An Update on Lovotibeglogene Autotemcel (Lovo-cel) Clinical Trials for Sickle Cell Disease (SCD) and Analysis of Early Predictors of Response to Lovo-Cel

Introduction: Lovo-cel gene addition therapy, which consists of transplantation of autologous hematopoietic stem and progenitor cells (HSPCs) transduced with the BB305 lentiviral vector (LVV) encoding a functional anti-sickling β-globin gene, is approved for patients ≥12 y with SCD and a history of vaso-occlusive events (VOEs). Here, we report updated outcomes for participants treated with lovo-cel from HGB-206 Group C (NCT02140554; phase 1/2) and HGB-210 (NCT04293185; phase 3) and examine how early pharmacodynamic (PD) parameters predict long-term clinical responses to lovo-cel in HGB-206 (Groups A, B, and C) and HGB-210.

Methods: Designs of HGB-206 and HGB-210 were previously described (Kanter, et al. Am J Hematol. 2023;98:11-22; Kanter, et al. N Engl J Med. 2022;386:617-628); upon completion, participants are eligible for the ongoing LTF-307 extension (NCT04628585). HGB-210 is ongoing and enrolling participants aged <12 y. Data shown are from lovo-cel trial participants treated as of Feb 2024; updated data will be presented. A post hoc, exploratory analysis utilized a univariate logistic regression model to estimate the association between post–lovo-cel PD parameters and clinical outcomes. PD parameters: % of LVV transduced cells (%LVV+), % of red blood cells expressing HbA^T87Q, and % of peripheral blood (PB) globins that are HbA^T87Q. Outcomes: complete resolution of adjudicated VOEs (VOE-CR), severe VOEs (sVOE-CR) including VOEs requiring hospitalization, and globin response (GR). A receiver operating characteristic (ROC) curve determined the optimal HbA^T87Q cutoff value, maximizing predictive power for VOE-CR/sVOE-CR. A linear regression model examined the association between drug product (DP) attributes and PB HbA^T87Q.

Results: As of Feb 2024, 36 HGB-206 Group C and 19 HGB-210 participants (n=55; male, 61.8%; median [range] age, 21 [9-38] y) received lovo-cel per the current HSPC mobilization and manufacturing process, including 14 participants <18 y. Median (range) time to last follow-up was 41.9 (0.9-72.3) mo. Of 34 VOE-evaluable participants (≥18 mo follow-up and ≥4 VOEs ≤2 y pre-enrollment), 30 (88.2%) and 32 (94.1%) achieved VOE-CR and sVOE-CR in the 6-18 mo post infusion. Among 46 evaluable participants, 41 (89.1%) achieved GR. Median (range) % PB HbA^T87Q 6 mo post infusion was 49% (26%-63%). Results were similar for adult (≥18 y) and pediatric (<18 y) subgroups. HbA^T87Q remained stable through last follow-up, as did PB vector copy number (median >1 c/dg). Median (range) total Hb level at last visit was 12.3 (8.4-15.3) g/dL and was stable without transfusion support post engraftment. The lovo-cel treatment regimen safety profile reflects the known effects of underlying SCD and myeloablative conditioning and is similar across age groups.

Prior multivariate analyses demonstrated significant correlation between DP characteristics, PD, and hematologic parameters (Kinney, et al. Transplant Cell Ther. 2024;30(Supp 2):S231) but did not evaluate clinical outcomes. Exploratory analyses of HGB-206 (all groups) and HGB-210 data showed long-term PD and biologic responses can be predicted early, with measurements at 6 mo significantly correlated to those at last follow-up for both Hb (r=0.8, P<0.001) and HbA^T87Q (r=0.97, P<0.001). PD parameters are highly predictive (P<0.05) of clinical outcomes, including achieving GR and resolution of VOEs 6-18 mo post infusion. Increased HbA^T87Q is associated with VOE-CR (odds ratio [OR], 1.09; 95% CI, 1.03-1.15) and sVOE-CR (OR, 1.16; 95% CI, 1.06-1.27). A univariate logistic regression model identified a highly significant correlation between achieving VOE-CR (area under the curve [AUC]=0.78) and sVOE-CR (AUC=0.97) above 33% PB HbA^T87Q. The HbA^T87Q level is established at the DP level, as PD response is significantly correlated with the DP %LVV+ cells (r=0.86, P<0.001). In participants from HGB-206 Group C and HGB-210, the median (range) %LVV+ cells is 83% (63%-93%), corresponding to a median 47% PB HbA^T87Q. Among participants who are not VOE evaluable (<18 mo follow-up; n=12), the median (range) % PB HbA^T87Q is 53% (46%-59%) at 6 mo.

Conclusions: One-time lovo-cel treatment results in sustained HbA^T87Q production and eliminated VOEs and sVOEs in a majority of participants. Models developed in this post hoc analysis enable prediction of the likelihood of VOE-CR and sVOE-CR using measurements as early as 6 mo post treatment.