Introduction Gemtuzumab ozogamicin (GO) is approved for use in combination with cytarabine plus daunorubicin (7+3) for the treatment of de novo acute myeloid leukemia (AML) with favorable or intermediate risk cytogenetics. This study aimed to describe real-world evidence comparing the safety and efficacy of 7+3 alone or with GO in patients (pts) with AML.

Patients and methods This retrospective study included adult pts with AML and favorable/intermediate risk cytogenetics who received 7+3 induction alone or with GO at the Princess Margaret Cancer Center (Toronto) from October 1st 2018 to October 1st 2024. GO was dosed at 3 mg/m2 on days 1, 4 and 7 with 7+3 per the ALFA-0701 protocol. Endpoints included composite complete remission (CRc) rates, measurable residual disease (MRD) negativity rates, overall survival (OS) and relapse-free survival (RFS) with analysis of key genetic subgroups (core binding factor [CBF], NPM1-mutated [NPM1mut], intermediate-risk cytogenetics without NPM1mut [INT]). Propensity score matching with inverse probability treatment weighting (IPTW) compared outcomes with 7+3 vs 7+3+GO, matching for age, sex, ECOG, WBC, NPM1mut and CBF-AML. Veno-occlusive disease (VOD), bleeding, and hepatotoxicity were adverse events (AEs) evaluated with severity graded by CTCAE v5.0.

Results In total, 162 pts were included, 62 received 7+3+GO and 100 received 7+3. Median age was 58 years (range, 22-85), 51% were male, 85% had ECOG of 0-1 and median WBC was 9.0 x 109/L (range, 0.5 – 288.0), with no differences between groups. Twenty-six (42%) pts who received 7+3+GO had CBF AML (15 [24%] t(8;21), 11 [18%] inv(16)), vs 10 (10%) pts who received 7+3 (3 [3%] t(8;21), 7 [7%] inv(16)) (p<0.05). Twenty-eight (45%) pts who received 7+3+GO had intermediate-risk cytogenetics vs 81 (81%) pts who received 7+3 (p<0.05), including 14 (23%) and 45 (45%), respectively, without NPM1mut (INT). In pts who received 7+3+GO, 17 (27%) had NPM1mut AML vs 41 (41%) in pts who received 7+3 (p=0.20).

CRc was achieved in 56 (90%) pts with 7+3+GO and in 77 (77%) pts with 7+3 (p<0.05). CRc rates were similar within each genetic subgroup: 96% vs 100% (CBF); 88% vs 90% (NPM1mut); and 57% vs 51% (INT) for pts who received 7+3+GO vs 7+3, respectively. MRD negativity by flow cytometry was achieved in 46 (81%) vs 58 (58%) in pts after 7+3+GO and 7+3, respectively (p<0.05). MRD negativity rates were not significantly different within each genetic subgroups: 92% vs 80% (CBF, p=0.30); 65% vs 73% (NPM1mut, p=0.54); and 57% vs 42% (INT, p=0.54) for pts who received 7+3+GO vs 7+3, respectively.

With a median follow-up of 28.6 months (7+3+GO) and 44.1 months (7+3), the 2-year OS was 78% (95% CI, 68-90) and 64% (95% CI, 55-75), respectively, (HR, 0.71, 95% CI 0.38-1.33, p=0.29). In pts with inv(16) AML, the 2-year OS was 91% (95% CI, 75-100) with 7+3+GO and 86% (95% CI, 63-100) with 7+3 (p=0.87). In pts with t(8;21) AML, the 2-year OS was 66% with 7+3+GO and 3/3 pts with 7+3 were alive at last follow-up. In pts with INT AML, the 2-year OS was 78% (95% CI, 59-100) with 7+3+GO vs 58% (95% CI, 44-76) with 7+3 (p=0.40). In pts with NPM1mut AML, the 2-year OS was 74% (95% CI, 54-100) with 7+3+GO and 70% (95% CI, 57-86) with 7+3 (p=1.00). Among patients achieving CRc, the 2-year RFS was 68% (95% CI, 56-83) with 7+3+GO and 65% (95% CI, 55-77) with 7+3 (p=1.00). There were no significant differences in 2-year RFS between pts who received GO or not within each pre-specified subgroup.

The frequency of AEs was higher in pts treated with 7+3+GO: VOD 4 (6%) vs 0 (0%); grade 3-4 AST/ALT elevation 9 (15%) vs 10 (10%); and grade 3-4 bleeding 15 (24%) vs 11 (11%). Grade 5 bleeding was reported in 3 (5%) and 1 (1%) pts, respectively. The 60-day mortality rate was 7% both with 7+3+GO and with 7+3 (p=1.00).

With propensity score matching with IPTW, 2-year OS was 81% (95% CI, 71-92) with 7+3+GO and 68% (95% CI, 59-79) with 7+3 (p=0.39), (HR 0.73, 95% CI 0.36-1.5, p=0.39). The 2-year RFS was 73% (95% CI, 60-87) and 69% (CI 95%, 58-81) with 7+3+GO and 7+3, respectively.

Conclusion Within key subgroups of AML (CBF, NPM1mut and INT), the use of GO did not result in significant improvements in CRc, MRD negativity rates and survival outcomes in this retrospective study. VOD, liver enzymes elevation, and bleeding were more frequent with GO. Further research, including larger sample size is needed to identify pts who benefit from GO and to optimize clinical outcomes.

This content is only available as a PDF.
Sign in via your Institution