Chronic Lymphocytic Leukemia

Abstract

Patients with purine analogue-refractory chronic lymphocytic leukemia (CLL) have short survival and limited treatment options. Defining the best salvage strategies for this population is challenging, because limited data are available from clinical trials, and because studies have enrolled mixed populations (patients with recurrent and refractory disease or patients with refractory disease and Richter transformation). Moreover, patients with refractory CLL have a high incidence of unfavorable molecular and clinical features, such as high-risk genomic profiles, unmutated immunoglobulin heavy-chain genes, expression of zeta-chain-associated protein kinase 70, and bulky lymphadenopathies. These patients are also severely immunosuppressed because of the underlying disease and the treatments received, and experience a high rate of infectious complications that pose an additional difficulty in selecting treatment. Despite these challenges, in parallel with better characterizations of the biologic features of refractory CLL, the number of available treatment modalities for this population has increased. Several chemoimmunotherapy combinations have been developed, and novel agents with a different mechanism of action are being investigated in clinical trials. Furthermore, allogeneic stem cell transplantation with nonmyeloablative conditioning regimens is a therapeutic strategy that is increasingly offered to patients with refractory CLL.

Chronic lymphocytic leukemia (CLL), the most common form of adult leukemia, is characterized by a monoclonal proliferation of mature-appearing lymphocytes with distinctive phenotypic features. CLL is a heterogeneous disease with variable prognosis; some patients have an indolent course and a virtually normal life expectancy, others have aggressive disease and a short survival.¹ Several prognostic factors (including genomic abnormalities such as 17p and 11q deletions, unmutated immunoglobulin [Ig]V_H gene, expression of CD38 and ZAP70, and high β-2-microglobulin levels) have been identified and correlated with resistance to chemotherapy and more aggressive clinical behavior.^2–5

During the last decade, the treatment approach to CLL has dramatically changed from palliative to potentially curative. The goal of treatment has shifted from controlling leukocytosis and disease-related symptoms to achieving eradication of minimal residual disease (MRD). This shift has occurred because achieving eradication of MRD has been reported to be associated with improved remission duration and more recently improved overall survival (OS). Eradication of MRD can be observed after chemoimmunotherapy or can be achieved by pursuing further consolidation therapy, as suggested by the experience with alemtuzumab.^6–8 We believe that refractory and recurrent CLL should be considered 2 distinct entities, because they have different prognosis. However, in many studies these patient populations are not separated, which makes interpreting the results of these trials challenging. Refractory CLL is commonly defined as disease that fails to respond to a therapy (does not achieve a complete response [CR] or partial response [PR]) or that progresses within 6 months of such therapy.⁹ Recurrent CLL is defined as evidence of disease progression after ≥6 months in patients who had achieved a CR or a PR.⁹ It is important to acknowledge that over the past years, some of the characteristics of patients with refractory CLL have changed. For example, currently, the majority of these patients not only have failed treatment with single-agent fludarabine, but have failed treatment with a chemoimmunotherapy combination; therefore, refractory CLL indicates a disease that is not only resistant to purine analogues or alkylators used as single agents, but also to combinations that include purine analogues, alkylators, and monoclonal antibodies (in particular, rituximab) such as fludarabine, cyclophosphamide, and rituximab (FCR); fludarabine and cyclophosphamide (FC); fludarabine and rituximab; and pentostatin, cyclophosphamide, and rituximab (PCR).

Patients with refractory CLL also have a high incidence of unfavorable molecular and clinical features. Greater than half of these patients carry unfavorable genetic abnormalities, such as 17p13 and 11q23 deletions, unmutated IgV_H gene and bulky disease (lymph node[s] measuring >5 cm in dimension). Also, patients with refractory disease often have severe immunosuppression derived from the underlying disease and recent immunosuppressive treatments. According to retrospective studies, the incidence of severe infections in this population is as high as 30% with any given treatment, and the median survival duration is <10 months.^10,11

In this review, we will summarize the results that have been achieved with purine analogue-based therapies, monoclonal antibody treatment, allogeneic stem cell transplantation, and newer agents in clinical development in patients with refractory CLL.^10–14

