A Stake Through the Heart of Cardiolipin
By Charles Abrams, MD
Dr. Abrams indicated no relevant conflicts of interest.
Galli M, Borrelli G, Jacobsen EM, et al. Clinical significance of different antiphospholipid antibodies in the WAPS (warfarin in the antiphospholipid syndrome) study. Blood. 2007;110:1178-83.
Some patients with antiphospholipid antibodies have a clinical syndrome composed of thrombosis, fetal wastage, and, frequently, autoimmune cytopenias, livedo reticularis, and marantic endocarditis. In this report by Galli and colleagues, the authors analyzed the significance of a variety of antibodies found in patients with persistent antiphospholipid antibodies who participated in an international and prospective registry. Confirming the observations of previous studies, this analysis found that antibodies directed against ß2-glycoprotein I (ß2-GPI), or prothrombin, were associated with recurrent thrombosis and pregnancy loss. This analysis also confirmed some, but not all, previous studies demonstrating that anticardiolipin antibodies were not significantly associated with either of these two adverse events.
|
Figure 1: Anticardiolipin antibodies alone are not a risk factor for thrombosis. Some patients with lupus anticoagulants have thrombi and some also have anticardiolipin antibodies. Anticardiolipin antibodies are not associated with thrombi unless a lupus anticoagulant is also present.
|
Antiphospholipid antibodies are a heterogenous group of autoantibodies directed against phospholipid-binding proteins. Some antiphospholipid antibodies will induce their target protein to soak up the small amount of phospholipid added to plasma during the PT and aPTT assays. These particular antibodies will prolong phospholipid-dependent coagulation tests and are called lupus anticoagulants. Other phospholipid antibodies are directed against cardiolipin, a phospholipid component of mitochondria originally described in bovine cardiac cells.
Approximately five to 10 percent of healthy people will have detectable antiphospholipid antibodies in their blood, although only a minority of people will have the persistent presence of these antibodies and have thrombotic or adverse pregnancy events. Since only a minority of patients with antiphospholipid antibodies have symptoms, several studies have investigated whether certain types of phospholipid antibodies better predict the subset of patients that will develop the full syndrome. The study by Galli, et al. confirms previous studies showing that the correlation between the antiphospholipid antibody syndrome and autoantibodies is most prevalent for lupus anticoagulants and anti-ß2-GPI (a common phospholipid-binding protein). But, what is the significance of antibodies against cardiolipin?
The Physicians’ Health Study produced the most widely quoted evidence that anticardiolipin antibodies are associated with venous thrombosis (Annals of Internal Medicine 1992;117:997-1002). In this large cohort of 22,071 male physicians, subjects with anticardiolipin antibody titers greater than the 95th percentile had an approximately five-fold increased risk of DVTs and PEs. Since this publication, the significance of these antibodies has been readdressed in numerous trials. Several of these trials are quite noteworthy because of their large size and findings that contradicted the Physicians’ Health Study. The LITE study, another very large trial of 21,680 subjects, found no increased risk of venous thrombosis in participants with greater than the 95th percentile of anticardiolipin antibodies (Br. J. Haematology 2002;119:1005-1010). Also notable is the lack of association between anticardiolipin antibodies and venous thrombosis in the HUNT trial that derived data from the 66,140 participants in the Health Study of Nord-Trondelag in Norway (J. Thrombosis and Haemostasis 2006;4:44-49). In a meta-analysis, again anticardiolipin antibodies were not associated with venous thrombosis (Blood 2003;101:1827-1832).
What should we do with patients who have anticardiolipin antibodies and thrombosis? Certainly, in patients with recurrent thrombosis, fetal loss, livedo reticularis, and autoimmune cytopenias, the presence of anticardiolipin antibodies merely confirms the clinical diagnosis of an autoimmune-mediated thrombotic disorder. A much tougher decision has to be made when we find the presence of anticardiolipin antibodies in patients who have their first DVT, but no other laboratory tests or symptoms consistent with the antiphospholipid syndrome. If we let evidence be our guide, there is little to support treating such a patient any differently from the patient with a DVT and no laboratory abnormalities.
Return to Top
Targeting Multiple RTKs in Cancer
By Kenneth Anderson, MD
Dr. Anderson indicated no relevant conflicts of interest.
Stommel JM, Kimmelman AC, Ying H, et al. Coactivation of receptor tyrosine kinases affects the response of tumor cells to targeted therapies. Science. 2007;318:287-90.
