Several decades of large, randomized, prospective trials of previously untreated patients have demonstrated statistically significant improvements in response rates, event-free survival (EFS), and progression-free survival (PFS) with comparison of combinations of drugs versus single-agent alkylators,[1,2] but only two trials have shown statistically significant improvement in overall survival (OS).[3,4]
The first trial, a comparison of chlorambucil versus fludarabine, after 15 years of median follow-up, showed improved median OS for patients on the fludarabine regimen at 63 months versus 59 months (P = .04), and an improved percentage of patients were alive at 8 years (31% vs. 19%, P = .04).[3,5][Level of evidence: 1iiA]
The second trial, which had 817 patients, compared FCR (fludarabine + cyclophosphamide + rituximab) versus FC (fludarabine + cyclophosphamide) with a median follow-up of 38 months and showed improved OS at 3 years for the rituximab combination (i.e., 87% vs. 83%, P = .01.[Level of evidence: 1iiA] Yet neither fludarabine nor FCR has been compared in a randomized study against watchful waiting in asymptomatic or minimally affected patients.
The improvements in response rates from more intensive regimens have maximized the clearance of minimal residual disease (MRD). In one prospective trial of 493 patients, clearance of MRD was an independent predictor of OS by multivariate analysis. The surrogate endpoint of such clearance of residual disease, while prognostic, has not been shown to improve survival in a randomized prospective trial; the necessary study would take patients who fail to completely clear the marrow with induction therapy and randomly assign them to further alternative treatment versus the same treatment later at relapse, looking at OS as the primary endpoint.[1,2]
The sequencing of the following treatment options cannot be determined from the current set of completed clinical trials. When patients become symptomatic or require treatment, FCR has become the most frequently chosen option outside of a clinical trial, mostly on the basis of the previously described prospective study.
Note: Standard options are roughly ordered by level of toxic effects, starting with the least toxic options. More recently discovered options are mentioned at the end of the list.
- Observation in asymptomatic or minimally affected patients. Outside of the context of a clinical trial, treatment for asymptomatic or minimally affected patients with chronic lymphocytic leukemia (CLL) is observation. No data exist as yet to suggest any harm with a delay in therapy until the patient becomes symptomatic or develops serious cytopenias despite growth factor support. Because the rate of progression may vary from patient to patient, with long periods of stability and sometimes spontaneous regressions, frequent and careful observation is required to monitor the clinical course.
- Rituximab, a murine anti-CD20 monoclonal antibody.[8,9,10,11,12] When used alone, higher doses of rituximab or increased frequency or duration of therapy is required for comparable responses to those seen for other indolent lymphomas.
- Ofatumomab is a human anti-CD20 monoclonal antibody. A trial of 138 patients, who were previously treated with fludarabine and alemtuzumab, showed overall response rates around 50% in patients refractory to fludarabine and with prior exposure to rituximab.[13,14][Level of evidence: 3iiiDiv]
- Oral alkylating agents with or without corticosteroids. The French Cooperative Group on CLL randomly assigned 1,535 patients with previously untreated stage A disease to receive either chlorambucil or no immediate treatment and found no survival advantage for chlorambucil.[Level of evidence: 1iiA] A meta-analysis of six trials of immediate versus deferred therapy with chlorambucil (including the aforementioned trial by the French Cooperative Group) showed no difference in OS at 10 years.[Level of evidence: 1iiA]
- Fludarabine, 2-chlorodeoxyadenosine, or pentostatin as seen in the CLB-9011 trial, for example.[17,18,19,20,21,22]
Several randomized trials have compared the purine analogs with chlorambucil; with cyclophosphamide, doxorubicin, and prednisone; or with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) in previously untreated patients.[5,23,24,25,26] All of these trials showed higher or equivalent response rates for the purine analog and most showed an improvement in PFS, with one reaching significance in OS favoring fludarabine.[3,23,24,25,26,27][Level of evidence: 1iiDiii]
A comparison of chlorambucil versus fludarabine, after 15 years' median follow-up, showed patients with improved median OS with fludarabine at 63 versus 59 months (P = .04) and an improved percentage of patients alive at 8 years (31% vs. 19%, P = .04).[Level of evidence: 1iiA] All of the trials demonstrated higher toxic effects with the purine analogs, especially granulocytopenic infections, herpes infections, autoimmune hemolytic anemia, and persistent thrombocytopenia. The increased risk of infection may persist for months or years after treatment with a purine analog.[27,29]
Although empiric evidence is lacking, some investigators recommend prophylaxis with trimethoprim-sulfa during therapy and for 6 to 12 months afterwards to prevent pneumocystis infection. In a similar way, other investigators employ prophylaxis (e.g., acyclovir) for the herpes viruses. Purine analogs cause less hair loss or nausea than combination chemotherapy, including alkylators and anthracyclines.
