| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
CONCISE REVIEW |
a Department of Pediatrics, Mattel Children's Hospital at UCLA, Gwynne-Hazen Cherry Memorial Laboratories, and the UCLA Jonsson Comprehensive Cancer Center, Los Angeles, California, USA;
b Department of Cellular and Molecular Pathology, UCLA School of Medicine, Los Angeles, California, USA;
c Division of Biology, California Institute of Technology, Pasadena, California, USA
Key Words. Molecular targeting • Leukemia • Lymphoma • Monoclonal antibodies
Correspondence:
Kathleen M. Sakamoto, M.D., Department of Pediatrics, Mattel Children's Hospital at UCLA, 10833 Le Conte Avenue, Los Angeles, California 90095-1752, USA. Telephone: 310-794-7007; Fax 310-206-8089; e-mail: kms{at}ucla.edu
 |
ABSTRACT |
|---|
 Top Abstract BackgroundMolecular AntibodiesSmall MoleculesConclusionReferences |
|---|
|
BACKGROUND |
|---|
|
TopAbstractBackgroundMolecular AntibodiesSmall MoleculesConclusionReferences |
|---|
The concept of targeted therapy for patients with cancer has intrigued researchers for years. In 1953, Pressman and Korngold showed that antibodies could specifically target tumor cells [2]. Unfortunately, it was not until 1975, when Kohler and Milstein described their Nobel Prize-winning work in hybridoma technology, that a continuous supply of monoclonal antibodies (mAbs) that targeted specific antigens became available [3]. By 1979, Nadler et al. treated the first patient with mAb therapy [4]. Today, more than 3,000 cancer patients have received mAb therapy.
In general, there are three main classes of cytotoxic mAbs that have been developed. The first consists of unconjugated mAbs, where the antibody itself mediates cell killing. The remaining two classes involve mAbs that are conjugated to either a potent drug/toxin, or a radioisotope. Despite the progress that has been made with developing targeted therapies, there are numerous obstacles to successful therapy (Table 1
).
|
This review outlines the progress of STI571 in CML therapy and the therapeutic use of four distinct mAbs: namely, rituximab (anti-CD20) in the treatment of non-Hodgkin's lymphoma (NHL); campath-1H (anti-CD52) in the treatment of chronic lymphocytic leukemia (CLL); Mylotarg® (anti-CD33) and its use in acute myelogenous leukemia (AML); and 131I anti-CD45 and its use in combination with chemotherapy prior to stem cell transplant. We have chosen to discuss these four antibodies given that they are among the best characterized antibodies that have been studied in hematologic malignancies. They also represent each of the main classes of antibodies to cell surface antigens, namely those that are unconjugated, conjugated to a potent toxin, and radioimmunoconjugated.
|
MOLECULAR ANTIBODIES |
|---|
|
TopAbstractBackgroundMolecular AntibodiesSmall MoleculesConclusionReferences |
|---|
The first mAb clinical trial for NHL was reported by Nadler et al. in 1980 and was unsuccessful, secondary to binding of the antibody on soluble antigen in the patient's serum [4]. The goal was to find an antigen on most, if not all, tumor cells and preferably not on critical or nonrenewable host cells, which would lead to dose-limiting toxicity [11]. After much trial and error with numerous B-cell surface antigens, CD20 became the target of mAb therapy for NHL [12, 13]. CD20 is expressed on pre-B cells and mature B cells, including up to 95% of B-cell NHLs. Of importance, it is absent from stem cells and plasma cells. It does not shed, internalize, or have much modulation upon binding by an anti-CD20 antibody [14, 15]. The function of CD20 is still controversial, but it is felt to regulate the cell cycle through calcium channel regulation in the cell membrane [9]. The expression of CD20 is variable, being somewhat low in CLL and higher in follicular lymphomas [11].
Rituximab (IDEC-C2B8) is a chimeric murine/human anti-CD20 mAb that has a human gamma-1-kappa antibody with variable regions isolated from a murine anti-CD20 mAb. This antibody has been approved by the U.S. Food and Drug Administration (FDA) for use in patients with NHL. Murine antibodies induce a host anti-mouse antibody (HAMA) immune response, which may prevent further administration when used [16]. In contrast, chimeric antibodies are cleared slowly from the circulation and can be given repeatedly. Rituximab was designed to interact with the human immune system, yet avoid a host HAMA immune response. In vitro results have shown that rituximab acts by complement-mediated cell lysis, antibody-dependent cell-mediated cytotoxicity (ADCC), and induction of apoptosis [1, 15, 17]. Despite the in vitro evidence, it is still unclear by which mechanism(s) rituximab works clinically.
Rituximab is among the most widely studied and available therapeutic mAbs for the treatment of NHL and has become incorporated into therapeutic clinical use. Initial clinical trials evaluated rituximab in patients with refractory indolent B-cell lymphoma, which is still one of the largest groups being studied for the antibody [18–20]. FDA approval of the antibody was based on five single-agent studies performed on patients with indolent or follicular NHL. The largest of the five was the pivotal phase II trial of 166 previously treated patients [20]. Rituximab was given as weekly doses of 375 mg/m2 for 4 weeks. A 48% overall response (OR) rate was seen, with 6% complete responses (CR) and 42% partial responses. This was consistent with results from previous studies, and the median duration of response at last follow-up was 13.2 months. The patients in this study had either relapsed or refractory indolent or follicular NHL and had received a median of three previous chemotherapy regimens. Patients with follicular lymphoma had much higher response rates (60%) than did patients with small lymphocytic lymphoma (CLL), who only had a 13% response rate. The reason for this difference is not entirely understood, but most patients with small lymphocytic lymphoma CLL express lower density CD20 antigen on their malignant lymphocytes, which could decrease rituximab efficacy [21]. Based on this, it seems reasonable to decrease the number of circulating tumor cells in CLL by using chemotherapeutic agents before therapy with rituximab [22]. Several strategies are being studied in this group of patients to optimize the efficacy of rituxamib, for example, greater frequency of dosing, higher doses, and use as first-line therapy in CLL [23–25]. In summary, rituximab has shown activity against a range of B-cell malignancies, including large cell lymphoma, mantle cell lymphoma, and posttransplant lymphoproliferative disorder [26–28].
