H&O How commonly does chronic lymphocytic leukemia (CLL) spread to the central nervous system (CNS)?
TC Our study found a prevalence of CNS involvement in CLL of 0.14%,1 and previous studies have reported a prevalence of approximately 1% to 2%.2 It is extremely difficult to say what the exact prevalence is because the prevalence can be either underestimated or overestimated, and it is difficult to conduct a comprehensive series to measure the frequency of this complication reliably. However, we can say that CNS involvement in CLL is extremely rare, probably less than 1%.
H&O What symptoms should cause a clinician to suspect CNS involvement in CLL?
TC Any CNS sign or symptom should prompt investigation; these include limb weakness, visual impairment, headache, nausea and vomiting, neck stiffness, cranial nerve palsies, and mental status changes. In our study, the most common signs and symptoms were motor impairment (38.3% of patients), visual impairment (31.9%), headache (27.7%), mental status changes (25.5%), and sensory impairment (25.5%).
CNS involvement is not at the top of the list of differential diagnoses when a patient with CLL has a CNS symptom, but it is one of the things that should be checked when other more common CNS pathologies have been excluded.
H&O Is CNS disease a primary complication of CLL itself, or is it more often a treatment-related phenomenon?
TC We do not consider CNS disease to be a treatment-related phenomenon. CNS symptoms that are due to treatment rather than to direct CLL involvement typically point toward other etiologies, such as hemorrhage or infection. In both our series and a previous series from France,3 we found that CNS involvement can occur in treatment-naive patients.
H&O What diagnostic workup do you recommend when CNS involvement is suspected?
TC When CNS involvement is suspected, we need to conduct imaging. The best test is magnetic resonance imaging (MRI) of the brain with contrast to detect leptomeningeal enhancement or, less commonly, parenchymal lesions. MRI of the spine may also be useful, depending on the symptoms. If MRI is not possible, computed tomography (CT) is also useful. Depending on the findings of imaging, a spinal tap may be useful to see whether the cerebrospinal fluid contains any CLL cells. Based on the MRI and spinal tap findings, a CNS biopsy may also be needed.
H&O How does Richter transformation factor into CNS disease?
TC Richter transformation develops in approximately 3% to 5% of patients with CLL, and these patients are at elevated risk for CNS involvement. When CNS involvement occurs in a patient with Richter transformation, we need to assess whether the involvement is caused by the transformation or by the CLL.
When researchers at the Mayo Clinic assessed 172 patients with CLL who had CNS symptoms, they found that 18 of them had CNS CLL and 15 had CNS Richter transformation.4 This finding indicates that CNS disease is far more common in patients with Richter transformation (as in those with other aggressive lymphomas) than in those with CLL, given how uncommon Richter transformation is. The study also showed that patients with Richter transformation had additional related signs, such as disease of the lymph nodes and B symptoms.
H&O Which patients are at the highest risk for CNS involvement?
TC It is difficult to say which patients with CLL are at the highest risk for CNS involvement. In the case of aggressive lymphomas, we have prognostic scores that help to determine which patients are at high risk for CNS involvement due to the lymphoma. However, because CNS involvement is so rare in CLL, we have not been able to formulate comparable prognostic scores to identify patients who are at risk for this complication. When we look at the retrospective data we have, we are unable to identify a symptom, biomarker, or other factor that can reliably predict the development of this complication in a given patient. Further complicating matters is that CNS involvement may occur either before or after treatment.
H&O What treatments or other options are available when CNS disease is confirmed?
TC Guidelines lack specific treatment recommendations for patients with CNS involvement. We do, however, know that Bruton tyrosine kinase (BTK) inhibitors are highly effective in CLL and cross the blood-brain barrier,5 so these are among the preferred options for patients who have CLL with CNS involvement. Preferred treatment should involve a BTK inhibitor, either as monotherapy or in combination with another agent. The agent that we most commonly add to BTK inhibition is the BCL2 inhibitor venetoclax (Venclexta, AbbVie), which also crosses the blood-brain barrier and can be effective in patients with CNS involvement.6
H&O What is typically the prognosis of a patient who has CLL with CNS involvement?
