W. G. Wierda performed research, analyzed data, and wrote the article; T. J. Kipps, M. J. Keating, J. R. Brown, and J. G. Gribben performed research; M. Browning, L. Z. Rassenti, and A. W. Greaves collected data; D. Neuberg analyzed data; and S. M. O'Brien designed the study, performed research, analyzed data, and wrote the article.
Alemtuzumab is highly effective at treating chronic lymphocytic leukemia (CLL) in bone marrow, which is the usual site of residual disease after fludarabine-based treatment. Eliminating residual disease potentially is associated with longer remission and overall survival. The authors of this report evaluated the ability of subcutaneous alemtuzumab to treat residual disease.
Patients in partial remission (PR), nodular PR (nPR), or complete remission (CR) who had disease in bone marrow established by 2-color flow cytometry analysis were enrolled and received alemtuzumab 30 mg subcutaneously 3 times weekly for 4 weeks, and patients had the option to self-administer alemtuzumab. Responders were patients in PR who converted to an nPR or a CR, patients in nPR who converted to a CR, and patients in CR who had no evidence of disease on 2-color flow cytometry analysis after treatment.
There were 31 patients enrolled, of whom 29 were evaluable, and there were 23 responders (4 of 4 patients who achieved a CR, 8 of 9 patients who achieved an nPR, and 11 of 16 patients who achieved a PR. Nonresponders had significantly lower plasma alemtuzumab levels at the end of treatment. Furthermore, higher plasma alemtuzumab levels at the end of treatment were correlated with a longer response duration. Compared with the results from an historic group that received intravenous alemtuzumab for residual disease, there was a trend toward a higher response rate but a shorter response duration with subcutaneous alemtuzumab.
Patients with chronic lymphocytic leukemia (CLL) who achieve complete remission (CR) or nodular partial remission (nPR) according to 1996 National Cancer Institute Working Group (NCI-WG) criteria have longer remission duration and survival than patients who achieve partial remission (PR) or who fail to respond to treatment.1, 2 Molecular analysis for the immunoglobulin heavy-chain variable (IGHV) gene and 4-color flow cytometry analysis are more sensitive than morphologic evaluation of the bone marrow used to establish CR according to 1996 NCI-WG criteria. A study by Moreton et al3 of alemtuzumab administered to previously treated patients with CLL demonstrated that those who achieved minimal residual disease (MRD)-free status according to 4-color flow cytometry analysis had longer survival. This suggested that evaluating for MDR-free status could add to response evaluation.
Studies in fludarabine-refractory, previously treated, and treatment-naive patients with CLL clearly demonstrated the efficacy of intravenous alemtuzumab.3–5 Subcutaneous alemtuzumab is effective in fludarabine-refractory patients and as initial treatment for patients with CLL.6, 7 Previously, we reported results from a clinical trial evaluating the efficacy of alemtuzumab intravenously 3 times weekly for 4 to 8 weeks in eliminating residual disease after treatment with chemotherapy or chemoimmunotherapy.8 That study demonstrated an overall response rate (ORR) of 46% in 41 patients; the median response duration was not reached at a median follow-up of 18 months.
Here, we report the results with subcutaneous alemtuzumab for residual disease and include a retrospective evaluation of bone marrow for MRD by 4-color flow cytometry analysis and end-of-treatment plasma alemtuzumab levels. Furthermore, we compare outcomes between the subcutaneous alemtuzumab group and an updated historic group of 58 patients who received intravenous alemtuzumab.
