- Treating Acute Myeloid Leukemia (AML) in Older Adults
- Survival rate in myeloid leukemia treatment
- Relapsed and Refractory Acute Myeloid Leukemia: What Next?
- Monoclonal Antibody Therapy in Acute Myeloid Leukemia
- Can Supportive Care Enhance the Efficacy of AML Treatment Regimens?
Treating Acute Myeloid Leukemia (AML) in Older Adults
Since the median age at presentation for AML is 64 years, most patients fall into the unfavorable prognostic risk group by virtue of age alone. Moreover, older adults tend to have other comorbid conditions that make treatment even more difficult by the very nature of the cytotoxic drugs employed in most therapeutic regimens.
Clearance of such cytotoxic drugs may also be impacted by baseline hepatic or renal function. Previous SWOG data have disclosed that multidrug resistance overexpression is found in more than 70% of de novo AML patients over the age of 55 years and is highly predictive for failure to achieve complete remission.
Also, as discussed previously, there is a greater incidence of AML arising from prior myelodysplasia, which, in turn, is accompanied by less favorable cytogenetic anomalies. All of these factors contribute to the bleak prognosis for AML in older adults.
The large, retrospective ECOG review that examined long-term survival among 2882 patients with newly diagnosed AML, treated with various protocols between 1973 and 1996, included 944 patients over the age of 55 years. Compared with an overall median survival of 11 months, the median survival in older patients was only 6 months, with a 5-year survival rate of 7.6% (compared with 15% for the entire patient cohort).
The European Organization for the Research and Treatment of Cancer (EORTC) recently updated their AML9 trial in which a 9-month median survival and an 8% 5-year survival rate were observed in patients older than 60 years of age.
A 5-year survival rate of less than 5% was reported during the 4th International Workshop on Chromosomes and Leukemia for patients older than 60 years of age.
Survival rate in myeloid leukemia treatment
It is likely that the real survival rate may be even worse than these sample studies suggest, since many older patients are not even entered into large treatment center-based clinical trials due to exclusion criteria, selection bias, and other complicating factors. Perhaps this is why only 30% of those patients included in the ECOG retrospective were 55 years old or older when 55% of all newly diagnosed AML patients are 55 years old or older.
The ideal induction regimen for older patients has not yet been defined. While cytarabine/anthracycline-containing induction regimens produce complete remission rates of approximately 70% in patients under the age of 60 years, those same regimens result in complete remission rates of 45% to 50% in patients older than 60 years of age.
In younger patients, as discussed previously, disease-free as well as overall survival advantages have been demonstrated in trials that have incorporated idarubicin or mitoxantrone with higher doses of cytarabine. On the other hand, these same trials show no advantage within older populations when compared with standard 7 + 3 therapy.
An ECOG trial (E3993), which compared daunorubicin vs mitoxantrone vs idarubicin, showed no clinical advantage with either mitoxantrone or idarubicin.
Few studies have focused specifically on older adults in terms of the best postremission consolidation therapy, but high-dose cytarabine, with or without mitoxantrone, has not been found to confer clear benefit.
In the previously cited CALGB trial, which randomized patients to 3 consolidation arms, only 29% of patients over the age of 60 years were able to receive all 4 courses of high-dose cytarabine compared with a 60% completion rate among younger patients.
This has led some investigators to use low-dose chemotherapy in both induction and consolidation phases (the focus of several small trials, not all of them prospectively randomized), while others have opted to use no therapy at all.
One study illustrates the “no therapy” approach.
In this EORTC trial, in which the median patient age was 71, individuals were randomized to receive standard therapy at time of diagnosis vs no therapy until clinical deterioration was apparent. There were profound differences between treatment arms, with 58% of those treated up front achieving complete remissions, while no patients in the treatment delayed group achieved remissions.
This translated into substantially different survival rates at 2 years (17% vs 0%).
Another multi-institutional European trial randomized elderly patients to receive either intensive induction chemotherapy vs low-dose cytarabine induction therapy.
Study results demonstrated a clear advantage for those patients receiving the more intensive regimen with a complete remission rate of 52% compared with a 32% remission rate in the low-dose cytarabine arm. A higher overall survival rate was noted in the more chemointensive treatment arm (12.8 months vs 8.8 months).
This improvement in response and survival was, however, associated with a 31% induction mortality rate vs 10% in the low-dose cytarabine arm, as well as an increased need for transfusion support.
