Chlorambucil for the treatment of patients with chronic lymphocytic leukaemia, or small lymphocytic lymphoma
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To assess the effect of chlorambucil compared to other therapeutic regimens on overall survival, progression free survival, and response rates in patients with CLL/SLL requiring therapy. We will include patients receiving chlorambucil as first‐line therapy and patients with relapsed or refractory CLL/SLL receiving salvage therapy with chlorambucil.
Background
Description of the condition
The World Health Organization (WHO) classification, published in 2001 and updated in 2008, attempts to group tumours of lymphoid tissue by cell type using phenotypic, molecular and cytogenetic characteristics (Swerdlow 2008). Chronic lymphoid leukaemia (CLL) or small lymphocytic lymphoma (SLL) is one of the mature B cell neoplasms. CLL is the leukaemic phase of SLL (Rai 1975; Swerdlow 2008; Tsimberidou 2007). Nowadays we know that CLL and SLL are the same disease, with the same cell of origin, the same response to therapy and the same prognosis, and is referred to as CLL/SLL. For that reason throughout the text we use the terms CLL and CLL/SLL interchangeably.
CLL/SLL is a chronic lymphoproliferative B cell neoplasm characterized by a progressive accumulation of monoclonal lymphocytes (Rai 1975; Swerdlow 2008; Tsimberidou 2007). CLL is one of the most common types of adult leukaemia with a lifetime risk of 0.47% (Horner 2009). The incidence of CLL may be higher than that estimated by cancer registry data (Seftel 2009; Zent 2001). In the European Union the 1‐, 5‐ and 10‐year prevalence estimates are 0.2/10,000, 0.9/10,000 and 2/10,000 individuals, respectively (Watson 2008).
More than 70% of diagnosed patients are 65 years of age or older, and the median age for diagnosis is 72 years (Horner 2009). The diagnosis of CLL/SLL is based on a complete blood count which shows absolute lymphocytosis that equals or is above 5 × 109/L; examination of the peripheral smear where leukaemia cells that are typically small, mature lymphocytes with a narrow border of cytoplasm and a dense nucleus are seen; and flow cytometry of the peripheral blood with confirmation of clonality. The diagnosis of SLL should be confirmed by histopathologic evaluation of a lymph node biopsy in a patient with lymphadenopathy or splenomegaly, or both, but without peripheral lymphocytosis (Hallek 2008).
The natural history of CLL/SLL is extremely variable, with survival from the initial diagnosis ranging from two to 20 years. The first staging system for CLL was developed by Rai and colleagues (Rai 1975). It is based on the concept that in CLL there is a gradual and progressive increase in the burden of leukaemic lymphocytes. While median survival of patients with CLL Rai stage 0 (lymphocytosis alone) was 150 months, that of patients with Rai stage III or IV (anaemia, thrombocytopenia) was 19 months. The Rai as well as the Binet staging system (Binet 1981) (which uses the number of involved lymphoid sites, anaemia or thrombocytopenia to define disease stage) are both simple yet accurate predictors of survival and are widely used by clinicians and researchers. Additional prognostic factors were suggested including genetic abnormalities and immunoglobulin (Ig) mutation status (Hamblin 1999; Kienle 2010).
Description of the intervention
Chlorambucil, a derivative of nitrogen mustard, is an alkylating agent that can be given orally as it is rapidly absorbed from the gastrointestinal tract. Metabolism of chlorambucil is primarily hepatic and excretion of its metabolites is through renal clearance. Monotherapy with chlorambucil has served as the initial, front‐line therapy for CLL/SLL for several decades. Two commonly used protocols for the chlorambucil dosage and schedule are low‐dose continuous treatment (0.08 mg/kg/day) or intermittent dose (0.8 mg/kg once every 2 to 4 weeks) (Eichhorst 2009; Gribben 2010). Its most commonly occurring adverse effects are cytopenias. Chlorambucil has also been associated with myelodysplasia and secondary leukaemia.
How the intervention might work
No benefit was demonstrated with treatment at an early stage of CLL/SLL (CLL Trialists 1999; French Cooperative Group 1990). Accepted indications for treatment are symptomatic disease, bulky progressive adenopathy, autoimmune haemolytic anaemia or thrombocytopenia poorly responsive to corticosteroid therapy, or marrow failure (anaemia or thrombocytopenia, or both). There has been a considerable debate regarding the best therapy for symptomatic patients with CLL/SLL (Gribben 2010; Hallek 2008).
