INTRODUCTION — Chronic lymphocytic leukemia (CLL) is a chronic lymphoproliferative disorder (lymphoid neoplasm) characterized by the progressive accumulation of functionally incompetent lymphocytes, which are usually monoclonal in origin.
CLL is considered to be identical (ie, one disease with different manifestations) to the mature (peripheral) B cell neoplasm small lymphocytic lymphoma (SLL), a clinically indolent non-Hodgkin lymphoma. The term CLL is used when the disease manifests primarily in the blood, whereas the term SLL is used when involvement is primarily nodal.
The staging and prognosis of CLL/SLL will be reviewed here. The pathobiology, clinical manifestations, diagnosis, and treatment of CLL/SLL are discussed separately, as is the related condition monoclonal B cell lymphocytosis.
●(See "Pathobiology of chronic lymphocytic leukemia".)
●(See "Clinical features and diagnosis of chronic lymphocytic leukemia/small lymphocytic lymphoma".)
●(See "Overview of the treatment of chronic lymphocytic leukemia".)
●(See "Monoclonal B cell lymphocytosis".)
NATURAL HISTORY — The natural history of CLL/SLL is extremely variable, with survival times from initial diagnosis that range from approximately 2 to 20 years, and a median survival of approximately 10 years [1,2].
Some patients experience a rapid deterioration and die within two to three years from complications or causes directly related to CLL/SLL. Others follow a clinically benign course for 5 to 10 years, followed by a terminal phase lasting one to two years. A minority of patients (<30 percent) follow a clinically benign course for 10 to 20 years, and the eventual cause of death may be unrelated to CLL/SLL. Spontaneous clinical regression has been reported in the absence of therapy, but it is rare [3].
During the initial asymptomatic phase, patients maintain their usual lifestyles, but during the terminal phase the performance status can be poor, with recurring need for hospitalization. The most frequent causes of death are severe systemic infection (especially pneumonia and septicemia), bleeding, and inanition with cachexia. During this terminal phase there can be considerable morbidity, both from the disease itself and from complications of therapy. (See "Overview of the complications of chronic lymphocytic leukemia".)
CLINICAL STAGING
Prognostic value — The Rai and Binet staging systems use physical examination and complete blood count to stratify patients into three risk groups (low, intermediate, and high) with different predicted overall survival (OS) (table 1 and table 2) and to identify patients who would benefit from treatment (eg, Rai stage III or IV, Binet stage C).
Survival curves for Rai low risk (stage 0), intermediate risk (stage I or II), and high risk (stage III or IV) groups correspond to those of the Binet stage A, B, and C, respectively. While increasing stage is associated with shorter survival, the absolute survival estimates reported in the original publications are significantly shorter than expected survival with modern therapy.
●Stage reflects burden of disease – Increasing stage reflects a gradual and progressive increase in the body burden of leukemic lymphocytes, starting in the blood and bone marrow (lymphocytosis), progressively involving lymph nodes (lymphadenopathy), spleen and liver (organomegaly), with eventual compromise of bone marrow function (anemia and thrombocytopenia).
With the passage of time, patients tend to progress from an early stage (low risk), to an intermediate stage, and eventually into an advanced stage (high risk) [1]. However, if a patient's stage shifts from a higher risk to a lower risk category with successful therapy, the outlook for survival improves accordingly [4]. These systems were developed with data from patients treated 40 to 50 years ago. Therapy in CLL has markedly improved during the past decades. As such, the prognosis for each stage has improved with the advent of new therapies [5].
●Imaging for selected patients – Imaging is not used for routine staging, but it should be performed in patients with signs or symptoms concerning for enlarged abdominal or pelvic nodes (eg, obstructive jaundice or obstruction of the inferior vena cava or ureters). (See "Overview of the treatment of chronic lymphocytic leukemia", section on 'Indications for treatment ("active disease")'.)
●Impact of autoimmune cytopenias – While there may be other causes of low blood counts (eg, autoimmune cytopenias, prior therapy, coexisting illness), these are not accounted for when determining stage. However, patients with cytopenias due to autoimmune processes appear to have a better prognosis than patients with cytopenias due to bone marrow failure [6,7].
This issue was addressed by a single institution retrospective study of 960 patients with CLL/SLL that identified an autoimmune cytopenia in 70 patients (7 percent) at some point during the disease process [7]. Patients who were classified as Binet stage C due to autoimmune cytopenias had longer median survival than patients with cytopenias due to bone marrow infiltration (7 versus 4 years) and a worse median survival than those with stage A disease (7 versus 10 years). Many patients with autoimmune cytopenias were able to normalize their blood counts after therapy directed at the autoimmune process and thereby postpone the initiation of chemotherapy.
Rai staging system — The Rai staging system uses physical examination and complete blood count to stratify patients into three risk groups, which can be further subdivided into five stages with different estimated OS (table 1). While the prognostic value of the Rai staging system remains, the estimated median OS by stage has improved dramatically as treatments have evolved.
In the original series published in 1975, the percentage of patients in each stage and estimated median OS were as follows [1]:
●Low risk
•Stage 0 – Lymphocytosis alone (18 percent; historical median OS >150 months)
●Intermediate risk
•Stage I – Lymphadenopathy (23 percent; historical median OS 101 months)
•Stage II – Enlarged liver and/or spleen (31 percent; historical median OS 71 months)
●High risk
•Stage III – Hemoglobin <11 g/dL (17 percent; historical median OS 19 months)
•Stage IV – Platelet count <100,000/microL (11 percent; historical median OS 19 months)
The percentage of patients with stage 0 disease at diagnosis may differ depending on the use of "routine" laboratory testing with complete blood counts.
