Can Lymphoma Spread to the Brain?

Lymphoma can spread to the brain, and when it does, the condition is called secondary central nervous system lymphoma (SCNSL). It is distinct from primary CNS lymphoma, which starts in the brain itself without evidence of disease elsewhere in the body at the time of diagnosis.¹ The overall risk of this happening depends heavily on the subtype of lymphoma, where it originated, and certain biological features of the tumor cells. For some patients the risk is vanishingly small; for others it is high enough that doctors take preventive steps from the start.

How Lymphoma Gets Past the Blood-Brain Barrier

The brain is shielded by a tightly sealed layer of cells lining its blood vessels, commonly known as the blood-brain barrier. This barrier keeps most large molecules, immune cells, and pathogens out of the brain tissue. It also keeps many chemotherapy drugs out, which is one reason brain involvement in lymphoma is so hard to treat once it happens.² For lymphoma cells to reach the brain, they need to breach or exploit weaknesses in that barrier.

Research has found that when lymphoma cells do infiltrate the brain, proteins involved in maintaining the barrier’s tight junctions are measurably disrupted. Patients with CNS involvement had far higher levels of molecules like MMP-9 and certain tight-junction proteins in their spinal fluid compared with lymphoma patients whose disease stayed outside the brain.³ Separately, studies using two-photon imaging in mouse models have shown that lymphoma growth in the brain is associated with barrier disruption and the deposition of the blood-clotting protein fibrinogen around blood vessels. Lymphoma cells appear to cluster at these leaky vascular sites, using fibrinogen as an adhesive foothold.⁴ In other words, the lymphoma cells don’t just randomly land in the brain. They exploit spots where the barrier is already compromised and essentially glue themselves to the surrounding tissue.

The brain’s status as an “immune-privileged” site adds another layer of complexity. Because the brain’s normal immune surveillance is limited, tumor cells that manage to get in face less resistance than they would in a lymph node or the bone marrow.⁵ This helps explain why brain relapses, once they occur, tend to be aggressive and difficult to control.

Which Lymphoma Subtypes Are Most Likely to Spread to the Brain

Not all lymphomas carry the same risk. Diffuse large B-cell lymphoma (DLBCL), the most common aggressive type, is the one most closely studied for CNS relapse. Within DLBCL, the risk varies dramatically depending on where the lymphoma first appeared. Testicular, breast, and uterine DLBCL share a genetic profile that resembles primary CNS lymphoma itself, including mutations in genes called MYD88 and CD79B. These mutations seem to give the tumor cells a built-in tendency to home to the brain, regardless of how advanced the disease is at diagnosis.⁶ In contrast, DLBCL arising in the stomach, thyroid, or bone carries a much lower risk of brain relapse and generally responds well to standard treatment.

Beyond the primary site, certain clinical features push the risk higher. A large study of nearly 1,700 patients with aggressive lymphoma found that elevated LDH (a blood marker of cell turnover) and involvement of more than one organ outside the lymph nodes were independent predictors of brain relapse. Elderly patients who had both elevated LDH and lymphoma in the liver, bladder, or adrenal glands faced up to a fifteen-fold increase in the risk of the disease reaching the brain.⁷ Kidney or adrenal involvement specifically has been flagged as a strong signal. In one series, four out of five patients whose lymphoma involved the kidney or adrenal glands went on to relapse in the brain.⁸

That same body of research found that certain gene rearrangements involving BCL6 and PD-L pushed the five-year cumulative rate of brain relapse to around 41% in affected patients.⁸ These are not numbers that apply to every person with DLBCL, but they illustrate how sharply the risk can spike when biology and clinical features align unfavorably.

How Doctors Estimate the Risk

Oncologists use a scoring system called the CNS International Prognostic Index (CNS-IPI) to sort patients into low, intermediate, and high risk groups. The model considers clinical factors like age, disease stage, LDH level, performance status, the number of organs involved, and whether the kidneys or adrenal glands are affected. In a large international study of over 1,500 patients, the three-year cumulative risk of brain relapse was about 0.4% for those classified as low risk, roughly 3% for intermediate risk, and about 11% for high risk.⁹

A separate analysis found two-year CNS relapse rates of about 0.8%, 1.9%, and 8.9% for the same three risk tiers.¹⁰ These models are better at ruling out risk than at pinpointing who will relapse. They have high negative predictive value, meaning a low score is genuinely reassuring, but only modest positive predictive value, meaning many patients flagged as high risk never actually develop brain disease.⁹ Incorporating additional biological information, like the cell-of-origin subtype of the tumor, can improve the prediction, but no scoring system yet catches every patient who will relapse.

