33 - Chronic lymphocytic leukaemia

Chapter 33

In chronic lymphocytic leukaemia:

  • FLCs are the most commonly detected monoclonal proteins, with an abnormal sFLC ratio present in 30 - 40% of patients at diagnosis.
  • Elevated polyclonal FLCs are present in a further 15% of patients at diagnosis.
  • A monoclonal or polyclonal FLC elevation is associated with a shorter time to first treatment and reduced overall survival.
  • Summated κ + λ FLCs is an important new prognostic marker that identifies patients requiring early treatment.

33.1. Introduction

With an incidence of 4.2 per 100,000 per year, chronic lymphocytic leukaemia (CLL) is the most common type of leukaemia in the Western world [726] The median age at presentation is 72 years and the incidence is higher in men than women [726]. CLL is a clinically heterogeneous disease. Approximately two-thirds of patients are asymptomatic at diagnosis, and CLL is frequently diagnosed incidentally, following a routine full blood count. Other patients present with lymphadenopathy, systemic symptoms (such as tiredness, night sweats and weight loss), or infection. The clinical outcome of CLL is variable. Some patients survive for decades without requiring treatment, whilst others experience an aggressive form of the disease and may die shortly after diagnosis, either of disease- or therapy-related complications [726].

The diagnosis of CLL is based on the presence of an absolute monoclonal B-cell count of >5 x 109/L in the peripheral blood (persisting for >3 months) with a characteristic immunophenotype and morphology [727]. Two related disorders—monoclonal B-cell lymphocytosis (MBL) and small lymphocytic lymphoma (SLL) —share many clinical and diagnostic features in common with CLL. A diagnosis of MBL requires a monoclonal B-cell count of <5 x 109/L, in the absence of lymphadenopathy or organomegaly, cytopenias or disease-related symptoms [727]. SLL is also characterised by a B-cell count of <5 x 109/L, with the addition of lymph node or other tissue infiltration by cells characteristic of CLL [727].

Two well-established CLL clinical staging systems (Rai and Binet) are in routine use. These are particularly useful for predicting outcome in patients presenting with lymphadenopathy, hepatosplenomegaly or bone marrow failure [726]. However, significant clinical heterogeneity exists within patients classified as early CLL (Binet stage A or Rai stage 0/1). Therefore, additional prognostic markers are required to identify patients at risk of clinical progression. A number of prognostic biomarkers have been shown to predict progression and survival in CLL. These include immunoglobulin heavy chain variable region gene (IGHV) mutation status, serum β2-microglobulin, CD38/ZAP-70 expression, and cytogenetic abnormalities [728]. However, these techniques have several limitations. For example, assessment of IGHV mutation status is complex, expensive and is not widely available. Other techniques (e.g. ZAP-70 expression) are poorly standardised and suffer from significant inter-laboratory variation [729].

33.2. Monoclonal and polyclonal sFLCs in CLL

CLL is thought to originate from the expansion of an antigen-activated B-cell clone. Persistent immune stimulation and polyclonal B-cell activation/dysfunction may play an important role in its pathogenesis [730]. This is supported by research demonstrating that increased concentrations of both monoclonal and polyclonal free light chains (FLCs) exist up to 10 years prior to CLL diagnosis [731].

FLCs are the most commonly detected serum monoclonal proteins in CLL, with an abnormal sFLC ratio being reported in 30 - 40% of patients [675][671][732][673]. By comparison, Maurer et al. [673] reported 16% of patients with a monoclonal protein detected by serum protein electrophoresis. In several published plasma cell disease screening studies, addition of sFLC analysis to the testing panel has identified additional CLL patients (Chapter 23) [501][188][129]. Therefore, identification of abnormal sFLC ratios in screening samples should prompt the consideration of other lymphoproliferative disorders, particularly CLL, in addition to plasma cell dyscrasias.

Maurer et al. [673] characterised both monoclonal and polyclonal FLC abnormalities in 339 newly diagnosed, untreated CLL patients. They defined three different types of FLC abnormality: monoclonal FLC elevation, polyclonal FLC elevation and ratio-only abnormality (Table 33.1).

Abnormality κ and λ concentration κ/λ sFLC ratio Number of patients
Monoclonal sFLC elevation
Elevated κ and/or λ
57 (17%)
Polyclonal sFLC elevation
Elevated κ and/or λ
52 (15%)
Ratio-only abnormality
Normal κ and λ
54 (16%)
Any abnormality
163 (48%)

Table 33.1. sFLC abnormalities in newly diagnosed CLL patients [673].

In patients with an abnormal sFLC ratio (ratio-only abnormality or a monoclonal FLC elevation), the involved sFLC type matched the CLL B-cell light chain restriction in 92% and 96% of cases, respectively [673]. Such findings suggest that an abnormal sFLC ratio is disease-related in the majority of cases. Morabito et al. [675] studied the distribution of monoclonal and polyclonal FLC synthesis in the CLL tumour microenvironment by immunohistochemistry. Lymph node infiltrates comprised a prominent population of plasmacytoid lymphocytes expressing the involved FLC, along with a smaller number of lymphocytes expressing the corresponding uninvolved FLC. These lymphocytes populated the same infiltrates in the lymph node and bone marrow, and were associated with scattered FLC-producing plasma cells (Figure 33.1).

Polyclonal B-cell activation found in CLL also underlies the pathogenesis of some inflammatory/autoimmune conditions (Chapter 35). Interestingly, both CLL and inflammatory/autoimmune conditions may be associated with lymphomatous transformation. This occurs in approximately 5 - 15% of CLL patients, and histologically resembles diffuse large B-cell lymphoma or Hodgkin lymphoma (Chapter 31) [726].

