1033P - Quantitative Assessment of Inotuzumab Ozogamicin (InO) Response Relative to Investigator’s Choice of Chemotherapy (ICC) in Adults With Relapsed or...

Date 09 September 2017
Event ESMO 2017 Congress
Session Poster display session
Topics Cancer in Adolescents
Leukaemia
Haematologic Malignancies
Presenter Ana Ruiz-Garcia
Citation Annals of Oncology (2017) 28 (suppl_5): v355-v371. 10.1093/annonc/mdx373
Authors A. Ruiz-Garcia1, E. Vandendries2, D.J. Deangelo3, H.M. Kantarjian4, J. Boni5
  • 1Clinical Pharmacology, Pfizer Inc, 92121 - San Diego/US
  • 2Clinical Research And Development Hematology, Pfizer Inc, Cambridge/US
  • 3Center For Hematologic Oncology, Dana-Farber Cancer Institute, Boston/US
  • 4Department Of Leukemia, MD Anderson Cancer Center, Houston/US
  • 5Clinical Pharmacology, Pfizer Inc, Collegeville/US

Abstract

Background

InO is a humanized CD22-targeted antibody conjugated to N-acetyl-γ-calicheamicin, a potent cytotoxic antibiotic. InO was administered to adult patients with R/R CD22+ B-cell ALL in a phase 1/2 (B1931010) study, and a phase 3 (INO-VATE) study that compared single-agent InO with ICC. The goal of this analysis was to quantify differences in response for the endpoints complete response (CR)/CR with incomplete hematologic recovery (CRi) and minimal residual disease-negativity (MRD (-)) for patients treated with InO relative to ICC.

Methods

The efficacy endpoints analyzed were CR/CRi per investigator’s assessment and MRD (-). The modeling analyses were performed using generalized binomial logistic regression, which allows constructing a linear continuous predictor for probabilities of response ranging from 0%–100%. 2 treatment arms were considered: single agent InO or ICC (fludarabine, cytarabine, granulocyte colony-stimulating factor; cytarabine with mitoxantrone; or high-dose cytarabine). Additional potential predictors of response (eg, baseline demographic/patient characteristics, laboratory values) were also tested.

Results

For the CR/CRi efficacy endpoint, only 3 variables were statistically significant predictors of achieving CR/CRi: treatment arm, baseline ECOG (BECOG) performance status, and baseline absolute blasts in peripheral blood (BLSTABL). For the MRD (-) endpoint, 5 variables were statistically significant predictors of achieving MRD(-): treatment arm, BECOG performance status, baseline cytogenetic characteristics, prior hematopoietic stem cell transplant before study therapy, and BLSTABL.Table:

1033P

Endpoint, n (%)CategoryStudy B1931010 InO (n = 72)INO-VATE InO Arm (n = 162)INO-VATE ICC Arm (n = 143)Total (N = 377)
MRD-negativity*No31 (43)59 (36)115 (80)205 (54)
Yes41 (57)97 (60)23 (16)161 (43)
Missing0 (0)6 (4)5 (3)11 (3)
CR/CRiNo23 (32)42 (26)95 (66)160 (42)
Yes49 (68)120 (74)48 (34)217 (58)
*

When CR/CRi was not achieved and MRD was missing, MRD was considered not achieved.

Conclusions

The odds of achieving CR/CRi and MRD(-) with InO were approximately 7 and 13 times higher, respectively, than ICC.

Clinical trial identification

NCT01363297, NCT01564784

Legal entity responsible for the study

Pfizer Inc

Funding

Pfizer Inc

Disclosure

A. Ruiz-Garcia, E. Vandendries: Employee of and owns stocks in Pfizer Inc. D.J. DeAngelo: Served on advisory boards for Pfizer Inc. H.M. Kantarjian: Received research grants from Pfizer, Amgen, Astex, Novartis, and Bristol-Myers Squib. J. Boni: Was an employee of Pfizer Inc at the time the study was conducted.