MATERIALS AND METHODS
This was a multicenter, phase 2 clinical trial conducted by the CLL Research Consortium. Patients were enrolled from March 2005 to May 2007, including 16 patients at The University of Texas M. D. Anderson Cancer Center, 11 patients at University of California-San Diego, and 4 patients at Dana Farber Cancer Institute. All respective institutional review boards approved the protocol and the informed consent form. All participants provided informed consent according to institutional guidelines. Patients with CLL/small lymphocytic leukemia were aged >18 years, had a performance status between 0 and 2, had adequate liver and kidney function (creatinine, total bilirubin, and aspartate and alanine aminotransferase levels all ≤2 times the upper limit of normal), and achieved CR, nPR, or PR according to 1996 NCI-WG criteria with their most recent treatment. Patients in CR were required to have a residual population of cluster of differentiation 5 (CD5)-positive/CD19-positive cells that comprised ≥10% of the bone marrow mononuclear population or CD5-positive/CD19-positive cells that comprised <10% of the bone marrow mononuclear population with a kappa/lambda ratio of >6 or <0.33, respectively. Patients were excluded for active infection, history of anaphylaxis with rodent-derived monoclonal antibodies, prior treatment with alemtuzumab, pregnancy, corticosteroids, and uncontrolled autoimmune hemolytic anemia. Patients could not have received treatment for CLL within 2 months of enrollment; there was no upper limit for the time from last chemotherapy.
The trial intended to enroll 25 patients in nPR or PR and 25 patients in CR. Accrual was completed to the nPR/PR group, but the CR group was very slow to accrue and, as a result, was closed. The current analysis includes 26 patients who were enrolled in the nPR/PR group and 5 patients who were enrolled in the CR group. Patients were registered in the CLL Research Consortium database, and all data were collected through the web-based clinical database system. Upon enrollment, patients underwent complete history and physical examination, pretreatment laboratory evaluation (including blood chemistries and blood count), and bone marrow aspirate and biopsy. Some patients had prognostic factors evaluated, including IGHV mutation status9 and leukemia cell expression of ζ chain-associated protein kinase 70 (ZAP-70)9 and CD38 (in respective clinical laboratories), by flow cytometry.
Treatment consisted of single-agent, subcutaneous alemtuzumab at doses of 3 mg, 10 mg, and 30 mg on Days 1, 2, and 3, respectively, followed by 30 mg subcutaneously 3 times weekly for a total of 12 doses, including the initial dose escalation. The research nurses with previously prepared educational materials taught patients self-administration. Once patients demonstrated self-administration to the staff, this was continued for the duration of treatment. Dosing, injection site, and toxicities were documented by patients with diaries, which were submitted for review and documentation at the end of treatment. Premedications consisted of 25 mg to 50 mg diphenhydramine and 650 mg oral acetaminophen. In addition, trimethoprim/sulfamethoxazole twice daily, 3 times weekly, or he equivalent for Pneumocystisjiroveci prophylaxis and valacyclovir 500 mg daily or valganciclovir 450 mg twice daily for viral prophylaxis were given for the duration of treatment and for a minimum of 3 months after completing alemtuzumab. Responses to treatment were assessed within 4 weeks of the end of treatment, and patients could receive a second 4-week course if residual disease was documented at their first response assessment.
Response Assessment and Follow-Up
Criteria for response were as follows8: Patients who enrolled in CR were considered responders if they had no disease detected on 2-color flow cytometry bone marrow analyses after treatment; specifically, 2-color flow cytometry bone marrow analysis could not demonstrate any kappa:lambda skewing (3:1 or 1:3) regardless of the proportion of cells that coexpressed CD5 and CD19. For patients who enrolled in nPR, responders must have achieved CR according to 1996 NCI-WG criteria. For patients who enrolled in PR, responders must have achieved either nPR or CR according to 1996 NCI-WG criteria. For patients who received 2 courses of alemtuzumab, response was assessed after the second course. Patients had a physician visit at least every 3 months after response assessment and underwent bone marrow aspirate and biopsy at least every 6 months to evaluate for loss of response. Time to loss of response and survival were dated from study enrollment. Loss of response was defined for patients who enrolled in CR as the loss of 2-color flow cytometry response in bone marrow, or developing nodules, or ≥30% lymphocytes in bone marrow, or lymphadenopathy, or hepatosplenomegaly; for patients who enrolled in nPR, loss of response was defined as redeveloping criteria for nPR or for PR; and, for patients who enrolled in PR, loss of response was defined as redeveloping criteria for PR.