TC The prognosis of patients who have CLL with CNS involvement is excellent in comparison with that of patients who have other lymphomas with CNS involvement. Most of the patients who have been examined in retrospective case series—both our series of 48 patients and a French series with 30 patients3—have done very well, especially those who were treatment-naive. We see very few CNS relapses among these patients, which is similar to what we see in patients who have CLL without CNS involvement. We may learn about some small subgroups of patients who do not have such a good prognosis, but as far as we know now, CNS involvement seems to be a manageable complication.
H&O Can you describe in detail the design and results of the study that you published in January?
TC This was an international retrospective study from the European Research Initiative on CLL (ERIC), in which we looked at patients with CLL and identified those who also had CNS involvement. Our goals were to assess the prevalence of CNS involvement, analyze the characteristics of these patients before their diagnosis and after their treatment, and carefully observe their outcomes.
We received data from 26 centers around the globe for a total of 48 patients with CNS involvement. We included only patients who had a confirmed diagnosis of CNS involvement, which we defined as radiologic evidence of disease on MRI or CT, the presence of CLL cells in the cerebrospinal fluid or CNS on the basis of a tissue biopsy specimen, and no other explanation for the neurologic findings. More than half (62%) of these patients were treatment-naive, showing that CNS involvement can occur before treatment and is not limited to patients who have received multiple lines of treatment.
We restricted our analysis of the prevalence of CNS involvement to patients in a registry from eastern Denmark. Because the registry included nearly all patients in this region, we thought that our study would provide the most accurate calculation of the prevalence of this disease. As I mentioned earlier, the prevalence was 0.14%, so it is extremely rare.
Treatment consisted of BTK inhibition as monotherapy or as part of combination therapy in 36% of the patients, chemoimmunotherapy in 26%, a CNS lymphoma regimen in 8%, venetoclax with or without an anti-CD20 antibody in 13%, and other in 17%. When we assessed the outcomes of these patients, we found that they were excellent in those treated with BTK inhibitors. Initial treatment was highly effective, leading to a complete resolution of neurologic symptoms and imaging findings in 71% of the patients. The estimated 5-year overall survival rate after a diagnosis of CNS involvement was 77.1%. The 5-year time to next treatment or death was longer in the patients treated with a BTK inhibitor than in those who received chemoimmunotherapy (94% vs 64%). The people who received a BTK inhibitor generally did well whether it was a first-line or a later-line treatment. The patients with a complete response in the CNS had better outcomes than those without a complete response.
H&O Is there anything that you would like to add?
TC I would like to add that this is an extremely difficult topic to study because CLL accounts for just 1% of all cancers, and CNS involvement occurs in fewer than 1% of patients with CLL. As a result, it is nearly impossible to conduct prospective studies. We should keep in mind that all our data regarding this topic are from retrospective series.
Disclosures
Dr Chatzikonstantinou has received research support to his institution from Johnson & Johnson; has received consulting fees from AbbVie; has served on the speaker bureaus of AbbVie, AstraZeneca, BeOne Medicines, and Lilly; and has received travel support from AstraZeneca and Johnson & Johnson.
References
1. Chatzikonstantinou T, Mikhaleva M, Brieghel C, et al. Central nervous system involvement in CLL: an international retrospective study by ERIC, the European Research Initiative on CLL. Blood Adv. 2026;10(2):428-438.
2. Hanse MC, Van’t Veer MB, van Lom K, van den Bent MJ. Incidence of central nervous system involvement in chronic lymphocytic leukemia and outcome to treatment. J Neurol. 2008;255(6):828-830.
3. Wanquet A, Birsen R, Bonnet C, et al. Management of central nervous system involvement in chronic lymphocytic leukaemia: a retrospective cohort of 30 patients. Br J Haematol. 2017;176(1):37-49.
4. Strati P, Uhm JH, Kaufmann TJ, et al. Prevalence and characteristics of central nervous system involvement by chronic lymphocytic leukemia. Haematologica. 2016;101(4):458-465.
5. Yu H, Kong H, Li C, et al. Bruton’s tyrosine kinase inhibitors in primary central nervous system lymphoma-evaluation of anti-tumor efficacy and brain distribution. Transl Cancer Res. 2021;10(5):1975-1983.
6. Soumerai JD, Takvorian RW, Sohani AR, Abramson JS, Ferry JA. Venetoclax activity in a patient with central nervous system involvement by chronic lymphocytic leukaemia. Lancet Haematol. 2022;9(10):e796.