A prospective phase 2 study testing high dose post-transplant cyclophosphamide as sole GHVD prophylaxis after matched allotransplant using baltimore-based reducedintensity conditioning regimens and PBSC as source of graft

Introduction: The use of high-dose post-transplant cyclophosphamide (PTCY) has revolutionized graft-versus-host disease (GVHD) prophylaxis and allowed to successfully reconsider haplotransplant in recent years. As this strategy significantly reduces the incidence of both acute and chronic GVHD, PTCY has been thereafter considered not only in matched settings but also as sole GVHD prophylaxis, at least when considering myeloablative allotransplant using matched sibling (MSD) or unrelated (MUD) donors and bone marrow as source of graft. Here, PTCY, as a sole GVHD prophylaxis, was tested in a reduced-intensity conditioning (RIC) setting, using peripheral blood stem cells (PBSC) as source of graft considering that this platform is currently broadly used worldwide in adults. Methods: This prospective monocentric phase 2 study was designed with the main objective to demonstrate the feasibility and safety of using only PTCY (without cyclosporine A nor mycophenolate mofetyl after transplant) in adults (18-70 years old) eligible for a RIC PBSC transplant with MSD or MUD. The Baltimore platform with 2 days of PTCY 50mg/kg/day on days 3 and 4 post infusion was considered as conditioning regimen, using fludarabine for lymphoid disease or clofarabine for myeloid disease. The primary objective was to appreciate the incidence of corticosteroid-resistant acute grade 3-4 GVHD (CR 3-4 GVHD) within 100 days post-transplant. According to statistical rules, patients have to be included in a step by step fashion (3, 3, 6, 15, 15 and 17 patients) for a total of 59 evaluable patients (meaning having received PTCY), in order to stop the protocol soon enough in case of excessive rate of deleterious severe acute GVHD (graded according to Mount Sinai International Consortium). Thus, the trial had to be stopped in case of documentation of > 2 CR 3-4 GVHD for the first 3 patients, >3 CR 3-4 GVHD for the first 6 patients, > 4 CR 3-4 GVHD for the first 12 patients, > 6 3-4 CR GVHD for the first 27 patients, > 8 CR 3-4 GVHD for the first 42 patients and finally as soon as > 9 CR 3-4 GVHD for the last included patients. All patients gave informed consent. The trial was registered at ClinicalTrials.gov Identifier: NCT03263767. Results: The results of the first 27 first patients (males n=17 and female n=10;median age: 59 years old (yo), range: 26-70) are reported here. They were included between February 2018 and November 2020. Diagnoses were AML (N=8), MDS (N=5), CMML (N=2), myelofibrosis (N=5), CML (N=1), DLBCL (N=1), T-cell lymphoma (N=1), Philadelphia positive B-ALL (N=1), CLL (N=1), lymphoblastic lymphoma (N=1) and mixed phenotype acute leukemia (N=1). Donors were MSD in 10 cases and MUD in 17. Only one primary graft failure was documented in a 61 yo MDS patient with active disease at transplant. He is however still alive in response after autologous reconstitution. With a median follow-up of 17.6 months (range: 10-42) for alive patients at the time of analysis (July 2021), 1-year and 2-year survivals were 80.9+7% and 74.7+9%, respectively, for both OS et DFS. GVHD-free/relapse-free survival (GRFS) at 1-year and 2-year was 58.7+9% and 52.2+10%, respectively. Three relapses (11%) and 6 deaths occurred. Deaths were due to acute GVHD in 4 patients (including 1 with sepsis and 1 with SARS-COVID 19 infection) and relapse in 2. Grade 2, 3 and 4 acute GVHD occurred in 11, 1 and 4 patients, respectively, for a total of 59% of grade 2-4 acute GVHD. CR 3-4 GVHD was observed in all of 5 patients with acute grade 3-4 GVHD and 4 died related to GVHD. Moderate/severe chronic GVHD occurred in 5/22 (22.7%) evaluable patients, including 4 still on immunosuppressive therapy at 40, 28, 25 and 16 months post-transplant. Overall non-relapse mortality (NRM) was 14.8% and related to acute GVHD. However, the number of cases conducting to stop the protocol was not reached. Conclusion: PTCY as a sole GVHD prophylaxis is here demonstrated as possible and relatively safe for adults receiving a matched PBSC Baltimore-based RIC allograft. The very good survivals reported he e may be related to a strong GVL effect associated with the high incidence of acute GVHD. However, because of this high incidence and the fact that NRM was related to GVHD after this first analysis, we have now made an amendment to test the addition to PTCY of one day of anti-thymoglobulin (ATG) 2.5 mg/kg on day-2 for the next 32 patients to be included. This second cohort receiving PTCY+ATG as a sole prophylaxis is ongoing.