Although the initial report noted that there were only three distinct actuarial survival patterns (stage 0, stages I and II combined, and stages III and IV combined), they recommended that the five-stage system be maintained to investigate prospectively whether biological and clinical differences would emerge between stages I and II and between stages III and IV. It became apparent in the 1980s, however, that it was impractical to stratify patients in five categories for prospective randomized trials and the staging system was modified to consist of three risk groups (low, intermediate, and high) [8].
Binet staging system — The Binet staging system uses physical examination and complete blood count to stratify patients into three risk groups with different estimated OS (table 2). While the prognostic value of the Binet staging system remains, the estimated median OS by stage has improved dramatically as treatments have evolved.
Data from the original series published in 1981 were as follows [2,9]:
●Stage A (low risk) – <3 involved lymphoid sites (median OS comparable to age-matched controls)
●Stage B (intermediate risk) – ≥3 involved lymphoid sites (historical median OS 84 months)
●Stage C (high risk) – Hemoglobin <10 g/dL and/or platelets <100,000/microL (historical median OS 24 months)
To determine Binet stage, physical examination evaluates five potential sites of lymphoid involvement: cervical, axillary, and inguinal lymph nodes (whether unilateral or bilateral, each area is counted as one), plus spleen and liver. Lymph node enlargement is defined as ≥1 cm.
Computed tomography (CT) of the chest, abdomen, and pelvis is not routinely performed as part of the staging of patients with CLL/SLL outside of a clinical trial, but it may be indicated based on the patient's symptoms [10]. CT scans may reveal enlargement of retroperitoneal or mediastinal nodes that are not included for determining palpable enlargement among the five sites of enlarged lymphoid areas.
PROGNOSTIC VERSUS PREDICTIVE MARKERS — Prognostic and predictive markers are measurable variables associated with disease outcome. They may be a single measurement (eg, TP53 mutation) or a combination of measurements (eg, Rai staging system). Whether a marker is prognostic, predictive, or both impacts its potential role in patient care [11,12].
●Prognostic markers – Prognostic markers are associated with clinical outcomes (eg, progression-free survival, overall survival) independent of therapy received. Such markers usually reflect tumor burden, distribution, and growth rate.
As an example, in CLL/SLL, the Rai staging system is a prognostic marker (table 1). Patients with a higher Rai stage will have a worse outcome than those with a lower Rai stage. This difference in predicted outcome is seen regardless of therapy chosen (eg, in patients receiving chemoimmunotherapy and in patients receiving targeted therapies). (See 'Rai staging system' above.)
●Predictive markers – Predictive markers inform the likelihood of response to a given treatment and therefore impact choice of therapy. Such markers may be the target of the treatment or may modify the expression and/or function of the target.
As an example, in CLL/SLL, del(17p) and TP53 mutation are used as predictive markers to guide therapy. CLL/SLL tumors with del(17p) or TP53 mutation have an impaired DNA damage response pathway and do not respond to chemoimmunotherapy but do respond to therapies that act through mechanisms independent of this pathway (eg, targeted therapies). (See 'Del(17p) or TP53 mutations' below.)
●Markers may be both prognostic and predictive – Some markers are both prognostic and predictive.
As an example, in CLL/SLL, del(17p) and TP53 mutation appear to be both predictive and prognostic. While targeted therapies are superior to chemoimmunotherapy in this population, del(17p) and TP53 mutation retain their prognostic impact; patients with del(17p) or TP53 mutation treated with targeted agents have inferior outcomes when compared with other patients without either del(17p) or TP53 mutations and treated with these same agents [13].
PREDICTIVE MARKERS — Predictive markers inform the likelihood of outcome to a given treatment, and therefore impact choice of therapy. We have incorporated predictive markers into our preferred initial therapy (algorithm 1) and management of relapsed or refractory CLL (algorithm 2 and table 3).
Del(17p) or TP53 mutations — TP53 is a tumor-suppressor gene located on the short arm of chromosome 17; it can be inactivated by deletion and/or point mutation. Testing for del(17p) and TP53 mutations is a routine part of the pretreatment evaluation of CLL/SLL before every new line of therapy to detect clonal evolution, and it impacts our choice of therapy (algorithm 1 and algorithm 2 and table 3). In CLL/SLL, del(17p) and TP53 mutation are both predictive of worse clinical outcomes with chemoimmunotherapy as compared to targeted therapies (eg, BTK or BCL2 inhibitors) (figure 1), and they are prognostic for patients treated with all available therapies.
●Detection – Del(17p) is evaluated with fluorescence in situ hybridization (FISH) of the peripheral blood. TP53 mutation can be evaluated with Sanger sequencing or next-generation sequencing (NGS). While NGS increases sensitivity, it also increases the false positive rate.
●Frequency of abnormalities – Using FISH techniques, deletions of 17p are found in 7 to 10 percent of treatment-naïve patients with CLL/SLL and in a much higher proportion of those with relapsed or refractory disease [14,15]. TP53 gene mutations are identified in 10 to 47 percent of patients with CLL/SLL, depending on disease phase and prior therapy [16-22].