Retreatment with rituximab in patients who show disease progression and who were initially treated with a 4-week course has shown some benefit. In a group of 39 patients with refractory lymphoma who responded initially to rituximab, there was a 40% response rate to a repeat 4-week course [29]. The median duration of second response was 16 months, which is slightly longer than the first response. There has been evidence of some lymphomas becoming resistant to rituximab, yet most lymphomas remain CD20 positive [30]. Therefore, retreatment with rituximab until the time of resistance seems a reasonable strategy in patients who continue to respond. An alternative strategy is to use an extended course of rituximab therapy, since the 40% response rate with retreatment suggests that patients retain sensitivity to rituximab. In a group of 37 patients with refractory low-grade lymphoma who received rituximab for 8 consecutive weeks, there was a 60% OR rate and a median response duration greater than 13 months [31]. These results are significant, especially for patients with large tumor bulk or large numbers of circulating lymphocytes, which may predispose them to lower rituximab levels after a standard 4-week course due to antibody saturation.
One can conclude from the five single-agent studies that the majority of adverse events with rituximab were infusion-related and primarily associated with the first infusion [20]. The most common events were fever, chills, myalgias, nausea, angioedema, and pruritus, and were probably due to cytokine release. Unlike standard chemotherapy, blood count suppression was not a major problem, nor was the incidence of infection. The recovery of B lymphocytes appears to occur within 6 to 9 months after therapy is discontinued [32]. The most serious toxicity with rituximab was a syndrome of rapid tumor lysis, which was reported most frequently in patients with high circulating malignant cell counts [33]. Symptoms occur during the first rituximab infusion and include severe infusion-related reactions, such as rigors, fever, bronchospasm, and hypoxemia in addition to thrombocytopenia, decreasing lymphocyte counts, and electrolyte abnormalities. The syndrome is reversible with appropriate supportive care, and some patients can receive rituximab at a lower infusion rate the following dose. These effects can be life threatening, and therefore, caution should be taken when treating patients with high counts, as in CLL. Based on the initial results, patients with CLL and more than 50 x 109/l circulating cells should not be treated with rituximab [22]. Given the relatively safe toxicity profile, combination therapy with rituximab and cytotoxic drugs was explored.
After considering combination therapy, the next question became that of scheduling considerations for combining mAbs with chemotherapy. Essentially, there are two possibilities: simultaneous administration, with the goal of achieving a more potent cytotoxic effect, or sequential administration, in which the goal is for antibody to attack any minimal residual disease following chemotherapy. Given that rituximab has not had any overlapping toxicity with chemotherapy, it could be given before, during, or after chemotherapy. One must consider the interactions of the two therapies due to their differing mechanisms of action. There are data to support simultaneous administration, given that antibody binding to CD20 can enhance the cytotoxic effects of chemotherapy. There is also the possibility that any agent that affects the immune response (e. g., corticosteroids/chemotherapy) could interfere with the ADCC, which would favor sequential therapy administration [11]. Other important considerations are the amount of resistance, and the potential for cross-resistance. There are not enough data to support whether or not resistance to rituximab can cause resistance to chemotherapy or vice versa. Most studies have used patients who were previously treated with chemotherapy prior to trying rituximab. Of note, in the pivotal rituximab trial, response rates were much lower (38% versus 57%) in patients who received three or more chemotherapy regimens than in those who received only one course prior to rituximab [20]. Based on the above uncertainties, several different schedules must be investigated prior to a final administration decision being made.
The first published phase II trial, by Czuczman et al. in 1999, combined CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) with rituximab in 38 patients with indolent or follicular NHL [34]. The majority of patients had not been treated previously. Patients received six doses of rituximab and six cycles of CHOP (Table 2). The rationale behind this regimen was to saturate CD20 binding sites with two doses of rituximab initially, and then to use the final two rituximab infusions to target minimal residual disease. Overall, there was no evidence of additive toxicity, and very little dose modification was needed. The OR rate was 100%, with 58% CRs and 42% partial responses. Twenty-five of 38 patients had not progressed after a median observation time of 40 months of follow-up. In addition, this study was the first to show that rituximab/CHOP therapy can induce a molecular remission, defined as clearance of polymerase chain reaction (PCR)-detectable cells carrying the bcl-2/IgH t(14;18) translocation from the blood and bone marrow. Of the initial eight patients who were PCR-positive at baseline, seven converted to PCR-negative after combined therapy. The significance of this is uncertain, however, in single-agent rituximab therapy, the clearance of PCR-detectable cells was associated with a significant increase in response duration compared with patients who remained PCR-positive after therapy [11]. The long remission duration and observation of molecular CR (rare with CHOP alone) are both very encouraging for the increased benefit of this combined therapy.
|
Although the majority of clinical trials have studied CHOP chemotherapy, the ideal cytotoxic therapy with rituximab has yet to be established. Additional studies will be needed to evaluate the best regimen, schedule, and dose in different clinical settings. For example, other groups are studying the effect of rituximab therapy after completion of CHOP with the hypothesis that rituximab may prolong the duration of response by eliminating minimal residual disease. Based on the ability of combined therapy with rituximab to achieve a molecular remission, the use of rituximab with high-dose chemotherapy and autologous peripheral blood stem cell transplant is also under investigation. Preliminary results have shown that rituximab in this setting can cause a decrease in graft contamination with ex vivo purging and an increase in molecular remissions following transplant [36]. However, follow-up is limited and larger trials will definitely be required.
