Salvage Therapy
Remission Duration and Overall Survival in Relapsed ALL
The general prognosis for adults with relapsed or refractory ALL is poor, with the median OS ranging between 4 and 7 months (Figure 2).[42,43] As would be expected, patients who are resistant to initial induction therapy, or those who have a CR duration of less than 12 months, do particularly poorly. The expected 5-year OS for adults with relapsed ALL is 5% to 10%, although some groups report improved outcomes for patients with a first CR of greater than 2 years.[44,45] There are several drugs and regimens for the salvage setting endorsed by the current guidelines, but most of these strategies are inadequate, and patients should always be considered for enrollment in a clinical trial first. There are several very promising new agents under investigation, and patients should be referred to centers that are accruing on these protocols.
Conventional strategies
For patients who do not qualify for participation in a clinical trial, the decision regarding salvage strategy is largely based on duration of first remission, performance status, and organ function. It is also very important to consider drug classes and agents that the patient has not yet been exposed to. In addition, one must revisit the cytogenetic and molecular characteristics of the leukemia to devise an appropriate treatment plan.
An augmented version of the hyperCVAD regimen has been developed for use in adults with relapsed ALL.[43] The program includes intensified doses of dexamethasone(Drug information on dexamethasone) and vincristine, and the addition of pegylated asparaginase(Drug information on asparaginase) to the traditional hyperCVAD backbone. Ninety patients were enrolled in a trial of this regimen, most of them in first relapse, with 80% having received standard hyperCVAD prior to experiencing recurrent disease. The overall response rate in the study was 64%, with 47% of the patients meeting the definition of CR. The 30-day mortality was less than 10%. This may thus be an ideal regimen for relapsed patients who have not received prior L-asparaginase.
Clofarabine (Clolar) is a nucleoside analog approved for the treatment of pediatric ALL in the third-line setting.[46] Several combination approaches have also been explored in the pediatric and adult patient populations. Recently, a group from France tested two clofarabine-containing regimens in patients with relapsed ALL.[47] The first regimen consisted of clofarabine, dexamethasone, mitoxantrone, etoposide, and pegylated asparaginase. The CR rate was 41%, with an early death rate of 14%. Five of 37 patients experienced grade 3/4 neurologic toxicity. The second regimen tested was a combination of clofarabine and cyclophosphamide(Drug information on cyclophosphamide). This program also led to a relatively high CR rate of 50%, with very little early mortality and minimal unexpected toxicities. A multicenter phase II trial was also conducted in a group of pediatric patients using clofarabine combined with etoposide(Drug information on etoposide) and cyclophosphamide.[48] These patients were heavily pretreated, with 84% receiving the regimen as second or higher salvage. The CR rate in this study was 28%, and several patients were able to move on to alloSCT. Patients who had undergone alloSCT prior to being enrolled in this study seemed to be predisposed to severe liver toxicity (eg, veno-occlusive disease [VOD]), and the study was eventually amended to exclude this group. The Programa Español de Tratamiento en Hematologia (PETHEMA) group has published their experience with several clofarabine-based regimens in adults.[49] They have reported a CR rate of 31% and manageable toxicity, with several patients being able to proceed to alloSCT. Most of these patients had been treated with two or more chemotherapy regimens.
For patients with T-ALL, nelarabine as a single agent has shown activity in the pediatric and adult settings.[26,50] One of the concerning adverse effects of nelarabine is neurotoxicity, which can manifest peripherally or centrally. For this reason, alternative dosing strategies are being explored to optimize the risk-to-benefit ratio. A study is underway evaluating the safety of giving nelarabine via continuous infusion over 5 days.[51] As mentioned previously, nelarabine would ideally be incorporated into the frontline management of these patients, and studies are ongoing.
On August 9, 2012, the US Food and Drug Administration (FDA) approved the labeling for the use of vincristine sulfate(Drug information on vincristine sulfate) liposomes (Marqibo) in patients with relapsed and/or refractory Ph-negative adult ALL. The drug was studied in relapsed patients who had received at least two previous chemotherapy regimens. It is given as a weekly intravenous infusion at 2.25 mg/m2.
Monoclonal Antibodies for ALL
Investigational and/or targeted strategies
As discussed earlier, using rituximab(Drug information on rituximab) and ofatumumab as examples, target-directed therapy against antigens expressed on the surface of leukemic cells represents an attractive strategy for fighting this disease and improving patient outcomes. A description of currently available and investigational monoclonal antibodies can be found in Table 2. In addition to CD20, CD19 and CD22 are also antigens that are highly expressed on B lymphoblasts.[52] Monoclonal antibodies against CD19 and CD22 are moving into advanced stages of development, and have demonstrated a high degree of activity, even in the most refractory settings.