Therapies targeting receptor tyrosine kinases (RTKs) have provided remarkable responses in both hematologic cancers and solid tumors, but their clinical efficacy has been limited in many cases, and few, if any, patients are cured. In a recent seminal work that has identified obstacles to meaningful responses to single-agent therapies targeting RTKs, Stommel and colleagues have utilized a glioblastoma (GBM) model to examine coactivation of RTKs triggering downstream phophatidylinositol 3-kinase (PI3K) signaling in cell lines, xenotransplants, and primary patient tumor cells. Remarkably, using an antibody array to assess for phosphorylation of 45 RTKs, they found three or more activated RTKs in each tumor, and up to 10 activated RTKs in some cases. In elegant experiments to assess the biologic and clinical relevance of this upstream multiple RTK coactivation, they utilized combinations of RTK inhibitors and/or RNA interference to examine the effects on PI3K signaling, cell survival, and anchorage-independent GBM growth. Importantly, combinations of RTK inhibitors, but not single agents, inhibited PI3K signaling and these functional sequelae.
These studies have fundamental implications for our translational studies to treat cancer. First, in the laboratory, they suggest that individual patients’ tumor cells may need to be profiled in order to identify specific activated RTKs and then select cocktails of targeted therapies in order to achieve meaningful inhibition of cell signaling. Alternatively, it may be possible to choose a single agent targeting multiple RTKs selected for individual patients. A third possibility is to target downstream molecules in cell signaling cascades, thereby theoretically resulting in significant inhibition of relevant signaling pathways and sequelae, regardless of the number of upstream RTKs coactivated. Importantly, translation of these findings will inform the design of clinical trials to evaluate targeted therapies in patients most likely to respond, avoiding the development of drug resistance and allowing for fewer side effects.
The paradigm of combination therapies to cure childhood acute lymphocytic leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, and testicular cancer is well established. Most importantly, the current study suggests that combinations of targeted therapies directed at activated RTKs, or single agents targeting multiple activated RTKs, may allow for more broad application of this paradigm of successful combination therapies in cancer.
Return to Top
Transplantation Leads to Fertility
By Nelson Chao, MD
Dr. Chao indicated no relevant conflicts of interest.
Lee HJ, Selesniemi K, Niikura Y, et al. Bone marrow transplantation generates immature oocytes and rescues long-term fertility in a preclinical mouse model of chemotherapy-induced premature ovarian failure. J Clin Oncol. 2007;25:3198-204.
Infertility is a devastating adverse effect of systemic chemotherapy. Hematopoietic cell transplantation (HCT) is now a treatment for chemotherapy-induced ovarian failure. HCT is the treatment of choice for many malignancies and non-cancerous conditions. Full donor chimerism is frequently seen in the hematopoietic system. Recent experimental models and studies in HCT recipients have documented donor cells in vascular endothelium, various organs, and even in the recipient’s fingernails. This recent report from Lee, et al. adds another twist to these observations.
We have known that the preparatory regimens for HCT are associated with near absolute ovarian failure and infertility due to the ablative regimens utilized (although, there have been reports of pregnancies especially in younger women). The current dogma is that women are born with a fixed number of oocytes that are not replenished and that the chemotherapy with or without radiation destroys the oocytes. This dogma was challenged in 2005 when investigators reported in a controversial manuscript that adult mice continued to produce new eggs and that these eggs could have arisen from bone marrow cells.1 In this current study, Lee, et al. tested if and how HCT (in this case using murine marrow cells) could restore fertility in mice treated with chemotherapy. Females administered nonlethal doses of chemotherapy (12 mg/kg of busulfan and 120 mg/kg of cyclophosphamide) without HCT became infertile, with the majority (10 of 13 mice) achieving three or fewer live-birth pregnancies and none achieving six live-birth pregnancies. On the other hand, 90 percent of the mice that received HCT one week after the same chemotherapy achieved at least four live-birth pregnancies, 80 percent (eight of 10 mice) achieved five pregnancies, and 70 percent (seven of 10 mice) achieved six pregnancies. It is important to note that in this preclinical model, the mice were six to 10 weeks old. (They were quite young.)
|
Figure 1: (A) Percentage of mice receiving vehicle (n = 10), nonlethal chemotherapy (CTx-low, n = 13), or nonlethal CTx followed by bone marrow (BM) transplantation (BMT) 1 week later (CTx-low + BM/1 week, n = 10) that achieved the indicated number of successful (live-birth) pregnancies over a 7-month period when mating was initiated coincident with BMT.