What It Feels Like When Lymphoma Reaches the Brain

Symptoms of CNS lymphoma, whether primary or secondary, are often frustratingly nonspecific. The clinical presentation is frequently vague enough that it mimics other neurological conditions, and confirming the diagnosis requires a biopsy or other tissue-based test.¹¹ The most common complaints include signs of increased pressure inside the skull (headache, nausea, visual changes), personality or cognitive changes, unsteadiness, and weakness on one side of the body.¹² Some patients develop focal neurological problems tied to wherever the tumor is located, while others experience a more diffuse cognitive decline.¹³

A subtle but important point: some patients with lymphoma in the brain are neurologically asymptomatic. In a study of patients with vitreoretinal lymphoma (a type that involves the eye and is closely related to CNS lymphoma), about 17% had suspicious findings in their cerebrospinal fluid without any abnormalities on brain MRI. Of those patients, 86% had no neurological symptoms at all.¹⁴ This underscores why doctors sometimes look for CNS involvement even when patients feel fine, especially in high-risk subtypes.

Getting the Diagnosis Right

MRI with contrast dye is the first step and is highly sensitive. At diagnosis, contrast-enhancing lesions appear on MRI in about 99% of immunocompetent patients and 96% of HIV-related cases.¹⁵ However, the appearance on imaging alone is not enough to make the call. Primary and secondary CNS lymphoma can look similar on scans, though secondary disease more often appears below the tentorium (in the lower back part of the brain), while primary CNS lymphoma tends to cause more surrounding swelling.¹⁶

Cerebrospinal fluid analysis is the next tool, but its sensitivity varies widely depending on the technique used. Simple cytology (looking for cancer cells under a microscope) picks up disease in only about 2% to 32% of cases. Combining it with flow cytometry, which detects abnormal cell markers, improves the detection rate. Newer methods analyzing proteins and tiny RNA fragments in the spinal fluid push specificity above 95%.¹⁵ Still, the gold standard remains either a positive spinal fluid cytology result, a vitreous biopsy (for eye involvement), or a direct brain biopsy.

Treatment When Lymphoma Is in the Brain

High-dose methotrexate-based chemotherapy is the backbone of treatment for CNS lymphoma, whether primary or secondary.¹⁷ Methotrexate is one of the few conventional chemotherapy drugs that can cross the blood-brain barrier when given at high enough doses. The challenge is that the intact barrier still limits how much drug actually reaches the tumor, so dosing needs to be aggressive and is often accompanied by intensive hydration and monitoring for kidney toxicity.

Whole-brain radiation therapy was once the standard follow-up to chemotherapy, but its role has become increasingly controversial. Radiation is effective at controlling tumor growth, but it carries a serious cost to long-term brain health. In one phase 2 trial, delayed neurotoxicity including memory loss, personality changes, walking difficulty, and urinary incontinence appeared in roughly 15% of patients, with those over 60 being most vulnerable.¹⁸ The overall five-year risk of developing treatment-related neurotoxicity was estimated at about 30%, and even younger patients had a roughly 20% risk when followed long enough. Comprehensive neuropsychological testing in survivors strongly suggests that radiation is the main driver of cognitive damage, though chemotherapy itself contributes too.

Because of this, many treatment centers now reserve whole-brain radiation for patients who do not respond to chemotherapy, or use reduced doses. In a study comparing reduced-dose whole-brain radiation with autologous stem cell transplant as consolidation after chemotherapy, both groups showed cognitive improvement during the first three years. After that, attention, executive function, and memory began to decline in both groups, though patients who received radiation developed more white matter changes on brain imaging over time.¹⁹

Newer Targeted Drugs

One of the most promising developments has been the use of BTK inhibitors, a class of drugs that block a signaling molecule critical to the survival of many B-cell lymphomas. Ibrutinib, the first of these drugs, has been evaluated in patients with relapsed or treatment-resistant primary CNS lymphoma, with overall response rates ranging from roughly 52% to 89%.²⁰ These are encouraging numbers for a disease where second-line options used to be grim. Lab studies have confirmed that both ibrutinib and tirabrutinib distribute well into brain tissue, a crucial requirement for any drug aimed at CNS disease.²¹

Combination regimens are being explored as well. A retrospective study of zanubrutinib (a newer-generation BTK inhibitor) combined with lenalidomide and rituximab found promising results in relapsed or resistant CNS lymphoma, particularly in the MCD subtype that carries the MYD88 and CD79B mutations discussed earlier.²² That genetic profile, which was flagged as a risk factor for CNS spread, turns out to also predict sensitivity to BTK-targeted treatment, a useful silver lining.

