Impact on Tumour Treatment and Outcome of Sars-Cov-2 Infection in 91 Patients with Primary CNS Lymphoma: A Real-Life Study of the International PCNSL Collaborative Group

Introduction. A number of recent studies have described outcomes of patients (pts) with hematological malignancies affected by SARS-CoV-2, however data according to a particular tumour type and its specific treatments are lacking. Aggressive non-Hodgkin lymphomas for which dose intensity is crucial, represent one of the most common and vulnerable populations infected by SARS-CoV2. Herein, we report data from an international real world study of pts with primary central nervous system lymphoma (PCNSL) and concurrent SARS-CoV-2 infection.
Methods. Key data on clinical presentation, management and outcome of pts were collected and analyzed to determine the impact of SARS-CoV-2 infection on the delivery of anti-lymphoma treatment and overall outcome. SARS-CoV-2 infection was defined by naso-pharyngeal swab or by broncho-alveolar lavage. Pts were grouped in 1st, 2nd and 3rd pandemic waves using July 31, 2020 and January 1, 2021 as cut-offs for SARS-CoV-2 diagnosis.
Results. Ninety-one pts from 27 centers of 5 countries (France, Israel, Italy, United Kingdom, USA) were registered. The enrolled cohort was dominantly in the pre-vaccination era, but subsequently included 16 vaccinated pts. SARS-CoV-2 was diagnosed before/during 1st-line PCNSL treatment in 64 (70%), during follow-up in 21 (23%), and during salvage therapy in 6 (7%) pts. Thirty-eight (59%) of the 64 pts infected before/during first-line developed pneumonia; this complication was more common if (1) non-vaccinated, (2) steroids before viral infection for at least two weeks or with a cumulative dose >100 mg of dexamethasone, and (3) high-dose cytarabine before virus detection. Eighteen (47%) pts with pneumonia cleared the virus; 15 of whom resumed anti-lymphoma treatment, with a median delay of 30 days, and 17 were alive at the last visit. Seventeen of the 20 pts with pneumonia who did not clear virus died early (<30 days) of COVID-19 or related infections. Twenty (77%) of the 26 pts without pneumonia cleared virus and resumed/initiated first line therapy, with a median delay of 22 days; 15 of whom were alive at the last follow up. Overall, 43 (67%) pts initiated/resumed or completed first-line treatment, with a median delay of 22 (range 0-116) days. Resumption of anti-lymphoma treatment was more common among those who did not develop pneumonia, cleared the virus and/or did not receive steroids during infection. Resumption of chemo despite viral persistence was associated with a poorer survival, with a 6-month OS of 70% (95%CI= 67-73%) for the 23 pts who initiated/resumed chemo despite viral persistence and 87% (95%CI= 86-87%) for the 20 pts who waited the virus clearance (p= 0.07). Eight (38%) of the 21 pts infected during lymphoma follow-up developed pneumonia; all cleared virus and 20 of 21 were alive at last follow-up. Conversely, 4 (67%) of the 6 pts with relapsed PCNSL infected during salvage therapy died of COVID-19 or related infections.
At a median follow-up since virus detection of 185 days (18-534), 61 pts cleared virus and 59 (65%) are alive, with tumor remission in 34. The 1- and 6-month OS were 80% (95% CI= 79-81%) and 66% (95% CI= 64-68%), respectively; virus persistence and pre-COVID-19 steroid therapy were independently associated with poor OS. Mortality during the 3 considered waves was not significantly different. Notably, however, deaths in the 3rd wave occurred exclusively in non-vaccinated pts (p= 0.0009). Moreover, overall outcome of the 16 pts infected after November 2021, when Omicron variant became prevalent appears favourable: 14 of these patients were vaccinated, all of them are alive and resumed/completed chemotherapy with a median delay of 3 days (range 0-28); there were only 2 COVID-19-related deaths, which occurred in the 2 non-vaccinated pts. Finally, this study did not show a significant association between comorbidities and COVID-19-related mortality.
Conclusions. COVID-19 was a strong outcome-defining event, especially in pts receiving anti-PCNSL treatment and in those who had received a cumulatively high steroid dose before viral infection. Vaccination was associated with a lower incidence of pneumonia and in-hospital mortality. Chemo initiation or resumption during active infection should be indicated cautiously. For pts in follow up, SARS-CoV-2 was not associated with severe symptoms and did not affect OS. These data will inform clinical management of pts with PCNSL in the context of SARS-CoV-2 infection.