Figure from Lee HJ, Selesniemi K, Niikura Y, et al. J Clin Oncol. 2007;25:3198-204. Reprinted with permission from the American Society of Clinical Oncology. |
The data demonstrate that if HCT is performed within one week of the chemotherapy, the procedure leads to restoration of long-term fertility. To the investigators’ surprise, while they were able to track a few donor-derived oocytes in the ovaries, the pups were of recipient origin (none were of donor origin). What happened in these studies? Is the ovary similar to the testes where the germ cells continuously generate new gametes? The data suggest that the infusion of donor bone marrow cells led to the rescue of a significant number of recipient follicles. These data are reminiscent of effects of other cell populations, such as mesenchymal stem cells, where the effect is that of repair or recruitment of endogenous host cells that allow for repair of the cell of interest or of the stromal support cells. The data raise many questions. Is the infusion of bone marrow cells simply enhancing clearance of damaged cells, decreasing inflammation, re-establishing stromal/nurse cells, enhancing overall health, recruiting important host cells to help with fertility, or restarting the production of oocytes? Regardless of the answers that will come with further research, if bone-marrow-derived cells can help preserve fertility, this could go a long way in mitigating one of the most devastating adverse events for young women receiving systemic chemotherapy.
- Powell, K. Skeptics demand duplication of controversial fertility claim. Nat Med. 2005:11;911.
Return to Top
NF-κB Profiling in Multiple Myeloma
By Steven Grant, MD
Dr. Grant indicated no relevant conflicts of interest.
Annunziata CM, Davis RE, Demchenko Y, et al. Frequent engagement of the classical and alternative NF-kappaB pathways by diverse genetic abnormalities in multiple myeloma. Cancer Cell. 2007;12:115-30.
The NF-κB pathway involves a family of transcription factors (p50, p52, c-Rel, p65/RelA, and RelB) involved in the regulation of diverse cellular processes, including survival, proliferation, differentiation, and inflammatory responses, among numerous others. This cascade can be subdivided into three components: the classical or canonical pathway, the alternative or non-canonical pathway, and the atypical pathway. In the classical pathway, an inciting stimulus, such as activation of TNF-related cell surface receptors, leads to activation of the IKK complex, which phosphorylates the IκBα protein, resulting in its proteasomal degradation. Under normal conditions, IκBα traps p65/RelA in the cytoplasm; hence, IκBα degradation results in RelA nuclear translocation and DNA binding, culminating in the transcription of numerous NF-κB-dependent genes, including those encoding survival (for example, XIAP, Bcl-xL) and antioxidant proteins (for example, MnSOD2), among others. The NF-κB pathway is hyperactivated in diverse neoplastic diseases, particularly those of hematopoietic origin. For example, human leukemia cells and leukemia stem cells have been shown to require an intact NF-κB pathway for survival. Notably, the survival of multiple myeloma cells appears to be particularly dependent upon NF-κB signaling, and it has long been suggested that targeting the NF-κB pathway might represent a very logical therapeutic strategy in this disease. The success of the proteasome inhibitor bortezomib, which among other actions spares IκBα from proteasomal degradation, in patients with advanced multiple myeloma provides strong support for this notion.
In a recent study appearing in Cancer Cell, Annunziata, et al. surveyed a large number of myeloma cell lines and patient samples for evidence of NF-κB-activating mutations. They found a very high incidence of such mutations (i.e., 15-20 percent), which took multiple forms, including translocations involving or amplifications of NF-κB-activating genes such as NIK, or mutations, deletions, or silencing of NF-κB negative-regulatory genes such as TRAF3 or CYLD. Interestingly, there was a correlation between the presence of such genetic abnormalities and NF-κB hyperactivation with the susceptibility of cells to the antiproliferative and death-inducing effects of agents, such as IKKß inhibitors (e.g., MLN120B), or proteasome inhibitors (e.g., bortezomib). In a companion study in the same journal, Keats, et al. reported a high percentage (e.g., ~20 percent) of such mutations in primary patient samples, particularly those involving TRAF3, many of which were associated with activation of the non-canonical NF-κB pathway.1
The significance of these studies lies not only in their implications for our understanding of the pathogenesis of multiple myeloma and related hematologic malignancies, but also in their potential to have profound ramifications for attempts to develop more rational targeted therapy in these disorders. In this context, genetic profiling of diffuse lymphocytic B-cell lymphoma (DLBCL) has recently identified specific subtypes (for example, GC vs. ABC), which differ in their response to therapy, and perhaps not coincidentally, their dependence upon the NF-κB pathway.2 It is therefore plausible to propose that in the future, genetic profiling of multiple myeloma will not only provide us with important prognostic information, but may also facilitate the development of more rational targeted approaches.
