Rituximab maintenance after bendamustine-based treatment for follicular lymphoma and mantle cell lymphoma may exert a negative influence on SARS-CoV-2 infection outcomes

Different chemoimmunotherapy strategies have been employed as first-line treatment for patients diagnosed with mantle cell lymphoma (MCL) and follicular lymphoma (FL). Cyclophosphamide-based regimens (RCVP/RCHOP) or bendamustine-containing protocols followed by the option of maintenance treatment with an anti-CD20 agent, are the approaches most commonly preferred. We analyzed outcomes of SARS-CoV-2 infection in 215 patients diagnosed with FL or MCL treated with chemoimmunotherapy and subsequent rituximab maintenance in six tertiary Spanish centers. Of note, temporary interruptions or dose delays during maintenance due to SARS-CoV-2 infection were documented in 44% of patients, with definitive suspension of treatment in 22% of patients. Patients receiving maintenance treatment after bendamustine-based regimens presented inferior SARS-CoV-2 infection outcomes compared to patients in maintenance after cyclophosphamide-containing regimens. The former cohort presented higher rates of severe disease, increased hospitalizations, and mortality related to the SARS-CoV-2 infection, leading to a shorter overall survival (OS), compared to the cyclophosphamide cohort.

R-CHOP/R-CVP treatment strategies have been widely employed as first-line immunochemotherapy regimens for indolent B-cell non-Hodgkin lymphomas (NHL).¹ While two randomized clinical trials (Study group indolent Lymphomas, StiL, and BRIGHT) demonstrated enhanced progression-free survival (PFS) and reduced toxicity with bendamustine-rituximab (BR) compared with RCHOP/RCVP,^2,3 “real-world” evaluations of the BR approach have shown a trend for increased hospitalizations and infections, albeit without an impact on OS.⁴ Moreover, in the GALLIUM study, evaluating the combination of rituximab or obinutuzumab with CHOP, CVP or bendamustine for previously untreated FL patients, a higher rate of infections was reported in the bendamustine arm compared to the CHOP/CVP arm. Also, there was a higher rate of infections in the bendamustine arm, independent of the combination with obinutuzumab or rituximab, and this was particularly evident during the maintenance and follow-up phases. Nevertheless, CHOP was associated with higher rates of grade 3 neutropenia during the induction.⁵ Rituximab maintenance (RM) after front-line treatment has significantly improved PFS in FL,⁶ and both PFS and OS in MCL.⁷, ⁸ However, the consensus on the use of RM following front-line BR remains elusive, primarily due to the absence of randomized data demonstrating a clear benefit for RM in this particular setting. While retrospective data suggest the potential safety of RM after BR,⁹ it has not been systematically evaluated in the context of the SARS-CoV-2 pandemic. Published reports have shown that COVID-19 disease presents a high mortality rate in immunocompromised patients, including patients with an active hematologic disease.^10,11 This is particularly evident in individuals diagnosed with B-cell NHL treated with immunochemotherapy (ICT) including anti-CD20 monoclonal antibodies (MoAb). In addition, the administration of anti-CD20 MoAb deeply diminishes the seroconversion rate after vaccination.^12-15 The aims of our study were to investigate the incidence and severity of SARS-CoV-2 infection and the seroconversion rate in patients diagnosed with FL and MCL who were undergoing maintenance after first-line immunochemotherapy treatment based on cyclophosphamide or bendamustine.

In this retrospective analysis, we included all patients diagnosed with FL and MCL who received upfront RCHOP/ RCVP or BR and sequential RM between March 2020 and March 2022 at six Spanish centers. Patients included in the study had initiated maintenance therapy during the study period or were already on maintenance by March 2020. This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Clinical Research Ethics Committee of the Vall d’Hebron Institute (study number: PR(AG)179/2022). An inverse probability of treatment weighting (IPTW) average treatment effect analysis was performed according to the type of first-line treatment (bendamustine or cyclophosphamide-containing regimens) to adjust for potential imbalances in other prognostic variables between both groups in binary outcomes (Figure 1B) and SARS-CoV-2 survival (Figure 2B). A standardized mean difference (SMD) with a threshold of 0.10 was used to assess the balance of co-variates between the two groups. COVID-19 disease events were collected from the beginning of RM while COVID-19 disease events before lymphoma onset and during induction therapy were excluded from the analysis. Survival of patients diagnosed with SARS-CoV-2 infection was calculated from the beginning of RM until death by SARS-CoV-2 or last follow-up. The severity of SARS-CoV-2 infection was graded according to the need for either hospitalization or admission to an intensive care unit (ICU). The seroconversion rate was evaluated based on the positivity of spike glycoprotein antibody titers after vaccination within the entire cohort. All statistical analyses were performed using R software version 4.2.2.

