Myeloid Neoplasia after CD19 Directed CAR-T Cells : Cumulative Incidence, Risk Factors and Outcome

Introduction:

CD19 directed chimeric antigen receptor T cells (CAR-T cells) are now being administered earlier in the management of relapse, and more patients (pts) with non-Hodgkin B cell lymphomas (NHL) are receiving this treatment.

Recently, non-relapse mortality (NRM) following CAR-T cell therapy was reported at 6.8%, with 50% attributed to infections and 7.8% to secondary primary malignancies (Cordas dos Santos, 2024). These malignancies are mainly treatment-related myeloid neoplasia (tMN), such as acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS).

We described in this study, cumulative incidence, risk factors and the outcome of tMN after CD19 directed CAR-T cells therapy for NHL in two tertiary departments of Hematology.

Methods:

Cumulative incidence of tMN and NRM was estimated by treating relapse or death as competing risk. The effects of covariates on the occurrence of tMN were assessed using the Fine and Gray model.

Results:

We retrospectively analyzed data from 511 pts treated between September 2017 and August 2023 with a median follow-up of 26.4 months (95% CI, 24 – 30). The median age was 63 (range 17 – 87). Pts were mostly treated for diffuse large BCL (n=264, 51.7%), transformed lymphoma from indolent NHL (n=98, 19.2%), follicular lymphoma (n=56, 10.9%) or mantle cell lymphoma (n=30, 5.9%). They were treated with either axi-cel (n=295, 57.8%), tisa-cel (n=140, 27.3%), liso-cel (n=43, 8.4%), brexu-cel (n=30, 5.8%), or an investigational CAR-T cell (n=3, 0.6%). The median number of previous lines of therapy was 2 (range 1 – 11), median LDH level prior to lymphodepletion (preLD) was 254U/I (IQR 205 – 329), median preLD CAR-HEMATOTOX score was 2 (range: 0-6). A total of 441 (86%) pts experienced CRS (7% ≥ grade 3), and 179 (35%) pts developed ICANS (8.2% ≥ grade 3).

The median OS and PFS were 48.8 and 9.5 months, respectively. Cumulative incidences of NRM at 1, 2, 3 and 4 years were 5.2% (95% CI, 3.5 – 7.5), 9.3% (95% CI, 6.8 – 12.0), 12% (95% CI, 8.5 – 15.0) and 14.0% (95% CI, 10.0 – 18.0), respectively. Cumulative incidence of tMN at 1, 2, 3 and 4 years were 1.4% (95% CI, 0.64 – 2.8), 4.2% (95% CI, 2.6 – 6.5), 7.1% (95% CI, 4.5 – 10) and 9.7% (95% CI, 6.1 – 14), respectively. NRM was mostly due to infection (n=21, 38%) and tMN (n=18, 33%).

In univariate analysis, age at reinfusion was significantly associated with the occurrence of tMN (HR 1.04; 95% CI 1.01 – 1.07, p=0.01), as was the number of previous lines of therapy (HR 1.18; 95% CI 1.05 – 1.32, p=0.005). No statistical association was found between occurrence of tMN and CAR-T-cell product, CAR-HEMATOTOX score, CRS and ICANS grade.

A total of 28 pts were diagnosed with tMN, 10 pts with AML (35.7%) and 18 with MDS (64.3%). At tMN diagnosis, the median age was 68 (IQR 57 – 71). Median time between CAR-T cell and tMN was 15.6 months (range 1.0 – 56.9). At tMN diagnosis, median hemoglobin level was 91 g/L (IQR 78 – 99), MCV was 96 fl, absolute neutrophil count was 0.9 G/L (IQR 0.47 – 1.52) and platelet count was 47 G/L (IQR 28 – 59).

Among the 10 AML cases, 90% were classified as intermediate (n=3) or unfavorable (n=6) risk according to ELN 2022 classification, and one pt showed an inv(16) karyotype. Of the 18 MDS cases, 11 (61%) had a high or very high R-IPSS score, with 80% showing a TP53 mutation and 33% a DNMT3A or PPM1D mutation. A majority of tMN (n=10/16, 63%) pts harbored monosomal or complex karyotypes.

Nine pts with tMN (32%) received intensive treatment, including 7 who underwent allogeneic hematopoietic cell transplantation. The median survival of this population after tMN diagnosis was 5.5 months (95% CI, 3.4 – 14.0), with a 1-year survival rate of 30.1% (95% CI, 15.9 – 59.0).

Conclusion:

Unlike registry-based data, our analysis utilized manually verified data, using cumulative incidences and considering death or relapse as competing risks. We observed 3-year NRM and tMN incidences of 12% and 7.1%, respectively. Our findings indicate a higher mortality from tMNs than previously reported, potentially due to missing data on tMN deaths in some studies.

We observed that pts age and number of prior therapies were significant risk factors for the development of tMNs. These tMNs exhibited a poor prognosis with more frequent adverse molecular and cytogenetic profiles compared to other tMN previously reported (Belhabri, 2023).

To further explore the prognosis of tMNs after CAR-T cell therapy, comparisons with a control group is ongoing and will be presented at ASH.