- Keats JJ, Fonseca R, Chesi M, et al. Promiscuous mutations activate the noncanonical NF-kappaB pathway in multiple myeloma. Cancer Cell. 2007;12:131-44.
- Bea S, Zettl A, Wright G, et al. Diffuse large B-cell lymphoma subgroups have distinct genetic profiles that influence tumor biology and improve gene-expression-based survival prediction. Blood. 2005;106:3183-90.
Return to Top
Improved Resolution of Stem and Progenitor Cells Reveals Increased Complexity Within the Hematopoietic Hierarchy
By Diane Krause, MD, PhD
Dr. Krause indicated no relevant conflicts of interest.
Pronk C, Rossi D, Månsson R, et al. Elucidation of the phenotypic, functional, and molecular topography of a myeloerythroid progenitor cell hierarchy. Cell Stem Cell. 2007;1:428-42.
This manuscript defines the surface phenotype of different hematopoietic stem and progenitor cell subpopulations with higher resolution than achieved previously. With this further refinement, the investigators reveal more plasticity within the hematopoietic differentiation hierarchy than has been previously recognized.
Prior excellent studies have shown enrichment for murine hematopoietic stem cells (HSCs) and various partially committed progenitor cells, such as common lymphoid progenitors and common myeloid progenitors. These investigators further purify these populations using additional cell surface proteins, including CD105 (endoglin) and CD150 (signaling lymphocytic activation molecule family member 1, SLAM1), which play roles in signal transduction. HSCs, previously known to be Lin-Kit+Sca+, were recently shown by Sean Morrison’s group to be further enriched by selection for CD150 expression. To test the hypothesis that CD105 and CD150 might also be differentially expressed within hematopoietic progenitor subsets, the investigators used flow cytometry to sort subpopulations based on surface expression of lineage markers Sca, Kit, CD150, CD105, and CD41. They extensively analyzed each subpopulation for morphology, in vitro and in vivo function, and gene expression profile.
Among the important findings presented, they show that the Lin-Kit+Sca1- population, which was previously known to be enriched for bipotent megakaryocytic erythroid precursors (MEPs), is CD41- and is comprised of at least three subpopulations based on CD105 and CD150 expression. The CD105+CD150- and CD105+CD150+ cells within this population are already erythroid-committed, and the CD105-CD150+ cells are truly biphenotypic with the ability to differentiate down the erythroid and megakaryocytic lineages. This CD105-CD150+ population represents less than 20 percent of the cells previously called MEP and is referred to as PreMegE. In vitro assays of colonies expanded from single PreMegE cells, a highly rigorous and labor-intensive approach, demonstrated that they were biphenotypic megakaryocytic and erythroid precursors. It remains to be seen whether the corresponding human hematopoietic subpopulations share the phenotypes identified.
Additional analysis of the different blood cell types that differentiated from single cells of the different sorted subpopulations adds to the complexity of current models of the hematopoietic hierarchy. For example, Lin-Kit+Sca+CD150+ cells, which are highly enriched for HSC, would be expected to be polyploid, and thus give rise to multiple different cell types. However, 25 percent of the time, these cells differentiated exclusively into megakaryocytes. This doesn’t fit with our current understanding of hematopoietic lineage steps and suggests that megakaryocytes may differentiate from HSCs without going through multiple intermediate stages.
Gene expression analyses were consistent with the lineage commitments observed in vitro. However, one surprising observation was that common myeloid and common lymphoid progenitors had many genes in common, which could underlie the lineage plasticity that has been found during initial phases of lineage commitment.
