| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
CONCISE REVIEW |
Department of Adult Oncology, Dana-Farber Cancer Institute and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts, USA
Key Words. Peripheral blood stem cell • CD34+ • Allogeneic • Transplantation • Hematologic malignancies • Apheresis
Joseph H. Antin, M.D., Dana-Farber Cancer Institute, 44 Binney St., Boston, Massachusetts 02115, USA. Telephone: 617-632-2525; Fax: 617-632-5175; e-mail: Joseph_Antin{at}dfci.harvard.edu
 |
ABSTRACT |
|---|
 Top Abstract IntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
Although the precise identification of PBSCs remains elusive, they can be phenotypically identified as a subset of all circulating CD34+ cells. There are important phenotypic and biologic distinctions between PBSCs and bone marrow (BM)-derived progenitor cells. PBSCs express more lineage-specific antigens but are less metabolically active than their BM-derived counterparts.
The use of PBSCs for allogeneic transplantation has been compared to BM in several randomized trials and cohort studies. The use of PBSCs in leukemia, myeloma, non-Hodgkin's lymphoma, and myelodysplasia has resulted in shorter times to neutrophil and platelet engraftment at the expense of increased rates of chronic graft-versus-host disease. The increase in graft-versus-host disease is mainly due to a log-fold increase in donor T cells transferred with the graft. Relapse rates after transplantation may be lower after PBSC transplantation but a convincing survival advantage has not been demonstrated overall. It is possible that a stronger graft-versus-tumor effect may exist with PBSCs when compared with BM although the mechanisms leading to this effect are not clear.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
PBSC transplantation (PBSCT) has become increasingly common in the autologous setting, with PBSC largely replacing bone marrow (BM) as the preferred stem cell source due largely to quicker engraftment kinetics and ease of collection. The use of PBSC in allogeneic transplantation has increased greatly as well, albeit more cautiously.
In this review, we will discuss the identification and immunophenotypic characteristics of PBSCs, describe methods for collection of these cells and discuss outcome issues such as engraftment kinetics and graft-versus-host disease (GVHD). The clinical experience of PBSCT in various hematological malignancies with both traditional and nonmyeloablative approaches will be discussed in detail.
|
IDENTIFICATION OF PBSCS |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
PBSCs represent a subpopulation of all CD34+ cells found in the circulation. Although morphologically difficult to identify, these cells can be distinguished to some degree by their immunophenotypic patterns. PBSCs are CD34+/CD38– and do not express a full complement of either myeloid or lymphoid lineage-specific markers (Lin–) but do express the Thy-1 differentiation antigen. These CD34+/CD38–/Lin–/Thy-1+ cells are the cells responsible for initiating long-term culture initiating colony (LTC-IC) assays [2].
PBSCs that are mobilized by colony-stimulating factors (i.e., recombinant human [rHuG-CSF]) are neither phenotypically nor immunologically identical to BM-derived stem cells. In comparison with BM-derived stem cells, mobilized PBSCs have been found to express more lineage-specific differentiation antigens (i.e., CD13, CD33), have a lower proportion of cells in S phase (i.e., are less active in cellular cycling) and are less metabolically active (in rhodamine retention assays and by demonstrating less CD71 positivity) [3]. Furthermore, PBSCs demonstrate higher clonogenicity in LTC assays [3].
Currently, the most reproducible method of stem cell quantification after collection is by flow cytometric evaluation of CD34+ cell numbers. Enumeration of CD34+/CD38–, CD34+/CD33–, and CD34+/Thy-1+ cell subsets may prove to be a more useful technique of estimation of stem cell numbers [4]. The percentage of colony-forming units-granulocyte-macrophage from a stem cell harvest has also been used to estimate stem cell numbers. This method is much less reliable and can vary widely due to differences in culture technique and media as well as several human factors. Furthermore, not all transplantation centers are equipped to perform these labor-intensive cultures, which require 10 to 14 days of incubation. The standardized method for enumerating CD34+ cell counts has been published in greater detail elsewhere [2, 5].
|
MOBILIZATION AND COLLECTION OF PBSCS |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
CSF doses of between 2 and 24 µg/kg administered s.c. daily have been given to healthy donors [6-8], including donors over the age of 60 years [9]. The induced leucocytosis, when maintained at levels below 70,000 cells/µl, has not been shown to be detrimental to the donor's health for short periods of time, however, important morbidity, including splenic rupture and death, have rarely been reported [10, 11]. Common minor side effects caused by the administration of growth factors include bone pain, myalgias, headache, and fever, which respond to mild analgesics in over 80% of cases. Longer follow-up (up to six years) has confirmed the safety of administration of rHuG-CSF to healthy donors [11, 12].
