The largest study yet to examine high-dose methotrexate in aggressive lymphoma has found it does not prevent brain relapse, even in the highest-risk patients.

First author Dr Matthew Wilson, Beatson West of Scotland Cancer Centre.

Researchers combined data from two international retrospective analyses to produce a cohort of 1,923 patients with large B-cell lymphoma (LBCL) and ultra high-risk features treated with and without the controversial prophylaxis.

Findings revealed no significant reduction in CNS relapse with high-dose methotrexate (HD-MTX) across all ultra high-risk patients and subgroups, backing up research that failed to show a benefit in the wider LBCL population.

Writing in the Journal of Clinical Oncology [link here], the researchers, including leading Australian haematologists and oncologists such as Professor Chan Cheah, Professor Michael Dickinson and Professor Eliza Hawkes, pointed out that many clinicians had already significantly reduced or stopped HD-MTX prophylaxis. But analyses were underpowered for ultra high-risk patients and international guidelines continued to recommend consideration of the drug, they said.

All patients in their study had been treated with R-CHOP or R-CHOP-like therapy.

“Ultra high risk” was defined as having at least one of the following criteria:

• CNS international prognostic index (IPI) score 5-6
• Involvement of ≥3 extranodal sites
• Testicular, renal/adrenal, or breast involvement

The researchers also performed a 6-month landmark analysis to mitigate potential immortality bias favouring the HD-MTX group, which included patients alive, progression free, and in partial or complete response to first-line chemotherapy.

What the study found

• No significant difference in the 3-year CNS relapse rate between no HD-MTX (n = 1,051) and HD-MTX patients (n = 872) in multivariable analyses (3-year rate 9.3% vs 8.1%; adjusted hazard ratio [aHR], 1.13 [95% CI, 0.82 to 1.57]).
• This finding was confirmed in analyses restricted to isolated CNS relapse (5.9% vs 5.7%; aHR, 1.03 [95% CI, 0.69 to 1.53]).
• No difference in 3-year CNS relapse in the landmark analysis (no HD-MTX n = 782, HD-MTX n = 773) (6.7% v 6.6%, aHR, 0.95 [95% CI, 0.62 to 1.44]).
• Patients in the no HD-MTX group had inferior 3-year event-free survival and overall survival, but the differences were less marked in the landmark population and no longer observed following propensity matching.
• No significant difference in CNS relapse rate according to number of HD-MTX cycles or cumulative dose.

The researchers said their findings provided “practice-informing data in an area where no adequately powered prospective or randomised studies guide clinicians”.

“There is an urgent need to design prospective, randomised trials aimed exclusively at ultra high-risk patients, using novel therapies other than HD-MTX,” they said.

Editorial author Professor Jonathan Friedberg.

In an accompanying editorial [link here], US haematologic-oncologist Professor Jonathan Friedberg agreed with the authors’ conclusions, going even further to say the results “should definitively end” the use of HD-MTX as CNS prophylaxis.

“From a therapeutic standpoint, we are now ready to move beyond methotrexate to prevent CNS relapse of DLBCL. BTK inhibitors cross the blood-brain barrier and represent a therapeutic modality with particular efficacy in certain genetically defined subsets of DLBCL,” said Professor Friedberg, director of Wilmot Cancer Institute and editor-in-chief of the journal.

“Immunologic approaches using bispecific T-cell engagers or CAR T-cell treatment are active in CNS DLBCL and offer particular promise as novel strategies for CNS prophylaxis. The outcomes of ongoing large randomised studies incorporating these agents with standard chemoimmunotherapy for de novo DLBCL are eagerly awaited; it will be very interesting to evaluate the rates of CNS progressions in these studies.”