The improved ability to identify and purify specific hematopoietic subpopulations will help investigators to better understand hematopoietic differentiation in general, and myelopoiesis specifically. The identification of successive lineage-restricted progenitors can be used to study the mechanisms of proliferation and maturation down specific lineages. For example, the highly purified biphenotypic PreMegE population can be purified to study lineage fate decisions and to design approaches for enhancing growth and differentiation of erythrocytes and/or megakaryocytes for in vivo enhancement of these lineages or for transfusion therapy.
Return to Top
A Bank Without Assets: Ex Vivo Storage Depletes RBC NO Bioactivity
By Michael Linenberger, MD
Dr. Linenberger indicated no relevant conflicts of interest.
Reynolds JD, et al. S-nitrosohemoglobin deficiency: A mechanism for loss of physiological activity in banked blood. Proc Natl Acad Sci USA. 2007;104:17058-62.
Bennett-Guerrero E, et al. Evolution of adverse changes in stored RBCs. Proc Natl Acad Sci USA. 2007;104:17063-8.
Tissue oxygen (O2) delivery is dependent on two interacting processes: red blood cell (RBC) O2 content and microcirculatory blood flow. Under normoxic conditions, vasomotor tone and tissue perfusion are regulated by endothelium-derived nitric oxide (NO). In hypoxic tissues, RBCs transfer bioactive NO to the vessel wall to cause vasodilation and increase oxygenation. The chemistry of NO biosynthesis and export from RBCs is not yet completely understood; however, the mechanistic roles of hemoglobin (Hb) have recently been clarified.1,2 Important intermediates in the process include S-nitrosothiol (SNO) adducts and S-nitrosohemoglobin (SNO-Hb), the concentration of which directly correlates with the ability of RBCs to transfer NO and induce vasodilation in laboratory models of tissue hypoxia. Reynolds, et al. and Bennett-Guerrero, et al. were prompted by observational studies showing an association between worse clinical outcomes in critically ill medical and surgical patients who had received RBC transfusions or transfusions of RBC units with prolonged storage times.3 They sought to determine whether collection and storage of blood under standard blood bank conditions resulted in changes in the content of RBC NO equivalents and hypoxic vasodilatory function, which, in turn, might explain a loss of efficacy for RBC transfusions in patients at high risk for ischemic tissue injury.
Both of these studies demonstrated rapid depletion of RBC SNO-Hb and membrane SNO concentrations that began as early as three hours after blood draw (prior to leukofiltration and processing) and persisted for six weeks. In concert with SNO-Hb depletion, stored RBCs exhibited significantly impaired vasodilatory activity in hypoxic rabbit aortic ring bioassays. The mechanism for rapid SNO-Hb loss was not defined nor was a direct correlation between SNO-Hb depletion and other biochemical RBC storage lesions identified. Bennett-Guerrero, et al. did observe a slow decline in RBC deformability that became more marked at two to six weeks of storage, possibly relevant to reports of adverse outcomes after transfusion with older banked blood.3 Reynolds, et al. showed that RBCs stored for up to six weeks could be renitrosylated by exposure to aqueous NO and adjustment of pH, and this reconstituted the SNO-Hb concentrations and hypoxic vasodilatory functions in organ chamber and live-animal canine coronary artery models.
These observations identify an important storage lesion in banked blood that profoundly affects the ability of RBCs to function as an O2-responsive vasoregulator. The reason for rapid SNO-Hb loss ex vivo is not defined nor is it known whether SNO-Hb levels and functional activity can be recovered in RBCs after transfusion. For patients with critical illness and tissue hypoxia, transfusion of SNO-Hb depleted RBCs could theoretically displace enough native RBCs to impair normal vasodilatory function and augment ischemic injury, resulting in increased morbidity and mortality. Altered RBC SNO-Hb levels have been implicated in the vasculopathic processes associated with sickle cell disease, diabetes, and pulmonary hypertension. Transfusion practices for these patients, and perhaps others with chronic cardiac, pulmonary, and vascular diseases, should be re-evaluated in light of these findings. The observation that SNO-Hb levels and experimental vasodilatory activity could be reconstituted by renitrosylation raises the intriguing notion that blood banks in the future might be able to provide "recharged" RBCs for anemic critically ill patients, for red cell exchange to treat severe sickle cell crisis, and for other transfusion indications in selected high-risk individuals.
- Angelo M, Singel DJ, Stamler JS. An S-nitrosothiol (SNO) synthase function of hemoglobin that utilizes nitrite as a substrate. Proc Natl Acad Sci USA. 2006;103:8366-71.