Preliminary results of the FILO phase 2 trial for untreated fit patients with intermediate risk chronic lymphocytic leukemia comparing ibrutinib plus venetoclax (IV) vs. FCR

With the emergence of targeted therapies, defining the best strategy for the treatment of previously untreated CLL patients remains challenging. The aim of this phase 2 study was to compare the efficacy of an association with ibrutinib and venetoclax (IV) to the standard FCR regimen in fit patients with intermediate-risk CLL defined by either unmutated IGHV status, 11q deletion, or complex karyotype in the absence of TP53 abnormality. Patients were randomized 1:1 between two treatment arms, i.e. FCR 6 cycles or IV. After a lead-in phase of ibrutinib as a single agent from the month (M)1 to M3, the total duration of treatment with IV was based on the response achieved at M9;if bone marrow (BM) MRD was <0.01% using flow cytometry, the treatment was continued for 6 additional months until M15 and then stopped;if BM MRD at M9 was ≥0.01%, the treatment with IV was continued for 18 additional months until M27. The primary endpoint was the percentage of patients with BM MRD <0.01% at M27 in both arms. We present here the preliminary results on the first evaluation done at M9 including CT-scan, BM biopsy, and MRD assessment in PB and BM after the inclusion of all the 120 patients as initially planned. One hundred and twenty patients were enrolled from September 2019 to February 2021. The median age was 59 [34-72] and 61 [34-74] years in the FCR and IV arms, respectively. The characteristics of the patients were well-balanced between the 2 arms in terms of gender (male 72% FCR, 74% IV), PS ECOG 0-1 (59% FCR, 68% IV), and Binet stage (A, B, and C 15, 64, 21% for FCR;8.5, 59, and 32% for IV). No major difference in terms of cytogenetic features was noted, all patients but one had unmutated IGHV. At the time of data cut-off for this interim analysis, the median follow-up for all cohort was 12.7 [4.5.9-21.4] months. The frequency of patients presenting all grades adverse events (AE) so far was 90% (grade ≥3: 45%) in the FCR arm and 80% (grade ≥3: 45%) in the IV arm. The rate of infusion-related reactions (IRR) in the FCR arm was 35% on cycle 1-day 1 (14% grade 3-4);for the IV arm, 5% of patients experienced tumor lysis syndrome (TLS) (grade 3 for 1 patient). Ibrutinib doses were reduced for seven patients (four permanently stopped and three resumed at a lower dose because of toxicities (digestive, hepatic, or hematological). Venetoclax was permanently discontinued before M9 in four patients (digestive toxicities and grade 4 neutropenia). Fifty-two serious adverse events were reported of which 22 were in the IV arm (among them one sudden death, one ischemic stroke, one acute coronary syndrome, two atrial fibrillations, two TLS, two acute renal failures, one hepatitis, one neutropenia, two COVID pneumonitis, and one osteoporotic fracture) and 30 in the FCR arm (among them five febrile neutropenia, one hemolytic anemia, one thrombocytopenia, three IRR, three TLS, three COVID pneumonitis, one acute myeloid leukemia, one myelodysplasic syndrome). All patients with COVID pneumonitis had a favorable evolution with the need for intensive care and convalescent plasma for three of them. The first 85 patients included in the study have reached M9 and among them, nine prematurely discontinued the study, (one active hemolysis, one ischemic stroke, one TLS, one hepatitis, and one sudden death in the IV arm;three hematologic toxicities and one early progression in the FCR arm). In the evaluated patients (n=74), 69% of patients in the FCR arm and 43% of patients in the IV arm achieved bone BM MRD <0.01%. The complete (CR, CRi) and partial response rates were 56 and 44% in the FCR arm and 74 and 26% in the IV arm, respectively. In conclusion, preliminary results show a lower BM MRD rate in the IV arm compared to the FCR arm at M9, with toxicity that remains significant and relatively similar between the two arms. However, BM MRD rate may improve after longer exposure to the IV combination and the analysis of the primary endpoint at M27 will be decisive in determining the best therapeutic strategy.

