicated no relevant conflicts of interest.
Huston JM, Ochani M, Rosas-Ballina M, et al. Splenectomy inactivates the cholinergic antiinflammatory pathway during lethal endotoxemia and polymicrobial sepsis. J Exp Med 2006;203:1623-8.
A research team led by Kevin Tracy at North Shore-LIJ Health System in New York has recently discovered and characterized an anti-inflammatory pathway mediated by acetylcholine (Ach) interactions with specific α7 nicotinic receptors on macrophages. They termed this the “cholinergic anti-inflammatory pathway” and showed that the central nervous system participates by releasing Ach from the efferent vagus nerve. In this paper, they showed that the spleen is the principal recipient of these efferent vagal signals. Utilizing well accepted rodent models of systemic sepsis, they first showed, perhaps surprisingly, that the spleen is the major source of TNF in the blood in systemic sepsis. In response to infusion of lethal doses of endoxin, TNF production in the spleen increased 30-fold, while liver and lung production increased modestly by two- to six-fold. Vagus nerve stimulation decreased splenic TNF production by >90 percent with little or no effect on lung or liver TNF production, and animals that were splenectomized prior to receiving endotoxin had 80 percent less circulating TNF than animals with intact spleens. Furthermore, vagus nerve stimulation did not inhibit systemic TNF production in splenectomized animals or in animals subject to selective vagotomy of the common celiac branches that enervate the spleen. Using genetic knockout mice, they showed that vagus nerve-mediated down-regulation of splenic TNF production and circulating TNF levels required α7 nicotinic receptors and involved both macrophages and splenocytes. The cholinergic anti-inflammatory pathway could be activated pharmacologically as well as by electrical vagal stimulation. Administration of nicotine to mice protected against the lethality of bacterial sepsis or endotoxin infusion. This beneficial result was lost if the mice were splenectomized or if the α7 nicotinic receptor gene was knocked out.
As all practicing hematologists know, severe sepsis syndromes are a significant cause of morbidity and mortality in patients with hematologic malignancies or in patients undergoing hematopoietic progenitor cell transplantation. It is well recognized that uncontrolled cytokine production and release plays a critical role in the pathophysiology of these syndromes. These new studies point to an important role for a “neuro-splenic-immuno axis” in providing counterregulatory anti-inflammaotry signals to blunt the response of the innate immune system to bacterial sepsis. Their work shows that Ach signals mediated by α7 nicotinic receptors on macrophages and splenocytes inhibit activation of the “master” inflammatory transcription factor NF-κB and thereby down-regulate production of potent pro-inflammatory cytokines including TNF. Somehow the brain, via the afferent vagus nerve, senses the presence of potentially lethal sepsis and sends a signal via the efferent vagus nerve to the spleen to shut down production of inflammatory cytokines. Since circulating leukocytes traverse the spleen rapidly and regularly, this mechanism has the capacity to provide systemic “education” to the innate immune system. These studies may explain in part why patients without spleens are more likely to develop sepsis syndromes in the setting of certain bacterial infections. Whether pharmacologic therapy (e.g., nicotine patch) or implanted vagus nerve stimulators can be of benefit in ameliorating or preventing sepsis syndromes is an idea that could be readily tested in clinical trials. The authors provocatively point out that more than 25,000 people have been treated with implantable vagus nerve stimulator devices for intractable seizures without significant complications.
Return to Top
Vaccine Therapy for Lymphoma: A Smart Role for Idiotypes?
Michael Williams, MD
Dr. Williams indicated no relevant conflicts of interest.
Redfern CH, Guthrie TH, Bessudo A, Densmore JJ, et al. Phase II trial of idiotype vaccination in previously treated patients with indolent non-Hodgkin's lymphoma resulting in durable clinical responses. J Clin Oncol 2006;24:3107-12.
Each B-cell non-Hodgkin lymphoma (NHL) expresses a unique surface immunoglobulin that may serve as a target for immunotherapy. In this paper, Redfern and colleagues report the findings of a phase II trial wherein lymph node biopsies obtained from patients with relapsed indolent lymphoma were utilized for the isolation and cloning of the unique immunoglobulin “idiotype” (Id) sequences. These Id proteins were conjugated to KLH as an immunostimulant and administered subcutaneously on day 1 of each treatment, with injections of the cytokine GM-CSF at the vaccine site on days 1-4. Vaccine was administered monthly for six months, with booster doses every other month for the next year and then every three months for patients without disease progression. This trial, unlike most other vaccine studies, did not utilize induction chemotherapy or rituximab therapy prior to vaccination and thus provides insight as to single-agent activity and the ability to generate vaccine-specific humoral and cellular immune responses in these relapsed NHL patients. Patient-specific Id/KLH vaccines were generated for each of the 40 eligible patients, although eight patients did not receive vaccine due to disease progression or initiation of other therapy prior to vaccine generation. Of the 32 patients who received at least one dose of vaccine, one complete and three partial remissions were noted; the median time to response was 5.9 months. An additional 21 patients were reported to have stable disease. The median time to progression for all patients was 13.5 months, and at least 28 months for the four responders with ongoing response at 44+ months for the CR patient. The most common side effect of treatment was transient mild or moderate injection site reaction. Among the subset of patients assessed for immune response, 80 percent developed a T-cell and 89 percent a humoral anti-KLH response, with 67 percent and 20 percent developing T-cell or humoral anti-Id responses, respectively.