PURINE ANALOGUE-CONTAINING REGIMEN

FCR

Fludarabine is the purine analogue most commonly used in the treatment of CLL. Cyclophosphamide is an alkylating agent with some single-agent activity in CLL. The rationale for using it in combination with fludarabine is based on the ability of fludarabine to inhibit the excision repair of DNA interstrand cross-links induced by cyclophosphamide and therefore to potentiate its cytotoxic activity in CLL cells.¹⁵ O'Brien et al¹⁶ reported that the FC combination was efficacious in patients with fludarabine-refractory CLL, yielding an overall response (OR) rate of 38%; most of the remissions were partial, with a median time to progression of <1 year. The addition of rituximab to fludarabine or other purine analogues has been associated with improved efficacy, possibly owing to the ability of fludarabine to down-regulate the expression of complement-resistance proteins CD55 and CD59 and therefore potentiate rituximab complement–mediated cytotoxicity. Rituximab also has the ability to down-regulate the expression of the antiapoptotic protein Bcl-2, thereby enhancing the sensitivity of CLL cells to fludarabine-induced apoptosis. In a study that evaluated FCR in previously treated patients with CLL, CR was achieved in 25% of 177 patients; nodular PR (nPR) and PR were achieved in 16% and 32% of the patients, respectively; the OR rate was 73%, and molecular disease remission was achieved in approximately 33% of patients achieving CR.¹⁷ There were 37 fludarabine-refractory patients in this study; 5% achieved CR, 11% achieved nPR, and 43% achieved PR, with an OR rate of 59% (Table 1).¹⁷

Table 1.

Chemoimmunotherapy and Combination Therapy Regimens for Patients With Recurrent or Refractory CLL

Reference	Treatment	No. of Patients	Refractory %	CR %	OR %	Refractory OR %
Wierda 200517	Fludarabine, 25 mg/m2 iv, d 2–4 + cyclophosphamide, 250 mg/m2 iv, d 2–4 + rituximab, 375–500 mg/m2 iv, d 1 for 6 cycles	177	21	25	73	59
Lamanna 200620	Cyclophosphamide, 600 mg/m2 iv, d 1 + pentostatine, 4 mg/m2 iv, d 1 + rituximab, 375 mg/m2 iv, d 1 for 6 cycles	32	25	25	75	75
Wierda 200622	Cyclophosphamide, 250 mg/m2 iv, d 3–5 + fludarabine, 25 mg/m2 iv, d 3–5 + alemtuzumab, 30 mg iv, d 1, 3, and 5 + rituximab, 375–500 mg/m2 iv, d 2 for 6 cycles	74	43	26	S6	51
O'Brien 200726	Fludarabine, 25 mg/m2 iv, d 5–7 + cyclophosphamide, 250 mg/m2 iv, d 5–7 + oblimersen, 3 mg/kg iv, d 1–7 for 6 cycles	120	57	9	17	10
Bosch 200228	Fludarabine, 25 mg/m2 iv, d 1–3 + cyclophosphamide, 200 mg/m2 iv, d 1–3 or 600 mg/m2 iv, d 1 + mitoxantrone, 6 or 8 mg/m2 iv, d 1 for 6 cycles	60	58	50	78	34
Elter 200532	Fludarabine, 30 mg/m2 iv, d 1–3 + alemtuzumab, 30 mg iv, d 1–3 for 6 cycles	36	33	30	83	66
Tsimberidou 200833	Oxaliplatin, 17.5, 20, 25 mg/m2 iv, d 1–4 + fludarabine, 30 mg/m2 iv, d 2, 3 + cytarabine, 1 g/m2 iv, d 2, 3 + rituximab, 375 mg/m2 iv, d 1 (d 3, cycle 1), 6 cycles	50	6O	12	40	33
Mauro 200229	Fludarabine, 25 mg/m2 iv, d 1–3 + cytarabine, 1000 mg/m2 iv, d 1–2 + mitoxantrone, 10 mg/m2 iv, d 1 + dexamethasone, 20 mg iv, d 1–3 for 6 cycles	31	45	58	81	50
Byrd 200652	Lumiliximab, 375 or 500 mg/m2 iv, d 1 + fludarabine, 25 mg/m2 iv, d 2–4 + cyclophosphamide, 250 mg/m2 iv, d 2–4 + rituximab, 375–500 mg/m2 iv, d 1 iv for 6 cycles	31	NR	48	71	NR

CLL indicates chronic lymphocytic leukemia; CR, complete response; OR, overall response; iv, intravenously; NR, not reported.