●Predictive value – CLL/SLL tumors with del(17p) or TP53 mutation have an impaired DNA damage response pathway and do not respond durably to chemoimmunotherapy, but they do respond to therapies that act through mechanisms independent of this pathway (eg, targeted therapies). While the predictive value of del(17p) and TP53 mutation has not been formally evaluated in a randomized trial, it is widely accepted as a predictive marker based on the magnitude of effect in nonrandomized comparisons. (See "Pathobiology of chronic lymphocytic leukemia", section on 'Impaired DNA damage response'.)
The choice of therapy in patients with del(17p) or TP53 mutation is discussed in detail separately (algorithm 1). (See "Selection of initial therapy for symptomatic or advanced chronic lymphocytic leukemia/small lymphocytic lymphoma", section on 'Del(17p) and/or TP53 mutations (very high risk)'.)
●Prognostic value – While targeted therapies are superior to chemoimmunotherapy in this population, del(17p) and TP53 mutation retain their prognostic impact; patients with del(17p) or TP53 mutation treated with targeted agents have inferior outcomes when compared with other patients treated with these same agents [13].
IGHV mutation status, ZAP-70, and CD38 — Testing for immunoglobulin heavy chain variable region (IGHV) gene mutation status is a routine part of the pretreatment evaluation of CLL/SLL, as IGHV mutation status can impact the choice of therapy (algorithm 1 and algorithm 2 and table 3). IGHV gene mutation status does not change with disease evolution and does not need to be retested at progression.
●Detection – IGHV mutation status can be tested on the peripheral blood or bone marrow. IGHV is usually evaluated with Sanger sequencing, and it is compared with germline genes available in immunoglobulin databases [23,24]. Mutated IGHV genes are defined in most studies as having a >2 percent difference in nucleotide sequence compared with germline DNA [25,26].
●Frequency of abnormalities – Approximately half of CLL/SLL tumors have mutated IGHV genes [5,25,26].
●Predictive value – In CLL/SLL, the IGHV gene mutation status is predictive of outcomes following chemoimmunotherapy as compared to targeted therapy. In prospective trials with long follow-up, a subset of patients with IGHV-mutated CLL/SLL obtain prolonged durable remissions after chemoimmunotherapy [27-29]. These durable remissions are not seen in patients with IGHV-unmutated CLL/SLL. (See "Selection of initial therapy for symptomatic or advanced chronic lymphocytic leukemia/small lymphocytic lymphoma", section on 'Immunoglobulin heavy chain variable mutated (standard risk)'.)
In the CLL14 trial, unmutated IGHV was identified as a predictive biomarker for increased benefit of venetoclax plus obinutuzumab over obinutuzumab plus chlorambucil [13].
●Prognostic value – Unmutated IGHV genes are associated with shorter time to initial therapy, higher risk of relapse following chemoimmunotherapy and hematopoietic cell transplantation, and shorter overall survival [5,25,26,30-38]. In one study of 325 consecutive patients with CLL/SLL, the presence of unmutated IGHV genes was associated with a hazard ratio of death of 5.3 (95% CI 2.7-10), as compared with those with mutated IGHV genes [33].
Expression of zeta chain associated protein 70 (ZAP-70) and CD38 are imperfect surrogates for IGHV mutation status. In multiple studies, increased expression of ZAP-70 and CD38 are associated with unmutated IGHV genes, higher clinical stage, greater tendency for disease progression, a poor response to treatment, and shorter survival [25,39-48]. However, correlation is not perfect. In addition, the best cutoff value for determining ZAP-70 positivity is unclear and levels appear to change over time, while IGHV mutation status remains constant [49].
The relationships among lymphocyte developmental stage and morphology, CD38 and ZAP-70 positivity, genetic studies (eg, chromosomal changes, gene expression profiling, IGHV gene mutation status, microRNA signatures), and other cellular variables as important prognostic features are complex and the subject of considerable ongoing interest [50-63].
NOTCH1 mutations — Testing for NOTCH1 mutations is not routinely performed in CLL/SLL, and the predictive and prognostic value of NOTCH1 mutations is controversial. Confirmatory trials are required to determine whether NOTCH1 status should impact treatment choice.
●Frequency of abnormalities – Mutations in NOTCH1 can be found in 4 to 13 percent of patients with CLL/SLL [22,64-70]. NOTCH1 mutations are virtually mutually exclusive with SF3B1 mutations [30].
●Predictive value – Some studies suggest that NOTCH1 mutations are predictive of worse clinical outcomes following therapies that target CD20, while other have not. The impact may differ with the specific anti-CD20 therapy used. In a subgroup analysis of the CLL8 study, CLL/SLL with NOTCH1 mutation did not appear to benefit from the addition of rituximab to treatment with fludarabine plus cyclophosphamide [30]. However, other studies suggest that CLL/SLL with NOTCH1 mutation does benefit from obinutuzumab [13].
●Prognostic value – NOTCH1 mutations are associated with unmutated IGHV and it is unclear whether it is an independent prognostic marker [71].