Plasma Alemtuzumab Levels and 4-Color Flow Cytometry
Plasma alemtuzumab levels were assessed retrospectively at the end of treatment using the flow cytometry-based method developed by Rebello and Hale.10 Bone marrow samples that were cryopreserved at response assessment were evaluated retrospectively for MRD by standardized 4-color flow cytometry11 (Genzyme Corporation, Cambridge, Mass). Patients also were evaluated for antialemtuzumab antibodies (BioAnalab Limited, Oxford Science Park, United Kingdom), and none were positive after treatment (data not shown).
Historic Intravenous Alemtuzumab Comparison Group
We previously conducted a trial of intravenous alemtuzumab for residual disease that enrolled 41 patients.8 The trial was revised to enroll an additional 17 patients among CLL Research Consortium clinical sites. We updated follow-up of all 58 patients to include response duration and survival (Table 1). This group was used as an historic intravenous alemtuzumab comparison group.
Table 1. Patient Characteristics Before Alemtuzumab
Median time to last Tx before alemtuzumab [range], mo
Follow-up [range], mo
Associations between categorical patient characteristics were evaluated with the Fisher exact test. Comparisons of measured data between the subcutaneous cohort and the intravenous cohort, as well as comparisons between responders and nonresponders, used either the 2-sample t test or its nonparametric alternative, the Wilcoxon rank-sum test.
The distribution of time to loss of response and survival were estimated with the Kaplan-Meier method.12 Kaplan-Meier curves for categorical variables were plotted for recurrence-free survival. The P values for testing differences between subgroups/levels for each variable were calculated by using log-rank tests.13 Time intervals were measured from the first day of treatment until progression, recurrence, or death; and deaths from all causes were included.
Patient Characteristics and Treatment
There were 31 patients enrolled who received subcutaneous alemtuzumab, including 17 patients in PR, 9 patients in nPR, and 5 patients in CR (Table 1). The median age was 65 years (range, 48-84 years), the median number of previous treatments was 1 (range, 1-6 previous treatments), and the time from last treatment to alemtuzumab was 7.6 months (range, 2-45 months). All patients who had previously received a fludarabine-based regimen were at least 4 months from treatment (data not shown). Enrollment status according to the last treatment regimen and response to that regimen are shown in Table 2.
Table 2. Response to Last Treatment Before Subcutaneous Alemtuzumab
NCI-WG Response to Last Tx
No. of Evaluable Patients
NCI-WG indicates National Cancer Institute Working Group; Tx, treatment; CR, complete remission; nPR, nodular partial remission; PR, partial remission; CVP-R, cyclophosphamide, vincristine, prednisone, and rituximab; R-GM-CSF, rituximab and granulocyte-macrophage–colony-stimulating factor; FR, fludarabine and rituximab; FCR, fludarabine, cyclophosphamide, and rituximab; HDMP-R, high-dose methylprednisolone, and rituximab.
Response to Subcutaneous Alemtuzumab
The ORR was 74% (90% exact binomial confidence interval, 58%-86%). Seven patients received a second alemtuzumab course (Table 3). There was 1 death (a patient in PR) that occurred within 1 month of the last alemtuzumab dose because of refractory autoimmune hemolytic anemia (AIHA), and 1 patient was lost to follow-up (a patient in CR); neither patient could be evaluated for response, and both were included in the denominator for ORR. All 4 evaluable patients who enrolled in CR were responders to subcutaneous alemtuzumab according to specified criteria (Table 4). Eight of 9 patients who enrolled in nPR and 11 of 17 patients who enrolled in PR responded to subcutaneous alemtuzumab treatment. Four-color flow cytometry analysis was performed retrospectively on cryopreserved bone marrow samples that were obtained at response assessment for 22 of the evaluable patients (Table 4). Two of 4 patients who enrolled in CR had post-treatment bone marrow MRD evaluations, and neither was MRD-negative at end of treatment. Twenty-nine percent of 7 patients who enrolled in nPR and 18% of 13 patients who enrolled in PR were MRD-negative according to 4-color flow cytometry analysis at response assessment.