PCR

Pentostatin (2-deoxycoformicin) is a purine analogue reported to have single-agent activity in CLL. Pentostatin has the potential advantage of being less myelosuppressive than fludarabine.¹⁸ Weiss et al¹⁹ reported an OR rate of 77% with the combination of pentostatin plus cyclophosphamide in 13 patients with refractory CLL. The PCR combination has been investigated in patients with recurrent or refractory CLL. The study population reported by Lamanna et al²⁰ consisted of 32 patients; only 8 patients had disease that was refractory to fludarabine. Among the refractory patients, responses were achieved in 6 (75%), including 1 CR. This treatment was well tolerated and easy to administer, consisting of a 1-day regimen with pentostatin at a dose of 4 mg/m², cyclophosphamide at a dose of 600 mg/m², and rituximab at a dose of 375 mg/m² starting on Cycle 2 for a total of 6 cycles. A PCR regimen with a lower dose of pentostatin (2 mg/m²) has also been investigated,²¹ but to our knowledge, there are no data available with the lower dose of pentostatin in the refractory disease population.

Cyclophosphamide, fludarabine, alemtuzumab, and rituximab

Cyclophosphamide, fludarabine, alemtuzumab, and rituximab (CFAR) combines FCR with the monoclonal antibody alemtuzumab.²² The rationale for this combination is based on the synergy demonstrated between alemtuzumab and rituximab,²³ and on the unique properties of alemtuzumab in terms of its efficacy in patients with 17p abnormalities. Furthermore, the mechanism of action of alemtuzumab appears different and independent from those of purine analogues. In the CFAR regimen, alemtuzumab is given intravenously in 30-mg doses on Days 1, 3, and 5 of each 28-day treatment course for 6 intended courses. As with similar regimens associated with severe lymphopenia, antiviral and anti–Pneumocystis carinii prophylaxes are required. In addition, the risk for cytomegalovirus (CMV) reactivation needs to be considered. The CFAR combination is associated with significant myelo-suppression, and growth factors are routinely used. As with all chemoimmunotherapy combinations, allopurinol should be given routinely for tumor lysis prophylaxis during the first cycle.²² The CFAR study of Wierda et al²² enrolled patients with fludarabine-refractory disease, and preliminary results indicated an OR rate of 66% and a CR rate of 26%, suggesting clinical activity with this combination.

Fludarabine, cyclophosphamide, and oblimersen

The expression of Bcl-2 protein is associated with chemotherapy resistance and decreased survival in patients with CLL.²⁴ Recently, O'Brien et al²⁵ demonstrated that the addition of the Bcl-2 antisense oblimersen sodium to FC significantly increased the rate of CR/nPR in patients with recurrent or refractory CLL. The improved response rate appears to almost exclusively benefit the patients with fludarabine-sensitive disease according to recently updated data by O'Brien et al.²⁶

Fludarabine, cyclophosphamide, and mitoxantrone

In an attempt to increase the apoptotic effect induced by fludarabine in CLL cells, the combination of fludarabine, cyclophosphamide, and mitoxantrone has been investigated. Bellosillo et al²⁷ demonstrated a synergistic effect of fludarabine and mitoxantrone on CLL cells. On the basis of these data, Bosch et al²⁸ conducted a trial in which 37 patients with recurrent or refractory CLL were treated with FCM. Fludarabine was given intravenously at a dose of 25 mg/m² on Days 1 to 3, cyclophosphamide at a dose of 200 mg/m² was given on Days 1 to 3, and mitoxantrone at a dose of 6 mg/m² was given on Days 1 to 3 in 4-week intervals for up to 6 courses. An additional 23 patients received FCM with cyclophosphamide at a dose of 600 mg/m² and mitoxantrone at a dose of 8 mg/m² on Day 1. CRs were achieved in 50% of patients, MRD negativity was observed in 17%, and PRs were achieved in 28%. The median duration of response was 19 months. The median survival from the initiation of chemotherapy was 41 months, whereas the median survival for patients who achieved MRD negativity and CR was not reached; it was 25 months for patients not achieving CR.