Bendamustine is a cytotoxic agent with bifunctional properties of an alkylator and a purine analog. In previously treated and untreated patients, bendamustine with rituximab has shown response rates around 70% to 90%.[31,32][Level of evidence: 3iiiDiii]
In a randomized comparison with chlorambucil in 319 previously treated patients, bendamustine showed a better response rate (68% vs. 31%, P < .0001) and PFS (21.6 months vs. 8 months) with a median follow-up of 35 months.[Level of evidence: 1iiDiii] The German CLL Study Group is comparing bendamustine plus rituximab versus FCR as first-line therapy in patients with CLL who require therapy.
Lenalidomide is an oral immunomodulatory agent with response rates over 50%, with or without rituximab, for patients with previously treated and untreated disease.[34,35,36,37,38,39][Level of evidence: 3iiiDiv] Prolonged, lower-dose approaches and attention to prevention of tumor lysis syndrome are recommended with this agent.[34,40] Combination therapy and long-term toxicities from using lenalidomide (such as increased myelodysplasia, as seen in myeloma patients) remain undefined for patients with CLL.
- Combination chemotherapy.
A trial of 817 patients comparing FCR versus FC with a median follow-up of 38 months showed improved OS at 3 years for the rituximab combination (87% vs. 83%, P = .01).[Level of evidence: 1iiA] FCR has never been compared with watchful waiting up front in asymptomatic or minimally affected patients. The improvements in response rates from more intensive regimens have maximized the clearance of MRD. However, the surrogate endpoint of MRD clearance has not been proven to be a valid surrogate for improved survival in a randomized prospective trial; the necessary study would take patients who fail to completely clear the marrow with induction therapy and randomly assign them to further alternative treatment versus the same treatment later at relapse looking at OS as the primary endpoint.[1,2] A cumulative incidence of 6% to 8 % for myelodysplasia is seen at 5 to 7 years in patients who received fludarabine plus cyclophosphamide, with or without rituximab.[41,42]
Other combination chemotherapy regimens include the following:
- Fludarabine plus cyclophosphamide plus rituximab.[4,43,44,45,46]
- Fludarabine plus rituximab as seen in the CLB-9712 and CLB-9011 trials.
- Fludarabine plus cyclophosphamide versus fludarabine plus cyclophosphamide plus rituximab.[4,48]
- Pentostatin plus cyclophosphamide plus rituximab as seen in the MAYO-MC0183 trial, for example.[49,50]
- Ofatumumab plus fludarabine plus cyclophosphamide.
- CVP: cyclophosphamide plus vincristine plus prednisone.
- CHOP: cyclophosphamide plus doxorubicin plus vincristine plus prednisone.
- Fludarabine plus cyclophosphamide versus fludarabine as seen in the E2997 trial [NCT00003764] and the LRF-CLL4 trial, for example.[54,55]
- Fludarabine plus chlorambucil as seen in the CLB-9011 trial, for example.