Figure 1.Impact on SARS-CoV-2 infection according to the first-line treatment administered: bendamustine-based versus cyclophosphamide-based strategies. (A) Differences according to SARS-CoV-2 infection outcomes of the patients from the total cohort. (B) Differences according to SARS-CoV-2 infection outcomes of the patients from the total cohort using a propensity analysis based on inverse probability of treatment weighting (IPTW). BR + RM: bendamustine plus rituximab and rituximab maintenance; RCHOP/RCVP + RM: RCHOP/RCVP and rituximab maintenance; ICU: intensive care unit.

The full population included 215 patients: 178 (83%) with FL and 37 (17%) with MCL. Baseline characteristics were analyzed according to the first-line treatment they received (Table 1). In the FL group, 14 (7%) patients were treated with BR induction while 164 (76%) patients had received cyclophosphamide-containing regimens. The MCL cohort included 6 (3%) patients treated with bendamustine and 31 (14%) patients with cyclophosphamide regimens. Median age for the full cohort was 59 years (interquartile range [IQR], 52-68), without significant differences between groups (62 years bendamustine vs. 59 years cyclophosphamide). In the cyclophosphamide group, 8 (4%) patients received the RCVP regimen (2 MCL and 6 FL) due to cardiac comorbidities while 187 (96%) received RCHOP. Patients treated with BR received this treatment by the center’s choice, and only one patient was given this regimen because of cardiac comorbidities. Response to the first-line of treatment was similar between cohorts, with a complete response (CR) rate after induction of 80% versus 77% for bendamustine and cyclophosphamide, respectively. Au tologous stem cell transplantation (ASCT) was performed in 23 patients (11%): 3 patients with FL and 20 patients with MCL. Following the IPTW analysis, baseline variables, including those associated with an impact on SARS-CoV-2 infection outcomes, such as age, vaccination status, and prognostic score (Follicular Lymphoma International Prognostic Index [FLIPI] or Mantle Cell Lymphoma International Prognostic Index [MIPI]) were similar in both groups (SMD <0.1). Response after maintenance was similar between cohorts, with an overall response rate of 100% versus 84% for bendamustine and cyclophosphamide, respectively (P=0.654). Temporary interruptions or dose delays during maintenance due to SARS-CoV-2 infection were reported in 95 (44%) cases and definitive suspensions of treatment in 47 (22%) patients. No significant differences in the rate of maintenance delays (55% bendamustine vs. 43% cyclophosphamide, P=0.24) or definitive interruptions (30% bendamustine vs. 21% cyclophosphamide, P=0.204) were observed between groups. SARS-CoV-2 infection was reported in 77 (36%) patients in the full patient population, with a higher rate of infection in the bendamustine group compared to the cyclophosphamide group (60% vs. 33%, P=0.026). Thirty-five (16%) patients were hospitalized due to severe COVID-19 disease, and 9 (4%) patients required ICU admission, with higher rates of both endpoints in the bendamustine group, compared to the cyclophosphamide group: 53% versus 15% hospitalization episodes, respectively (P<0.001) and 26% versus 2% of ICU admission (P<0.001) (Figure 1A). These results were also maintained in the IPTW analysis: 50% versus 15% hospitalization episodes (P=0.003) and 26% versus 2% of ICU admission (P<0.001) (Figure 1B). The severity of SARS-CoV-2 infection was evaluated across different years during the study period, and a sensitivity analysis on the FL cohort was performed (Online Supplementary Table S1). In terms of previous vaccination status, 195 (91%) patients had received at least one dose of SARS-CoV-2 vaccine and 141 (66%) patients had received 3 or more doses; mRNA-1273 SARS-CoV-2 was the vaccine most frequently used. No differences in vaccination status were observed between groups in the raw data (85% bendamustine vs. 92% cyclophosphamide, P=0.397) or IPTW analysis (87% bendamustine vs. 92% cyclophosphamide, P=0.482). The seroconversion rate was 22% in the overall cohort, with no differences according to the type of first-line treatment (35% bendamustine vs. 21% cyclophosphamide, P=0.156); these results were confirmed in the IPTW analysis (32% bendamustine vs. 21% cyclophosphamide, P=0.315) (Figure 1A, B). OS according to the induction therapy was determined for the overall cohort. A shorter survival was observed in the bendamustine group, compared to the cyclophosphamide group (HR: 0.2, 95% CI: 0.08-0.56, P=0.0019). Regarding the OS of patients diagnosed with SARS-CoV-2 infection, a shorter survival was observed in the bendamustine group compared to the cyclophosphamide group (HR: 0.08, 95% CI: 0.02-0.31, P<0.001) (Figure 2A); these results were confirmed in the IPTW analysis (HR: 0.11, 95% CI: 0.03-0.41, P=0.001) (Figure 2B). With a median follow-up of 20.9 months, 36 (17%) patients had relapsed (31 FL and 5 MCL) and 16 patients had died. Regarding patients who received the RCVP regimen due to significant cardiac comorbidities, no deaths due to SARS-CoV-2 infection were observed. Among the causes of death, 9 were due to SARS-CoV-2 infection, with significant differences between the bendamustine and cyclophosphamide groups (25% vs. 2%, respectively, P<0.001). In our series, the COVID-19 pandemic significantly affected therapy outcomes in patients diagnosed with FL and MCL, causing treatment delays in 44% and suspensions in 22% of patients receiving rituximab as maintenance treatment. This modification in the maintenance regimen due to SARS-CoV-2 infection could have an impact on long-term disease control. The previous induction regimen seemed to carry a significant impact on the SARS-CoV-2 infection outcomes, with higher rates of severe disease, hospitalization, ICU admission, and death due to COVID-19 disease in the bendamustine group. The depletion and impaired cellular function of CD4⁺ T cells might have been the cause behind the differences observed between induction therapies, with probably a longer impact of T-cell suppression in the bendamustine group.