Fludarabine, cytarabine, mitoxantrone, and dexamethasone

Mauro et al²⁹ investigated the combination of fludarabine, cytarabine, mitoxantrone, and dexamethasone in patients with recurrent disease. This combination was developed based on potential synergy between cytarabine and fludarabine and the known activity of mitoxantrone and dexamethasone in lymphoproliferative disorders. Because of its myelosuppressive effect, this treatment was offered only to patients aged ≤60 years. Thirty-one patients were treated, including 14 with fludarabine-refractory disease. OR in the patients with refractory CLL was 50%, with 29% achieving a CR.

Fludarabine and alemtuzumab

Single-agent alemtuzumab is active in patients with fludarabine-refractory CLL, with disease in blood and bone marrow being particularly sensitive to this agent.³⁰ Kennedy et al³¹ and Elter et al³² reported their results with fludarabine and alemtuzumab (FluCam) in patients with recurrent or refractory CLL. The latter study demonstrated an OR rate of 83% with 11 CRs, 19 PRs, 1 patient with stable disease, and 5 patients with progressive disease (of 36 treated patients). Notably, the patient responses observed in this trial were achieved with a cumulative FluCam dose that was lower than the single-agent dose.

Oxaliplatin, fludarabine, cytarabine, and rituximab

It is known that fludarabine and cytarabine act in a synergistic fashion based on the ability of fludarabine to inhibit the resynthesis step of excision repair and increase the level of arabinofuranosylcytosine triphosphate in CLL leukemia cells.¹³ The third-generation platinum compound oxaliplatin was added because platinum compounds induce DNA interstrand cross-links. Oxaliplatin also has a lower renal and auditory toxicity than other platinum compounds. Tsimberidou et al³³ studied the oxaliplatin, fludarabine, cytarabine, and rituximab combination in a phase 1/2 study that enrolled patients with fludarabine-refractory disease as well as patients with Richter transformation. The phase 1 component was based on increasing doses of oxaliplatin (17.5 to 25 mg/m²) and fixed doses of fludarabine, cytarabine, and rituximab. Of the 30 patients with fludarabine-refractory disease, 28 were evaluable; 2 patients achieved CR, and 8 achieved PR. It is interesting to note is that the responses were noted mostly at the 25-mg dose of oxaliplatin.

MONOCLONAL ANTIBODIES

Alemtuzumab (Campath-1H) is a humanized immunoglobulin-1 monoclonal antibody against the human CD52 antigen. Alemtuzumab has shown encouraging activity as a single agent and in combination regimens.^{23,30–32,34–42} The OR and CR rates across trials with alemtuzumab as a single agent were 41% (range, 31%–54%) and 11% (range, 0%–36%), respectively (Table 2). Clinical activity was demonstrated in heavily pretreated patients, including those with fludarabine-refractory CLL. Keating et al³⁵ reported the results of a large, international study of alemtuzumab in 93 patients exposed to alkylating agents who had failed to respond to fludarabine therapy. The alemtuzumab dose was 30 mg, 3 times a week, for a maximum of 12 weeks. The overall median survival was 16 months for the overall population and 32 months for responders, with a median time to disease progression of 4.7 months and 9.5 months, respectively. Of note, the antitumor effect of alemtuzumab was more significant in blood and bone marrow than in lymph nodes. Patients with lymph nodes measuring ≥5 cm in dimension were significantly less likely to respond to alemtuzumab. Alemtuzumab has clinical activity in patients with deletion 17p13, a known unfavorable genomic aberration. Lozanski et al⁴² reported the results with alemtuzumab treatment in 36 patients with recurrent CLL, including 15 patients with p53 mutations or deletion, and clinical responses were observed in 40%. In 2005, More-ton et al⁶ in a nonrandomized clinical trial reported that prolonged treatment with alemtuzumab with the goal of eradicating residual disease was associated with improved OS and treatment-free survival. In this study, 91 patients received alemtuzumab at a dose of 30 mg 3 times a week until maximum response was achieved; 44 patients with fludarabine-refractory CLL were included in this trial. The median OS was longer in MRD-negative patients than in those who had an MRD-positive CR; the median survival was not reached for patients with MRD-negative CRs, was 60 months for patients with MRD-positive CRs, was 70 months for patients with PRs, and was 15 months for patients in whom the response was not reported. Recently, Stilgenbauer et al³⁸ reported the results of the CLL2H trial of the German CLL Study Group with subcutaneous alemtuzumab in fludarabine-refractory patients. In this study, 103 patients received alemtuzumab at a dose of 30 mg 3 times a week. The OR rate was 34% (4% CR rate and 30% PR rate), the median progression-free survival (PFS) was 7.7 months, and the median OS was 19.1 months.

