TP53 Mutations Increased the Risk of CD19 Negative Relapse Following Blinatumomab in Adults with B-Cell Acute Lymphoblastic Leukemia (B-ALL)

Introduction: Blinatumomab (blina) is a bispecific CD3xCD19 T cell engager that redirects autologous T cells to target CD19+ lymphoblasts. Despite the success of blina in B-ALL, treatment failure is common and can manifest with unique patterns of relapse such as antigen loss and extramedullary disease (EMD). While some mechanisms have been proposed to explain the emergence of CD19-negative (CD19-) disease and relapse as EMD, the impact of intrinsic factors such as disease biology remains largely unknown.

Methods: We retrospectively reviewed 267 adult patients (pts) with B-ALL treated at City of Hope who received at least 1 cycle of blina between 2012 and 2024. We examined ALL genetics using next generation sequencing, cytogenetics and/or microarray. We analyzed response to blina and patterns of failure according to pts and clinical factors as well as leukemia genetic alterations, the presence of myeloid-type mutations, and IKZF1-alterations.

Results: Blina was administered for relapsed/refractory (r/r) disease (n=150, 56%), minimal residual disease (MRD+) (n=88, 33%), upfront as induction (n=10, 4%) and as consolidation in MRD negative state (n=19, 7%). The median age was 40 (range: 18-88) years and 186 (70%) pts were Hispanic. The median number of prior lines of therapies was 1 (range: 0-5) and 26 patients (10%) had allogeneic hematopoietic cell transplantation (alloHCT) before blina. Among all treated pts, 50 (19%) had Ph+, 80 (30%) had Ph-like, 39 (15%) had normal karyotype (NK), 30 (11%) had TP53 mutations (TP53m), 14 (5%) complex karyotype without TP53m, 7 (3%) with KMT2Ar, 9 (3%) with PAX5-alterations. Among pts who underwent comprehensive genetic testing, 80 (44%) had detectable myeloid mutations (frequency in order: TP53, KRAS, NRAS, FLT3, DNMT3A, TET2, ASXL1, RUNX1, IDH1, IDH2, PHF6, PTPN11) and 61 (33%) had IKZF1 alterations. For pts treated for r/r B-ALL, the overall complete remission (CR)/CR with incomplete hematological recovery (CRi) rate was 59% (n=89); rates among genetic subgroups were 54% for Ph+, 66% for Ph-like, 74% for NK, and 59% for TP53m (p=NS). CR/CRi rates for pts with and without myeloid mutations were 60% and 73% (p=0.29), respectively, and for pts with and without IKZF1 alteration were 61% and 68% (p=0.74), respectively. There were 99 pts who underwent alloHCT consolidation after blina in CR/CRi. Only pretreatment high disease burden (≥50% bone marrow [BM] blasts; p<0.01) and active EMD (p<0.01) were associated with inferior CR/CRi rate.

Among 169 pts in CR/CRi post blina, 90 (53%) remained alive and in remission and 79 (47%) relapsed, including 3 with unknown CD19 expression status, 22 (13%) pts with CD19- relapse and 54 (32%) pts with CD19+ relapse. The median times to CD19- and CD19+ relapse were 6 and 7.4 months, respectively. The presence of myeloid mutations (p<0.01), TP53m (p<0.01), blina administered in r/r setting (p<0.01), the use of blina as an earlier line of therapy (0-1) (p<0.01), pre-infusion high-disease burden (p=0.03) and not receiving alloHCT after response to blina (p<0.01) were associated with higher risk for CD19- relapse in univariate analysis. CD19- relapse represented 77% (n=13) and 75% (n=3) of relapsed cases with TP53m and KMT2Ar, respectively; CD19- disease did not occur for relapsed pts with Ph+ (n=10), complex cytogenetics without TP53m (n=7), Ph-like with non-CRLF2r (n=3) or PAX5a (n=2). CD19- relapse was observed among 56% and 10% of relapsed pts with and without myeloid mutations, respectively, and in 11% and 42% of pts with and without IKZF1-alterations, respectively.

In multivariate analysis, TP53m was associated with increased risk of CD19- relapse (HR=5.03; 95%CI:1.4-18.1, P<0.01), while post blina alloHCT consolidation was associated with lower risk of CD19- relapse (HR= 0.11; 95%CI: 0.03-0.44, P<0.01).

Among pts in CR/CRi post blina, 22% had EMD +/- BM relapse, 25% had only BM relapse, and 53% remained alive and did not relapse. The median times for EMD relapse and BM only relapse were 5.9 and 8.7 months, respectively. In multivariate analysis, alloHCT consolidation was associated with lower risk for EMD relapse (HR=0.17; 95%CI: 0.08-0.37, P<0.01).

Conclusion: Leukemia biology may predict patterns of blina failure after response despite lack of clear impact on initial response. The presence of TP53m was associated with significantly higher risk for CD19- relapse as an immune escape mechanism following blina and warrants further study.