Alemtuzumab therapy was highly effective at eliminating morphologic evidence of residual disease from blood and bone marrow and was less effective at treating residual lymphadenopathy (Fig. 1). The median plasma alemtuzumab concentration at the end of treatment was significantly higher (P = .003) for responders (n = 19; mean concentration, 9.3 μg/mL) than for nonresponders (n = 6; mean concentration, 0.1 μg/mL) (Fig. 2). Among the responders who were evaluated, the end-of-treatment plasma alemtuzumab concentration was correlated with response duration (Fig. 3A). The overall median plasma level was 6 μg/mL. Significantly longer response duration was observed for responders who had alemtuzumab levels above the median (P = .05). The mean alemtuzumab plasma concentration at the end of treatment for the 19 patients who received 1 course of alemtuzumab was 7.7 ± 4.7 μg/mL compared with 2.6 ± 3.8 μg/mL for the 6 patients who received 2 courses of alemtuzumab. The response rate for the 6 patients who received 2 courses was 50% versus 84% for the 19 patients who received 1 course of subcutaneous alemtuzumab.
Responses were evaluated according to pretreatment prognostic factor status, and none of the factors correlated with response or response duration, possibly because of the small patient numbers (Table 5). The median follow-up is 31 months, and the median response duration is 13 months. Five patients have died, including 1 patient with refractory AIHA, as mentioned above, and 4 patients during follow-up 27 to 31 months after treatment on this study, all with progressive, refractory CLL.
Table 5. Response to Subcutaneous Alemtuzumab by Pretreatment Characteristics in Evaluable Patients
No. of Patients
IGHV indicates immunoglobulin heavy chain variable region; ZAP-70, ζ chain-associated protein kinase 70; CD38, cluster of differentiation 38 (also known as cyclic adenine diphosphate ribose hydrolase).
No. of alemtuzumab courses
No. of prior treatments
Mutated (<98% germline)
Unmutated (≥98% germline)
Negative (<20% of cells)
Positive (≥20% of cells)
Negative (<30% of cells)
Positive (≥30% of cells)
Safety and Toxicity
The majority of treatment-related adverse events were grade 1 or 2 (Table 6). Patients were able to self-administer alemtuzumab without complication or difficulties, as documented by their diaries. Injection site reactions (grade 1 or 2), consisting of erythema, pruritus, and induration that typically lasted 2 or 3 days with complete resolution, were reported by 24 patients (most were grade 1). Four patients experienced serious infections associated with neutropenia; 2 patients developed pneumonia, 1 patient developed gram-positive bacteremia, and 1 patient developed fever of unknown origin. All serious infections occurred within 2 months of completing alemtuzumab and were not associated with plasma alemtuzumab levels greater than the median (6 μg/mL) at the end of treatment. Two patients developed symptomatic cytomegalovirus (CMV) reactivation that required ganciclovir treatment, and 1 patient developed a fever and was treated empirically for presumed CMV reactivation. Hematologic toxicities consisted of neutropenia, which was grade 3 in 19% of patients and grade 4 in 19% of patients. Grade 3 anemia was noted in 3% of patients, and no grade ≥3 thrombocytopenia was observed. These hematologic toxicities were attributed to treatment with alemtuzumab and were not associated with increased end-of-treatment plasma alemtuzumab levels (Table 6).
Table 6. Adverse Events With Subcutaneous Alemtuzumab
No. of Patients (%)
CMV indicates cytomegalovirus; URI, upper respiratory tract infection; FUO, fever of unknown origin; LFT, liver function test; Bili, bilirubin; SOB, shortness of breath.
Injection site reaction/rash
Hepatic, LFT or Bili
There were 4 serious adverse events, all of which occurred at the end of alemtuzumab treatment or within 1 month of completion. One patient developed uncontrolled hemolytic anemia during treatment and died (Table 6), as mentioned above. One patient developed neutropenic fever and was admitted to the hospital; blood cultures grew Staphylococcushominis, and the infection was treated and resolved. One patient was admitted and treated with empiric antibiotics for neutropenic fever. Radiographs revealed no source of infection, and cultures remained negative; the fever resolved with antibiotics (Table 6). The same patient experienced a second event of pneumonia with an infiltrate observed on a chest x-ray in the setting of neutropenia that was treated and resolved. There were no reported late infections or infection-related deaths.