PBSCs are obtained by apheresis, generally via peripheral venous access; occasionally central venous access may be required. Up to two to three donor blood volumes are processed per session by extracorporeal continuous-flow apheresis machines. Stem cells and granulocytes are separated from the red blood cell and plasma fractions of blood by centrifugation and the latter two components are returned to the donors during the apheresis procedure itself. The process generally requires between three and five hours per session. The target range of peripheral stem cells of 2 x 106 cells/kg (recipient weight) can generally be achieved in one to two sessions.
The efficacy of stem cell collection comparing BM and PBSCs has been examined. In a randomized trial, a single PBSC apheresis procedure yielded 3.7 times more CD34+ cells than a standard BM harvest. Furthermore, BM harvests were 6.8 times more likely to be insufficient for transplantation (defined as less than 2 x 106 cells/kg of recipient weight, p < 0.001) [13].
Potential risks associated with stem cell apheresis procedure include complications related to central line placement when necessary (such as pneumothorax) and cytopenias due to the apheresis procedure itself. Leukopenia or lymphopenia frequently occur and can last a variable period of time. Thrombocytopenia (<100 x 109 cells/l) is not uncommon, however, bleeding complications are extraordinarily rare. Similarly, anemia that requires transfusion is very uncommon. Anderlini et al. provide a complete review of donor safety issues [14].
PBSC donation is generally preferred by donors over traditional BM donation, which entails either general or spinal anesthesia, a brief hospital stay, and more post-procedure discomfort. Several phase II studies measuring outcomes (including quality of life and anxiety scores) comparing the two procurement procedures confirmed this fact [15, 16].
|
ENGRAFTMENT AND KINETICS |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
A convincing reduction in time to engraftment after both autologous and allogeneic PBSCT when compared to traditional BMT has been demonstrated. In the allogeneic setting, neutrophil engraftment (to 0.5 x 109 cells/l on two consecutive days) occurred between one and six days earlier with PBSCT when compared with BMT in randomized trials (mean time to engraftment: 14 versus 15 days and 15 versus 21 days) [17, 18]. Unsupported platelet counts of 20 x 109/l occurred between four and seven days earlier (mean time to platelet engraftment: 15 versus 19 days and 11 versus 18 days) [17, 19]. The results of a large database review are consistent with the results of the randomized trials (mean time to neutrophil engraftment 14 versus 19 days, mean time to platelet engraftment 18 versus 25 days, p < 0.001 for both comparisons) [20].
Typical doses of CD34+ stem cells used for PBSCs are 2 x 106 cells/kg of recipient body weight or greater. Doses lower than this threshold are associated with prolonged cytopenias and increased early mortality [21].
A relationship between the dose of CD34+ stem cells delivered with the transplant and the tempo of hematologic recovery has been demonstrated for both BMT [21] as well as PBSCT [22]. The use of higher doses of CD34+ cells may lead to quicker engraftment, particularly when doses are greatly increased [23, 24]. Platelet recovery appears to be more sensitive to CD34+ doses than neutrophil recovery [24].
Efforts to enrich PBSCT by ex-vivo CD34+ cell selection (positive selection) have resulted in increased rates of GVHD, possibly by altering the cytokine expression patterns of transplanted cells or changing lymphocyte subsets delivered with the graft [25]. These efforts do not appear to alter engraftment kinetics significantly [25, 26]. Negative selection (T cell depletion) clearly leads to lower rates of GVHD, at the expense of a slightly higher incidence of graft rejection.
Other strategies used to shorten the time to engraftment include the use of combined PBSCT and BMT [27] and the use of rHuG-CSF-mobilized BM for transplantation [28-30]. The latter approach may have the advantage of GVHD rates similar to traditional BMT [29] and neutrophil engraftment kinetics similar to those of PBSCT [30].
The earlier engraftment seen after PBSCT has lead to earlier discharge from hospital [17, 19, 31] and total lower immediate costs associated with the transplant procedure [18, 32]. The reduction in costs associated with the procedure is primarily due to fewer dollars spent on hospital room charges, blood products and other supportive measures. The costs of stem cell mobilization and collection procedures, however, are greater for PBSCT than for traditional BMT, primarily due to the use of recombinant human hematopoietic growth factors [32]. Long-term cost issues are more difficult to predict and will be influenced by GVHD outcomes after PBSCT (see below).
|
GVHD |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
Acute GVHD occurs within the first weeks following transplantation. Chronic GVHD occurs later, and is arbitrarily defined as the presence or persistence of GVHD beyond 100 days since transplantation. The occurrence and severity of GVHD is related to the degree of HLA-mismatching between donor and recipient, the type of conditioning regimen used, the use of degree of immunosuppression of the recipient (GVHD prophylaxis), the viral exposure of the donor and recipient, the underlying malignant disease, and to the passage of mature, immunocompetent T cells with the stem cell transplant [34]. PBSCT contain roughly a log-fold increase in the concentration of CD3+ T cells when compared to BMT. This large increase in T cell dose is largely responsible for the increased risk of GVHD seen after PBSCT. PBSCs were avoided for years because of concerns over the higher risk of GVHD, however, these concerns appear to be less severe than originally anticipated.