Table 2.

Alemtuzumab in Patients With Recurrent/Refractory CLL

Reference	Treatment	No. of Patients	Refractory %	CR %	OR %	Refractory OR %
Osteborg 199734	Alemtuzumab, 30 mg iv, 3× weekly up to 12 wk	29	NR	4	42	NR
Keating 200235	Alemtuzumab, 30 mg iv, 3× weekly up to 12 wk	93	48	2	33	29
Rai 200236	Alemtuzumab, 30 mg iv, 3× weekly up to 16 wk	24	71	0	33	29
Ferrajoli 200337	Alemtuzumab, 30 mg iv, 3× weekly up to 12 wk	42	55	5	31	26
Stilgenbauer 200838	Alemtuzumab, 30 mg sc, 3× weekly up to 12 wk	103	100	4	34	34
Moreton 20056	Alemtuzumab, 30 mg iv, 3× weekly, to maximum response	91	48	36	55	50
Cortelezzi 20053S	Alemtuzumab, 10 mg sc, 3× weekly, up to 18 wk	16	88	25	50	50
Osuji 200540	Alemtuzumab, 30 mg iv, 3× weekly, to maximum response	24	62	18	54	NR
Lozanski 200442	Alemtuzumab, 30 mg iv, 3× weekly up to 12 wk	36	81	6	31	24

CLL indicates chronic lymphocytic leukemia; CR, complete response; OR, overall response; iv, intravenously; NR, not reported; sc, subcutaneously

To our knowledge, the optimal dose and schedule of alemtuzumab remain to be defined. The most common infusion-related adverse events are rigors, fever, dyspnea, skin rash, and hypotension, and these symptoms may be significantly reduced by using the subcutaneous route of administration. Other common toxicities include cytopenias and opportunistic infections, particularly Pneumocystis carinii pneumonia, as well as reactivation of herpes viruses. Most patients receive trimethoprim-sulfa twice daily 3 times a week and daily antiherpes virus treatment with acyclovir, famcyclovir, or valacyclovir. CMV reactivation may occur in patients receiving alemtuzumab-based therapy. Careful and frequent monitoring of CMV status is mandatory in patients receiving this therapy to institute prompt treatment of patients who develop CMV reactivation. Oral valganciclovir given daily has been found by O'Brien et al to be an efficient strategy to prevent CMV reactivation and associated complications. However, need for daily dosing, the potential for myelo-suppression, and the significant cost of this treatment may limit its routine use.⁴³

Faderl et al²³ explored the use of rituximab and alemtuzumab in combination. In a population of 48 heavily pretreated patients with lymphoproliferative disorders, greater than half of whom were refractory to fludarabine, the OR rate was 52%, with an 8% CR rate, indicating good clinical activity in a group of patients with advanced disease. This combination was also tested by Nabhan et al in a small pilot study⁴¹ of 12 patients with recurrent and/or refractory CLL. In this study, 3 different doses of alemtuzumab were explored in combination with rituximab. This combination was found to be safe, with only 1 patient achieving a PR and all other patients having stable disease.

It is known that rituximab when used as a single agent has only minimal activity in patients with refractory CLL.^44–47 Recently, Castro et al⁴⁸ reported a small experience with the combination of rituximab and high-dose methylprednisolone (HDMP) in 14 patients with fludarabine-refractory CLL. The OR rate was 93%, and the CR rate was 36%. The median time to progression was 15 months in this study. Similar results with the same dose and schedule of HDMP-rituximab were reported by Bowen et al⁴⁹ in 37 patients with recurrent CLL and unfavorable cytogenetic features. The OR rate was 78%, the CR rate was 22%, and the PR rate was 57%; the median PFS was 1 year in this group of patients. Dungarwalla et al⁵⁰ also explored the HDMP-rituximab combination in 14 patients with advanced CLL that was resistant to fludarabine therapy. The OR rate was 93%, CRs were observed in 2 patients, and the median response duration was 7 months. This combination was associated with a high rate of severe infectious complications, such as invasive aspergillum and Candida infections likely related to the use of steroids in the experiences reported by Bowen et al and Dungarwalla et al,^49,50 whereas in the study by Castro et al,⁴⁸ only 1 patient had infection-related complications. This difference in the occurrence of infection complications between the studies may be because of the small number of patients studied.