Many trials have compared chlorambucil to different regimens, including other alkylating agents, purine analogues, monoclonal antibodies and novel agents, for the treatment of patients with CLL/SLL (Hillmen 2007; Jaksic 1997; Morrison 2001; Robak 2000). So far, none of these regimens have improved overall survival compared to chlorambucil (CLL Trialists 1999). Though treatment with purine analogues improved response rates and progression free survival it did not improve overall survival and was associated with significant adverse effects (Eichhorst 2009; Steurer 2006). Also early inclusion of an anthracycline did not improve survival rates (Jaksic 1997).
A previous meta‐analysis that compared chlorambucil to combination chemotherapy and included 10 randomised controlled trials (2022 patients) was published in 1999 (CLL Trialists 1999). No difference in mortality was found. No update of this meta‐analysis has been published although clinical trials with newer agents regarding this issue have been conducted since.
Why it is important to do this review
In practice today, young and fit patients are usually treated with fludarabine‐based regimens while old or frail patients are treated with other first‐line treatments including alkylating agents, mainly chlorambucil. In the United States chlorambucil is less often chosen for the treatment of patients with CLL/SLL while in European Union it is still widely used (Gribben 2010; NCCN guidelines).
The role of chlorambucil in the management of young and old patients with CLL/SLL has to be re‐evaluated.
Objectives
To assess the effect of chlorambucil compared to other therapeutic regimens on overall survival, progression free survival, and response rates in patients with CLL/SLL requiring therapy. We will include patients receiving chlorambucil as first‐line therapy and patients with relapsed or refractory CLL/SLL receiving salvage therapy with chlorambucil.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials.
Types of participants
Patients with CLL/SLL with either a lymphocyte count above or equal to 5 × 109/L and proven by peripheral blood smear and, where available, by flow cytometry (positive CD20, CD19, CD5) or histopathology compatible with CLL/SLL.
We will include trials using any criteria system that defines CLL/SLL as long as it follows the above minimal criteria. Criteria systems comprise the following: the National Cancer Institute sponsored Working Group (NCI‐WG) on chronic lymphocytic leukaemia (CLL) published in 1988 and updated in 1996; the International Workshop on Chronic Lymphocytic Leukemia (IWCLL) published in 1989 and updated in 2007; and the WHO classification of lymphomas (Hallek 2008; Swerdlow 2008).
We will include patients with CLL/SLL in the following situations (considered as indications for chemotherapy):
Binet stage A (or Rai stage 0‐II) progressive disease or symptomatic (weight loss, night sweats, fever, asthenia);
Binet stage B (or Rai stage III);
Binet stage C (or Rai stage IV).
We will include both patients receiving chlorambucil as first‐line therapy and patients with relapsed or refractory CLL/SLL receiving salvage therapy which includes chlorambucil.
Types of interventions
Investigational intervention
-
Chlorambucil with or without steroids. Chlorambucil with steroids will be considered as chlorambucil alone for the purpose of this review
-
Chlorambucil as part of combination chemotherapy (this may include other types of alkylating agents, purine analogues, anthracyclines, vinca alkaloids and other chemotherapeutic agents)
-
Chlorambucil in combination with monoclonal antibodies (i.e., rituximab, alemtuzumab)
-
Chlorambucil as part of combination chemotherapy combined with monoclonal antibodies
Comparison interventions (not including chlorambucil)
-
Combination chemotherapy or corticosteroids or both, excluding chlorambucil
-
Monoclonal antibodies
-
Combination chemotherapy combined with monoclonal antibodies
If chlorambucil (with or without steroids) is given in combination with other chemotherapeutic drugs, monoclonal antibodies or high dose chemotherapy, then the treatment protocol in both arms should be the same except for chlorambucil (or chlorambucil and steroids).
For example:
-
chlorambucil versus fludarabine ‐ chlorambucil is compared to fludarabine;
-
chlorambucil versus rituximab;
-
rituximab and chlorambucil versus rituximab and fludarabine‐cyclophosphamide ‐ chlorambucil is compared to fludarabine‐cyclophosphamide;
-
chlorambucil versus cyclophosphamide, vincristine, prednisone ‐ cyclophosphamide, vincristine, prednisone protocol is compared to chlorambucil;
-
chlorambucil and prednisone versus cyclophosphamide, vincristine, prednisone versus chlorambucil, prednisone ‐ chlorambucil is compared to cyclophosphamide, vincristine.