BTK, PLCg2, and BCL2 mutations — Testing for mutations of genes in the B cell receptor (BCR) pathway (BTK and PLCg2) and BCL2 can inform treatment at the time of disease progression (algorithm 2 and table 3). Mutations in these pathways are uncommon at the time of CLL/SLL diagnosis, and testing at diagnosis is not indicated [72]. However, a percentage of CLL/SLL tumors that progress on targeted therapy will have mutations in the target, which impact choice of subsequent therapy. (See "Treatment of relapsed or refractory chronic lymphocytic leukemia", section on 'Choice of therapy'.)
●BTK and PLCg2 mutations – Following treatment that targets the BCR pathway, mutations in BTK and PLCg2 are assumed to be predictive of worse clinical outcomes with BCR inhibitors.
●BCL2 mutations – Following treatment with venetoclax, mutations in BCL2 are assumed to be predictive of worse outcomes upon retreatment with venetoclax-based therapy.
Initial studies suggest that many CLL/SLL progressions on ibrutinib, acalabrutinib, and zanubrutinib are characterized by point mutations (eg, BTK C481S) that disrupt the covalent binding of these agents to BTK or point mutations in phospholipase Cy2 (PLCg2), which plays a role in BCR signaling [73-77]. These mutations are not detectable before therapy but are selected for during BTK inhibitor treatment in a proportion of patients. The BTK C481 mutation does not block binding of noncovalent, reversible-binding BTK inhibitors such as pirtobrutinib; however, CLL/SLL progressions on pirtobrutinib have demonstrated novel, non-C481 BTK mutations that confer resistance to noncovalent BTK inhibitors and some covalent BTK inhibitors [78,79].
Similarly, BCL2 gene mutations can be detected in a percentage of CLL/SLL tumors that progress on long-term BCL2-directed therapy (ie, venetoclax) [80-83].
PROGNOSTIC MARKERS — Prognostic markers are associated with clinical outcomes (eg, progression-free survival [PFS], overall survival [OS]) independent of therapy received. Such markers usually reflect tumor burden, distribution, and growth rate.
As described separately, the Rai and Binet staging systems use physical examination and complete blood count to stratify patients into three risk groups (low, intermediate, and high) and provide prognostic information for patient care. Both systems are in wide use in clinical practice and are described in detail separately. (See 'Clinical staging' above.)
However, clinical staging is imperfect. Some patients classified as having early-stage disease progress rapidly. Other prognostic markers have been investigated to identify these patients (table 4). Some have been incorporated into prognostic scores.
The main markers that are used clinically are:
●Rai and Binet clinical staging (see 'Clinical staging' above)
●Del(17p) and TP53 mutation status (see 'Del(17p) or TP53 mutations' above)
●Immunoglobulin heavy chain variable region (IGHV) gene mutation status (see 'IGHV mutation status, ZAP-70, and CD38' above)
●Lymphocyte doubling time (see 'Lymphocyte doubling time' below)
●Beta-2 microglobulin (see 'Beta-2 microglobulin' below)
●Del(11q), del(13q), and trisomy 12 (see 'Genomic abnormalities' below)
The clinical use of other prognostic markers (eg, CD38, ZAP-70, NOTCH1, BIRC3, and SF3B1 mutation) is less clear.
Several prognostic factors have been tested and found to be of limited, if any, usefulness in predicting the course of nonadvanced stages of CLL [3,84]. These prognostic factors include: morphologic features of blood lymphocytes (ie, the relative proportion of prolymphocytes [85,86]), phenotypic profile of B-lymphocytes, and levels of serum immunoglobulins [87,88].
Prognostic scores
Scores for early-stage CLL — Prognostic scores have been developed to predict time to first treatment for patients presenting with early-stage CLL (Rai stage 0, I, or II or Binet stage A).
●CLL1-PM – The CLL1 prognostic model (CLL1-PM) uses a weighted point system with six variables to stratify patients with early-stage CLL/SLL observed at diagnosis [89,90]:
•Del(17p) – 3.5 points
•Unmutated IGHV – 2.5 points
•Del(11q) – 2.5 points
•Serum beta-2 microglobulin >3.5 mg/L – 2.5 points
•Lymphocyte doubling time <12 months – 1.5 points
•Age >60 years – 1.5 points
These variables separate patients into four risk groups with differing likelihood of treatment-free survival (TFS) at 5 and 10 years:
•Very low risk (0 to 1.5 points) – 86 and 67 percent 5- and 10-year TFS; median not reached
•Low risk (2 to 4 points) – 52 and 26 percent 5- and 10-year TFS; median 64 months
•High risk (4.5 to 6.5 points) – 28 and 3 percent 5- and 10-year TFS; median 41 months
•Very high risk (7 to 14 points) – 11 and 0 percent 5- and 10-year TFS; median 28 months
The CLL1-PM was developed using data from 539 patients with Binet stage A CLL enrolled on the prospective CLL1 trial. These factors were also independent predictors of OS, although standard therapy at the time was chemoimmunotherapy.
●IPS-E – The International Prognostic Score for Early-stage CLL (IPS-E) uses three variables to stratify patients with early-stage CLL/SLL at diagnosis [91]:
•Unmutated IGHV
•Lymphocytes >15,000/microL (>15 x 109/L)
•Palpable lymph nodes
These variables separate patients three risk groups with differing likelihood of requiring treatment at one and five years:
•Low risk (no risk factors) – <1 percent treated at 1 year; 8 percent treated at 5 years
•Intermediate risk (one risk factor) – 3 percent treated at 1 year; 28 percent treated at 5 years
•High risk (two or three risk factors) – 14 percent treated at 1 year; 61 percent treated at 5 years
The IPS-E was developed and validated using individual patient data from 11 international cohorts of approximately 5000 patients with early-stage CLL (Rai stage 0, I, or II or Binet stage A) initially managed with active surveillance [91]. Unmutated IGHV, lymphocytes >15,000/microL, and palpable lymph nodes consistently and independently correlated with time to first treatment with similar magnitude allowing for equal weighting. In contrast, del(17p) and TP53 mutation were not prognostic for time to first treatment even though these are important predictors of treatment response and help guide the choice of therapy once indicated.