- Basu S, Grubina R, Huang J, et al. Catalytic generation of N2O3 by the concerted nitrite reductase and anhydrase activity of hemoglobin. Nat Chem Biol. 2007;3:785-94.
- Tinmouth A, Fergusson D, Yee IC, et al. Clinical consequences of red cell storage in the critically ill. Transfusion. 2006;46:2014-27.
Return to Top
Monoallelic Gene Expression in Health and Disease
By Roberto H. Nussenzveig, PhD, and Josef T. Prchal, MD
Drs. Nussenzveig and Prchal indicated no relevant conflicts of interest.
Alexander MK, Mlynarczyk-Evans S, Royce-Tolland M, et al. Differences between homologous alleles of olfactory receptor genes require the Polycomb Group protein Eed. J Cell Biol. 2007;179:269-76.
Scholl C, Bansal D, Döhner K, et al. The homeobox gene CDX2 is aberrantly expressed in most cases of acute myeloid leukemia and promotes leukemogenesis. J Clin Invest. 2007;117:1037-48.
Although most genes in the genome of diploid organisms are expressed from both alleles, genes in some tissues are transcribed preferentially from a single allele. Sex chromosome gene dosage compensation is the first described and best known example of such monoallelic expression. In order to balance expression of X-linked genes, which are present in two copies in females, the majority of one of the X-chromosome genes are transcriptionally silenced in a process called X-chromosome inactivation; however, it has subsequently been shown that monoallelic expression also occurs in some autosomal genes. Genomic imprinting, another example of monoallelic expression, is characterized by transcription in a parent-of-origin-dependent manner. There is clinical relevance of imprinting since a parenteral location of the same genetic lesion may be associated with disorders of entirely different phenotypes. Yet another mechanism of preferential allele usage is allelic exclusion, which occurs in specialized cell types (e.g., B-lymphocytes express only a single heavy and light chain and each olfactory neuron expresses a single odorant receptor gene from greater than 1,300 genes). In the examples given above, allele-specific transcription is not fully understood but may be attained by one or a combination of mechanisms, including asymmetric DNA methylation, replication timing, chromatin structure, non-coding RNAs, and nuclear localization.
Two recent papers have contributed to the elucidation and significance of the mechanism by which two homologous alleles will be targeted for inactivation in so-called random monoallelic expression. Alexander and colleagues employed mouse embryonic stem cell lines that have yet to choose the allele for inactivation and used fluorescence in situ hybridization probes targeted to specific genes. These investigators could then detect and differentiate between the active and inactive states of specific loci during the S-phase by the presence of a single fluorescent pinpoint in replicated loci that was characteristic of the active state, while a doublet on the homologous chromosome indicated its tendency for inactivation. In their paper, Alexander, et al. noted that alleles subjected to monoallelic expression on both autosomes and X-chromosomes often flipped between the active and inactive states in the embryonic stem cells. The inability to assay the state of activity of monoallelic loci in methanol fixed cells suggested a requirement for intact chromatin structure. Based on this, the authors investigated whether embryonic stem cells carrying mutations in either Eed or Dnmt1, proteins that are known to play a role in chromatin modification, would affect the allelic inactivation, and their results showed a pivotal role for Eed gene in the establishment of the active/inactive state.
A second paper by Scholl and colleagues unveiled a pivotal role for the clustered homeobox (HOX) genes in normal hematopoiesis and observed their abnormal expression in the majority of patients with acute myeloid leukemia (AML). A normal human genome carries 39 HOX transcription factors clustered on four chromosomes (HOXA - 7p15, HOXB - 17q21, HOXC - 12q12, and HOXD - 2q31); expression of the genes within each cluster is developmentally, temporally, and spatially correlated to their location within each loci. Gene expression from these loci is under complex transcriptional and epigenetic regulation. One family of transcription factors, the Caudal-like homeobox genes (CDX1, CDX2, and CDX4), plays a critical role in regulating expression of HOX genes. Normal expression of CDX2 is restricted to intestinal development in the adult, while aberrant expression has been linked to gastrointestinal malignancies. The authors of this report first showed that the ectopic expression of CDX2 in a mouse was sufficient to induce malignant leukemic transformation. In their study of human AML, they showed that 153 of 170 patients (90 percent) and eight of 15 human myeloid leukemia cell lines (53 percent) expressed CDX2 regardless of their cytogenetic status. These investigators then studied the mechanism of aberrant CDX2 expression and discovered that six of seven patients expressed CDX2 from a single allele. However, the molecular basis of this monoallelic expression of CDX2 remains unclear as no mutations in the promoter, its exons, or hypomethylation of CpG island of CDX2 exon-1 were found.