CAR T-Cell Therapy for CNS Lymphoma

CAR T-cell therapy, in which a patient’s own immune cells are engineered to recognize and attack lymphoma, was initially off-limits for patients with brain involvement because of fears that the treatment’s known side effects would be amplified in the brain. A meta-analysis of 128 patients with CNS lymphoma who received anti-CD19 CAR T-cells found that the toxicity profile was actually comparable to what is seen in patients treated for systemic disease, with no increased signal of neurotoxicity. About 56% of primary CNS lymphoma patients and 47% of secondary CNS lymphoma patients achieved a complete remission. At six months, 37% in both groups remained in remission.²³ These are early results, and the durability of responses is still being studied, but the fact that CAR T-cells can be used safely in the brain at all represents a significant shift from just a few years ago.

Can Prophylaxis Prevent Brain Relapse

For patients deemed high-risk, oncologists often give preventive (prophylactic) methotrexate, either injected directly into the spinal fluid or given intravenously at high doses alongside the main chemotherapy. The rationale is straightforward: if you treat potential microscopic brain disease before it grows, you might prevent a full relapse. In practice, the evidence for this approach is frustratingly ambiguous.

A study comparing the two routes found that the five-year CNS relapse rate was about 5.6% with intrathecal methotrexate and 5.2% with high-dose intravenous methotrexate, a negligible difference. The authors concluded that whatever benefit methotrexate prophylaxis provides appears to be temporary, and that more effective preventive approaches are needed.²⁴ Another comparison of the two methods found identical relapse rates of about 18% in each arm, with no statistical difference. The high-dose intravenous route caused more side effects and was substantially more expensive.²⁵ A Japanese study using propensity-score matching similarly found no superiority of high-dose over intrathecal methotrexate, with three-year CNS relapse rates of about 4% versus 6%.²⁶

The honest assessment is that CNS prophylaxis remains one of the most contentious topics in lymphoma treatment. There is still no randomized trial definitively proving that either route of methotrexate prophylaxis prevents brain relapse in the long run, and several studies suggest it at best delays rather than prevents it. Despite this, most guidelines still recommend it for high-risk patients, partly because the consequences of untreated CNS relapse are severe enough that even uncertain protection feels worthwhile.

The Cognitive Cost of Treatment

Even when treatment works, many survivors of CNS lymphoma face lasting cognitive problems. In one study, 63% of patients who had achieved a complete remission from primary CNS lymphoma still showed cognitive impairment on formal testing, compared with 11% of matched controls. About a fifth had severe deficits. Only 42% of patients returned to work, versus 81% of controls.²⁷ The degree of impairment correlated with cortical atrophy visible on brain imaging and with age, not with depression or fatigue, suggesting it was a genuine neurological consequence rather than a psychological one.

A systematic review confirmed that the tumor itself initially causes cognitive problems, and that chemotherapy tends to improve cognition at first by shrinking the disease. In the longer term, though, patients who received both chemotherapy and whole-brain radiation had worse cognitive outcomes and lower quality of life scores than those treated with chemotherapy alone.²⁸ Survivors treated with radiation had significantly lower scores in attention, executive function, and motor skills, and their white matter abnormalities on MRI were more than double those of patients who had not received radiation.²⁹ This trade-off between tumor control and long-term brain function is one of the defining tensions in CNS lymphoma care. It is also a major reason the field has moved toward chemotherapy-only strategies and newer consolidation approaches like stem cell transplant where possible.

Liquid Biopsies and Early Detection

One area generating real optimism is the use of “liquid biopsies” of the cerebrospinal fluid. Instead of looking for intact cancer cells, which are often absent, researchers are now testing for fragments of tumor DNA floating freely in the spinal fluid. These cell-free circulating tumor DNA (ctDNA) fragments can pick up specific genetic fingerprints of CNS lymphoma, such as MYD88 mutations, and may serve as a complementary diagnostic marker.³⁰

More intriguingly, ctDNA levels in the spinal fluid appear to track treatment response in real time: they fall when therapy is working and rise when the tumor returns. In at least two documented cases, ctDNA detected impending CNS relapse three to eight months before it was visible on MRI or conventional spinal fluid analysis.³¹ If validated in larger studies, this kind of early warning system could allow doctors to intervene before a relapse becomes clinically apparent, potentially changing outcomes for high-risk patients. For now, it remains a research tool rather than a standard clinical test, but the science is moving fast.

Children Versus Adults

Primary CNS lymphoma in children and adolescents is extremely rare. A German registry study identified just 22 pediatric patients over a 25-year span. Despite the rarity, the outcomes were strikingly better than in adults. The five-year event-free survival was about 86%, and overall survival was about 96%, with only two patients relapsing (both within the brain).³² These children were treated using protocols designed for systemic lymphoma rather than the high-dose methotrexate regimens used in adults, suggesting that the biology of pediatric CNS lymphoma may be inherently less aggressive. The numbers are too small to draw firm conclusions, but the pattern is encouraging and distinct from the adult experience, where outcomes have historically been much worse.