The incidence and severity of GVHD after PBSCT may be influenced by rHuG-CSF administered to donors prior to stem cell collection. Polarization of T cells to the Th2 class has been demonstrated to occur in response to rHuG-CSF in mouse models [35, 36]. This polarization is related to the subtype of antigen-presenting cell (dendritic cell) initiating the immune response. rHu-G-CSF has been shown to increase the circulating numbers of type 2 dendritic cells, which in turn influence the predominance of a Th2 cellular response and elucidated cytokines [37, 38]. Th2 cells are considered GVHD neutral in comparison to Th1 cells.
There have been five randomized controlled trials [17-19, 39, 40] and many cohort studies [20, 31, 41-50] that have examined the relative risks (RR) of acute and chronic GVHD after PBSCT when compared to traditional BMT (Table 1
). No single trial has concluded that the incidence of acute GVHD incidence is higher after PBSCT when compared to traditional BMT. Only one of the trials, however, was statistically powered to detect small differences in the incidence of acute GVHD [40]. RR for acute GVHD after PBSCT varied from 0.67 to 1.5 compared to BMT [43, 44]. In a meta-analysis of 15 trials, a small but statistically significant increase in the RR of acute GVHD after PBSCT was demonstrated (RR 1.13, 95% confidence interval [CI] 1.01-1.26) [51].
|
The occurrence of acute GVHD has been linked to higher treatment-related morbidity and early mortality, however, chronic GVHD has been shown to be protective against relapse [51, 52] and may be associated with better long-term survival, particularly when used in high-risk patients [20, 40]. It remains to be determined whether the reduction in hospital stay during the acute phase is cost-effective compared with the increased costs and morbidity of chronic GVHD.
|
USE OF PBSCS FOR SPECIFIC HEMATOLOGICAL MALIGNANCIES |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
Acute and Chronic Leukemia
BMT for acute myelogenous leukemia (AML) is recommended for patients with refractory/resistant leukemia, patients in second clinical remission, and patients with high-risk cytogenetic features in first remission. The Medical Research Council AML10 trial revealed significant differences in long-term disease-free survival when autologous transplantation was compared with consolidative chemotherapy in individuals without an HLA-matched sibling [55]. At least one trial has demonstrated higher long-term survival after allogeneic transplantation compared with autologous transplantation [58], but others have demonstrated equivalence [57, 59]. Allogeneic transplantation has become the standard of care for patients with high-risk cytogenetic features in first remission, patients with resistant or refractory disease and patients in second remission, provided the patient is a candidate for the procedure and a suitable donor is available. Acute lymphoblastic leukemia is treated largely the same way in the adult population.
In stable phase CML, allogeneic transplantation is currently the curative therapy of choice when feasible. Autologous transplantation with Philadelphia chromosome-negative-purged marrow or stem cells has recently re-emerged as an investigational treatment option [60], but there is no established benefit of autologous transplantation over interferon-based therapy.
PBSCT has been used most extensively in acute and chronic leukemias. In a prospective, randomized trial comparing allogeneic BMT and PBSCT in 111 patients, Blaise et al. did not demonstrate significantly different overall and leukemia-free survival between PBSCT and BMT groups [18]. The trial did demonstrate the expected significantly decreased time to neutrophil and platelet engraftment after PBSCT. The rate of chronic GVHD was noted to be significantly higher after PBSCT when compared with BMT (30% versus 55%, p < 0.03).
In a study reported by Schmitz et al., the authors reported the outcomes of 66 patients with acute or chronic leukemia who received BMT or PBSCT from HLA-matched siblings [17]. The majority of patients entered in this trial had low-risk disease, referring to AML in first remission or CML in chronic, stable phase. The mean time to platelet recovery (>20 x 109/l) was reduced by four days after PBSCT when compared to BMT. No differences in rates of GVHD or overall survival were reported in this trial.