Preliminary Results of the Filo Phase 2 Trial for Untreated Fit Patients with Intermediate Risk Chronic Lymphocytic Leukemia Comparing Ibrutinib Plus Venetoclax (IV) Versus FCR

With the emergence of targeted therapies, defining the best strategy for the treatment of previously untreated CLL patients remains challenging. The aim of this phase 2 study was to compare the efficacy of an association with ibrutinib and venetoclax (IV) to the standard FCR regimen in fit patients with intermediate risk CLL defined by either unmutated IGHV status, 11q deletion or complex karyotype in the absence of TP53 abnormality. Patients were randomized 1:1 between two treatment arms, ie FCR 6 cycles or IV. After a lead-in phase of ibrutinib as a single agent from month (M)1 to M3, the total duration of treatment with IV was based on the response achieved at M9;if bone marrow (BM) MRD was < 0.01% using flow cytometry, the treatment was continued for 6 additional months until M15 and then stopped;if BM MRD at M9 was ≥ 0.01%, the treatment with IV was continued for 18 additional months until M27. The primary endpoint was the percentage of patients with BM MRD < 0.01% at M27 in both arms. We present here the preliminary results on the first evaluation done at M9 including CT-scan, BM biopsy and MRD assessment in PB and BM after the inclusion of all the 120 patients as initially planned. One hundred and twenty patients were enrolled from September 2019 to February 2021. The median age was 59 [34-72] and 61 [34-74] years in the FCR and IV arms, respectively. The characteristics of the patients were well balanced between the 2 arms in terms of gender (male 72% FCR, 74% IV), PS ECOG 0-1 (59% FCR, 68% IV) and Binet stage (A, B and C 15%, 64%, 21% for FCR;8.5%, 59% and 32% for IV). No major difference in terms of cytogenetic features was noted, all patients but one had unmutated IGHV. At the time of data cut-off for this interim analysis, the median follow-up for the all cohort was 11 [2.9 - 19.8] months. The frequency of all grades adverse events (AE) observed so far was 53% (grade 3-4, 24%) in the FCR arm and 47% (grade 3-4, 17%) in the IV arm. The rate of infusion-related reactions (IRR) in the FCR arm was 35% on cycle 1-day 1 (14% grade 3-4);for the IV arm, 6% of patients experienced tumor lysis syndrome (TLS) (grade 4 for 4 patients). ibrutinib doses were reduced for 7 patients (4 permanently stopped and 3 resumed at a lower dose because of toxicities (digestive, hepatic or haematological)). Venetoclax was permanently discontinued before M9 in 4 patients (digestive toxicities and grade 4 neutropenia). Forty serious adverse events were reported of which 15 in the IV arm (1 sudden death, 1 ischemic stroke, 2 atrial fibrillations, 2 clinical TLS, 1 hepatitis, 1 neutropenia, 4 COVID pneumonitis and one osteoporotic fracture) and 25 in the FCR arm (2 neutropenias, 1 anemia, 1 thrombocytopenia, 1 autoimmune haemolytic anemia, 3 IRR, 4 TLS, 2 COVID pneumonitis, 4 fever episodes of undetermined origin, 1 community-acquired pneumonia, 1 gastrointestinal toxicity, 1 confusion, 2 chest pains, 1 acute myeloid leukemia, 1 myelodysplasic syndrome). The patients with