A meta-analysis of ten trials comparing combination chemotherapy (before the availability of rituximab) with chlorambucil alone showed no difference in OS at 5 years.[Level of evidence: 1iiA]
- Involved-field radiation therapy. Relatively low doses of radiation therapy will affect an excellent response for months or years. Sometimes radiation therapy to one nodal area or the spleen will result in abscopal effect (i.e., the shrinkage of lymph node tumors in untreated sites).
- Alemtuzumab, the monoclonal antibody directed at CD52, shows activity in the setting of chemotherapy-resistant disease or high-risk untreated patients with 17p deletion or p53 mutation.[57,58,59] As a single agent, the subcutaneous route of delivery for alemtuzumab is preferred to the intravenous route in patients because of the similar efficacy and decreased adverse effects, including less acute allergic reactions that were shown in some nonrandomized reports.[59,60,61,62,63]
In a combination regimen, subcutaneous alemtuzumab plus fludarabine (with or without cyclophosphamide) or intravenous alemtuzumab plus alkylating agents have resulted in excess infectious toxicities and death, with no compensatory improvement in efficacy in phase II trials.[64,65,66][Level of evidence: 3iiiDiv]
In a randomized prospective study, 335 previously treated patients received intravenous alemtuzumab plus fludarabine versus fludarabine alone; with a median follow-up of 30 months, the combination of fludarabine plus intravenous alemtuzumab had better PFS, with a median of 23.7 months versus 16.5 months (hazard ratio [HR], 0.61; 95% confidence interval [CI], 0.47–0.80; P = .0003); and better OS, with a median not reached, versus 52.9 months (HR, 0.65; 95% CI, 0.45–0.94; P = .021).[Level of evidence: 1iiA] Profound and long-lasting immunosuppression has been seen, which mandates monitoring for reactivation of cytomegalovirus and prophylaxis for pneumocystis and herpes virus infections.[68,69] Antibiotic prophylaxis includes trimethoprim and sulfamethoxazole, itraconazole, and acyclovir (or ganciclovir) for asymptomatic cytomegalovirus viremia.
- Bone marrow and peripheral stem cell transplantations are under clinical evaluation.[70,71,72,73,74,75,76]
In a prospective randomized trial, 241 previously untreated patients younger than 66 years with advanced-stage disease received induction therapy with a CHOP-based regimen followed by fludarabine. Complete responders (105 patients) were randomly assigned to undergo autologous stem cell transplantation (ASCT) or observation, while the other 136 patients were randomly assigned to receive dexamethasone, high-dose aracytin, and cisplatin reinduction followed by either ASCT or fludarabine plus cyclophosphamide. Although the 3-year EFS favored ASCT in complete responders, there was no difference in OS in any of the randomized comparisons.[Level of evidence: 1iiDi]
Patients with adverse prognostic factors are very likely to die from CLL. These patients are candidates for clinical trials that employ high-dose chemotherapy and immunotherapy with myeloablative or nonmyeloablative allogeneic peripheral stem cell transplantation.[70,71,72,73,74,75,78,79,80,81,82,83,84,85] Although most patients who attain complete remission after ASCT eventually relapse, a survival plateau for allogeneic stem cell support suggests an additional graft-versus-leukemia effect.
- Ibrutinib is a selective irreversible inhibitor of Bruton tyrosine kinase, a signaling molecule located upstream in the B-cell receptor signaling cascade. Other trials of relapsed and refractory CLL have shown durable responses to the oral agent in phase I and II studies.[Level of evidence: 3iiiDiii] A phase Ib–II trial (NCT01105247) of 85 patients with relapsed or refractory CLL showed a 26-month PFS rate of 76%, including patients with 17p- or unmutated IgVH fluorescence in situ hybridization testing.[Level of evidence: 3iiiDiii]
- Autologous T-cells were modified by a lentiviral vector to incorporate antigen receptor specificity for the B-cell antigen CD19 and then infused into a previously treated patient. A dramatic response lasting 6 months has prompted larger trials of this concept.[Level of evidence: 3iiiDiv] Ongoing clinical trials are testing this concept.