Anti-CD22 antibodies. The rapid internalization of CD22 upon receptor binding makes it an excellent target for monoclonal antibody–cytotoxic chemotherapy conjugates.[53] Once inside the cell, the toxic component is released, ultimately leading to cell destruction and death. Theoretically, this allows for a minimal dose of chemotherapy to be delivered directly to the leukemic blasts, thus minimizing off-target toxicity.
Inotuzumab ozogamicin is a monoclonal antibody against CD22 that is linked to calicheamicin, a potent cytotoxin that induces double-stranded DNA breaks. The initial clinical trial of this agent was a dose-ranging study conducted in patients with B-cell non-Hodgkin lymphoma.[54] Objective responses were documented in 39% of patients who underwent treatment. The maximum tolerated dose was determined to be 1.8 mg/m2 administered every 3 to 4 weeks. Reversible thrombocytopenia was the most frequently encountered toxicity.
A phase II study was subsequently conducted in patients with relapsed ALL.[55] An initial dose of 1.3 mg/m2 was given to the first three patients to ensure safety, but most patients went on to receive 1.8 mg/m2. Patients with CD20-positive disease could have rituximab added to inotuzumab starting on cycle 3 if they exhibited stable disease or no improvement. Forty-nine patients were enrolled, and the majority (73%) received inotuzumab as second or higher salvage. Fourteen percent of patients had previously undergone alloSCT. After a median of 2 cycles of therapy, the overall response rate was 57%, with most patients achieving complete marrow response with incomplete recovery of platelets or peripheral blood cell counts. Fever within 48 hours of drug administration was the most common nonhematologic toxicity, occurring in 59% of patients. Other important toxicities included elevations in bilirubin and hepatic transaminases. Of 49 patients, 22 were treated and subsequently went to alloSCT. Of concern post-transplant was the development of clinical VOD in five patients (23%). Several of these patients had also received thiotepa(Drug information on thiotepa) or clofarabine as part of the preparative regimen, which are known to be potentially hepatotoxic. Two of the four patients undergoing a second alloSCT had clinical evidence of VOD post-transplant. Importantly, alloSCT did not appear to confer a survival benefit, potentially due to the refractoriness of the disease in the patients studied or to transplant-associated complications. Also, the addition of rituximab did not appear to benefit patients who received it according to protocol. MRD was assessed in most patients who responded to treatment, but MRD negativity did not correlate with improved survival.
A weekly schedule of inotuzumab was recently explored in an attempt to optimize the benefit-to-risk ratio based on the pharmacodynamics and pharmacokinetics of the drug.[56] Twenty patients were enrolled, and inotuzumab was given according to the following schedule: 0.8 mg/m2 IV on day 1, 0.5 mg/m2 IV on day 8, and 0.5 mg/m2 IV on day 15. Of note, this is the same cumulative dose per cycle as was given in the every-3-to-4-week schedule. Patients received a median of 2 cycles, with an overall response rate of 50%. The toxicity profile was similar to that of the previous study, with transient elevations of bilirubin and transaminases occurring in 35% of patients. Notably, however, there have been no instances of clinical VOD, including in the four patients who proceeded with alloSCT after receiving inotuzumab.
Overall Response Rates (ORR) of Inotuzumab vs Conventional Chemotherapy
These data are encouraging given the refractory nature of the disease in the patients who were treated. The response rates seen with inotuzumab compared to the rates seen with conventional chemotherapy in the salvage setting are presented in Table 3.
Anti-CD19 antibodies. CD19 is a surface receptor with nearly universal expression on B-ALL cells.[57] The receptor also internalizes sufficiently upon binding to make it reasonable to target with immunoconjugated compounds.[58] Harnessing the immune system as a cancer-fighting modality has been studied extensively. Recruiting T cells directly to leukemic blasts using monoclonal antibody technology may lead to synergistic effects and improved outcomes.
Blinatumomab is in a class known as bispecific T-cell–engaging (BiTE) molecules, which actually contain components of two monoclonal antibodies. One arm of blinatumomab is designed to bind CD3-positive cytotoxic T cells, while the other recognizes CD19. Upon binding to CD19, the T cell becomes activated, thereby leading to the death of the malignant cell. Because of its short half-life, blinatumomab is given as a continuous infusion for 4 weeks, followed by a 2-week treatment break. The agent was initially used in adult patients with ALL who had persistent or resurgent MRD after induction or consolidation therapy.[59] This population would be expected to be at very high risk for clinical relapse with continuation of conventional chemotherapy alone. Of 20 patients enrolled, 16 converted to MRD-negative status, and 8 patients subsequently went on to alloSCT. One patient had to be taken off study after the development of a seizure disorder, which was reversible after discontinuation of the blinatumomab. Other common toxicities included fever and lymphopenia.