Comparison of Subcutaneous Versus Intravenous Alemtuzumab
Response and response duration, as defined for the subcutaneous population, for 58 patients who received intravenous alemtuzumab for residual disease was updated (Table 1).8 The characteristics of this group were similar to those of the 31 patients who received subcutaneous alemtuzumab with the after noted differences: The intravenous group received fewer courses of alemtuzumab, was younger, had lower platelet count, and had higher β-2 microglobulin levels at enrollment (Table 1). The proportion of patients with lymphadenopathy before alemtuzumab was similar between the groups (Table 1). Among the patients who had lymphadenopathy, 60% of those in the subcutaneous group and 52% of those in the intravenous group had more than 1 involved lymph node site. Furthermore, the largest lymph node measured ≥2 cm in 50% of those in the subcutaneous group and in 31% of those in the intravenous group. The ORR using the same criteria was lower (P = .07) for the patients who received intravenous alemtuzumab (Table 1). However, among the responders, the proportion of patients who had MRD-negative bone marrow status was significantly higher (P = .005) in the intravenous group, although the method used to evaluate MRD differed between trials; 4-color flow cytometry analysis was used in the subcutaneous trial, and a polymerase chain reaction (PCR)-based assay for the clonal IGHV gene was used in the intravenous trial. There was a trend (P = .07) for shorter response duration, when using the same criteria for both trials, for patients who received subcutaneous alemtuzumab versus intravenous alemtuzumab (median response duration, 13 months vs 31 months, respectively) (Fig. 3B). There was no difference in overall survival between the trials at the most recent follow-up (data not shown).
Investigators have evaluated strategies to achieve MRD-free status by incorporating alemtuzumab into the approach with the expectation that this may translate into longer remission duration and survival.8, 14–16 The phase 3 trial of the German CLL Study Group clearly demonstrated longer progression-free survival for patients who received alemtuzumab.14, 16 That trial demonstrated a high response rate of 74%, although the response duration was disappointing. There was an apparently shorter response duration with subcutaneous alemtuzumab compared with the that reported in the historic intravenous trial. The characteristics of the patients were similar in our historic comparison group, making it less likely that the shorter response duration was caused by differences in patient characteristics. The shorter response may have been caused by the reduced ability of subcutaneous alemtuzumab to clear MRD when given for the defined treatment period. In both trials, patients received a 4-week course of alemtuzumab (12 doses) with the option for a second 4-week course for patients who had residual CLL after the first course. Fewer patients received a second course of alemtuzumab on the intravenous trial. Also, the proportion of responders who were negative for MRD was lower for patients who received subcutaneous alemtuzumab versus intravenous alemtuzumab. One difficulty in interpreting this observation is that the methods for evaluating MRD were different between trials. For the subcutaneous trial, 4-color flow cytometry was used to evaluate MRD, which potentially has different sensitivity than the PCR-based assay that was used to evaluate patients in the intravenous trial. Also, these trials were not contemporaneous but sequential; therefore, unappreciated trial effects may account in part for the differences in response duration between the patient populations.