Four factor prognostic model for ibrutinib — The four factor prognostic model for ibrutinib assigns one point for each of the following variables (table 5) [92]:
●Del(17p) or TP53 mutation
●Serum beta-2 microglobulin ≥5 mg/L
●Lactate dehydrogenase >250 units/L
●Relapsed or refractory disease
The sum of these points determines the prognostic score, which separates patients into three risk groups with different estimated PFS and OS following treatment with ibrutinib:
●Low risk (0 to 1 points) – 87 percent 3-year PFS; 93 percent 3-year OS
●Intermediate risk (2 points) – 74 percent 3-year PFS; 83 percent 3-year OS
●High risk (3 to 4 points) – 47 percent 3-year PFS; 63 percent 3-year OS
The scoring system was developed and validated using data from patients treated with ibrutinib on several industry sponsored trials and underwent external validation using data from a single center investigator-initiated phase II trial. The prognostic value was maintained when applied to individual cohorts of patients with treatment-naïve or relapsed/refractory disease, and it was superior to the CLL international prognostic index (CLL-IPI) in this population. The three risk groups also had different likelihoods of developing mutations in BTK and PLCg2 (17, 40, and 50 percent, respectively) and undergoing histologic transformation (0, 5, and 17 percent, respectively). Other groups have also validated the four factor prognostic score in other cohorts treated with ibrutinib [93].
CLL-IPI — The CLL international prognostic index (CLL-IPI) incorporates data regarding disease stage, biology, and patient-related factors. It was developed using data from patients receiving chemoimmunotherapy (CIT) and it is unknown whether the prognostic value will be maintained with targeted therapies.
A weighted point system is used to assess the following prognostic variables [71]:
●Del(17p) or TP53 mutation – 4 points
●Serum beta-2 microglobulin >3.5 mg/L – 2 points
●Unmutated IGHV – 2 points
●Intermediate- or high-risk Rai or Binet stage – 1 point
●Age >65 years – 1 point
The sum of these points determines the prognostic score, which separates patients into four risk groups with different estimated OS in the original cohort:
●Low risk (0 to 1 points) – 91 and 87 percent 5- and 10-year OS; median not reached
●Intermediate risk (2 to 3 points) – 80 and 40 percent 5- and 10-year OS; median 104 months
●High risk (4 to 6 points) – 53 and 16 percent 5- and 10-year OS; median 63 months
●Very high risk (7 to 10 points) – 19 and 0 percent 5- and 10-year OS; median 31 months
The scoring system was developed and validated using data from 3472 previously untreated patients enrolled on one of eight phase 3 clinical trials of CIT completed between 1997 and 2007 [71]. The type of first-line CIT treatment was not identified as an independent factor in this analysis; however, the studies did not include treatment with novel oral inhibitors (eg, idelalisib, acalabrutinib, ibrutinib, venetoclax), which have demonstrated activity in patients with del(17p) or TP53 mutations.
The ability of this scoring system to prognosticate OS and time to first treatment has also been validated in independent cohorts of patients with newly diagnosed and previously untreated patients with CLL/SLL [94-97]. In a meta-analysis of several trials, the percentage of patients remaining treatment-free at five years after diagnosis decreased with increasing CLL-IPI risk group (82 percent low risk, 45 percent intermediate risk, 30 percent high risk, and 16 percent very high risk) [96].
Individual factors
Lymphocyte doubling time — The lymphocyte doubling time (LDT) is the number of months it takes the absolute lymphocyte count to double. An actual or projected LDT in untreated patients <12 months is associated with more aggressive disease, and a longer LDT is associated with an indolent course [43,44,98]. While LDT is useful in estimating disease tempo in patients not requiring immediate treatment, application in more advanced disease is limited because it takes time to measure. Clinical application is described separately. (See "Overview of the treatment of chronic lymphocytic leukemia", section on 'Initial observation as standard care'.)
Beta-2 microglobulin — Beta-2 microglobulin (B2M) levels correlate with disease stage and tumor burden in patients with CLL/SLL, with increasing levels associated with a poorer prognosis [5,30,99-101]. B2M is incorporated into several prognostic scores for CLL. (See 'Prognostic scores' above.)
B2M may be regulated, at least in part, by exogenous cytokines [102]. The source of these elevated cytokines in CLL/SLL is unclear, although interleukin (IL)-6, which inhibits apoptosis in CLL/SLL cells, may be released from vascular endothelium [103]. B2M levels also rise with worsening renal dysfunction leading some investigators to suggest a measure of B2M adjusted for the glomerular filtration rate [104].