Table 1.Main characteristics of the overall cohort of patients included in the study and an evaluation of these variables using a propensity score analysis based on inverse probability of treatment weighting.

Figure 2.Survival of patients diagnosed with SARS-CoV-2 infection. (A) Survival of the overall cohort of patients included in the study diagnosed with SARS-CoV-2 infection. (B) Survival of patients diagnosed with SARS-CoV-2 infection using a propensity analysis based on inverse probability of treatment weighting. BR + RM: bendamustine plus rituximab and rituximab maintenance; RCHOP/RCVP + RM: RCHOP/RCVP and rituximab maintenance; adj: adjusted; N: number.

The limitations of this report are based on the retrospective nature of data collection. The seroconversion rate was assessed after vaccination but no data regarding antibody status at the time of SARS-CoV-2 infection were available. Moreover, given the lack of a cohort receiving BR induction without sequential RM, we were unable to confirm if the negative impact observed on SARS-CoV-2 infection outcomes in the BR+RM group was due to the bendamustine induction itself or the RM in this particular setting.

In summary, in the context of the COVID-19 pandemic, cyclophosphamide-based regimens coupled with rituximab maintenance appeared to be better tolerated than bendamustine-rituximab followed by maintenance with this MoAb. Immunocompromised patients with hematologic malignancies should follow the SARS-CoV-2 vaccination recommendations very closely to reduce the risk of severe infections.

Footnotes

• Received February 5, 2024
• Accepted July 2, 2024

Correspondence

P. ABRISQUETA - pabrisqueta@vhio.net

Disclosures

ASe has received honoraria from Janssen, Incyte, Roche, and AstraZeneca. GI has received honoraria from BMS/Celgene, Novartis, Janssen, AstraZeneca, AbbVie, Miltenyi, and Roche. J-MS has received honoraria from Roche, Gilead, Janssen, Celgene, Servier, Novartis, Mundipharma, Sanofi, and Kern-Pharma, and has held an advisory role for Roche, Janssen, Gilead, Celgene, Bristol Myers, and Celltrion. EG-B has received lecture fees and advisory board fees from Janssen, AbbVie, Gilead, Kiowa, EUSAPharma, Incyte, Lilly, Beigene, Novartis, Takeda, and Roche. AL-G has held an advisory role and has been a speaker for Beigene, Roche, Janssen, AstraZeneca, and AbbVie, and has received travel grants from these companies. RC has received research funding from Pfizer, has held an advisory role for Kyowa-Kirin, Roche, AbbVie, ADC Therapeutics, Bristol-Myers Squibb, Incyte, Takeda, Janssen, and Kite, and has sat on the Speakers Bureau for Roche, AbbVie, Bristol-Myers Squibb, Takeda, Janssen, and Kite. TG has held a consultancy or advisory role from Janssen, Roche, Gilead, and Celgene, and has received research funding from Janssen and AbbVie. AC has received honoraria from Jazz, AstraZeneca, and Janssen. MJ has received honoraria from GSK and Sanofi. AM-N has held a consultancy or advisory role for Lilly, Kite, Kyowa-Kirin, Takeda, AstraZeneca, Roche, and Janssen, and has received honoraria from Lilly, Kite, Takeda, and Janssen. FB has held a consultancy or advisory role for Novartis, Celgene/BMS, Roche, AstraZeneca, Takeda, Mundipharma, Gilead, AbbVie, Lilly, and Beigene, and has received honoraria from Novartis, Celgene/BMS, Roche, AstraZeneca, Takeda, Karyospharm, Mundipharma, Gilead, AbbVie, Lilly, and Beigene. PA has held a consultancy or advisory role for Roche, AbbVie, BMS, AstraZeneca, Janssen, and Incyte, and has received honoraria from Roche, AbbVie, BMS, AstraZeneca, Janssen, and Sandoz. All the other authors have no conflicts of interest to disclose.