Trials comparing chlorambucil (without high dose therapy (HDT) and stem cell transplantation (SCT)) to HDT and SCT will be excluded.
Types of outcome measures
Primary outcomes
Overall survival
Secondary outcomes
-
Complete response
-
Partial response
-
Progression free survival (PFS) defined as the time from study entry until objective disease progression or death. We will use definitions of progression as defined in each trial. If PFS is not available we will use event free survival (defined as the time from study entry till progression, adverse event, or death). If both PFS and event free survival are not available we will use time to next treatment (time from end of treatment until next chemotherapy or radiotherapy)
-
Quality of life
-
Adverse events requiring discontinuation of therapy
-
Grade 3 or 4 adverse events
-
Infection‐related adverse events
-
Secondary malignancies
Search methods for identification of studies
Electronic searches
We will search the following databases:
-
Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, latest issue),
-
MEDLINE (1966 to present) (through PubMed, see Appendix 1),
-
LILACS (1982 to present ),
-
clinical trials in haematological malignancies (www.hematology‐studies.org).
We will cross the term 'chronic lymphocytic leuk?emia' and similar with the term 'chlorambucil' and similar.
The search terms will be combined with the highly sensitive search strategy for identifying reports of randomised controlled trials (Robinson 2002) in the MEDLINE search.
We will search the conference proceedings of the American Society of Hematology (1995 to 2010), conference proceedings of the American Society of Clinical Oncology Annual Meeting (1995 to 2010), and proceedings of the European Hematology Association for relevant abstracts.
We will search databases of ongoing and unpublished trials: www.controlled‐trials.com/, www.clinicaltrials.gov/ct, http://clinicaltrials.nci.nih.gov/.
Searching other resources
We will contact the first or corresponding author of each included study and the researchers active in the field for information regarding unpublished trials or complementary information on their own trial.
We will check the citations of included trials and major reviews for additional studies.
Data collection and analysis
Selection of studies
Two review authors (LV, RG) will inspect the title and, when available, the abstract of each reference identified in the search and apply the inclusion criteria. Where relevant articles are identified, the full article will be obtained and inspected independently by two review authors.
We will include trials regardless of publication status, date of publication, and language.
Data extraction and management
Two review authors (LV, RG) will independently extract the data of included trials. Any disagreement between the two review authors will be discussed. If the disagreement remains unsolved a third review author (AG) will independently extract the data. The data extraction will be discussed, disagreement and resolution documented and, where necessary, the authors of the studies will be contacted for clarification. If this is unsuccessful, disagreements will be reported.
The justification for excluding studies from the review will be documented. All data will be collected on an intention‐to‐treat basis, where possible.
The following data will be extracted, checked, and recorded.
1. Characteristics of trials
-
Publication status: published; published as abstract; unpublished
-
Year (defined as recruitment initiation year) and country or countries of study
-
Trial sponsor (academic, industry)
-
Intention‐to‐treat analysis: performed; possible to extract; efficacy analysis
-
Design (method of allocation generation and concealment; blinding)
-
Unit of allocation (patient, episodes, cluster)
-
Duration of study follow up
-
Response definition, event definitions
-
Case definitions used (inclusion and exclusion criteria)
-
Assessment of mortality (primary outcome, secondary outcome, safety)
2. Characteristics of patients
-
Number of participants in each group
-
Age (mean and standard deviation)
-
Disease status (untreated; previously treated)
-
Number of previous courses
-
Number of patients with performance status ≤ 2, > 2
-
Number of patients with Rai stage II, III, IV or Binet stage B, C
-
Number of patients with bulky disease
-
Number of patients with elevated lactate dehydrogenase (LDH)
-
Number of patients with haemolytic anaemia
-
Number of patients with unmutated, and number of patients with mutated immunoglobulin Vh genes
-
Number of patients with deletion 17p or p53 abnormalities
3. Characteristics of interventions
-
Experimental intervention
-
-
Dose, number of administered doses, and total duration of therapy
-
Additional drugs (combination)
-
-
Treatment of control group
-
Regimen, dose, number administered doses, and total duration of therapy
-
4. Characteristics of outcome measures (extracted for each group, and total events)
-
Overall survival (OS)
-
Number of deaths at 12, 36, 60 months, end of follow up
-
Number of patients available for follow up at the time of evaluation of survival risk
-
Hazard ratio (HR) of OS and its standard error (SE), confidence interval (CI) or P value
-
Kaplan‐Meier curve (yes or no)
-
-
HR of progression free survival (PFS) and its SE, CI or P value
-
Number of patients achieving complete response
-
Number of patients achieving partial response
-
Change and end of treatment quality of life assessment, and the tool used
-
Adverse events (any, grade 3 or 4, requiring discontinuation of treatment, infection related, grade 3 or 4 neutropenia, secondary malignancies)
-
Number of patients excluded from outcome assessment after randomisation, and the reasons for their exclusion
Assessment of risk of bias in included studies
Two review authors (LV, RG) will independently assess the trials for methodological quality. Allocation concealment, sequence generation, blinding, incomplete outcome data, and selective outcome reporting will be individually described and assessed according to The Cochrane Collaboration's tool for assessing bias (Higgins 2008).
