Hypomethylating Agents Are Associated with High Rates of Hematologic Toxicity in Patients with Secondary MDS/AML That Develops after Acquired Aplastic Anemia

Abstract

Acquired aplastic anemia (AA) is an autoimmune bone marrow failure (BMF) associated with depletion of hematopoietic stem and progenitor cells. Approximately 15-20% of AA patients treated with immunosuppressive therapy (IST) develop late complications of myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). Little is known about managing patients with post-AA myeloid neoplasms (MN). We hypothesized that patients with post-AA MN may be particularly susceptible to hematologic toxicities from cytotoxic therapy because of stem cell deficits. To investigate our hypothesis, we retrospectively analyzed post-AA MN patients treated at our 2 institutions over the past 15 years. Fourteen post-AA MN patients were identified: 11 with MDS, 1 with AML, and 2 with clonal cytopenia of undetermined significance (CCUS). Patients with inherited BMF or allogeneic stem cell transplant (SCT) prior to MN diagnosis were excluded. The median age at MN was 56.5 years (range 5-75) with a median time of 5 years (range 0.25-30) between AA diagnosis and MN. At MN diagnosis, 12 of 14 patients (86%) had a partial or complete response of AA with 5 receiving cyclosporine (CSA) maintenance. Two patients (14%) were on CSA within 6 months of IST without response. After MN diagnosis, CSA was discontinued. Ten adults received hypomethylating agents (HMA) as first-line treatment in preparation for SCT. Three pediatric patients were treated with SCT with no prior HMA. One patient died before receiving treatment. The 10 post-AA patients who received HMA were matched in a 3:1 ratio with a similarly aged, randomly selected non-AA MDS cohort treated with HMA at our center during the same period (Table 1). Compared to patients with non-AA MDS, post-AA MN patients tolerated HMA poorly with frequent, severe complications. Their median per-cycle duration of grade 4 neutropenia was longer (9 v. 1.5 days, p = 0.044), as was median duration of grade 4 thrombocytopenia (13 v. 0 days, p = 0.003). Post-AA patients notably had lower baseline platelets prior to HMA (median 36.5 v. 115 k/mL, p =0.007). Following HMA, the post-AA cohort had higher rates of febrile neutropenia (80% v. 17%, RR 4.8, p < 0.001) and infections ≥ grade 3 (90% v. 13%, RR 6.8, p < 0.001). They also had higher rates of ≥ grade 3 bleeding (40% v. 7%, RR 6.0, p = 0.026) with 2 patients (20%) experiencing intracranial hemorrhage on HMA; no such events occurred in the non-AA cohort. Post-AA patients had more hospital admissions - 18 in 25 total chemotherapy cycles (72%) compared to 12 in 207 cycles (6%) in the matched cohort (RR 12.4, p < 0.001). Post-AA MN patients had more treatment delays >2 weeks (28% v. 8% of planned treatment cycles, RR 3.4, p = 0.01) and dose reductions (40% v. 13%, RR 3.0, p = 0.089). HMA was discontinued due to treatment-emergent adverse events (TEAEs) in 70% of post-AA cases v. 3% of non-AA patients (RR 21.0, p < 0.001). Consequently, post-AA MN patients received fewer cycles of HMA compared to non-AA patients (median 2.5 v. 6, p = 0.021). Death occurred following TEAEs in 20% of post-AA MN patients; no deaths on HMA occurred in the non-AA cohort (RR 14.1, p = 0.058). Among surviving patients, SCT was delayed in 20% of the non-AA cohort due to TEAEs, while no delays occurred in the matched cohort. The median months from final chemotherapy cycle to SCT were 4 (range 1-12) in post-AA v. 1 (1-3) in non-AA, while median months from MN diagnosis to SCT were 8.5 (4-21) in post-AA v. 5 (4-15) in non-AA. For all 14 post-AA MN patients, including those not treated with HMA, overall survival (OS) from MN diagnosis was 71% (95% CI 51-99) at 1 year and 56% (95% CI 34-90) at 3 years. OS of post-AA MN patients differed significantly (p = 0.011) based on treatment with SCT. Among the 5 who did not receive SCT, 4 died within 3 years of MN, while 1 patient is alive 13 months from diagnosis receiving supportive care after intracranial hemorrhage during the first cycle of HMA. In contrast, 7 of 9 (78%) patients who received SCT were alive 3 years after MN diagnosis. Of the 6 patients treated with SCT after HMA, 5 stopped HMA due to toxicity, all had morphologic dysplasia at SCT, and 2 were transplanted with ≥ 5% marrow blasts. Our study shows that patients with MN following antecedent autoimmune AA are at high risk of severe toxicities with standard HMA regimens and have difficulty tolerating repeated cycles. Our results do not support routine use of HMA prior to SCT in this patient population and suggest that early SCT may be the most suitable strategy.