Lumiliximab is a macaque-human primatized monoclonal antibody that targets CD23, an antigen expressed on activated B cells and CLL cells. In a phase 1 study, lumiliximab was found to decrease absolute lymphocyte counts in 91% of patients with refractory or recurrent CLL and reduced lymph node size in approximately 52%.⁵¹ Although none of the patients attained a CR or a PR, lumiliximab appeared to have clinical activity and to be well tolerated. Therefore, based on the activity of lumiliximab as a single agent, the combination of FCR plus lumiliximab has been explored. Byrd et al⁵² reported an OR rate of 71% and a CR rate of 52% in 31 evaluable patients. No significant toxicity was noted with the concomitant administration of the 2 antibodies beginning with Cycle 2.

Ofatumumab is a fully human anti-CD20 monoclonal antibody. Ofatumumab binds to a different epitope on the CD20 antigen and therefore has the potential to overcome some of the limitations of rituximab, such as low antigen density and shedding. Coiffier et al⁵³ published the results of a phase 1/2 study using ofatumumab in patients with recurrent or refractory CLL. Patients were divided into 3 cohorts receiving 4 weekly infusions of ofatumumab at 3 different doses. The OR rate in the higher dose cohort was 50%, with 12 PRs and 1 nPR reported. Recently, Osteborg et al⁵⁴ conducted an international, multicenter, pivotal study of ofatumumab in CLL patients with disease that was refractory to fludarabine and alemtuzumab (double refractory) or refractory to fludarabine with bulky lymphadenopathies (bulky fludarabine-refractory). One patient achieved CR; the OR rate was 51% for the double refractory patients and 44% for the bulky fludarabine-refractory patients. The median time to next therapy and the median OS for the 2 patient groups were 9 months and 8 months and 14 months and 15 months, respectively. In the initial experience, the toxicity of ofatumumab appears to be similar to that of rituximab, consisting mainly in infusion reactions. Ofatumumab is currently undergoing further evaluation in clinical trials both as a single agent and in combination with chemotherapy.

AGENTS WITH A DIFFERENT MECHANISM OF ACTION: BENDAMUSTINE, FLAVOPIRIDOL, AND LENALIDOMIDE

Bendamustine is a chemotherapeutic agent with both an alkylating and purine analogue–like structure that has been investigated in patients with CLL both as monotherapy and in combination with other agents (Table 3).^55,56 Bremer⁵⁷ reported that bendamustine is able to induce a high number of disease remissions in patients with CLL, including patients with prior treatment with alkylators. In 2005, the German CLL Study Group published the results of a phase 1–2 study of 16 patients with recurrent or refractory CLL.⁵⁸ The OR rate was 56%; 2 patients achieved a CR, 5 patients achieved a PR, and the median duration of response was 42.7 months. The combination of bendamustine, mitoxantrone, and rituximab was also tested in patients with recurrent and refractory lymphoproliferative disorders.⁵⁹ Twenty-one patients with CLL, 6 of whom (27%) had received prior treatment with fludarabine, were treated in this study. Patients with CLL received a lower dose of bendamustine (80 mg/m²) on Days 1 to 2; mitoxantrone at a dose of 10 mg/m² on Day 1; and rituximab at a dose of 375 mg/m² on Days 8, 15, 22, and 29. The administration of bendamustine was repeated on Days 36 through 37, and mitoxantrone was repeated on Day 36 and every 4 weeks thereafter up to a maximum of 6 cycles. The OR rate in patients with CLL was 95%, with 23% of patients achieving a CR and 73% of patients achieving a PR.

Table 3.