COVID pneumonitis had a favorable evolution with the need for intensive care and convalescent plasma for 3 of them. The first 60 patients included in the study have reached M9 and among them, 6 prematurely discontinued the study, 3 in each arm (active hemolysis, ischemic stroke and sudden death in the IV arm;2 grade 4 hematologic toxicities and 1 early progression in the FCR arm). In the evaluated patients (n=54), 71% of patients in the FCR arm and 48% of patients in the IV arm achieved bone BM MRD < 0.01%. The complete (CR, CRi) and partial response rates were 54% and 46% in the FCR arm and 76% and 24% in the IV arm respectively. In conclusion, the preliminary results show a lower BM MRD rate in the IV arm compared to the FCR arm at M9, with a toxicity that remains significant and relatively similar between the two arms. However, BM MRD rate should improve after longer exposure to the IV combination and the analysis of the primary endpoint at M27 will be decisive in determining the best therapeutic strategy. Disclosures: Quinquenel: Abbvie: Honoraria;Jansse : Honoraria;AstraZeneca: Honoraria. Laribi: Le Mans Hospital: Research Funding;Novartis: Other: Personal Fees, Research Funding;Takeda: Other: Personal Fees, Research Funding;BeiGene: Other: Personal Fees;IQONE: Other: Personal Fees;AbbVie: Other: Personal Fees, Research Funding;Astellas Phama, Inc.: Other: Personal Fees;AstraZeneca: Other: Personal Fees;Jansen: Research Funding. Cymbalista: Lilly-LOXO: Honoraria, Membership on an entity's Board of Directors or advisory committees;Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees;ASTRA ZENECA: Honoraria, Membership on an entity's Board of Directors or advisory committees;Abbvie: Honoraria, Membership on an entity's Board of Directors or advisory committees;Roche: Honoraria, Membership on an entity's Board of Directors or advisory committees. Leblond: AstraZeneca: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees;Lilly: Consultancy;AbbVie: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel Support;Roche: Honoraria;Amgen: Honoraria;Beigene: Honoraria, Membership on an entity's Board of Directors or advisory committees;Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees. Dartigeas: Astra-Zeneca: Membership on an entity's Board of Directors or advisory committees, Other: travel grants/Congress;Abbvie: Membership on an entity's Board of Directors or advisory committees, Other: travel grants/Congress;Janssen: Membership on an entity's Board of Directors or advisory committees, Other: travel grants/Congress. Ferrant: Janssen: Other: Travel, Accommodations, Expenses;AbbVie: Honoraria, Other: Travel, Accommodations, Expenses;AstraZeneca: Honoraria. de Guibert: Janssen: Consultancy, Honoraria;AbbVie: Consultancy, Honoraria;Gilead: Consultancy, Honoraria. Feugier: Astrazeneca: Consultancy, Honoraria;Gilead: Consultancy, Honoraria;Abbvie: Consultancy, Honoraria;Amgen: Honoraria;Janssen: Consultancy, Honoraria. Cartron: Roche, Celgene-BMS: Consultancy;Danofi, Gilead, Novartis, Jansen, Roche, Celgene-BMS, Abbvie, Takeda: Honoraria. Ysebaert: Abbvie, AstraZeneca, Janssen, Roche: Other: Advisory Board, Research Funding.