33.3. Prognostic value of sFLCs at baseline

The concentrations of sFLCs observed in CLL patients are typically much lower than those found in multiple myeloma (Figure 33.2A), and are similar to patients with B-cell lymphomas (Chapter 31).

In a retrospective study of 259 CLL patients comprising 181 untreated/pre-treatment and 78 treated individuals, an abnormal κ/λ sFLC ratio was associated with shorter time to first treatment; median 48 months vs. 117 months for a normal ratio (p=0.001) (Figure 33.2B) [671]. Patients were allocated to one of three groups based on their normal or abnormal κ/λ sFLC ratios (analysing abnormally high or low ratios separately). An abnormal sFLC ratio was prognostic for reduced overall survival (OS) from diagnosis (Figure 33.2). The sFLC ratio remained prognostic for OS when the groups of 181 untreated/pre-treatment patients were considered separately [671]. Similar findings have been reported by Morabito et al. [675]. On multivariate analysis, four independent prognostic variables for OS were identified: 1) κ/λ sFLC ratio (p=0.024); 2) β2-microglobulin concentration (p=0.01); 3) IGHV mutation status (p=0.017); and 4) ZAP-70 expression (p=0.0001) [671].

Maurer et al. [673] studied the prognostic value of monoclonal sFLC elevation, polyclonal sFLC elevation and ratio-only abnormality in a cohort of 339 newly diagnosed, untreated CLL patients. After a median follow-up of 47 months, 26% had been treated and 10% had died. All three sFLC abnormalities were associated with a reduced time to first treatment (Figure 33.3A). A monoclonal or polyclonal sFLC elevation was also associated with poor overall survival, compared with patients with normal sFLC concentrations (Figure 33.3). The group with monoclonal sFLC elevation had the worst prognosis, and was shown to have distinct clinical characteristics associated with more aggressive disease. These included a higher prevalence of high-risk biologic characteristics (CD38+, CD49d+, ZAP-70+, IGHV unmutated, and high-risk cytogenetic abnormalities) compared with patients with normal sFLC. Consistent with this finding, the most common cause of death in the monoclonal sFLC elevation group was progressive CLL. In contrast, most deaths in the polyclonal sFLC group were due to other causes, and the authors concluded that polyclonal FLC elevations may serve as a marker of host “fitness”.

33.4. Prognostic value of combined FLC measurements

Summated κ + λ sFLCs (ΣFLC) is a measure of both monoclonal tumour cell and polyclonal “bystander” FLC production. Morabito et al. [675] studied the prognostic utility of ΣFLC in 449 untreated CLL patients. After a median follow-up of 3 years, 33% of patients had received treatment. Receiver operating characteristic (ROC) analysis defined an optimal cut-off of ΣFLC (60.6 mg/L) that identified patients with inferior outcome. The percentages of patients not requiring treatment at 3 years were 84.1% and 51.8% for patients with ΣFLC ≤60.6 mg/L or >60.6 mg/L, respectively (Figure 33.4A). A further study by Sarris et al. [676] confirmed that baseline ΣFLC >60 mg/L were associated with shorter TTFT, and also demonstrated that ΣFLC >60 mg/L correlated with shorter OS. Morabito et al. [675] demonstrated that on multivariate analysis, ΣFLC >60.6 mg/L, ZAP-70 expression, cytogenetic abnormalities, and Binet stage B+C remained significantly associated with inferior TTFT. The prognostic significance of ΣFLC >60.6 mg/L was significantly higher than the sFLC κ/λ ratio, with a 3-fold higher risk of early treatment requirement for patients with ΣFLC >60 mg/L compared to those with an abnormal κ/λ sFLC ratio (hazard ratios of 3.5 [p<0.0001] versus 1.5 [p=0.015]).

Morabito et al. [675] proposed a novel prognostic scoring system was proposed in which ΣFLC >60.6 mg/L, ZAP-70 expression, cytogenetic abnormalities and Binet stage were combined [675]. In this model, 1 point was assigned for each unfavourable marker present. The percentages of patients not requiring treatment at 3 years were 94.8%, 84.5%, 61.6% and 21.1% for patients scoring 0, 1, 2 or 3/4 respectively (p<0.0001) (Figure 33.4B). The authors concluded that the cumulative amount of ΣFLC, irrespective of their clonality, represents a strong and independent prognostic predictor in CLL. An alternative CLL risk-stratification model that incorporates ΣFLC concentrations and Hevylite immunoparesis was proposed by Tadmor et al. [1203].

33.5. Monitoring CLL with sFLCs

A preliminary study by Aue et al. [678] highlighted the potential utility of sFLCs for monitoring CLL. Prior to treatment, a total of 8/11 patients with a κ CLL clone had κ sFLC concentrations above the normal range. After 6 months of ibrutinib therapy, there was a 76% reduction in κ sFLCs (p<0.01), and in 7/8 patients sFLC concentrations had normalised. By contrast, λ FLCs were initially low and then increased to normal levels following therapy. Similar findings were reported for 8 of 9 patients with a λ clone. The authors suggested that sFLC monitoring might provide an insight into both the killing of the tumour cells and the recovery of normal B-cell function.


  1. What is the source of sFLCs in CLL?


  1. FLCs may be produced by the tumour clone, polyclonal bystander lymphocytes and plasma cells in the tumour microenvironment (Section 33.2).