Multiple Myeloma
Autologous BMT for multiple myeloma has been shown to be superior to traditional chemotherapy, offering both a disease-free and overall survival advantage [61]. However, autologous transplants are frequently contaminated by residual tumor cells when assessed by sensitive molecular techniques [53, 62]. The risk of contamination as well as the demonstration of a pronounced graft-versus-myeloma effect [63, 64] has prompted the study of allogeneic stem cell sources as an alternative to autologous cells. This approach is not without risk; the Cancer and Leukemia Group B recently has had to stop accrual to a trial of non-T cell-depleted allogeneic transplantation in myeloma due to concerns over toxicity.
Majolino et al. described 10 patients (including four who had already received PBSC autografts) undergoing allogeneic PBSCT for multiple myeloma [65]. The median time to engraftment was 13 days. Eight of 10 patients had a complete remission (CR) while the remaining two patients had a partial remission (PR). None of the patients achieving CR had evidence of recurrence of myeloma 7 to 28 months after transplantation. Two patients died of treatment-related causes.
Corradini et al. described the molecular remission status of 51 patients after autografting or allogeneic transplantation with either BM or PBSCs [66]. All 17 patients who received allografts entered a CR state after transplantation. In patients with molecular markers available, there was a significantly increased proportion of patients who achieved a molecular remission after allogeneic transplantation compared to autologous transplantation. Furthermore, only one of five allogeneic BMT patients became polymerase chain reaction-negative for clonal immunoglobulin gene rearrangements compared with six of nine patients who received allogeneic PBSCT. Similarly, Cavo et al. described five patients who had both complete clinical and molecular remissions after PBSCT from HLA-matched siblings [67].
Myelodysplasia
Allogeneic transplantation is the only therapy for myelodysplastic syndromes (MDS) with curative potential available today. Patients with fewer cytogenetic abnormalities [68], less latency since the time of diagnosis [69], and patients with refractory anemia or refractory anemia with ringed sideroblasts [70, 71] have better outcomes after transplantation. The International Prognostic Scoring System score is also useful for predicting outcome after transplantation [72]. The optimal timing of stem cell transplantation for this disease is not yet known.
The experience in transplantation with PBSC for MDS is more limited than that for the acute and chronic leukemias. Patients with MDS have comprised only a small proportion of patients in trials comparing PBSCT and BMT and therefore no conclusions can be drawn on the relative advantages and disadvantages of PBSCT.
Non-Hodgkin's Lymphoma (NHL)
Autologous transplantation for relapsed intermediate and high-grade NHL has clear survival advantages over chemotherapy [73]. Allogeneic transplantation has been used for patients with relapse after BMT or refractory disease. In a preliminary report by Körbling et al., four patients with refractory NHL underwent allogeneic PBSCT from HLA-matched siblings [74]. Three patients had a CR and the fourth, a PR. One patient died of infectious complications 82 days after transplantation and the other three patients were reported to be alive greater than 31 to 100 days post-transplant.
Khouri et al. studied the effects of allogeneic transplantation in patients with mantle cell lymphoma, an intermediate-grade lymphoma with a uniformly poor prognosis [75]. Sixteen patients were studied, 11 of whom received allogeneic PBSCs as a stem cell source. Nine patients remained alive at the time of publication, with eight patients in CR. Molecular minimal residual disease was assessed in seven patients. Five of the seven patients had no evidence of molecular residual disease between 3 and 30 months after transplantation. Five of seven patients had evidence of chronic GVHD. Other case reports have documented the efficacy of allogeneic PBSCT for NHL [76, 77].
A graft-versus-lymphoma effect has been noted in both animal models [78] and human studies [75-77]. In a murine model of NHL, Ito and Shizuru demonstrated a unique graft-versus-lymphoma activity which was separable from GVHD effects and mediated by CD8+ T cells and perforin-dependent cytolysis [78].
BM Aplasia and Donor Lymphocyte Infusion (DLI)
The use of unstimulated PB buffy coat cells for aplastic anemia was first reported by Storb et al. in 1982 [79]. At this time, the addition of buffy coat cells to BMT increased the risk of GVHD, likely as a result of the immense T cell load in the buffy coat compared with the relative sparse concentration of progenitor cells. Graft rejection was diminished and survival for patients treated with both marrow and buffy coat cells was improved compared with marrow alone; however, it was not clear whether this improvement could be attributed to the stem cells infused or prevention of graft rejection by the infusion of large numbers of allogeneic T cells [80].
More recently, infusions of buffy coat cells, referred to as DLI have been used as post-transplantation immunotherapy for malignant disease. Given at variable times after transplantation of allogeneic BM or PBSC, DLI can induce potent graft-versus-tumor effects and is useful as treatment for relapsed malignancies or pre-emptive therapy for prevention of relapse. The procedure is associated with a 50% risk in acute GVHD and a 20% risk of marrow aplasia [81]. The effects of DLI are most prominent in CML, where single infusions of donor lymphocytes have induced long-lasting remissions in relapsed patients [82].