Agents With Different Mechanisms of Action in Patients With Recurrent/Refractory CLL

Reference	Treatment	No. of Patients	Refractory %	CR %	OR %	Refractory OR %
Koppler 200455	Bendamustine, 80–240 mg/m2 iv, d 1–3 + mitoxantrone 8–10 mg/m2 iv, d 1 for 6 cycles	22	NR	27	86	NR
Kath 200156	Bendamustine, 50–60 mg/m2 iv, d 1–5 for 8 cycles	23	NR	26	75	NR
Bremer 200257	Bendamustine, 60 mg/m2 iv, d 1–5, up to 11 cycles	15	NR	7	93	NR
Bergmann 200558	Bendamustine, 70–100 mg/m2 iv, d 1–2 for 6 cycles	16	25	12	56	0
Weide 200459	Bendamustine,* 80 mg/m2 iv, d 1–2 + mitoxantrone* 10 mg/m2 iv, d 1 + rituximab 375 mg/m2 iv, d 8, 15, 22, and 29	21	NR	23	95	NR
Phelps 200960	Flavopiridol, 30, 40, 50 mg/m2 iv bolus-CIVI weekly, for 4 wk for 6 cycles	52	83	0	40	40
Chanan-Khan 200665	Lenalidomide, 25 mg po, d 1–21, until disease progression, unacceptable toxicity, or CR	45	51	9	47	NR
Ferrajoli 200866	Lenalidomide, 10–25 mg po daily, until disease progression or unacceptable toxicity	44	27	7	32	25

CLL indicates chronic lymphocytic leukemia; CR, complete response; OR, overall response; iv, intravenously; NR, not reported; CM, 4-hour continuous iv infusion; po, orally.

^*Bendamustine/mitoxantrone therapy was repeated every 4 weeks from Day 36, for a maximum of 5 additional cycles.

Flavopiridol is a cyclin-dependent kinase inhibitor with preclinical activity in CLL. This agent also appears to down-regulate Mcl-1. Interestingly, flavopiridol-induced apoptosis is independent of p53 and caspase activation, and therefore this agent is potentially able to overcome some of the mechanisms associated with fludarabine resistance. Furthermore, flavopiridol has shown activity in patients with bulky lymph node disease. Recently, Phelps et al⁶⁰ reported the final results of a phase 1 study of flavopiridol in 52 patients with recurrent CLL and high-risk features. Partial responses were observed in 40% of the patients. Activity was observed both in patients with fludarabine-refractory CLL and in those with high-risk genomic aberrations. Treatment with flavopiridol has been accompanied by the development of acute tumor lysis and cytokine release syndrome; therefore the use of flavopiridol requires careful patient selection (excluding patients with a high lymphocytes count) and close monitoring during the initial treatment phase. The first treatment with flavopiridol should be administered in the inpatient setting, with close monitoring of electrolyte levels in a center with the capability to perform emergency dialysis if needed.

Lenalidomide is a derivative of thalidomide and belongs to a new class of drugs called immunomodulating agents. Lenalidomide has antitumor activity in multiple myeloma and myelodysplastic syndromes.^61–63 To our knowledge, the exact mechanisms of action of lenalidomide in CLL are not known, but are likely to be different from those of classic chemotherapeutic agents. This is a field of active investigation; available data suggest that immunomodulation may be important. Ramsay et al were able to demonstrate improved immunologic synapse formation after T cells and autologous CLL cells were exposed to lenalidomide in vitro.⁶⁴

Two phase 2 trials in patients with recurrent or refractory CLL suggest its clinical activity as a single agent. Chanan-Khan et al⁶⁵ administered lenalidomide orally at a dose of 25 mg on Days 1 through 21 of a 28-day cycle to 45 patients with recurrent CLL (51% of the patients were refractory to fludarabine). The OR rate was 47%, with 9% of the patients attaining a CR. Ferrajoli et al⁶⁶ reported the results of a phase 2 study of low-dose lenalidomide in patients with recurrent and refractory CLL. Lenalidomide was given orally at a dose of a 10 mg/day for 4 weeks, followed by dose escalation by 5 mg every 28 days to a maximum dose of 25 mg/day. Forty-four patients were enrolled; 27% of the patients had disease that was refractory to fludarabine, 88% had unmutated IgV_H genes, and 59% carried a poor prognostic genomic abnormality (deletion 11q or 17p). The OR rate was 32%, and 7% of patients achieved a CR. Several clinical studies with this agent are currently ongoing to define the optimal dosing schedule and to explore the use of lenalidomide in combination. The most common toxicity observed with this agent has been myelosuppression, in particular neutropenia. Tumor flare reaction, consisting of the rapid development of enlarged and tender lymph nodes, has also been observed and has been more common with the starting dose of 25 mg and in patients with bulky lymphoadenopathies.⁶⁷ Tumor lysis syndrome has also been observed; predisposing factors appear to be marked lymphocytosis and compromised renal function. Adequate hydration, the use of allopurinol, and close laboratory and clinical monitoring are required while patients are undergoing treatment with lenalidomide.