Preliminary information is available regarding the results of first-line therapy with rituximab in patients with low-grade NHL. A study group of 41 patients with stage II-IV untreated indolent lymphoma were treated with a 4-week course of rituximab, and those with stable disease or an objective response received another 4-week course at 6-month intervals [37]. The results after the first course of rituximab showed that the response rate was 54%, with an additional 36% of patients achieving stable disease. With further treatment, the OR rate increased to 72%, with 18% CR. At 1 year, the progression-free survival was 77%. These results were similar to those of another study, by Solal-Celigny et al., in a group of 50 patients with follicular NHL [38]. There was a 71% objective response rate, and 17 of 30 patients (57%) showed clearing of their initial bcl-2 rearrangement detected by PCR. Further studies are definitely necessary to evaluate rituximab as a first-line therapy versus standard chemotherapy. Preliminary data show that the rate and duration of response to first-line rituximab therapy appear superior to those seen with standard chemotherapy, with a much shorter and less toxic therapeutic course.
In summary, rituximab is becoming more and more established as one of the most active single agents in the treatment of indolent NHL. It lacks dose-limiting toxicity and has not demonstrated overlapping toxicity with chemotherapy. It has shown a beneficial effect when used either simultaneously or sequentially with chemotherapy. At this point, some investigators feel that rituximab should be offered to all patients with indolent lymphoma at some point during their therapy. It is still unclear whether rituximab acts synergistically with chemotherapy or is best administered following chemotherapy to promote a more intact immune system. One question is whether use of the drug as second- or third-line therapy may be preferable to use as salvage therapy for patients that have proven to be refractory to chemotherapy. In addition, some patients have benefited from repeat therapy with rituximab, which could result in control of disease for long periods with limited toxicity. First-line therapy with rituximab might also be tolerated better in patients who are elderly or who are symptomatic, given its low toxicity profile. Patients who are asymptomatic but desire therapy rather than observation alone might also benefit from first-line rituximab. Large clinical trials are still required for further follow-up and to determine the ideal combination therapy.
Anti-CD52 (Campath-1H)
The campath-1 family of mAbs recognizes the CD52 antigen. This antigen is present on at least 95% of all normal human B and T lymphocytes, monocytes, and macrophages, as well as in most B-cell and T-cell lymphomas [39]. Current use of anti-CD52 is approved by the FDA in patients with CLL. The functions of CD52 are not yet known, but the antigen is not expressed on stem cells. Early studies in the 1980s using rat monoclonal IgM or IgG showed response in a range of B-cell malignancies [40]. The humanized version, alemtuzumab (campath-1H) has shown efficacy in CLL, although with significant toxicity [41]. Also, given its ability to deplete T cells, campath-1H has been used to prevent graft-versus-host disease (GVHD) following transplantation in acute leukemia [42].
Campath-1H, which is a human IgG anti-CD52 mAb, works by binding to the cell membrane CD52 antigen on lymphocytes. The antigen has not been detected on erythrocytes, platelets, or stem cells, therefore normal hematopoiesis should not be impaired with campath-1H. Campath-1H works through two mechanisms: causing cell lysis by using complement fixation in the host and ADCC [22]. Initial clinical trials with campath-1H showed limited activity in patients with previously treated indolent lymphoma. In a multicenter study with 50 patients, only 14% achieved a partial remission [43]. Phase II trials with campath-1H in refractory CLL have shown much better results. In two large multicenter phase II trials with a total of 76 patients with refractory or relapsed CLL, 30 mg of Campath-1H were given i.v. three times per week for a maximum of 12 weeks. The OR rate was 33%, with 5% complete remission and 28% partial remission. Another 36% of patients had stable disease [44]. Of note, blood and marrow involvement responded better than bulky lymph node disease. Rapid elimination of CLL cells in the peripheral blood was seen in 97% of patients, while resolution of lymphadenopathy was seen in only 7%. The median time to progression was projected to be >9 months. Acute infusion-related toxicity occurred in 77% of patients, primarily within the first week of therapy. The symptoms included: fever, chills, nausea, and rigors. Similar infusion-related side effects were seen in patients with CLL after treatment with the anti-CD20 antibody rituximab (Table 3). This was most likely associated with the rapid lysis of peripheral tumor cells and cytokine release. Of great significance, in the previously mentioned phase II trials with campath-1H, 25% of patients experienced grade 3 or 4 infections. In a study of previously untreated CLL patients, infections during therapy with campath-1H seemed to occur less often [45]. A small percentage of patients had grade IV neutropenia or thrombocytopenia of short duration.
|
Experience with campath-1H underscores the importance of issues such as cross-reactivity with normal tissues and tumor bulk in the efficacy and safety of mAb therapy. It also proves that toxicity can often be used as a therapeutic benefit. For example, researchers have used the lymphocyte-depleting ability of campath-1H for prevention of GVHD after transplantation.