A major concern for the use of alemtuzumab in residual disease is infection, including CMV reactivation and opportunistic infections.14, 17 The German CLL Study Group had to terminate their phase 3 trial, and Cancer and Leukemia Group B terminated their phase 2 trial of alemtuzumab for MRD, both because of unacceptable, infection-related toxicity. One possible explanation for this observation was a shorter duration between the completion of chemotherapy and the initiation of alemtuzumab. Nevertheless, the current study demonstrated that alemtuzumab can be self-administered safely by patients with residual disease, with the expected reactivation of CMV in a minority of patients. The incidence of symptomatic CMV reactivation on this trial was lower than previously reported, likely because some patients received valganciclovir for CMV prophylaxis. Reactivation of CMV can be minimized, if not eliminated, with prophylactic valganciclovir.18 It is possible that trials of alemtuzumab for residual disease that reported serious and, in some cases, fatal infectious complications were the result of allowing too little time for immune reconstitution between fludarabine-based chemotherapy and the initiation of alemtuzumab. In our previous trial of 58 patients who received intravenous alemtuzumab and in the current trial with 31 patients who received subcutaneous alemtuzumab for residual disease, there was no treatment-related, infectious mortality. The median time from fludarabine-based therapy to alemtuzumab was >6 months, and we recommend waiting at least 4 to 6 months before initiating alemtuzumab to minimize infectious complications.
To our knowledge, this is the first trial to report the self-administration of subcutaneous alemtuzumab. Patients were taught to self-administer alemtuzumab and were able to complete the treatment without complication or incident. Subcutaneous alemtuzumab was associated with local injection-site reactions, but these were mild and were minimized by supportive measures, such as applying ice before and after the injection and administering acetaminophen and diphenhydramine premedication.
In the current trial, plasma alemtuzumab levels were correlated with response and response duration. There are limited reports of single-agent alemtuzumab pharmacokinetics. It is noteworthy that Hale et al19 evaluated the pharmacokinetics of alemtuzumab in 30 patients with recurrent disease who received intravenous alemtuzumab and in 20 untreated patients who received subcutaneous alemtuzumab. Their analysis indicated significant interpatient variability in pharmacokinetic parameters; however, the mean cumulative dose of alemtuzumab that was needed to reach a trough serum concentration of 1 μg/mL was 90 mg for the intravenous group and was substantially higher at 551 mg for the subcutaneous group. Furthermore, higher trough serum alemtuzumab concentrations were correlated significantly with a better clinical response and with achieving MRD-free status. Montillo et al15 evaluated pharmacokinetics in 16 patients who received alemtuzumab 10 mg subcutaneously 3 times weekly for 6 weeks as treatment for residual disease. Their data indicated that patients who achieved an area under the receiver operating characteristic curve on Day 15 of therapy that was >5 μg·hour/mL were more likely to achieve CR compared with individuals who had lower levels. In our study, patients who did not achieve a response had lower plasma levels of alemtuzumab at the end of treatment compared with responders. Furthermore, among the responders, those with end-of-treatment alemtuzumab levels lower than the median of 6 μg/mL had significantly shorter response duration than patients who had levels higher than the median. Given the significantly higher cumulative dose of subcutaneous alemtuzumab needed to achieve blood concentrations associated with response, we hypothesize that optimized treatment and improved safety and responses, specifically MRD-free status, could be achieved better by evaluating for MRD in a real-time fashion and administering individualized alemtuzumab subcutaneously or intravenously based on blood levels. Because blood reveals only 1 disease compartment, a time factor to allow for compartment equilibration is yet another consideration.
Further work to optimize this treatment strategy may make use of a new trial design in which patients have blood alemtuzumab levels measured and undergo real-time evaluations for MRD. Patients could begin with a 4-week course (12 doses of 30 mg) of subcutaneous alemtuzumab followed by evaluations of response, MRD, and blood drug levels. For patients with residual disease, treatment could continue, and the subcutaneous alemtuzumab dose or schedule could be adjusted, depending on blood drug levels; then, treatment would continue with monthly bone marrow evaluations either until patients achieved negative MRD status or a predetermined total duration of treatment is reached for patients who remain positive with potentially adequate blood alemtuzumab levels. This remains an interesting, active, and promising therapeutic strategy for patients with CLL. Continued use of this strategy should be undertaken only in the setting of a well designed, controlled clinical trial.
CONFLICT OF INTEREST DISCLOSURES
This clinical trial and correlative laboratory studies were supported by Bayer Healthcare. W. G. Wierda is a Leukemia and Lymphoma Society Clinical Scholar. This study was conducted with the infrastructure and support of the CLL Research Consortium.