Despite the infusion of hematopoietic progenitor cells with the lymphocyte population, even rHuGM-CSF-mobilized donor lymphocytes are not completely protective against marrow aplasia when used in relapsed CML [81]. This implies that aplasia seen after DLI may not be solely due to destruction of native hematopoietic elements, but also involves suppression of donor engraftment and hematopoiesis.
Nonmyeloablative Transplantation and PBSCs
The DLI experience led to the notion that a transplantation could rely primarily on the immune system of the donor to eradicate the leukemia and that sublethal conditioning might allow stable engraftment. Clearly, depending on whether the targeted disease was a stem cell disorder (e.g., CML) or a disease in which stem cells are not involved (e.g., NHL), it may or may not be necessary to establish complete chimerism of hematopoiesis. The first trial of nonmyeloablative or reduced-intensity conditioning followed by allogeneic PBSCT was reported by Khouri et al. in 1998 [83]. This technique employs the graft-versus-tumor effect as the primary therapeutic modality and omits high-dose conditioning regimens. Instead, low-dose immunosuppressive regimens containing drugs such as the nucleoside analogue fludarabine, generally in combination with cyclophosphamide or busulfan, are used to permit donor progenitor cell engraftment even without host myeloablation. The graft-versus-host/graft-versus-leukemia response might result in eradication of host hematopoiesis, therefore it is necessary to include a source of stem cells to sustain hematopoiesis.
Potential advantages of a nonmyeloablative approach to allogeneic transplantation include the inclusion of an older patient population and patients with other comorbid conditions that would be otherwise excluded from allogeneic transplantation. As well, the procedure itself is not limited by chemotherapy-induced toxicity and can often be performed in an outpatient setting. Rates of GVHD after nonmyeloablative transplantation are similar to those seen after myeloablative transplantation.
In the first published phase II study [83], 15 patients with lymphoid malignancies were transplanted following nonmyeloablative conditioning. Eleven of 15 patients demonstrated evidence of donor-derived hematopoiesis at one month post-transplant. Mixed chimerism (defined as the coexistence of recipient and donor-derived hematopoiesis) was noted in most subjects. DLIs in the post-transplant setting were able to convert some patients from mixed chimeric hematopoiesis to full donor-derived hematopoiesis.
A nonmyeloablative approach to PBSCT for NHL was explored by Nagler et al. [84]. Nineteen patients with heavily-pretreated NHL were given a nonmyeloablative conditioning regimen consisting of fludarabine, busulfan, and antithymocyte globulin. Eight of the 19 patients were alive between 15 and 37 months post-transplant. There were four patients who relapsed after treatment. An attempt at post-transplantation immunomodulation with immunosuppressive therapy withdrawal and/or DLI was not successful at inducing remissions in these patients.
|
CONCLUSIONS AND FUTURE DIRECTIONS |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
Although the time to engraftment of PBSCs appears to be shorter than BM-derived stem cells, newer BM mobilization techniques may obviate this difference in the future. The specific biological factors leading to a more rapid engraftment are yet to be identified fully and represent a gap in the basic understanding of stem cell biology.
Although the incidence of GVHD was initially thought to be similar for both BMT and PBSCT, recent evidence from ongoing clinical trials and a meta-analysis has demonstrated that both acute and chronic GVHD occur with greater frequency after PBSCT than BMT. Efforts to control GVHD with novel immunophyllin inhibitors and other immunomodulatory agents may reduce rates of GVHD after both procedures to minimize treatment-related mortality and long-term morbidity. The ability to harness the useful effects of the graft-versus-tumor activity and separate them from GVHD will further increase the beneficial effects of high-dose therapy and stem cell transplantation. The addition of post-transplantation cellular immunotherapy with either donor lymphocyte or dendritic cell preparations may enhance graft-versus-tumor activity.
The role of PBSCT for individual malignancies remains to be determined. At the present, it is reasonable to reserve the use of PBSC to clinically high-risk scenarios, such as second remission AML or CML in blast crisis, where PBSCT has shown a survival advantage over BMT in a database review. For low-risk malignant conditions such as stable phase CML, it is reasonable to continue using BM-derived stem cells for transplantation until clinical trials demonstrate the superiority of one stem cell source over the other.
|
REFERENCES |
|---|
|
TopAbstractIntroductionIdentification of PBSCsMobilization and Collection of...Engraftment and KineticsGVHDUse of PBSCs for...Conclusions and Future...References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