Genomic abnormalities — Genomic abnormalities as detected by fluorescence in situ hybridization (FISH) are present in most CLL/SLL tumors at diagnosis, and additional abnormalities are acquired with disease evolution. Our understanding regarding the prognostic value of specific cytogenetic and molecular findings in CLL/SLL is rapidly changing. While initial attempts have been made to incorporate this information into clinical practice, these will undoubtedly change as the impact of combinations of molecular findings is analyzed further.
In the pretreatment evaluation of patients with CLL/SLL, we routinely perform FISH of the peripheral blood for the following four common chromosomal abnormalities that can be detected in approximately 80 percent of CLL tumors [14,105,106]:
●Del(13q14) – (See 'Del(13q14)' below.)
●Trisomy 12 – (See 'Trisomy 12' below.)
●Del(17p12) – (See 'Del(17p) or TP53 mutations' above.)
●Del(11q22-23) – (See 'Del(11q)' below.)
In addition, we assess the following genetic abnormalities which help guide therapy (algorithm 1):
●TP53 mutation – (See 'Del(17p) or TP53 mutations' above.)
●IGHV gene mutation status – (See 'IGHV mutation status, ZAP-70, and CD38' above.)
Studies evaluating the prognostic impact of genetic abnormalities in CLL/SLL have had disparate results, which may be due to small sample size, variation in the timing of sample collection, and heterogeneity of treatments used.
Experts agree on the prognostic value of del(17p), TP53 mutation, del(11q), unmutated IGHV, del(13) as a sole abnormality, and trisomy 12. Further study is needed to confirm the prognostic value of other biomarkers, including mutations in BIRC3, SAMHD1, RPS15, NFKBIE, EGR2, KRAS, and POT1, in particular in the era of targeted therapy.
For each biomarker, it is important for future studies to assess whether the prognostic value remains in the context of new therapies.
Del(17p) — Del(17p) is found in 7 to 10 percent of treatment-naïve patients with CLL/SLL and in a much higher proportion of those with relapsed or refractory disease [14,15]. Del(17p) is both predictive of worse clinical outcomes with chemoimmunotherapy as compared to targeted therapies (eg, BTK or BCL2 inhibitors) (figure 1) and prognostic for patients treated with all available therapies. (See 'Del(17p) or TP53 mutations' above.)
Del(11q) — Del(11q22-q23) is present in 10 to 20 percent of patients with CLL/SLL [14,46]. CLL/SLL with del(11q) has an intermediate risk prognosis.
Historically, patients with del(11q) have been at high risk of either not responding to initial treatment or relapsing soon after achieving remission. The prognosis of patients with del(11q) has improved with the use of fludarabine, cyclophosphamide, rituximab. Even so, outcomes are expected to be better with targeted therapies.
An analysis of patients enrolled on the randomized CLL8 trial evaluating the addition of rituximab to fludarabine plus cyclophosphamide (FCR) in previously untreated CLL/SLL, identified 17p deletion, TP53 mutation, 11q deletion, SF3B1 mutation, unmutated IGHV, and elevated thymidine kinase as independent markers that were associated with short PFS following initial treatment [30]. All but 11q deletion were also prognostic factors of short OS along with older age, poor performance status, and elevated B2M. Median PFS rates following FCR for those with del(17p), TP53 mutations, del(11q), or normal cytogenetics were approximately 12, 15, 50, and 50 months, respectively.
In another study of patients treated with non-targeted therapy, CLL/SLL with del(11q) in the absence of TP53 and BIRC3 was associated with an estimated survival at 5 and 10 years of 66 and 37 percent, respectively (compared with 78 and 58 percent for low-risk disease) [107]. The 10-year life expectancy was approximately half of that expected in a matched sample from the general population without CLL/SLL.
Chromosome 11q contains the ataxia telangiectasia mutated (ATM) gene (mapped to chromosome 11q22.3). The ATM gene product is involved in the detection of DNA damage and plays an important role in cell cycle progression. CLL/SLL cells without normal ATM function are unable to respond appropriately to chemotherapy-induced DNA damage. (See "Pathobiology of chronic lymphocytic leukemia", section on 'Impaired DNA damage response'.)
Del(13q14) — Del(13q14) is detected in 50 to 60 percent of CLL/SLL tumors [14,46]. When present as the sole abnormalities, monoallelic or biallelic 13q abnormalities appear to be associated with a favorable outcome [14,108].
In a large study of outcomes following non-targeted therapy, patients with del(13q14) as the sole genetic lesion had an estimated 10-year life expectancy that was similar to that expected in a matched sample from the general population without CLL/SLL [107].
The deleted region contains miR15A and miR16A microRNA [109]. These microRNA regulate the expression of proteins involved in apoptosis and the cell cycle. Without them, the cells do not respond appropriately to stress signals that should induce apoptosis. The B cell leukemia/lymphoma 2 (BCL2) gene encodes the antiapoptotic protein BCL2, and is one of the genes upregulated when these microRNA are deleted [109]. (See "Pathobiology of chronic lymphocytic leukemia", section on 'Upregulation of BCL2 proto-oncogene'.)
Trisomy 12 — Trisomy 12 is detected in 10 to 20 percent of CLL/SLL tumors. The prognostic value of trisomy 12 detected by FISH is not clear.
One study using FISH found an association of trisomy 12 with advanced disease and higher proliferative activity [110], while another study found that median survival in patients with CLL/SLL and trisomy 12 detected by FISH was similar to that of patients with CLL/SLL and a normal karyotype [14].
