Does High-Dose Methotrexate Protect the Central Nervous System in High-Risk DLBCL?

A retrospective study found that treatment with high-dose methotrexate did not lower the risk of central nervous system (CNS) relapse in patients with high-risk diffuse large B cell lymphoma (DLBCL), bringing into question whether current recommendations for prophylactic intravenous high-dose methotrexate are appropriate for all patients at high risk of a CNS relapse. Results of the study were presented during the 2020 ASH Annual Meeting by Robert Puckrin, MD, of the University of Calgary in Canada.

Although intravenous high-dose methotrexate is the most widely adopted method for CNS relapse prevention, Dr. Puckrin noted that there are no randomized data supporting its use.
In this study, researchers retrospectively reviewed charts of adults with DLBCL who were treated with curative intent at two academic medical centers in Alberta, Canada, between 2012 and 2019. During this time, international guidelines recommended high-dose methotrexate for patients at high risk of CNS relapse, including those with CNS international prognostic index (IPI) scores of 4 to 6, double hit lymphoma, testicular involvement, elevated lactate dehydrogenase, Eastern Cooperative Oncology Group performance status score greater than 1, and more than one extranodal site of involvement.

The present study included patients who met criteria for high-risk disease and could be offered higher-intensity chemoimmunotherapy or R-CHOP (rituximab, cyclophosphamide, doxorubicin hydrochloride, vincristine sulfate, and prednisone) followed by consolidative autologous hematopoietic cell transplantation (AHCT).

In total, the charts of 906 patients were included, which covered a median follow-up duration of 35.3 months (range = 0.29-105.7 months). The median CNS-IPI score was 3, and 36% of patients met Alberta Lymphoma Guideline (ALG) criteria for high-risk disease.
Approximately one-third of patients with high-risk disease (n=115; 35.3%) received prophylactic high-dose methotrexate, for a median of two doses. Those who received methotrexate were more likely to be younger, have kidney/adrenal involvement, or have double-hit lymphoma.

Of the total patient population, 44 (6.1%) developed CNS relapse, and those with higher CNS-IPI scores were at a significantly greater risk, compared with those in the intermediate- and low-risk categories: 12.2% vs. 4.9% vs. 1.9% (p<0.001). “CNS relapse appeared to occur early, at a median of 7.4 months from diagnosis,” Dr. Puckrin reported, “and outcomes were poor, with a median survival from time of CNS relapse of only seven months.”

Factors included in the ALG system were found to be the strongest predictors of CNS relapse, followed by CNS-IPI:

  • ALG high-risk criteria: 11.8% vs. 3.0% (hazard ratio [HR] = 4.7; p<0.001)
  • CNS-IPI 4-6: 12.2% vs. 3.5% (HR=4.3; p<0.001)
  • testicular involvement: 20.6% vs. 5.7% (HR=3.5; p=0.039)
  • double-hit lymphoma: 9.3% vs. 5.8% (HR=2.1; p=0.084)

Looking at the effect of prophylactic high-dose methotrexate, the researchers found that the incidence of CNS relapse was 11.2% in the methotrexate-treated group, compared with 12.2% for patients who did not receive high-dose methotrexate (HR=0.92; p=0.82), suggesting that prophylactic methotrexate does not significantly reduce CNS relapse risk in this population.

In multivariate analysis, the researchers observed that patients who received higher-intensity chemoimmunotherapy (n=35) or consolidative AHCT (n=68) had a lower risk of CNS relapse compared with patients who received conventional R-CHOP (HRs=0.38 and 0.30, respectively), but the difference was not statistically significant.

They also found that treatment with high-dose methotrexate and higher-intensity chemoimmunotherapy showed no association with CNS relapse, progression-free survival, or overall survival.

The findings are limited by the study’s retrospective, nonrandomized design.

The authors report no relevant conflicts of interest.


Puckrin R, Darsa HE, Ghosh S, Peters A, and Stewart DA. Lack of effectiveness of intravenous high-dose methotrexate for prevention of CNS relapse in patients with high-risk DLBCL: A retrospective analysis from Alberta, Canada. Abstract #477. Presented at the 2020 American Society of Hematology Annual Meeting, December 6, 2020.

For many years, it has been appreciated that more is not always better in the frontline treatment of DLBCL. At the 2020 ASH Annual Meeting, the spotlight focused on whether the same applies for intravenous high-dose methotrexate in preventing CNS recurrence – an uncommon but devastating complication of DLBCL. Dr. Puckrin and colleagues performed a retrospective analysis of more than 900 patients treated in two academic institutions in Alberta, Canada, 326 of whom had high risk for CNS recurrence. Of those 326, approximately one third received at least one of three recommended doses of intravenous high-dose methotrexate (3.5 g/m2). In those high-risk patients who received high-dose methotrexate, the CNS relapse risk was 11.2% versus 12.2% in those that did not. On multivariate analysis, high dose methotrexate was not associated with a lower risk of CNS. Interestingly, consolidative autologous stem cell transplantation was of borderline significance, suggesting that effective systemic control maybe important.

Where does this leave us? The removal of an intervention aimed at reducing the risk of a feared complication is difficult to do without high-quality data. A retrospective analysis is, by its nature, prone to bias as the patient groups will be different. While a prospective randomized study in this setting would be ideal, the rare nature of CNS recurrence events makes this very challenging. In the absence of better-quality data, we can either stop using high-dose methotrexate for high-risk patients, we can use more or different CNS prophylactic strategies, or we can continue with the current practice. I suspect most will go with the latter option.

Graham Collins, MBBS, FRCP, FRCPath, DPhil
Oxford University Hospitals NHS Foundation Trust
Oxford, U.K.