Treatment (Tx) options for pts with ND AML include high-intensity (HI) chemotherapy (7+3 regimen) followed by consolidation therapy or allogeneic stem cell transplantation, low/intermediate-intensity (LI) therapies (low-dose cytarabine or hypomethylating agents), best supportive care (BSC), or no Tx. While several studies have examined HCRU and costs among pts with ND AML, these studies focused on claims data, and few studies have examined HCRU based on Tx regimen received.
To examine HCRU in ND AML pts receiving different Tx modalities in the Connect® MDS/AML Disease Registry (NCT01688011).
Pt demographics, disease characteristics, and Tx information were reported by sites at screening and every 3 months for AML pts enrolled from December 2013 to December 2018 in the Connect® MDS/AML Disease Registry, a large, US, multicenter, prospective observational cohort study of pts with ND AML (aged ≥55 yrs) or myelodysplastic syndromes (aged ≥18 yrs). HCRU was assessed in ND AML pts according to Tx groups defined as HI, LI, BSC, or no Tx initiated ≤45 days post-diagnosis. Unplanned hospitalizations due to disease or Tx-related complications were assessed at 3 and 6 months post-enrollment. Response to Tx was analyzed in the HI and LI groups only. Number of transfusion episodes was calculated 6 months post-enrollment and at each month in the HI and LI groups.
A total of 510 AML pts, treated at 22 academic and 92 community/government centers, were included in the analysis. Median age was 70 years (range 55–92), 63% were male, and 84% were white. 98 (19.2%) pts received BSC/no Tx, 186 (36.5%) received LI therapy, and 226 (44.3%) received HI therapy. The percentage of pts with ≥1 hospitalization was similar between the HI and LI groups (59.3% vs 53.8%), but was lower in the BSC/no Tx group (24.5%; P < 0.05). During months 1–3 post-enrollment, pts in the HI group spent a mean of 15.9 days in hospital versus 8.9 days in the LI group and 10.3 in the BSC/no Tx group; this decreased in all Tx groups to 11.6, 8.2, and 6.5 days, respectively during months 4–6 post-enrollment. Mean monthly transfusion burden over the first 6 months in the HI and LI groups was 4.9 versus 2.8 transfusion episodes/month (P < 0.01). During month 1 post-enrollment, mean transfusion episodes in the HI group was 8.8 versus 3.2 in the LI group (P < 0.01). Over the 6-month study period, the transfusion rate decreased in both groups (P < 0.01). Transfusion rates in the HI and LI groups were similar after Month 4 (Figure), coinciding with the end of consolidation chemotherapy in the HI group. In logistic regression analyses, pts in the HI group were more likely to achieve complete remission (CR) versus pts in the LI group, after adjusting for age, comorbidities, and disease risk (odds ratio 5.2, 95% CI 2.0–13.5; P < 0.01). Mortality rates were significantly higher in the LI group versus the HI group at 60 (P < 0.05) and 180 days (P < 0.01), but not at 30 days. Median overall survival (OS) was longer in the HI group (21 months; 95% CI 16–32) versus the LI group (9 months; 95% CI 6–11; P < 0.01); however, Tx group was not a significant independent predictor of OS in multivariate analysis.
These findings highlight the level of HCRU in ND AML pts treated with different intensity therapies. While HCRU was highest in the HI group and may be secondary to cytopenia-related complications associated with HI regimens, pts in this group were significantly more likely to achieve CR, have improved OS, and lower mortality rates at 60- and 180-day landmarks.