Sars-Cov-2 T-Cell Response in Allogeneic Hematopoietic Stem Cell Recipients Following Two Doses of BNT162b2 Vaccine

Introduction: Virus-specific humoral and cellular immune responses act synergistically to protect from viral infection. In our recent observational monocentric study of 117 hematopoietic stem cell adult recipients, we found that 54% and 83 % patients achieved a humoral response after two doses of BNT162b2 anti-SARS-CoV-2 messenger RNA vaccine (Pfizer BioNTech), respectively. Here, we evaluated the T-cell response against the SARS-Cov-2 spike protein after two doses of BNT162b2 vaccine in some allografted patients from the same cohort and compared these results to those from healthy controls. Methods: To quantify SARS-CoV-2 specific T-cells, we used an INFg ELISpot assay that detects these cells after activation of peripheral blood mononuclear cells (PBMC) with 3 peptide pools covering the whole protein sequence of the spike glycoprotein (Prot _S1;_S+ and _S PepTivator peptide pools, Miltenyi Biotec, Bergisch Gladbach, Germany). EBV and CMV specific T-cells were also quantified as controls. The immunophenotype of PBMC was determined by flow cytometry, after dead cell exclusion, with monoclonal antibodies identifying the following surface antigens: CD45, CD3, CD14, CD19 and HLA-DR. The frequencies of spot-forming units (SFU) were reported as per 10 6 CD3+ T-cells. Results: Samples from 46 allografted patients (acute myeloblastic leukemia, N=27, myelodysplastic syndrome, N=19) and 16 healthy controls were available. Characteristics of the population are given in Table 1. All fully vaccinated healthy donors became seropositive and developed a positive T-cell response to spike peptide pools even though variable frequencies were observed. The median response was 195 SFU/10 6 T-cells. By comparison, the frequency of EBV-specific T-cells was 774 SFU/10 6 T-cells (Figure 1). In the group of patients, 78% (n=36/46) had achieved a humoral response after the second dose of vaccine. Among these humoral responders (HR), 89% (n=32/36) also had a positive anti-spike T-cell response with variable frequencies (median =119 SFU/10 6 T-cells. For 8 patients, this T cell response was higher than that of controls (>800 SFU/10 6 T-cells) (Figure 1), which is equivalent to more than 1 specific T-cell per microliter of blood (Figure 2). The humoral responders (HR) who did not develop a T-cell response (11%, n=4/36) had a median time from transplant to vaccination of 523 days compared to 1032 days for cellular responder patients. Among the 10 patients who were non humoral responders (NHR) (22%, n=10/46), 4 (40%) developed a cellular immunity, including one with a very high T cell response (1333 SFU/10 6 T-cells). As expected, the absence of humoral response was observed in patients who were within one year of the transplant. Of note, somehow unexpectedly, patients often presented a high frequency of EBV- and CMV-specific T cells (Figures 1 & 2). As expected, PBMC immunophenotypic analysis revealed that CD3+ frequencies were lower in patients compared to those of controls but were similar between HR and NHR. NHR had very low frequencies of B cells and interestingly, they had an elevated frequency of CD14+ monocytes with low/neg HLA-DR expression potentially corresponding to myeloid-derived suppressor cells (MDSCs) (Figure 3). Conclusion: In this series, 89% of allografted patients who developed an anti-spike humoral response also presented an anti-SARS-Cov-2 cellular immunity. Interestingly, anti-SARS-Cov-2 specific T-cells could be detected in 40% of NHR patients. Although a larger group of patients is required to confirm these results, it remains to be determined whether this T-cell response is protective against SARS-Cov-2 infection as previously demonstrated for CMV (Litjens et al, 2017). Finally, the role of potential immunosuppressive MDSCs must be explored in patients who develop no sign of T-cell response after vaccination. [Formula presented] Disclosures: Moreau: Oncopeptides: Honoraria;Celgene BMS: Honoraria;Sanofi: Honoraria;Abbvie: Honoraria;Janssen: Honoraria;Amgen: Honoraria.

