Distinct Prognostic Effects of TP53 Mutations in Newly Diagnosed Versus Relapsed/Refractory (R-R) Patients (pts) with B-Acute Lymphoblastic Leukemia (ALL) Treated with Mini-Hcvd-Inotuzumab Ozogamicin with or without Blinatumomab Regimens

Introduction

B-ALL with TP53 mutations (TP53^MT) at diagnosis is associated with older age and frequent low-hypodiploid karyotype. We have previously shown that hyper-CVAD-based regimens negate the poor prognostic impact of TP53^MT. However, the prognostic effect of TP53^MT in B-ALL treated with mini-HCVD-inotuzumab with or without blinatumomab in the frontline and R-R settings have not been explored.

Methods

Older (≥60 yrs) pts with newly-diagnosed ALL and adults (≥18 yrs) with R-R ALL treated with mini-HCVD-inotuzumab ozogamicin with/without blinatumomab underwent testing for TP53^MT. We compared the clinico-pathologic & outcome differences between TP53^MT and TP53^wt cases in both settings. Germline origin was inferred based on persistence of TP53^MT at high variant allele frequency (VAF) during remission. Multiallelic (>1) TP53 alteration was defined as a concurrent second TP53^MT or deletion (TP53^del).

Results

Of the 48 pts with newly diagnosed B-ALL, 19 (40%) pts had TP53^MT at diagnosis; in 2 of 14 (14%), TP53^MT was of germline origin. All pts were Philadelphia chromosome (Ph) negative; 1 pt had CRLF2 rearrangement (Ph-like immunophenotype); none had KMT2A rearrangement (KMT2Ar). Fifteen (83%) pts had 1 TP53^MT, 2 had 2 mutations, and one pt had 3 mutations. The majority were missense (15, 83%), the rest included non-sense, frame-shift, and splice site mutation in 1 patient, each. The median TP53^MT VAF was 46% (range, 4-80). Nine (56%) had multiallelic TP53 alteration, with 6 (38%) showing TP53^del. Compared with TP53 wild-type (TP53^wt), TP53^MT had a higher frequency of low-hypodiploidy (40% vs. 8%, p=0.04), lower median platelet counts (34 vs. 55, p=0.02), and a trend for older age (median 71 vs. 68, p=0.09). There was no difference in morphologic response rates (89% vs. 100%; p=0.11) and minimal residual disease negativity [MRD neg] rates (94% vs. 75%; p=0.13) between mutated and wild type groups. Over a median follow-up (f/u) of 46 months, 21 (44%) patients died [9 (50%) mutated vs. 12 (44%) wild type]. There was no significant difference in the overall survival (median OS, 33 vs. 38 months, p=0.28; Fig 1A); the 3-year survival rates were 45% and 63% in TP53^MT and TP53^wt, respectively.

Of the 47 pts with R-R B-ALL, 16 (34%) had TP53^MT. Three (3/12, 25%) had germline TP53^MT. All were Ph-negative, and 1 patient was Ph-like ALL. All 16 patients had a single TP53^MT; 9 (56%) had multiallelic TP53 alteration due to concurrent TP53^del. Majority were missense (12, 75%), the rest included 2 non-sense, 1 frame-shift and 1 in-frame duplication. The median TP53^MT VAF was 33% (range, 1-78). Except for a higher frequency of low-hypodiploidy (30% vs. 0%, p=0.01), no other significant clinico-pathologic differences were noted between the TP53^MT and TP53^wt groups. Pts with TP53^MT had a higher morphologic response rate (100% vs. 67%, p=0.02) with no difference in MRD neg rates (78% vs. 85%, p=0.65). Over a median f/u of 33 months, 15 TP53^MT pts (94%) died compared to 14 TP53^wt pts (45%; p=0.001). Pts with TP53^MT had significantly shorter median survival (5 months vs not reached, p=0.0003; Fig 1B); the 3-year survival rates were 0% and 52% in TP53^MT and TP53^wt, respectively.

Despite significant differences in outcome of TP53^MT at diagnosis vs. relapse, there were no notable differences in TP53^MT characteristics, except for enrichment of KMT2Ar in R-R (0% vs. 31%; p=0.04). We retrospectively reviewed the cytogenomic features at the time of diagnosis of 7 R-R B-ALL with TP53^MT (non-germline). None had TP53^del at diagnosis. TP53 NGS profiling at diagnosis on one KMT2Ar pt was negative for TP53^MT (1% sensitivity) and TP53^del (Fig 1C). These findings suggest that TP53^MT/TP53^del clones (undetectable at diagnosis) expanded following chemotherapy and relapsed with cytogenetic evolution. TP53^MT clones that survived the chemotherapy insult were inherently more aggressive than treatment naïve clones at diagnosis, and may explain the observed differences in outcome.

Conclusions

TP53^MT is seen in up to 40% of B-ALL at diagnosis and ~35% at relapse, and are associated with low-hypodiploid karyotype. B-ALL with TP53^MT at diagnosis shows no significant difference in OS compared to TP53^wt in patients treated with frontline mini-HCVD-inotuzumab with or without blinatumomab regimen. In contrast, TP53^MT confers a poor prognosis in pts with R-R disease. Sequencing studies on additional R/R TP53^MT patients at diagnosis are underway.