Aplastic anemia treatment in older adults in resource-limited settings: Can rabbit ATG bridge the gap?

Abstract Background The treatment of choice for aplastic anemia (AA) is triple immunosuppressive therapy with horse anti-thymocyte globulin (h-ATG), cyclosporine (CsA), and eltrombopag (ETB), as it has shown to improve overall response rates (ORR) and relapse-free survival, even in the small subgroup of patients ≥65 years old with ECOG ≤2. However, this combination is often unavailable for patients in Latin American countries due to lack of access to h-ATG. In our country, the only available alternative is rabbit ATG (r-ATG), which is often avoided in patients ≥60 years due to drug toxicity and patient comorbidities. Our objective was to determine whether combinations including r-ATG improve ORR in patients ≥60 years with severe (SAA) or very severe aplastic anemia (VSAA) compared to CsA/ETB and other therapies. Methods We conducted a multicenter study in five referral centers in Mexico. Three treatment groups were defined: r-ATG plus immunosuppressive therapy (CsA/ETB or CsA), CsA/ETB, and other treatments (O). Overall response (ORR), defined as the sum of complete response (CR) and partial response (PR), was assessed at 3, 6, and 12 months after treatment initiation. We analyzed differences in response among groups, predictors of response, overall survival (OS), and predictors of OS. Results From 1990 to 2024, we enrolled 70 patients ≥60 years with AA: 30 (42.9%) from the National Institute of Medical Sciences and Nutrition, 16 (22.9%) from the General Hospital of Mexico, 12 (17.1%) from ISSSTE, 7 (10%) from the University Hospital of Monterrey, and 5 (7.1%) from PEMEX Central North Hospital. The median age at diagnosis was 67 years (range: 60–87), and 40 patients (57.1%) were female. Fifty-two patients (74.3%) had at least one comorbidity, the most common being hypertension (42.9%), and 27 had an ECOG score >2 (39.1%). Forty-five patients (64.3%) were classified as SAA. The median time to treatment initiation was 8.2 days (IQR: 0–25.5). At diagnosis, 17 patients (24.3%) had an active infection. Seventeen patients (24.2%) received r-ATG, almost all as part of triple therapy; 16 received CsA/ETB (22.9%); and 37 received other therapies (52.9%). The main reason for not administering r-ATG was patient age (70.37%). ORR was higher in the CsA/ETB group at all time points: at 3 months, CsA/ETB 12 (75%) vs r-ATG 4 (23.5%) vs O 8 (21.6%), p<0.001; at 6 months, 13 (81.2%) vs 5 (29.4%) vs 14 (37.8%), p=0.004; and at 12 months, 13 (81.2%) vs 6 (35.3%) vs 15 (40.5%), p=0.011. r-ATG did not show a significant improvement in ORR at 3 (p=0.92), 6 (p=0.493), or 12 months (p=0.679). In multivariate analysis, the only predictor of ORR was CsA/ETB use (3 months OR 0.09, 95%CI 0.01–0.5, p=0.009; 6 months OR 0.07, 95%CI 0.01–0.47, p=0.006; 12 months OR 0.09, 95%CI 0.01–0.61, p=0.013). Median follow-up was 22.5 months (IQR 6.38–68.8), and median OS was 85.9 months (95%CI 50.3–NA). The main cause of death was infection (50%), most commonly pulmonary. The 36-month OS was 15.4% for r-ATG, 80.2% for CsA/ETB, and 74.7% for O (p=0.0014). Among r-ATG recipients, 52.94% died due to AA, compared to only 18.75% in the CsA/ETB group. In multivariate Cox regression analysis, predictors of OS were CsA/ETB use (HR 0.22, 95%CI 0.06–0.84, p=0.027), use of other therapies (HR 0.34, 95%CI 0.13–0.9, p=0.03), age ≥70 (HR 2.95, 95%CI 1.14–7.6, p=0.025), AA severity (HR 6.23, 95%CI 2.18–17.75, p=0.001), and achieving OR at any time (HR 0.09, 95%CI 0.03–0.24, p<0.001). Discussion The utility of r-ATG in patients ≥60 years with AA has not been well described in the literature, and our results do not support its systematic use. We found that adding r-ATG to immunosuppressive therapy was associated with lower response rates, worse OS, and more infection-related deaths. Surprisingly, CsA/ETB consistently showed higher ORR and longer OS compared to r-ATG. These findings call for a reassessment of the optimal treatment strategy for AA in patients ≥60 years in countries without access to h-ATG.