A large, retrospective analysis, which examined 53 different publications including 751 patients treated with low-dose cytarabine, failed to demonstrate significant improvement in remission or survival rates with a minimum of 15% treatment-related deaths. Hematologic toxicity, when reported, was quite dramatic, affecting 254 of 289 patients (88%). The author concluded that in the absence of more favorable data, low-dose cytarabine was neither as benign nor as efficacious as numerous case reports suggested.
As is the case in younger patients, the number of postremission chemotherapy cycles remains undefined for older patients.
Once again, the question of prolonged maintenance therapy has been raised, particularly since standard or dose-intensive consolidation regimens do not seem to be beyond tolerable for many older patients. A recently published study by Löwenberg and colleagues, for the Leukemia Cooperative Group of the EORTC, demonstrated an improved disease-free survival among elderly patients, who were randomized to receive low-dose cytarabine maintenance therapy (10 mg/m2 subcutaneously every 12 hours for 12 days every 6 weeks) vs observation, after completion of induction therapy with either mitoxantrone/cytarabine or daunorubicin/cytarabine.
While no difference in overall survival was noted, a 5-year disease-free survival advantage of 13% vs 7% (P = .006) was reported in the low-dose cytarabine-treated cohort, compared with those patients who were merely observed.
Clearly, newer treatment strategies for management of AML in the elderly are needed. Given the higher incidence of multidrug resistance gene expression (particularly the gp170 multidrug resistance protein) in this population, it is logical to develop therapies that target this entity. A recent SWOG trial has shown positive results targeting gp170 multidrug resistance with cyclosporine, although problems with immunosuppression tended to complicate therapy.
Work with PSC 833, a nonimmunosuppressive, multidrug resistance inhibitor, has not shown improvement in outcome in several randomized clinical trials.
Nonmyelosuppressive immunotherapeutic approaches to treatment include immunostimulatory agents such as IL-2, monoclonal antibody therapy, and vaccine therapy. IL-2 can instigate a graft vs leukemia effect via activation of T and natural killer cells and has demonstrated both efficacy and feasibility in clinical trials.
A phase 3 CALGB study continues to randomize patients 60 years of age and older, who remain in remission following induction and consolidation therapy for leukemia, to either observation or a 90-day course of subcutaneous low-dose IL-2.
The subject of monoclonal antibody therapies will be discussed in detail below, but conceptually, they offer a less toxic, less myeloablative approach.
Vaccine therapy as acute leukemia treatment
Vaccine therapies, using murine leukemic cells that have been transfected with adenovirus vectors expressing immunologically detectable antigens, have already been developed, and have demonstrated an ability to protect vaccinated subjects against inoculations of wild-type leukemia cells.
Mice with active leukemia have gone into remission following inoculation with these vaccines. Other forms of vaccine manipulation have stimulated AML cells to express dendritic cell antigens or have fused AML cells to autologous dendritic cells in order to make them amenable to intrinsic immune responses.
Supportive treatment, using colony stimulating factors, seems like a logical therapeutic adjunct for this patient population, but it will also be discussed below in more detail.
Relapsed and Refractory Acute Myeloid Leukemia: What Next?
For the majority of non-APL AML patients treated with chemotherapy alone, the issue of disease recurrence remains a major obstacle to overcome. Several factors must be taken into consideration, including patient age, duration of first remission, and cytogenetic findings, but the most important of these is duration of first remission.
Those patients whose remission lasted for 2 years or more will achieve a second remission in 50% to 60% of cases, when the same initial treatment regimen is repeated. If the first remission lasted 12-14 months, then the patient has a 40% chance of attaining a second remission.
For those patients whose remissions were less than 1 year in duration or who failed to achieve a first remission (primary refractory disease), only 10% to 20% attain complete remission. Long-term survival at 3 years in the long remission group is approximately 20% to 25%, while the shorter duration remission groups have virtually no 3-year survivals.
Therefore, treatment decisions must be based, in large part, upon an individual’s potential for obtaining and maintaining a second remission.
Those at greater risk for failure should not be offered standard therapies as a matter of course.
Whether high-dose cytarabine should be used vs investigational therapy is an important consideration. Dr. Estey and colleagues suggested that high-dose cytarabine was better able to induce second remissions following both short-, intermediate-, and long-term first remission relapses when compared with a variety of investigational therapies.
This assumption was based on the fact that achievement of second complete remissions translates into survival improvements, an assumption based on studies that link complete remission to survival in newly diagnosed patients with AML.