These two papers contribute to our understanding of the cellular mechanism of mono- or bi-allelic gene expression and also show that deregulation of these processes can have profound organismal consequences. The discovery that CDX2 is aberrantly expressed from a single allele in the majority of AML patients underscores the importance of disturbances in monoallelic and bi-allelic gene expression and, in addition, its novelty as a mechanism of disease pathophysiology, about which much more remains to be learned.
Return to Top
Autologous Stem Cell Transplantation for Follicular Lymphoma: Evolution or Extinction?
By Michael Williams, MD
Dr. Williams indicated no relevant conflicts of interest.
Montoto S, Canals C, Rohatiner AZ, et al. Long-term follow-up of high-dose treatment with autologous haematopoietic progenitor cell support in 693 patients with follicular lymphoma: an EBMT registry study. Leukemia. 2007; 21:2324-31.
Follicular lymphoma (FL) is an indolent non-Hodgkin lymphoma responsive to a variety of chemoimmunotherapeutic approaches. However, the lack of cure with these treatments has led to a number of clinical trials over the past 20 years investigating the role of high-dose therapy (HDT) and autologous stem cell transplantation (ASCT) in first or subsequent remission, with no consensus benefit for overall survival (OS) and with concerns regarding early and late toxicities. Montoto and colleagues report a retrospective analysis and long-term follow-up of 693 patients with FL from the European Blood and Marrow Transplantation Registry who received HDT-ASCT in first remission or after relapse and subsequent treatment. As a registry-based analysis, induction treatments and the type of HDT conditioning regimens varied. Fifty-eight percent received total-body irradiation (TBI) as part of HDT, 81 percent of patients were treated after 1990, and all patients in the analysis had at least five years of follow-up; 30 percent of patients had received only a single line of treatment prior to HDT-ASCT, and 19 percent were in first complete remission (CR1). Half of the patients relapsed following HDT-ASCT at a median of 1.5 years, with lower relapse rates among those patients who received TBI conditioning regimens (53 percent vs. 43 percent; p=.05). Progression-free survival (PFS) and OS were improved for patients transplanted in CR1 (Figures 1a and 1b). Ninety-three patients (13 percent) died within one year of HDT from a cause other than lymphoma, most due to infection, and 40 (6 percent) died of a non-lymphoma cause more than one year post-HDT, most due to second malignancy. The latter was significantly more common in TBI-treated patients and contributed to an inferior long-term survival for TBI versus non-TBI conditioning regimens despite the lower relapse rates.
These results support previous findings that HDT-ASCT may result in durable remission and possible cure for a subset of patients with FL. Patients who received a transplant in CR1, and those who had received fewer lines of pre-transplant chemotherapy or were younger than 45 at transplant, had improved outcomes. The analysis also verifies excess late mortality due to second malignancies including MDS/AML, primarily in those treated with TBI; as such, TBI should no longer be utilized in conditioning regimens and has been largely abandoned. Although this is the largest series to date evaluating HDT-ASCT in FL and having the benefit of very long follow-up, caveats include the retrospective and non-randomized nature of this study, the lack of central pathology review, and the varying induction and conditioning regimens utilized. Previous prospective randomized clinical trials of HDT-ASCT for FL in CR1 failed to show a benefit in OS and also showed excess late, second malignancies.1 However, these prospective trials and the current retrospective analysis only included patients treated in the pre-chemoimmunotherapy era. Further prospective clinical trials will be necessary to assess whether HDT-ASCT will evolve to become a reliably curative approach via induction chemoimmunotherapy and "in vivo purging" with monoclonal antibody therapy prior to stem cell collection, or will become extinct if comparably long PFS and OS is achievable with more effective and less toxic induction therapies, consolidative immunotherapy with monoclonal antibodies or idiotype vaccines, or the availability of more effective regimens for patients who relapse.
1. Bowles K, Hodson D, Marcus R. Follicular lymphoma. Lymphoma: Pathology, Diagnosis and Treatment, R Marcus et al., Eds. pp 119-122. Cambridge Univ. Press, 2007.
Return to Top
Return to Table of Contents
|