Interest of a Third Dose of BNT162b2 Anti- Sars-Cov-2 mRNA Vaccine in Allogeneic Hematopoietic Stem-Cells Recipients

Introduction In a previous observational study of 117 allogeneic hematopoietic stem cell transplant (Allo-HSCT) recipients, we found that 83 % of them achieved a specific humoral response after two doses (V1 and V2) of BNT162b2 anti-SARS-CoV-2 messenger RNA vaccine (Pfizer BioNTech). However, although 61.5% of the patients achieved the highest detectable IgG titers, this proportion remained significantly lower than what was observed in healthy controls, where 100% reached these highest antibody titers. Here, we investigated whether a third dose of vaccine would improve the anti- SARS-CoV-2 response in Allo-HSCT recipients. Methods This monocentric retrospective study aimed at evaluating the efficacy of a third vaccine (V3) of BNT162b2 in a cohort of Allo-HSCT adult recipients. Patients with previous clinical or asymptomatic biological COVID-19 infection at V1 were excluded from the study. A cohort of healthy volunteers (caregivers from the Clinical Hematology Department) who had also already received V1 and V2 was considered as controls. All participants were vaccinated between January 20 and June 1, 2021. Analyses were performed in July 2021. Antibody response to the SARS-CoV-2 spike protein receptor-binding domain was tested after V2 for all subjects (Serology post V2, SpV2) using the Roche Elecsys® assay. All subjects benefited later from another evaluation of specific serum antibodies as monitoring (Serology post V2+, SpV2+) or after V3 (Serology post V3, SpV3). Various serological methods were used for these later assays because performed outside of our hospital for some patients. Considering thresholds of negativity and positivity as well as highest values for each test, we were able nevertheless to distinguish 4 sub-groups: i) negativity at both SpV2 & SpV2+/SpV3, ii) increase of the IgG titer between SpV2 & SpV2+/SpV3, including patients showing seroconversion, iii) decreased or stable IgG titer between SpV2 & SpV2+/SpV3 and iv) highest IgG titers at both SpV2 and SpV2+/SpV3. Results A cohort of 25 controls and 114 patients, including 91 who received V3 (V3+) and 23 who did not (V3-) was considered for the purpose of this study. The characteristics of participants and delays from SpV2 to SpV2+ or SpV2 to SpV3 are reported in Tables 1 and 2. The serological methods used for the latest assays are reported in Table 2 with criteria of negativity, positivity and highest IgG titer values. V3- patients were younger, with less myeloid disease than V3+ cases and had not received myeloablative conditioning. However, both V3+ and V3- groups shared similar median intervals between Allo-HSCT and V1, incidence of previous graft versus host disease (GVHD), proportions of patients under chemotherapy or immunosuppressive drugs and median lymphocyte counts at V1, suggesting similar immune status. The reasons for not receiving V3 were forgetting, refusal or surveillance after detection of the highest IgG titer at SV2. Samples from controls, all evaluated by Roche Elecsys®, showed the highest anti-spike antibody value (>250U/mL) at both SpV2 and SpV2+, suggesting a persistent response without the need of a third vaccine in this healthy population. The proportion of patients still negative at SpV2+/SpV3 was similar between V3- and V3+ patients (17% vs 12%, p=0.74). However, the proportion of patients showing a decreased/stable IgG titer between SpV2 and SpV2+/SpV3 was significantly higher for V3- cases (35% vs 4%, p=0.0001) (Table 2). Moreover, the proportion of patients with the highest IgG titer at SpV2+/SpV3 was significantly higher in the V3+ sub-group (80% vs 43%, p=0.001), even if it remained significantly lower than in controls (p=0.03). The proportion of patients showing an IgG titer increase between SpV2 and SpV2+/SpV3 was higher in V3+ vs V3- patients (24% vs 4%, p=0.06). The difference was not significant as surprisingly one V3- case showed a seroconversion without any argument for SARS-CoV-2 infection between SpV2 and SpV2+. Three patients out of 14 (21%), with a negative SpV2, showed a seroconversion after V3. Fi ally, with a median follow up from V1 of 106 days in V3+ patients, 138 days in V3- patients and 154 days in controls, no COVID-19 infection was documented in any participant. Conclusion This study shows the interest of a third dose of BNT162b2 anti-SARS-CoV-2 messenger RNA vaccine after allograft as more patients are documented with less decrease of IgG titers and the highest IgG values after V3. [Formula presented] Disclosures: Moreau: Abbvie: Honoraria;Amgen: Honoraria;Janssen: Honoraria;Sanofi: Honoraria;Celgene BMS: Honoraria;Oncopeptides: Honoraria.