This is, in fact, not supported by the survival data among these individuals attaining second complete remission. In the less than 12 month first remission group, high-dose cytarabine was no better than investigation strategies at extending survival, even though higher initial rates of complete remission were achieved. These results were attributed to both short second remission durations as well as higher mortality rates associated with dose-intensive chemotherapy. Therefore, the data do not support the use of conventional high-dose cytarabine regimens in such cases.
On the other hand, the data do support the use of high-dose cytarabine in patients whose first remissions lasted for more than 1 year.
Transplants in Acute Myeloid Leukemia treatment
Ideally, patients with short duration first remissions, barring other complicating factors, should be considered as candidates for alloBMT and ABMT, which can achieve approximately 30% long-term survival rates.
Both Petersen and colleagues and Buckner and colleagues have shown that alloBMT and ABMT can be offered at time of relapse without benefit of prior chemotherapy with similar survival, whether chemotherapy preceded transplantation.
For those patients younger than 55 years of age with primary refractory disease, alloBMT seems to be superior to ABMT, as demonstrated in data from the City of Hope Cancer Center and from the International Bone Marrow Transplant Registry.
For patients who relapse at 1 year and less than 2 years after first remission, following standard or high-dose cytarabine chemotherapy, alloBMT is still the preferred therapy. Comparing alloBMT with chemotherapy in first relapse, the International Bone Marrow Transplant Registry identified a leukemia-free survival advantage (41% vs 17%) among transplanted patients who were younger than 30 years old with at least 1 year or more of initial complete remission.
A second group of patients, older than 30 years of age with less than 1 year of initial complete remission, also achieved a higher 3-year leukemia-free survival compared with chemotherapy (18% vs 7%). Similar results were noted in patients who underwent ABMT.
In order to compare treatment outcomes from transplant vs chemotherapy regimens, multiple prognostic factors would have to be taken into account, as well as the influence of selection bias. To date, this has not been done, lending further credence to the argument that these patients might benefit even more from investigational approaches.
Patients with APL who relapse following treatment with ATRA and an anthracycline-based chemotherapy regimen may be retreated with ATRA, as long as resistance has not been demonstrated. For those patients who fail to respond, arsenic trioxide was approved by the US Food and Drug Administration (FDA) in September 2000. The original Chinese investigators recently reported data on patients treated with arsenic trioxide between 2005 and 2012.
A total of 104 patients received arsenic trioxide, 67 who relapsed after ATRA treatment, 10 who relapsed following standard chemotherapy, 17 who relapsed after prior arsenic treatment, and 10 patients who relapsed post-ABMT/alloBMT treatment. Half of the patients received only arsenic trioxide monotherapy, while the other half were given both arsenic and serial chemotherapy.
Complete remission for the arsenic trioxide-only arm was 55% and the median survival of complete responders was 4.8 years. A total of 22 patients (43%) eventually relapsed. For the arsenic plus chemotherapy group, the relapse rate was lower (31%), but the median complete remission duration was longer (7 years). Many other studies regarding the use of arsenic trioxide have been conducted in China, but a recent US multicenter pilot study confirms their findings.
Soignet and colleagues reviewed the initial American pilot study results in which 12 heavily pretreated patients received arsenic until best response. Marrow blasts were eliminated in 11 of 12 patients after a median of 33 days. Eight of 11 patients became t(15;17) negative by reverse transcriptase-polymerase chain reaction (RT-PCR) assay. Median duration of remission was 5 or more months.
Thereafter, a multicenter confirmatory study was conducted that reported a complete remission rate of 89% among 40 patients with relapsed APL.
Given the high complete response rate seen in first-line APL with the use of ATRA, it is doubtful that arsenic trioxide will supplant ATRA as initial therapy. Still, as resistance to retinoic acid administration occurs, arsenic trioxide remains a highly successful salvage alternative, one that is well-tolerated with acceptable toxicity and is potentially useful among older patients.
Monoclonal Antibody Therapy in Acute Myeloid Leukemia
One of the more innovative approaches to treatment of malignancy within the last decade has been the development of monoclonal antibody therapy. By targeting features unique to malignant cells, these treatments conceptually allow for eradication of malignant clones while sparing normal tissue. Since 1997, 4 monoclonal antibodies have been introduced into the clinical arsenal against malignancy, and a host of others are either pending for FDA approval or are in advanced phases of clinical testing.
