A start, stop, GO story in AML – gemtuzumab ozogamicin

Gemtuzumab ozogamicin (GO) is a monoclonal antibody-targeted chemotherapeutic agent consisting of a recombinant humanized murine CD33 antibody which is linked to calicheamicin, an anti-tumor antibiotic.¹

CD33, a 67-kDa type I transmembrane receptor, is expressed by cells of myeloid origin but not by normal hematopoietic cells. In Acute Myeloid Leukemia (AML), CD33 surface antigen is expressed in approximately 90% of leukemic blast cells.  GO binds specifically to, and is internalized by, tumor cells expressing CD33. As a result, this Antibody Drug Conjugate (ADC) can then dispense the calicheamicin to CD33-expressing tumor cells.¹

Frederick R. Appelbaum, a member of the AGP North-American Steering Committee and Irwin D. Bernstein, both of the Fred Hutchinson Cancer Research Center, Seattle, WA, have written an insightful review on the history of the recently US Food & Drug Administration approved GO in AML, specifically on the pivotal clinical studies leading to the initial approval, the re-approval of GO by the FDA and the challenges moving forward. The review was published in Blood on 11^th October 2017.²

Initial Approval of GO in AML

GO was first approved by the FDA via accelerated review in May 2000 for patients age > 60 years with CD33+ AML who are not candidates for intensive chemotherapy and it was the first ADC approved for human use. This approval was based on early clinical studies of GO monotherapy in AML patients. In a phase I dose escalation study, 40 patients with relapsed or refractory (R/R) AML were administered GO with doses ranging from 0.25–9 mg/m². Data from this study demonstrated that 20% of patients achieved a reduction in morphological detectable AML with 12.5% achieving complete remission (CR). Adverse Events (AEs) that occurred include infusion-related fevers and chills and mylosuppression. The recommended phase II dose was 9 mg/m² given at Days 1 and 14.³

Based on the data of this phase I study, three open-label multicenter phase II studies with GO at 9 mg/m² were started. Preliminary analysis of 142 patients (median age = 61 years) with CD33+ AML in first relapse who were enrolled in these three studies demonstrated an overall response rate of 30%.¹ This lead to the accelerated approval of GO by the FDA.

Withdrawal of GO

As part of the post-approval commitment, a randomized phase III Southwest Oncology Group (SWOG) S0106 study (NCT00085709) was designed to compare two different regimens of chemotherapy with or without a single dose of GO in adult patients less than 60 years old with previously untreated de novo AML commenced. Six-hundred and thirty-seven AML patients were randomized to receive daunorubicin and cytarabine (DA) plus GO (6 mg/m² on day 4) versus DA alone. Findings from this study demonstrated that there was no improvement in survival in patients treated with GO. Remarkably, there was an increase in treatment-related mortality in patients treated in the GO arm compared to the DA arm (5% vs 1% respectively, P = 0.0062).⁴ As a result of lack of clinical benefit and increased mortality, GO was voluntarily withdrawn from market on 15^th October 2010.

Re-approval of GO in CD33+ AML

Subsequent to the withdrawal of GO from the market, the pivotal phase III randomized ALFA-0701 (NCT00927498) study began along with other studies with GO. In this ALFA-0701 study, 278 untreated de novo AML patients were randomized 1:1 to receive induction therapy [DA] with daunorubicin (60 mg/m² for 3 days) and cytarabine (200 mg/m² for 7 days) plus fractionated doses of GO (3 mg/m² on days 1, 4, 7) versus DA alone. The CR rates were similar in patients between the GO arm and the control arm as was the treatment-mortality rates. However, the primary endpoint of the study which was 3-year event free survival was significantly longer in patients in the GO arm compared to the control arm (15.6 vs 9.7 months respectively, HR = 0.66, P = 0.0026).⁵

Conversely, in the phase III GOELAMS AML 2006IR study (NCT00860639), 238 adults with de novo AML were randomized to receive either standard chemotherapy with or without GO (6 mg/m²). Similar to results from the ALFA-0701 study, there were no differences in CR rates and treatment mortality rates. However, the EFS was significantly longer in patients in the GO arm (53.7% vs 27%, P = 0.03).⁶

Additionally, findings from the UK MRC AML-15 and UK NCRI AML-16 trials, which evaluated the role of GO (3 mg/m²) with induction therapy among younger and older patients respectively showed that GO significantly improved the survival and reduced relapse risk with little increase in toxicity.⁶

A meta-analysis of the ALFA-0701, S0106, GOELAMS AML2006IR, AML-15 and AML-16 studies which comprised of 3325 patients demonstrated that GO significantly reduced the risk of relapse and improved 5-year relapse free survival (OR = 0.87, P = 0.005) and overall survival (OR = 0.81, P = 0.0001). Additionally, lower doses (3 mg/m²) of GO were as effective as higher doses and had improved safety.⁶

GO was approved by the FDA on 1^st September 2017, for the treatment of adult patients with newly diagnosed CD33-positive AML. GO was also approved for adults and pediatric patients 2 years and older with R/R CD33-positive AML. The approval was based on results from the ALFA-0701 study, the phase III AML-19 study,⁷ which demonstrated that GO significantly improved the OS compared to best supportive care and the phase II Mylofrance-1 study.

Schematic representation of the start, stop, GO story of gemtuzumab ozogamicin

Conclusions

Addition of GO to induction therapy has been shown to be safe and effective in both newly diagnosed with CD33+ AML thus “establishing this as a new standard of care”. The potential of GO is still being explored in other settings, including acute promyleocytic leukemia, in which GO monotherapy and GO in combination with ATRA-ATO has been shown to induce durable response.⁸

The authors noted that future areas that might be of interest for GO include studies investigating the implications of CD33 splice variations and also studies that can help understand the mechanisms of GO better. In a recent article by Lamba et al. in July 2017, CD33 single nucleotide polymorphism, rs12459419 genotype frequency was able to predict response in patients with de novo AML. Treatment of patients with the rs12459419 CC genotype (binding to GO) lead to a significantly lower relapse risk and higher disease free survival but had no effects on patients with the CT and TT genotype. This offers the prospect that knowledge of CD33 genotype and prediction of response to GO provides opportunities to use patient genotypes for selecting CD33-targeted therapies.⁹

In conclusion, the story of GO has been a captivating one: challenges have been responded to and averted by more clinical research, and the story is set to continue with the use of genetically targeted therapy.

 Abstract

On September 1, 2017, the U.S. Food and Drug Administration (FDA) approved gemtuzumab ozogamicin for the treatment of adults with newly diagnosed CD33+ acute myeloid leukemia and for patients aged 2 years and older with CD33+ AML who have experienced a relapse or who have not responded to initial treatment. This signals a new chapter in the long and unusual story of gemtuzumab ozogamicin (GO), which was the first antibody-drug conjugate approved for human use by the FDA.