Anti-CD33 (HuM195 and Mylotarg®)
AML is the most common type of acute leukemia in adults, with an annual incidence of 2.7 per 100,000 adults in the U.S. A very important statistic is that the incidence increases to 14.1 per 100,000 adults over the age of 65 [47]. Approximately 60%-80% of newly diagnosed AML patients achieve CR, unfortunately, the median duration of this remission is 15 months. A much smaller percentage of patients with relapsed or refractory AML will achieve a second CR, and the duration of this response is often only 4-8 months [48, 49]. Another difficulty in treating older patients with AML is that the chemotherapy is often not tolerated. For example, due to the cumulative cardiotoxicity of anthracyclines, they are usually contraindicated in patients with a prior history of congestive heart failure [50, 51]. This obviously greatly affects those patients over 65 years of age. Allogeneic or autologous hematopoietic stem cell transplantation is also often not offered to elderly patients due to the excessive transplant-related mortality involved in this age group. Given the above difficulties and the fact that less than 20% of patients who achieve a CR are disease free for at least 2 years, new therapies have been explored for both older patients and those in relapse [52].
The goal is to find a therapy for AML that is capable of achieving remissions yet has limited side effects. The idea of mAb therapy is attractive, as is the idea of a drug-antibody conjugate, i.e., a cytotoxic agent that is conjugated to an mAb that binds to a specific cell antigen and delivers targeted therapy. The CD33 antigen is a good target, as it is expressed on the surface of leukemic blasts in greater than 80% of patients with AML. The antigen is also expressed on myeloid precursor cells, resulting in universal neutropenia and thrombocytopenia associated with anti-CD33 antibody; however, there is also hematologic recovery, since CD33 is not expressed on pluripotent stem cells [53–55].
Unmodified antibodies are limited in their activity against acute leukemia for several reasons. Binding of an mAb to the CD33 antigen does not seem to transform an intracellular signal, and therefore, the use of nonconjugated anti-CD33 would not cause targeted cell death. The most likely reason for this relies on the fact that unconjugated anti-CD33 antibody results in saturation of CD33 sites and rapid internalization of the antigen-antibody complex [56]. Another limitation is that murine antibodies are not well recognized by human antigen-presenting cells, which makes the murine antibodies weak in ADCC [57]. Despite this, there were promising in vitro data that lead to HuM195 use in AML. Humanized mouse IgG1, called HuM195, is a nonconjugated humanized IgG1 antibody derived from the murine IgG2 M195, which attacks the CD33 antigen [58]. In a phase II study, 35 patients with refractory or relapsed myeloid leukemia were treated with HuM195 4 days a week for 4 weeks. There were decreases in the blast count, yet only two CRs were seen. The two patients who achieved a CR also were among the patients who had less than 30% blasts in their marrow before receiving the antibody [59]. The conclusion that came from this study was that unlabeled antibodies show clinical efficacy in patients with low tumor burdens, and therefore, might be more effective in settings of minimal residual disease. This led investigators to use HuM195 during consolidation or maintenance therapy for acute promyelocytic leukemia (APL) [60]. APL cells are strongly positive for CD33. In a study of 27 patients with APL in CR, who were treated with all-trans retinoic acid (ATRA) and/or chemotherapy already, HuM195 was given two times a week for 3 weeks, followed by consolidation chemotherapy. This was followed with 6 months of maintenance therapy with HuM195 given twice a week every 4 weeks. Bone marrow was analyzed after each phase for minimal residual disease using the reverse transcriptase-polymerase chain reaction (RT-PCR) to amplify the promyelocytic leukemia/retinoic acid receptor (PML/RAR) mRNA seen in APL. Of the patients treated with ATRA and/or chemotherapy in induction, 2 of 27 became PCR-negative. After the 3-week course of HuM195, 11 of 22 evaluable patients (50%) were PCR-negative. Also, 25 of the 27 patients who received HuM195 remained in clinical CR with a median duration of 29 months. This showed that humanized unconjugated mAbs have some efficacy, but whether there will be an impact on disease-free survival is yet to be seen.