In a large study of outcomes following non-targeted therapy, patients with trisomy 12 and those without genetic lesions were considered to have low-risk disease, although 10-year life expectancy was approximately 70 percent of that expected in a matched sample from the general population without CLL/SLL [107].
Complex karyotype — Metaphase cytogenetic testing is not routinely performed on CLL/SLL cells. It is technically challenging and requires specifically stimulated cultures since the majority of CLL/SLL cells are in G0/G1 phase of the cell cycle. However, accumulating evidence suggests that complex karyotype may be an independent marker of worse prognosis [111-116]. Many studies have defined complex karyotype as the presence of at least three structural and/or numerical chromosomal aberrations on chromosome-banding analysis.
The largest study to evaluate the impact of complex karyotype was a multicenter, retrospective study of >5000 patients with CLL/SLL using stimulated cultures [113]. The presence of five or more chromosomal aberrations was a marker of worse survival, independent of clinical stage, IGHV mutation status, and TP53 aberration status. In contrast, the prognostic impact of having three or four aberrations was limited to those with TP53 aberrations. Among those without TP53 aberrations, survival was similar among those with zero to five aberrations.
These results suggest heterogeneity among those with traditionally defined complex karyotypes. However, this study had several limitations. It was a retrospective cohort with data collected outside of clinical trials with heterogeneous treatment not using modern targeted therapy. In addition, the number of patients with complex karyotype and without TP53 aberrations was small. As such, correlations between survival and number of karyotype abnormalities are exploratory. Further study is needed to better understand whether the identification of a complex karyotype should impact treatment decisions.
TP53 mutations — TP53 gene mutations are identified in 10 to 47 percent of patients with CLL/SLL, depending on disease phase and prior therapy [16-22]. TP53 mutations are both predictive of worse clinical outcomes with chemoimmunotherapy as compared to targeted therapies (eg, BTK or BCL2 inhibitors) (figure 1) and prognostic for patients treated with all available therapies. (See 'Del(17p) or TP53 mutations' above.)
SF3B1 mutations — Testing for SF3B1 mutations is not routinely performed in CLL/SLL outside of a clinical trial.
SF3B1 mutations (most commonly K700E) can be identified in 15 to 18 percent of patients with CLL/SLL and are associated with a poor prognosis [22,30,45,68,117-120].
In an analysis of patients enrolled on the randomized CLL8 trial, SF3B1 mutation was identified as an independent marker associated with worse PFS and OS following initial chemoimmunotherapy [30].
In another study of outcomes following non-targeted therapy, patients with SF3B1 mutation in the absence of TP53 and BIRC3 abnormalities were considered to have intermediate-risk disease with estimated survival at 5 and 10 years of 66 and 37 percent, respectively (compared with 78 and 58 percent for those with low-risk disease) [107]. The 10-year life expectancy was approximately half of that expected in a matched sample from the general population without CLL/SLL.
The SF3B1 gene encodes for part of a nuclear ribonucleoprotein that complexes with other nuclear ribonucleoproteins to create the spliceosome that is responsible for splicing messenger RNA. (See "Pathobiology of chronic lymphocytic leukemia", section on 'Gene mutations'.)
Other markers — Several other prognostic markers have been studied, but their actual place in clinical practice has not been established. A few of particular interest are described below. The prognostic value of some of these markers (eg, CD49d) are likely to be dependent on the specific therapies employed.
●Free light chain assay – A prospective study of 339 patients with CLL/SLL reported that the 49 percent of patients who had abnormality in free light chain assays (eg, monoclonal, polyclonal, or kappa/lambda ratio abnormalities) had a shorter time to initial therapy than those with normal free light chain assays [121]. Patients with either monoclonal or polyclonal abnormalities had inferior OS when compared with those with normal or kappa/lambda only abnormalities. These results have been confirmed in another cohort [122].
●Vitamin D levels – Patients with CLL/SLL have a similar rate of vitamin D deficiency as that seen in the general population (approximately 30 to 40 percent). A prospective cohort analysis identified low vitamin D levels (<25 ng/mL) as a marker of poor prognosis in 390 patients with newly diagnosed CLL/SLL and confirmed this prognostic value in a second cohort of 153 patients [123]. On multivariate analysis, vitamin D deficiency was associated with a shorter time to initiation of treatment (HR 1.47; 95% CI 1.11-1.96) and a trend toward shorter survival (HR 1.47; 95% CI 0.97-2.23). (See "Vitamin D deficiency in adults: Definition, clinical manifestations, and treatment".)
●Pattern of bone marrow infiltration – While not routinely performed, studies suggest that the pattern of lymphocyte infiltration in the bone marrow biopsy specimen is an independent prognostic marker. A diffuse pattern is associated with a progressive course and a non-diffuse (interstitial and nodular) pattern is associated with a more indolent course [124-127].
●MicroRNAs – MicroRNAs (miRNAs) are small noncoding RNAs that modulate the expression of genes at the post-transcriptional level. The decreased expression of specific microRNAs (eg, miR-223 and miR-29c) and the increased expression of others (eg, miR-155) has been associated with reduced OS in retrospective analyses [128-130]. Confirmation in prospective trials is required to validate their prognostic value.
●CD49d expression – Increased expression of CD49d as detected by flow cytometry is associated with decreased TFS and OS [131-134]. The integrin CD49d interacts with receptors on endothelial cells (eg, VCAM-1) to modulate the homing of CLL cells to secondary organs and their interaction with the microenvironment [135].