Measures of treatment effect
We will estimate relative risks (RR) for dichotomous data, and hazard ratios (HR) for time to event outcomes. We will estimate standardised mean difference (SMD) for quality of life assessment.
Unit of analysis issues
Cross‐over trials
We do not expect trials with a cross‐over design as the effect of the chemotherapy in the first period is assumed to continue through the second period.
Multiple observations at different time points for the same outcome
(Cochrane Handbook for Systematic Reviews of Interventions; Chapter 9)
For time to event meta‐analysis we will use data at the longest follow up.
For dichotomous data we will analyse outcomes data at 12 and 60 months separately. Response rates will be analysed at the end of chemotherapy.
Events that may re‐occur
(Cochrane Handbook for Systematic Reviews of Interventions; Chapter 9)
Adverse events may occur more than once in the same individual, mainly during different treatment cycles. We will extract the number of patients in each arm and the number of patients who experienced at least one event. We will count each patient once even if repeated events occur and will analyse the data as dichotomous data.
Dealing with missing data
We will impute missing data for patients who were lost to follow up after randomisation (dichotomous data) assuming a poor outcome (worst case scenario) for missing individuals (Higgins 2008a).
We will perform a sensitivity analysis of the primary outcome without trials that have > 20% losses to follow up.
Assessment of heterogeneity
Heterogeneity (degree of difference between the results of different trials) in the results of the trials will be graphically inspected and assessed by applying a test of heterogeneity. We will assess the heterogeneity of the trial results by the Chi2 test of heterogeneity and the I2 statistic for inconsistency (Deeks 2008; Higgins 2003). We will define statistically significant heterogeneity as P < 0.1 or an I2 statistic greater than 50%.
Assessment of reporting biases
If at least 10 studies are included in the meta‐analysis of overall survival, we will examine a funnel plot of the treatment effect against the precision of trials (plot of the log of the relative risk for efficacy against the standard error) in order to estimate potential asymmetry that may indicate selection bias (the selective publication of trials with positive findings) or methodological flaws in the small studies (Sterne 2008).
Data synthesis
We will pool log HR for time to event outcomes using an inverse variance method (Review Manager (RevMan) version 5 for Windows; The Cochrane Collaboration, Oxford, United Kingdom). If not enough data are available, we will estimate HRs indirectly using methods described by Parmar 1998. An HR < 1.0 is in favour of chlorambucil treatment. We will estimate relative risks (RR) and their CIs for dichotomous data using the Mantel‐Haenszel method. We will use a fixed‐effect model. We will repeat the primary analysis using a random‐effects model (DerSimonian and Laird method) (Der Simonian 1986) in a sensitivity analysis. We will estimate the standardised mean difference (SMD) for ordinal data (quality of life scales) using the inverse of variance method.
Subgroup analysis and investigation of heterogeneity
Potential sources of heterogeneity will be explored through stratifying the patient subgroups given below, allocation concealment, blinding, and size of studies.
We will perform subgroup analysis by:
-
disease stage;
-
treatment line (first‐line, treatment of previously treated patients);
-
age of patients (up to 65 years, above 65 years; trials that included patients above 65 years, trials that included patients up to 65 years);
-
type of comparator protocol (other alkylating agent, purine analogue, immuno‐chemotherapy etc.).
Differences between subgroups will be formally assessed using the Chi2 test for difference between subgroups (Deeks 2001).
Sensitivity analysis
We will perform sensitivity analyses using the method of allocation concealment (Schulz 1995), blinding (patients, caregivers, and assessors), allocation generation (Higgins 2008), reporting of all cause mortality (primary analysis, secondary outcomes, safety), and the size of trials.