Antibody Response after One and/or Two Doses of BNT162b2 Anti- Sars-Cov-2 mRNA Vaccine in Patients Treated By CAR T-Cells Therapy

Introduction: Data regarding the efficacy of anti-severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) messenger RNA vaccines in immunocompromised hosts are scarce and no data yet appear to be available for patients with hematological malignancies who also received chimeric antigen receptor-T (CAR-T) cells therapy. Methods: The efficacy and safety of one and/or two injections of the BNT162b2 (Pfizer-BioNTech) vaccine was evaluated retrospectively in 23 CAR-T recipients in our Hematology Department, compared to a cohort of 25 healthy caregivers, vaccinated concomitantly between January 28 and May 31, 2021. None of these individuals had a previous clinical history of COVID-19. Results: Overall, the patients (14 males and 9 females) had a median age of 62 years old (range: 21-79) and had received CAR-T for high-grade lymphoma (n= 20) or acute lymphoblastic leukemia (n= 3). Eight and 3 had been respectively previously autografted or allografted and two were allografted after CAR-T. All patients were pretreated for lymphodepletion by fludarabine + cyclophosphamide before CAR-T infusion. The CAR-T provided were axicabtagene ciloleucel (Yescarta, Kite/Gilead, n=16, tisagenlecleucel (Kymriah, Novartis Pharma, n=5 and brexucabtagene autoleucel (KTE-X19, Tecartus, Kite/Gilead, n=1). One additional patient received allogeneic UCART19 (Servier). The median delay between CAR-T administration and the first vaccine (V1) was 401 days (d;range: 113-819). All patients but 2 were in complete remission at V1 and 3 were still on therapy (revlimid n=1, tafasitamab n=1, chemotherapy n=1). After V1, antibody response to the SARS-CoV-2 spike protein receptor-binding domain was tested by the Roche Elecsys® assay at a median time of 29 d (range: 16-32) in 19 patients and 23 d (range:18-32) in controls. At that time, only 4/23 patients (21%) but all controls (100%) had a positive anti-spike antibody response (p<0.001). Among seropositive cases, median IgG titers were higher in controls (35.1 U/mL, range 2.2->250) than in patients (5.9 U/mL range 4.1-41.6, p=0.06). The highest IgG titer (>250) was obtained in 2 controls. The median delay between V1 and the second vaccine (V2) was 28 d (range: 14-46) for patients and 23 d (range: 18-32) for controls. Among the 20 patients tested after V2, 17 had also been tested after V1 while 3 were tested only after V2. All controls were tested after V2. The second serology assay was performed at a median interval from V2 of 52 d (range: 21-99) for patients and 58 d (range: 32-71) for controls. This serology assay was positive in 6 patients (30%), while all controls (100%, p<0.001) had again a positive response. Three out of these 6 patients (15% of all patients) achieved the highest IgG titer according to the serology assay used. Among the 4 patients with positive antibody titers after V1, 3 remained positive including one reaching the highest IgG titer. The fourth patient has not yet received V2. Median IgG titers could not be compared with controls because various methods of detection were used after V2. However, all controls tested again by Roche Elecsys® displayed the highest IgG titer (>250) after V2. The two patients in relapse and treated by chemotherapy or tafasitamab did not develop antibodies after V2 conversely to the patient under maintenance by revlimid. The delay between CAR-T infusion and vaccine did not influence the antibody response nor did the rate of lymphopenia as almost all patients remained under a lymphocyte threshold of 1x10 9/L. Finally, with a median follow up from V1 of 77 d (range: 49-127) in patients and 81 d (range: 62-95) in controls, no COVID-19 infection has been documented in any of these participants. Conclusion: This study shows that the administration of two doses of BNT162b2 anti-SARS-CoV-2 messenger RNA vaccine provides a low rate of seroconversion (30%) in recipients of CAR-T therapy. This is likely related to the profound B-cell depletion induced by this treatment precisely targeting CD19+ cells. Investigation of the development of specific T-cell response in these individuals could provide more information about the efficacy of vaccination in this context. Disclosures: Moreau: Celgene BMS: Honoraria;Sanofi: Honoraria;Janssen: Honoraria;Abbvie: Honoraria;Amgen: Honoraria;Oncopeptides: Honoraria.