●Serum VEGF – An Italian study found that Binet stage A patients with low serum concentrations of vascular endothelial growth factor (VEGF) had a median PFS that was not reached at 40 months, significantly longer than patients with higher serum VEGF concentrations (median PFS 33 months) [136]. CLL cells have been shown to produce VEGF, which may be important for expansion of lymphoid tissues, migration of CLL cells from blood to tissues, and/or prolongation of the survival of these cells [137].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic lymphocytic leukemia/small lymphocytic lymphoma".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient education" and the keyword(s) of interest.)
●Beyond the Basics topics (see "Patient education: Chronic lymphocytic leukemia (CLL) in adults (Beyond the Basics)" and "Patient education: Hematopoietic cell transplantation (bone marrow transplantation) (Beyond the Basics)")
SUMMARY
●Natural history – Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) has a heterogeneous natural history with survival times from initial diagnosis that range from approximately 2 to 20 years, and a median survival of approximately 10 years. Patients are often asymptomatic at diagnosis and able to maintain their usual lifestyle. CLL/SLL progresses at a variable rate to a terminal phase with worsening performance status and frequent complications from cytopenias. (See 'Natural history' above.)
●Clinical staging – The Rai and Binet clinical staging systems use physical examination and complete blood counts to stratify patients into three risk groups with differing expected disease tempo and overall survival (table 1 and table 2). Imaging is not used for routine staging, but it should be performed in patients with signs or symptoms concerning for enlarged abdominal or pelvic nodes (eg, obstructive jaundice or obstruction of the inferior vena cava or ureters). (See 'Clinical staging' above.)
●Predictive markers – The following predictive markers inform the likelihood of response to a given treatment and therefore impact choice of therapy (algorithm 1):
•Del(17p) and TP53 mutation – CLL/SLL tumors with del(17p) or TP53 mutation have an impaired DNA damage response pathway and have poor outcome with chemoimmunotherapy, which is markedly improved by therapies that are independent of this pathway (eg, targeted therapies). While targeted therapies are superior to chemoimmunotherapy in this population, patients with del(17p) or TP53 mutation treated with targeted agents have inferior outcomes when compared with other patients treated with these same agents. Testing for del(17p) and TP53 mutations is a routine part of the pretreatment evaluation of CLL/SLL both at diagnosis and at progression to detect clonal evolution. (See 'Del(17p) or TP53 mutations' above.)
•IGHV mutation status – CLL/SLL tumors with mutated immunoglobulin heavy chain variable region (IGHV) genes have an indolent clinical course, and a subset obtain prolonged durable remissions after chemoimmunotherapy. In contrast, these durable remissions are not seen in patients with IGHV-unmutated CLL/SLL. IGHV gene mutation status does not change with disease evolution and does not need to be retested at progression. (See 'IGHV mutation status, ZAP-70, and CD38' above.)
●Prognostic markers – Prognostic markers are associated with clinical outcomes (eg, progression-free survival, overall survival) independent of therapy received. Such markers usually reflect tumor burden, distribution, and growth rate. Several prognostic models have been proposed that combine several prognostic markers to predict outcomes in different populations:
•Observation of early-stage CLL/SLL – The CLL1 prognostic model (CLL1-PM) and the International Prognostic Score for Early-stage CLL (IPS-E) stratify patients with early-stage CLL/SLL to predict likelihood of requiring treatment at five years. (See 'Scores for early-stage CLL' above.)
•Ibrutinib – The four factor prognostic model for ibrutinib identifies those with three or four variables (17p aberration, beta-2 microglobulin ≥5 mg/L, lactate dehydrogenase >250 units/L, and prior treatment) as those at high risk of early progression on ibrutinib (table 5). (See 'Four factor prognostic model for ibrutinib' above.)
•Chemoimmunotherapy – The CLL international prognostic index (CLL-IPI) incorporates clinical stage, age, IGHV mutation status, del(17p), TP53 mutation status, and serum beta-2 microglobulin to predict time to first treatment and overall survival following chemoimmunotherapy. It is unknown whether the prognostic value will be maintained with targeted therapies. (See 'CLL-IPI' above.)
Prior to treatment, we routinely test for del(17p), TP53 mutation, IGHV mutation status, beta-2 microglobulin, and lactate dehydrogenase. We repeat each of these tests prior to each new line of therapy with the exception of IGHV mutation status. We also test for mutations of genes in the B cell receptor (BCR) pathway (BTK and PLCg2) and BCL2 for patients progressing after therapies targeting BCR and BCL2, respectively (algorithm 2 and table 3).
In addition, we routinely perform fluorescence in situ hybridization (FISH) of the peripheral blood for del(11q), trisomy 12, and del(13q). Of these, del(13q) and trisomy 12 are favorable prognostic findings. Del(11q) is intermediate risk. (See 'Genomic abnormalities' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Michael J Keating, MD, who contributed to earlier versions of this topic review.
15 : Presence of multiple recurrent mutations confers poor trial outcome of relapsed/refractory CLL.
55 : The LPL/ADAM29 expression ratio is a novel prognosis indicator in chronic lymphocytic leukemia.
56 : A MicroRNA signature associated with prognosis and progression in chronic lymphocytic leukemia.
Do you want to add Medilib to your home screen?