ALLOGENEIC STEM CELL TRANSPLANTATION

The development of reduced-intensity conditioning (RIC) regimens has improved the tolerability of allogeneic hematopoietic stem cell transplantation (HCT), and therefore this strategy is increasingly offered to older patients with refractory disease. Moreover, RIC allows preservation of the graft-versus-leukemia effect and further immunomanipulation.^68–70 Schetelig et al⁷¹ published the results of RIC HCT in 30 patients with advanced CLL. After a median follow-up of 2 years, 23 patients were alive, 40% of patients had achieved a CR, 53% of patients had achieved a PR, and acute and chronic graft-versus-host disease (GVHD) was observed in 56% and 75% of patients, respectively. Dreger et al⁷² retrospectively compared 73 cases of RIC HCTs with 82 conventional allogeneic transplants. RIC HCTs were associated with a reduced treatment-related mortality compared with conventional allogeneic transplants; however, RIC transplants were also associated with a higher incidence of disease recurrence. Sorror et al⁷³ reported on 82 patients with advanced CLL who were treated with nonmyeloablative conditioning (2-gray total body irradiation with or without fludarabine) and HCT from related or unrelated donors. The OR rate was 70% (55% CR rate). The 5-year incidences of nonrecurrence mortality, disease progression/recurrence, OS, and PFS were 23%, 38%, 50%, and 39%, respectively. The incidence of grades 2, 3, and 4 acute and chronic GVHD was 39%, 14%, and 2%, respectively, among related recipients and was 43%, 20%, and 3%, respectively, among unrelated recipients. The 5-year cumulative incidence of chronic extensive GVHD was 49% for related and 53% for unrelated recipients. Unrelated HCT resulted in higher CR and lower recurrence rates than related HCT, suggesting more effective graft-versus-leukemia activity. Khouri et al⁷⁴ reported on the importance of adding rituximab to RIC HCT and on the role of immunomanipulation in this setting; 9 of 17 patients had disease that was refractory to fludarabine and underwent RIC HCT. The OR rate was 94%. The addition of rituximab was accompanied by a survival advantage (100% vs 14%) with respect to the small number of patients who received RIC HCT without rituximab. In 7 of 10 patients with persistent disease after HCT, a CR was successfully induced with immunomanipulation that included withdrawal of immunosuppressive agents, donor lymphocyte infusion, and treatment with rituximab. Even if encouraging, these results need to be confirmed in larger studies. Recently, the European Group for Blood and Marrow Transplantation transplant consensus published the indications for HCT in chronic lymphocytic leukemia.⁷⁵ Allogeneic HCT is a reasonable treatment option for eligible patients with previously treated, poor-risk CLL. Criteria for poor-risk disease are lack of response or disease recurrence within 12 months after purine analogue-containing therapy, recurrence within 24 months after purine analogue combination therapy or treatment of similar efficacy (ie, autologous HCT), and patients with recurrent/refractory disease with evidence of p53 deletion or mutation.

Conclusions

Multiple treatment options have emerged for patients with recurrent and refractory CLL. Current data from consolidation strategies (such as alemtuzumab) reported by several groups,^6–8 as well as the correlation between quality of response and response duration after allogeneic HCT,⁷⁶ have suggested that achieving a negative MRD status is associated with longer PFS and OS. Thus, strategies that are able to eliminate MRD are also likely to improve the survival outcome of patients with refractory disease. Allogeneic HCT, which has a therapeutic basis in the immune response against leukemia, should be considered as a preferred treatment option for younger patients with poor-risk disease and in patients with primary treatment failure. Developing a better understanding of the biology of the disease, including the microenvironment interactions, might enable researchers to identify effective and targeted treatments and potentially curative strategies for patients with refractory CLL.