No Difference in CNS Relapse Rate According to Prophylaxis Route

According to results from a retrospective analysis of central nervous system (CNS) prophylaxis in patients with aggressive non-Hodgkin lymphomas (NHLs), rates of CNS relapse did not differ according to intrathecal (IT) or intravenous (IV) administration. The analysis, presented at the 2020 ASH Annual Meeting by Victor M. Orellana-Noia, MD, from Emory University in Atlanta, Georgia, also highlighted factors predictive of CNS relapse that are missing from conventional scoring systems.

These results accompanied other data presented at the meeting that demonstrated that high-dose methotrexate did not lower the risk of CNS relapse in high-risk diffuse large B-cell lymphoma, despite current recommendations for prophylactic IV high-dose methotrexate in this population.

To determine whether route of prophylaxis affected CNS relapse, Dr. Orellana-Noia and colleagues performed a multicenter retrospective analysis of patients with aggressive NHL (excluding Burkitt lymphoma) who received single-route CNS prophylaxis during frontline anthracycline-based therapy between 2013 and 2019 across 19 U.S. academic institutions.

The authors identified a total of 1,056 patients who received CNS prophylaxis. Twenty-eight were excluded after they switched routes due to toxicity, leaving 1,024 patients in the analysis: 216 who received IV prophylaxis and 808 who received IT prophylaxis.

Patients’ median ages were 57 years in the IV cohort and 59 years in the IT cohort. Incidence of renal impairment was slightly lower in the IV group (12% vs. 16%), while incidence of double-hit lymphoma was slightly higher in the IT group (8.3% vs. 16%).

The CNS International Prognostic Index (IPI) predicted a relapse rate of 6.13% across the entire population, however, the observed CNS relapse rate was 5.5%, regardless of type of prophylaxis (5.3% in the IT group vs. 7.1% in the IV group; p=0.18).

This lack of difference persisted across all subgroups, including age, stage, CNS-IPI score, and double-hit status, the researchers noted.

Looking at the predictive value of the six CNS-IPI components, only two [elevated serum lactate dehydrogenase (LDH) and number of extranodal sites involved] were significantly associated with CNS relapse. In addition, while CNS-IPI discriminated CNS relapse risk among those with high versus moderate risk, it was a poor predictor for some groups with low risk (TABLE). “Our low-risk group had a significantly higher observed CNS relapse risk, which was fascinating to see,” Dr. Orellana-Noia reported.

Based on these data, the researchers were able to identify four factors that were significantly associated with higher CNS relapse risk in univariate analysis: testicular involvement, elevated LDH, high extranodal disease burden, and liver involvement.

In a multivariate model, only testicular involvement and elevated serum LDH continued to predict for CNS relapse:

  • testicular involvement: 5.38 (p=0.067)
  • LDH > upper limit of normal: 2.70 (p=0.018)
  • >1 extranodal site: 1.39 (p=0.32)
  • liver involvement: 1.95 (p=0.067)

However, Dr. Orellana-Noia noted that median CNS-IPI scores were higher among patients with liver involvement, suggesting that the CNS risk with liver involvement may already be captured by conventional scoring systems. “In contrast, those with testicular involvement had, on average, lower CNS-IPI scores, [which] appears to show a shortcoming in the models we are using to risk-stratify patients,” he said.

Among those who developed CNS relapse despite prophylaxis, there was no significant difference in time to relapse with either administration, but each group had a somewhat longer time to event than historical references, Dr. Orellana-Noia noted. “There are still a number of early CNS relapse events despite prophylaxis … and overall survival remains very poor when relapse occurs (at 7.1 months) with no significant benefit from salvage therapy,” he added.

“CNS relapse is a rare, but devastating complication, [which] highlights the need to optimize and standardize our approach to CNS prophylaxis,” Dr. Orellana-Noia concluded. In that vein, future studies are planned to evaluate single- versus dual-route of administration, as well as the value of prophylaxis versus no prophylaxis.

The authors report no relevant conflicts of interest.


Orellana-Noia VM, Reed DR, Sen JM, et al. CNS prophylaxis during front-line therapy in aggressive non-Hodgkin lymphomas: real-world outcomes and practice patterns from 19 US academic institutions. Abstract #478. Presented at the 2020 American Society of Hematology Annual Meeting, December 6, 2020.