Central nervous system (CNS) relapse occurs in roughly one in 20 patients with diffuse large B-cell lymphoma (DLBCL), but in some disease subtypes that risk increases substantially.
“In certain subsets of patients, the risk is quite low, but in others it can be quite high so it is better to focus on the high-risk groups,” according to Joshua Brody, MD, of the Tisch Cancer Institute at Mount Sinai in New York City.
Groups at higher risk for CNS relapse — in some cases up to 50% — include those with double-hit or triple-hit lymphoma and patients with adrenal/kidney, testes, or breast involvement.
There have been efforts to stratify the risk for secondary CNS relapse based on clinical presentation using the CNS International Prognostic Index (IPI), explained Jakub Svoboda, MD, of Perelman School of Medicine at the University of Pennsylvania in Philadelphia. The CNS IPI calculates risk based on factors including the number of extranodal sites involved, IDH status, age, stage, and performance status.
Brody said the CNS IPI is “not perfect, but it can stratify folks into groups where risk is less than 5% or up to 25%.”
Currently, not every patient with DLBCL will be screened for CNS involvement, but those patients considered to be at high-risk should be screening.
“Any patient with aggressive lymphoma at diagnosis and subsequently who has unexplained neurological symptoms such as new headaches, double vision, or other deficits needs to undergo work-up for CNS involvement urgently,” Svoboda said. “This usually includes brain imaging and spinal tap for CSF [cerebral spinal fluid] analysis and possibly an eye exam. For patients with some high-risk features like testicular involvement, I would screen for CNS involvement even without any symptoms.”
Things get “controversial” and “frustrating,” when it comes to the use of prophylactic treatment for CNS relapse, according to the experts.
In testicular cancer, prophylactic treatment for CNS involvement has been standard of care for decades, Brody noted, and patients will typically undergo prophylactic treatment with intrathecal (IT) or IV methotrexate.
“High-dose methotrexate is in some ways gentler than other chemotherapy that we give, but for people with bad kidneys or those above a certain age, it can be tough on the kidneys and is associated with other toxicities,” he explained. “It is not uncommon with a 75-year-old [patient] to try to give three cycles of a high-dose methotrexate, and have the patient unable to get through it.”
Use of prophylactic therapy in other intermediate- or high-risk groups is more controversial, he said, adding that it has been an open debate for decades.
“While some studies showed benefit of IT chemotherapy or IV methotrexate, there are now several other studies showing no clear benefit,” Svoboda noted. “Since the downside of using prophylaxis is relatively low, I tend to use it in high-risk patients.”
Once a patient is diagnosed with CNS relapse, management is usually tailored to the specific situation and will depend on several factors: Is CNS the only site of relapse? Is the patient having concurrent progression systemically? Is this early on or even at diagnosis or in someone who already had many prior therapies?
“The main issue is that most of our conventional chemotherapy agents do not easily penetrate through the blood-brain barrier and, therefore, we need to use agents like IV methotrexate or cytarabine, which have various toxicities and may not be very effective systemically,” Svoboda said. “Finally, radiation is effective in treating lymphoma including CNS involvement, but the toxicities including cognitive decline are of concern to providers and patients.”
All of these unknowns point to the fact that many questions remain related to both the prophylaxis and management of CNS relapse of DLBCL.
Svoboda said that more work is needed to define which patients with DLBCL would benefit from prophylaxis, and to determine the best prophylactic strategy.
“Ultimately, we hope to learn more about the heterogeneity of the DLBCL and why some of these malignant lymphocytes travel to sites like [the] CNS and some do not,” he said. In terms of treatment, Svoboda said, it is encouraging that novel agents like Bruton’s tyrosine kinase inhibitors or chimeric antigen receptor (CAR) T-cell therapy, are effective in both the CNS and the rest of the body. “Future treatment strategies of CNS involvement may rely more on these types of products.”
Brody noted another exciting area for future research “is the ability to use next-generation sequencing to better risk stratify people.”
A small 2021 study in Blood Advances used a next-generation sequencing-based assay to try to detect tumor-derived DNA in the CSF as a marker of CNS invasion. The assay detected clonotypic DNA in 100% of the samples with known CNS involvement and in 36% of newly diagnosed aggressive lymphomas.
“Using a simple test, if we did see CNS involvement we would no longer call the treatment ‘prophylaxis,’ but would instead be seeing lymphoma and treating it,” Brody said.
Brody disclosed relationships with Gilead/Kite, Merck, SeaGen, Roche/Genentech, ADC Therapeutics, Epizyme, AstraZeneca, and Bristol Myers Squibb.
Svoboda disclosed relationshiops with, and/or support from, Adaptive, ADC Therapeutics, AstraZeneca, Atara, Bristol Myers Squibb, Genmab, Incyte, Merck, Pharmacyclics, SeaGen, and TG Therapeutics.