Sievers et al., in collaboration with Wyeth-Ayerst Pharmaceuticals (Philadelphia, PA), have studied CMA-676, a conjugate of the antitumor antibiotic, calicheamicin, and a humanized anti-CD33 antibody [61, 62]. CMA-676, also known as gemtuzumab ozogamicin or Mylotarg®, was given FDA approval in May of 2000 for the treatment of CD33-positive AML in first relapse in patients who are not candidates for other cytotoxic chemotherapy or who are older than 60 years of age (Table 4). Gemtuzumab and CD33 antigen binding cause the formation of a complex that is internalized, which then, upon release of the calicheamicin, results in DNA binding, DNA double-strand breaks, and cell death via apoptosis [63]. An initial phase I study had 40 patients with relapsed or refractory AML (some of whom had already been treated with stem cell transplant) who were then treated with escalating doses of gemtuzumab [61]. Leukemia was eliminated from the blood in 8 of the 40 patients (20%) and blood counts normalized in 8% of patients. The most common side effect in the phase I trial was grade 1 or 2 fever and chills, seen in 80% of patients. Some patients who received the higher doses of antibody had reversible elevations of their liver transaminases and hyperbilirubinemia. The FDA approval of gemtuzumab was based on multicenter phase II trials involving 142 patients with CD33-positive AML in first untreated relapse (after a remission that lasted at least 6 months). The median age of patients was 61 years. Gemtuzumab was given at a dose of 9 mg/m2 i.v. every 2 weeks for two doses [64]. The OR rate was 30%, which is similar to the range of rates seen with various cytotoxic regimens in relapsed AML. Of the 30% with an OR, 16% had a complete remission and 13% achieved remission with incomplete platelet recovery (less than 100,000/mm3). The complete remission was characterized by no blasts in the peripheral blood, less than 5% blasts in the bone marrow, neutrophils >1,500/mm3, and independence from platelet transfusion for 1 week. The median duration of response was 170 days (range 21 to 267). There was also only a small difference in the OR rate for patients <60 years old (34% OR) and >60 years old (26% OR). There were also fewer hospital days in patients treated with gemtuzumab (24 days) versus the control group of patients treated with conventional chemotherapy (31-38 days) [65]. The patients who achieved remission were then offered autologous or allogeneic hematopoietic stem cell transplant, consolidation chemotherapy, or no further therapy. Approximately 40% of these patients remained in remission for at least 6 months. Sievers et al. also reported the outcomes of 27 patients who received gemtuzumab and then underwent stem cell transplantation [66]. Of the 10 patients who underwent allogeneic transplant during remission, nine were still alive 100 days after transplant. Of the five who underwent autologous stem cell transplant, all were alive 100 days after transplant. The other 12 patients were not in remission after gemtuzumab, but three of nine were alive 1 year after allogeneic transplant, and all three of the patients who underwent autologous transplant were alive 1 year later. Of importance, there was no evidence of cardiotoxicity with this agent. The principal adverse effect is myelosuppression, not surprisingly, given that normal myeloid and megakaryocytic precursors were eliminated. Grade 3-4 neutropenia and thrombocytopenia were noted in almost all patients in the postinfusion period. The most common other adverse effects were: fever, chills, gastrointestinal (GI) symptoms, dyspnea, epistaxis, and anorexia. National Cancer Institute Grade 3 and 4 mucositis, sepsis, and pneumonia were 4%, 16%, and 7%, respectively [64]. Of note, in contrast to other studies that have found HAMA immune responses after using immunoconjugates containing murine-derived mAbs, there were no patients in the phase II trials with gemtuzumab that developed an immune response [67]. This is not surprising since gemtuzumab is a humanized mAb.
|
In summary, antibody-targeted therapy provides an important new therapeutic option in patients with relapsed AML. Not only do these antibodies function to block signals leading to cell proliferation, but they may also function as ligands for pro-apoptotic signals. Gemtuzumab had comparable outcomes with cytotoxic regimens, yet a much more favorable toxicity profile. Despite significant progress with mAb therapy for AML, there are still unanswered questions, including its use in combination with chemotherapy and the optimal dose. In the phase II trials, there were still approximately two-thirds of patients who failed to achieve complete remission. At the current dose, there was consistent saturation of the anti-CD33 antibody. This suggests that other drug resistance mechanisms may explain treatment failure [56].
Anti-CD45 with 131I
For many years, bone marrow transplantation has offered the best chance of cure for patients with advanced acute leukemia. However, the conditioning regimens of high doses of total body irradiation (TBI) and/or chemotherapy to induce a remission often cause life-threatening toxic effects. Even despite high-dose therapy, many patients will relapse after transplant, and most studies have looked at increasing the TBI dose or using more chemotherapy to combat this. Unfortunately, this often results in higher transplant-related mortality, which is one of the reasons that transplant is often not offered to older patients.
In two randomized studies from the Fred Hutchinson Cancer Research Center, results showed radiation sensitivity of the myeloid leukemias. The first study had 71 patients with AML in first remission who received cyclophosphamide and either 12-Gy or 15.75-Gy TBI prior to their matched transplant [69]. The results showed significantly lower relapse rates (35% versus 12%) in the group that received the higher TBI dose. However, there was a higher rate of transplant-related mortality with the higher TBI dose, and therefore, there was no significant difference in disease-free survival. A second trial, including patients with CML in chronic phase, showed a lower relapse rate with the higher TBI dose (25% versus 0%) [70]. Because these studies demonstrated the radiation dose-response effect in myeloid leukemias, it was proposed that delivering targeted radiation may lower relapse rates without increasing mortality because normal tissues are spared.
Radioimmunoconjugates seem to be an effective delivery system given their binding specificity. The first clinical trials using radiolabeled antibodies in leukemia studied anti-CD33 antibodies [71, 72]. These studies had varying success and had limitations with dose escalation and unfavorable biodistribution. Matthews et al. began to focus on the CD45 antigen because it is the most broadly expressed of the hematopoietic antigens. CD45 is a cell surface antigen that is expressed on essentially all leukocytes, including myeloid and lymphoid precursors in the bone marrow, and lymphocytes in the lymph nodes [73]. It is found in more than 90% of AML samples and most acute lymphoblastic leukemia (ALL) patient samples, which makes it an ideal target for a radioimmunoconjugate [74]. In contrast to CD33, which is an excellent target for drug-conjugated therapy given that it internalizes upon antibody binding, the CD45 antigen does not internalize after antibody binding [75]. Therefore, antibody-delivered 131I is less likely to be cleaved and released into the circulation. The expression of CD45 allows for the delivery of anti-CD45 antibody to patients in remission or active relapse.
Preclinical studies with this radioimmunoconjugate by Matthews' group in mice and nonhuman primates showed specific delivery of radiation to different hematopoietic tissues. There was two to four times more radiation to marrow, three to twelve times more radiation to spleen, and two to eight times more radiation to lymph nodes than to the liver, lung, or kidney [76, 77].