Profound B-Cell Lymphopenia Is a Major Factor Predicting Poor Humoral Response after BNT162b2 mRNA Sars-Cov-2 Vaccines in Recipients of Allogeneic Hematopoietic Stem Cell Transplantation

Background Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection results in poor outcome in patients with hematologic malignancies. Moreover, the efficacy of anti-SARS-CoV-2 mRNA vaccines appears lower in immunocompromised patients, including recipients of allogeneic stem cell transplantation (Allo-HSCT). In this population, data are scarce regarding factors predicting the response to mRNA vaccines. Methods This retrospective study aimed to decipher which factors, including immune status at time of vaccine and recipient/donor blood groups, might influence the antibody response after two injections (V1 and V2) of BNT162b2 (Pfizer-BioNTech) vaccine in a cohort of allografted patients with no previous symptomatic nor asymptomatic COVID-19 infection. Possible previous asymptomatic COVID-19 infection was investigated in pre-V1 samples by testing for anti-nucleocapsid (N) antibodies (anti-SARS-CoV-2 immunoassay, Roche Elecsys®, Rotkreuz, Switzerland). Antibody response to the SARS-CoV-2 spike protein receptor-binding domain was tested post-V2 (Roche Elecsys®). As recommended by the manufacturer, titers ≥0.8 U/mL were considered positive, the highest value being >250 U/mL. Blood samples were also collected before V1 and at distance from V2 to evaluate, by flow cytometry, total lymphocyte (Ly) counts and quantitative Ly subsets (CD3, CD4 and CD8 T cells, B and NK cells). Statistical analyses were performed on R (version 4.0.3). Patient characteristics were compared by using the Χ² test for discrete variables and the Wilcoxon test for continuous variables. Generalized linear models were used to conduct multivariate analyses. Results Samples were available from 117 Allo-HSCT patients who had been vaccinated between January 20 and April 17, 2021. Patient characteristics are provided in Table 1. The average interval from Allo-HSCT day 0 (D0) to V1 (D0-V1) was 654 (IQR: 372-1367) days (d). S-antibody response rate post-V2 was 82.9% for the entire cohort. Non-humoral responders (NHR) post-V2 (n = 20) had a lower D0-V1 interval (median 271 vs 914 d, p <10 -5) and a lower pre-V1 median total Ly count (0.62 vs 1.61x10 9/L, p < 10 -4). Lymphocyte subsets possibly predictive of antibody response were then investigated. NHR were associated with lower median CD3 (0.39 vs 0.97 x10 9/L, p = 0.01), CD4 (0.13 vs 0.35 x10 9/L, p=<10 -3), and B-cell (0.00 vs 0.28 G/L, p <10 -6) counts. NK and T CD8 counts were not statistically different between NHR and HR (respectively p=0.14 and p=0.06). No influence either was observed when considering the age of donors (p=0.39) or recipients (p=0.55), underlying disease (p=1), Allo-HSCT conditioning (p=0.11), blood groups (donor, p=0.55;receiver, p=0.39) or a previous history of graft versus host disease (GVHD;83.1 vs 83.6%, p=1). Conversely, ongoing immunosuppressive (IS)/chemotherapy treatment and a haploidentical source of graft were associated with lower responses to vaccination (respectively 62.5 vs 90.5%, p<10 -3, and 69.4 vs 88.6% for patients with matched donors, p=0.02). In multivariate analysis (Fig.1) also including D0-V1 interval, donor source, current IS/chemotherapy treatment and TCD4 Ly count, only B cell aplasia remained statistically associated with lack of antibody response after two vaccine injections (OR 0.01, 95%CI [0.00 - 0.10], p <10 -3). The possible modification in terms of lymphocyte counts between pre-V1 and post-V2 times has been also investigated showing that only CD4 lymphocytes counts improved significantly (0.31 vs 0.34 x10 9/L, p=0.01) between this interval. Conclusion B cell aplasia appears as a major predictor of anti SARS-CoV-2 mRNA vaccine failure after Allo-HSCT. It may be suggested from this result that a close immune monitoring should be proposed after allotransplant to propose the vaccine at the most appropriate time, meaning after of B cell detection, regardless of the delay from Allo-SCT or the presence of an IS/chemotherapy treatment. The possibility for these patients to have mounted a cellular response should also be considered, which was not investigated here. [Formula presented] Disclosures: Moreau: Celgene BMS: Honoraria;Sanofi: Honoraria;Abbvie: Honoraria;Oncopeptides: Honoraria;Amgen: Honoraria;Janssen: Honoraria.