Trastuzumab was the first monoclonal antibody approved for clinical use in the United States for the treatment of HER-2/neu receptor positive metastatic breast cancer. This was followed shortly by the approval of rituximab, an unconjugated monoclonal antibody that targets the CD20 receptor expressed by many B-cell non-Hodgkin’s lymphomas as well as other hematologic malignancies.
Denileukin diftitox, directed against the CD25 antigen on the IL-2 receptor expressed, among others, by cutaneous T-cell lymphoma cells, also received FDA approval, followed recently by approval of a conjugated anti-CD33 antibody for treatment of relapsed AML in older patients.
Early in hematopoiesis, it is believed that a pluripotent stem cell gives rise to committed precursor cells that are responsible for production of granulocytes, monocytes, erythrocytes, and platelets. Both stem cells and precursor cells express the CD34 antigen. In contrast, the CD33 antigen is expressed in committed myeloid precursor cells but not in hematopoietic stem cells.
The CD33 antigen is also expressed by leukemic blasts in at least 90% of patients with AML.
The notion of employing “naked” unconjugated monoclonal antibody as therapy against AML did not make good scientific sense, since unconjugated antibody therapy relies in part upon the patient’s relatively intact immune system to mediate antibody-dependent cellular cytotoxicity. AML is considered to be, potentially, a disease deriving from dysregulation of the immune response.
Unlike solid tumors, which present penetration problems in the presence of bulky disease, AML seems to be an ideal target for delivery of monoclonal antibody therapy because leukemia cells are generally well vascularized, allowing antibody access to virtually all neoplastic cells.
One role for unlabeled CD33 antibody might be in minimal residual disease states, where it is assumed that the immune system is relatively intact. Researchers at the Memorial Sloan-Kettering Cancer Center have used unlabeled HuM195 (anti-CD33 antibody) in patients with APL before consolidation chemotherapy and found that 44% of patients were RT-PCR negative, compared with 34 similarly treated patients who had achieved clinical remission with ATRA alone in previous studies.
Among these patients, 21% were found to be RT-PCR negative before consolidation therapy. Of note, unconjugated antibody may also be effective as a purging agent for ABMT or PBSCT.
Several investigators have conjugated anti-CD33 antibody to a number of different radioisotopes, including iodine-131, yttrium-90, and bismuth-213, as sole treatment or in combination with marrow transplant regimens for AML. While none of these conjugated radioisotopes produced durable responses, significant marrow penetration was achieved with both yttrium and bismuth compounds.
Perhaps the lack of efficacy can be explained in part by the speed in which the antibody-antigen complex moves from the cell surface into the cellular interior. This process of rapid internalization might weaken the effect of radioisotope activity on surrounding cells. Alternatively, if a toxin were conjugated to the antibody core, its rapid internalization might lead to quick and irreversible cellular death. And, thus, the immunoconjugate CMA-676 was generated.
CMA-676 is composed of an anti-CD33 antibody, complexed to calicheamicin, an antitumor antibiotic that generates double-stranded DNA breaks, resulting in cellular death.
Phase 1 testing in 40 patients with relapsed or refractory AML established the phase 2 dose at 9 mg/m2 every other week for 2 doses.
A reduction of marrow blasts to less than 5% was observed in 20% of the patients and toxicity was mostly infusion-related, with 80% of patients experiencing fevers and chills. Prolonged neutropenia occurred more often at the 9-mg/m2 dose, while transient liver transaminase elevation was noted in 20% of those treated. There was no dose-limiting toxicity. An interim analysis of the first 23 patients in the ongoing phase 2 trial for untreated first-relapse AML showed elimination of peripheral and marrow blasts in 10 patients with considerably less toxicity than traditionally seen with standard chemotherapy regimens, requiring significantly fewer days of hospitalization.
Combining the results of several trials that studied the use of CMA-676, a total of 142 patients were treated, and complete remissions were observed in 30% of them. On the basis of this data, CMA-676 was recently FDA approved for relapsed/refractory AML in patients older than 60 years of age.
The CD45 antigen is expressed on all leukocytes and leukocyte precursors and is found in more than 90% of AMLs as well as in most ALLs. Unlike the CD33 antigen, CD45 does not internalize after antibody binding. Phase 1 testing of the radioimmunoconjugate I-131 BC8 (murine anti-CD45), in combination with total body irradiation and cyclophosphamide as a pretransplant marrow-ablative regimen among 34 patients, defined the maximum tolerated dose of antibody-delivered radiation.