The preclinical data led to a phase I study of 44 patients with high-risk acute leukemias or myelodysplastic syndrome (MDS) who received cyclophosphamide, TBI, and 131I-BC8 (murine anti-CD45) antibody prior to transplantation [78]. The results showed that 84% of patients had good localization of antibody; there was at least twice as much radiation delivered to the marrow and spleen than to the other vital organs. Unacceptable Grade III regimen-related toxicity (severe mucositis) was seen in two patients treated at the 12.25-Gy dose. One of six patients treated at the 10.5-Gy dose to the liver experienced a Grade III veno-occlusive episode, from which the patient recovered. Therefore, the study concluded that a maximum dose of 10.5 Gy to the liver delivered by radiolabeled antibody could be tolerated in addition to the 12-Gy TBI and cyclophosphamide. Even though the study was not designed to measure efficacy, 7 of 25 patients with AML or MDS remained disease-free 26-100 months (median 58 months) posttransplant.
A similar study by Matthews et al. incorporated 131I anti-CD45 within a preparative regimen of busulfan and cyclophosphamide for 25 patients with AML in first or second remission [79]. Of the 25 patients, 90% had favorable biodistribution. Of the 24 patients with AML in first remission, 18 patients were disease-free 10 to 63 months (median 42 months) after allogeneic transplant. Unfortunately four patients died from transplant-related mortality, but only two patients (10%) relapsed. For comparison, AML patients with busulfan and cyclophosphamide conditioning, who then undergo transplant in first remission, have a 30% expected relapse rate. Given the encouraging relapse rate in patients with AML transplanted in first remission, there appears to be a significant clinical effect with this therapeutic approach. Whether it will further impact outcome will require additional studies, especially a phase III trial using the standard combined conditioning (i.e., busulfan and cyclophosphamide) versus the standard plus 131I-anti-CD45. Many questions remain in determining the optimal use of mAbs. For example, what biological characteristics of the malignant cells are associated with a favorable response to antibody treatment? What is the percentage of leukemia cells that would need to express the particular antigen for an mAb to be successful? Which combination of drugs and antibodies should be used?
A potentially interesting, but difficult, dilemma in treating AML is the possibility that the myeloid leukemia stem cells do not express the same cell surface antigens as the cells which are used to characterize the leukemia immunophenotype. In this case, unconjugated or drug-antibody conjugates would be unable to target the leukemic clone. Recent work by John Dick suggests that the leukemia clone is maintained by a rare subpopulation of leukemic stem cells with increased proliferative and self-renewal capacity. Furthermore, the AML clone appears to be organized as a hierarchy that is maintained by stem cells that undergo differentiation [80]. Thus, targeting one cell surface antigen following differentiation using antibodies may not affect the leukemia stem cell clone.
|
SMALL MOLECULES |
|---|
|
TopAbstractBackgroundMolecular AntibodiesSmall MoleculesConclusionReferences |
|---|
The cytogenetic hallmark of the majority of CML cases and a minority of ALL cases is the Ph chromosome. The Ph chromosome is caused by a specific chromosomal translocation between chromosomes 9 and 22, which results in the fusion of the c-abl oncogene from chromosome 9 with the breakpoint cluster region (bcr) of chromosome 22. This fusion results in the bcr-abl gene, which encodes different fusion proteins depending on the site of the breakpoint in bcr [82]. The two most common fusion proteins are p210 (210 kDa) and p185 (185 kDa). The p210 protein is seen in 95% of patients with CML and up to 20% of adult patients with ALL, whereas p185 is seen in 10% of adults with ALL and in the majority of pediatric patients with Ph+ ALL (only 5% of all pediatric ALL cases).
The bcr-abl fusion proteins have constitutive tyrosine kinase activity, which causes activation of intracellular signaling pathways causing alterations in the proliferation, adhesion, and survival of CML cells [83]. This activity of the bcr-abl fusion proteins is essential for transformation of the malignant CML cells. Transduction of bcr-abl into murine hematopoietic stem cells followed by transplantation of these stem cells into syngeneic mice produces a CML-like syndrome, which implicates bcr-abl as the causative molecular abnormality in CML and Ph+ ALL [84]. Therefore, the tyrosine kinase activity has become an ideal target for inhibition as a potential therapy for CML and Ph+ ALL. In 1988, Yaish et al. reported the first synthetic tyrosine kinase inhibitors, the tyrophostins [85]. After further drug development and screening of a large library of compounds, signal transduction inhibitor, STI571, (formerly CGP57 148B) or GleevecTM was developed initially as a specific platelet-derived growth factor receptor inhibitor, but was also found to be a selective inhibitor of abl tyrosine kinases, including bcr-abl [86]. STI571 functions by blocking ATP binding to the bcr-abl tyrosine kinase, which inhibits its action. Without tyrosine kinase activity, phosphorylation of substrates does not occur, which causes disruption of downstream cellular events. Upon further study, it was found that STI571 specifically inhibits proliferation of myeloid cells containing the p210 bcr-abl, and clonal growth of myeloid cells from CML patients [87, 88]. The above findings were also found in p185 bcr-abl cell lines and patient samples from Ph+ ALL cases. The only other tyrosine kinase inhibited by STI571 is c-kit, which is associated with GI stromal tumors.