The GMALL group subsequently initiated a study to evaluate blinatumomab in patients with clinically relapsed disease. The results were updated this year at the American Society of Clinical Oncology annual meeting.[60] This is an ongoing phase II trial evaluating multiple dosing strategies in order to minimize toxicity (all regimens have been given as 28-day infusions with 14 days of rest). By the end of 2 cycles of therapy, 68% of patients achieved the study definition of CR or CR with incomplete hematological recovery. The final dose selected was 5 mcg/m2/d for 7 days, followed by 15 mcg/m2/d for 21 days. The most common adverse events included fever, headache, and tremor. Seizures were also noted in this study, as well as other CNS events, such as confusion. Premedication and supportive care strategies are being developed to prevent and manage these issues.
Case presentation. The current status and potential of the above options is best seen through a brief case presentation.
KA is a 30-year-old man who complained of a several-week history of fever, fatigue, and shortness of breath. His white blood cell count at presentation was 265,000/μL. A bone marrow biopsy was performed. The biopsy results and physical findings were consistent with a diagnosis of mixed-phenotype leukemia with B and myeloid features. Cytogenetics include t(4;11)(q21;q23). The immunophenotype analysis revealed a clonal population of cells positive for CD19, CD22, CD33, CD56, CD38, CD15, CD66, and CD184. Because of the mixed phenotype, the patient was induced with a combination of idarubicin(Drug information on idarubicin), cytarabine(Drug information on cytarabine), vincristine, and dexamethasone. He achieved CR after one cycle. The patient then received one cycle of consolidation, followed by a matched–related donor alloSCT. Seven months post-transplant, the patient was found to have relapsed disease. The patient was enrolled on a clinical trial using the anti-CD22 antibody inotuzumab ozogamicin. After one cycle, the patient achieved CR, as well as MRD negativity by multiparameter flow cytometry. He went on to receive one additional cycle of inotuzumab, followed by a second matched–related donor alloSCT. Two years after transplant, the patient remains in CR, with no evidence of the t(4;11)(q21;q23) or MRD by flow cytometry.
This case describes a dire situation encountered far too often in the clinic. This patient had very high-risk disease at baseline, and was appropriately taken directly to alloSCT after achieving a first CR. Relapsing within 1 year of the transplant normally would place this patient in a situation in which further therapy might be offered for palliation, but in which the possibility of cure, or even of attaining another CR, would be considered bleak.
Future Directions and Perspective
It is clear that the incorporation of monoclonal antibodies is changing the treatment paradigm for adults with ALL. Thus far, the only antibody that has been evaluated as part of the frontline treatment strategy is rituximab, and its benefit was demonstrated when added to an accepted chemotherapy regimen. The use of monoclonal antibodies against CD20 is potentially hampered by the varying degrees of expression of this antigen on lymphoblasts. An interesting concept that has recently been studied is the potential for corticosteroid-induced up-regulation of CD20 expression, which would broaden the applicability of these agents.[61]
Some of the most promising agents are still in the initial stages of clinical development, and thus are currently being tested in patients with relapsed and refractory disease. As data continue to accumulate, it will be important to move the most active agents to the frontline management setting. Inotuzumab ozogamicin is able to induce molecular remission in a large number of patients, which is critical to outcomes for patients undergoing induction and consolidation chemotherapy. Monoclonal antibodies also appear to be less toxic than conventional cytotoxic agents, making them particularly interesting for elderly patients with ALL. A strategy utilizing inotuzumab in combination with reduced doses of cyclophosphamide, vincristine, and corticosteroids (alternating with methotrexate(Drug information on methotrexate) and cytarabine) in elderly patients is being studied in an ongoing trial.[62] The anthracycline is omitted completely from this protocol, as it tends to cause significant problems for this age group.
A far-reaching goal is to one day incorporate multiple monoclonal antibodies into the frontline approach, thereby minimizing the use of cytotoxic chemotherapy. The challenge will be to determine the optimal way to sequence such combinations, and the amount of cumulative exposure necessary to maintain durable responses (ie, cure). Monoclonal antibodies are large molecules, so it will be important to monitor the impact increased use has on isolated CNS disease. The prophylactic approach may also have to be modified.
Targeted approaches for T-ALL need to be improved. Monoclonal antibodies active in this subtype are lacking, and this should be a focus of drug development going forward. Gamma-secretase inhibitors that block NOTCH receptor signaling are exciting biologically, but experience in the clinic has been far less promising.[63] ❍
Financial Disclosure: Dr. Kantarjian has received research grants from Sanofi, Micromet, and Pfizer. Drs. Mathisen and Jabbour have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
A hematologically stable CML patient developed a solitary, exquisitely painful ulceration, 7 cm × 5 cm, located on the mid-medial foreleg. There was never any evidence of venous insufficiency. The patient denied the possibility of a spider bite. What is the likely cause of this lesion?