The liver was determined to be the dose-limiting organ at a maximum tolerated dose of 10.5 Gy. An average of 24 Gy to marrow and 50 Gy to spleen could be delivered at the maximum tolerated liver dose. This radioimmunoconjugate is now in ongoing phase 2 testing. Radiolabeled I-131anti-CD45 antibody has also been given to patients with AML in first remission, again as a preparative regimen with cyclophosphamide and busulfan for ABMT.
Whether the addition of the radioimmunoconjugate confers a benefit to high-dose busulfan and cyclophosphamide or whether straight busulfan/cyclophosphamide followed by PBSCT alone is just as efficacious, remains to be determined in randomized phase 3 testing.
Can Supportive Care Enhance the Efficacy of AML Treatment Regimens?
Despite the findings of several major clinical trials, which examined the effects of growth factors following induction therapy for AML, especially for older patients, considerable controversy still exists, even though the study conclusions are generally suggestive of benefit.
The majority of these trials seem to indicate that growth factors are safe and somewhat useful, depending upon individual circumstances. Each of these trials differs according to the type of growth factor used, the specific drugs given as treatment for leukemia, the trigger point for starting and stopping growth factor support, whether growth factors were given concurrently with or following completion of chemotherapy, and the type of leukemia investigated.
Two large American trials conducted by the CALBG and ECOG administered GM-CSF at the end of chemotherapy, and only after development of documentable marrow hypoplasia (day 10 in the ECOG trial).
In the ECOG trial, a modest but significant improvement in the complete remission rate was shown, and there was some reduction in mortality secondary to reduced treatment toxicity, as well as a trend toward improved long-term survival. GM-CSF was given before, during, and after chemotherapy to some patients in 2 European trials, allowing investigators to test whether the growth factors can actually prime leukemic blasts, making them more sensitive to cytotoxic chemotherapy agents, although in 2 other perspective randomized studies, no such benefit was demonstrated.
In all of these trials, the reduction in duration of neutropenia following growth factor support was approximately 2-3 days. None of the aforementioned trials reported stimulation of leukemic clones as a result of growth factor use.
The role of granulocyte colony-stimulating factor (G-CSF) has also been examined in 3 large clinical trials.
Even though they were conducted at different study centers, the characteristics as well as the results of these trials were quite similar, demonstrating a 5-day reduction in the duration of neutropenia, no reduction in toxic death, and no improvement in disease-free or overall survival. The complete remission rate was higher in G-CSF-treated patients in the French-sponsored trial, while 1 of the American trials showed a statistically significant reduction in days of hospitalization.
Platelet growth factors have also been considered in the context of acute leukemia treatment because of the complicating effects of profound and prolonged thrombocytopenia during induction, consolidation, and transplantation regimens. IL-11, the first commercially available platelet growth factor, and recombinant megakaryocyte growth and development factor (MGDF, thrombopoietin) have been studied in patients with both solid and malignant tumors, following administration of cytotoxic chemotherapy with transfusion reduction as the end point. In a placebo-controlled trial in which women with breast cancer received dose-intensive cyclophosphamide/doxorubicin therapy, the use of recombinant human IL-11 (rhIL-11) enabled 68% of patients to avoid platelet transfusions during the course of treatment, compared with 41% of individuals receiving placebo.
Gordon and associates also demonstrated activity in breast cancer patients in a phase 1 trial that showed rhIL-11 activity at all doses studied. Another placebo-controlled, randomized, multicenter trial in 93 patients who had already received platelet transfusions following chemotherapy for treatment of solid tumors demonstrated that rhIL-11 eliminated the need for platelet transfusions in 30% of evaluable patients treated at a dose of 50 mcg/kg daily.
A small trial, in which rhIL-11 was administered to patients with leukemia and preleukemias following allogeneic stem cell transplant, demonstrated a reduction in platelet transfusion requirements without incurring any significant dose-related side effects.
Three trials, in which patients received MGDF following induction chemotherapy for AML, showed no complications or increased incidence of leukemic blasts, but also failed to demonstrate any discernible effect on platelet transfusion requirements.
Cripe and colleagues did confirm the safety of concurrent GM-CSF and rhIL-11 use in the setting of acute leukemia, and, perhaps, future trials should examine the role of combination growth factors with larger patient populations before definitive conclusions are drawn. Given the significance of death from hemorrhage during treatment of patients with acute leukemia, such research is likely warranted.