Currently, there are multiple options for primary treatment of patients with CML diagnosed in chronic phase. These include allogeneic stem cell transplantation, hydroxyurea, busulfan, and interferon-alpha (IFN)-based regimens [89]. Hydroxyurea is used initially as cytoreductive therapy and is effective in controlling blood counts. Unfortunately, cytogenetic responses are rare with hydroxyurea, and the onset to blast crisis is not delayed. Allogeneic stem cell transplantation has the potential to cure selected patients with CML. Unfortunately, there is a significant risk for illness or death as a direct consequence of transplant, which precludes many older patients from undergoing this therapy. Physicians are currently studying less intense pretransplant conditioning or nonmyeloablative stem cell transplants, also referred to as "minitransplants," as a means of exploiting the "graft versus leukemia" effect of a transplant [90]. Whether the minitransplant approach will impact how many older individuals are considered for transplant remains to be seen. Irrespective of the conditioning regimen, donor availability still remains a problem.
For most patients, IFN prolongs life, with 5-year survival rates of 57% compared with 42% with hydroxyurea and busulfan [91]. Hematological responses are seen in up to 80% of patients, with cytogenetic responses in 30%-50%. Unfortunately IFN has significant toxicity and is poorly tolerated in up to 20% of patients. Studies have shown an increase in cytogenetic responses when IFN is added to cytarabine compared with IFN alone [92]. These results have led to this combination becoming the standard for nontransplant candidates; unfortunately there is increased GI toxicity. There has also been much concern that prior treatment with IFN might adversely affect the results of a subsequent transplant due to increased graft rejection or GVHD. A multicenter study in Germany implied that if IFN was stopped at least 90 days prior to transplant, there were no adverse effects on the transplant [93].
Given that the standard treatment options involve less than optimal responses and significant toxicity, clinical trials of STI571 were begun. In June 1998, the initial phase I study of STI571 included 83 Ph+ CML patients in chronic phase with less than 15% blasts in their blood or bone marrow and who were either resistant or refractory to IFN [5, 94]. The threshold for an effective dose was 300 mg/day. Fifty-four patients were treated at or above the dose, and almost all (98%) patients achieved a complete hematologic response. Major cytogenetic responses (31%) and complete cytogenetic responses (13%) were observed. Ongoing responses were seen in 96% of patients with follow-up of at least 8 months. There was no dose-limiting toxicity, but 8% had grade 2 and 3 myelosuppression, which might be a therapeutic effect since most of hematopoiesis is contributed by the Ph+ clone.
Based on the effectiveness of STI571 in patients who did not respond to IFN in chronic phase, the phase I studies were expanded to include CML patients in myeloid and lymphoid blast crisis and patients with relapsed or refractory Ph+ ALL [95, 96]. Twenty-four of 38 (55%) patients in myeloid blast crisis responded with decreases in the amount of blasts in their marrow to less than 15%; and 8 of 38 (21%) had less than 5% blasts in their marrow. Seven of 38 (18%) of the myeloid blast crisis patients continued to remain in remission on STI571 with follow-up ranging from 100-365 days. Unfortunately, the patients with lymphoid phenotype disease do not have as much success with maintaining remission. Initially 14 of 20 (70%) with either CML in lymphoid blast crisis or Ph+ ALL responded, and 11 of 20 had marrows cleared to less than 5% blasts. All but one of the lymphoid phenotype patients had relapsed between days 42 and 123.
Given the success of the phase I trials, phase II trials were initiated in 1999 and focused on STI571 being administered in all phases of CML. In the study with 532 chronic-phase patients who did not respond to IFN, 47% achieved major and 28% achieved complete cytogenetic responses [97]. Among the 233 patients in accelerated phase, 63% achieved a complete hematologic response but not full recovery of peripheral blood counts [98]. Of note, only 44% of patients had a complete hematologic response with blood count recovery. Cytogenetic responses were seen in 41%, with complete cytogenetic responses in 14%. Of patients in the accelerated phase, 65% were free of progression at 1 year. Lastly, there were 260 patients in the myeloid blast crisis phase who were evaluated [99]. The OR rate was 64% with STI571, with 11% achieving less than 5% blasts in their marrow. In addition, 38% of patients were either returned to chronic phase or had partial responses. Major cytogenetic responses were seen in 15%, with 6% achieving complete cytogenetic responses. The most significant result was that of median survival of 6.8 months; 8.6 months if STI571 was first-line therapy and 4.4 months when STI571 was used as second-line therapy. In comparison, for patients in myeloid blast crisis treated with chemotherapy, median survival is approximately 3 months. Toxicity has remained low in STI571 trials, with most common side effects including nausea, diarrhea, and myalgias; less common side effects include myelosuppression, primarily in the accelerated and blast crisis trials.
Currently, a phase III trial randomizing newly diagnosed patients to either IFN with cytosine arabinoside (Ara-C) versus STI571 is ongoing. Also, the question of whether patients treated with STI571 in combination with other agents will do better than patients treated with STI571 alone remains to be addressed. In vitro studies have provided evidence of an additive or synergistic effect between STI571 and other agents, including IFN and Ara-C [100].
STI571 is still in the early stages of clinical development, yet it is clear that it works most effectively when used early in the disease course. The results also emphasize the importance of the tyrosine kinase activity of bcr-abl in the pathogenesis of CML. It is most likely that early in the chronic phase, bcr-abl is the only molecular abnormality. Therefore, STI571 could be sufficient therapy in early chronic phase patients. In late chronic phase, where there may be more molecular abnormalities, STI571 may not be sufficient therapy as a single agent [86]. This may be an area for combination therapy with STI571 and other agents to maximize the therapeutic benefits. The development of resistance to STI571 could be a problem, especially in advanced accelerated phase or blast crisis. The cause of this could be multifactorial, including mechanisms such as: bcr-abl amplification, multidrug resistance, drug efflux, or clonal evolution [101–103].