Lymphoma Id vaccine therapy was originally developed by Levy and colleagues at Stanford University, who showed durable clinical responses in a subset of NHL patients. Although the original techniques for creation of these personalized vaccines was labor-intensive and could take six months or more, current recombinant technologies have high success rates for generating the Id proteins while shortening the time required for manufacture to only two to three months. Preliminary results of phase II Id vaccine trials recently have been presented (Table), and two multicenter phase III trials recently completed accrual with a third nearing completion. The results of the present study confirm the ability to generate Id vaccines for the majority of patients with relapsed indolent NHL. Furthermore, the vaccines are well tolerated and may lead to clinical responses, and both cellular and humoral immune responses can be generated. Weng, Czerwinski, and Levy recently summarized the Stanford experience for 180 follicular lymphoma patients treated with Id vaccination between 1988 and 2002; 36 percent developed a humoral and 24 percent a cellular immune response1. The development of a cellular response was enhanced when either dendritic cells or GM-CSF were employed as an adjuvant for vaccination. Additional approaches to enhance therapeutic efficacy as well as cellular and humoral immune responses are currently being pursued. The results of the multicenter phase III Id vaccine studies noted above are awaited with interest to determine their ability to improve long-term disease control for indolent lymphomas.
1. Weng WK, Czerwinski D, and Levy R. Use of dendritic cells and
GM-CSF adjuvant are associated with anti-idiotype cellular immune
response following idiotype vaccination in follicular lymphoma
patients. Blood (ASH Annual Meeting Abstracts), Nov 2005;106:771.
Return to Top
D-Dimer and a Little “Horse-Sense”
Charles Abrams, MD
Dr. Abrams indicated no relevant conflicts of interest.
Kearon C, Ginsberg JS, Douketis J, et al.; Canadian Pulmonary Embolism Diagnosis Study (CANPEDS) Group. An evaluation of D-dimer in the diagnosis of pulmonary embolism: a randomized trial. Ann Intern Med 2006;144:812-21.
In this paper, Kearon et al. describe the findings of the Canadian Pulmonary Embolism Diagnostic Study (CANPEDS) group. This study was designed to determine if patients with a low clinical suspicion for a pulmonary embolism who had a normal D-dimer level required any further diagnostic testing. Previous trials have suggested that omission of lower extremity Doppler ultrasounds, chest CT scans, or ventilation–perfusion scans is acceptable in low-risk patients, but this has not been demonstrated in a randomized controlled trial.
The CANPEDS trial enrolled 1126 patients with suspected pulmonary embolisms. Erythrocyte agglutination D-dimer tests were normal in 456 of these patients. Using a common risk-assessment score, 373 of the D-dimer negative patients were considered to have a low clinical probability of a pulmonary embolism. The subjects were randomized to undergo no further testing or undergo additional diagnostic testing such as lung scans, serial ultrasonography, or pulmonary angiography. The incidence of a pulmonary embolism was extremely low and not significantly different between the two randomized groups. The CANPEDS also analyzed whether D-dimer testing was useful in patients considered to have moderate or high clinical probability of a pulmonary embolism, but accrual of these subjects was too small to make any definitive conclusions.
Without the assistance of good old-fashioned horse sense by a physician, no single diagnostic test can effectively diagnose or exclude the possibility of a pulmonary embolism. Although CT technology is improving, it is better at “ruling-in” a pulmonary embolism than definitively excluding one. In contrast, D-dimer levels are better at “ruling-out” a pulmonary embolism than they are at clinching the diagnosis. Thus the physician must first determine the clinical probability of this diagnosis before actually choosing the most appropriate diagnostic test. Since only 10-25 percent of patients considered to potentially have a pulmonary embolism actually have one, it would be useful to find a way to quickly dismiss this possibility in a large proportion of patients.
Wells and colleagues proposed a method to divide patients into low-, intermediate-, and high-risk groups1. A simplified version is shown in Figure 1A. Over time, several investigators proposed that merely dividing patients into low-risk (score 4 or less) or high-risk categories is sufficient. Although risk assessment methods such as these are helpful, they still heavily depend on clinical acumen to determine whether a deep venous thrombosis is likely, and whether a diagnosis other than pulmonary embolism is also probable.
The D-dimer assay has been proposed as the ideal complement to the physician’s clinical assessment. D-dimers are one type of degradation product of polymerized fibrin and are elevated in the plasma of the majority of patients with deep venous thrombi or pulmonary embolisms. It has been proposed, but never previously demonstrated in a randomized trial, that patients with a low clinical suspicion of a pulmonary embolism and a normal D-dimer do not need additional testing to exclude this disease.
The results of the trial by CANPEDS group were predictable, and therefore reassuring. At last there is Level 1 evidence that clinicians can stop evaluating patients for a pulmonary embolism if the clinical suspicion is low and the D-dimer is normal. This is applicable to the approximately two-thirds of patients who are thought to be in the low-risk category for pulmonary embolisms. Regardless of the D-dimer, the moderate- and high-risk patients will still need additional testing with CT scans or low extremity Doppler studies. This approach is outlined in Figure 1B. However, despite the utility of D-dimer testing, the diagnosis of pulmonary embolism still relies heavily on the clinical acumen and “horse sense” of the treating physician.
1. Wells PS, Anderson DR, Rodger M, et al. Derivation of a simple clinical model
to categorize patients’ probability of pulmonary embolism: increasing the models’
utility with the SimpliRED D-dimer. Thromb Haemost 2000;83:416-20.
Return to Top
Return to Table of Contents