The future of STI571 will involve many more clinical trials, but it should be restated that allogeneic stem cell transplantation is still the only treatment known to cure patients with CML. In a recent article by Goldman and Druker, contrasting approaches were given for the treatment of patients with newly diagnosed CML [104]. One approach recommended giving STI571, IFN, or a combination of both to every newly diagnosed patient with CML. Those patients who failed to respond to therapy, or who had a matched donor, would then receive an allogeneic stem cell transplant (allo-SCT). The problems with this approach involve the difficulty in defining who is failed by STI571 treatment and the risk of delaying a transplant. The second approach outlined separating patients initially, after diagnosis, by whether or not they were good candidates for allo-SCT. The patients deemed to be at low risk for transplant-related mortality would proceed to transplant. The patients with a higher risk of treatment-related mortality would be offered a trial of STI571 or combination therapy, and then be assessed for response after 6 or 12 months. The patients who had a less than desirable response would then be offered a transplant. The last patient group in whom transplant would not be offered would receive primary treatment with STI571 or a combination therapy initially after diagnosis.
There is currently much debate of when to consider transplantation in a patient who is responding to STI571. The age of the recipient, suitability of the donor (matched versus unmatched, sibling versus unrelated), physical condition of the recipient, and molecular status of the disease are all important variables to be considered. As more effective nontransplant treatment options, such as STI571, become available, the idea might become less desirable because of transplant-related mortality. It is currently unknown whether initial treatment with STI571 will compromise outcomes by delaying transplant.
In summary, STI571 has been proven to be effective in treating patients with CML, and this has supported the idea of targeted therapy through inhibition of tyrosine kinases. This drug has received FDA approval for use in patients with CML who are in chronic phase in addition to certain classes of GI tumors. It will continue to be an important therapeutic tool in the treatment of CML in the future. Furthermore, STI571 is proof of principle that directed therapy is possible against the action of proteins resulting from translocations in leukemias and lymphomas. In fact, previous work has demonstrated that retinoids, i.e., ATRA, promote differentiation of leukemia cells in patients with APL [105]. ATRA is currently approved for use in patients with APL. Interestingly, the action of ATRA is specific for leukemia cells that contain the translocation t(15;17) or other fusions that include the RAR
gene. In contrast, arsenic trioxide appears to be specific for the PML portion of the PML/RAR fusion protein [106]. Studies are under way to understand the precise mechanisms of action of ATRA and arsenic trioxide in leukemia cell differentiation.
|
CONCLUSION |
|---|
|
TopAbstractBackgroundMolecular AntibodiesSmall MoleculesConclusionReferences |
|---|
). A great deal has been learned from studies using mAbs in leukemia and lymphoma patients. Although significant research is still needed, targeted therapy promises to provide new options for the treatment of patients with hematologic malignancies (Table 6).
|
|
|
ACKNOWLEDGMENT |
|---|
|
TopAbstractBackgroundMolecular AntibodiesSmall MoleculesConclusionReferences |
|---|
This work was supported by the National Institutes of Health CA068821, American Cancer Society RPG-99-081-01-LBC, and California Cancer Research Program 99-00557V-10021. K.M.S. is also a Scholar of Leukemia and Lymphoma Society of America and recipient of awards from CapCURE and UC BioSTAR/Celgene -Signal Pharmaceuticals 00-10107.
|
REFERENCES |
|---|
|
TopAbstractBackgroundMolecular AntibodiesSmall MoleculesConclusionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  U. Sahin, M. Koslowski, K. Dhaene, D. Usener, G. Brandenburg, G. Seitz, C. Huber, and O. Tureci Claudin-18 Splice Variant 2 Is a Pan-Cancer Target Suitable for Therapeutic Antibody Development Clin. Cancer Res., December 1, 2008; 14(23): 7624 - 7634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Picchio, E. Scala, D. Pomponi, E. Caprini, M. Frontani, I. Angelucci, A. Mangoni, C. Lazzeri, M. Perez, D. Remotti, et al. CXCL13 Is Highly Produced by Sezary Cells and Enhances Their Migratory Ability via a Synergistic Mechanism Involving CCL19 and CCL21 Chemokines Cancer Res., September 1, 2008; 68(17): 7137 - 7146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Koslowski, U. Sahin, K. Dhaene, C. Huber, and O. Tureci MS4A12 Is a Colon-Selective Store-Operated Calcium Channel Promoting Malignant Cell Processes Cancer Res., May 1, 2008; 68(9): 3458 - 3466. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Alfonso-Perez, S. Lopez-Giral, N. E. Quintana, J. Loscertales, P. Martin-Jimenez, and C. Munoz Anti-CCR7 monoclonal antibodies as a novel tool for the treatment of chronic lymphocyte leukemia J. Leukoc. Biol., June 1, 2006; 79(6): 1157 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Christiansen, S. A. Rajasekaran, L. Inge, L. Cheng, G. Anilkumar, N. H. Bander, and A. K. Rajasekaran N-glycosylation and microtubule integrity are involved in apical targeting of prostate-specific membrane antigen: implications for immunotherapy Mol. Cancer Ther., May 1, 2005; 4(5): 704 - 714. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Christiansen and A. K. Rajasekaran Biological impediments to monoclonal antibody-based cancer immunotherapy Mol. Cancer Ther., November 1, 2004; 3(11): 1493 - 1501. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
